Arkansas Rice Research Study 2015

25th Anniversary Edition
B.R. Wells
Arkansas
Rice Research
Studies 2015
Bobby Wells
R.J. Norman and K.A.K. Moldenhauer, editors
ARKANSAS AGRICULTURAL EXPERIMENT STATION

August 2016 Research Series 634
This publication is available on the Internet at http://arkansasagnews.uark.edu/1356.htm
Cover Photo: Bobby Wells evaluating a rice study for nutrient deficiencies (circa 1985). Photo credit: University
of Arkansas System Division of Agriculture.
Layout and editing by Marci A. Milus

Technical editing and cover design by Gail Halleck
Arkansas Agricultural Experiment Station, University of Arkansas System Division of Agriculture, Fayetteville.
Mark J. Cochran, Vice President for Agriculture. Clarence E. Watson, Associate Vice-President for
AgricultureResearch and Director, AAES. MG300CC2015/CS6.
The University of Arkansas System Division of Agriculture offers all its Extension and Research programs and
services without regard to race, color, sex, gender identity, sexual orientation, national origin, religion, age, dis-
ability, marital or veteran status, genetic information, or any other legally protected status, and is an Affirma-
tive Action/Equal Opportunity Employer.
ISSN:1941-2177 CODEN:AKAMA6
25th Anniversary Edition
B.R. Wells
ARKANSAS RICE
Research Studies
2015
R.J. Norman and K.A.K. Moldenhauer, editors
University of Arkansas System

Division of Agriculture
Arkansas Agricultural Experiment Station
Fayetteville, Arkansas 72701
DEDICATED IN MEMORY OF
Bobby R. Wells
Bobby R. Wells was born July 30, 1934, at Wickliffe,

Ky. He received his B.S. degree in agriculture from Murray
State University in 1959, his M.S. degree in agronomy from
the University of Arkansas in 1961, and his Ph.D. in soils
from the University of Missouri in 1964. Wells joined the
faculty of the University of Arkansas in 1966 after two years as an assistant professor
at Murray State University. He spent his first 16 years at the University of Arkansas
System Division of Agricultures Rice Research and Extension Center near Stuttgart. In
1982, he moved to the University of Arkansas Department of Agronomy in Fayetteville.
Wells was a world-renowned expert on rice production with special emphasis
on rice nutrition and soil fertility. He was very active in the Rice Technical Working
Group (RTWG), for which he served on several committees, chaired and/or moderated
Rice Culture sections at the meetings, and was a past secretary and chairman of the
RTWG. He loved being a professor and was an outstanding teacher and a mentor to
numerous graduate students. Wells developed an upper-level course in rice production
and taught it for many years. He was appointed head of the Department of Agronomy
in 1993 and was promoted to the rank of University Professor that year in recognition
of his outstanding contributions to research, service, and teaching.
Among the awards Wells received were the Outstanding Faculty Award from the
Department of Agronomy (1981), the Distinguished Rice Research and/or Education
Award from the Rice Technical Working Group (1988), and the Outstanding Researcher
Award from the Arkansas Association of Cooperative Extension Specialists (1992). He
was named a Fellow in the American Society of Agronomy (1993) and was awarded,
posthumously, the Distinguished Service Award from the RTWG (1998).
Wells edited this series when it was titled Arkansas Rice Research Studies from
the publications inception in 1991 until his death in 1996. Because of Wells contribu-
tion to rice research and this publication, it was renamed the B.R. Wells Rice Research
Studies in his memory starting with the 1996 publication.
FOREWORD
Research reports contained in this publication may represent preliminary or only

a single year of results; therefore, these results should not be used as a basis for long-
term recommendations.
Several research reports in this publication will appear in other Arkansas Agri-
cultural Experiment Station publications. This duplication is the result of the overlap in
research coverage between disciplines and our effort to inform Arkansas rice producers
of all the research being conducted with funds from the rice check-off program. This
publication also contains research funded by industry, federal, and state agencies.
Use of products and trade names in any of the research reports does not constitute
a guarantee or warranty of the products named and does not signify that these products
are approved to the exclusion of comparable products.
ACKNOWLEDGMENTS
Most of the research results in this publication were made possible through
funding provided by the rice farmers of Arkansas and administered by the Arkansas
Rice Research and Promotion Board. We express sincere appreciation to the farmers
and to the members of the Arkansas Rice Research and Promotion Board for their vital
financial support of these programs.
The Arkansas Rice Research and Promotion Board

Joe Christian Jonesboro
Jay Coker Stuttgart (Vice-Chairman)
David Gairhan Jonesboro
Marvin Hare Jr. Newport
Rich Hillman Carlisle
Bobby Hoard Marianna (Secretary/Treasurer)
Bryan Moery Wynne (Chairman)
Roger Pohlner Fisher
Wayne Wiggins Jonesboro
CONTENTS
OVERVIEW AND VERIFICATION
Trends in Arkansas Rice Production, 2015

J.T. Hardke........................................................................................................... 13
2015 Rice Research Verification Program

R. Baker, R.S. Mazzanti, J.T. Hardke, K.B. Watkins, and R. Mane...................... 24
BREEDING, GENETICS, AND PHYSIOLOGY
Evaluation of Advanced Semi-Dwarf Medium-Grain and Long-Grain

Breeding Lines at Three Arkansas Locations
B.A. Beaty, J.M. Bulloch, M.W. Duren, Y.D. Liyew,
S.D. Clark, and X. Sha......................................................................................... 47
Kompetitive Allele-Specific Polymerase Chain Reaction (KASPTM)

Marker-Assisted Selection for the Development of Rice Varieties
V.A. Boyett, V.I. Thompson, X. Sha, K.A.K. Moldenhauer, D.K.A. Wisdom,
J.M. Bulloch, and H.H.M. Moldenhauer............................................................. 56
Titan, a Very Early-Maturing, High-Yielding, and High-Quality Conventional

Medium-Grain Rice Variety
J.M. Bulloch, B.A. Beaty, X. Sha, K.A.K. Moldenhauer, J.W. Gibbons,
J.T. Hardke, R.J. Norman, C.E. Wilson Jr., Y.A. Wamishe,
T.J. Siebenmorgen, D.K.A. Wisdom, M.M. Blocker, D.L. McCarty,
V.A. Boyett, D.L. Frizzell, and E. Castaneda-Gonzalez....................................... 62
Testing Uniform Regional Rice Nursery Varieties for Resistance to

Rice Blast Disease
C. Feng, and J.C. Correll.................................................................................... 66
Diamond, a High Yielding, Very Short Season, Long-Grain Rice Variety

K.A.K. Moldenhauer, X, Sha, J.T. Hardke, R.J. Norman, Y.A. Wamishe,
M.M. Blocker, D.L. McCarty, C.H. Northcutt, D.G. North,
D.K.A. Wisdom, V.A. Boyett, D.L. Frizzell, J.M. Bulloch,
E. Castaneda-Gonzalez, B.A. Beaty, C.D. Kelsey, and S.B. Belmar.................... 73
Genetic Basis of Altered Grain Quality in Different Rice Cultivars Under
High Nighttime Termperature
V. Ramegowda, S. Srivastava, J. Thomas, C. Gupta, S. Basu, P. Counce,
Y.-J. Wang, T.J. Siebenmorgen, K.A.K. Moldenhauer, and A. Pereira................. 79
Development of Superior Medium-Grain and Long-Grain Rice Varieties

for Arkansas and the Mid-South
X. Sha, K.A.K. Moldenhauer, B.A. Beaty, J.M. Bulloch, D.K.A. Wisdom,
M.M. Blocker, D.L. McCarty, V.A. Boyett, D.L. Frizzell, J.T. Hardke,
and C.E. Wilson Jr............................................................................................... 86
Development of Male Sterile Line for Hybrid Rice Production

E. Shakiba, K.A.K. Moldenhauer, G.L. Berger, D.G. North, V.A. Boyett,
V.I. Thompson, M.M. Blocker, D.L. McCarty, and C.H. Northcutt...................... 95
Development of Aromatic Rice Varieties

D.K.A. Wisdom, K.A.K. Moldenhauer, C.E. Wilson Jr., X. Sha,
J.M. Bulloch, B.A. Beaty, M.M. Blocker, and V.A. Boyett.................................. 100
PEST MANAGEMENT: DISEASES
Rice Breeding and Pathology Technical Support

C.D. Kelsey, S.B. Belmar, K.A.K. Moldenhauer, and Y.A. Wamishe.................. 103
Studies on Rice Bacterial Panicle Blight Diseases Relative

to Soil Depth, Crop Residue, Dew Environment, and Potassium Levels
Y.A. Wamishe, T. Mulaw, T. Gebremariam, S.B. Belmar, and C.D. Kelsey.........110
Monitoring Bacterial Panicle Blight Disease of Rice and Germplasm

Evaluation for Resistance in Arkansas in 2015
Y.A. Wamishe, T. Mulaw, Y. Jia, T. Gebremariam,
S.B. Belmar, and C.D. Kelsey............................................................................ 120
Understanding Autumn Decline, Evaluating Rice Varieties for

Resistance, and Identifying New Strategies to Reduce
the Phenomenon in Problematic Rice Fields
Y.A. Wamishe, T.L. Roberts, J.T. Hardke, J. Allen, T. Gebremariam,
C.D. Kelsey, T. Mulaw, and S.B. Belmar........................................................... 126
PEST MANAGEMENT: INSECTS
Efficacy of Selected Insecticides for Control of Rice Stink Bug, Oebalus pugnax,
in Arkansas, 2015
H.M. Chaney, G.M. Lorenz, N.M. Taillon, W.A. Plummer,
and J.L. Black.................................................................................................... 132
Evaluation of Rice Kernel Damage and Yields Due to Rice

Stink Bug, Oebalus pugnax, Population and Infestation Timing
T.L. Clayton, G.M. Lorenz, J.T. Hardke, G.J. Lee,
E. Castaneda-Gonzalez, D.L. Frizzell, and H.M. Chaney................................. 137
Potential Exposure of Honey Bees to Neonicotinoid Insecticides in Rice

G.M. Lorenz, J.T. Hardke, T.L. Clayton, N.M. Taillon, D.L. Frizzell,
J. Black, W.A. Plummer, and H.M. Chaney....................................................... 146
Value of Insecticide Seed Treatments in Arkansas Rice

N.M. Taillon, G.M. Lorenz, W.A. Plummer, H.M. Chaney, and J. Black........... 152
PEST MANAGEMENT: WEEDS
Residual Weed Control and Crop Response to Pethoxamid Systems in Rice

R.C. Doherty, L.T. Barber, J.K. Norsworthy, and Z.T. Hill................................ 159
Evaluation of Very Long-Chain Fatty Acid-Inhibiting

Herbicides in Arkansas Rice
J.A. Godwin Jr., J.K. Norsworthy, R.C. Scott,
L.T. Barber, M.L. Young, and M.W. Duren........................................................ 163
Sharpen Tank-Mixtures with Rice Herbicides for Barnyardgrass

Control in ProvisiaTM Rice
R.R. Hale, J.K. Norsworthy, J.A. Godwin Jr., N.R. Stepping,
C.J. Meyer, R.C. Scott, L.T. Barber, and J. Schultz............................................ 169
Obey and Command Herbicide Programs

Z.T. Hill, L.T. Barber, J.K. Norsworthy, and R.C. Doherty................................ 174
Optimizing Quizalofop Rate Structure for Sequential

Application in ProvisiaTM Rice
Z.D. Lancaster, J.K. Norsworthy, S.M. Martin, M.L. Young,
R.C. Scott, and L.T. Barber................................................................................ 180
Effects of CruiserMaxx Rice on Rice Tolerance to Pre-Emergence Herbicides
S.M. Martin, J.K. Norsworthy, R.C. Scott, G.M. Lorenz, and J.T. Hardke........ 185
Evaluation of ProvisiaTM Herbicide Tank Mixtures

M.S. McCown, L.T. Barber, J.K. Norsworthy, Z.D. Lancaster,
J. Green, and M.R. Miller.................................................................................. 190
Influence of Herbicide Rate, Application Volume, and Adjuvant

Use on Efficacy of Rinskor Active
M.R. Miller and J.K. Norsworthy...................................................................... 196
Evaluation of Topramezone (Armezon/Impact) for Rice Weed Control

R.C. Scott, J.K. Norsworthy, J.C. Moore, and B.M. Davis................................ 201
Rice Flatsedge Resistance to Acetolactate Synthase-Inhibiting Herbicides

P. Tehranchian, J.K. Norsworthy, R.C. Scott, and L.T. Barber.......................... 206
Evaluation of a Benzobicyclon Plus Halosulfuron Premix

for Weed Control in Arkansas Ric
M.L. Young, J.K. Norsworthy, C.A. Sandoski, R.C. Scott, and L.T. Barber....... 212
RICE CULTURE
2015 Degree Day 50 Thermal Unit Thresholds for New Rice Cultivars
and Seeding Date Studies
E. Castaneda-Gonzalez, J.T. Hardke, D.L. Frizzell, G.J. Lee,
T.L. Clayton, R.J. Norman, K.A.K. Moldenhauer, and X. Sha........................... 218
Continued Validation of the Nitrogen Soil Test for Rice

on Clay Soils in Arkansas
J.T. Davidson, T.L. Roberts, R.J. Norman, C.E. Greub,
N.A. Slaton, and J.T. Hardke............................................................................. 226
Evaluation of Alternative Nitrogen Fertilizer Application

Timings in Four Water Management Regimes
D.L. Frizzell, J.T. Hardke, T.L. Roberts, R.J. Norman,
E. Castaneda-Gonzalez, G.J. Lee, and T.L. Clayton......................................... 233
Effect of Delayed Nitrogen Fertilization into the Floodwater

D.L. Frizzell, J.T. Hardke, T.L. Roberts, R.J. Norman,
E. Castaneda-Gonzalez, G.J. Lee, and T.L. Clayton......................................... 239
Validation of Soil-Test Based Fertilizer Recommendations for Rice
M. Fryer, N.A. Slaton, T.L. Roberts, J. Hardke, R.J. Norman,
R.E. DeLong, Y. Liyew, D.L. Frizzell, J. Pace, and M.W. Duren....................... 243
Effects of Three Different Alternate Wetting and

Drying Regimes in Rice Cultivation on Yield, Water Use,
and Water Use Efficiency in a Clay Soil During a Wet Year
J.P. Gaspar, C.G. Henry, M.W. Duren, A.P. Horton, and H. James................... 251
Grain Yield Response of Four New Rice Cultivars to Seeding Rate

J.T. Hardke, D.L. Frizzell, E. Castaneda-Gonzalez,
T.L. Clayton, G.J. Lee, and R.J. Norman........................................................... 261
Arkansas Rice Performance Trials, 2013-2015

J.T. Hardke, D.L. Frizzell, E. Castaneda-Gonzalez,
G.J. Lee, K.A.K. Moldenhauer, X. Sha, Y. Wamishe,
R.J. Norman, M.M. Blocker, J.A. Bulloch, B.A. Beaty,
R.S. Mazzanti, R. Baker, W. Kirkpatrick, M. Duren, and Y. Liyew.................... 267
Methane Emissions from Direct-Seeded, Delayed-Flood Rice

Production as Influenced by Cultivar and Water Management
J.J. Humphreys, K.R. Brye, A.D. Smartt, J.T. Hardke,
D.L. Frizzell, E. Castaneda-Gonzalez, G.J. Lee, and R.J. Norman.................. 275
Utilization of On-Farm Testing to Evaluate Rice Cultivars, 2015

G.J. Lee, E. Castaneda-Gonzalez, D.L. Frizzell, J.T. Hardke,
Y.A. Wamishe, and R.J. Norman........................................................................ 282
Grain Yield Response of Six New Rice Cultivars to Nitrogen Fertilization

R.J. Norman, T.L. Roberts, J.T. Hardke, N.A. Slaton,
K.A.K. Moldenhauer, X. Sha, D.L. Frizzell, M.W. Duren,
E. Castaneda-Gonzalez, and G.J. Lee............................................................... 295
Response of Two Rice Cultivars to Midseason Nitrogen Fertilizer

Application Timing in the Presence of a Zinc Deficiency
R.J. Norman, J.T. Hardke, T.L. Roberts, N.A. Slaton,
D.L. Frizzell, A.D. Smartt, E. Castaneda-Gonzalez, and G.J. Lee.................... 306
Rice Cultivar Yield Response to Delayed Preflood Nitrogen

Application and Flood Establishment Time
N.A. Slaton, T. Richmond, J.T. Hardke, T.L. Roberts,
R.J. Norman, Y. Liyew, and D.L. Frizzell........................................................... 312
Response of Two Rice Cultivars to Midseason Nitrogen
Fertilizer Application Timing
A.D. Smartt, R.J. Norman, J.T. Hardke, T.L. Roberts, N.A. Slaton,
D.L. Frizzell, M.W. Duren, E. Castaneda-Gonzalez, and G.J. Lee................... 319
Summary of Nitrogen Soil Test for Rice (N-STaR)

Nitrogen Recommendations in Arkansas During 2015
S.M. Williamson, T.L. Roberts, C.L. Scott, R.J. Norman,
N.A. Slaton, and J. Shafer.................................................................................. 326
RICE QUALITY AND PROCESSING
Impact of Storage Duration, Temperature, and Moisture Content

on Mold Growth on Hybrid Rice
G.G. Atungulu and S. Thote............................................................................... 334
Simulation and Validation of On-Farm In-Bin Drying

and Storage of Rough Rice
G.G. Atungulu and H. Zhong............................................................................. 342
Optimization of Process Parameters in Rough Rice Drying

Using Industrial Microwave
G.G. Atungulu, G.A. Olatunde, D.L. Smith, S. Sadaka, and S. Rogers............. 353
Impacts of Rough Rice Temperature and Moisture Content

on Laboratory Milling Yields
B.C. Grigg, C.D. Shook, and T.J. Siebenmorgen............................................... 362
Effects of Temperature, Moisture Content, and Rough Rice Storage

Duration on Milling Properties and Discoloration
K.N. Haydon and T.J. Siebenmorgen................................................................. 369
Variable Impact of Nighttime Air Temperatures on Rice Chalk

and Milling Properties Due to Heading Date
K.N. Haydon, T.J. Siebenmorgen, and P.A. Counce.......................................... 378
Characterization of Broken Rice Kernels Caused by

Moisture-Adsorption Fissuring
S. Mukhopadhyay and T.J. Siebenmorgen......................................................... 386
Observing Fissures in Rough Rice Kernels Using X-Ray

Imaging: Preliminary Observations
Z.R. Odek, B. Prakash, and T.J. Siebenmorgen................................................. 395
Kernel and Starch Properties of United States-Grown and
Imported Medium- and Short-Grain Rice Cultivars
J. Patindol, J.-R. Jinn, Y.-J. Wang, and T.J. Siebenmorgen............................... 405
ECONOMICS
Trans-Pacific Partnership: What Can it Mean for the U.S. Rice Sector?
A. Durand-Morat and E.J. Wailes..................................................................... 414
Rice Enterprise Budgets and Production Economic Analysis

W.A. Flanders.................................................................................................... 422
Cost-Effective Use of Water? Factors Influencing the Use of Irrigation

Technologies and Water Management Practices by Arkansas Producers
Q. Huang, K. Kovacs, and Y. Xu........................................................................ 426
Economic Simulation Analysis of Margin Protection Crop

Insurance in Arkansas Rice Production
R.U. Mane and K.B. Watkins............................................................................. 433
World and U.S. Rice Outlook: Deterministic and Stochastic

Baseline Projections, 2015-2025
E.J. Wailes and E.C. Chavez.............................................................................. 441
The Economics of Methane Emissions in Arkansas Rice Production

F. Tisboe, L. Nalley, K. Brye, B. Dixon, and A. Shew........................................ 456
Trends in Arkansas Rice Production, 2015

J.T. Hardke1
ABSTRACT
Arkansas is the leading rice-producing state in the United States. The state represents
49.0% of total U.S. rice production and 50.0% of the total acres planted to rice in 2015.
Rice cultural practices vary across the state and across the U.S. However, these practices
are also dynamic and continue to evolve in response to changing political, environmental,
and economic times. This survey was initiated in 2002 to monitor and record changes in
the way Arkansas rice producers approach their livelihood. The survey was conducted
by polling county extension agents in each of the counties in Arkansas that produce
rice. Questions included topics such as tillage practices, water sources and irrigation
methods, seeding methods, and precision leveling. Information from the University of
Arkansas System Division of Agricultures Rice Degree-Day 50 (DD50) Program was
included to summarize variety acreage distribution across Arkansas. Other data was
obtained from the USDA National Agricultural Statistics Service.
INTRODUCTION
Arkansas is the leading rice-producing state in the United States in terms of
acreage planted, acreage harvested, and total production. Each year, rice planting typi-
cally ranges from late March into early June with harvest occurring from late August
to early November. Rice production occurs across a wide range of environments in
the state. The diverse conditions under which rice is produced leads to variation in the
adoption and utilization of different crop management practices. To monitor and better
understand changes in rice production practices, including adoption of new practices, a
survey was initiated in 2002 to record annual production practices. Information obtained
through this survey helps to illustrate the long-term evolution of cultural practices for
rice production in Arkansas. It also serves to provide information to researchers and
extension personnel about the ever-changing challenges facing Arkansas rice producers.
1
Rice Extension Agronomist, Department of Crop, Soil, and Environmental Sciences, Stuttgart.
13
AAES Research Series 634
PROCEDURES
A survey has been conducted annually since 2002 by polling county agriculture
extension agents in each of the counties in Arkansas that produce rice. Questions were
asked concerning topics such as tillage practices, water sources and irrigation methods,
seeding methods, and precision leveling. Acreage, yield, and crop progress information
was obtained from the USDA National Agricultural Statistics Service (USDA-NASS,
2016). Rice cultivar distribution was obtained from summaries generated from the
University of Arkansas System Division of Agricultures Rice Degree-Day 50 (DD50)
program enrollment.
RESULTS AND DISCUSSION

Rice acreage by county is presented in Table 1 with distribution of the most
widely produced cultivars. RiceTec CLXL745 was the most widely planted cultivar in
2015 at 19.9% of the acreage, followed by RiceTec XL753 (14.5%), Jupiter (14.4%),
Roy J (13.1%), CL151 (12.4%), LaKast (5.0%), Mermentau (4.1%), CL111 (3.8%),
RiceTec CLXL729 (3.2%), and Wells (1.6%). Additional cultivars of importance in
2015, though not shown in the table, were CL271, RiceTec XL723, Francis, Taggart,
CL152, and Caffey.
Arkansas producers planted 1,306,000 acres of rice in 2015 which accounted for
50.0% of the total U.S. rice crop in 2015 (Table 2). The States average yield of 7340
lb/acre (163 bu/acre) represented a 220 lb/acre reduction compared to 2014. Despite the
decline, average yields in Arkansas still represented the second highest average in the
U.S. behind California in 2015. In addition, 2015 average yields for Arkansas (163 bu/
acre) were the fourth highest on record behind 2013 (168 bu/acre), 2014 (168 bu/acre),
and 2012 (166 bu/acre). The total rice produced in Arkansas during 2015 was 94.34
million hundredweight (cwt). This represents 49.0% of the 192.3 million cwt produced
in the U.S. during 2015. Over the past 3 years, Arkansas has produced 47.3% of all
rice produced in the U.S. The six largest rice-producing counties in Arkansas during
2015 included Poinsett, Lawrence, Arkansas, Cross, Jackson, and Lonoke, representing
41.7% of the states total rice acreage (Table 1).
Planting in 2015 started behind the 5-year state average due to cold, wet condi-
tions throughout March and April (Fig. 1). Planting progress was only 37% by 26 April
in 2015 compared to 59% planting progress averaged across the previous five years.
Dry conditions resulted in significant planting progress during the last week of April
and first week of May as planting progress jumped from 37% to 86% over this 2-week
period. Planting was almost fully complete by 1 June. While planting progress was no-
tably delayed by early-season weather, mild and extremely dry conditions led to harvest
progressing at a rate almost identical to the 5-year average (Fig. 2). About 57% of the
crop had been harvested by 20 September compared with 56% harvest progress on the
same date in the 5-year average. Harvest progress was complete (100%) by 1 November.
Over 60% of the rice produced in Arkansas was planted using conventional
tillage methods in 2015 (Table 3). This usually involves fall tillage when the weather
cooperates, followed by spring tillage to prepare the seedbed. The remainder of rice
14
B.R. Wells Arkansas Rice Research Studies 2015
acres were planted using stale seedbed (30.1%) or no-till (6.3%) systems. True no-till
rice production is not common but is done in a few select regions of the state.
The majority (53.6%) of rice is still produced on silt loam soils (Table 3). Rice
production on clay or clay loam soils (20.6% and 20.9%, respectively) has become
static over recent years after steadily increasing through 2010. These differences in
soil type present unique challenges in rice production such as tillage practices, seeding
rates, fertilizer management, and irrigation.
Rice most commonly follows soybean in rotation, accounting for 72.3% of the
rice acreage (Table 3). Approximately 21% of the acreage in 2015 was planted following
rice, with the remainder made up of rotation with other crops including cotton, corn,
grain sorghum, wheat, and fallow. The majority of the rice in Arkansas is produced uti-
lizing a dry-seeded, delayed-flood system with only 5.5% using a water-seeded system.
Annually, approximately 85% of all the Arkansas rice acreage is drill-seeded with the
remaining acreage broadcast-seeded (dry-seeded and water-seeded).
Irrigation water is one of the most precious resources for rice producers in Arkan-
sas. Reports of diminishing supplies have prompted many producers to develop reservoir
and/or tailwater recovery systems to reduce the waste by collecting and re-using all
available water. Simultaneously, producers have tried to implement other conservation
techniques to preserve the resource vital to continued production. Groundwater is used
to irrigate 76.4% of the rice acreage in Arkansas with the remaining 23.6% irrigated
with surface water obtained from reservoirs or streams and bayous (Table 3).
During the mid-1990s, the University of Arkansas System Division of Agricul-
ture began educating producers on multiple-inlet irrigation which uses poly-tubing as
a means of irrigating rice to conserve water and labor. As of 2015, rice farmers utilize
this practice on 40.6% of the rice acreage (Table 3). Most remaining acreage is still
irrigated with conventional levee and gate systems. A small percentage of rice acreage
is produced in more upland conditions utilizing furrow irrigation systems. Intermittent
flooding is another means of irrigation increasing in interest recently as a means to reduce
pumping costs and water use; but the practice accounts for little acreage at this time.
Stubble management is important for preparing fields for the next crop, particularly
in rice following rice systems. Several approaches are utilized to manage the rice straw
for the next crop, including tillage, burning, rolling, and winter flooding. In 2015, 43.5%
of the acreage was burned, 39.0% was tilled, 26.7% was rolled, and 20.4% was winter
flooded (Table 3). Combinations of these systems are used in many cases. For example,
a significant amount of the acreage that is flooded during the winter for waterfowl will
also be rolled. Some practices are inhibited by fall weather, but in 2015 acreage where
the stubble was burned noticeably increased as dry fall conditions permitted more of
this stubble management practice to take place.
Pest management is vital to preserve both yield and quality in rice. Foliar fun-
gicide applications were made on 52.5% of rice acres in 2015 (Table 3). This number
was lower than the previous year as warm and dry late season conditions in southern
Arkansas reduced disease development. Nearly 36% of rice acres received a foliar
insecticide application due to rice stink bug infestation levels which were low and
similar to 2014. Insecticide seed treatments were used on 67.4% of rice acreage as
producers continue to adopt this technology more widely each year due to its benefits
15
for both insect control and improved plant growth and vigor. However, in 2015, use of
insecticide seed treatments was likely lower than indicated in this survey as growers
attempted to reduce input costs.
Clearfield rice continues to play a significant role in rice production in Arkansas.
This technology (all cultivars combined) accounted for 44% of the total rice acreage
in 2015 (Fig. 3). This represents a 5% decrease in Clearfield rice acreage compared to
2014 and the fourth consecutive year of acreage decline. Proper stewardship of this
technology will be the key to its continued success on the majority of rice acres. In
areas where stewardship has been poor, imadazolinone-resistant barnyardgrass has
been discovered. Evidence of these resistant populations may have served to reduce the
number of Clearfield acres by emphasizing the negative effects of improper technol-
ogy management. In addition, multiple years of this technology and crop rotation have
likely cleaned up many red rice fields to the point where they can be safely returned to
conventional rice production.
SIGNIFICANCE OF FINDINGS
State average yields over the past 20 years in Arkansas have increased from an
average of 120 bu/acre in 1993-1995 to an average of 166 bu/acre in 2013-2015, an
increase of 46 bu/acre. This increase can be attributed to the development and adop-
tion of more productive cultivars and improved management practices, including
better herbicides, fungicides, and insecticides, improved water management through
precision-leveling and multiple-inlet irrigation, improved fertilizer efficiency, and in-
creased understanding of other practices such as seeding dates and tillage. Collecting
this kind of information regarding rice production practices in Arkansas is important
for researchers to understand the adoption of certain practices as well as to understand
the challenges and limitations faced by producers in field situations.
ACKNOWLEDGMENTS
The author would like to extend thanks to the rice farmers of Arkansas who provide
support through the rice check-off program administered by the Arkansas Rice Research
and Promotion Board and the support provided by the University of Arkansas System
Division of Agriculture; all of the county agents who participated in this survey; and the
members of the Rice Agronomy crew: Donna Frizzell, Eduardo Castaneda-Gonzalez,
Garrett Lee, Tara Clayton, Chuck Pipkins, Ralph Mazzanti, and Ron Baker.
LITERATURE CITED
USDA-NASS. 2016. United States Department of Agriculture National Agricultural
Statistics Service Crop Production 2015 Summary. Access date: 12 Feb. 2016.
Available at: http://usda.mannlib.cornell.edu/usda/current/CropProdSu/Crop-
ProdSu-01-12-2016.pdf
16
This page left blank intentionally to facilitate the following tables.
17
Table 1. 2015 Arkansas

Harvested acreagea Medium-grain Long-grain-
County 2014 2015 Jupiter Othersb CL111
Arkansas 91,155 86,669 6212 3451 2172
Ashley 11,182 9105 981 0 0
Chicot 34,839 27,057 0 617 630
Clay 81,506 69,905 9086 1249 529
Conway 1703 1149 0 0 0
Craighead 71,509 66,874 10,239 7591 14,620
Crittenden 51,036 43,842 3422 3092 0
Cross 88,036 84,001 17,214 480 3073
Desha 25,266 17,226 3539 2196 0
Drew 11,312 9550 1895 474 0
Greene 78,405 66,208 5958 0 0
Independence 12,747 9974 1596 84 2045
Jackson 104,194 82,216 18,707 13,979 1515
Jefferson 72,463 64,767 3607 709 0
Lafayette 4434 3546 0 0 355
Lawrence 99,922 91,554 19,290 0 6699
Lee 29,920 21,744 1431 0 0
Lincoln 21,516 21,016 1782 0 0
Lonoke 89,732 80,916 4799 662 1564
Mississippi 53,540 47,953 1431 0 0
Monroe 59,492 48,728 5711 697 233
Phillips 32,643 16,094 0 543 0
Poinsett 121,569 110,824 34,132 11,351 3503
Pope 2205 2186 0 0 0
Prairie 63,640 61,743 8631 1974 5932
Pulaski 4168 3799 962 0 0
Randolph 35,657 30,009 10,907 0 4503
St. Francis 38,443 37,462 7046 24 275
White 13,192 10,073 2029 0 0
Woodruff 61,925 50,874 4302 1902 1104
Othersc 7868 2746 0 0 0
Unaccountedd 4781 5596
2015 Total 1,286,000 184,910 51,076 48,751
2015 Percent 100 14.38 3.97 3.79
2014 Total 1,480,000 191,915 19,990 73,412
2014 Percent 100 12.97 1.35 4.96
a
Harvested acreage. Source: USDA-NASS, 2016.
b
Other varieties: AB647, Antonio, Caffey, Cheniere, Cocodrie, Della-2, Francis, Jazzman,
Jazzman-2, RiceTec CL XL746, RiceTec CL XP4534, RiceTec XL723, RiceTec XP4523, and
Taggart.
c
Other counties: Clark, Franklin, Faulkner, Hot Spring, Little River, Perry, and Yell.
d
Unaccounted for acres is the total difference between USDA-NASS harvested acreage
estimate and preliminary estimates obtained for each county from the USDA Farm Service
Agency.
18
harvested rice acreage summary.

Long-grain
CL151 LaKast Mermentau CLXL729 CLXL745 XL753 Roy J Wells Othersb
8315 1747 640 2344 29,216 17,064 7422 0 8086
0 0 0 0 8124 0 0 0 0
4093 0 1350 2528 7914 6775 1786 0 1364
26,138 2929 467 206 23,299 4001 1451 550 0
763 0 0 0 0 0 386 0 0
3457 1898 16,079 0 6842 963 5185 0 0
1572 389 0 3276 11,763 11,182 6161 0 2984
13,097 6568 1249 711 6495 12,437 21,012 0 1724
608 893 1662 0 1153 5096 2078 0 0
0 0 0 0 7181 0 0 0 0
14,655 0 1256 0 21,691 20,377 657 0 1519
2045 0 0 0 2045 818 818 409 0
13,107 4894 519 228 7228 8539 8864 0 4558
8968 6675 0 0 34,851 1241 8717 0 0
355 0 0 0 1418 709 355 0 355
25,054 3228 8942 0 11,751 3083 11,880 0 1628
1341 2041 121 2220 1662 5166 7316 0 445
0 0 0 0 989 15,078 3279 0 0
7536 1033 1564 11262 20,831 18,211 9340 0 4151
4215 3901 1246 0 2098 21,244 2749 11,226 0
838 7017 1071 652 7036 3176 20,399 0 1900
0 2239 11,321 0 1990 0 0 0 0
16,634 6398 103 0 6867 4027 20,646 5504 1692
0 0 0 109 2077 0 0 0 0
1454 2067 2429 4113 12,169 12,199 4938 0 6180
294 0 0 0 1472 589 294 294 0
0 0 0 3659 0 6530 0 0 4409
1922 1446 2237 0 9337 258 11,751 1131 2068
676 0 0 1528 2919 1685 0 0 1236
2294 9045 758 7967 5260 5666 11,080 0 1496
406 51 0 0 815 560 113 775 26
5596
159,837 64,460 53,015 40,804 256,492 186,673 168,677 19,889 51,415
12.43 5.01 4.12 3.17 19.94 14.52 13.12 1.55 4.00
186,518 3,143 72,426 62,445 326,016 174,626 186,022 42,156 141,330
12.60 0.21 4.89 4.22 22.03 11.80 12.57 2.85 9.55
19
Table 2. Acreage, grain yield, and production of rice in the United States from 2013 to 2015.a
20
Area planted Area harvested Yield Production
State 2013 2014 2015 2013 2014 2015 2013 2014 2015 2013 2014 2015
------------------------------- (1,000 acres)----------------------------- -------------(lb/acre)----------- ----------- (1,000 cwtb)-------
Arkansas 1076 1486 1306 1070 1480 1286 7560 7560 7340 80,888 111,957 94,341
California 567 445 423 562 442 421 8480 8580 8890 47,641 37,936 37,441
Louisiana 418 466 420 413 462 415 7300 7130 6940 30,135 32,944 28,791
Mississippi 125 191 150 124 190 149 7400 7420 7110 9176 14,096 10,594
Missouri 159 216 182 156 213 174 7030 6830 7020 10,968 14,540 12,212
Texas 145 150 133 144 146 130 7740 7360 6900 11,145 10,742 8964
U.S. 2490 2954 2614 2469 2933 2575 7694 7576 7470 189,953 222,215 192,343
a
Source: USDA-NASS, 2016.
b
cwt = hundredweight.
Table 3. Acreage distribution of selected cultural practices for Arkansas rice production.a
2013 2014 2015
Cultural practice Acreage % of total Acreage % of total Acreage % of total
Arkansas rice 1,070,000 100.00 1,480,000 100.00 1,286,000 100.00
acreage

Soil texture
Clay 209,251 19.6 290,508 19.6 264,441 20.6
Clay loam 252,702 23.6 311,721 21.1 268,398 20.9
Silt loam 547,386 51.2 825,486 55.8 689,012 53.6
Sandy loam 45,733 4.3 41,474 2.8 53,116 4.1
Sand 14,928 1.4 10,811 0.7 11,033 0.9
Tillage practices
Conventional 654,647 61.2 883,586 59.7 818,368 63.6
Stale seedbed 329,807 30.8 482,323 32.6 386,620 30.1
No-till 85,546 8.0 114,090 7.7 81,011 6.3
Crop rotations
Soybean 759,792 71.0 1,069,283 72.2 930,396 72.3
Rice 225,690 21.1 317,662 21.5 273,627 21.3
Cotton 5586 0.5 4030 0.3 3718 0.3
Corn 45,006 4.2 41,093 2.8 42,343 3.3
Grain sorghum 6810 0.6 11,532 0.8 15,450 1.2
Wheat 13,107 1.2 7222 0.5 852 0.1
Fallow 13,705 1.3 29,178 2.0 19,613 1.5
Other 305 0.0 0 0.0 0 0.0
Seeding methods
Drill seeded 881,172 82.4 1,250,157 84.5 1,074,460 83.6
Broadcast seeded 183,112 17.1 229,843 15.5 211,540 16.4
Water seeded 32,570 3.0 61,221 4.1 70,302 5.5
Irrigation water sources

Groundwater 848,435 79.3 1,145,847 77.4 982,419 76.4
Stream, rivers, etc. 111,743 10.4 155,345 10.5 146,202 11.4
Reservoirs 109,822 10.3 178,807 12.1 157,379 12.2
Irrigation methods
Flood, levees 698,139 65.2 885,796 59.9 731,614 56.9
Flood, multiple inlet 368,092 34.4 585,658 39.6 521,689 40.6
Intermittent (AWD) -- -- -- -- 21,241 1.7
Furrow 3769 0.4 6203 0.4 11,456 0.9
Sprinkler 0 0.0 458 0.0 0 0.0
Other 0 0.0 1885 0.1 0 0.0
Stubble management
Burned 303,204 28.3 414,650 28.0 559,736 43.5
Tilled 430,519 40.2 537,686 36.3 501,329 39.0
Rolled 316,705 29.6 548,333 37.0 343,383 26.7
Winter flooded 203,971 19.1 294,729 19.9 262,846 20.4
continued
21
Table 3. Continued.
2013 2014 2015
Cultural practice Acreage % of total Acreage % of total Acreage % of total
Land management
Contour levees 345,944 32.3 402,239 27.2 625,600 48.6
Precision-level 603,039 56.4 896,041 60.5 519,907 40.4
Zero-grade 121,016 11.3 181,720 12.3 141,897 11.0
Precision agriculture
Yield monitors 553,505 51.7 877,850 59.3 847,603 65.9
Grid sampling 240,490 22.5 437,759 29.6 386,143 30.0
Variable-rate 202,822 19.0 367,045 24.8 336,228 26.1
fertilizer
Pest management
Insecticide seed 653,049 61.0 1,047,204 70.8 867,242 67.4
treatment
Fungicide 578,201 54.0 853,570 57.7 674,727 52.5
(foliar application)
Insecticide 457,649 42.8 526,939 35.6 462,302 35.9
(foliar application)
a
Data generated from surveys of county agriculture extension agents.
Fig. 1. Arkansas rice planting progress during 2015

compared to the five-year state average (USDA-NASS, 2016).
22
Fig. 2. Arkansas rice harvest progress during 2015

compared to the five-year state average (USDA-NASS, 2016).
Fig. 3. Percentage of rice planted in Arkansas

to Clearfield rice cultivars between 2001 and 2015.
23
2015 Rice Research Verification Program

R. Baker1, R. Mazzanti2, J.T. Hardke2, K.B. Watkins3, and R. Mane4
ABSTRACT
The 2015 Rice Research Verification Program (RRVP) was conducted on 16 commercial
rice fields across Arkansas. Counties participating in the program included Arkansas,
Ashley, Chicot, Clay, Cross, Desha, Independence, Lawrence, Lee, Lonoke, Mississippi,
Monroe, Phillips, Randolph, St. Francis, and White counties for a total of 1013 acres.
Grain yield in the 2015 RRVP averaged 176 bu/acre ranging from 119 to 237 bu/acre.
The 2015 RRVP average yield was 13 bu/acre greater than the estimated Arkansas state
average of 163 bu/acre. The highest yielding field was in Randolph County with a grain
yield of 237 bu/acre. The lowest yielding field was in Lee County and produced 119
bu/acre. Milling quality in the RRVP was comparable with milling from the Arkansas
Rice Performance Trials and averaged 58/72 (head rice/total white rice).
INTRODUCTION
In 1983, the University of Arkansas System Division of Agricultures Coopera-
tive Extension Service established an interdisciplinary rice educational program that
stresses management intensity and integrated pest management to maximize returns.
The purpose of the Rice Research Verification Program (RRVP) was to verify the
profitability of Cooperative Extension Service (CES) recommendations in fields with
less than optimum yields or returns.
The goals of the RRVP are to: 1) educate producers on the benefits of utilizing
CES recommendations to improve yields and/or net returns, 2) conduct on-farm field
trials to verify research-based recommendations, 3) aid researchers in identifying areas
of production that require further study, 4) improve or refine existing recommendations
which contribute to more profitable production, 5) incorporate data from RRVP into
CES educational programs at the county and state level. Since 1983, the RRVP has been
conducted on 431 commercial rice fields in 33 rice-producing counties in Arkansas.
1
Rice Verification Program Coordinator, Department of Crop, Soil, and Environmental Science, Piggott..
2
Program AssociateRice Verification Coordinator and Rice Extension Agronomist, respectively, De-
partment of Crop, Soil, and Environmental Science, Stuttgart.
3
Professor, Rice Research and Extension Center, Stuttgart.
4
Program Associate, Rice Research and Extension Center, Stuttgart.
24
The program has typically averaged 20 bu/acre better than the state average yield. This
increase in yield over the state average can mainly be attributed to intensive cultural
management and integrated pest management.
PROCEDURES
The RRVP fields and cooperators are selected prior to the beginning of the grow-
ing season. Cooperators agree to pay production expenses, provide expense data, and
implement CES recommendations in a timely manner from planting to harvest. A des-
ignated county agent from each county assists the RRVP coordinator in collecting data,
scouting the field, and maintaining regular contact with the producer. Weekly visits by
the coordinator and county agents are made to monitor the growth and development of
the crop, determine what cultural practices needed to be implemented and to monitor
type and level of weed, disease and insect infestation for possible pesticide applications.
An advisory committee, consisting of CES specialists and university researchers
with rice responsibility, assists in decision-making, development of recommendations,
and program direction. Field inspections by committee members are utilized to assist
in fine tuning recommendations.
Counties participating in the program during 2015 included Arkansas, Ashley,
Chicot, Clay, Cross, Desha, Independence, Lawrence, Lee, Lonoke, Mississippi, Mon-
roe, Phillips, Randolph, St. Francis, and White Counties. The sixteen rice fields totaled
1013 acres enrolled in the program. Eight different cultivars were seeded (CL151,
CL271, RiceTec CLXL745, Jupiter, LaKast, Mermentua, Roy J, and RiceTec XL753)
and CES recommendations were used to manage the RRVP fields. Agronomic and pest
management decisions were based on field history, soil test results, cultivar, and data
collected from individual fields during the growing season. An integrated pest manage-
ment philosophy was utilized based on CES recommendations. Data collected included
components such as stand density, weed populations, disease infestation levels, insect
populations, rainfall, irrigation amounts, dates for specific growth stages, grain yield,
milling yield, and grain quality.

Yield
The average RRVP yield was 176 bu/acre with a range of 119 to 237 bu/acre (Table
1). All grain yields of RRVP fields are reported in dry, 12% moisture, bushels. The RRVP
average was 13 bu/acre more than the estimated State yield of 163 bu/acre. This differ-
ence has been observed many times since the program began and can be attributed in
part to intensive management practices and utilization of CES recommendations. The
Randolph County field, seeded with RiceTec XL753, was the highest yielding RRVP
field at 237 bu/acre. Eight of the sixteen fields enrolled in the program exceeded 170
bu/acre. Lee County had the lowest yielding field with LaKast producing 119 bu/acre.
25
Milling data was recorded on all of the RRVP fields. The average milling yield for
the sixteen fields was 57/72 (head rice/total white rice) with the highest milling yield
of 66/74 with Roy J in Cross County (Table 1). The lowest milling yield was 50/72
with RiceTec CLXL745 in Chicot County. The milling yield of 55/70 is considered the
standard used by the rice milling industry.
Planting and Emergence

Planting began with Arkansas County on 30 March and ended with White County
planted 7 May (Table 1). Two of the verification fields were planted in March, twelve
in April, and two in May. An average of 74 lb seed/acre was planted for pure-line
varieties and 25 lb seed/acre for hybrids. Seeding rates were determined with the CES
RICESEED program for all fields. An average of 12 days was required for emergence.
Stand density averaged 19 plants/ft2 for pure-line varieties and 7 plants/ft2 for hybrids.
The seeding rates in some fields were higher than average due to planting method,
soil texture, and planting date. Broadcast seeding and clay soils generally require an
elevated seeding rate to achieve desired plant populations.
Fertilization
The Nitrogen Soil Test for Rice (N-STaR) was utilized on 9 of 16 RRVP fields.
The difficult early-season conditions of 2015 required, for some fields, a deviation from
the N-STaR recommendations. Nitrogen recommendations were based on a combina-
tion of factors including soil texture, previous crop, and cultivar requirements (Table
2). Nitrogen rates can appear high in some fields with a clay soil texture and when rice
was the previous crop. These factors increase the nitrogen requirements compared to
a silt loam soil where soybeans were the previous crop. The Lee County field received
ammonium sulfate (21-0-0-24) at the 2- to 3-lf stage as a management tool to increase
plant growth and shorten the time required to get the rice to flood stage. The White
County field received a poultry litter application in the fall to improve soil conditions
following land leveling.
Phosphorus, potassium, and zinc fertilizer were applied based on soil-test analy-
sis recommendations (Table 2). Phosphorus was applied preplant to Arkansas, Ashley,
Chicot, Cross, Independence, Lawrence, Lee, Randolph, St. Francis, and White County
fields. Potassium was applied to Arkansas, Ashley, Clay, Cross, Desha, Independence,
Lawrence, Lee, Randolph, St. Francis, and White Counties. Zinc was applied as a
preplant fertilizer to fields in Cross, Desha, Independence, Lawrence, Lee, and White
Counties, while zinc seed treatment was used with hybrid rice cultivars at a rate of 0.5
lb zinc/60 lb seed. The average cost of fertilizer across all fields was $120.46.
Weed Control
Command was utilized in 13 of the 16 fields for early-season grass control (Table
3). Facet was applied in 10 of 16 fields either pre-emergence or early post-emergence.
26
Four fields (Ashley, Chicot, Clay, and Lonoke Counties) were seeded in Clearfield
cultivars (Table 1). Only the Mississippi County field did not require a post-emergence
herbicide application for grass weed control (Table 3).
Disease Control
Foliar fungicides were applied to 2 of the 16 fields in 2015 for management of
sheath blight and prevention of kernel smut (Table 4). Fungicide rates were determined
based on cultivar, growth stage, climate, disease incidence/severity, and disease history.
Fourteen fields had a seed treatment containing a fungicide.
Insect Control
Nine fields (Arkansas, Ashley, Chicot, Cross, Desha, Independence, Lee, Missis-
sippi, and Phillips Counties) were treated with a foliar insecticide application for rice
stink bug in 2015 (Table 4). Thirteen fields received an insecticide seed treatment in
the form of CruiserMaxx Rice.
Irrigation
Well water was used to irrigate 10 of the 16 fields in the 2015 RRVP while 6 fields
were irrigated with surface water. Three fields (Ashley, Chicot, and Lonoke Counties)
were zero-grade. Ten fields used Multiple Inlet Rice Irrigation (MIRI) either by utiliz-
ing irrigating tubing or by having multiple risers or water sources. Typically, a 25%
reduction in water use is observed when using MIRI. Flow meters were used in 14
of the fields to record water usage throughout the growing season (Table 5). In fields
where flow meters were not utilized, the average across all irrigation methods of 30
acre-inches was used. The difference in water used was due in part to rainfall amounts
which ranged from 6.2 to 26.1 inches.
Economic Analysis
This section provides information on production costs and returns for the 2015
RRVP (Tables 6 and 7). Records of field operations on each field provided the basis for
estimating production costs. The field records were compiled by the RRVP coordina-
tors, county Extension agents, and cooperators. Production data from the 16 fields were
applied to determine costs and returns above operating costs, as well as total specified
costs. Operating costs and total costs per bushel indicate the commodity price needed
to meet each cost type.
Operating costs are those expenditures that would generally require annual cash
outlays and would be included on an annual operating loan application. Actual quan-
tities of all operating inputs as reported by the cooperators are used in this analysis.
27
Input prices are determined by data from the 2015 Crop Enterprise Budgets published
by the CES and information provided by the cooperating producers. Fuel and repair
costs for machinery are calculated using a budget calculator based on parameters and
standards established by the American Society of Agricultural and Biological Engineers.
Machinery repair costs should be regarded as estimated values for full-service repairs,
and actual cash outlays could differ as producers provide unpaid labor for equipment
maintenance.
Fixed costs of machinery are determined by a capital recovery method which
determines the amount of money that should be set aside each year to replace the value
of equipment used in production. Machinery costs are estimated by applying engineering
formulas to representative prices of new equipment. This measure differs from typical
depreciation methods, as well as actual annual cash expenses for machinery.
Operating costs, fixed costs, costs per bushel, and returns above operating and
total specified costs for each RRVP field are presented in Table 6. Costs in this report
do not include land costs, management, or other expenses and fees not associated with
production. Operating costs ranged from $411.49/acre for Lonoke County to $749.15
for White County, while operating costs per bushel ranged from $2.24/bu for Missis-
sippi County to $4.62/bu for White County. Total costs per acre (operating plus fixed)
ranged from $526.87/acre for Lonoke County to $846.50/acre for Chicot County, and
total costs per bushel ranged from $2.76/bu for Mississippi County to $5.35/bu for Lee
County. The average return above operating costs for the 16 fields was $350.18/acre
and ranged from $38.29/acre for Lee County to $680.11/acre for Mississippi County
(Tables 6 and 7). The average return above total costs for the 16 fields was $249.20/acre
and ranged from -$48.22/acre for Lee County to $579.75/acre for Mississippi County.
A summary of yield, rice price, revenues, and costs by cost type for each RRVP
field are presented in Table 7. The average rice yield for the 2015 RRVP was 176 bu/
acre but ranged from 119 bu/acre for Lee County to 237 bu/acre for Randolph County
(Table 1). An Arkansas average long-grain cash price of $4.92/bu was estimated using
USDA, National Agricultural Statistics Service (USDA-NASS, 2016) U.S. long-grain
price data for the months of August through October. The RRVP had four fields planted
to the medium-grain cultivar Jupiter. The average medium-grain price contracted in
Arkansas was estimated to be $5.68/bu and represented the average long-grain price plus
an average medium-grain premium of $0.76/bu. The average medium-grain premium
was estimated based on the average difference in Arkansas milled rice value between
medium- and long-grain rice obtained from the Arkansas Weekly Grain Review for the
period 3 August through 2 November, converted to a rough rice equivalent. A premium
or discount was given to each field based on the milling yield observed for each field
and standard milling yields of 55/70 for long-grain rice and 58/69 for medium-grain
rice. Broken rice was assumed to have 70% of whole grain price value. If milling yield
was higher than the standard, a premium was made while a discount was given for
milling less than the standard. Estimated long-grain prices adjusted for milling yield
varied from $4.83/bu in Ashley County to $5.38/bu in Cross County (Table 7). The
medium-grain price adjusted for milling yield varied from $5.59/bu for Desha County
to $5.84/bu for Independence County.
28
The average operating cost for the 16 RRVP fields was $571.43/acre (Table 7).
Fertilizer and nutrients accounted for the largest share of operating costs on average
(21.1%) followed by post-harvest expenses (20.4%), seed (15.0%), and chemicals
(13.3%). Although seeds share of operating expenses was 15.0% across the 16 fields,
its average cost and share of operating costs varied depending on whether a Clearfield
hybrid was used ($195.36/acre; 27.9% of operating expenses), a non-Clearfield hybrid
was used ($130.56/acre; 18.6% of operating expenses), a Clearfield non-hybrid (pure-
line) variety was used ($95.62/acre; 19.7% of operating expenses) or a non-Clearfield
non-hybrid (pure-line) variety was used ($44.45/acre; 8.6% of operating expenses).
Table 8 provides select variable input costs for each field and includes rice type; seed,
fertilizer, diesel fuel, and irrigation costs; as well as chemical costs for herbicides,
insecticides, and fungicides.
Field Summaries
The 48-acre Arkansas County field was located southeast of Stuttgart on a Stuttgart
and Dewitt silt loam soil. The previous crop was soybean. Conventional tillage practices
were used for field preparation and a preplant fertilizer based on soil test results was
applied at a rate of 0-30-90-10 (lb/acre N-P2O5-K2O-Zn). RiceTec hybrid XL753 was
drill-seeded on 30 March at 22 lb/acre. CruiserMaxx Rice seed treatment was used in
addition to the companys standard seed treatment. Rice emerged on 13 April with a
stand density of 6.4 plants/ft2. Glyphosate herbicide was used as a burndown on 4 April.
Command and League were applied as pre-emergence herbicides on 13 April. Excellent
residual herbicide activity was observed for 30 days. Facet and Permit Plus herbicides
were applied 15 May and provided post-emergence control of dayflower and nutsedge.
Using the N-STaR recommendation, nitrogen in the form of urea plus an approved N-
(n-butyl) thiophosphoric triamide (NBPT) product was applied preflood at a rate of 235
lb/acre on 15 May. Multiple-inlet rice irrigation was utilized to achieve a more efficient
permanent flood. On 7 July urea at 70 lb/acre was applied at the late-boot stage. The
field was clean throughout the year and a deep flood was maintained. No fungicides
were needed for disease but rice stink bugs reached threshold levels and were treated
with Karate insecticide on 14 July. The field was harvested on 8 August with a yield of
213 bu/acre5. The average harvest moisture was 18% and the milling yield was 59/71.
This was the second-highest yield in the 2015 RRVP. Irrigation water use totaled 31.9
acre-inches with rainfall amounts totaling 13.25 inches.
The zero-grade, 79-acre Ashley County field was located east of Montrose on a
Grubbs silt loam and Jackport silty clay loam soil. Preplant fertilizer was applied at a
rate of 0-40-60 (lb/acre N-P2O5-K2O). The previous crop was soybean and conventional
tillage practices were utilized in the spring. RiceTec hybrid CL XL745 was drill-seeded
at 23 lb/acre on 8 April. The seed was treated with CruiserMaxx Rice seed treatment in
addition to the companys standard seed treatment. A pre-emergence application of Com-
mand herbicide was made on 11 April. Rice emergence was observed on 19 April with 7.6
5
The yield for this and all other RRVP fields is reported in dry, 12% moisture, bushels.
29
plants/ft2. Ammonium sulfate was applied on 8 May as a starter fertilizer at 100 lb/acre.
On 26 May Clearpath and Permit Plus herbicides were applied. Nitrogen in the form of
urea plus an approved NBPT product was applied on 14 May at 330 lb/acre according to
N-STaR recommendations. The late boot nitrogen application of urea at 70 lb/acre was
applied on 17 July. No fungicides were necessary since disease never reached threshold
levels. Stink bugs reached threshold level and Lambda-Cy insecticide was applied on 31
July. The field was harvested on 29 August yielding 166 bu/acre. The milling yield was
53/69 and the average harvest moisture was 18%. The irrigation water use totaled 28.4
acre-inches and rainfall for the growing season was 19.75 inches.
The 69-acre zero-grade Chicot County field was located north of Lake Village
on a Perry clay soil. On 3 April, RiceTec hybrid CL XL745, treated with CruiserMaxx
Rice seed treatment in addition to the companys standard seed treatment, was drill-
seeded at 33 lb/acre. Preplant fertilizer was applied on 5 April at the rate of 0-69-0
(lb/acre N-P2O5-K2O) according to soil-test recommendations. Glyphosate, Command,
and League were applied on 4 April as burndown and pre-emergence herbicides. Con-
tinuous rainfall on a weekly basis provided residual weed control for 58 days. Field
emergence was recorded on 30 April with a stand density of 2.5 plants/ft2 that eventually
increased to 3.5 plants/ft2. On 8 June, Ricestar HT herbicide was applied post-emergence
for sprangletop escapes. Based on N-STaR recommendations, nitrogen in the form of
urea was applied preflood at 250 lb/acre on 8 May. Late boot urea fertilizer at 70 lb/
acre was applied on 14 June. Rice stink bugs reached treatment levels and Lambda-Cy
insecticide was applied on 16 July. The field was harvested 14 August with a yield of
196 bu/acre, milling yield of 50/72, and an average harvest moisture of 12%. The grower
was pleased with the yield considering the low stand count and unfavorable weather in
April and May. Irrigation water used totaled 30.0 acre-inches. Rainfall amounts were
6.2 inches for the season.
The precision-graded Clay County field was located 7.5 miles southwest of Corn-
ing on a Jackport silty clay loam soil. The field was 78 acres and the previous crop grown
was soybean. In late April, conventional tillage practices were used for field preparation
and a preplant fertilizer based on soil-test analysis was applied at a rate of 0-0-60 (lb/acre
N-P2O5-K2O). On 29 April, the medium-grain variety CL271 with CruiserMaxx Rice
seed treatment was drill-seeded at a rate of 65 lb/acre. Rice emergence was observed
on 12 May and consisted of 13.4 plants/ft2. Command was applied as a pre-emergence
herbicide plus glyphosate as a burndown herbicide prior to crop emergence. This was
followed by a post-emergence application of Clearpath followed later by Newpath.
Excellent pre- and post-emergence control of weeds was achieved. Using the N-STaR
recommendation, a single preflood N application was made with the intention of ex-
cluding a midseason N application. Urea plus an approved NBPT product was applied
at a rate of 207 lb/acre on 27 June. However, nitrogen loss prior to flood-up made it
necessary to apply midseason N to correct the problem. Urea at a rate of 100 lb/acre
was applied on 16 July. Once the permanent flood was established, flood levels were
maintained well throughout the season. Although sheath blight lesions were found in
the field, they remained low on the plant and cool nighttime temperatures helped hold
30
the disease well below treatment threshold levels. Continued field evaluations resulted
in no treatments for sheath blight or any other disease. Rice stink bugs were present in
the field but remained below treatment threshold levels and no insecticide treatments
were required. On 21 September, a sodium chlorate harvest aid treatment was applied
at the rate of 1 gal/acre. The rice was harvested on 24 September, yielding 202 bu/acre,
an average harvest moisture of 17%, and a milling yield of 57/74. Total irrigation water
use was 32.9 acre-inches and total rainfall for the season was 18.5 inches.
The traditionally contoured Cross County field was located 2.4 miles southwest of
Hickory Ridge on Crowley and Hillemann silt loam soils. The field was 27 acres and the
previous crop grown was soybean. Conventional tillage practices were used for spring
field preparation and a preplant fertilizer based on soil-test analysis was applied at a rate
of 0-40-60-2 (lb/acre N-P2O5-K2O-Zn). On 8 April, the variety Roy J with CruiserMaxx
Rice seed treatment was broadcast-seeded at a rate of 90 lb/acre. Rice emergence was
observed on 16 April and consisted of 21 plants/ft2. Command herbicide was applied
pre-emergence followed by a post-emergence application of Facet plus propanil fol-
lowed by Ricestar HT. Levees were sprayed with 2,4-D plus Grandstand. Good pre- and
post-emergence control of weeds was achieved. Using the N-STaR recommendation,
urea plus an approved NBPT product was applied preflood on 29 May at 174 lb/acre.
Multiple-inlet rice irrigation was utilized to achieve a more efficient permanent flood.
Once the permanent flood was established, flood levels were maintained well throughout
the season. A midseason application of urea was made on 22 June at the rate of 100 lb/
acre. No fungicide applications were required. However, rice stink bugs reached treat-
ment level on 25 July and were treated with Karate insecticide. Rice was harvested on
12 September with a yield of 139 bu/acre. The low yield was notably similar to other
fields in that part of the state, including two other RRVP fields that were in the same
vulnerable stages of development during unfavorable weather conditions. Moisture at
harvest was 15% and the milling yield was 66/74. Total irrigation water use was 24.4
acre-inches and total rainfall for the season was 22.8 inches.
The zero-grade, 31-acre Desha County field was located just east of Tiller on
a Herbert silt loam and Perry clay soil. No tillage practices were performed from the
previous rice crop. Preplant fertilizer at 0-0-90-10 (lb/acre N-P2O5-K2O-Zn) was applied
30 March. The medium-grain variety Jupiter was drill-seeded at a rate of 70 lb/acre on
31 March. The seed was treated with CruiserMaxx Rice seed treatment. Glyphosate,
Command, and League herbicides were applied for burndown and pre-emergence weed
control on 1 April. Rice emergence was observed on 14 April with 19 plants/ft2. A post-
emergence herbicide application of Permit was made on 7 May. Nitrogen in the form
of urea plus an approved NBPT product was applied preflood at 270 lb/acre according
to N-STaR recommendations. Multiple-inlet rice irrigation was utilized to achieve a
more efficient permanent flood. The spreader buggy application left the field streaked
and on 4 May 100 lb urea/acre was applied by air to correct the problem. No midseason
urea application was necessary according to GreenSeeker (Trimble Navigation Limited,
Sunnyvale, Calif.) technology. Stink bugs reached threshold level and Lambda-Cy
was applied 18 July. The field was harvested 2 September yielding 170 bu/acre with a
31
milling yield of 57/68. The average harvest moisture was 18%. The yield was slightly
disappointing but characteristic of the 2015 growing season. The irrigation water use
totaled 27 acre-inches and the rainfall amount for the growing season was 18.7 inches.
The precision-graded Independence County field was located 1.5 miles southwest
of Oil Trough. The soil combination was a Jackport silty clay loam, Engham silt loam,
and Hontas silt loam. The field was 43 acres and the previous crop grown was soybean.
Conventional tillage practices were used in the fall and spring for field preparation and
a preplant fertilizer based on soil-test analysis was applied at a rate of 0-46-96-0.5 (lb/
acre N-P2O5-K2O-Zn). On 9 April, the medium-grain variety Jupiter with CruiserMaxx
Rice plus Release seed treatment was drill-seeded at a rate of 70 lb/acre. Rice emergence
was observed on 21 April and consisted of 20 plants/ft2. Command herbicide was ap-
plied pre-emergence 9 April followed on 29 May by a post-emergence application of
Command plus Sharpen providing excellent pre- and post-emergence control of weeds
except on the edges of the field. An additional treatment of Permit and Grasp was ap-
plied to field edges for full weed control. Using the N-STaR recommendation, a single
preflood N application was made with the intention of excluding a midseason N ap-
plication. Urea plus an approved NBPT product was applied at a rate of 185 lb/acre on
6 June. Unfortunately, some nitrogen loss occurred before flood-up took place making
it necessary to apply midseason N to correct the problem. Urea at a rate of 100 lb/acre
was applied on 27 June. Once the permanent flood was established, flood levels were
maintained well throughout the season. Although sheath blight lesions were present in
the field, they remained low on the plant and cool nighttime temperatures helped hold
the disease well below threshold treatment levels. Continued field evaluations resulted
in no treatments for sheath blight or any other disease. Rice stink bugs were found to
overwhelm natural predators in the field, exceeding the threshold level for treatment.
Control of the pest was accomplished with a single treatment of 2 oz/acre Karate on
31 July. No further insecticide treatments were required. The rice was harvested on 25
September yielding 196 bu/acre with a harvest moisture of 18%. The milling yield was
55/73. Total irrigation water use was 35.2 acre-inches and total rainfall for the growing
season was 26.1 inches.
The precision-graded Lawrence County field was located north of Alicia on a
Dubbs silt loam soil. The field was 50 acres and the previous crop grown was soybean.
Conventional tillage practices were used for field preparation in the spring. A preplant
fertilizer based on soil-test analysis was applied on 24 April at the recommended rate
of 0-40-48-8-4 (lb/acre N-P2O5-K2O-Zn-S). On 27 April, the variety Mermentau was
drill-seeded at a rate of 80 lb/acre. Rice emergence was observed on 9 May and con-
sisted of 25 plants/ft2. Command pre-emergence herbicide plus glyphosate as a burn-
down herbicide were applied on 28 April. On 30 May a post-emergence application
was made of Grasp Xtra plus Command to extend its residual activity. Good pre- and
post-emergence control of weeds was provided. Using the N-STaR recommendation,
urea plus an approved NBPT product at the rate of 250 lb/acre was applied preflood
on 12 June followed by a normal midseason application of 100 lb of urea/acre on 29
June. Multiple-inlet irrigation was utilized to achieve a more efficient permanent flood.
32
Once the permanent flood was established, flood levels were maintained sufficiently
throughout the season but not without some difficulty due to the permeable nature of
portions of the field. Lesions caused by sheath blight fungus reached treatment level.
An application of Quadris fungicide was made on 29 July and no further fungicide ap-
plications were required. Rice stink bugs were present in the field but remained below
threshold levels the entire season and no insecticide treatments were required. Harvest
began on 22 September and the yield average was 163 bu/acre, with a harvest moisture
of 18%, and a milling yield of 64/73. Total irrigation water use was 82.0 acre-inches, re-
flecting highly permeable areas of the field. Total rainfall for the season was 18.7 inches.
The 54-acre Lee County field was located just east of Moro on a Calloway and
Henry silt loam soil. Soybean was the previous crop grown on the field and no tillage
practices were performed on the contour field. A preplant fertilizer blend of 21-40-75-
10-1 (lb/acre N-P2O5-K2O-Zn-S) was applied according to the soil sample analysis. On
8 April the variety LaKast, treated with CruiserMaxx Rice seed treatment and zinc, was
broadcast at a rate of 75 lb/acre. Sharpen, glyphosate, and Command were applied on
8 April as burndown and pre-emergence herbicides. Ammonium sulfate was applied at
100 lb/acre as a starter fertilizer. Emergence was observed on 28 April with 12 plants/
ft2. Facet and Permit were applied on 6 May as post-emergence herbicides. Based on
N-STaR recommendations, nitrogen in the form of urea plus an approved NBPT product
was applied at 240 lb/acre on 26 May. A minimal flood was maintained throughout the
growing season with MIRI. Using GreenSeeker technology, midseason urea fertilizer
was applied 16 June on the south half of the field at 100 lb/acre. Plant nitrogen on the
north half of the field was sufficient without applying midseason urea. Stink bugs reached
threshold levels and Lambda-Cy insecticide was applied on 28 July. The field was har-
vested on 2 September yielding 119 bu/acre with a milling yield of 51/72. Broadcast
planting, low stand counts, and excessive cloudiness during early season to midseason all
contributed to a decreased yield. The average harvest moisture was 15%. The irrigation
water use totaled 33.9 acre-inches and the season-long rainfall total was 15.8 inches.
The 36-acre zero-grade Lonoke County field was located south of England on a
Perry silty clay soil. No tillage practices were performed on the field from the previous
rice crop. Based on soil-test analysis, no preplant fertilizer was needed. The variety
CL151, treated with CruiserMaxx Rice seed treatment and zinc, was drilled-seeded at
65 lb/acre on 11 April. Glyphosate, Command, and League herbicides were applied 25
April. Rice emergence was observed on 26 April with 16 plants/ft2. Due to continual
rainfall, residual herbicide activity was observed for 35 days. On 1 June, Command
and Ricestar HT herbicides were applied for continued grass control. Nitrogen in the
form of urea plus an approved NBPT product was applied at 190 lb/acre on 19 May
according to N-STaR recommendations. An adequate flood was maintained throughout
the growing season. No midseason fertilizer was necessary according to GreenSeeker
technology. The field was harvested on 2 September yielding 167 bu/acre with a milling
yield of 59/72. For comparison, the 2014 yield in the same field with the same variety
was 188 bu/acre. The 21 bu/acre difference from 2014 to 2015 was consistent with 2015
growing season results. Irrigation water use totaled 48 acre-inches and the rainfall for
the growing season totaled 15 inches.
33
The precision-graded Mississippi County field was located 3 miles east of Dyess
on a Sharkey silty clay/Sharkey-Steele complex soil. The field was 30 acres which lay
fallow the previous year. Conventional tillage practices were used for field prepara-
tion in the spring. Based on soil-test analysis, no preplant fertilizer was needed. On 6
April, the medium-grain variety Jupiter was drill-seeded at a rate of 90 lb/acre. Rice
emergence was observed on 15 April and consisted of 16 plants/ft2. Prowl H2O herbi-
cide plus Roundup WeatherMAX plus Facet L was applied pre-emergence on 6 April
and provided excellent control of weeds; no post-emergence herbicide application
was needed. Urea plus an approved NBPT product at a rate of 250 lb/acre was applied
preflood on 24 May followed by a split midseason application of urea at 150 lb/acre
(75 lb on 15 June followed by 75 lb on 22 June). It should be noted that applying mid-
season N in a split application is no longer the preferred recommendation by the CES.
Proper preflood N fertilization followed by a single midseason N application has been
found to be the most cost effective and efficient means of split N fertilization in rice.
Once the permanent flood was established, flood levels were maintained sufficiently
throughout the season. Although sheath blight lesions were present in the field, they
remained low on the plant and cool nighttime temperatures helped hold the disease well
below threshold treatment levels. Continued field evaluations resulted in no treatments
for sheath blight or any other disease. Rice stink bugs exceeded the threshold level for
treatment and Karate was applied for control on 27 July. No further insecticide treat-
ments were required. Harvest began on 1 September and the yield averaged 191 bu/
acre with a harvest moisture of 21%. The milling yield was excellent at 65/68. Total
irrigation water use was 21 acre-inches and total rainfall for the season was 22.3 inches.
The precision-graded, 60-acre Monroe County field was located east of Clarendon
on a Grubbs silt loam and Jackport silty clay loam soil. Conventional tillage practices
were used for field preparation in the spring and soybean was the previous crop. Based
on soil-test analysis, no preplant fertilizer was needed. The medium-grain variety Jupi-
ter, treated with CruiserMaxx Rice seed treatment and Release, was drill-seeded at 72
lb/acre on 16 April. Emergence was observed on 28 April at 15 plants/ft2. Glyphosate,
Command, and League herbicides were applied on 17 April giving 25 days residual
control. Facet and RiceBeaux were applied 12 May as post-emergence herbicides.
Nitrogen fertilizer in the form of urea plus an approved NBPT product was applied 12
June at 300 lb/acre according to N-STaR. An adequate permanent flood was maintained
throughout the growing season using MIRI. No fungicide or insecticide applications were
necessary due to careful scouting and no midseason nitrogen was necessary according
to GreenSeeker technology. The field was harvested 24 September with a grain yield
of 184 bu/acre and a milling yield of 57/69. The grower was very pleased stating the
yield was 20 to 24 bu/acre better than his other two Jupiter fields yet with only a frac-
tion of the input costs. Those input cost savings included lower chemical cost from the
absence of fungicide and insecticide applications and no midseason nitrogen application.
Irrigation water use totaled 15.8 acre-inches and rainfall amounts totaled 14.7 inches.
The contoured 47-acre Phillips County field was located southeast of Marvell
on a Calloway silt loam soil. Conventional tillage was used after the previous soybean
34
crop. Based on soil-test analysis, no preplant fertilizer was needed. The variety LaKast
was treated with CruiserMaxx Rice seed treatment plus zinc and was drill-seeded at
72 lb/acre on 17 April. Emergence was observed on 24 April at 16 plants/ft2. No pre-
emergence herbicides were able to be applied at planting, but Facet, RiceBeaux, and
Permit herbicides were applied post-emergence on 14 May. Nitrogen in the form of
urea plus an approved NBPT product was applied on 25 May at 250 lb/acre. Stink bugs
reached threshold levels and Karate insecticide was applied 30 July. Multiple-inlet
rice irrigation was utilized to achieve a more efficient permanent flood. No midseason
nitrogen was needed according to GreenSeeker technology. The field was harvested on
7 September with a yield of 160 bu/acre with a milling yield of 52/71. Again, the yield
was consistent for the 2015 growing season. The irrigation amount was 30 acre-inches
and the rainfall amount was 17.8 inches.
The precision-graded Randolph County field was located 2.5 miles northeast of
Pocahontas on Amagon and Dundee silt loam soils. The field was 235 acres and the
previous crop grown was soybean. Spring conventional tillage practices were used for
field preparation and a preplant fertilizer based on soil-test analysis was applied at a rate
of 0-46-120 (lb/acre N-P2O5-K2O). On 6 May, RiceTec hybrid XL753 with CruiserMaxx
Rice seed treatment, in addition to the companys standard seed treatment, was drill-
seeded at a rate of 22 lb/acre. Rice emergence was observed on 15 May and consisted
of 5.7 plants/ft2. Command herbicide was applied pre-emergence on 7 May followed on
5 June by a post-emergence application of Prowl H2O plus Grasp followed on 30 June
by a post-flood application of Ricestar HT providing excellent pre- and post-emergence
control of weeds. Using the N-STaR recommendation, urea plus an approved NBPT
product was applied preflood at a rate of 217 lb/acre on 13 June. Multiple-inlet irriga-
tion was utilized to achieve a more efficient permanent flood. Even so, due to the very
large size of the field there was an extended flood-up period. This delay, combined with
weather conditions, resulted in nitrogen loss ultimately requiring additional nitrogen
to correct the problem. Urea at a rate of 100 lb/acre was applied on 1 July. The normal
65 lb/acre of urea at late boot for straw strength was also applied on 27 July. Once the
permanent flood was established, flood levels were maintained well throughout the
season. Although sheath blight lesions were present in the field, they remained low
on the plant and cool nighttime temperatures helped hold the disease below threshold
treatment levels. However, a preventative treatment for smut disease was applied us-
ing Quilt Xcel. Rice stink bugs were present in the field but remained below treatment
levels. The rice was harvested on 6 October, yielding 237 bu/acre, the highest RRVP
yield in 2015. Moisture at harvest was 14.3% and the milling yield was 53/71. Total ir-
rigation water use was 28.4 acre-inches and total rainfall for the season was 19.9 inches.
The traditionally contoured St. Francis County field was located 2.5 miles south-
west of Palestine and consisted of Henry, Calloway, and Loring silt loam soils. The
field was 84 acres and the previous crop grown on the field was soybean. Conventional
tillage practices were used in the fall for field preparation and a preplant fertilizer
based on soil-test analysis was applied at a rate of 0-47-77 (lb/acre N-P2O5-K2O). On
6 April, the variety LaKast with Apron XL LS seed treatment was drill-seeded at a rate
of 70 lb/acre. Rice emergence was observed on 15 April and consisted of 26 plants/ft2.
35
Sharpen plus glyphosate were applied in early spring as a burndown treatment. Com-
mand, glyphosate, and League were applied pre-emergence on 9 April followed on 27
April by a post-emergence application of Facet plus Sharpen. An additional application
of Sharpen was used to control weed escapes on levees. Good pre- and post-emergence
control of weeds was achieved. Urea plus an approved NBPT product was applied
preflood on 29 May at a rate of 272 lb/acre. Multiple-inlet rice irrigation was used;
however, permanent flood levels were difficult to establish and maintain due to a failing
older well. A new well was drilled midseason providing much improved flood control.
A midseason application of urea was made at the rate of 100 lb/acre on 20 June. Based
on field evaluations and established pest threshold treatment levels, no fungicide or
insecticide applications were required. The rice was harvested on 31 August yielding
143 bu/acre with a harvest moisture of 15%, and a milling yield of 62/72. Total irriga-
tion water use was 37 acre-inches and total rainfall for the season was 15.7 inches.
The fresh-cut, precision-graded White County field was located 1.5 miles east
of Kensett on Calloway and Immanuel silt loam soils. The field was 42 acres and the
previous crop grown was soybean. A fall application of poultry litter at a rate of 1.5
tons/acre was made to improve the fresh-cut soil conditions. Spring conventional till-
age practices were used for field preparation and a preplant fertilizer based on soil-test
analysis was applied at a rate of 0-30-90-10 (lb/acre N-P2O5-K2O-Zn). On 7 May, Rice-
Tec hybrid XL753 with CruiserMaxx Rice seed treatment, in addition to the companys
standard seed treatment, was drill-seeded at a rate of 24 lb/acre. Rice emergence was
observed on 15 May and consisted of 11 plants/ft2. Prowl H2O herbicide was applied
pre-emergence followed by a post-emergence application of Broadhead plus Facet plus
Londax, providing excellent pre- and post-emergence control of weeds. Using the N-
STaR recommendation, urea plus an approved NBPT product was applied preflood at
a rate of 295 lb/acre on 13 June. Due to weather conditions, nitrogen loss occurred on
12 acres of the field making it necessary to apply additional N to correct the problem.
Urea at a rate of 100 lb/acre was applied on the affected acres on 23 June. The entire
field received the normal 65 lb/acre of urea at late boot on 23 July. Once the permanent
flood was established, flood levels were maintained well throughout the season using
MIRI. Based on field evaluations and established pest threshold treatment levels, no
fungicide or insecticide applications were required. The field was harvested on 18
September and yielded 162 bu/acre with a harvest moisture of 16%, and a milling
yield of 59/76. Total irrigation water use was 11.4 acre-inches and total rainfall for the
season was 18.7 inches.
Data collected from the 2015 RRVP reflects the general trend of decreasing rice
yields and average returns in the 2015 growing season. Analysis of this data showed that
the average yield was higher in the RRVP compared to the state average and the cost
of production was equal to or less than the Cooperative Extension Service-estimated
rice production costs.
36
ACKNOWLEDGMENTS
This research was supported by the Arkansas Rice Research and Promotion Board
and the University of Arkansas System Division of Agriculture.
LITERATURE CITED
USDA-NASS. United States Department of Agriculture, National Agricultural Sta-
tistics Service. 2016. Quick States (Searchable Database). Access Date: 31 May
2016. Available at: https://quickstats.nass.usda.gov/
37
Table 1. Agronomic information for fields enrolled in the 2015 Rice Research Verification Program.
38
Field location Field Previous Seeding Stand Planting Emergence Harvest Milling Harvest
by county Cultivar size crop Rate density date date date Yield yielda moisture
(acres) (lb/acre) (plants/ft2) (bu/acre) (%)
Arkansas XL753 48 Soybean 22 6 03/30/15 04/13/15 08/13/15 213 59/71 18
Ashley CL XL745 79 Soybean 23 7 04/08/15 04/19/15 08/29/15 166 53/69 18
Chicot CL XL745 69 Rice 33 3 04/03/15 04/30/15 08/14/15 196 50/72 12
Clay CL271 78 Soybean 65 13 04/29/15 05/12/15 09/24/15 202 57/74 17
Cross Roy J 27 Soybean 90 21 04/08/15 04/16/15 09/12/15 139 66/74 15
Desha Jupiter 31 Soybean 70 19 03/31/15 04/14/15 09/22/15 170 57/68 18
Independence Jupiter 43 Soybean 70 20 04/09/15 04/21/15 09/25/15 196 55/73 18
Lawrence Mermentau 50 Soybean 80 25 04/27/15 05/09/15 09/22/15 163 64/73 18
Lee LaKast 54 Soybean 75 12 04/08/15 04/28/15 09/02/15 119 51/72 15
Lonoke CL151 36 Rice 65 16 04/11/15 04/25/15 09/02/15 167 59/72 16
Mississippi Jupiter 30 Fallow 90 16 04/06/15 04/15/15 09/01/15 191 65/68 21
Monroe Jupiter 60 Soybean 72 15 04/16/15 04/28/15 09/24/15 184 57/69 13
Phillips LaKast 47 Soybean 72 16 04/17/15 04/24/15 09/07/15 160 52/71 15
Randolph XL753 235 Soybean 22 6 05/06/15 05/15/15 10/06/15 237 53/71 14
St. Francis LaKast 84 Soybean 70 26 04/06/15 04/15/15 08/31/15 143 62/72 15
White XL753 42 Soybean 24 11 05/07/15 05/15/15 09/18/15 162 59/76 16
b c
Average 64 176 58/72 16
a
Head rice milling yield / total rice milling yield.
b
Seeding rates averaged 74 lb/acre for conventional cultivars and 25 lb/acre for hybrid cultivars.
c
Stand density averaged 19 plants/ft2 for conventional cultivars and 7 plants/ft2 for hybrid cultivars.
Table 2. Soil test results, fertilization program, and soil classification for fields enrolled in the 2015 Rice Research Verification Program.
Applied fertilizer
Urea (46% N)
Field location Soil test
Prefloodb rates applied Total
by county pH Pa Ka Zna N-P-K-Zna by timingc,d N ratee Soil classification
------------- (lb/acre)------------ -------------------(lb/acre)------------------- (lb N/acre)
Arkansas 6.3 40 145 7.5 0-30-90-0 235-0-70 140* Stuttgart and Dewitt silt loam
Ashley 8.1 43 487 4 0-40-60-0 330-0-70 184* Grubbs silt loam and Jackport silty clay loam
Chicot 6.9 32 788 5 0-60-0-0 250-0-70 147* Perry clay
Clay 5.8 96 278 19.6 0-0-60-0 207-100-0 141 Jackport silty clay loam
Cross 6.2 42 194 3.8 0-40-60-2 174-100-0 126 Crowley and Hillemann silt loams
Desha 5.7 51 179 2.2 0-0-90-10 270-100-0 170 Herbert silt loam and Perry clay
Independence 6.8 24 180 4 0-46-96-.5 185-100-0 131 Jackport silty clay loam/ Engham silt loam
Lawrence 6.6 56 232 3.4 0-40-48-8 250-100-0 161* Dubbs silt loam
Lee 7.8 32 185 3 21-40-75-10 240-100-0f 156* Calloway and Henry silt loam
Lonoke 6.7 47 713 5 0-0-0-0 190-0-0 87* Perry silty clay
Mississippi 6.9 56 844 9.6 0-0-0-0 250-150-0 184 Sharkey silty clay/Sharkey-Steele complex
Monroe 6.7 58 320 6.2 0-0-0-0 300-0-0 138* Grubbs silt loam and Jackport silty clay loam
Phillips 6 86 375 2 0-0-0-0 250 115* Calloway silt loam
Randolph 6 28 118 7.4 0-46-120-0 217-100-65g 176 Amagon/Dundee silt loams

St. Francis 7.2 52 184 8 0-47-77-0 272-100-0 171* Henry/Calloway/Loring silt loams
White 5.8 42 178 2.6 24-129-170-10h 295-65-0 203 Calloway/Immanuel silt loams
a
N = nitrogen, P = phosphorus, K = potassium, and Zn = zinc.
b
N-P2O5-K2O-Zn-S (includes seed treatments and preplant applications).
c
Timing: preflood - midseason - boot.
d
All preflood Urea applications utilized an approved NBPT product.
e
Column values with an (*) were fertilized according to N-STaR recommendations.
f
Only half of the field received a midseason N application.
g
Not a true midseason application but a pre-midseason correction for N loss due to adverse weather conditions.
h
Includes analysis established from 1.5 ton of chicken litter plus 0-30-90-10 fertilizer application.
39
Table 3. Herbicide rates and timings for fields enrolled in the 2015 Rice Research Verification Program.
40
Field location
by county Pre-emergence herbicide applications Post-emergence herbicide applications
a, b
(trade name and product rate/acre)
Arkansas Command (11 oz) + League (3.2 oz) Facet L (32 oz) + Permit Plus (0.75 oz) + COCc (1 pt)
Ashley Newpath (4 oz) + Aim (1 oz) Clearpath (0.5 lb) + Permit Plus (0.75 oz) + COC (1 pt)
Chicot Glyphosate (1.5 pt) + Command (24 oz) + Ricestar HT (24 oz) + COC (0.5 pt)
League (3.2 oz)
Clay Command (12.8 oz) + glyphosate (32 oz) Clearpath (0.5 lb) + COC (1 pt) fb Newpath (4 oz) + COC (1 pt)
Cross Command (12.8 oz) Facet (0.33lb) + propanil (3 qt) fb Ricestar (24 oz)
fb levee treatment 2,4-D (1 qt) + Grandstand (1 pt)
Desha Command (11 oz) + League (3.2 oz) Permit (1 oz) + COC (1 pt)
Independence Command (11 oz ) Command (8 oz) + Sharpen (1 oz) + COC (12.8 oz) fb spot spray
application Permit (1 oz) + Grasp (2 oz) + COC (1 qt)
Lawrence Command (12.8 oz) + glyphosate (48 oz) Command (8 oz) + Grasp Xtra (18 oz) + COC (1 qt)
Lee Glyphosate (32 oz) + Command (11 oz) Facet L (32 oz) + Permit (1 oz)
+ Sharpen (2 oz)
Lonoke Glyphosate (32 oz) + Command (12 oz) Ricestar HT (24 oz) + Command (12 oz)
+ League (3.2 oz)
Mississippi Prowl H2O (2.1 oz) + Roundup WeatherMAX (22 oz) None
+ Facet L (32 oz)
Monroe Glyphosate (32 oz) + Command (11 oz) + League (3.2 oz) RiceBeaux (4 qt) + Facet L (32 oz)
Phillips RiceBeaux (4 qt) + Facet L (8 oz) + Permit (1 oz)
Randolph Command herbicide (12.8 oz) Prowl H2O (2 oz) + Grasp (2 oz) + COC (1 qt)
fb post-flood application of Ricestar HT (24 oz)
St. Francis Early Spring Burndown: Sharpen (2 oz) + Facet (0.4 lb) + Sharpen (1 oz) + COC (1 pt) fb levee treatment:
glyphosate (26 oz) fb Sharpen (1 oz) + COC (1 pt)
Pre-emerge: Command (12.8 oz) +
glyphosate (28 oz) + League (3.2 oz)
White Prowl H2O (2.1 oz) Broadhead (9.2 oz) + Facet (2 oz) + Londax (1 oz) + COC (12.8 oz)
a
All rates specified are on a per-acre basis.
b
The abbreviation fb = followed by; used to separate herbicide application events.
c
COC = crop oil concentrate.
Table 4. Seed treatments used and foliar fungicide and insecticide
applications made on fields enrolled in the 2015 Rice Research Verification Program.
Seed treatments Foliar fungicide and insecticide treatments
Fungicide and/or Insecticide Fungicide Applications Fungicide Insecticide Insecticide
seed treatment for control for control of applications applications for applications for
Field location of diseases and insects sheath blight/kernel for control of control of rice control of rice stink
by county attacking seedling rice smut/false smut rice blast water weevil bug/chinch bug
(trade name and product rate/cwt seed) ------------------------------- (trade name and product rate/acre)--------------------------------
Arkansas RTSTa + CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Karate (3.7 oz)
Ashley RTST + CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Lambda-Cy (4 oz)
Chicot RTST + CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Lambda-Cy (4 oz)
Clay CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ ------
Cross CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Karate (2 oz)
Desha CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Lambda-Cy (5 oz)
Independence CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ Karate (2 oz)

Lawrence ------ Quadris (10 oz) ------ ------ ------
Lee CruiserMaxx Rice (7 oz/cwt) + Zinc ------ ------ ------ Lambda-Cy (4 oz)
Lonoke CruiserMaxx Rice (7 oz/cwt) + Zinc ------ ------ ------ ------
Mississippi ------ ------ ------ ------ Karate (2 oz)

Monroe CruiserMaxx Rice (7 oz/cwt) + Release ------ ------ ------ ------
Phillips CruiserMaxx Rice (7 oz/cwt) + Zinc ------ ------ ------ Karate (2 oz)
Randolph RTST + CruiserMaxx Rice (7 oz/cwt) Quilt Xcel (16 oz) ------ ------ ------
St. Francis Apron XL LS (0.64 oz/cwt) ------ ------ ------ ------
White RTST + CruiserMaxx Rice (7 oz/cwt) ------ ------ ------ ------
a
RTST refers to RiceTec Seed Treatment and is used to define those fields whose seed was treated by RiceTec, Inc. prior to seed purchase.
Seed is treated with compounds intended to enhance germination and early-season plant growth.
41
Table 5. Rainfall and irrigation information for fields

enrolled in the 2015 Rice Research Verification Program.
Field location by county Rainfall Irrigationa Rainfall + Irrigation
(inches) (acre-inches) (inches)
Arkansas 13.3 31.9 45.2
Ashley 19.8 28.4 48.2
Chicot 6.2 30.0* 36.2
Clay 18.5 32.9 51.4
Cross 22.8 24.4 47.2
Desha 18.7 27.0 45.7
Independence 26.1 35.2 61.3
Lawrence 18.7 82.0 100.7
Lee 15.8 33.9 49.6
Lonoke 15.0 48.0 63.0
Mississippi 22.3 21.0 43.3
Monroe 14.7 15.8 30.5
Phillips 17.8 30.0* 47.8
Randolph 19.9 28.4 48.3
St. Francis 15.7 37.0 52.7
White 18.7 11.4 30.1
17.8 32.3 50.1
a
Not all fields were equipped with flow meters to monitor water use for irrigation. Therefore, the
average irrigation amount used in fields with flow meters was calculated and this average was
used for fields with no irrigation data. Irrigation amounts using this calculated average are fol-
lowed by an asterisk (*).
42
Table 6. Operating costs, total costs, and returns for fields
enrolled in the 2015 Rice Research Verification Program.
Returns Returns
above above
Operating costs operating Fixed Total total Total
County Per acre Per bushel costs costs costs costs costs
($/acre) ($/bu) ------------------------------------ ($/acre)------------------------------------ ($/bu)
Arkansas 654.11 3.07 425.80 97.63 751.74 328.17 3.53
Ashley 670.49 4.04 131.29 80.75 751.23 50.55 4.53
Chicot 731.20 3.73 233.12 115.30 846.50 117.82 4.32
Clay 561.32 2.78 485.04 92.08 653.40 392.96 3.23
Cross 522.59 3.76 225.23 92.80 615.39 132.43 4.43
Desha 514.46 3.03 435.84 71.44 585.90 364.40 3.45
Independence 607.02 3.10 537.62 123.72 730.73 413.91 3.73
Lawrence 536.95 3.29 323.69 164.46 701.41 159.23 4.30
Lee 549.57 4.62 38.29 86.51 636.08 -48.22 5.35
Lonoke 411.49 2.46 443.55 115.38 526.87 328.17 3.15
Mississippi 427.69 2.24 680.11 100.36 528.05 579.75 2.76
Monroe 470.51 2.56 569.09 82.93 553.44 486.16 3.01
Phillips 474.40 2.97 311.20 88.17 562.57 223.03 3.52

Randolph 708.63 2.99 459.78 109.35 817.98 350.43 3.45
St. Francis 553.29 3.87 188.88 101.43 654.72 87.45 4.58
White 749.15 4.62 114.31 93.36 842.51 20.95 5.20
Average 571.43 3.32 350.18 100.98 672.41 249.20 3.91
43
Table 7. Summary of revenue and costs per acre for fields enrolled in the 2015 Rice Research Verification Program.
44
Receipts Arkansas Ashley Chicot Clay Cross Desha Independence Lawrence
Yield (bu) 213 166 196 202 139 170 196 163
Price received 5.07 4.83 4.92 5.18 5.38 5.59 5.84 5.28
Total crop revenue 1079.91 801.78 964.32 1046.36 747.82 950.30 1144.64 860.64

Operating costs
Seed 126.72 164.76 225.96 94.06 42.30 32.90 32.90 37.60
Fertilizers and nutrients 117.17 128.52 147.39 82.83 108.71 146.29 123.10 141.23
Chemicals 73.47 88.28 77.83 69.36 98.50 55.10 76.78 76.08
Custom applications 51.45 58.10 56.00 48.28 53.18 53.90 45.95 38.50
Diesel fuel 21.56 16.56 17.42 23.45 23.47 14.45 27.25 23.13
Repairs and maintenance 31.41 26.85 36.17 29.21 28.38 23.35 36.88 46.00
Irrigation energy costs 61.63 54.87 17.36 54.25 47.18 52.34 103.11 31.30
Labor, field activities 11.99 9.42 9.08 10.47 12.21 8.53 13.48 14.06
Other inputs and fees, pre-harvest 17.38 13.00 13.95 15.39 16.43 14.80 17.52 20.90
Post-harvest costs 141.33 110.14 130.05 134.03 92.23 112.80 130.05 108.15
Total operating expenses 654.11 670.49 731.20 561.32 522.59 514.46 607.02 536.95
Returns to operating costs 425.80 131.29 233.12 485.04 225.23 435.84 537.62 323.69

Capital recovery and fixed costs 97.63 80.75 115.30 92.08 92.80 71.44 123.72 164.46
Total specified costsa 751.74 751.23 846.50 653.40 615.39 585.90 730.73 701.41

Returns to specified costs 328.17 50.55 117.82 392.96 132.43 364.40 413.91 159.23

Operating costs/yield unit 3.07 4.04 3.73 2.78 3.76 3.03 3.10 3.29
Total costs/yield unit 3.53 4.53 4.32 3.23 4.43 3.45 3.73 4.30
continued
Table 7. Continued.
St.
Receipts Lee Lonoke Mississippi Monroe Phillips Randolph Francis White Average
Yield (bu) 119 167 191 184 160 237 143 162 176
Price received 4.94 5.12 5.80 5.65 4.91 4.93 5.19 5.33 5.25
Total crop revenue 587.86 855.04 1107.80 1039.60 785.60 1168.41 742.17 863.46 921.61

Operating costs
Seed 60.38 97.18 42.30 57.10 57.96 126.72 37.59 138.24 85.92
Fertilizers and nutrients 143.79 44.51 79.00 70.27 58.56 150.82 135.45 249.68 120.46
Chemicals 77.57 44.77 42.76 92.21 88.52 111.78 70.63 72.65 76.02
Custom applications 51.80 27.30 45.50 28.00 31.50 56.19 53.04 54.65 47.08
Diesel fuel 16.16 16.47 26.22 18.58 20.98 24.36 23.50 23.85 21.09
Repairs and maintenance 28.18 35.73 30.65 27.49 26.19 32.53 31.96 29.87 31.30
Irrigation energy costs 65.38 18.40 12.15 30.52 57.96 16.43 71.48 33.33 45.48
Labor, field activities 10.00 9.36 9.93 10.69 11.58 9.76 12.17 13.55 11.02
Other inputs and fees, pre-harvest 17.37 6.98 12.46 13.56 14.99 22.79 22.58 25.84 16.62
Post-harvest costs 78.96 110.80 126.73 122.08 106.16 157.25 94.88 107.49 116.44
Total operating costs 549.57 411.49 427.69 470.51 474.40 708.63 553.29 749.15 571.43
Returns to operating costs 38.29 443.55 680.11 569.09 311.20 459.78 188.88 114.31 350.18

Capital recovery and fixed costs 86.51 115.38 100.36 82.93 88.17 109.35 101.43 93.36 100.98
Total specified costsa 636.08 526.87 528.05 553.44 562.57 817.98 654.72 842.51 672.41

Returns to specified costs -48.22 328.17 579.75 486.16 223.03 350.43 87.45 20.95 249.20

Operating costs/yield unit 4.62 2.46 2.24 2.56 2.97 2.99 3.87 4.62 3.32
Total costs/yield unit 5.35 3.15 2.76 3.01 3.52 3.45 4.58 5.20 3.91
a
Does not include land costs, management, or other costs and fees not associated with production.
45
Table 8. Selected variable input costs per acre for fields enrolled in the 2015 Rice Research Verification Program.
46
Fertilizers Fungicides and Diesel Irrigation
County Rice type Seed and nutrients Herbicides Insecticides other inputs fuel energy costs
Arkansas XL753 126.72 117.17 66.83 6.65 --- 21.56 61.63
Ashley CL XL745 164.76 128.52 83.88 4.40 --- 16.56 54.87
Chicot CL XL745 225.96 147.39 73.21 4.62 --- 17.42 17.36
Clay CL271 94.06 82.83 65.86 --- 3.50 23.45 54.25
Cross Roy J 42.30 108.71 94.91 3.59 --- 23.47 47.18
Desha Jupiter 32.90 146.29 48.54 6.57 --- 14.45 52.34
Independence Jupiter 32.90 123.10 69.69 3.59 3.50 27.25 103.11
Lawrence Mermentau 37.60 141.23 56.68 --- 19.40 23.13 31.30
Lee LaKast 60.38 143.79 73.17 4.40 --- 16.16 65.38
Lonoke CL151 97.18 44.51 44.77 --- --- 16.47 18.40
Mississippi Jupiter 42.30 79.00 39.16 3.60 --- 26.22 12.15
Monroe Jupiter 57.10 70.27 92.21 --- --- 18.58 30.52
Phillips LaKast 57.96 58.56 84.93 3.59 --- 20.98 57.96
Randolph XL753 126.72 150.82 81.37 --- 30.41 24.36 16.43
St. Francis LaKast 37.59 135.45 70.63 --- --- 23.50 71.48
White XL753 138.24 249.68 72.65 --- --- 23.85 33.33
Average --- 85.92 120.46 69.90 4.56 14.20 21.09 45.48
Evaluation of Advanced Semi-Dwarf Medium-Grain and

Long-Grain Breeding Lines at Three Arkansas Locations
B.A. Beaty1, J.M. Bulloch1, M.W. Duren2, Y.D. Liyew3, S.D. Clark3, and X. Sha1
ABSTRACT
A precisely controlled yield trial under the most representative soil and environmental
conditions is critical for rice breeders to identify the ideal genotypes for potential varietal
releases. To bridge the gap between the single location, 2-replication preliminary yield
trials and the multi-state Uniform Regional Rice Nursery (URRN) and/or the multi-
location statewide Arkansas Rice Performance Trial (ARPT) which only accommodate
a very limited number of entries, an advanced yield trial (AYT) of 60 entries with 3
replications was initiated in 2015. This trial is conducted at the University of Arkansas
System Division of Agricultures Rice Research and Extension Center, (RREC), Stutt-
gart, Ark.; the Pine Tree Research Station, (PTRS), near Colt, Ark.; and the Northeast
Research and Extension Center, (NEREC), in Keiser, Ark. This new trial will help us
to select the best and the most uniform breeding lines for advancement into the URRN
and/or ARPT trials, and ultimately will improve the quality of those yield trials.
INTRODUCTION
Complicated rice traits, such as yield and quality can only be evaluated effectively
under small plot yield trials. Once reaching a reasonable uniformity, rice breeding lines
are bulk-harvested and tested in the single location, 2-replication preliminary yield tri-
als, which include the Clearfield Stuttgart Initial Trial (CSIT) or Conventional Stuttgart
Initial Trial (SIT). Each year, about 1000 new breeding lines are tested in CSIT or SIT
trials. About 10% of the tested breeding lines, which yield numerically higher than com-
mercial checks and possess desirable agronomical characteristics, need to be tested in
replicated and multi-location advanced yield trials. However, the current advanced yield
trials include the multi-state Uniform Regional Rice Nursery (URRN) and statewide
Arkansas Rice Performance Trial (ARPT) which only accommodate about 20 entries
1
Program Associate Program Associate I, and Associate Professor, respectively, Rice Research and
Extension Center, Stuttgart.
2
Program Technician III, Northeast Research and Extension Center, Keiser.
3
Research Program Technician and Resident Director in Charge, respectively, Pine Tree Research
Station, near Colt.
47
from each breeder each year. Obviously, a new replicated and multi-location trial is
needed to accommodate those additional breeding lines. In addition to the verification
of the findings in the previous preliminary trials, the new trial will result in purer and
more uniform seed stock for URRN and ARPT trials.
PROCEDURES
A total of 60 entries were tested in the 2015 AYT trial, which included 2 Louisi-
ana experimental hybrids, 48 Arkansas experimental lines (20 Clearfield long-grain, 4
Clearfield medium-grain, 15 semi-dwarf long-grain, 8 medium-grain, and 1 aromatic
long-grain line), and 10 commercial check varieties. Twenty-six of the experimental
lines were also concurrently tested in 2015 URRN and/or ARPT trials. The experimen-
tal design for all three locations is a randomized complete block with three replica-
tions. Plots measuring 5 ft wide (7 rows with 8-inch row spacing) and 14 ft long were
drill-seeded at 75 lb/acre rate. The soil types at the Northeast Research and Extension
Center (NEREC), the Pine Tree Research Station (PTRS), and the Rice Research and
Extension Center (RREC) are Sharkey clay, Calloway silt loam, and DeWitt silt loam,
respectively. Planting dates at NEREC, PTRS, and RREC were 4 May, 1 May, and 22
April, respectively. A single preflood application of 150 lb nitrogen (N) in the form
of urea was applied to a dry soil surface at 4- to 5-lf stage, and a permanent flood was
established 2 days later. At maturity, the six rows (including a border row) of each plot
were harvested by using a Wintersteiger plot combine (Wintersteiger AG, 4910 Ried,
Austria), and the moisture content and plot weight were determined by the automated
weighing system Harvest Master that is integrated into the combine. A small sample
of seed was collected from the combine for each plot for later milling yield determina-
tion. Milling evaluations were conducted by Riceland Foods, Inc., Stuttgart, Ark. Grain
yields were calculated as bushel per acre at 12% moisture.
Data were analyzed using the General Linear Model procedure of SAS software,
v. 9.2 (SAS Institute, Inc., Cary, N.C.). Analysis of variance for grain yield, milling
yields, days to 50% heading, plant height, and seedling vigor were performed for each
location, and a combined analysis was conducted across the three locations. The means
were separated by Fishers protected least significant difference (LSD) test at the 0.05
probability level.

The average grain yield of all genotypes across 3 locations is 185 bu/acre. Among
3 locations, RREC has the highest yield of 192, followed by 190 and 172 bu/acre of
NEREC and PTRS, respectively. Of the 60 entries, experimental hybrid CLH161
(15AYT011) had the highest averaged grain yield of 229 bu/acre, followed by 217,
210, and 207 bu/acre of LaKast (15AYT009), 15AR1111 (15AYT023), and Titan
(15AYT014), respectively (Table 1). Milling yields are very high for all locations with
the overall head rice of 67% and total rice of 72%. The average seedling vigor is 4.1
48
which is normal, the average days to 50% heading is 81 days, and the average plant height
is 41 inches (Table 2). Lodging was observed on two Louisiana experimental hybrids
(15AYT011 and 15AYT012) with the lodging incidence of 1% and 17%, respectively.
Two Clearfield long-grain lines, 15AYT015 (15AR1024) and 15AYT032
(15AR1170), had a numerically higher grain yield than the check CL111 and CL151
(Table 1), while all four Clearfield medium-grain lines, 15AYT023 (15AR1111),
15AYT016 (15AR1027), 15AYT022 (15AR1099), and 15AYT021 (15AR1096), had
either a statistically or numerically higher grain yield than the check CL271. Two
conventional medium-grain lines, Titan (15AYT014) and 15AYT019 (15AR1050) had
a numerically higher grain yield than the commercial check Jupiter. All conventional
long-grain lines yielded lower than the check LaKast which has the conventional
height, however four of them (15AYT048, 15AYT051, 15AYT052, and 15AYT053)
had a numerically higher grain yield than the semidwarf check Mermentau. Some of
these lines were selected for purification and increase in the winter nursery in Lajas,
Puerto Rico in winter 2015.
The new AYT trial successfully bridged the gap, between the single location
preliminary yield trials with numerous entries and the multi-state or statewide advanced
yield trial which can only accommodate a very limited number of entries, by offering
the space for the trial of additional elite breeding lines. Our results enable us to verify
the findings from other yield trials, and to identify the outstanding breeding lines which
were excluded from URRN or ARPT trials due to insufficient space.
ACKNOWLEDGMENTS
This research is partially funded by the Arkansas Rice Research and Promotion
Board. Technical supports from Nicholas Murphy, Emily Carr, and Jenifer Clemons
are greatly appreciated. Support also provided by the University of Arkansas System
Division of Agriculture.
49
Table 1. Grain yield of 60 semi-dwarf long- and medium-grain breeding lines and
50
commercial checks in the advanced yield trial (AYT) conducted at Northeast Research and Extension Center (NEREC) at
Keiser, Ark.; Pine Tree Research Station (PTRS), near Colt, Ark.; and Rice Research and Extension Center (RREC), Stuttgart, Ark., 2015.
Grain yield
Entry Pedigree NEREC PTRS RREC Mean
---------------------- (bu/acre)------------------------
15AYT001 CL111 199 188 205 197
15AYT002 CL151 215 170 202 195
15AYT003 CL152 172 170 197 180
15AYT004 CL163 189 174 197 187
15AYT005 CL172 183 171 197 184
15AYT006 CL271 206 162 195 187
15AYT007 Mermentau 195 176 201 191
15AYT008 Roy J 178 158 215 184
15AYT009 Lakast 225 209 218 217
15AYT010 Jupiter 215 178 189 194
15AYT011 CLH161 239 207 241 229
15AYT012 09A/R608 222 152 208 194
15AYT013 CYBT/LM1/4/WLLS/PI597049/3/RSMT//NWBT/KATY/5/9901133/JEFF 170 161 190 174
15AYT014 Titan 213 199 210 207
15AYT015 CL111/3/CCDR//9502008/LGRU 204 186 210 200
15AYT016 NPTN//BNGL/CL161 222 172 200 198
15AYT017 CPRS/KBNT//9502008-A 186 180 183 183
15AYT018 RU0902125/CL131 197 163 172 177
15AYT019 RU0501136/RU0902162 214 207 185 202
15AYT020 CL111//CCDR/0502085 205 162 166 178
15AYT021 STG07IMI-01-129/JPTR 206 188 182 192
15AYT022 BNGL/CL161/4/9502065/3/MERC//MERC/ 210 189 185 194
15AYT023 BNGL/CL161/4/9502065/3/MERC//MERC/ 225 197 208 210
15AYT024 CL111/CCDR 195 176 181 184
15AYT025 KATY/CPRS//JKSN/3/AR1188/CCDR/4/CFX-29/CCDR 186 171 197 185
15AYT026 CCDR/RU0801167 173 163 186 174
continued
Table 1. Continued.
Grain yield
---------------------- (bu/acre)------------------------
15AYT027 CPRS/KBNT//CFX29/CCDR/3/06CFP952 157 165 190 171
15AYT028 CPRS/KBNT//WLLS/CFX18/5/9502008/3/CPRS//82CAY21/TBNT/4/CPRS// 181 167 190 179
15AYT029 RU0902174/RU0902134 173 175 197 182
15AYT031 RU0902125/CL131 182 166 191 180
15AYT032 CHNR/4/CPRS/9502008-A/3/CFX 29//AR1142/LA2031 185 203 209 199
15AYT033 CPRS/KBNT//9502008-A/3/TRNS/4/CPRS/KBNT//WLLS/CFX18 155 162 199 172
15AYT035 CYBT/LM1/4/WLLS/PI597049/3/RU0901145 176 142 173 164
15AYT036 MRMT/4/9502008//AR1188/CCDR/3/CCDR 183 164 195 181
15AYT037 EARL/4/9502065/3/BNGL//MERC/RICO 208 186 182 192
15AYT038 RU0902162/RU0801124 216 174 180 190
15AYT039 TACAURI/3/CPRS//82CAY21/TBNT/4/CFX18/5/REX 193 184 197 191
15AYT040 CHNR/4/CPRS/9502008-A/3/CFX 29//AR1142/LA2031 179 165 191 178
15AYT041 CL181/CL111 186 136 138 153
15AYT042 CPRS/KBNT//WLLS/CFX18/3/CHNR 183 163 183 176

15AYT043 TACAURI/3/CPRS//82CAY21/TBNT/4/CFX18/5/CPRS/KBNT//WLLS/CFX18 196 162 208 188
15AYT044 CL111/3/CCDR//9502008/LGRU 177 154 177 169
15AYT045 CCDR/JEFF//CFX-18//CCDR/9770532 DH2 177 139 171 162
15AYT046 KATY/CPRS//JKSN/3/AR1188/CCDR/4/CFX-29/CCDR 182 163 171 172
15AYT047 CYBT/LM1//CHNR/3/ADAR/JDON//JEFF 168 165 186 173
15AYT048 CYBT/LM1//CHNR/3/RU0901102 209 194 190 198
15AYT049 CCDR/RU0801167 165 182 189 179
15AYT050 CCDR/RU1002183 187 177 195 186
15AYT051 FRNS/RU0902137 179 188 218 195
15AYT052 MRMT/RU0802134 186 187 204 192
15AYT053 MRMT/4/9502008//AR1188/CCDR/3/CCDR 190 177 207 191
15AYT054 CCDR//CCDR/JEFF 173 177 178 176
15AYT055 0402022/3/9502008//AR1142/MBLE/4/CTHL 168 149 210 176
continued
51
Table 1. Continued.
52
Grain yield
---------------------- (bu/acre)------------------------
15AYT056 M207/JPTR//JPTR 185 179 191 185
15AYT057 CFFY/STG07M-07-096 203 162 154 173
15AYT058 JPTR/RU1001102 187 153 199 179
15AYT059 BNGL//ORIN/BNGL 199 184 191 191
15AYT060 L202/Leah//Toro/3/IR67016 132 127 174 144

c.v.(%)a 6.6 10.3 5.8 7.7
LSD0.05 20 29 18 13
a
c.v. = coefficient of variance; LSD = least significant difference.
Table 2. Average seedling vigor, days to 50% heading, plant height, and milling yields of 2015
advanced yield trial (AYT) conducted at Northeast Research and Extension Center (NEREC) at Keiser, Ark.;
Pine Tree Research Station (PTRS), near Colt, Ark.; and Rice Research and Extension Center (RREC), Stuttgart, Ark.
Seedling 50% Plant Milling
Entry Pedigree vigora heading height yieldb
(days) (in.)
15AYT001 CL111 3.3 79 43 68-71
15AYT002 CL151 3.3 80 41 66-70
15AYT003 CL152 3.9 84 41 67-70
15AYT004 CL163 3.7 84 41 66-70
15AYT005 CL172 3.8 82 39 67-70
15AYT006 CL271 3.3 81 40 67-71
15AYT007 Mermentau 3.4 82 41 67-70
15AYT008 Roy J 4.1 86 42 64-69
15AYT009 Lakast 3.9 79 46 67-71
15AYT010 Jupiter 3.7 81 37 68-70
15AYT011 CLH161 4.1 79 53 65-70
15AYT012 09A/R608 3.8 87 54 60-66
15AYT013 CYBT/LM1/4/WLLS/PI597049/3/RSMT//NWBT/KATY/5/9901133/JEFF 4.7 82 40 66-71
15AYT014 Titan 4.1 77 40 66-71

15AYT015 CL111/3/CCDR//9502008/LGRU 4.2 81 41 67-71
15AYT016 NPTN//BNGL/CL161 4.2 81 40 68-71
15AYT017 CPRS/KBNT//9502008-A 4.7 83 40 69-71
15AYT018 RU0902125/CL131 3.4 82 38 68-71
15AYT019 RU0501136/RU0902162 4.0 80 37 68-72
15AYT020 CL111//CCDR/0502085 3.9 79 40 66-71
15AYT021 STG07IMI-01-129/JPTR 3.7 82 39 67-70
15AYT022 BNGL/CL161/4/9502065/3/MERC//MERC/ 4.0 79 41 67-72
15AYT023 BNGL/CL161/4/9502065/3/MERC//MERC/ 4.2 79 41 68-72
15AYT024 CL111/CCDR 4.2 80 42 66-70
15AYT025 KATY/CPRS//JKSN/3/AR1188/CCDR/4/CFX-29/CCDR 4.1 81 44 67-71
15AYT026 CCDR/RU0801167 4.8 82 43 67-71
15AYT027 CPRS/KBNT//CFX29/CCDR/3/06CFP952 4.4 82 42 66-69
15AYT028 CPRS/KBNT//WLLS/CFX18/5/9502008/3/CPRS//82CAY21/TBNT/4/CPRS// 4.7 83 40 68-71
53
continued
Table 2. Continued.
54
(days) (in.)
15AYT029 RU0902174/RU0902134 4.0 83 42 69-71
15AYT031 RU0902125/CL131 4.1 81 40 66-69
15AYT032 CHNR/4/CPRS/9502008-A/3/CFX 29//AR1142/LA2031 4.3 84 41 68-71
15AYT033 CPRS/KBNT//9502008-A/3/TRNS/4/CPRS/KBNT//WLLS/CFX18 4.3 82 41 68-71
15AYT035 CYBT/LM1/4/WLLS/PI597049/3/RU0901145 4.6 82 44 67-71
15AYT036 MRMT/4/9502008//AR1188/CCDR/3/CCDR 4.2 82 42 67-70
15AYT037 EARL/4/9502065/3/BNGL//MERC/RICO 4.2 80 38 68-71
15AYT038 RU0902162/RU0801124 4.4 81 40 68-72
15AYT039 TACAURI/3/CPRS//82CAY21/TBNT/4/CFX18/5/REX 4.0 79 43 67-70
15AYT040 CHNR/4/CPRS/9502008-A/3/CFX 29//AR1142/LA2031 4.0 82 39 69-72
15AYT041 CL181/CL111 4.3 82 39 63-69
15AYT042 CPRS/KBNT//WLLS/CFX18/3/CHNR 4.2 81 42 69-72
15AYT043 TACAURI/3/CPRS//82CAY21/TBNT/4/CFX18/5/CPRS/KBNT//WLLS/CFX18 4.2 82 41 67-70
15AYT044 CL111/3/CCDR//9502008/LGRU 4.3 82 41 66-70
15AYT045 CCDR/JEFF//CFX-18//CCDR/9770532 DH2 4.1 79 40 68-70
15AYT046 KATY/CPRS//JKSN/3/AR1188/CCDR/4/CFX-29/CCDR 3.8 79 41 67-71
15AYT047 CYBT/LM1//CHNR/3/ADAR/JDON//JEFF 4.1 81 44 65-70
15AYT048 CYBT/LM1//CHNR/3/RU0901102 4.6 77 41 61-71
15AYT049 CCDR/RU0801167 4.2 79 47 66-71
15AYT050 CCDR/RU1002183 4.1 82 42 67-71
15AYT051 FRNS/RU0902137 4.2 83 43 66-70
15AYT052 MRMT/RU0802134 3.9 82 40 67-70
15AYT053 MRMT/4/9502008//AR1188/CCDR/3/CCDR 4.2 80 40 67-70
15AYT054 CCDR//CCDR/JEFF 4.1 81 40 67-71
15AYT055 0402022/3/9502008//AR1142/MBLE/4/CTHL 4.3 84 44 69-72
15AYT056 M207/JPTR//JPTR 4.1 82 40 67-70
15AYT057 CFFY/STG07M-07-096 4.4 78 42 68-71
15AYT058 JPTR/RU1001102 4.2 81 38 66-70
continued
Table 2. Continued.
(days) (in.)
15AYT059 BNGL//ORIN/BNGL 4.0 80 39 67-71
15AYT060 L202/Leah//Toro/3/IR67016 4.0 80 42 63-70

c.v.(%)c 14.1 1.8 4.4 3-1
LSD0.05 0.5 1 2 2-1
a
A subjective rating 1 to 7 taken at emergence, 1 = excellent stand and 7 = no stand.
b
Milling yield = % head : % total.
c
c.v. = coefficient of variance; LSD = least significant difference.
55
Kompetitive Allele-Specific Polymerase Chain Reaction (KASPTM)

Marker-Assisted Selection for the Development of Rice Varieties
V.A. Boyett1, V.I. Thompson1, X. Sha1, K.A.K. Moldenhauer1,
D.K.A. Wisdom1, J.M Bulloch1, and H.H.M Moldenhauer1
ABSTRACT
Researchers in molecular genetics at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center (RREC) have been performing DNA
marker-assisted selection (MAS) for over 15 years. The vast majority of DNA markers
have been rice microsatellite or simple sequence repeat (SSR) and insertion-deletion
(InDel) markers; recently wth the new equipment, a few of the more informative mark-
ers in rice, single nucleotide polymorphism (SNP) markers, are being used more often.
Microsatellite and InDel markers are analyzed easily by capillary electrophoresis, but
SNP markers can be particularly challenging. Cross-priming and preferential amplifi-
cation have been significant problems, often rendering the data useless for interpreta-
tion. A newer technology, Kompetitive Allele-Specific Polymerase Chain Reaction
(KASPTM) resolves these issues. In 2015, the Molecular Genetics lab worked on five
major projects for breeding involving DNA marker-assisted selection for the important
traits of cooking quality, rice blast disease resistance, and Clearfield resistance. One
project consisted of molecular quantitative trait loci (QTL) mapping, four other smaller
projects were conducted for the breeding program and four small proprietary projects
for Extension clients. The lab processed 3213 mostly bulked genomic DNA samples,
generating 21,277 data points. Over 30% (6555 data points) of the total data generated
for the year were derived from KASP marker analysis.
INTRODUCTION
Currently there are four rice breeding programs and cooperative extension ac-
tivities which utilize the laboratory. Much of the effort over the last 15 years has been
devoted to the genotypic characterization of parental lines and progeny in the areas of
new long-grain and medium-grain cultivar development, hybrid rice breeding, back-
1
Program Associate II, Program Technician II, Associate Professor, Professor, Program Associate II, Pro-
gram Associate I, and student worker, respectively, Rice Research and Extension Center, Stuttgart.
56
cross populations, aromatic rice breeding, genomic mapping of specific traits, and seed
purification. Single nucleotide polymorphisms (SNPs) are distributed throughout the
rice genome in high abundance (Liu and Zhang, 2006; Mammadov et al., 2012; Nasu et
al., 2002; and Singh et al., 2013). Liu and Zhang (2006) reported that a total of 80,127
SNP sites had been identified in the rice genome, with one SNP/154 bp found between
indica and japonica rice subspecies. Seven years later, Singh et al. (2013) reported that
the incidence of SNPs was found to be one SNP/140 bp in the rice genome.
These SNPs are valuable molecular markers for rapid varietal identification (Singh
et al., 2013), generating high-resolution genetic maps (Liu and Zhang, 2006; Nasu et al.,
2002; Singh et al., 2013), studying population structure, and discovering marker-trait
relationships in association-mapping experiments (Singh et al., 2013), and for use in
marker-assisted selection (MAS) in rice breeding (Liu and Zhang, 2006).
In addition to their abundance and distribution in the rice genome, SNP-based mark-
ers are gaining in popularity for genotyping due their amenability for high-throughput
detection formats and platforms (Liu and Zhang, 2006; Mammadov et al., 2012; Nasu
et al., 2002; Singh et al., 2013). Many different genotyping platforms and chemistries
have been developed, making the analysis of SNP markers more rapid and efficient
(Mammadov et al., 2012; Nasu et al., 2002). One of these platforms is Kompetitive
Allele-Specific Polymerase Chain Reaction (KASP) chemistry.
Developed over 10 years ago, KASP is a Fluorescence Resonance Energy Transfer
(FRET)-based endpoint detection platform capable of detecting SNP and InDel markers
(LGC Genomics, Beverly, Mass.). The KASP Assay mix contains three assay-specific
non-labelled oligos: two allele-specific forward primers and one common reverse
primer. The allele-specific primers each have a unique tail sequence that corresponds
with a quenched universal FRET cassette. One cassette is labelled with FAM dye
and the other with HEX dye. The KASP Master Mix contains the universal FRET
cassettes, ROX passive reference dye, Taq polymerase, free nucleotides and MgCl2
in an optimized buffer. In the first round of PCR, the relevant allele-specific primer
binds to the DNA template and elongates. This attaches the tail sequence to the newly
synthesized strand. The complement of the allele specific tail is then generated during
subsequent rounds of PCR. This enables the FRET cassette to bind to the DNA. The
FRET cassette is no longer quenched and emits fluorescence. Bi-allelic discrimination
is achieved through the competitive binding of the two allele-specific forward primers.
If the genotype at a given SNP is homozygous, only one of the two possible fluorescent
signals (HEX or FAM) will be generated. If the genotype is heterozygous, a mixed
fluorescent signal will be generated (Green).
In 2015, materials from the RREC Rice Breeding Programs were screened with
KASP markers linked to the traits of amylose (Conaway-Bormans et al., 2003; Mc-
Clung et al., 2004), relative viscosity (McClung et al., 2004), gelatinization temperature
(McClung et al., 2004), leaf surface texture (Fjellstrom, pers. comm.), and Clearfield
herbicide resistance (Kadaru et al., 2008; Rosas et al., 2014).
The objective of this ongoing study is to apply DNA marker technology to assist
with the mission of the RREC Rice Breeding Programs. The goals include (i) characteriz-
ing parental materials on a molecular level for important agronomic traits and purity, (ii)
57
performing DNA marker-assisted selection of progeny to confirm identity and track gene
introgression, and (iii) ensuring seed quality and uniformity by eliminating off types.
PROCEDURES
Leaf tissue from individually tagged field plants or greenhouse-grown seedlings
was collected in manila coin envelopes and kept in plastic bags on ice until being placed
in storage at the molecular genetics lab. In some instances, seeds were germinated in
Petri dishes to obtain leaf tissue. The leaf tissue was stored at -80 C until sampled.
Total genomic DNA was extracted from the embryo using a Sodium hydroxide/Tween
20 buffer and neutralized with 100mM TRIS-HCl, 2 mM EDTA (Xin et al., 2003).
Each set of DNA samples was arrayed in a 96-well format and processed through
a OneStep-96 PCR Inhibitor Removal system (Zymo Research Corporation, Irvine,
Calif.). Eleven samples on the plate were assessed for DNA concentration and purity at
the wavelengths 260 and 280 nm using an Eppendorf BioPhotometer spectrophotom-
eter. Using the median DNA concentration of those 11 samples, the DNA of the entire
96-well plate was diluted in water to 7-8 ng/l.
The KASP reactions were prepared by adding 5 l of each DNA sample and 5
l of the 2X Master Mix + 0.14 l Assay Mix to the wells of a 96-well opaque qPCR
plate (LGC Genomics, Beverly, Mass.). The plate was then sealed with qPCR film (LGC
Genomics, Beverly, Mass.), and the KASP reactions were cycled in a Mastercycler
Gradient S thermal cycler (Eppendorf North America, Inc., Westbury, N.Y.) using a 61-
55 C Touchdown protocol. The plates were then allowed to cool to room temperature
prior to reading on a BMG Labtech FLUOstar Omega SNP plate reader (LGC Genom-
ics, Beverly, Mass.). Detected fluorescence was analyzed using KlusterCaller software
(LGC Genomics, Beverly, Mass.).

Amplification was robust and the data was interpreted easily without any of the
previous issues such as cross-priming (Fig. 1). The KASP marker analysis provided
simple and straightforward allele calls, allowing more confidence that the interpretation
of the data was accurate. Of the 6555 data points generated with KASP markers, 95%
of the analysis was conducted on five major rice breeding projects for the purpose of
marker-assisted selection for the development of new rice varieties. The remaining 5%
of the KASP analysis was for identification purposes.
Waxy Exon 1 and Waxy Exon 6, two markers linked to amylose content, confirmed
the previous amylose potential determined by the SSR marker RM190 in all the popula-
tions. There was a 100% correlation between the two different chemistries, giving the
rice breeder a higher degree of confidence in the phenotype prediction. Waxy Exon 10,
a marker linked to relative viscosity (RVA) of the rice grain, revealed that the all the
breeding populations in the RREC program in 2015 have a weak RVA, which is typi-
cal of southern long- and medium-grain cooking quality. Only a high amylose DNA
58
control sample amplified the strong RVA allele. Alk, a marker linked to gelatinization
temperature, was used to complete the grain cooking quality profile in tested populations.
Clearfield herbicide resistance trait was determined by the KASP markers for the
SNPs of S653D, G654E, and A122T. The S653D SNP is the one in the vast majority of
southern U.S. Clearfield varieties, but there is the possibility that materials exist in the
breeding programs that have the G654E SNP in their pedigree. The A122T SNP is found
in germplasm from South America, but the marker was ordered and validated to have for
collaborative efforts with the scientists at Dale Bumpers National Rice Research Center.
To determine leaf surface texture, the GlabSNP KASP marker was validated for
use in the program at the RREC, but has not been used for MAS at this time. Work is
ongoing in obtaining a viable KASP marker for the rice blast resistance gene Pi-ta.
Marker screening of breeding materials revealed that progress is being made in
the RREC Rice Breeding Programs in reducing trait segregation and identifying prom-
ising lines to advance. Applying molecular marker technology to the Rice Breeding
Programs enabled the breeders to assess the status of the populations, and eliminate
those materials that are not desirable for inclusion in future rice breeding efforts, saving
time, resources, and expenses.
ACKNOWLEDGMENTS
This work was supported by funding to the RREC provided by the rice farmers
of Arkansas and administered by the Arkansas Rice Research and Promotion Board.
The authors thank Greg Berger for assistance in obtaining funding for installing KASP
technology in the molecular genetics lab at the RREC. We thank Aaron Jackson for
Clearfield and Pi-ta gene sequence information. Support also provided by the University
LITERATURE CITED
Conaway-Bormans, C.A., M.A. Marchetti, C.W. Johnson, A.M. McClung, and W.D.
Park. 2003. Molecular markers linked to the blast resistance gene Pi-z in rice for
use in marker -assisted selection. Theor. Appl. Genet. 107:1014-1020.
Kadaru, S., W. Zhang, A. Yadav, and J. Oard. 2008. Development and application
of allele-specific PCR assays for imazethapyr resistance in rice (Oryza sativa).
Euphytica 160:431-438.
Liu, C.G. and G.Q. Zhang. 2006. Single nucleotide polymorphism (SNP) and its ap-
plication in rice. Yi Chuan 28(6):737-744.
Mammadov, J., R. Aggarwal, R. Buyyarapu, and S. Kumpatla. 2012. SNP markers
and their impact on plant breeding. Intl. Jour. Plant Genom. 2012:728398.
59
McClung, A.M., M. Chen, H. Bockelman, R. Bryant, W. Yan, and R.G. Fjellstrom.

2004. Characterization of a core collection of rice germplasm and elite breeding
lines in the U.S. with genetic markers associated with cooking quality. PAG XII;
10-14 Jan. 2004; San Diego, Calif.
Nasu, S., J. Suzuki, R. Ohta, K. Hasegawa, R. Yui, N. Kitazawa, L. Monna, and
Y. Minobe. 2002. Search for and analysis of single nucleotide polymorphisms
(SNPs) in rice (Oryza sativa, Oryza rufipogon) and establishment of SNP mark-
ers. DNA Res. 9:163-171.
Rosas, J.E., V. Bonnecarrre, and F.P. de Vida. 2014. One-step, codominant detection
of imidazolinone resistance mutations in weedy rice (Oryza sativa L.). Electron. J.
Biotechnol.
Singh, N., D.R. Choudhury, A.K. Singh, S. Kumar, K. Srinivasan, R.K. Tyagi, N.K.
Singh, and R. Singh. 2013. Comparison of SSR and SNP markers in estimation
of genetic diversity and population structure of Indian rice varieties. PLoS ONE
8(12).
Xin, Z., J.P. Velten, M.J. Oliver, and J.J. Burke. 2003. High-Throughput DNA Ex-
traction Method Suitable for PCR. BioTech. 34:4:820-825.
60
Fig. 1. KlusterCaller image of S653D Clearfield single nucleotide polymorphism (SNP).
61
Titan, a Very Early-Maturing, High-Yielding, and

High-Quality Conventional Medium-Grain Rice Variety
J.M. Bulloch1, B.A. Beaty1, X. Sha1, K.A.K. Moldenhauer1,
J.W. Gibbons1, J.T. Hardke2, R.J. Norman3, C.E. Wilson Jr.1,
Y.A. Wamishe4, T.J. Siebenmorgan5, D.K.A. Wisdom1, M.M. Blocker1,
D.L. McCarty1, V.A. Boyett1, D.L. Frizzell1, and E. Castaneda-Gonzalez1
ABSTRACT
Titan (Oryza sativa L.) is a high-yielding, very early-maturing, and short stature
medium-grain rice variety developed at the University of Arkansas System Division
of Agricultures Rice Research and Extension Center, (RREC) near Stuttgart, Ark. It
was officially approved for the formal release by the Division of Agriculture in Febru-
ary 2016. In yield trials conducted during 2012-2015, Titan had a 5 to 10 bu/acre yield
advantage over the predominant medium-grain cultivar Jupiter. Compared with Jupiter,
Titan matures about 5 days earlier, has a better resistance to leaf blast, and slightly
better lodging tolerance. It also has typical southern medium-grain quality but a much
larger kernel size than Jupiter. Successful development of medium-grain variety Titan
certainly will provide rice producers the better option in their choice of variety and
management systems for Arkansas rice production.
INTRODUCTION
Medium-grain rice is the important component of Arkansas rice. Arkansas ranks
second in medium-grain rice production in the United States only behind California.
During 2005-2014, an average of 0.15 million acres of medium-grain rice was grown
annually, which makes up about 10% of total state rice acreage (USDA-ERS, 2015).
The current predominant medium-grain variety Jupiter was released 10 years ago by
Rice Research Station, Louisiana State University Agricultural Center in Crowley, La.
1
Program Associate I, Program Associate, Associate Professor, Professor, Former Rice Breeder, Center
Director, Program Technician I, Program Associate III, Program Technician I, Program Associate I,
Program Associate III, and Program Associate I, respectively, Rice Research and Extension Center,
Stuttgart.
2
Rice Extension Agronomist, Department of Crop, Soil, and Environmental Science, Stuttgart.
3
Professor, Department of Crop, Soil, and Environmental Science, Fayetteville.
4
Assistant Professor, Department of Plant Pathology, Stuttgart.
5
Distinguished Professor, Department of Food Science, Fayetteville.
62
Jupiter has started to show a breakdown in disease resistance especially for leaf blast.
The small and short kernel size of Jupiter always has been a concern for medium-grain
rice end-users other than Kelloggs. Therefore, a new conventional medium-grain variety
is urgently needed to complement Jupiter.
PROCEDURES
Titan was originally selected from the cross M-206//Bengal/Lafitte/3/Jupiter made
in 2006. M-206 is a Calrose-type medium-grain rice cultivar developed by California
Rice Research Station at Biggs, Calif., while Bengal, Lafitte, and Jupiter are southern
medium-grain rice cultivars developed by Rice Research Station, Louisiana State
University Agricultural Center in Crowley (Linscombe et al., 1993; Linscombe et al.,
1997; Sha et al., 2006). Jupiter is the current predominant southern medium-grain cul-
tivar which accounts for the majority of the southern medium-grain rice acreage. Titan
was introduced as a F4 bulk of a single progeny row STG09PR-81-087 in the winter
nursery in Lajas, Puerto Rico in spring 2010. It was evaluated in the preliminary yield
trial (experimental designation 2010 PREL1179) in 2010, advanced to 2011 Stuttgart
initial trial (SIT) as entry JG SIT2060, entered the Arkansas Rice Performance Trial
(ARPT) in 2012, and the Cooperative Uniform Regional Rice Nurseries in 2013 with
the experimental designation RU1301021.

Titan appears to have an outstanding yield potential, good milling and grain qual-
ity, and improved lodging and blast resistance compared with the current commercial
cultivar Jupiter. In 53 statewide and regional trials during 2012-2015, the average
grain yield of Titan was 9054 lb/acre or 201 bu/acre compared with 8829 or 196 for
Jupiter. Average milling yields (g kg-1 whole milled kernels : g kg-1 total milled rice)
at 120 g kg-1 moisture in 45 state and regional tests from 2012-2015 were 594:691 for
Titan, and 624:685 for Jupiter. Titan has a semi-dwarf plant type and is moderately
resistant to lodging. It averaged 98 cm in height in yield tests across the mid-South and
is slightly taller than the 95 cm of Jupiter. However, Titan matures much earlier than
Jupiter. The average number of days from emergence to 50% heading is 80 compared
with Jupiter at 86.
Titan has the typical medium-grain shape, and its kernels appear much larger
and longer than that of Jupiter. Based on the analyses conducted by Riceland Foods,
Inc. (Stuttgart, Ark.) on 12 different sets of samples collected across Arkansas during
2012-2015, the length and width (mm), length/width ratio, and kernel weight (mg) of
milled whole kernels were 5.91, 2.68, 2.21, and 23.20 for Titan as compared with 5.57,
2.66, 2.09, and 21.03 for Jupiter, respectively. Average apparent amylose content of
Titan is 150 g kg-1 compared with 156 g kg-1 of Jupiter. Titan also has a low gelatiniza-
tion temperature of 62.8 C similar to the 62.7 C of Jupiter. These results indicate that
Titan has typical U.S. medium-grain rice cooking characteristics.
63
Results from the upland rice blast nursery for leaf blast [caused by Pyricularia
grisea (Cooke) Sacc.] indicated that Titan has moderate resistance with a rating of 1.8 on
a disease scale of 0 = immune, 9 = highly susceptible, as compared with 4.8 of Jupiter.
Molecular markers also confirmed that Titan possesses both blast resistant genes Pi-z
and Pi-ks as compared to Jupiters Pi-ks gene. In a greenhouse inoculated test, Titan was
susceptible to blast races IB-1, IB-33, and IB-49, but resistant to IC-17, IE-1, IG-1, and
IE1-K. Under natural infestation or inoculated evaluation, Titan appeared moderately
susceptible to sheath blight (caused by Rhizoctonia solani Khn) and susceptible to
bacterial panicle blight (caused by Burkholderia glumae).
The flag leaf of Titan is longer than that of Jupiter and well above the panicle
canopy at maturity. The leaves, lemma, and palea are glabrous. The spikelet is straw
colored. The apiculus is red or purple at heading and the color fades as grains approach
maturity. The grain is non-aromatic.
Variants observed and removed from increase fields of Titan were primarily taller
and earlier. Other variants included any combination of the following: pubescent, earlier,
shorter, long-grain and intermediate grain types, and gold hull. The total number of
variants numbered less than 1 per 5000 plants.
About 790 hundred weight foundation seeds have been produced at the Rice
Research and Extension Center, and they are available to seed rice growers for the
registered seed production in 2016.
Successful development of the new medium-grain rice variety Titan offers pro-
ducers options in their choice of variety and management systems for Arkansas rice
production. Continued utilization of new germplasm through exchange and introduction
remains important for Arkansas rice improvement.
ACKNOWLEDGMENTS
The authors wish to express their appreciation for funding and support for this
project from the rice producers of Arkansas through the monies administered by the
Arkansas Rice Research and Promotion Board and the University of Arkansas System
Division of Agriculture. Technical support from Nickolas Murphy, Emily Carr, and
Jenifer Clemons is greatly appreciated.
LITERATURE CITED
Linscombe, S.D., F. Jodari, K.S. McKenzie, P.K. Bollich, D.E. Groth, L.M. White,
R.T. Dunand, and D.E. Sanders. 1993. Registration of Bengal rice. Crop Science
33:645-646.
Linscombe, S.D., F. Jodari, P.K. Bollich, D.E. Groth, L.M. White, R.T. Dunand, and
D.E. Sanders. 1997. Registration of Lafitte rice. Crop Science 37:631.
64
Sha, X., S.D. Linscombe, Q. Chu, D.E. Groth, L.M. White, J. Bond, R.T. Dunand,
and H. Utomo. 2006 Registration of Jupiter rice. Crop Science 46:1811-1812.
USDA-ERS. 2015. 2015 Rice Yearbook. Available online at: http://www.ers.usda.
gov/data-products/rice-yearbook-2015.aspx
65
Testing Uniform Regional Rice Nursery

Varieties for Resistance to Rice Blast Disease
C. Feng1 and J.C. Correll1
ABSTRACT
In 2015, 198 Uniform Regional Rice Nursery (URRN) lines were tested with 11 refer-
ence isolates representing 9 races of the rice blast pathogen. Isolate 24 (race IG 1) and
IB54 (race IB 54) were the least virulent with most rice lines being resistant to these
two isolates, while 49D (race IB49), TM2 (race k), and IB33 (race IB 33) were the most
virulent isolates. Nine lines were susceptible to all isolates, 17 lines were susceptible
to 10 isolates, 17 lines were susceptible to 9 isolates, and more than 20% of the lines
tested were very susceptible to the rice blast pathogen. However, 4 lines (RU1501030,
RU1303138, RU1501050, and RU1502115) were resistant to all 11 isolates tested, 16
lines were resistant to 10 isolates, 22 lines were resistant to 9 isolates, and more than
20% lines were very resistant to the rice blast pathogen. The most resistant and most
susceptible lines are listed. The results of this research will be useful to help breed-
ers make decisions on cultivar release and to choose parental lines for their breeding
programs to improve rice blast disease management.
INTRODUCTION
Rice is one of the most important staple food crops worldwide, feeding over half
of the worlds population. Although the United States is a relatively small rice producer
growing about 3 M acres of rice annually, producing about 10 M tons of rice (account-
ing for <2% of world total), it is the fifth largest rice exporter, which occupies about
10% of the world rice export market. The annual value of rice in the United States is
$3 billion. Rice blast disease, caused by the fungus Magnaporthe oryzae (anamorph:
Pyricularia oryzae), is one of the most destructive diseases of rice, threatening the rice
production in the U.S. The most economic and effective way to manage this disease is
growing resistant cultivars. Research revealed that multiple races exist in the Magna-
porthe oryzae population in the U.S., for example, race IB49 and IC-17 remain the most
prevalent in Arkansas (Correll et al., 2000; Xia et al., 2000), with occasional epidemics
due to race K type isolates (Lee et al., 2005). It is necessary to know the resistance
1 Program Associate II and Professor, respectively, Department of Plant Pathology, Fayetteville.
66
spectrum of new cultivars to the current rice blast pathogen population before they are
released. This study involved testing the Uniform Regional Rice Nursery (URRN) lines
with 11 U.S. reference isolates of Magnaporthe oryzae, which are representative of the
pathogen populations in Arkansas.
PROCEDURES
The 198 rice breeding lines developed by the rice breeders from Arkansas, Loui-
siana, Mississippi, and Texas were tested with 11 rice blast reference isolates (Table
1). The rice cultivar Francis was included in each test as the susceptible control. Rice
seed were planted in plastic trays filled with river sand mixed with potting soil in the
greenhouse at the University of Arkansas System Division of Agricultures Agricultural
Experiment Station, Fayetteville, Ark. Each tray was planted with 38 cells of URRN
entrees and 2 cells of the susceptible control Francis. Iron sulfate was applied to the
newly emerged seedlings. Then plants were fertilized with Miracle Gro All-Purpose
Plant Food 20-20-20 once a week during each test. Plants were inoculated approximately
14 to 20 days after planting. Each isolate was grown on rice bran agar (RBA) (Correll
et al., 2000) for approximately 7 to 10 days, then re-inoculated on new rice RBA for
7 to 10 days. Spores were collected in cool water, and adjusted to a concentration of
200,000 spores/ml per isolate. Each tray was inoculated with 50 ml of inoculum mixed
with 0.02% Tween 20 with an air compressor sprayer. After inoculation, the plants
were incubated at 100% relative humidity in a mist chamber at approximately 22 C
for 24 h, and allowed to dry for 2 to 3 h before being moved to the greenhouse. The
inoculated plants were incubated in the greenhouse for 6 days. On the seventh day after
inoculation, the plants were scored according to a standard 0 to 9 disease rating scale
(Correll et al., 1998). Lines rated 0 to 3 were considered resistant whereas those rated
4 to 9 were considered susceptible.

The 198 URRN lines were tested with 11 U.S. reference isolates of Magnaporthe
oryzae. The isolate IB33, originally recovered from rice under greenhouse conditions
by F.N. Lee, and isolates 49D (race IB49) and TM2 (race k) were the most virulent
isolates, with only 56, 24 and 44 lines (about 28%, 12% and 22%, respectively, of total)
resistant to these three isolates. Over 70% of the lines were resistant to isolates #24 (race
IG1), IB54 (race IB54), and ZN15 (race IB-1). Isolates A119 and A598 were classified
as race IB49. However, 60% of the lines were resistant to isolate A119 and 35% lines
resistant to A598. Again, the difference in virulence of the three IB49 (A119, A598,
and 49D) isolates suggested there are differences in their virulence characteristics. The
number of lines that were resistant or susceptible to each isolate was shown in Fig. 1.
Four lines (RU1501030, RU1303138, RU1501050, and RU1502115) were re-
sistant to all tested isolates and sixteen lines (RU1402174, RU1401136, RU1302192,
RU1402134, RU1401081, RU1303153, RU1402051, RU1502071, RU1203190, CL172,
RU1003123, RU1502152, RU1403153, RU1401161, RU1502171, and RU1504194)
67
were resistant to 10 isolates; the 16 lines were only susceptible to one isolate of IB33,
49D, or TM2. A total of 22 lines were resistant to 9 isolates, and 19 lines were resistant
to 8 isolates, so about 30% of the tested lines showed some resistance, which is similar
to that of 2014 results. Nine lines (RU1501001, FRNS, RU1501056, RU1401070,
RU1501076, RU1501093, RU1504114, RU1501182, and RU1504196) were susceptible
to all isolates; Seventeen lines (RU1301084, RU1501010, RU1404122, RU1404156,
RU1404157, CHNR, RU1401067, RU1404194, RU1501081, RU1501127, RU1501139,
RU1501142, RU1501145, RU1504157, RU1501176, RU1501188, and RU1505001)
were only resistant to one isolate. Seventeen and 25 lines were only resistant to two or
3 isolates, respectively and accounted for 35% of the total tested. The 20 most resistant
and 26 most susceptible lines were listed in Table 2. The number of lines that were
resistant to certain number of isolates was shown in Fig. 2. A complete examination of
the entry by isolate interactions is available online at http://www.uark.edu/ua/jcorrell/
data/2015URRNfinal.xls.
SIGNIFICANCE OF FINDING
The results from this study suggested that the URRN varieties had a wide range
in resistance to the rice blast pathogen, which may help breeders to make decisions on
the releasing of new cultivars and the choice of parental lines in their future breeding
programs. The screening efforts will ultimately help the growers to select rice cultivars
for the most effective disease management of rice blast disease.
ACKNOWLEDGMENT
The authors would like to thank the rice producers of Arkansas for their financial
support of the research though monies administered by the Arkansas Rice Research and
Promotion board and for the continued support of the University of Arkansas System
Division of Agriculture.
LITERATURE CITED
Correll, J.C., T.L. Harp, and F.N. Lee. 1998. Large-scale screening of USDA rice
germplasm for qualitative and quantitative resistance to two common races (IC-17
and IB-49) of the rice blast pathogen, Pyricularia grisea. In: R.J. Norman and
T.H. Johnson (eds.). Rice Research Studies 1998. University of Arkansas Agricul-
tural Experiment Station Research Series 468:169-194. Fayetteville.
Correll, J.C., T.L. Harp, J.C. Guerber, R.S. Zeigler, B. Liu, R.D. Cartwright, and F.N.
Lee. 2000. Characterization of a population of Pyricularia grisea from rice in the
United States using independent genetic and molecular markers. Phytopathology
90:1396-1404.
Lee, F.N., R.D. Cartwright, Y. Jia, and J.C. Correll. 2005. Magnaporthe grisea
race shift for virulence to the major R gene, Pi-ta in Arkansas. Phytopathology
95:S175.
68
Xia, J.Q., J.C. Correll, F.N. Lee, W.J. Ross, and D.D. Rhoads. 2000. Regional
population diversity of Pyricularia grisea in Arkansas and the influence of host
selection. Plant Disease 84:877-884.
Table 1 Background information for the 11 U.S.

reference isolates of Magnaporthe oryzae used in this study.a
Vegetative
compatibility group MGR586 Mating
Isolate (VCG) group type Race Year Origin
A119 US-03 C I IB49 1992 AR
A264 US-02 B II IC17 1993 AR
A598 US-01 A I IB49 1992 AR
#24 US-02 B II IG-1 1992 AR
IB33 US-04 I IB33 AR
IB54 US-04 I IB54
49D US-03 E II IB49 1985 AR
ZN7 US-02 B II IE-1 1995 TX
ZN15 US-01 A I IB-1 1996 TX
ZN46 US-01 A I IC-1 1996 FL
TM2 US-02 B II race K TX
a
The reference isolates belong to different genetic groups based on vegetative compatibility
(US-01-US-08) which also correspond to different molecular fingerprint groups (MGR586 A-H).
69
Table 2. Disease reactions of the most resistant (20) and susceptible (26) lines tested.
70
A119 A264 A598 24 IB33 IB54 49D ZN7 ZN15 ZN46 TM2
Entry Variety IB-49 IC-17 IB-49 IG-1 IB33 IB54 IB-49 IE-1 IB-1 IC-1 Race K
30 RU1501030 0 0 0 0 0 0 1 0 0 0 1
32 RU1303138 0 0 3 0 3 0 3 0 0 3 0
50 RU1501050 2 0 3 0 0 0 3 0 0 3 2
115 RU1502115 0 0 0 0 0 0 0 0 0 0 3
8 RU1402174 0 0 3 0 3 0 3 0 0 0 6
13 RU1401136 0 0 3 0 0 0 2 0 0 0 6
28 RU1302192 1 0 3 0 3 0 8 0 1 3 0
34 RU1402134 0 0 0 0 3 0 0 0 0 0 5
41 RU1401081 0 0 0 0 0 0 1 0 0 0 6
46 RU1303153 0 0 0 0 0 0 5 0 0 0 0
51 RU1402051 0 0 0 0 0 0 0 0 0 0 8
71 RU1502071 0 0 0 0 0 0 8 0 0 0 0
78 RU1203190 1 0 3 0 0 0 8 0 0 0 3
118 CL 172 0 0 0 0 3 0 3 0 0 0 5
123 RU1003123 1 3 3 0 3 0 6 3 0 0 3
152 RU1502152 0 0 0 0 4 0 0 0 3 0 3
153 RU1403153 1 3 3 0 1 0 6 3 0 1 3
161 RU1401161 0 0 3 0 0 0 3 0 0 0 5
171 RU1502171 0 0 3 0 0 0 3 0 0 0 6
194 RU1504194 0 0 3 0 3 0 0 0 0 0 6
4 RU1301084 4 6 6 4 4 6 8 5 1 5 7
10 RU1501010 4 7 6 5 5 7 6 5 1 6 6
36 RU1404122 6 5 6 4 6 6 8 5 3 6 6
54 RU1404156 4 4 7 3 4 6 6 4 4 5 6
55 RU1404157 5 4 7 0 4 6 6 4 4 6 6
58 CHNR 4 6 6 4 4 0 6 5 4 5 6
67 RU1401067 4 7 6 4 5 6 7 5 4 1 4
74 RU1404194 4 6 6 4 6 0 6 5 4 4 8
81 RU1501081 5 1 6 5 4 6 8 5 4 6 8
94 RU1502094 5 4 0 4 5 6 8 5 4 5 8
127 RU1501127 4 5 6 0 5 6 7 6 4 6 6
continued
Table 2. Continued.
A119 A264 A598 24 IB33 IB54 49D ZN7 ZN15 ZN46 TM2
Entry Variety IB-49 IC-17 IB-49 IG-1 IB33 IB54 IB-49 IE-1 IB-1 IC-1 Race K
139 RU1501139 5 5 8 4 5 6 6 5 4 2 6
142 RU1501142 6 6 6 0 6 6 8 6 4 6 6
145 RU1401145 3 6 7 4 5 6 6 6 4 5 6
157 RU1504157 4 5 6 4 4 7 8 6 0 6 8
176 RU1501176 6 6 6 5 4 6 8 6 0 5 8
188 RU1501188 4 6 8 5 4 1 8 4 4 5 6
1 RU1505001 5 6 6 5 5 6 8 6 4 6 7
40 FRNS 4 6 6 4 5 6 6 5 4 6 8
56 RU1505056 6 7 7 4 5 7 7 6 4 6 6
70 RU1401070 6 6 6 4 5 6 7 5 5 6 7
76 RU1501076 5 6 6 6 6 6 7 6 5 5 7
93 RU1501093 5 4 7 4 5 5 8 5 5 5 8
114 RU1504114 4 5 6 4 4 5 8 4 5 4 6
182 RU1501182 5 6 6 5 6 4 6 6 5 6 7
196 RU1504196 5 6 6 6 6 5 6 5 5 5 6
71
Fig. 1. Proportion of the number of rice lines that were

resistant (rating scale 0 to 3, as 0 is most resistant) and susceptible
(rating scales 4 to 9, as 9 is the most susceptible) to a given reference isolate.
Fig. 2. Distribution of the number of rice lines that were resistant

to 0 isolates, 1 isolate, 2 isolates, etc. For example, 9 rice lines were not
resistant to any isolates and 4 lines were resistant to all 11 reference isolates.
72
Diamond, a High Yielding,

Very Short Season, Long-Grain Rice Variety
K.A.K. Moldenhauer1, X. Sha1, J.T. Hardke2, R.J. Norman3,
Y.A. Wamishe4, M.M. Blocker1, D.L. McCarty1, C.H. Northcutt1,
D.G. North1, D.K.A. Wisdom1, V.A. Boyett1, D.L. Frizzell1, J.M. Bulloch1,
B.A. Beaty1, E. Castaneda-Gonzalez1. C.D. Kelsey1, and S.B. Belmar1
ABSTRACT
Diamond, a new short season, very high yielding, long-grain rice cultivar, was derived
from the cross Francis/Roy J. Diamond has been approved for release to qualified
seed growers for the summer of 2016. The major advantages of Diamond are its high
yield potential, long kernel length, low chalk and its early maturity. Diamond is a non-
semidwarf standard long-grain rice cultivar with lodging resistance approaching that
of Roy J. Diamond is very susceptible to false smut, susceptible to rice blast, sheath
blight, and kernel smut, and moderately susceptible to bacterial panicle blight.
INTRODUCTION
Diamond was developed in the rice improvement program at the University of
Arkansas System Division of Agricultures Rice Research and Extension Center (RREC),
Stuttgart, Ark., and has been released to qualified seed growers for the 2016 growing
season. Diamond has very high rough rice grain yield, good milling yield, and earli-
ness compared to Roy J. It is approximately 1 to 2 days later in maturity than LaKast
and 4 to 5 days earlier than Roy J. It is similar in height to Roy J and LaKast, and has
straw strength approaching that of Roy J. Diamond was developed with the use of rice
grower check-off funds distributed by the Arkansas Rice Research and Promotion Board.
1
Professor, Associate Professor, Program Associate III, Program Technician I, Program Technician I,
Program Technician I, Program Associate II, Program Associate II, Program Associate III, Program
Associate I, Program Associate I, Program Associate I, Program Technician II, and Program Technician
II, respectively, Rice Research and Extension Center, Stuttgart.
2
3
4
73
PROCEDURES
Diamond rice (Oryza sativa L.), is a very high yielding, short season, long-grain
rice cultivar developed by the Arkansas Agricultural Experiment Station. Diamond
originated from the cross Francis/Roy J (cross no. 20082221), made at the University
of Arkansas System Division of Agricultures Rice Research and Extension Center,
(RREC), Stuttgart, Ark., in 2008. Diamond is named for the diamond state of Arkansas
and because it is a diamond of a variety. Francis is a high yielding long-grain rice de-
scribed by Moldenhauer et al. (2007). Roy J (Moldenhauer et al., 2010) is a long-grain
lodging resistant high yielding rice. The experimental designation for early evaluation
of Diamond was STG10L-08-129, starting with a bulk of F5 seed from the 2010 panicle
row P-08-129. Diamond was tested in the Arkansas Rice Performance Trials (ARPT) and
the Cooperative Uniform Regional Rice Nursery (URRN) during 2013 to 2015 as entry
RU1301084 (RU number indicated Cooperative Uniform Regional Rice Nursery; 13
indicates year entered was 2013; 01 indicates Stuttgart, Ark.; and 084 its entry number).
In 2013, the ARPT was conducted at five locations in Arkansas: RREC; Northeast
Research Extension Center, (NEREC), Keiser Ark.; Pine Tree Research Station, (PTRS),
near Colt, Ark.; a Clay County producer field (CCPF) near Corning Ark.; and a Desha
County producer field (DCPF) near Dumas, Ark. In 2014 the tests were conducted at
the RREC, PTRS, CCPF, and DCPF; in 2015 the trials were grown at RREC, NEREC,
PTRS, CCPF, and DCPF. The tests had four replications per location to reduce soil
heterogeneity effects and to decrease the amount of experimental error. Diamond was
also grown in the URRN at the RREC; Crowley, Louisiana; Stoneville, Mississippi;
Beaumont, Texas, and at Malden, Missouri during 2013 to 2015. This test has three
replications per location. Data collected from these tests included plant height, maturity,
lodging, percent head rice, percent total rice, grain yield adjusted to 12% moisture, and
disease reaction information. Cultural practices varied somewhat among locations, but
overall the trials were grown under conditions of high productivity as recommended
by the University of Arkansas System Division of Agricultures Cooperative Extension
Service Rice Production Handbook MP192 (CES, 2013). Agronomic and milling data are
presented in Tables 1 and 2. Disease ratings, which are indications of potential damage
under conditions favorable for development of specific diseases, have been reported on
a scale from 0 = least susceptible to 9 = most susceptible, or as very susceptible (VS),
susceptible (S), moderately susceptible (MS), moderately resistant (MR), and resistant
(R). Straw strength is a relative estimate based on observations of lodging in field tests
using the scale from 0 = very strong straw to 9 = very weak straw, totally lodged.

Rough rice grain yields of Diamond have consistently ranked as one of the highest
in the ARPT. In 14 ARPT tests (2013-2015), Diamond, LaKast, Roy J, Taggart, Wells,
Mermentau, and RiceTec XL753 averaged yields of 210, 189, 195, 191, 183, 177, and
239 bu/acre, respectively (Table 1). Data from the URRN conducted at Arkansas during
2013-2015, showed that Diamond had an average grain yield of 246 bu/acre which com-
74
pared favorably with those of LaKast, Roy J, Taggart, Francis, Wells, and Mermentau,
at 244, 214, 221, 228, 222, and 219 bu/acre, respectively (Table 2). Milling yields (mg
g-1 whole kernel:mg g-1 total milled rice) at 120 mg/g moisture from the ARPT, 2013-
2015, averaged 610:690, 600:700, 620:700, 580:700, 590:700, 650:700, and 570:700,
for Diamond, LaKast, Roy J, Taggart, Wells, Mermentau, and RT XL753, respectively.
Milling yields for the URRN in Arkansas during the same period of time, 2013-2015,
averaged 650:710, 650:730, 630:710, 630:720, 660:720, 660:720, and 680:720, for
Diamond, LaKast, Roy J, Taggart, Francis, Wells, and Mermentau, respectively.
Diamond is a short to very short season variety close to the maturity of LaKast
and about 4 to 5 days earlier than Roy J. Diamond has straw strength approaching
that of Roy J which is an indicator of lodging resistance. On a relative straw strength
scale (0 = very strong straw, 9 = very weak straw) Diamond, LaKast Francis, Wells,
LaGrue, Cocodrie, and Roy J rated 2, 4, 4, 3, 5, 2, and 1, respectively. Diamond is 40
inches in plant height which is similar to Roy J and Wells.
Diamond, like Francis, and LaKast, is susceptible to common rice blast [Pyricu-
laria grisea (Cooke) Sacc.] races IB-1, IB-33, IB-49, IC-17, IE-1, and IE-1K with sum-
mary ratings in greenhouse tests of 6, 6, 6, 6, 5, and 7, respectively, using the standard
disease scale of 0 = immune, 9 = maximum disease susceptibility. Diamond is rated S
to sheath blight (Rhizoctonia solani Khn) which compares with Francis (MS), Wells
(S), Roy J (MS), and LaKast (S), using the standard disease ratings of R = resistant,
MR = moderately resistant, MS = moderately susceptible, S = susceptible, and VS =
very susceptible to disease. Diamond is rated S for kernel smut [Tilletia barclayana
(Bref.) Sacc. & Syd. in Sacc.], which compares to Francis (VS), Roy J (VS), Wells (S),
Mermentau (S), and Taggart(S). Diamond is rated MS to bacterial panicle blight caused
by Burkholderia species compared to Francis (VS) and Roy J (S) and VS to false smut
[Ustilaginoidea virens (Cooke) Takah].
Plants of Diamond have erect culms, green erect leaves, and glabrous lemma,
palea, and leaf blades. The lemma and palea are straw colored with purple apiculi, many
of which fade to straw at maturity. Milled kernels of Diamond are long at 7.15 mm
compared to LaKast, Roy J, Wells, Taggart, and Mermentau at 7.47, 7.24, 7.16, 7.40,
and 7.06 mm, respectively. Individual milled kernel weights of Diamond, LaKast, Roy
J, Taggart, Wells, and Mermentau averaged 21.6, 21.8, 21.2, 22.7, 21.7, and 19.8 mg/
kernel, respectively, in the ARPT 2013-2014, data from the Riceland Quality Laboratory.
The endosperm of Diamond is nonglutinous, nonaromatic, and covered by a light
brown pericarp. Rice quality parameters indicate that Diamond has typical southern
U.S. long-grain rice cooking quality characteristics as described by Webb et al. (1985).
Diamond has an average apparent starch amylose content of 22.8 g kg-1 and an inter-
mediate gelatinization temperature (70 to 75 C), as indicated by an average alkali (17
g kg-1 KOH) spreading reaction of 3 to 5.
The release of Diamond provides producers with a very-high yielding, short
season, long-grain rice replacement for Wells or Francis. It has the added benefit of
75
yield stability over time, yielding an average of 10 to 15 bushels better than any other
pure-line variety for the past three years.
ACKNOWLEDGMENTS
The authors wish express their appreciation for continued support and funding
for this project by the producers of Arkansas through the monies administered by the
Arkansas Rice Research and Promotion Board. Support also provided by the University
LITERATURE CITED
CES. 2013. Cooperative Extension Service. Arkansas Rice Production Handbook.
J.T. Hardke (ed). University of Arkansas System Division of Agriculture, MP 192,
Little Rock, Ark.
Moldenhauer, K.A.K., J.W. Gibbons, F.N. Lee, J.L. Bernhardt, C.E. Wilson, R.D.
Cartwright, M.M. Anders, R.J. Norman, N.A. Slaton, M.M. Blocker, A.C. Tol-
bert, K. Taylor, and J.M. Bulloch. 2007. Registration of Francis rice. Crop Sci.
47:443-444.
Moldenhauer, K.A.K., J.W. Gibbons, F.N. Lee, J.L. Bernhardt, C.E. Wilson Jr., R.D.
Cartwright, R.J. Norman, M.M. Blocker, D.K. Ahrent, A.M. Stivers, V.A. Boyett,
J.M Bulloch, and E. Castaneda. 2010. Roy J, high yielding stiff-strawed, long-
grain rice variety. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells
Rice Research Studies 2009. University of Arkansas Agricultural Experiment
Station Research Series 581:53-59.
Webb, B.D., C.N. Bollich, H.L. Carnahan, K.A. Kuenzel., and K.S. McKenize. 1985.
Utilization characteristics and qualities of United States rice. pp. 25-35. In: Rice
grain quality and marketing. IRRI, Manila, Philippines.
76
Table 1. Three-year average for yield and three-year average for the agronomic data
from the 2013 to 2015 Arkansas Rice Performance Trials for Diamond and other cultivars.
Yieldb
Cultivar Grain 50% Chalky
Cultivar typea 2013 2014 2015 Mean Heightc headingd kernelse Millingf
------------------------------(bu/acre)--------------------------- (in.) (days) (HR:TOT)
Diamond L 226 218 186 210 40 81 1.40 61:69
LaKast L 203 202 162 189 41 80 1.04 60:70
Roy J L 210 207 169 195 41 85 1.13 62:70
Taggart L 205 200 167 191 43 84 1.02 60:70
Wells L 196 192 161 183 40 82 1.37 59:70
Mermentau L 190 181 161 177 37 80 1.88 65:70
RT XL753 L 245 259 212 239 44 78 2.13 57:70
a
Grain type L = long-grain.
b
Yield trials in 2013 consisted of five locations, Rice Research and Extension Center, (RREC), Stuttgart Ark.; Pine Tree Research Station, (PTRS),
near Colt, Ark.; Northeast Research and Extension Center, (NEREC), Keiser, Ark.; Clay County Farmer Field, (CCPF), Corning, Ark.; and Desha
County Producer Field (DCPF), in 2014 the successful trials were grown at RREC, PTRS, CCPF, and DCPF; in 2015 the trials were at RREC,
PTRS, NEREC, CCPF, and DCPF.
c
Height data is from 2013-2015.
d
Heading information from 2013-2015.

e
Data for chalk is from 2013-2014 Riceland Grain Quality Laboratory data.
f
Milling figures are head rice : total milled rice 2013-2015.
77
Table 2. Data from the 2013 to 2015 Uniform Regional Rice Nursery for Diamond and other check cultivars.
78
Yielda Arkansas yield 50%
Cultivar 2013 2014 2015 Mean 2013 2014 2015 Mean Heightb headingc Millingd
------------------------------------------------ (bu/acre)------------------------------------------------ (in.) (days) (HR:TOT)
Diamond 219 235 209 221 244 261 234 246 43 88 65:71
LaKast 218 243 202 221 241 258 232 244 44 86 65:72
Wells 199 213 183 198 227 251 189 222 44 90 66:72
Francis 195 229 186 203 217 259 208 228 43 89 66:72
Taggart 211 222 211 217 231 235 197 221 48 93 63:72
Roy J 215 229 184 209 240 221 180 214 45 95 63:71
Mermentau 201 208 194 201 235 221 202 219 40 89 68:72
a
Arkansas = Rice Research and Extension Center, Stuttgart, Ark.
b
Height data from Arkansas 2013-2015 only.
c
Heading data from Arkansas 2013-2015 only.
d
Milling figures are %Head Rice : %Total Milled Rice.
Genetic Basis of Altered Grain Quality in Different

Rice Cultivars Under High Nighttime Temperature
V. Ramegowda1, S. Srivastava1, J. Thomas1, C. Gupta1, S. Basu1, P. Counce2,
Y.-J. Wang3, T.J. Siebenmorgen3, K.A.K. Moldenhauer2, and A. Pereira1
ABSTRACT
This study was focused on understanding the genetic basis of grain quality, and testing
the chalkiness of diverse rice cultivars treated with high nighttime temperature (HNT).
Our results confirmed previous field and controlled-climate experiments on the effect
of HNT on the chalkiness of Bengal, M204, Cypress, LaGrue, and Wells, with Bengal
and Cypress being least affected. In addition, in our study the higher yield rice (HYR)-
Nipponbare transgenic line, Roy J, and Cheniere had low chalk, whereas Nipponbare
and CL151 had higher chalk. Gene expression and genetic variation analysis provided
valuable information on the potential role of several starch biosynthetic pathway genes
and the functional polymorphisms in regulating grain quality in the cultivars studied.
INTRODUCTION
High nighttime temperature (HNT) during grain filling is one of the major
causes of rice chalkiness in both field and controlled climate experiments (Cooper et
al., 2006; Counce et al., 2005; Lanning et al., 2011; Peng et al., 2004). Chalkiness is
a critical attribute of rice quality and chalk percentage is related to reduced head rice
yield. Especially in premium rice markets, chalk incidence reduces buyer acceptance,
appearance, and overall evaluations of rice quality. It has been suggested that the opaque
appearance of chalky grain is due to loosely packed amyloplasts and starch granules
(Singh et al., 2003). Existence of greater genetic variation for grain quality under HNT
has been observed in rice cultivars (Cooper et al., 2008). Cooper et al. (2008) reported
that rice cultivars subjected to HNT during grain filling under a controlled-environment
showed different degrees of chalkiness. Gene expression analysis of the developing
caryopses treated to high day/night temperatures revealed a decrease in transcript lev-
els of sucrose and starch synthesis genes and an increase in transcript levels of starch
1
Post Doctoral Associate, Former Post Doctoral Associate, National Science Foundation Post Doctoral
Fellow, Senior Graduate Assistant, Post Doctoral Associate, and Professor, respectively, Department of
Crop, Soil, and Environmental Science, Fayetteville.
2
Professor and Professor, respectively, Rice Research and Extension Center, Stuttgart.
3
Professor and Distinguished Professor, respectively, Department of Food Science, Fayetteville.
79
degradation genes (Yamakawa and Hakata, 2010). However, the genetic basis of al-
tered grain quality under HNT has not been explored so far. Therefore, this study was
focused on understanding the genetic control of altered grain quality in selected U.S.
rice cultivars. Here we describe the screening of an extended panel of rice cultivars for
altered grain quality parameters under HNT, using controlled growth chamber condi-
tions, and identify functional gene markers for rice grain quality by gene expression
and genetic variation analysis.
PROCEDURES
Plant Growth Conditions and Temperature Treatment
For grain quality measurements, plants at the R2 stage were treated to HNT of
28 C while controls were maintained at 22 C with constant day temperature of 30 C.
At physiological maturity, seeds were harvested and air dried. For gene expression stud-
ies, beginning at the R5 stage, rice plants were subjected to HNT; and the caryopses
at R6 stage (soft to hard dough stage) were harvested as soon as the lights came on in
the morning and frozen in liquid nitrogen. Grains were collected at grain maturity for
chalk measurements.
Chalk Measurement
Rough rice was de-hulled using a manually operated de-huller (Rice Husker
TR120, Kett Electric Laboratory, Tokyo, Japan). Chalkiness was measured using an
image analysis system WinSeedle Pro 2005a (Regent Instruments, Quebec, Canada)
and expressed as percent of grain projected area. Data are the means of two biological
replicates with each replicate measured twice using 100 grain. A significant difference
between treatments within the cultivar was determined by pairwise comparisons of
means using Students t-test.
Transcriptome Analysis
The total caryopsis RNA was isolated using TRIZol reagent at the milky dough
stage. Sequencing of RNA was carried out using Illumina High-Seq 2000 (Illumina,
San Diego, Calif.) platform using two biological replicates per sample (Michigan State
University Genomics core facility). Differentially expressed genes were identified as
described in Trapnell et al. (2010). For genome sequencing, genomic DNA was isolated
from Bengal, M204, Cypress, and LaGrue using the DNeasy Plant Mini Kit (Qiagen,
Inc, Valencia, Calif.). The genome sequencing was carried out using Illumina High-Seq
2000 platform and single nucleotide polymorphisms (SNPs) were predicted as described
in Srivastava et al. (2014).

High nighttime temperature effect on the chalkiness of 11 cultivars/genotypes
[Bengal, Cypress, LaGrue, M204, Nipponbare, HYR-Nipponbare a transgenic line
80
(Ambavaram et al., 2014), Cheniere, CL151, Roy J, Starbonnet, and Wells] was studied.
High nighttime temperatures increased chalkiness, and reduced grain width and length
compared to the control in all the cultivars (Fig. 1). The chalkiness of short- and medium-
grain cultivars Nipponbare and M204 were higher than in the HYR-Nipponbare line and
Bengal, respectively. Overall, Bengal, Cypress, Roy J, and Cheniere showed the lowest
chalk; whereas LaGrue showed the highest chalk among the cultivars studied. These
results corroborate previous findings that Bengal and Cypress showed least chalkiness
under HNT among the medium- and large-grain cultivars, respectively, when tested
under controlled environmental conditions as well as under field conditions (Cooper
et al., 2008; Ambardekar et al., 2011). In addition to chalkiness, HNT also reduced the
grain length of Bengal, M204, and Wells, and grain width of HYR-Nipponbare, Nip-
ponbare, Bengal, M204, Cypress, Roy J, and CL151 (Fig. 1B and C). Decrease in grain
width of Bengal, M204, and Cypress was also observed in previous studies, both under
controlled environment and field conditions (Counce et al., 2005; Cooper et al., 2008).
Gene expression analysis by RNA sequencing in the R6 stage caryopses of all
cultivars subjected to HNT showed differential expression of several starch biosynthetic
pathway genes among the low and higher chalk cultivars. For example, the expression
of GBSSI gene, the primary determinant of amylose content, in good quality cultivars
such as HYR-Nipponbare, Bengal, Cypress, and Roy J was higher compared to the
cultivars Nipponbare, M204, and LaGrue (Fig. 2). Similarly, AGPL2, another key gene
in starch biosynthesis also showed higher expression in Bengal and Cypress compared
to M204 and LaGrue (Fig. 2). These results suggest that the differential expression of
starch biosynthetic pathway genes under HNT determine grain quality. Cultivars having
better expression of one or more of these genes may have improved quality over those
with lower gene expression.
Further, genome sequence analysis of the four cultivars (Bengal, M204, Cypress,
and LaGrue) was analyzed for SNPs in starch biosynthetic pathway genes including
their promoters. The analysis identified several SNPs which are present in low chalk
cultivars but absent in cultivars with higher chalk (Table 1). For example, SSIIIa and
BEI genes from Bengal have multiple SNPs which are also present in Cypress but
absent in M204 and LaGrue. Similarly, BEIIb genes from Bengal and Cypress have
many SNPs, which are absent in M204 and LaGrue. Interestingly, several functional
nonsynonymous SNPs, potentially changing the protein, were found in genes of the
low chalk cultivars which were absent in genes of cultivars with high levels of chalk
(Table 1). These variations in SNPs might be contributing to the differential expression
of starch genes and their activity, resulting in altered amylose content and amylopectin
structure under HNT.
Our results comparing 11 genotypes under similar controlled conditions confirmed
the previous findings (Counce et al., 2005; Cooper et al., 2008) that under HNT, Ben-
gal and Cypress showed reduced chalkiness compared to M204, LaGrue, and Wells.
81
In addition, Roy J and Cheniere had lower chalk than CL151. Our results suggest that
controlled condition screens for HNT response are very valuable in identifying chalkiness
parameters before testing the cultivars under field conditions which can be unpredict-
able. Gene expression and genetic variation analysis provided useful indicators of the
genetic basis for differential grain chalk formation between the cultivars. However, the
results do not sufficiently explain the varying chalky phenotypes observed among differ-
ent cultivars under HNT. Therefore, further confirmation of functional isoforms of the
genes and validation of the SNPs can identify the real cause of differential chalkiness
in the tested cultivars that can be used in the breeding program.
ACKNOWLEDGMENTS
The authors thank the rice producers of Arkansas for their funding administered
by the Arkansas Rice Research and Promotion Board and for the continued support
from the University of Arkansas System Division of Agriculture
LITERATURE CITED
Ambavaram, M.M., S. Basu, A. Krishnan, V. Ramegowda, U. Batlang, L. Rahman,
N. Baisakh, and A. Pereira, 2014. Coordinated regulation of photosynthesis in rice
increases yield and tolerance to environmental stress. Nat. Commun. 5.
Ambardekar, A.A., T.J. Siebenmorgen, P.A. Counce, S.B. Lanning, and A. Mauro-
moustakos. 2011. Impact of field-scale nighttime air temperatures during kernel
development on rice milling quality. Field Crops Res. 122:179-185.
Cooper, N.T.W., T.W. Siebenmorgen, and P.A. Counce. 2008. Effects of nighttime
temperature during kernel development on rice physicochemical properties. Ce-
real Chem. 85:276-282.
Cooper, N.T.W., T.J. Siebenmorgen, P.A. Counce, and J.-F. Meullenet. 2006. Ex-
plaining rice milling quality variation using historical weather data analysis.
Cereal Chem. 83:447-450.
Counce, P.A., R.J. Bryant, C.J. Bergman, R.C. Bautista, Y.J. Wang, T.J. Siebenmor-
gen, K.A.K. Moldenhauer, and J.-F. Meullenet. 2005. Rice milling quality, grain
dimensions, and starch branching as affected by high night temperatures. Cereal
Chem. 82:645-648.
Lanning, S.B., T.J. Siebenmorgen, P.A. Counce, A.A. Ambardekar, and A. Mauro-
moustakos. 2011. Extreme nighttime air temperatures in 2010 impact rice chalki-
ness and milling quality. Field Crops Res. 124:132-136.
Peng, S., J. Huang, J.E. Sheehy, R.C. Laza, R.M. Visperas, X. Zhong, G.S. Centeno,
G.S. Khush, and K.G. Cassman. 2004. Rice yields decline with higher night tem-
perature from global warming. Proceedings of the National Academy of Sciences
of the United States of America 101:9971-9975.
82
Singh, N., N.S. Sodhi, M. Kaur, and S.K. Saxena. 2003. Physico-chemical, morpho-
logical, thermal, cooking and textural properties of chalky and translucent rice
kernels. Food Chem. 82:433-439.
Srivastava, S.K., P. Wolinski, and A. Pereira. 2014. A Strategy for genome-wide
identification of gene based polymorphisms in rice reveals non-synonymous
variation and functional genotypic markers. PLoS ONE 9, e105335.
Trapnell, C., B.A. Williams, G. Pertea, A. Mortazavi, G. Kwan, M.J. van Baren, S.L.
Salzberg, B.J. Wold, and L. Pachter. 2010. Transcript assembly and quantification
by RNA-Seq reveals unannotated transcripts and isoform switching during cell
differentiation. Nat. Biotechnol. 28:511-515.
Yamakawa, H. and M. Hakata. 2010. Atlas of rice grain filling-related metabolism
under high temperature: Joint analysis of metabolome and transcriptome demon-
strated inhibition of starch accumulation and induction of amino acid accumula-
tion. Plant Cell Physiol. 51:795-809.
Table 1. Single nucleotide polymorphisms (SNPs) in starch

biosynthesis genes of diverse rice cultivars differing in grain quality.
Gene name Bengal M204 Cypress LaGrue
OsAGPL1 0 0 0 0
OsAGPL2 2 0 2 0
OsAGPS1 0 0 0 0
OsAGPS2 2 0 27 4
OsGBSSI 0 0 2 (1 non-syn) 0
OsSSI 0 0 0 0
OsSSIIa 0 0 1 (non-syn) 0
OsSSIIIa 3 2 38 (7 non-syn) 8 (1 non-syn)
OsBEI 7 0 14 (1 non-syn) 1
OsBEIIb 25 (2 non-syn) 1 35 (2 non-syn) 5
OsISA1 0 0 0 0
83
Fig. 1. Effect of high nighttime temperatures (HNT) on chalkiness,

length, and width of brown grain of diverse rice cultivars. Plants at R2 stage
were treated to HNT of 28 C until maturity with controls maintained at 22 C. The daytime
temperature was kept constant at 30 C. At physiological maturity, seeds were harvested,
air dried, and de-hulled using a manually operated de-huller (Rice Husker TR120). (A)
Chalkiness, (B) grain length and (C) grain width was measured using an image analysis
system (WinSeedle Pro 2005a) and expressed as percent of grain projected area. Data
are the means of two biological replicates with each replicate measured twice using 100
grain. An * indicates significant difference at P < 0.05, Students t-test.
84
Fig. 2. Expression of starch biosynthetic pathway

genes in different rice cultivars treated to high nighttime
temperatures (HNT). Caryopses at R6 stage treated to HNT
were used to isolate total RNA and mRNA was sequenced using Illumina
High-Seq 2000 platform (Michigan State University Genomics core facility).
The reads were mapped to the rice reference genome sequence (MSU 7.0)
with Tophat 1.3.1 and mapped reads were assembled into transcripts by
Cufflinks. The differentially expressed genes were identified by using Cuffdiff.
The expression values of GBSSI and AGPL2 under HNT in rice cultivars is given.
85
Development of Superior Medium-Grain and

Long-Grain Rice Varieties for Arkansas and the Mid-South
X. Sha1, K.A.K. Moldenhauer1, B.A. Beaty1,
J.M. Bulloch1, D.K.A. Wisdom1, M.M. Blocker1, D.L. McCarty1,
V.A. Boyett1, D.L. Frizzell1, J.T. Hardke2, and C.E. Wilson Jr.1
ABSTRACT
To reflect the recent changes of the state rice industry and streamline the delivery of
new and improved rice varieties to the Arkansas rice growers, the new medium-grain
rice breeding project will expand its research areas and breeding populations to include
both conventional and Clearfield medium-grain and semi-dwarf long-grain rice, as
well as hybrid rice. Newest elite breeding lines/varieties from collaborating programs,
as well as lines with diverse genetic origins will be actively collected, evaluated, and
incorporated into the current crossing blocks for the programmed hybridization. To
improve the efficiency and effectiveness, maximum mechanized-operation, multiple
generations of winter nursery, and new technologies such as molecular marker-assisted
selection (MAS) will also be rigorously pursued.
INTRODUCTION
Medium-grain rice is the important component of Arkansas rice. Arkansas ranks
second in medium-grain rice production in the United States only behind California.
During 2005-2014, an average of 0.15 million acres of medium-grain rice was grown
annually, which makes up about 10% of total state rice acreage (USDA-ERS, 2015).
Planted acres of medium-grain rice in Arkansas in the last decade have varied from a
high of 243,000 acres in 2011 (21% of total rice planted in Ark.) to a low of 99,000
acres in 2008 (7% of total rice planted in Ark.).
A significant portion of Arkansas rice area was planted to semi-dwarf long-
grain varieties, such as CL111, CL151, and Mermentau. However, locally developed
semi-dwarf varieties offer advantages including better stress tolerance and more stable
yields. Improved semi-dwarf long-grain lines also can be directly adopted by the newly
1
Associate Professor, Professor, Program Associate I, Program Associate I, Program Associate II,
Program Associate III, Program Technician I, Program Associate I, Program Associate III, and Center
Director, respectively, Rice Research and Extension Center, Stuttgart.
2
86
established hybrid breeding program. Since genetic potential still exists for further
improvement of current varieties, rice breeding efforts should and must continue.
The inter-subspecies hybrids between indica male sterile lines and tropical ja-
ponica restorer/pollinator lines that were first commercialized in the United States in
1999 by RiceTec have a great yield advantage over conventional pure-line varieties
(Walton, 2003). However the further expansion of hybrid rice may be constrained by
its inconsistent milling yield, poor grain quality, lodging susceptibility, seed shattering,
and high seed cost. A public hybrid-rice research program that focuses on the develop-
ment of adapted lines (male sterile, maintainer, and restorer lines) will be instrumental
to overcome such constraints.
PROCEDURES
Potential parents for the breeding program are evaluated for the desired traits.
Cross combinations are programmed that combine desired characteristics to fulfill the
breeding objectives. Marker assisted selection will be carried out on backcross or top-
cross progenies on simply inherited traits such as blast resistance and physicochemi-
cal characteristics. Segregating populations are planted, selected, and advanced at the
University of Arkansas System Division of Agricultures Rice Research and Extension
Center (RREC), Stuttgart, Ark., and the winter nursery in Lajas, Puerto Rico. Pedigree
and modified single seed descent will be the primary selection technology employed.
A great number of traits will be considered during this stage of selection including
grain quality (shape and appearance), plant type, short stature, lodging resistance, dis-
ease (blast, sheath blight, and panicle blight) resistance, earliness, and seedling vigor.
Promising lines having a good combination of these characteristics will be further
screened in the laboratory for traits such as kernel size and shape, grain chalkiness,
and grain uniformity. Milling small samples, as well as the physicochemical analysis
at the USDA Rice Quality Laboratory at the Dale Bumpers National Rice Research
Center near Stuttgart, Ark., and at Riceland Foods, Inc. Research and Technology Cen-
ter, Stuttgart, Ark., will be conducted to eliminate lines with evident quality problems
and/or maintain standard U.S. rice quality of different grain types. Yield evaluations
include the Stuttgart Initial Yield Trial (SIT) and Clearfield SIT (CSIT) at the RREC;
the Advanced Yield Trial (AYT) at the RREC, the Pine Tree Research Station (PTRS),
near Colt, Ark. and the Northeast Research and Extension Center (NEREC), Keiser,
Ark.; the Arkansas Rice Performance Trials (ARPT) conducted by Jarrod Hardke, the
rice extension specialist, at six locations in rice-growing regions across the state; and
the Uniform Regional Rice Nursery (URRN) conducted in cooperation with public rice
breeding programs in California, Louisiana, Mississippi, Missouri, and Texas. Promis-
ing advanced lines will be provided to cooperating projects for the further evaluation
of resistance to sheath blight, blast, and panicle blight, grain and cooking/processing
quality, and nitrogen fertilizer requirements. All lines entered in the SIT or CSIT and
beyond will be planted as headrows for purification and increase purposes.
87

A great number of breeding populations have been created and rapidly advanced
since 2013 when the senior author was hired. The field research in 2015 included 427
transplanted F1 populations, 625 space-planted F2 populations, and 53,450 panicle rows
ranging from F3 to F6. Visual selection on approximately 625,000 individual space-
planted F2 plants resulted in a total of 31,000 panicles, which will be grown as F3 panicle
rows in 2016. From 53,450 panicle rows, 3981 were selected for advancement to the
next generation, while 1753 rows which appeared to be uniform and superior were
bulk-harvested as candidates for the 2016 SIT or CSIT trials. In the 2015 Clearfield
(CL) preliminary yield trial (CSIT), we evaluated 557 new breeding lines which in-
cluded 462 semidwarf CL long-grain and 95 CL medium-grain lines. In the SIT trial,
307 new semidwarf breeding lines were tested, which consist of 171 long-grain and
136 medium-grain lines. A new 60-entry Advanced Yield Trial (AYT) was initiated
and conducted at PTRS and NEREC in addition to RREC. A number of breeding lines
showed yield potential similar to or better than the check varieties (Tables 1-4). Twenty
six advanced breeding lines were evaluated in the ARPT and/or multi-state URRN tri-
als. Results of those entries and selected check varieties were listed in Table 5. Three
Puerto Rico winter nurseries of 10,500 rows were planted, selected, harvested, and/or
advanced throughout 2015. A total of 592 new crosses were made to incorporate desir-
able traits from multiple sources into adapted Arkansas rice genotypes, which included
211 CL long-grain, 59 CL medium-grain, 101 semidwarf conventional long-grain, 81
conventional medium-grain, and 16 hybrid line crosses, as well as 96 hybrid test crosses
and 28 hybrid backcrosses.
The conventional medium-grain line RU1301021 continued showing excellent
yield potential, good milling, and superior grain quality in trials across Arkansas and
the mid-South in 2015. It was approved for the formal release as Titan by the Univer-
sity of Arkansas System Division of Agriculture in February 2016. Foundation seed
was produced, and is available to seed rice growers for the 2016 growing season. The
semi-dwarf CL long-grain line 15AR1024 (RU1501024) and CL medium-grain line
15AR1111 (RU1501111) were selected for purification and increase in Lajas Puerto
Rico in winter 2015 for their superior yielding potential and excellent milling and grain
quality. One hundred seventy-five breeding lines that outperformed commercial check
varieties in AYT, CSIT, and SIT trials were selected and were further evaluated in the
laboratory before entering 2016 ARPT and/or URRN trials.
Successful development of medium-grain variety Titan offers producers op-
tions in their choice of variety and management systems for Arkansas rice production.
Continued utilization of new germplasm through exchange and introduction remains
important for Arkansas rice improvement.
88
ACKNOWLEDGMENTS
This research is partially funded by grower monies administered by the Arkansas
Rice Research and Promotion Board. Support also provided by the University of Ar-
kansas System Division of Agriculture. We thank Nickolas Murphy, Emily Carr, and
Jenifer Clemons for their technical support.
LITERATURE CITED
USDA-ERS. 2015. United States Department of Agriculture Economic Research
Service. 2015 Rice Yearbook. Available at: http://www.ers.usda.gov/data-prod-
ucts/rice-yearbook-2015.aspx
Walton, M. 2003. Hybrid rice for mechanized agriculture. pp. 97-102. In: S.S.
Virmani, C.X. Mao, and B. Hardy (eds.). Hybrid Rice for Food Security, Poverty
Alleviation, and Environmental Protection. International Rice Research Institute,
Los Baos, Philippines.
89
Table 1. Performance of selected Clearfield long-grain experimental lines and check varieties in the
90
Clearfield Stuttgart Initial Trial (CSIT) at the Rice Research and Extension Center near Stuttgart, Ark., 2015.
Milling yields
Seedling Days to 50% Plant
Variety/Line Pedigree vigora heading height Yield Head rice Total rice
(days) (cm) (bu/acre) ------------- (%)-----------
15CSIT409b RU1102192/4/WLLS/CFX-18/3/CFX-18//CCDR/ 3.0 78 116 234 68.8 70.9
15CSIT324b RU1102034/4/CCDR/JEFF/3/CFX-18//CCDR/ 5.0 78 118 222 69.2 71.7
15CSIT397b CL172/RU1102192 4.5 77 106 220 68.6 71.1
15CSIT343b RU1102034/RU1202082 5.5 81 119 220 64.7 70.5
15CSIT568c RU0902125/CL151 4.0 78 113 228 61.2 68.2
15CSIT532c CCDR//CCDR/JEFF/3/CL131/4/RU1302045 4.5 79 120 228 68.0 71.2
15CSIT550c STG10IMI-05-034//RU0902155/RU0902131 4.5 79 105 223 68.8 72.8
15CSIT515c RU1102034/RU1302045 4.0 79 113 222 69.2 71.6
15CSIT754d RU1302048/RU1302045 4.5 67 113 224 67.9 70.8
15CSIT708d RU1102028/CL111 3.0 67 111 222 68.2 71.1
15CSIT742d RU1302045/MRMT 4.0 69 114 222 67.4 69.6
CL151b CL151 3.5 78 106 185 66.1 70.9
CL151c CL151 4.0 78 113 212 66.8 70.2
CL151d CL151 3.0 68 106 197 69.8 72.5
CL172c CL172 5.5 80 105 213 66.3 70.0
a
A subjective 1 to 7 rating taken at emergence, 1 = perfect stand and 7 = no stand.
b
Planted on 22 April.
c
Planted on 9 May.
d
Planted on 23 May.
Table 2. Performance of selected Clearfield medium-grain experimental lines and check varieties in the
Clearfield Stuttgart Initial Trial (CSIT) at the Rice Research and Extension Center near Stuttgart, Ark., 2015.
Milling yields
(days) (cm) (bu/acre) ------------- (%)-----------
b
15CSIT465 RU1202065/JPTR 4.5 77 105 232 66.6 68.6
15CSIT475b RU1202168/JPTR 4.0 78 101 220 62.0 68.6
15CSIT429b RU1202168//9865216DH2/EARL 4.0 80 99 205 59.4 67.1
15CSIT463b JPTR/RU1202068 5.0 81 113 203 67.3 69.0
15CSIT472b RU1202168//RU0602162/RU0502031 4.0 82 106 202 69.1 71.1
15CSIT479b STG09PR-82-037/RU1202068 3.5 78 95 199 60.2 68.0
15CSIT757c RU1301124/CL261 5.5 68 117 222 64.3 67.7
15CSIT782c BNGL/CL161/4/9502065/3/MERC//MERC/ 5.0 72 113 220 64.5 67.0
15CSIT779c STG10PR-05-022/RU1201087//STG10PR-08-077/ 3.5 63 106 204 62.5 66.7
M207
15CSIT765c RU1202068/RU1301021 4.0 68 122 202 65.4 66.9
15CSIT758c RU1301021/RU1202168 5.0 73 115 200 63.4 67.4
CL172b CL172 4.0 79 102 187 66.3 70.0
CL271b CL271 4.0 79 97 179 64.8 67.1
CL151b CL151 3.0 68 106 197 67.6 70.9

CL271c CL271 3.0 79 101 183 63.4 71.2
a
A subjective 1-7 rating taken at emergence, 1 = perfect stand and 7 = no stand.
b
Planted on 22 April.
c
Planted on 23 May.
91
Table 3. Performance of selected conventional medium-grain experimental lines and check
92
varieties in the Stuttgart Initial Trial (SIT) at the Rice Research and Extension Center near Stuttgart, Ark., 2015.
Milling yields
(days) (cm) (bu/acre) ------------- (%)-----------
15SIT794b JPTR/RU1001099 5.0 97 106 238 65.4 66.1
15SIT818b RU1301121/JPTR 4.0 90 104 216 66.4 67.1
15SIT792b JPTR/RU1001099 5.5 98 100 213 65.1 65.8
15SIT828b BNGL/SRICO/4/9502065/3/BNGL//MERC/RICO 5.0 96 98 210 66.8 67.7
15SIT821b RU1301133/JPTR 4.0 93 101 209 65.5 66.6
15SIT802b EARL/4/ORIN//MERC/RICO/3/9602134 4.5 93 101 208 62.9 63.9
15SIT849b 9902028/3/BNGL//MERC/RICO/4/JPTR 6.0 95 95 208 66.9 68.0
15SIT786b JPTR/J062 5.0 97 97 208 66.0 66.7
15SIT898c EARL/9902028//JPTR 4.0 79 101 258 57.7 65.7
15SIT916c JPTR/RU1301021 5.0 76 101 244 60.8 67.8
15SIT886c EARL/9902028//RICO/BNGL 5.0 78 102 239 65.5 67.4
15SIT899c EARL/9902028//JPTR 4.5 78 95 232 61.0 66.1
15SIT917c JPTR/RU1301021 5.0 76 96 230 63.6 68.0
Jupiterc Jupiter 4.0 80 106 232 65.8 67.1
Titanb Titan 4.5 90 101 216 62.5 67.2
a
b
Planted on 8 April.
c
Planted on 9 May.
Table 4. Performance of selected conventional long-grain experimental lines and check
varieties in the Stuttgart Initial Trial (SIT) at the Rice Research and Extension Center near Stuttgart, Ark., 2015.
Milling yields
(days) (cm) (bu/acre) ------------- (%)-----------
b
15SIT710 MRMT/RU0502068 4.5 99 104 246 68.3 70.1
15SIT666b CTHL/RU1002192 5.0 98 105 235 69.4 71.6
15SIT758b RU1102034/RU1002128 5.0 98 112 233 70.2 72.1
15SIT636b CCDR/CPRS 4.5 99 107 232 65.7 67.9
15SIT726b CHNR/CTHL 5.5 99 109 228 68.1 70.4
15SIT757b RU1102034/RU1002128 4.0 99 111 228 68.5 70.6
15SIT703b CCDR/CHNR 5.0 97 114 225 68.0 70.3
15SIT633b RU0502068/RU1002128 5.0 101 110 224 68.3 69.9
15SIT683b RU0802134/RU0902125 6.0 99 111 223 68.7 70.7
15SIT669b RU0902034/RU0502068 5.5 100 105 220 68.0 70.4
15SIT883c RU1102192/RU1102034 4.0 79 108 225 60.5 68.8
CL151b CL151 5.0 95 109 205 67.6 70.9
LaKastb LaKast 5.0 95 117 245 63.1 69.0
Mermentaub Mermentau 5.0 99 104 201 64.0 68.3
Roy Jb Roy J 5.0 102 111 189 61.6 69.3

a
b
Planted on 8 April.
c
Planted on 9 May.
93
Table 5. Average yield, milling, and agronomic characteristics of selected experimental long-grain and medium-grain lines and
94
check varieties tested in the Uniform Regional Rice Nursery (URRN) in Arkansas, Louisiana, Mississippi, Missouri, and Texas, 2015.
Milling yields
Grain Days to 50% Plant
Variety/Line Pedigree typea heading height Yield Head rice Total rice
(days) (cm) (bu/acre) ------------- (%)-----------
RU1501024 CL111/3/CCDR//9502008/LGRU CL 88 99 186 63.7 71.6
RU1501027 NPTN//BNGL/CL161 CM 89 100 200 65.4 69.8
RU1501030 CPRS/KBNT//9502008-A L 86 100 186 63.7 71.7
RU1501061 CL111//CCDR/0502085 CL 86 97 181 62.1 70.3
RU1501111 BNGL/CL161/4/9502065/3/MERC//MERC/ CM 87 101 192 62.4 71.2
RU1501121 CL111/CCDR CL 85 102 185 62.9 70.2
RU1501124 KATY/CPRS//JKSN/3/AR1188/CCDR/4/CFX-29/ CL 85 106 200 64.4 71.3
CCDR
RU1501127 CCDR/RU0801167 L 88 104 187 65.8 71.7
RU1501130 CPRS/KBNT//CFX29/CCDR/3/06CFP952 CL 88 103 187 64.5 70.5
Jupiter Jupiter M 89 93 214 65.9 69.5
Titan Titan M 84 98 204 62.5 69.4
CL151 CL151 CL 86 96 209 63.0 71.2
CL172 CL172 CL 89 95 176 64.4 71.2
CL271 CL271 CM 90 98 188 63.9 70.4
LaKast LaKast L 86 106 202 57.2 71.2
Mermentau Mermentau L 86 95 194 64.5 71.3
a
CL = Clearfield long-grain, CM = Clearfield medium-grain, L = long-grain, and M = medium-grain.
Development of Male Sterile Line for Hybrid Rice Production

E. Shakiba1, K.A.K. Moldenhauer1, G.L. Berger1, D.G. North1,
V.A. Boyett1, V.I. Thompson1, M.M. Blocker1, D.L. McCarty1, and C.H. Northcutt1
ABSTRACT
Production of hybrid rice in Arkansas has been growing rapidly in the last decade due
to its net revenue advantage over inbred lines (Lyman and Nalley, 2013). Currently,
over 40% of Arkansas rice fields consist of hybrid rice production (Berger et al., 2014).
Since 2010, the University of Arkansas, as one of the major crop variety developers in
the state, has aimed to release hybrid rice cultivars with high yield and acceptable seed
quality. Despite high yield, poor eating quality has been one of the main issues in hybrid
rice production (Khush et al.,1988). One solution to address this issue is to develop male
sterile lines with desirable phenotypic characteristics and eating quality. Therefore, in
summer 2015, 9 BC2F2 populations resulting from 2 crosses between 236s, a male sterile
line developed at the University of Arkansas System Division of Agricultures Rice
Research and Extension Center (RREC) near Stuttgart, Ark., and 2 elite rice cultivars,
Francis and Cocodrie were grown in field conditions. After extensive phenotypic and
genotypic evaluations, 59 BC2F2 single plants possessing desirable genes associated
with eating quality were selected. In the next phase, the selected 59 BC2F3 lines will
be planted in summer 2016 and the best lines will be selected as new male sterile lines.
INTRODUCTION
Hybrid rice is defined as the first generation (F1) seeds produced by a cross between
two different types of rice: one is a male sterile line used as a female parent and an elite
cultivar as a male parent. Due to a phenomenon known as heterosis, hybrid rice variet-
ies outperform their parents; hybrid rice cultivars can often produce 15% to 20% seed
yield more than rice inbred cultivars. Moreover, hybrid rice shows better resistance to
unfavorable environmental conditions such as salinity and drought. To reach maximum
seed production and yield, F1 seed needs to be produced every planting season (Virmani
et al., 1997, 2003). In 2010, the University of Arkansas System Division of Agriculture
established the rice hybrid program to address the needs of very high yield with good
1
Assistant Professor, Professor, formerly Assistant Professor, Program Technician I, Program Associ-
ate II, Program Technician II Program Associate III, Program Technician I, and Program Technician I,
respectively, Rice Research and Extension Center, Stuttgart.
95
seed quality of hybrid rice cultivars in the mid-southern United States. Since then, several
male sterile lines have been developed (Yan et al., 2012). However, the significant issue
in hybrid rice production is low eating quality characteristics such as: milling quality,
chalkiness, high or low amylose content, and low gelatinization temperature. A major
goal in 2015 was to develop new male sterile lines containing desirable agronomic traits
which can be used in hybrid rice production.
PROCEDURES
In 2014, two BC2F1 populations resulting from crosses between 236s, an Arkansas
male sterile line, and Cocodrie and Francis, two Japonica cultivars developed from
Louisiana and Arkansas respectively, were tested via molecular markers to evaluate some
agronomic and eating quality traits including plant height, grain type, aroma, amylose
content, gelatinization temperature (GT hereafter) as well as blast resistance. Rice grain
types are classified as long, medium, and short; long-grain rice is the favorable trait in
hybrid rice. Amylose content is associated with texture of cooked rice. There are three
classes of amylose content: low, medium, and high (Khush et al., 1988). The market
demands a hybrid rice with the medium amylose content. Gelatinization temperature is
related to the time required to cook rice. There are 3 classes of GT: high, intermediate,
and low (Khush et al., 1988); intermediate temperature is the desirable one for long-
grain rice, both pure-line and hybrid rice. Several blast disease resistance genes have
been reported such as Pi-ta, Pi-z, and Pi-ks. Each gene confers resistance to some, but
not all, blast isolates.
The results showed that the alleles attributed to these traits were in heterozygous
condition that cause segregation in the next generation. Therefore, the foremost goal
was to identify and select single plants containing desirable agronomic traits in homo-
zygous condition. In 2015, nine BC2F2 lines including eight 236s Cocodrie and one
236s Francis were planted in field at the University of Arkansas System Division
of Agricultures Rice Research and Extension Center (RREC), Stuttgart, Ark. All se-
lected lines were semi-dwarf and non-aromatic rice. Each single plant from each line
was tested via Simple Sequence Repeat (SSR) and Single Nucleotide Polymorphism
(SNP) markers linked to the eating quality traits and blast resistance. In addition, four
common rice cultivars including Bengal, Jupiter, Katy, Cypress, Pokhareli, Masino,
and ZHE733 were used as checks.

A total of 589 BC2F2 plants were tested via molecular study. The results revealed
that 126 single plants (about 25% of whole population) were long-grain type with inter-
mediate amylose content, 25% medium-grain with low amylose content, and 50% were
segregating in grain size and amylose content. Only those 126 single plants exhibiting
long-grain type with intermediate amylose content were selected for further evalua-
tion. Molecular study on the GT trait showed that of 126 selected BC2F2 single plants,
about 43% of single plants were medium-high, 11% low, and 46% were segregating.
96
The molecular study showed that these 126 plants possessed at least 1 gene
conferring resistance to blast disease: all contain Pi-ta gene conferring resistance to
blast disease, 8 plants contain Pi-ks gene at homozygous stage, but there was no single
plant possessing Pi-z gene.
SIGNIFICANCE OF FINDING
After careful consideration, 59 BC2F2 single male sterile plants were selected,
ratooned, and transferred from the farm to a greenhouse under controlled light and tem-
perature conditions required for seed increase (Table 1). The amount of harvested seeds
from each plant varied. These seeds will be planted as 59 plots in summer 2016 and will
be evaluated for sterility and uniformity. The best plot will be selected as a new male
sterile line developed by RREC. The selected line will be crossed with elite cultivars
and the F1 hybrid seeds will be tested for yield, in 2017 in a preliminary yield trial test.
Three BC2F1 populations resulted from cross between 811s, a male sterile line
developed at the RREC; and three rice accessions Francis, RU0302143, and RU1201102
will be planted in fields and evaluated for agronomic and eating quality. Moreover, 811s
and 236s will be planted in growth chamber to identify the genetic sources of sterility
and determine the optimum day length and temperature threshold required by the plants
to show sterility or fertility.
ACKNOWLEDGMENTS
The authors would like to express their gratitude to Arkansas rice producers via
monies administered by the Arkansas Rice Research and Promotion Board; and the
University of Arkansas System Division of Agriculture. The authors extended their
appreciation to Xue Jin for marker analysis.
LITERATURE CITED
Berger, G.L., Z.B. Yan, X. Sha, K.A.K. Moldenhauer, J.T. Hardke, C.E. Wilson Jr.,
and C.W. Deren. 2014. Development of hybrid rice cultivars. In: R.J. Norman and
K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies 2013.
University of Arkansas Agricultural Experiment Station Research Series 617:58-
63. Fayettville, Ark.
Khush, G.S., I. Kumar, and S.S. Virmani. 1988. Grain quality of hybrid rice. pp. 201-
215. In: W.H. Smith (ed.). Hybrid Rice. Int. Rice Res. Inst. Manila, Philippines.
Lyman, N. and L.L. Nalley. 2013. Economic analysis of hybrid rice performance in
Arkansas. Agron. J. 105:977-988.
Virmani, S.S, Z.X. Sun, T.M. Mou, A. Jauhar, and C.X. Mao. 2003. Hybrid rice and
heterosis breeding two-line hybrid rice breeding manual. pp. 1-3. Int. Rice Res.
Inst. Makati City, Phillippines.
97
Virmani, S.S., B.C. Viraktamath, C.L. Casal, R.S. Toledo, M.T. Lopez, and J.O.
Manto.1997. Hybrid rice breeding manual. pp. 1-10. Int. Rice Res. Inst. Los
Baos, Laguna, Philippines.
Yan, Z.B., C.W. Deren, and W.G. Yan. 2012. Hybrid rice breeding. In: R.J. Norman
and K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies
2011. University of Arkansas Agricultural Experiment Station Research Series.
561:86-88. Fayetteville, Ark.
Table1. Fifty-nine BC2F2 single plants possessing genes associated with eating quality.
Plant number Progenya Grain type Amylose Gel temp Pi-k Pi-ta Pi-z
P19-1-14 236s x CCDR Long Intermed-High Med-High S R S
P19-1-33 236s x CCDR Long Intermed-High Med-High S b
S
P19-2-94 236s x CCDR Long Intermed-High Med-High Seg. S
P19-2-110 236s x CCDR Long Intermed-High S R S
P19-2-126 236s x CCDR Long Intermed-High Med-High Seg. R S
P19-2-134 236s x CCDR Long Intermed-High Med-High R R S
P19-2-145 236s x CCDR Intermed-High Med-High R R S
P19-2-148 236s x CCDR Intermed-High Med-High Seg. R S
P19-3-265 236s x CCDR Long Intermed-High Med-High S
continued
98
Table 1. Continued.
Plant number Progenya Grain type Amylose Gel temp Pi-k Pi-ta Pi-z
P19-3-307 236s x CCDR Long Intermed-High S S
P19-4-316 236s x FRNS Long Intermed-High Med-High S R S
P19-4-370 236s x FRNS Long Intermed-High Med-High S R S
a
FRNS = Francis, CCDR = Cocodrie, S = susceptible, R = resistance, and Seg = segregating.
b
Blank = no data.
99
Development of Aromatic Rice Varieties

D.K.A. Wisdom1, K.A.K. Moldenhauer1, C.E. Wilson Jr.1,
X. Sha1, J.M. Bulloch1, B.A. Beaty1, M.M. Blocker1, and V.A. Boyett1
ABSTRACT
Interest in aromatic rice has increased with the advent of nouveau cuisine causing a
rise in niche markets. Sales of aromatic rice have led rice imports to increase over
30% in the last ten years. The University of Arkansas System Division of Agricultures
Aromatic Rice Breeding Program at the Rice Research and Extension Center (RREC),
Stuttgart, Ark., was implemented to develop aromatic rice varieties for the southern
rice-producing regions. Evaluating cultural practices is essential for selecting advanced
lines in the breeding program as well as for growers. Information regarding successful
cultural practices of aromatic rice varieties is very limited for the southern United States
growing regions, and especially for Arkansas.
INTRODUCTION
Approximately 13.6 million metric hundredweight of milled rice were imported
to the United States in the fiscal year 2011/2012 (USA Rice Federation, 2009, 2012).
Of the 19% imported rice consumed domestically, 58% came from Thailand in the
2012/2013 milling year (USA Rice Federation, 2015). Thailand produces high quality
Jasmine rice and India, which provides the second largest amount of imported rice,
produces highly desired Basmati rice (USA Rice Federation, 2012, 2015). United States
consumers are purchasing more aromatic and/or specialty rices than in previous years.
It has been difficult for U.S. producers to grow the true Jasmine and Basmati varieties
due to environmental differences, photoperiod sensitivity, fertilizer sensitivity, and
low yields. These difficulties make aromatic rice an expensive commodity to produce.
Adapted aromatic rice varieties need to be developed for Arkansas producers which
meet the taste requirements for either Jasmine or Basmati.
1
Program Associate II, Professor, Center Director, Associate Professor, Program Associate I, Program
Associate I, Program Associate III, and Program Associate II, respectively, Rice Research and Exten-
sion Center, Stuttgart.
100
PROCEDURES
The aromatic rice breeding program collected parental material from the U.S.
breeding programs and the USDA World Collection. Crosses were made to incorporate
traits for aroma, yield, improved plant type, superior quality, and broad-based disease
resistance. The winter nursery in Puerto Rico is being employed to accelerate genera-
tion advance of potential varieties for testing in Arkansas during the summer of 2016.

In 2015, 87 cross-pollinations were successfully completed to produce aromatic
lines for future screening. The F1 plants from these crosses were grown in the green-
house during the winter to produce F2 seed. The F2 populations will be planted in 2016
at RREC for observation and selection.
Panicles were selected from 43 F2 populations in 2015. The parents in these crosses
were selected for their aromatic seed quality or high yield potential. Approximately
2570 F3 lines from 43 populations were shipped to the winter nursery in Puerto Rico to
advance. The harvested seed from Puerto Rico will be planted at the RREC for further
observation and selection in 2016. Panicle rows from 28 F4 and F5 populations will be
grown in 2016 for observation. Selections from these populations will be harvested
and samples from the 28 populations will undergo molecular marker analysis. Lines
that have the preferred markers for aroma, cooking quality, and blast resistance will be
entered in yield trials in 2017.
In 2015, 151 heterozygous lines from 36 F4, F5, and F6 populations were screened
through marker assisted selection for aroma and amylose content. Results of the screen-
ing helped to eliminate lines which did not meet breeding program requirements. The
entries which are homozygous aromatic will move forward into yield trials.
In a two-replication preliminary trial planted in 2015, 19 aromatic lines were
evaluated for yield. In the Aromatic Stuttgart Initial Test, which has four replications,
19 aromatic lines were evaluated for yield and potential release. Seed from the top
yielding 14 experimental lines with preferred plant types were milled and cooked in a
taste test during the winter 2015. The four experimental lines chosen as having the best
flavor and aroma have been entered in the Arkansas Rice Performance Trials (ARPT)
and are being grown in increase plots in 2016. Four aromatic experimental lines have
also been entered in the 2016 Uniform Regional Rice Nursery (URRN).
In 2015, four Jasmine type experimental lines were entered in the cooperative
URRN. The Arkansas mean yields for the four lines were: EXP14105, 174 bu/acre;
EXP15102, 152 bu/acre; EXP15105, 127 bu/acre; and EXP15108, 151 bu/acre. Also
in 2015, six experimental lines were entered in the Arkansas Rice Performance Trials.
One experimental line that showed promising potential in the ARPT and URRN was
EXP14105. This line originated from a cross between Jazzman and a plant introduction
line. EXP14105 has excellent flavor and will continue to be examined in the ARPT and
URRN in 2016. Head rows of EXP1405 will be planted for foundation seed increase
in 2016.
101
ACKNOWLEDGMENTS
The authors appreciate the financial support of the rice producers of Arkansas
through monies administered by the Arkansas Rice Research and Promotion Board.
Support also provided by the University of Arkansas System Division of Agriculture.
We thank Vetress Thompson and Xue Jin for conducting marker analysis.
LITERATURE CITED
USA Rice Federation. 2009. U.S. Rice Domestic Usage Report For the Milling Year
August 1, 2008 - July 31, 2009. Access date: January 2013. Available at: www.
USARice.com
USA Rice Federation. 2012. U.S. Rice Domestic Usage Report For the Milling Year
August 1, 2011 - July 31, 2012. Access date: January 2013. Available at: www.
USARice.com
USA Rice Federation. 2015. Domestic Usage Report: Per Capita Rice Consumption
Increases in 2014. November 24, 2015. Access date: March 2015. Available at:
www.USARice.com
102
Rice Breeding and Pathology Technical Support

C.D. Kelsey1, S.B. Belmar1, K.A.K. Moldenhauer1, and Y.A. Wamishe2
ABSTRACT
Development of disease resistant rice is one of the most important achievements rice
breeders attempt to accomplish at the University of Arkansas System Division of Agri-
cultures Rice Research and Extension Center (RREC) near Stuttgart, Ark. The centers
plant pathology group assists with this goal by screening rice germplasm, before the
lines become advanced breeding entries, in the greenhouse and field. Breeding materials
are mostly evaluated using artificial inoculation for blast and sheath blight diseases of
rice at the RREC and Pine Tree Research Station (PTRS). Artificial inoculation of these
pathogens on rice is essential for collecting disease severity data. Considerable amounts
of disease inocula are prepared in the laboratory and applied to the plants using specific
protocols. Screening for blast is conducted both in the greenhouse and the field. Screen-
ing for sheath blight is only in the field. Data from these tests are used by the breeding
program either to transfer genes for resistance into adapted high yielding varieties or to
advance entries for further agronomic testing. As part of the crucial responsibility for
the rice extension pathology program, screening for bacterial panicle blight is largely in
the field with some selected lines tested in the greenhouse. The breeding and pathology
technical group, therefore, assists the extension plant pathology program not only in
screening for bacterial panicle blight but with all applied research for finding answers
to manage major prevailing and newly emerging diseases, including collaborative
interdepartmental, industry and multi-state research endeavors.
INTRODUCTION
Disease resistance is an important element needed when developing new variet-
ies in any breeding program. At the Rice Research and Extension Center (RREC), rice
breeders and pathologists work together to develop varieties with desirable disease
resistance along with pertinent agronomic traits. Disease evaluation of crops for major
diseases starting in early generations has been important and is a required activity for a
successful breeding program. Lines that have some potential traits but have not reached
full expectation for release may possibly be used as parents to develop new varieties.
1
Program Technician II, Program Technician II, and Professor, respectively, Rice Research and Exten-
sion Center, Stuttgart.
2
103
Rice blast caused by Magnaportha grisea (T.T. Herbert) M.E. Barr is still an im-
portant disease Therefore, emphasis is given to evaluate breeding materials for both leaf
and neck/panicle blast. Screening for leaf blast in the greenhouse at the seedling stage
is more efficacious than field screening. However, field screening is still practiced due
to the differences in controlled and natural environments. Neck/panicle blast screening
is conducted only in the field where mature plants are easily maintained. Growing rice
to maturity under natural conditions also allows breeders to gain beneficial knowledge
about blast disease from a commercial growers point of view.
Sheath blight (Rhizoctonia solani Kuhn) also remains a major disease of rice.
Breeding materials are evaluated in the field at the RREC. Although there is no quali-
tative resistance to this pathogen, knowledge on quantitative resistance/tolerance, is
greatly important to the breeding program.
Bacterial panicle blight (BPB) caused largely by Burkholderia glumae (Kurita
and Tabei), formerly known as Pseudomonas glumae has shifted from being an emerg-
ing to an established disease since many of the conventional commercial rice varieties
are susceptible. There are still many unanswered questions relating to the survival of
this bacterium. Research is being conducted in the laboratory, greenhouse, and field
to further develop practical management techniques and to obtain more understanding
about this disease.
Screening for disease resistance under natural conditions may not be as reliable as
it is with artificial inoculation for any of the three diseases tested at the RREC. Screening
plants for blast requires desired environmental conditions prior to and after artificial
field inoculations for the pathogen to sporulate and cause infection. Inoculation usually
precedes a weather front that delivers dew or light rain conditions. The blast pathogen
is aseptically grown on agar dishes then homogenously mixed with a sterile corn cops/
ryegrass mixture for field dispersal within 24 hours. Blast inoculum preparation is done
multiple times throughout the season. Sheath blight inoculum also requires massive
amounts of the seed mixture to be produced; however, months of careful preparation
allow it to be stockpiled for a single field application.
Bacterial isolation and purification needs a semi-selective media. The production
of bacterial suspension for inoculation requires a constant supply of bacteria grown on
Kings B agar media. Common microbiological aseptic technique is used to produce
bacterial culture plates to obtain sizable volumes of a useable bacterial suspension.
Skilled assistants/associates within the breeding/pathology support group train
new employees, students, and seasonal workers in all aspects of the lab, field, and
greenhouse responsibilities. This allows flexibility in assigning daily activities to timely
accomplish responsibilities for the breeding and extension programs.
PROCEDURES
Evaluation of Breeding Materials for Blast Resistance in the Greenhouse
In 2015, nearly 700 entries comprised of the Uniform Rice Regional Nursery
(URRN), the Arkansas Rice Performance Trials (ARPT), advanced lines, the Stuttgart
Initial Tests (SIT)/Clearfield Stuttgart Initial Tests (CSIT), hybrid lines and preliminary
104
breeding materials were evaluated using 4 to 6 races of M. grisea individually per test.
Seven-day old individual race isolates were washed from agar plates with a xanthan gum
suspension to create a standardized spore suspension of 2.0 104 spores/ml. Each test
was replicated three times and each individual suspension was applied at the 3- to 4-leaf
growth stage (approximately 21 days after planting) using a Badger 250-2 basic spray
gun. Spray inoculated plants were immediately placed in a dew chamber for about 14
hours. Disease data were collected 7 to 10 days after the plants were removed from the
dew chamber and placed on a greenhouse bench. Disease evaluation was conducted 7
to 10 days after inoculation using a rating scale of 0 (no disease) to 9 (severe disease).
A single comprehensive greenhouse test for only one blast race required 28 to 30 days.
Evaluation of Breeding Materials for Blast and Sheath Blight

Field testing for blast disease was replicated four times. Inoculum preparation
for blast required sterilizing several hundred gallons of cracked corn (corn chops) and
ryegrass seed. Several gallons of milo were also sterilized for blast inoculum. The
sterilization protocol required 24 hours to process around 60 gallons for blast inoculum
production. Blast races are grown on agar for 7 days before being homogenously mixed
into sterile chops/milo and ryegrass seed. After mixing, the inoculum is distributed onto
the test plots within 24 hours. Test plots were established on 11 June at the Pine Tree
Research Station (PTRS) as hill plots surrounded by a spreader mixture of blast sus-
ceptible lines to encourage the buildup of spores for disease spread. Eight rows of corn
were planted on and around the levee as a partial windbreak. Rice plants at tillering and
heading were inoculated with semi-dried seed media which contained five races of the
pathogen. Test plots and spreader rows were inoculated once for leaf blast at the 4-leaf
stage, once at 5-leaf stage and at least twice for panicle blast starting at boot emergence.
An estimation of 24 g per 6 hill plots was hand-broadcasted. A similar test was also
initiated at the RREC on 8 May using entries only from the ARPT and URRN tests.
Sheath blight disease testing was replicated four times in the field. The inoculum
process also required several hundred gallons of sterilized corn chops and ryegrass seed.
Preparation of approximately 16 gallons of inoculum took 3 days to develop. The cultures
were grown on agar for 5 days before being added to sterile chops/ryegrass seed media
and allowed to grow at room temperature for a week. The sheath blight infected seed
media was then air-dried and kept in paper bags until ready for field inoculation. Sanitary
conditions were maintained throughout the entire process to avoid contamination. A
hill plot nursery was planted on 8 May at the RREC. Air-dried inoculum that contained
six isolates of the pathogen was applied to test entries at the panicle initiation growth
stage. Approximately 24 g per 6 hill plots of inoculum was also hand-broadcasted. A
rating scale of 0 (no disease) to 9 (severe disease) was used to collect disease severity
data from all plots approximately 7 weeks after the last inoculation.
Extension Pathology
Field testing for bacterial panicle blight on the URRN/ARPT was replicated
twice in hill plots that were planted on 23 May. One replication was inoculated using
105
a back-pack sprayer with a bacterial suspension (ca 106 to 108 colony forming units/
ml) directly on the plants between boot-split to flowering stage. The other replication
was observed for natural infection. Disease data were collected from all plots using a
rating scale of 0 (no disease) to 9 (severe disease).
The breeding/pathology technical group also provided assistance to applied
research conducted by the rice extension plant pathology program with greenhouse,
laboratory, and field activities. Studies on survival of B. glumae in soil, seeds, or crop
residues required production of over 5 gallons of CCNT, a selective media for the bac-
terium, to monitor its presence in colonized substrates. In addition, 4 gallons of Kings
B media was prepared to cultivate B. glumae for seed inoculations and mid-season rice
plant inoculations. The study on effects of potassium on Burkholderia glumae conducted
at PTRS required 35 pounds of Bengal seed inoculated with a bacterial suspension.
Greenhouse disease evaluation included mostly preliminary studies on various
methods to artificially inoculate plants at both the seedling and adult developmental
stage. Tested techniques included direct seed dip, foliar spray, syringe injection of culm,
a cut leaf dip, and a soil inoculation. An investigative test of 10 entries previously tested
in the 2012/2013 URRN/ ARPT with ratings of resistant (R) /moderately resistant (MR)
was also conducted in both the greenhouse and field. This testing began in 2014 using
126 selected entries.
Field tests were conducted in collaboration with chemical industries that included
5 products with a total of 36 treatments for sheath blight, one early-season seedling
disease containing 9 treatments and one false smut with 3 treatments. All of these tests
were done in 4 replications. Inoculum for sheath blight amounted to 32,400 g to meet
the needs of industry tests. The inoculum production endeavor for all tests required
a substantial amount of time to prepare the various inocula and to inoculate fields at
RREC and PTRS related to extension, industry, and potassium collaborative research.
Disease data were collected from the respective plots and summarized for each test as
each protocol required.
A false smut study was designed using the variety Roy J. A mist system was
constructed to automatically apply moisture for a two-hour period during both morn-
ing and evening to create favorable conditions. A very susceptible germplasm, GP2,
was planted for a border to determine the effect of disease initiation and enhancement.
Due to the lack of field protection using current fungicides, preliminary in vitro
studies were conducted to check the sensitivity of the kernel smut pathogen in chemically
amended agar. Colonies were estimated using a common microbiological procedure to
compare the sensitivity of the fungus to triazoles and strobulin combination fungicides.

As part to the breeding program, disease assessment of rice for resistance/toler-
ance to sheath blight, blast, and bacterial panicle blight was successfully completed
for 2015. A large number of candidates rated 0 to 4 for both leaf and neck/panicle blast
to the races tested. Entries tested in 2015 for sheath blight showed lower ratings than
106
the previous year which may be due to the dry and hot weather of late July and August
along with about 50 pounds less total nitrogen applied to plants.
Field blast and sheath blight evaluations were assessed for about 1600 experimen-
tal lines and checks (Table 1). These lines include ARPT, URRN, ARPT-Imidazoline
(IMI), Missouri lines, SIT-IMI, SIT, and preliminary materials from Dr. Moldenhauer,
aromatics, and five long-grain advanced entries, SIT and CSIT from Dr. Shas breeding
programs. A total of 10,256 hill plots were established that included all replications.
All were inoculated and evaluated for leaf and neck/panicle blast. For sheath blight,
the total number lines inoculated and evaluated was 8924 hill plots including replica-
tions. In addition, the support group also does field evaluations of breeding materials
for diseases under natural infection.
Of 688 experimental lines tested for leaf blast in the greenhouse, several lines
rated between 0 and 4 (Table 2). Experimental lines were screened using 6 blast races.
ARPT, URRN, 22 combined entries from Dr. Shas CSIT and SIT, and aromatics were
included in the tests. Four races of the six used in aforementioned tests were used for
preliminary lines. In three replications, all tests included, the total number of evalua-
tions for leaf blast in the greenhouse was 10,410.
Field evaluations for bacterial panicle blight resistance included 200 entries of
the URRN and 86 entries of the ARPT in hill plots replicated twice. Of these tested
entries, 27 showed low disease scores with 0 to 4 rating; 19 entries from the URRN
and 8 entries from the ARPT (Table 3). Given the subset of entries with a rating of 1,
some were considered to be late maturing: URRN had 5 entries and ARPT 4 entries
(Table 3, Wamishe et al., 2016). These later-maturing entries may be disease escapes
with their development coinciding with unfavorable weather for pathogen development
and spread. In addition, 10 entries were selected for re-evaluation in the greenhouse
with different types of inoculation methods after showing levels of consistent R/MR
ratings (Wamishe et al., 2015).
The breeding/pathology tech support group provided an immeasurable amount
of support to the success of research activities in extension pathology starting from
preliminary to full-fledged applied research, collaborative research with industries and
interdepartmental research along with evaluations of the breeding materials. Assistance
in disease evaluations is also provided for other research departments on- and off-site
upon request.
The goal of the rice breeding/pathology technical support group will always be
to provide support to increase the efficiency of rice breeders in developing maximum
yielding cultivars with expected levels of disease resistance. The group also plays a vital
role in extension plant pathology assisting with applied research. Disease evaluation
remains pivotal for breeding for disease resistance. A strong applied research approach
also provides dependable and practical solutions to rice producers in Arkansas and other
rice-producing states. Therefore, this technical support group is actively working with
the rice breeders and the extension pathology program to enhance rice productivity.
107
ACKNOWLEDGMENTS
The authors gratefully acknowledge the Arkansas rice producers and their support
through the monies administered by the Arkansas Rice Research and Promotion Board
and the University of Arkansas System Division of Agriculture through their continued
interest and funding. Thanks also go to the USDA-ARS for their cooperation, interest,
and evaluation of materials and to the other University of Arkansas System Division of
Agriculture Research Stations located throughout Arkansas for their continued support.
LITERATURE CITED
Wamishe, Y., Y. Jia, T. Gebremariam, C. Kelsey, S. Belmar, and T. Mulaw. 2015.
Characterization of rice germplasm for genetic resistance to bacterial panicle
blight and development of techniques for monitoring the bacterial pathogen. In:
R.J. Norman and K.A.K Moldenhauer (eds.). B.R. Wells Arkansas Rice Research
Studies 2014. University of Arkansas Agriculture Experiment Station Research
Series 626:109-117. Fayetteville, Ark.
Wamishe, Y., T. Mulaw, T. Gebremariam, S. Belmar, and C. Kelsey. 2016. Studies on
bacterial panicle blight disease relative to soil depth, crop residue, dew develop-
ment, and potassium levels. In: R.J. Norman and K.A.K. Moldenhauer (eds.).
B.R. Wells Arkansas Rice Research Studies 2015. University of Arkansas Agricul-
ture Experiment Station Research Series 634:110-119. Fayetteville, Ark.
Wamishe, Y., T. Mulaw, Y. Jia, T. Gebremariam, S. Belmar, and C. Kelsey. 2016.
Monitoring bacterial panicle blight disease of rice and germplasm evaluation for
resistance in Arkansas in 2015. In: R.J. Norman and K.A.K. Moldenhauer (eds.).
B.R. Wells Arkansas Rice Research Studies 2015. University of Arkansas Agricul-
ture Experiment Station Research Series 634:120-125. Fayetteville, Ark.
108
Table 1. Number of entries rated as disease tolerant in 2015 field disease nurseries.
Total Sheath Leaf blasta Neck/panicle blasta
Test entries blight RRECc PTRS RREC PTRS
(0-5 rating )
b
---------------------(0-4 rating )----------------------
b
ARPT 90 15 26 19 51 17
URRN 200 14 22 12 75 40
Aromatic 46 1 10 4 29 8
Missouri 100 1 NA 30 NA 18
ARPT-IMI 66 2 NA 23 NA 14
Molden-SIT 140 6 NA 71 NA 59
SIT-IMI 278 13 NA 79 NA 56
Sha-SIT 149 0 NA 89 NA 21
Sha-CSIT 149 2 NA 60 NA 20
Prelim 335 18 NA 202 NA 114
a
Five races used in bulk for screening of entries to blast under field conditions.
b
Rating scale of 0 (no disease) to 9 (severe disease) was used. Entries with two or more obser-
vations of a low disease rating were included.
c
Abbreviations: RREC = Rice Research and Extension Center; PTRS = Pine Tree Research
Station; ARPT = Arkansas Rice Performance Trials; URRN = Uniform Regional Rice Nursery;
ARPT-IMI = Arkansas Rice Performance Trials - Imidazoline; SIT = Stuttgart Initial Tests; and
CSIT = Clearfield Stuttgart Initial Tests.
Table 2. Number of entries rated as disease

tolerant (0 to 4a) for 2015 greenhouse leaf blast testing.
Test Entry total IB-1b IB-49 IC-17 IE-1K
URRNc 199 74 57 85 76
ARPT 89 27 18 30 11
Aromatic 43 26 12 20 21
Sha-SIT 22 7 6 11 10
Prelim 335 209 187 188 100
a
Rating scale of 0 (no disease) to 9 (severe disease) was used.
b
Data from most prominent Arkansas blast races IB-1, IB-49, IC-17 and IE-1K used for table.
c
Abbreviations: URRN = Uniform Regional Rice Nursery; ARPT = Arkansas Rice Performance
Trials; and SIT = Stuttgart Initial Tests.
Table 3. Number of entries with low bacterial

panicle blight ratinga from 2015 field evaluation.
Test Total entries Potential entries Late maturity entries
URRNb 198 19 5
ARPT 88 8 4
a
Rating scale of 0 (no disease) to 9 (severe disease) was used.
b
Abbreviations: URRN = Uniform Regional Rice Nursery and ARPT = Arkansas Rice Perfor-
mance Trials.
For additional information see Wamishe et al. (2016).
109
Studies on Rice Bacterial Panicle Blight Disease Relative to Soil

Depth, Crop Residue, Dew Environment, and Potassium Levels
Y.A. Wamishe1, T. Mulaw2, T. Gebremariam1, S.B. Belmar2, and C.D. Kelsey2
ABSTRACT
Bacterial panicle blight (BPB) disease of rice is mainly caused by Burkholderia glumae
and possibly other Burkholderia species. Bacterial panicle blight is thought to favor
hot and dry summers particularly under extended high night temperatures. Bacterial
panicle blight is one of the most threatening diseases for rice production in Arkansas
and other southern rice-producing states. In 2010 and 2011 late-planted conventional
rice fields were hit the hardest with BPB. Host resistance is believed to be the ultimate
solution to combat this disease. Early planting, adequate rates of nitrogen fertilizer, and
seeding rates have also been experimentally proven to reduce the disease incidence. The
role of dew on plants and soil potassium content were tested in 2015 in relation to BPB
disease development and/or the reduction of disease incidence. Research is on-going
to understand the survival and infectivity of the bacteria in soil and plant residue. To
be able to detect the lowest concentrations of B. glumae from soil, three culture media
(CCNT, CPG, SMART), and three vegetables (yellow onion, carrot, celery) were tested.
Among the culture media, CCNT was able to detect 3.1 103 cfu/g (colony forming
units/gram) of soil from natural soil and 49 cfu/g soil from sterile soil. Both soils were
artificially infested with a 48-h culture of B. glumae at 4 104 cfu/mL. Yellow onion was
able to detect the bacteria in both sterile and non-sterile soil at a rate of approximately
1.6 103 cfu/g soil and in one case down to 98 cfu/g soil. Further testing of onion with
B. glumae-infested soil is needed to further refine the technique, and tests also need
to be conducted with a B. gladioli culture. The former is largely seedborne while the
latter is believed to be more soilborne. Preliminary infectivity tests in the greenhouse
and growth chamber from soil and residue were inconclusive. B. glumae survived in
the greenhouse soil for up to four months with a substantial decrease in population size
over time. The field test for infectivity of a susceptible rice cultivar by rice residue was
negative. Infected residue buried or kept on the soil surface in the greenhouse tested
negative on CCNT culture medium in a month, while residue in a cloth bag kept on a
1
Assistant Professor and Program Associate - Plant Pathology, respectively, Department of Plant Pathol-
ogy, Stuttgart.
2
Plant Pathology Program Technician, Program Technician II, and Program Technician II, respectively,
Rice Research and Extension Center, Stuttgart.
110
greenhouse bench for six months tested positive. Substantial blight symptom develop-
ment occurred on rice panicles kept in a dew environment when sprayed with a bacterial
suspension at the flowering stage. However, plants that were needle-inoculated at earlier
growth stages (V4, V7, and V11) sometimes showed the visual BPB symptoms on the
panicles. Rice plants incubated in dew chamber from 24 to 48 h showed an enhanced
effect on panicle infection when they were needle-inoculated on the stem when the
antepenultimate and penultimate leafs were emerging. Field results from 2015 annual
potassium rates on incidence and severity of BPB were inconclusive. In some plots,
the denser the plants, the more panicles infected. There was no significant difference
between plants grown from inoculated and non-inoculated seeds. However, the amount
of potassium applied significantly affected grain yield.
INTRODUCTION
Bacterial panicle blight (BPB) disease of rice is sporadic which could be due to
changes in weather and environmental conditions along with multiple causal agents
that may survive in seeds, soil, or crop residues. The disease is primarily seedborne
and seems to favor hot summer nights. Historically, BPB was reported to cause up to
60% yield loss in susceptible rice cultivars under environmental conditions favorable
for pathogen development and spread. Panicle symptoms typically develop late in the
season during grain fill. In fields, infected panicles mainly have blighted florets which
first appear white to light gray with a dark-brown margin on the basal third of the tissue.
Later, these florets turn straw-colored and may further darken toward the end of the
season with growth of other opportunistic microorganisms. Heavily infected panicles
remain upright due to lack of grain fill. The bacteria most likely can be carried around
by wind during rain. Predicting BPB disease occurrence and severity level before grain
fill appeared difficult. The disease increased between 1995 and 2011. Historic epidemics
on conventional commercial rice occurred in 2010 and 2011 both in Arkansas and other
rice-producing states in the U.S. Bacterial panicle blight occurrence was relatively low
in 2012 when the weather was hot and dry during the growing season. In 2013 and 2014
conditions were wet and cold for most of the growing season and there were no great
issues with BPB. Although 2015 started wetter and colder than the prior two years, the
latter three weeks of July and August became hot and dry increasing the apprehension
of having another BPB epidemic. The hot and dry weather during this time essentially
caused a lot of panicle blanking followed by various types of kernel discoloration in
several rice fields of Arkansas. However, none of the panicle samples received from
commercial fields in Arkansas tested positive for BPB under our laboratory testing
procedure. Our research in recent years has proven that early planting, adequate seeding
and nitrogen fertilizer rates reduce the incidence of BPB. Adequate water supply has
been shown to increase the productivity of rice in spite of BPB incidence (Wamishe
et al., 2014). To date, there are no chemical options registered in the U.S. to protect or
salvage the crop from BPB disease. Efforts are being made to evaluate and develop rice
cultivars with resistance. Moreover, research is underway to understand the biology of
the causal pathogens and their interaction with rice plants.
111
This study briefly presents a single year progress report with the following ob-
jectives which were considered to be continuations of the prior research on short term
BPB management options: 1) to evaluate the survival of B. glumae and its infectivity
on rice in infected rice residues and infested soil; 2) to observe/evaluate the effect of
a dew/mist on the development of bacterial panicle blight disease and a study of the
association of dew in fields by tree lines and waterways; and 3) to evaluate the effect
of potassium fertilizer on bacterial panicle blight disease.
PROCEDURES
Evaluation of Survival and Infectivity of B. glumae from Soil and Rice Residue
Survival Test for B. glumae in the Soil. Burkholderia glumae is used for our stud-
ies because our isolate collection and surveys to date indicated it is the major causal
species of BPB rice disease in Arkansas. Although B. glumae is mainly seedborne,
there is no strong evidence against it inhabiting soil or rice residue. With continuous
rice cultivation on many zero grade fields in Arkansas, information on longevity and
infectivity of these bacteria in the soil or residue is desired.
Preliminary greenhouse tests were carried out to determine whether B. glumae
can survive incorporated into a sterile soil environment. Bacteria survival in drainage
water was also tested by studying bacterial leaching from rice root zones where the seed
inoculation was followed by rainy days after planting (2013 to 2015). Three different
soils were used: field soil from the University of Arkansas Division of Agricultures
Rice Research and Extension Center (RREC) near Stuttgart, Ark., greenhouse soil
mixed with vermiculite and greenhouse soil mixed with sand as 16 soil: 1:1 or 2 soil:
1:1, respectively. Approximately 2.2 lb of each soil was sterilized by pre-moistening
for a couple of hours in an autoclave bag before placing in autoclave for an hour expo-
sure time on solid cycle. Sterile soil was allowed to cool overnight prior to filling pots.
Four chemically cleaned plastic pots (4-in. diam.) were filled with each of the soils. For
each soil type, three pots of soil were prepared by thoroughly mixing the suspension,
of bacteria (5 mL of approx. 109 cfu/mL B. glumae in salt-sugar buffer), into the soil.
A non-inoculated control pot for each soil type was mixed with 5 mL sterile water.
The pots were placed in open ziplock bags to collect leachate from periodic watering
in the greenhouse. Initial soil sampling was done by gently pushing a drinking straw
through the soil profile to obtain approx.0.5 g of soil along the depth of a pot. Top and
bottom soil profiles were obtained by cutting the straw filled with soil in half. Soil was
carefully removed from each portion and vortexed with 5 mL sterile water. A 100 l
aliquot of suspension was spread onto CCNT medium and incubated at 39 C. Colony
forming units were counted after 48 h of incubation. Samples were collected and tested
as described above for all treatments in 4 week intervals for up to 5 months. Drainage
water was also tested by streaking 0.1 mL on CCNT culture medium. Pots were irregu-
larly watered with sterile distilled water thus allowing for the soil to dry intermittently.
A field experiment is underway to test the survival of B. glumae for a six-month
period representing the duration from rice harvest to planting. The experiment was
112
started in October 2015 and it will run up to 15 March 2016. Soil is sampled and tested
in monthly intervals. The procedure and results will be reported in the future.
Development of Technique to Isolate B. glumae from Soil. Comparative tests
were conducted between culture media and vegetable hosts. CCNT (Kawaradani et
al., 2000), CPG (Jeong et al., 2003), and SMART medium (Kawanishi et al., 2012)
were selected based on previous experience and a literature search. Among vegetables,
yellow onion, carrot, and celery were selected based on literature and availability. The
main objective in this study was to determine the best technique for detection of a low
concentration of B. glumae in the soil with the expectation of using the method later
for B. gladioli. A 15 mL bacterial suspension approx.4 104 cfu/mL was prepared in
a salt-sugar buffer (Streeter, 2007) and applied to 50 g sterilized or non-sterilized soil
that were handled separately. A gram of each soil was thoroughly suspended in 10 mL
of sterilized water. A 0.1 mL aliquot of soil suspension was transferred to CCNT media
for each of two replications. A 1:1 serial dilution of each soil suspension was carried
out 15 times (Table 1). Media were incubated for 48 h at 39 C before scoring the plate
for the characteristic cultural morphology of B. glumae. Fleshy onion scales and cross
sectioned pieces of celery and carrot were placed on moist filter paper in a petri dish,
scratched with sterile scalpel to create a wound, and then inoculated with 10 l of either
sterile or natural soil suspension. Vegetable hosts were incubated at 30 C for 7 days
until sunken water soaked lesions are formed. Tissue from the lesion was streaked back
to CCNT to confirm the identity of the bacteria using morphological characteristics.
Survival Test for B. glumae in Rice Residue. Four different preliminary tests were
utilized to study survival of B. glumae in rice residue. 1) B. glumae infected residue
collected from artificially inoculated plants were air dried in a cloth bag on a greenhouse
bench for 6 months and then tested on CCNT for B. glumae; 2) Seeds of Bengal sterilized
with oxolinic acid were planted with infected residue mixed in a 1:1 (soil: residue by
weight) ratio in the greenhouse. However, due to decaying residue, the test was modified
to 2:1 (soil: residue by weight) and pots were kept in a growth chamber conducive to
rice growth. From the latter test, rice plant tissues were collected at different growth
stages and were tested for the presence of B. glumae in a polymerase chain reaction
(PCR)-based assay; 3) In another test, infected rice residue was buried in pots filled
with sterilized field soil about an inch deep while in another set residue was left on the
surface of the soil. Pots were then watered irregularly for up to 4 months. 4) A replicated
infectivity test from residue was carried out at the RREC in a field where rice was not
planted in the previous 5 years. Seeds mixed with well chopped infected residue were
planted by hand in a 152-ft plot with four replications. Treatments included a 1: 2, 1:1,
1:0.5 (seed: residue ratio by weight) mixture. The control plots had no residue added.
A separate micro-bay was constructed for the control plots to avoid possible cross
contamination of B. glumae through flood water.
Observations/Tests on Effect of Dew/Mist on Bacterial Panicle Blight Disease

Observation 1. Seventy rice entries had staggered planting dates based on their
heading date to synchronize the timing of flowering for a mass inoculation with B.
glumae. Due to weather-induced variability, selected rice plants were pulled out of
113
the field and brought to the greenhouse to create a similar pre- and post-inoculation
environment. Each rice cultivar was kept in a 3-gal bucket. Plants were inoculated twice
in 4-day intervals in the greenhouse following a field standard procedure (Wamishe et
al., 2014). The pots were kept on the greenhouse floor to avoid drying out the panicles
from the circulating air fans or lights. Due to a failure in symptom development, the
pots were moved in dew chambers and incubated for 48 h. The initial purpose of this
study was to evaluate rice cultivars for resistance to bacterial panicle blight under the
same environment.
Observation 2. Ten selected rice entries from the 2014 Uniform Regional Rice
Nursey (URRN) and Arkansas Rice Performance Trials (ARPT) that rated as moderately
resistant to BPB and three control varieties, Jupiter (MR), Bengal (S), and CL151(S)
were planted in a RREC field, spray inoculated at flowering and misted from boot split
until grain-fill for four hours each day in late evening and early morning with a misting
system constructed in the field. The initial purpose of this study was to re-evaluate the
resistance level of the selected rice cultivars in the field under dew environment.
Preliminary Tests to Evaluate Effect of Dew on Bacterial Panicle Blight. A few
plants of the rice variety Wells were pulled from the field at flowering stage, brought to
the greenhouse and spray inoculated and then kept in dew chamber for 55 hours. The
same variety in the field with a thin plant stand was heavily sprayed with B. glumae
suspension and left under the dry-hot weather in the last week of July 2015.
A set of cultivars used in Observation 2 were pulled out and brought to the green-
house. They were inoculated with a bacterial suspension using a syringe at the 7-lf stage
and incubated in three conditions: a) in a dew chamber at night and on the greenhouse
bench during the day for three days; b) in a dew chamber continuously for 48 h; and c)
left continuously on greenhouse bench. Lesions were measured 7 days after inoculation.
Effects of Annual Potassium Fertilizer Rates on Rice Bacterial Panicle Blight

This is the first year of a test carried out in collaboration with Nathan Slaton on
long-term plots established for potassium yield evaluations at the University of Arkansas
System Division of Agricultures Pine Tree Research Station (PTRS), near Colt, Ark.
The test field consisted of five potassium fertilizer (0, 40, 80, 120, 160 lb K2O/acre)
rates in a randomized complete block design with 9 replications. Burkholderia glumae-
inoculated seeds and non-inoculated seeds of a susceptible rice variety of Bengal were
planted at a recommended seeding rate (88 lb/acre) on 23 April 2015 with a no-till
planter. Plots were grown in two bays. All other inputs and irrigation were maintained
in the same manner for all of the plots. Infected panicle counts per plot were collected
on 11 and 12 August 2015 and statistically analyzed.

Evaluation of Survival and Infectivity of B. glumae from Soil and Rice Residue
Survival Test for B. glumae in the Soil. Preliminary infectivity tests in the
greenhouse and growth chamber from soil and residue were inconclusive due to false
114
positives with classical PCR-based assays in B. glumae presumably clean control soils.
Since the control pots which contained sterilized soil were kept in the same growth
chamber, the contamination probably was caused during watering and by plant-to-plant
contact.
In a separate test, B. glumae survived in greenhouse soil for up to four months
with a substantial decrease in population size over time. The greenhouse temperature
ranged from 78 F to 83 F during the test period. Burkholderia glumae survived lon-
ger in the bottom and wet part of the sterilized field soil than the soil surface that was
mostly dry due to the irregular watering. Burkholderia glumae also survived longer
in pots filled with sterile field soil than the greenhouse soil. It was not clear why B.
glumae survived longer in the field soil than in the amended soils with vermiculite and
sand. However, the bacteria could more easily wash out from the sandier soil than the
field soil. This observation agreed with tests in different pots where B. glumae colonies
were captured in drained water from pots within a month of starting the experiment.
This may be indicative that leaching played a role in low BPB disease incidence in
experimental plots planted with inoculated seeds during past three years (2013-2015).
Residues for the following tests were collected from the very susceptible entries
of URRN/ARPT plots artificially inoculated with B. glumae in the previous season.
Infected residue buried or kept on the soil surface in the greenhouse tested negative on
CCNT culture medium when tested after four months. The field infectivity test from
B. glumae-infected residue was also negative on CCNT culture. The residue that was
kept in a cloth bag on a greenhouse bench from harvest to planting for about 6 months
tested positive on CCNT media, although the expected color intensity of the toxoflavin
was not as strong as the initial cultures tested immediately after harvest.
Field experiments are underway to test the survival of B. glumae in soil from
harvest to rice planting. The experiment started in October 2015 and will continue until
March 2016. Monthly sampling is currently underway.
Development of Technique to Isolate B. glumae from Soil. Use of CPG another
culture media and CCNT media provided recovery of B. glumae from a gram of
inoculated soil. However, CPG allowed more contaminates to grow especially at lower
dilutions of soil compared to CCNT. SMART medium failed to show any colony growth
in 48 h compared to CCNT. With longer incubation time of a week, The SMART media
produced distinctive bluish colonies ideal for single colony isolation and purification
of B. glumae.
Although carrot and celery were able to detect B. glumae at higher levels, they
were not as good as CCNT and yellow onion with lower levels. Further dilutions were
tested using only CCNT and yellow onion in later repeated experiments. The CCNT
media detected 49 cfu/g soil in sterile soil while detecting up to 3.1 103 cfu/g soil in
natural soil. In natural soil, the growth of other bacteria on CCNT appeared to mask
the visibility of the yellow toxin produced by B. glumae, a characteristic diagnostic
on the medium. A pure B. glumae colony was transferred to a plate where natural soil
contaminates existed to make observation of the capability of B. glumae to grow and
produce the toxin. No color of the toxin was detected after 55 h incubation at 39 C.
115
It is not clear from this study if growth and multiplication of B. glumae is suppressed
or masked by other contaminates. In this study B. glumae formed a sunken lesion on
yellow onion at concentrations of approx. 1.6 103 cfu/g soil. When a small tissue of
onion from the sunken area was transferred to CCNT after 7 days of incubation at 30 C,
bacterial growth and typical B. glumae cultural characteristics were shown. Although,
the vegetable host system needs more work to refine the technique, this preliminary
finding suggested the potential use of onion to compliment CCNT medium to detect
B. glumae from field soil.
Survival Test for B. glumae in Rice Residue. Results of the four preliminary tests
are summarized below: 1) Infected rice residue kept in cloth bags in a greenhouse tested
positive for B. glumae after six months. However, there was a substantial decline in the
B. glumae population on CCNT medium compared to the initial density; 2) Seeds of
Bengal sterilized with oxolinic acid and planted in infected residue mixed in a 1:1 (soil:
residue) ratio in the greenhouse showed decay which resulted in poor germination. The
experiment was modified with oxolinic acid-treated seed planted in 2:1 (soil: residue)
pots that were kept in a growth chamber. In this test, most of the plant samples collected
at different stages tested positive for the bacteria. However, plants from the residue-free
control pots also tested positive for B. glumae rendering inconclusive results; 3) Infected
residue left on the surface or buried in the soil in a greenhouse pot experiment tested
negative for B. glumae after four months; 4) Bengal seeds (susceptible to BPB) planted
and mixed with infected residue showed no symptom of BPB when grown in a field at
the RREC in 2015. RandomLy picked florets from random panicle samples also tested
negative on CCNT culture media. Sample collection and testing for the field experiment
on the survival of B. glumae is underway. Sampling is done in a monthly interval running
from October 2015 through 15 March 2016 representing the duration from rice harvest
to planting. A report will be ready upon completion of the test. Although B. glumae is
mainly seedborne, there is no strong evidence for its lack of ability to inhabit soil or
plant residue. With the continuous rice cultivation practice particularly in zero grade
rice fields of Arkansas, information on longevity and infectivity of these bacteria from
the soil or residue will be useful because currently B. glumae is the primary cause of
BPB disease of rice.
Observations/Tests on Effect of Dew/Mist on Bacterial Panicle Blight Disease

Observation 1. Ninety seven percent of the 70 entries that did not show symptoms
of BPB 10 days after the first spray inoculation showed differing levels of BPB symptoms
during the 48 hours of dew incubation.
Observation 2. Except for the hybrid rice included in the test, 50% of the entries
that were selected as moderately resistant shifted to moderately susceptible when placed
in the artificially elongated dew period.
Preliminary Tests to Evaluate Effect of Dew on Bacterial Panicle Blight. In
2015, a few plants of rice cultivar Wells that were spray-inoculated with B. glumae
suspension at flowering and kept in a dew chamber for 55 hours resulted in fully failed
116
grain with severe symptoms of BPB disease. In contrast, the same variety inoculated in
a similar way and on the same day in a field test during the dry week of July continued
filling grains. Bacterial panicle blight symptoms were not clear until two weeks after
the inoculation date.
The lesion lengths of the 13 rice cultivars which were needle-inoculated at the
7-lf stage and kept intermittently or continually in a dew chamber for up to 3 days did
not show substantial difference from those kept on the bench.
There are clear indications that moisture in the form of dew, mist, fog, and possibly
rain are an important factor in a BPB disease epidemic. The second observation and the
two preliminary tests described above, are in agreement with the field observation in
2012 where the season was hot and dry and BPB was expected in large scale. However,
only two commercial fields, one planted with Jazzman 2 and the other with CL111
were reported. In both fields, BPB incidence was higher close to a waterway or tree
lines suggesting the positive role of longer dew periods on BPB incidence and severity.
The fact that experimental plots established in an open field and planted with
artificially inoculated seeds revealed nearly 100% BPB incidence in a susceptible rice
cultivar, Bengal, within a week after the tropical storm Isaac passed through in 2012
strongly supports the role of wind with rain enhancing the panicle symptoms and/or
spreading of the bacteria within the plots and across the bay. There was severe BPB on
the non-inoculated rice cultivar Wells on the west side of these seed inoculated plots.
Wind direction at the time of the storm was largely from east to west.
Effects of Potassium Fertilizer Rates on Rice Bacterial Panicle Blight

Five potassium fertilizer rates (0, 40, 80, 120, 160 lb K2O/acre) were used in
this test. There was significant difference in yield response to annual K2O fertilizer
rates (Slaton, 2015, pers. comm.). Plots with lower potassium rate had a scant canopy
compared to those that received higher rates showing a positive relationship between
canopy cover and yield. Incidence of BPB disease increased as rice canopy increased
up to the 80 lb K2O/acre rate, and at higher potassium rates the disease was reduced
and then became erratic (Fig. 1). Previous studies on the effect of nitrogen fertilizer
and seeding rates showed that the rice canopy increased with increased nitrogen and
excessive seeding rates (Wamishe et al., 2014; 2015). Likewise, BPB incidence also
increased. It makes sense that plots receiving low rates of potassium had a lower plant
stand resulting in limited BPB incidence. Bacterial spread from plant to plant can be
limited in thin rice plots. It also is plausible that higher potassium rates showed rela-
tively low BPB incidence. Well managed rice plants with adequate levels of nutrition,
particularly potassium, were proven to increase rice plants tolerance to disease. Data
from 2015 were somewhat erratic in BPB response to potassium fertilizer rates and
need to be repeated before any conclusions can be reached. Moreover, there was no
significant difference in disease incidence due to seed inoculation with B. glumae. It
was possible that the bacteria may have been washed down by rain that was frequent
before germination. The field experiment will be repeated at least twice to clearly learn
the relationship between BPB rice disease and potassium fertilizer rates.
117
Managing bacterial panicle blight of rice is very important to reduce potential
yield losses. With lack of resistance in current commercial rice cultivars and absence
of chemical options, cultural management options are immensely important to rice
producers. To effectively manage the disease, understanding the biology of the pathogen
and its host is inevitable. Cultural management options can always be integrated with
host resistance. These studies and findings appear interesting both from a scientific and
practical point of view.
ACKNOWLEDGMENTS
Special appreciation is extended to the Rice Research and Extension Board for
providing funding and support for these projects. Support also provided by the Univer-
sity of Arkansas System Division of Agriculture. Appreciation is also extended to Rick
Cartwright of the Division. Special gratitude goes to Nathan Slaton for his unreserved
cooperation and allowing us to incorporate our potassium test in his field at Pine Tree
Research Station. Many thanks go to Chuck Wilson, Karen Moldenhauer, and Jarrod
Hardke for their positive feedback. We are grateful to Paul Counce and Howard Black
related to growth chamber providence and technical advice.
LITERATURE CITED
Jeong, Y., J. Kim, S. Kim, Y. Kang, T, Nagamatsu, and I. Hwang. 2003. Toxoflavin
produced by Burkholderia glumae causing rice grain rot is responsible for induc-
ing bacterial wilt in many field crops. Plant Disease 87:890-895.
Kawaradani, M., K. Okada, and S. Kusakari. 2000. New selective medium for isola-
tion of Burkholderia glumae from rice seeds. J. Gen. Plant Pathol. 66:234-237.
Kawanishi, T., T. Shiraishi, Y. Okano, K. Sugwara, M. Hashimoto, et al. 2012. New
detection systems of bacteria using highly selective media designed by SMART:
Selective medium-design algorithm restricted by two constraints. PLoS ONE
6(1):e16512. Doi:10.1371/journal.pone.0016512.
Streeter, J.G. 2007. Factors affecting the survival of Bradyrhizobium applied in
liquid cultures to soya bean [Glycine max (L.) Merr.] seeds. J. Appl. Microbiol.
103:1282-1290.
Wamishe, Y.A., T. Gebremariam, S. Belmar, C. Kelsey, and T. Mulaw. 2015. Stud-
ies on cultural management options for rice bacterial panicle blight disease. In:
R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice Research
Studies 2014. University of Arkansas Agricultural Experiment Station Research
Series 626:118-126. Fayetteville.
Wamishe, Y.A., Y. Jia, T. Gebremariam, C. Kelsey, S. Belmar, and T. Mulaw. 2014.
Characterization of rice germplasm for genetic resistance to bacterial panicle
blight disease and development of techniques for monitoring the bacterial patho-
118
gen. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice
Research Studies 2014. University of Arkansas Agricultural Experiment Station
Research Series 617:133-142. Fayetteville.
Table 1. Serial dilutions and approximate colony

forming units of Burkholderia glumae for inoculated soil.
Dilution Approx. cfu/g soil Dilution Approx. cfu/g soil
99 OD 400,000 1/256 1600
1/2 200,000 1/512 781
1/4 100,000 1/1,024 391
1/8 50,000 1/2,048 195
1/16 25,000 1/4,096 98
1/32 12,500 1/8,192 49
1/64 6200 1/16,384 12
1/128 3100 1/32,768 6
Fig. 1. Number of bacterial panicle blight diseased

panicles found in the plots of the rice cultivar Bengal for varying soil
potassium levels for both the inoculated and non-inoculated Bengal seed.
119
Monitoring Bacterial Panicle Blight Disease of Rice and

Germplasm Evaluation for Resistance in Arkansas in 2015
Y.A. Wamishe1, T. Mulaw1, Y. Jia2, T. Gebremariam1, S.B. Belmar3, and C.D. Kelsey3
ABSTRACT
Rice is a major cereal crop that contributes significantly to global food security. Rice
can be affected by both abiotic and biotic stresses. Rice bacterial panicle blight (BPB)
has been recognized as one of the major biotic factors that can cause severe yield loss
in southern U.S. rice-growing states and several other rice-growing countries. Cultural
practices such as early planting with adequate seed and nitrogen fertilizer rates have
been experimentally shown to reduce the disease incidence. To date, chemical options
are not available for use in the U.S. None of the current conventional commercial rice
cultivars appear to have complete resistance to BPB. Hybrid rice and Jupiter have
shown moderate resistance under field conditions in the epidemic years of 2010 and
2011. Breeding for disease resistance requires continuous efforts by breeders to enrich
their gene pools to effectively tackle yield robbing problems. In the past few years,
research efforts have been focused on understanding the causal bacterial species of
panicle blight and evaluating rice for resistance to BPB. In 2015, a total of 165 panicle
samples that were either fully or partially blank with or without floret discoloration
were collected from 9 rice-producing counties. Samples were collected largely, from
the Uniform Regional Rice Nursery (URRN), Arkansas Rice Performance Trials
(ARPT), or Producer Rice Evaluation Program (PREP) across Arkansas. Of the165
field samples collected, 72 samples were considered positive for B. glumae visually on
CCNT culture medium. However, the molecular approach using classical polymerase
chain reaction (PCR) with specific primers confirmed only 45 as positive to B. glumae.
When these 45 isolates were tested with a B. gladioli primer, none tested positive for B.
gladioli suggesting B. glumae as the major causal agent of BPB in these samples. Two
rice germplasm nurseries, namely ARPT and URRN consisting of 90 and 200 entries,
respectively were evaluated for BPB using artificial inoculation in an experimental
field at the University of Arkansas System Division of Agricultures Rice Research
1
Assistant Professor, Program Technician - Plant Pathology, and Program Associate - Plant Pathology,
respectively, Department of Plant Pathology, Stuttgart.
2
Research Plant Pathologist, Dale Bumpers National Rice Research Center, Stuttgart.
3
Program Technician and Program Technician II, respectively, Rice Research and Extension Center,
Stuttgart.
120
and Extension Center (RREC) near Stuttgart, Ark., in 2015. A 0 to 9 disease rating
scale was used where 0 is no disease and 9 severe disease with blank panicles. Based
on the reference reactions of Jupiter (MR) and Bengal (S), 27 entries were categorized
as R (resistant), 34 MR (moderately resistant), 36 MS (moderately susceptible), 126
S (susceptible), 62 VS (very susceptible), and none were rated zero (immune/highly
resistant). Late-maturing rice entries appeared relatively clean and were placed in the
resistant category. However, further investigation is required to ensure true resistance
to BPB in these late-maturing rice entries. False resistance response or disease escape
can possibly be due to the unfavorable weather conditions during and after inoculation
as the season tapered off.
INTRODUCTION
Rice constitutes the staple diet for more than 50% of the worlds population in
terms of cultivation and consumption (FAO, 2012). It is part of the everyday diet of
many households. Rice production and consumption are concentrated in Asia, other
parts of the world such as South America, Africa, Australia, some parts of Europe, and
southern United States (FAO, 2012). Rice production is constrained by many abiotic
and biotic factors among which bacterial panicle blight (BPB) has been ranked recently
as a major threat (Shahjahan et.al., 2000).
Bacterial panicle blight is caused by the gram-negative bacterial pathogens
Burkholderia glumae and B. gladioli and probably a couple others. Symptoms include
panicle discoloration, grain rot, and sterile florets. Bacterial panicle blight is favored by
prolonged high night temperatures during the heading and flowering stages (Nandakumar
et al., 2009). B. glumae is considered by and large as seedborne bacterium. It produces
yellow-colored phytotoxin (toxoflavin) that has been proved to be its major virulence
factor (Sato et al., 1989). To date, complete rice host resistance for BPB has not been
reported. Oxolinic acid is the only known commercial chemical agent for controlling
this disease. However, chemicals such Oxolinic acid and Kasugamycin used in some
countries are not labeled for U.S., and reports on antibiotic resistance limits their usage
(Hikichi et al., 2001). Our study in 2015, therefore, focused mainly on the distribution
of BPB in nine rice-producing counties of Arkansas and rice germplasm evaluation for
BPB resistance under field condition using artificial inoculation. Results from molecular
marker and greenhouse evaluations will be reported once tests are completed.
PROCEDURES
Isolation and Identification of Burkholderia Species from Arkansas Rice
One hundred sixty-five rice panicle samples that either showed some level of
blanking or brown floret discoloration were collected from nine rice-growing counties in
Arkansas. The counties included: Prairie, Arkansas, Lincoln, Desha, Clay, Mississippi,
Craighead, Jackson, and Woodruff. Samples were mainly collected from the counties
with field plots of the Uniform Regional Rice Nursery (URRN), Arkansas Rice Perfor-
121
mance Trials (ARPT), and Producer Rice Evaluation Program (PREP). About 100 florets
from each sample were randomly picked and kept at 4 C in a coin envelope until plated
on a culture medium. CCNT, a semi-selective medium for B. glumae (Kawaradani et
al., 2000) was used to plate the seeds. Cultures were then incubated at 39 C for 48 h
in the dark. Bacterial colonies that showed similar morphology to B. glumae on culture
were isolated and purified. DNA was extracted from pure cultures of each isolate and
a polymerase chain reaction (PCR)-based method was used to identify the isolate with
B. glumae and B. gladioli specific primers (Yukiko et al., 2006).
Evaluation of Rice for Resistance Against Bacterial Panicle Blight Disease

In 2015, the ARPT and URRN consisting of 90 and 200 entries, respectively,
were evaluated for BPB. The entries were tested using artificial inoculation under field
conditions at the RREC following the procedure in Wamishe et al. (2013). Hill plots
were used to plant two replicates of 290 lines interspaced with Jupiter and Bengal after
each 10 entries. Another bay was planted similarly to serve as a non-inoculated check.
Jupiter and Bengal were included as known references for moderately resistant and
susceptible reactions to BPB, respectively. Inoculation for each entry was carried out
between the boot-split to flowering growth stage twice in an interval of 4 days to reach
the panicles on both the primary and secondary tillers. Spray-inoculation was targeted
to the panicles. A 48-h old B. glumae culture grown on Kings B medium at 39 C was
used to produce the pathogen suspension. The suspension was ~106 to 108 cfu/mL (colony
forming units/milliliter). Disease reactions were evaluated four weeks after the last
inoculation using a 0 to 9 scale, where 0 is no disease and 9 is severe disease (Table 1).

Isolation and Identification of Burkholderia Species from Arkansas Rice in 2015
Of 165 field samples collected, 72 samples were considered positive for B. glu-
mae visually on CCNT culture medium. However, the molecular approach using PCR
with specific primers confirmed only 45 as positive for B. glumae. When tested with
a B. gladioli primer, none tested positive for B. gladioli indicating B. glumae was the
major causal agent of BPB in Arkansas. Based on this study, the largest proportion of
samples tested negative for B. glumae suggesting that there were other causes for panicle
blanking and discoloration during the hot and dry weeks of July and August in 2015.
These results also suggest that molecular technique are useful in verifying visual and
cultural identification of B. glumae.
Evaluation of Rice for Resistance Against Bacterial Panicle Blight Disease

None of the 290 rice entries tested were immune to BPB disease under field condi-
tions using artificial inoculation. Based on the reference reactions of Jupiter (MR) and
Bengal (S), 27 entries were categorized as R (resistant), 34 MR (moderately resistant),
36 MS (moderately susceptible), 126 S (susceptible), and 62 VS (very susceptible)
122
(Table 2). Late-maturing rice varieties appeared relatively clean, and were placed in the
resistant category (Tables 3 and 4). However, further tests are required to ensure true
resistance to BPB in late-maturing rice. Disease escape or a false resistance response
can possibly be due to the unfavorable weather conditions during and after inocula-
tion as the season tapered off. On the other hand, as reported in Pinson et al. (2010),
late maturity may be a confounding factor for genetic resistance to B. glumae. Further
testing with molecular markers coupled with greenhouse testing is recommended for
evaluation of genetic resistance to B. glumae for late-maturing rice. Note that the late-
maturing commercial rice cultivar Roy J was susceptible to BPB during the epidemic
years. However, in this test, Roy J was resistant indicating that it was a possible outcome
of environmental effects. Overall, the resistance groups from this and previous seasons
are encouraging for the identification of resistant rice cultivars.
Results from other on-going laboratory molecular marker and greenhouse resis-
tance evaluation activities will be presented at the completion of the tests.
Rice resistance to BPB would provide long-term control in years of increased
disease pressure and thus improve yield. Development of a better toolbox to evaluate
genetic resistance remains to be an important priority for combating BPB disease in
rice. The preliminary surveys of Burkholderia species across Arkansas are encouraging
suggesting that the major causal organism is primarily one species of Burkholderia.
Extensive surveys with different molecular markers will be needed to evaluate diseased
plants from commercial rice fields particularly in epidemic years. Efforts to understand
virulence, pathogenicity, and epidemiology of the Burkholderia pathogen must continue
in order to identify more effective control means to manage BPB.
ACKNOWLEDGMENTS
The support of Extension rice pathology research by the rice producers of Arkan-
sas through monies administered by the Arkansas Rice Research and Promotion Board
and funding from the University of Arkansas System Division of Agriculture is highly
appreciated. We are grateful to Rick Cartwright of University of Arkansas Division of
Agriculture for his valuable advice, Don Groth of Louisiana State University, and C.
Wilson, Karen Moldenhauer, and J. Hardke of University of Arkansas for their col-
laboration. We appreciate USDA Agricultural Research Service Dale Bumpers National
Rice Research Center for continued cooperation.
LITERATURE CITED
FAO. 2012. Food and Agriculture Organization of the United Nations. Rice Market
Monitor, Trade and Markets Division 15(1):33 pp. Available at: www.fao.org/
docrep/015/an459e/an459e00.pdf
123
Hikichi Y, K. Tsujiguchi, Y. Maeda, and T. Okuno. 2001. Development of increased

oxolinic acid-resistance in Burkholderia glumae. J Gen Plant Pathol. 67:58-62.
Kawaradani, M., K. Okada, and S. Kusakari. 2000. New selective medium for isola-
tion of Burkholderia glumae from rice seeds. J. Gen. Plant Pathol. 66:234-237.
Maeda, Y. H. Shinohara, A. Kiba, K. Ohnishi, N. Furuya, Y. Kawamura, T. Ezaki, P.
Vandamme, S. Tsushima, and Y. Hikichi. 2006. Phylogenetic study and multiplex
PCR-based detection of Burkholderia plantarii, Burkholderia glumae, and Burk-
holderia gladioli using gyrB and rpoD sequences. Int. J. System. Evolut. Micro-
biol. 56:1031-1038. DOI 10.1099/ijs.0.64184-0
Nandakumar, R., A.K.M. Shahjahan, X.L. Yuan, E.R. Dickstein, D.E. Groth, and
C.A. Clark. 2009. Burkholderia glumae and B. gladioli cause bacterial panicle
blight in rice in the southern United States. Plant Dis. 93(9):896-905.
Pinson, S.R.M., A.K.M. Shahjahan, M.C. Rush, and D.E. Groth. 2010. Bacterial
panicle blight resistance QTLs in rice and their association with other disease
resistance loci and heading date. Crop Sci. 50:1287-1297. doi: 10.2135/crop sci.
2008.07.0447.
Sato, Z., Y. Koiso, S. Iwasaki, I. Matsuda, and A. Shirata. 1989. Toxins produced by
Pseudomonas glumae. Ann Phytopathol Soc. J. 55(3):353-356.
Shahjahan, A.K.M, M.C. Rush, D. Groth, and C.A. Clark. 2000. Panicle blight. Rice
J. 15:26-29.
Wamishe, Y., Y. Jia, C. Kelsey, S. Belmar, and M. Rasheed. 2013. Development of
panicle diagnostic methods for monitoring rice bacteria panicle blight disease and
evaluation of rice germplasm for resistance. In: R.J. Norman and K.A.K. Mold-
enhauer (eds.). B.R.Wells Rice Research Studies 2012. University of Arkansas
Agricultural Experiment Station Research Series 609:103-108. Fayetteville.
Yukiko M., H. Shinohara, A. Kiba, K. Ohnishi, N. Furuya, Y. Kawamura, T. Ezaki, P.
Vandamme, S. Tsushima. and Y. Hikichi. 2006. Phylogenetic study and multiplex
PCR-based detection of Burkholderia plantarii, Burkholderia glumae and Burk-
holderia gladioli using gyrB and rpoD sequences. Int. J. System. Evol. Microbiol.
56:10311038. DOI 10.1099/ijs.0.64184-0
Table 1. The 0 to 9 disease rating scale used to evaluate

rice reactions to bacterial panicle blight under field conditions sprayed with
Burkholderia glumae between late-boot to flowering stage of rice development.
0-9 scale Reaction group 0-9 scale Reaction group
0 Immune 5 Moderately resistant (MR)
1 Resistant (R) 6 Moderately susceptible (MS)
2 Resistant (R) 7 Susceptible (S)
3 Resistant (R) 8 Susceptible (S)
4 Moderately resistant (MR) 9 Very susceptible (VS)
124
Table 2. Number of rice entries grouped in different relative resistance

categories for rice bacterial panicle blight disease using a 0 to 9 rating scale in 2015.
Reaction group URRNa ARPT Total
Highly resistant 0 0 0
Resistant 19 (8)b 8 (6) 27 (14)
Moderately resistant 30 4 35
Moderately susceptible 32 4 36
Susceptible 84 42 126
Highly susceptible 33 29 62
Total 198 88 186
a
URRN = Uniform Regional Rice Nursery; ARPT = Arkansas Rice Performance Trials.
b
Numbers in parentheses indicate number of late-maturing entries.
Table 3. Rice entries in the resistant group from Uniform Regional Rice
Nursery (URRN) to bacterial panicle blight (BPB) disease of rice in 2015.
Entry no. Accession BPB disease score Maturity
URRN-20 MRMT 2
URRN-23 RU1503023 1
URRN-25 RU1402125 2
URRN-33 RU1304156 1
URRN-35 RU1304122 2
URRN-61 RU1501061 2
URRN-63 RU1003153 2
URRN-68 RU1502068 2
URRN-71 RU1502071 2
URRN-79 ROYJ 1 Late
URRN-80 MM14 1 Late
URRN-91 RU1502091 1
URRN-92 RU1503092 2
URRN-108 RU1501108 2 Late
URRN-160 TGRT 1 Late
Table 4. Rice entries in the resistant group from

Arkansas Rice Performance Test (ARPT) to
bacterial panicle blight (BPB) disease of rice in 2015.
Entry no. Accession BPB disease score Maturity
ARPT-16 RTCLXL729 1
ARPT-17 RTCLXL745 2
ARPT-42 RU1501139 1 Late
ARPT-47 STG11F3-04-065 1 Late
ARPT-52 STG11P-23-073 1 Late
ARPT-56 STG12L-48-213 2 Late
ARPT-57 RU1501047 1 Late
ARPT-60 STG12L-47-256 2 Late
125
Understanding Autumn Decline, Evaluating

Rice Varieties for Resistance, and Identifying New
Strategies to Reduce the Phenomenon in Problematic Rice Fields
Y.A. Wamishe1, T.L. Roberts2, J.T. Hardke3, J. Allen2,
T. Gebremariam1, C.D. Kelsey4, T. Mulaw4, and S.B. Belmar4
ABSTRACT
Autumn decline and hydrogen sulfide toxicity have similar symptomology and are
often referred to as the same thing even if they are two different problems in rice fields.
Symptoms include black root rotting with stunted and yellowish rice foliage starting as
early as two weeks following permanent flood establishment. To understand the primary
causes of autumn decline in Arkansas, tests on redox of different soils that have had
autumn decline problem are underway. To search for practical methods to prevent or
reduce the problem, preliminary greenhouse pot experiments have been underway to
test three oxidizing agents and two bio-products (microbial oxidizers) using field soils
from two different fields in Woodruff County. To evaluate rice for degree of resistance
or tolerance to autumn decline under field conditions, a Producers Rice Evaluation
Program field trial consisting of 20 commercial cultivars was planted in 2015. To
prove the effect of soil drainage on autumn decline severity and cultivar survival rate,
greenhouse tests were carried out. Although inconsistent across replications, the field
study indicated various levels of susceptibility among the cultivars tested. From the
cultivar test data, a rating matrix scale was developed that combined root blackening and
crown blockage or discoloration. Greenhouse tests on practical methods to alleviate the
problem were not conclusive. Preliminary greenhouse soil drainage tests showed new
root growth within four days. Since this is the first study year, results are considered
too premature to draw conclusions.
1
Assistant Professor and Program Associate - Plant Pathology, respectively, Department of Plant Pathol-
ogy, Stuttgart.
2
Assistant Professor and Graduate Assistant, respectively, Department of Crop, Soil, and Environmental
Science, Fayetteville.
3
Research Extension Agronomist, Department of Crop, Soil, and Environmental Science, Stuttgart.
4
Program Technician II, Plant Pathology Program Technician, and Program Technician II, respectively,
126
INTRODUCTION
Autumn decline and hydrogen sulfide toxicity, also referred as akiochi, have
similar symptomology and are often referred to as the same thing even though they are
two different problems in rice fields. Symptoms include black root rotting with stunted
and yellowish rice foliage starting as early as two weeks following permanent flood
establishment. The problem is often most severe where cold well water first enters a
rice field and may spread throughout the field, except on levees. The phenomenon was
reported in Arkansas in a limited number of fields in 2004 (Delta Farm Press, 2004;
Wilson and Cartwright, pers. comm.). However, there were several reports of autumn
decline across the state of Arkansas in 2012, 2013, and 2014, and even more in 2015.
Although the problem may be aggravated in the anaerobic/flooded situation, there is no
clear understanding of why this phenomenon is occurring in different soil types across
several rice-growing counties in Arkansas. Observations have shown fields having a
clay loam soil texture are more prone to the autumn decline phenomenon than others
commonly cropped to rice. The root rotting symptoms often start a few weeks after flood
establishment and become progressively worse throughout the season if unmitigated. In
situations where root rotting is severe, fungi grow into the crown which limits function
of the whole root system and prevents translocation of water and nutrients from the soil
to the plant. In moderate to severe cases, tillers break off easily and plant death may
occur rapidly leading to significant yield losses. Ongoing field and greenhouse investi-
gations that started in 2015 have the following objectives: 1) to understand the primary
causes of autumn decline in Arkansas; 2) to search for practical methods to prevent or
correct the root blackening and rotting associated with autumn decline; 3) to evaluate
the degree of resistance or tolerance of common rice cultivars to autumn decline under
greenhouse and field conditions; and 4) to evaluate the effect of soil drainage (the cur-
rent preventative/rescue strategy) on autumn decline severity and cultivar survival rate.
PROCEDURES
To understand the primary causes of autumn decline in Arkansas, five fields known
to have a history of autumn decline were identified, with varying levels of disease his-
tory and severity. Soil samples from these locations have been collected, sterilized and
flooded to study redox differences between sterilized and non-sterilized soils across
soil types and locations. Redox potential is the inherent tendency of a compound to act
as an electron donor or electron acceptor as defined by Fuhrmann (1999). This portion
of the research is focused on identifying the soil chemical and physical attributes that
contribute to the occurrence of autumn decline. Development of protocols to identify
areas/fields that are prone to autumn decline and preventative measures that are more
effective in preventing yield loss associated with the disease are anticipated.
To search for practical methods to prevent or correct the root blackening and
rotting associated with autumn decline, preliminary greenhouse pot experiments have
been underway to test three oxidizing agents and two bio-products (microbial oxidizers)
using field soils from two different fields in Woodruff County. Both fields were known
127
to have a history of autumn decline in previous years. Soil from the Hillemann, Ark.,
area appeared light in color with visually noticeable iron content and was collected at
the beginning of October 2015, approximately two months after rice crop harvest. Soil
from the Hunter area appeared darker in color and was collected at the end of October
2015, about a month after the soybean harvest. The first experiment was conducted using
the soil collected from the Hillemann area. The second experiment is currently running
using the soil collected from Hunter area. Big plant residue materials were sieved out
and removed from the soil prior to the initiation of the trials. In both experiments, a very
susceptible rice cultivar Mermentau was used based on data collected from the 2015
field test. Taking into account the delayed and erratic start of symptoms in the first and
second experiments, pots for the third experiment were flooded for three weeks before
planting. CL151, another known susceptible variety for autumn decline was used for
the third experiment. Pre-germinated seeds were planted in muddy soil and pots were
left wet to keep the soil somewhat anaerobic. Pots were flooded to about a 2-in. depth
after the seedlings received their first nitrogen application at 5-lf stage and kept flooded.
All experimental pots were kept flooded throughout the experiment with refrigerated
water at 4 C until flowering. Treatments included: 1) a product which claimed to have
sulfur-philic anaerobic bacteria; 2) a product which claimed to have photosynthetic and
sulfur-philic bacteria; and three other oxidizing compounds, namely KNO3, H202, and
KMNO4. Two control treatments were included, one with continuous flood and the other
with intermittent flushing. The test was designed to be conducted in three replications.
The test will be repeated at least twice following the procedure that renders the best
desired symptoms. Field tests will be conducted possibly in 2017 using products that
may provide the best protection from autumn decline.
To evaluate rice for degree of resistance or tolerance to autumn decline under
field conditions, a Producers Rice Evaluation Program (PREP) field trial consisting of
20 commercial cultivars was planted in 2015 (Table 1). The trial was planted in four
replications in a field that had a history of autumn decline near Hillemann, Ark., in
Woodruff County with a plot size of 8 rows on 7-in. spacing, 15 ft in length. When the
early maturing cultivars were flowering, roots were pulled from the north-side outer
row of all plots, washed immediately, and rated for both root and crown discoloration
using a 0 to 5 and 0 to 9 rating scale, respectively. From these data, a disease matrix
was developed that combined the two ratings for the length of time they were under
flood. Note that these cultivars may not have similar heading and maturity dates. The
later the cultivar, the more flood exposure duration. The rating scale will be evaluated
in the years to come in both greenhouse and field studies.
To evaluate the effect of soil drainage on autumn decline severity and cultivar
survival rate, a preliminary greenhouse experiment was conducted back in 2013 using
field soil from the Hunter area in Woodruff County. The soil was collected immediately
prior to harvest. Pre-germinated seeds of 10 cultivars were planted without removing
the rice residue and pots were kept flooded with refrigerated water. Plants were drained
around the vegetative stage V11 (Counce et al., 2000), left to dry for four days and
then re-flooded. Similar greenhouse tests and different field tests will be carried out in
the upcoming years.
128

Some tests are currently being conducted to understand the primary causes of
autumn decline in Arkansas, and data was not available at the time of this report.
The first experiment using iron-rich soil from Hillemann, Ark., produced up to
50% blackening in some of the plants from continuously flooded control pots. Although
there were some indications of activity to the tested products, it appeared that the soil and
rice plants were not exposed to anaerobic conditions enough to give uniform symptoms
to all plants in the control. Plants were pulled to collect data after the rice reached the
flowering stage. Since the soil appeared to have high iron levels, and such soils may
require a longer duration of flooded conditions to reach an anaerobic state, results from
the first test are considered inconclusive. The second and the third experiments were
underway during the time of this reporting.
The field study to evaluate degree of resistance or tolerance of common rice
cultivars to autumn decline showed potential differences among cultivars. Rating in-
formation from this study was used to select potentially more susceptible cultivars for
further evaluation in greenhouse trials. However, due to inconsistencies in some of the
cultivars within replications, data are not shown for this reporting. The inconsistency
within replications maybe due to variation in flood depth within the field.
The fact that autumn decline in this field was recorded using two rating scales (a 0
to 5 and a 0 to 9) it was deemed necessary to develop a matrix index that best described
the severity level in each cultivar for comparison. Twenty rice cultivars were evaluated
in the field trial (Table 1) using descriptive rating scales (Table 2) to quantify the extent
of damage that was then summarized as a matrix scale (Table 3). In developing the rat-
ing matrix for autumn decline, crown discoloration has been given double the weight
compared to root blackening due to its higher effect on grain yield loss. Observations
and experiments have shown crown discoloration as irreversible damage compared to
the often reversible symptom of root blackening.
Preliminary tests in 2013 indicated both root blackening and crown discoloration as
reproducible in a greenhouse using soils with a history of autumn decline. Ten cultivars
tested in a greenhouse indicated cultivar differences in tolerance (data not shown). Root
vigor and oxidation power of roots may play a role in cultivar tolerance. However, the
experiment needs to be repeated to confirm these results.
In a preliminary greenhouse experiment to evaluate the effect of soil drainage on
autumn decline severity and cultivar survival rate, pots were drained and left to dry for
four days. They showed new root development just above the base of the crown. These
experiments need to be repeated in the greenhouse and tested in the field to determine
the length of time required for new roots to grow and root colors to reverse in different
soils types and environments/locations. Until then, these will be considered premature
results even though they are in agreement with field observations which showed new
root growth and the positive effect of draining and drying out the soil.
The idea behind the drain and dry strategy is to allow oxygen into the rhizosphere
to reaerate the soil and prevent hydrogen sulfide production. However, this strategy is
not compatible with limited water resources and can be costly, especially in large fields
129
where draining and re-flooding each take several days to complete. As a result the risk
of yield loss due to drought stress could be high. Therefore, searching for alternative
management options would be very beneficial.
From 2012 to 2015 root blackening and crown rotting, known as autumn decline
or often referred to as hydrogen sulfide toxicity of rice, appeared to be increasing
compared to the eight previous years. Although hydrogen sulfide toxicity is considered
the cause, a better understanding of this phenomenon is needed. Autumn decline and
hydrogen sulfide toxicity are often referred to as the same thing even if they are two
different problems in rice fields. In some fields, draining surface flooded water improved
the situation. However, in other fields the drain and dry approach did not improve the
situation enough to salvage the crop. A better understanding of this problem would
permit growers to make the best decisions possible to avoid losses due to the failure of
the drain and dry strategy. Additionally, the drain and dry approach does not work if a
field is not a manageable size. Knowledge of cultivars susceptibility/intolerance and
the discovery of additional management options could have prevented the significant
losses that have occurred to some rice fields in previous seasons.
ACKNOWLEDGMENTS
The extension rice pathology program appreciates the funding and support from,
the Arkansas rice producers through monies administered by the Arkansas Rice Research
and Promotion Board and support from the University of Arkansas System Division of
Agriculture. We are grateful to Rick Cartwright of the University of Arkansas for his
technical help. Our appreciation goes to rice producers that allowed us to use land and
soil to research. We do appreciate the Arkansas County agent Grant Beckwith for his
information on one of the products that we are testing and Dr. Anita Kelly for provid-
ing the product.
LITERATURE CITED
Counce, P.A., T.C. Keisling, and A.J. Mitchell. 2000. A uniform, objective and adap-
tive system for expressing rice development. Crop Sci. 40:436-443.
Delta Farm Press Sept 24, 2004 at http://deltafarmpress.com/rice-malady-identified-
hydrogen-sulfide-toxicity
Fuhrmann, J.J. 1999. Microbial metabolism. pp. 189-217. In: D.M. Sylvia et al.
(eds.). Principles and applications of soil microbiology. Upper Saddle River, N.J.:
Prentice Hall, Inc.
130
Table 1. Rice cultivars tested for autumn decline/hydrogen sulfide toxicity

resistance/tolerance in a field in Hillemann, Ark., at Woodruff County in 2015.
Rice cultivars tested
1 LaKast 6 CL111 11 CL153 16 RTXL753
2 Roy J 7 CL151 12 CL172 17 RTXL760
3 Taggart 8 CL163 13 CLX1102 18 Titan
4 Mermentau 9 CL271 14 RTCLXL729 19 Diamond
5 Jupiter 10 CL272 15 RT CLXL745 20 MSX4077
Table 2. Rating scales used to rate crown

discoloration and root discoloration in cultivars grown in soil with
history of autumn decline/hydrogen sulfide toxicity at Hillemann, Ark., in 2015.
% Crown length % Root mass
0 to 9 scale discoloreda 0 to 5 scale blackeneda
(%) (%)
0 0 0 Clean as in
levee roots
1 10 1 10
2 20 2 25
3 30 3 50
4 40 4 75
5 50 5 75 or >
6 60
7 70
8 80
9 90 or >
a
Roots need to be washed well and rated immediately, up to 10 root crowns need to be exam-
ined. Numbers shown under % columns refer to range of estimate. For instance: 10 refers to
discoloration percentage up to 10.
Table 3. Matrix to rate incidence and severity of

autumn decline/hydrogen sulfide toxicity in rice cultivars.
% Root blackening aligned with a 0 to 5 scale
% Crown 0 to 9 2X 0 10% 25% 50% 75 >
infection scale (0-9)a 0 1 2 3 4 5
0 0 0 R R R MR MR MS
10 1 2 R MR MR MS MS S
20 2 4 MR MS MS S S VS
30 3 6 MS S S VS VS VS
40 4 8 S VS VS VS VS VS
50 5 10 VS VS VS VS VS VS
a
The 0-9 scale was multiplied by 2 to give more weight to crown infection as it is the more seri-
ous and irreversible problem than the root blackening. Root crown is the upper part of the main
root system.
131
Efficacy of Selected Insecticides for Control of

Rice Stink Bug, Oebalus pugnax, in Arkansas, 2015
H.M. Chaney1, G.M. Lorenz2, N.M. Taillon2, W.A. Plummer2, and J.L. Black2
ABSTRACT
A trial was conducted to evaluate selected insecticides for the control of rice stink bug,
Oebalus pugnax. Results indicated that all treatments reduced rice stink bug below the
untreated check up to 13 days after application.
INTRODUCTION
Rice stink bug is a common and important pest in Arkansas rice. In the spring,
rice stink bugs feed and reproduce on a wide range of wild grasses. This enables the
rice stink bug to reproduce and increase in numbers before cultivated host plants are
available. Rice stink bugs normally do not occur in rice fields until heading has begun,
but may occur earlier if heading of wild grasses is present in or around field edges.
Early feeding from pre-fertilization through early milk stage causes the heads to blank or
abort resulting in yield reduction. Feeding during the milk-to-soft dough stage results in
kernel shrinkage or slight discoloration commonly referred to as pecky rice (Johnson
et al., 2002). This can result in deductions in quality or grade. The use of insecticides
gives producers the ability to lower rice stink bug numbers. When populations are at
moderate levels, a single insecticide application may be used to control rice stink bugs.
When populations are high, multiple applications may be required to achieve control
(Plummer et al., 2015; Thrash et al., 2012). Finding alternative insecticides is necessary
to reduce potential for resistance. A trial was conducted to evaluate current and potential
foliar treatments for control of rice stink bug.
PROCEDURES
The trial was conducted near Colt, Ark., at the University of Arkansas System
Division of Agricultures Pine Tree Research Station. Plot size was 15 ft 35 ft in a
1
Associate Director, Agriculture and Natural Resources, Cooperative Extension Service, Conway.
2
Associate Department Head / Extension Entomologist, Program Associate I - Entomology, Program
Associate - Entomology, and Graduate Assistant, respectively, Department of Entomology, Lonoke.
132
randomized complete block design with four replications. Foliar treatments included:
Endigo ZCX (5 oz/acre and 6 oz/acre); Karate Z (2.56 oz/acre); Tenchu (9 oz wt/acre);
Strafer (3.5 oz/acre); and A21120A (15.5 oz/acre). All treatments were compared to an
untreated check (UTC). Insecticide treatments were applied with a hand boom on 21
August. The boom was fitted with TX6 hollow cone nozzles at 19-inch nozzle spacing;
spray volume was 10 gal/acre, at 40 psi. Insect counts were taken at 3, 7, 11, and 13 days
following treatment by taking 10 sweeps per plot with a standard sweep net (15-inch
diameter). Data was processed using Agriculture Research Manager Version 9, analysis
of variance, and Duncans New Multiple Range Test (P = 0.10) to separate means.

At 3 days after application (DAA), all treatments reduced rice stink bug numbers
below the UTC; Endigo ZCX (6 oz/acre) had fewer rice stink bugs than A21120A,
Strafer, and Endigo ZCX (5 oz/acre) (Fig. 1). By 7 DAA, rice stink bug numbers in-
creased in the UTC; however in all treatments, rice stink bug numbers went down from
3 DAA; Endigo ZCX (6 oz/acre) and A21120A had fewer rice stink bugs than all other
treatments (Fig. 2). At 11 DAA, all treatments had fewer rice stink bugs than the UTC;
A21120A had fewer stink bugs than all other treatments; Endigo ZCX (5 oz/acre) had
fewer stink bugs than Karate Z and Tenchu (Fig. 3). At 13 DAA, all treatments had fewer
stinkbugs than the UTC but no difference between treatments was observed (Fig. 4).
Rice stink bug is one of the most damaging pests in Arkansas rice, it not only
affects yield but can affect the quality of the rice. We will continue to conduct studies
to find economic ways for growers to control rice stink bug.
ACKNOWLEDGMENTS
The authors wish to express their appreciation for funding and support for this
project from the rice producers of Arkansas through the monies administered by the
Division of Agriculture, Syngenta, Gowan, and Mitsui. We would also like to acknowl-
edge Shawn Clark, director, and his crew at Pine Tree Research Station for their help
with this trial.
LITERATURE CITED
Johnson, D.R., J. Bernhardt, J. Greene, G. Lorenz, and G. Studebaker. 2002. Rice stink
bug in Arkansas. University of Arkansas System Division of Agriculture Coopera-
tive Extension Service Fact Sheet FSA 7057. Available at: www.uaex.edu/publica-
tions/PDF/FSA-7057.pdf
133
Plummer, W.A., G.M. Lorenz, N.M. Taillon, H.M. Chaney, B.C. Thrash, D.L. Clark-
son, M.E. Everett, and L.R. Orellana Jimenez. 2015. Efficacy of selected insecti-
cides for control of rice stink bug, Oebalus pugnax, in Arkansas 2011-2013. In:
R.J. Norman and K.A.K. Moldenhaur (eds.). B.R. Wells Arkansas Rice Research
Thrash, B.C., G.M. Lorenz, N.M. Taillon, W.A. Plummer, C.K. Colwell, and J.W.
Fortner. 2012. Efficacy of selected insecticides for control of rice stink bug,
Oebalus pugnax, in Arkansas. In: R.J. Norman and K.A.K. Moldenhaur (eds.).
B.R.Wells Rice Research Studies 2011. University of Arkansas Agricultural Ex-
periment Station Research Series 600:138-141. Fayetteville, Ark.
Fig. 1. Rice stink bug counts taken 24 August, 3 days after application (DAA). UTC =
untreated check. Means followed by same letter do not significantly differ (P = 0.10,
Duncan's New Multiple Range Test). Mean comparisons performed only when analysis of
variance Treatment P (F) is significant at mean comparison observed significance level.
134
Fig. 2. Rice stink bug counts taken 28 August, 7 days after application (DAA). UTC =
untreated check. Means followed by same letter do not significantly differ (P = 0.10,
Fig. 3. Rice stink bug counts taken 1 September, 11 days after application (DAA). UTC
= untreated check. Means followed by same letter do not significantly differ (P =0.10,
135
Fig. 4. Rice stink bug counts taken 3 September, 13 days after application (DAA). UTC
= untreated check. Means followed by same letter do not significantly differ (P = 0.10,
136
Evaluation of Rice Kernel Damage and Yields Due to Rice

Stink Bug, Oebalus pugnax, Population and Infestation Timing
T.L. Clayton1, G.M. Lorenz2, J.T. Hardke3, G.J. Lee4,
E. Castaneda-Gonzalez4, D.L. Frizzell4, and H.M. Chaney2
ABSTRACT
The rice stink bug, Oebalus pugnax, is an important pest to rice that can reduce rough
rice and milling yields. Field cage studies were conducted in 2015 at the University
of Arkansas System Division of Agricultures Rice Research and Extension Center
near Stuttgart, Ark., to evaluate the effect of different populations of rice stink bugs
and infestation timings on kernel damage and grain yields. Our study did not show a
reduction in rough rice yield or milling yields in any population or infestation timing.
In the milk infestation timing, there was an increase in damaged kernels due to higher
rice stink bug population densities.
INTRODUCTION
The rice stink bug, Oebalus pugnax, feeds on rice, Oryza sativa L., from head-
ing to hard dough in the southern United States (Swanson and Newsom, 1962; Way
and Bowling, 1991). During early stages of kernel development, the piercing-sucking
stylet of the rice stink bug penetrates the rice hull and removes the content of the ker-
nels resulting in yield loss. In the later stages of grain development, feeding causes
discoloration of the kernel which is called pecky rice (Swanson and Newsom, 1962).
The rice inspection handbook allows for no more than 0.5% damaged grain in a 500
g sample to be considered U.S. grade 1 (USDA-FGIS, 2009). The stylets of rice stink
bugs can carry fungi that cause this discoloration, or the fungi can enter through the
wound at the feeding site after the rice stink bug has fed (Hollay et al., 1987). Farmers
are penalized when pecky rice causes breakage during the milling process resulting in
lower head rice yields (Way, 2003). This feeding can also result in reduced seed viability
(Swanson and Newsom, 1962; Patel et al., 2006).
1
Program Associate - Entomology, Department of Entomology, Lonoke.
2
Associate Department Head/Extension Entomologist and Program Associate, respectively, Department
of Entomology, Lonoke.
3
4
Program Technician - Rice Agronomy, Program Associate I, and Program Associate III, respectively,
137
Rice stink bugs usually move to rice from weeds or other rotational commodities
during heading (Way, 2003). Some of the alternate hosts for rice stink bug include grain
sorghum, oats, rye, wheat, barnyardgrass, bearded sprangletop, dallisgrass, lovegrass,
ryegrass, crabgrass, broadleaf signalgrass and several species of Panicum (Lorenz and
Hardke, 2013). Tindall et al. (2005) observed an increase in pecky rice with the presence
of these weeds and an increase in unfilled kernels in the plots with rice stink bug presence.
The rice stink bug is managed with foliar insecticide sprays (Way, 2003). The
economic threshold for stink bugs in Arkansas, during the first 2 weeks of heading, is
5 rice stink bugs in 10 sweeps with a sweep net, resulting in a recommended insecti-
cide application. During the next 2 weeks at soft dough, the threshold is 10 rice stink
bugs in 10 sweeps (Lorenz and Hardke, 2013). Some neighboring states have recently
lowered their thresholds to 2 to 3 rice stink bugs per 10 sweeps until rice kernels have
reached the hard dough stage.
Espino et al. (2007) showed the most susceptible stages of panicle development
to the rice stink bug feeding were milk and soft dough. The least desirable rice plants
were those in the preheading stage (Espino and Way, 2008). Bowling (1963) showed
that populations of rice stink bugs during the early stages of flowering increased the
number of non-filled seed, while others did not show any yield loss (Blackman, 2014).
Espino et al. (2007) found no differences in the amount and weight of complete ker-
nels of rice infested during heading. Awuni et al. (2015) found the highest decrease in
yield in the bloom stage and that milk and soft dough stages had the highest amount
of atrophied kernels.
It is important to understand the damage at different rice growth stages caused
by rice stink bug and the feeding densities that cause appreciable levels of damage to
better help growers effectively and economically control this pest. The objective of
our study was to determine the amount of yield loss and milling yield loss caused by
different rice stink bug densities at different kernel development stages.
PROCEDURES
Experiments were conducted at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center near Stuttgart. The cultivar in these
studies was Roy J, which was drill-seeded in early May 2015 and grown according to
standard agronomic practices for Arkansas. Plots were 70 63 inches on 7-inch drill
spacing. Cages were placed prior to heading to prevent natural infestations of rice stink
bugs. Infestations of stink bugs were initiated once > 50% of panicles had reached the
desired kernel development stage of flowering, milk, soft dough, or hard dough. Infested
stink bugs were allowed to remain for 7 days, after which time infestations were termi-
nated with foliar insecticide sprays and cages replaced with the plots remaining covered
until harvest. Infestations levels were 0, 4, 8, 25, and 42 rice stink bugs per plot. The
density of rice stink bugs/ft2 is 0, 0.1307, 0.2614, 0.8170, and 1.3725. The experimental
design was a randomized complete block with three replications per infestation timing.
Rice stink bug adults and late instar nymphs were collected with sweep nets in
heading rice fields and weedy areas surrounding rice fields. Insects were kept in small
138
cages with fresh plant material and a moist paper towel in a laboratory at 75 F for 48
h prior to infestation in field cages. Infestations in field cages were made early in the
morning or late in the evening to help the rice stink bugs acclimate under less stressful
environmental conditions to increase survival. Cage frames were 6 ft3 made of 1-in.
PVC pipe with 20 20 amber fabricated coverings (Lumite, Inc., Alto, Ga.) were used
to cover each plot.
Prior to harvest, 10 rice panicles were removed and placed in a brown paper bag
and stored in a dryer until moisture was 12%. The 10 panicles were harvested by hand
and separated into seed and blanks; partial filled seeds were counted as seeds. Blanks
and seeds were counted and the percentage of each plot was calculated. After the 10
panicles were removed, the center 5 rows of the plots were harvested with a plot com-
bine and seed was stored in a cloth bag and placed in a dryer until the moisture was
12%. A random 100-g sample of seed harvested with the plot combine was dehulled.
Using a light box, seed was separated into undamaged, rice stink bug damaged, kernel
smut, false smut, and other damage. The seed in each category was weighed and the
percentage of damage for each plot was calculated. After harvest, a random sample of
162 g of rough rice from each plot was used to evaluate grain milling quality. Rice was
milled to obtain percent head rice (whole kernels) and percent total white rice (whole
and broken kernels). Data were analyzed using analysis of variance, PROC GLM,
SAS v. 9.4 (SAS Institute, Inc., Cary, N.C.) with means separated using Fishers least
significant difference (P = 0.10).

A significant difference was found between rice stink bug densities when infested
at the rice milk development stage (P < 0.10, Table 1). Percent damaged kernels was
significantly higher for plots infested with 0.26, 0.82, and 1.37 rice stink bugs/ft2 com-
pared to plots infested with 0 or 0.13 rice stink bugs/ft2. No differences were observed
between rice stink bug densities infested at the bloom, soft dough, hard dough stages
(P > 0.10). In a greenhouse study, Espino et al. (2007) found differences between the
amount of damaged rice in rice stink bug infestations made to rice at heading, milk, and
soft dough. Comparing control plots to those infested with 15.79 rice stink bug/ft2, soft
dough had the highest percent of pecky rice, followed by milk, and then heading. Awuni
et al. (2015) found a significant difference in damaged kernels between the non-infested
plots and plots infested with 0.84 and 1.67 rice stink bug/ft2 (Table 2).
No rice stink bug density resulted in significantly lower grain yield compared to
the control (Table 3). Awuni et al. (2015) found uninfested plots yielded significantly
higher than plots infested with 0.84 and 1.67 rice stink bug/ft2 in their large cage study.
Bowling (1963) only found a significant difference between the non-infested cage and
the highest infested cage of 4 rice stink bug/ft2.
There were no significant differences in the amount of unfilled kernels at differ-
ent rice stink bug densities or infestation timings (Table 4). These results are similar to
Blackman (2014), he found no significant differences in unfilled kernels with rice stink
139
bug densities ranging from 0.15 to 3.0 rice stink bug/ft2. However, Espino et al. (2007),
in a greenhouse study, saw a significant increase in unfilled kernels for rice infested at
the heading stage compared to the uninfested control and rice infested at soft dough.
No decrease in total milled rice yield or head rice yield was observed for any
rice stink bug density or infestation timing (Tables 5 and 6). These results are similar
to Espino et al. (2007), who found no decrease in head rice milling yield for three of
their experiments, including two greenhouse cage experiments and one field cage study.
However, Bowling (1963) found varying decreases in milling yield associated with rice
stink bug infestation densities of 1.0, 2.0, and 4.0 rice stink bugs/ft2.
No decrease in rough rice yield resulted from increasing infestation densities of
rice stink bugs during the selected stages of kernel development. Total percent damage
only showed a significant difference when rice was infested during the milk stage at
rice stink bug densities that greatly exceed current action thresholds for Arkansas. A
wide range of injury to rice stink bug infestation has been found among current and
previous research studies. This has resulted in different thresholds established by neigh-
boring states throughout the mid-South and indicates the need for further study to more
accurately determine the effect of rice stink bug on grain yield loss and grain quality.
The rice stink bug is an important economic pest of rice. It is imperative that the
University of Arkansas System Division of Agriculture Cooperative Extension Service
provides growers with a threshold for control of this pest to avoid damage and/or quality
losses, but equally important to avoid making unnecessary applications for control to
maximize profit for rice growers.
ACKNOWLEDGMENTS
We would like to express our appreciation for funding and support from the
Arkansas rice growers administered through the Rice Research and Promotion Board
and the University of Arkansas System Division of Agriculture. We also wish to thank
Chuck Pipkins for his assistance with this project.
LITERATURE CITED
Awuni, G.A., J. Gore, D. Cook, F. Musser, A. Catchot, and C. Dobbins. 2015. Impact
of Oebalus pugnax (Hemiptera: Pentatomidae) infestation timing on rice yields
and quality. J. Econ. Entomol. 108(4):1739-1747.
Blackman, B.D. 2014. Evaluatin of economic injury levels and chemical control
recommendations for rice stink bug (Oebalus pugnax) in Louisiana [dissertation].
Baton Rouge, La., Louisiana State University.
Bowling, C.C. 1963. Cage tests to evaluate stink bug damage to rice. J. Econ. Ento-
mol. 56(2):197-200.
140
Espino, L., M.O. Way, and J.K. Olson. 2007. Most susceptible stage of rice panicle
development to Oebalus punax (Hemiptera: Pentatomidae). J. Econ. Entomol.
100(4):1282-1290.
Espino, L. and M.O. Way. 2008. Attractiveness of stages of rice panicle development
to Oebalus pugnax (Hemiptera: Pentatomidae). J. Econ. Entomol. 101(4):1233-
1237.
Hollay, M.A., C.M. Smith, and J.F. Robinson. 1987. Structure and formation of feed-
ing sheaths of rice stink bug (Heteroptera: Pentatomidae) on rice grains and their
association with fungi. Ann. Entomol. Soc. Am. 80:212-216.
Lorenz, G. and J.T. Hardke. 2013. Insect management in rice. In: J.T. Hardke (ed.).
Rice Production Handbook: MP 192. University of Arkansas System Division of
Agriculture Cooperative Extension Service, Little Rock, Ark.
Patel, D.T., M.J. Stout, and J.R. Fuxa. 2006. Effects of rice panicle age on quantita-
tive and qualitative injury by the rice stink bug (Hemiptera: Pentatomidae). Fla.
Entomol. 89:321-327.
Swanson, M.C. and L.D. Newsom. 1962. Effect of infestation by the rice stink bug,
Oebalus pugnax, on yield and quality in rice. J. Econ. Entomol. 55(6):877-879.
Tindall, K.V., B.J. Williams, M.J. Stout, J.P. Geaghan, B.R. Leonard, and E.P.
Webster. 2005. Yield components and quality of rice in response to graminaceous
weed density and rice stink bug populations. Crop Protection 24:991-998.
USDA-FGIS. 2009. United States Department of Agriculture Federal Grain Inspection
Service. United States Standards for Rice. USDA Grain Inspection, Packers, and
Stockyards Administration. Washington, D.C.
Way, M.O. and C.C. Bowling. 1991. Insect pests of rice. pp. 237-268. In: B. S. Luh
(ed.). Rice production. Van Nostrand Reinhold, N.Y.
Way, M.O. 2003. Rice arthrolpod pests and their management in the United States.
pp. 437-456. In: C.W. Smith and R. H. Dilday (eds.). Rice: Origin, history, tech-
nology, and production. Wiley, Hoboken, N.J.
141
Table 1. Percent damaged kernels based on weight in a 100-g brown

rice sample for each rice growth stage and rice stink bug infestation density.
Damaged kernels
Infestation density Bloom Milk Soft dough Hard dough
(ft2) ---------------------------------- (%)----------------------------------
0 1.20 0.91 b 1.64 1.97
0.13 1.20 1.07 b 1.63 1.49
0.26 1.03 1.42 a 1.45 1.61
0.82 1.10 1.49 a 1.57 1.82
1.37 1.12 1.33 a 1.80 1.78
LSD0.10 NS 0.2574 NS NS
CV 22.4831 13.6280 12.4328 18.8347
Means followed by the same letter within a column are not significantly different (P < 0.10).
142
Table 2. Calculated rice stink bug densities of previous caging studies.

Rice stink Calculated
Type of Caged Caged bug infested rice stink
Author study area area per cage bug density
(ft2) (ft2)
Swanson Field cage 6 ft 12 ft 66 0 0
and Newsom, 20 0.3030
1962 100 1.5152
500 7.5758
33 0 0
20 0.6061
100 3.0301
500 15.1515
Bowling, 1963 Field cage 4 ft 20 ft 80 0 0

1 1
2 2
4 4
Natural Natural
3 ft 20 ft 60 0 0
1 1
2 2
4 4
Natural Natural
Espino et al., Greenhouse 4 15-cm 0.76 12/cage 15.789

2007 cage diam. pots
Espino and Greenhouse 4 15-cm 0.76 10/cage 13.16

Way, 2008 cage diam. pots
Blackman, Field cage 0.62 m2 6.67 0/cage 0

2014 1/cage 0.1499
2/cage 0.2999
5/cage 0.7496
10/cage 1.4993
20/cage 2.9985
Awuni et al., Field cage 3.24 m2 34.875 0/m2 0

2015 9/m2 0.8361
18/m2 1.6723
Lorenz and Field cage 30.6 0 0

Hardke 4 0.1307
8 0.2614
25 0.8170
42 1.3725
143
Table 3. Grain yield for each rice growth stage and rice stink bug infestation density.
Grain yield
(ft2) ------------------------------- (bu/acre)-----------------------------
0 176.0 193.1 169.1 146.8
0.13 182.9 194.4 159.2 147.6
0.26 175.8 185.2 168.5 158.0
0.82 184.2 199.2 174.7 153.4
1.37 181.7 188.4 179.3 151.3
LSD0.10 NS NS NS NS
CV 4.0839 5.0108 6.4683 7.1455
Table 4. Percent blank kernels (based on kernel

count) attributed to rice stink bug feeding in a 10 panicle rough rice
sample for each rice growth stage and rice stink bug infestation density.
Blank kernels
(ft2) ---------------------------------- (%)----------------------------------
0 11.7 14.1 14.1 20.0
0.13 11.7 11.9 18.3 29.6
0.26 12.3 12.6 20.1 25.4
0.82 13.8 11.9 17.1 26.9
1.37 10.4 12.4 13.8 20.5
LSD0.10 NS NS NS NS
CV 17.9375 19.6384 25.7167 21.1187
Table 5. Percent milled total white rice yield in a 162-g sample

for each rice growth stage and rice stink bug infestation density.
Milled total white rice yield
(ft2) ---------------------------------- (%)----------------------------------
0 74.34 74.26 73.50 73.29
0.13 74.26 74.34 73.31 73.19
0.26 74.34 74.36 73.62 73.37
0.82 73.83 73.97 73.91 72.72
1.37 74.20 74.22 73.64 72.74
LSD0.10 NS NS NS NS
CV 0.3876 0.4071 0.5530 0.5587
144
Table 6. Percent milled head rice yield in a 162-g sample

for each rice growth stage and rice stink bug infestation density.
Milled head rice yield
(ft2) ---------------------------------- (%)----------------------------------
0 68.40 a 66.98 67.26 66.93
0.13 67.78 b 67.39 66.83 67.04
0.26 68.64 a 66.28 66.93 67.43
0.82 67.84 b 66.42 67.06 66.46
1.37 68.27 a 66.36 67.67 66.60
LSD0.10 0.4042 NS NS NS
CV 0.3654 0.9422 0.9448 0.7761
Means followed by the same letter within a column are not significantly different (P < 0.10).
145
Potential Exposure of Honey Bees

to Neonicotinoid Insecticides in Rice
G.M. Lorenz1, J.T. Hardke2, T.L. Clayton3, N.M. Taillon1,
D.L. Frizzell4, J. Black1, W.A. Plummer1, and H.M. Chaney1
ABSTRACT
Insecticide seed treatments and foliar clothianidin applications were evaluated for
expression in the flag leaf and floral parts of rice. Data analysis of samples indicate
that neonicotinoid insecticides used as seed treatments or applied as early-season foliar
treatments were expressed at very low levels or were nonexistent when samples were
taken. Also, observations of bees visiting rice indicated extremely low levels of honey
bees in rice fields.
INTRODUCTION
Recently, neonicotinoid insecticides used in agronomic crops have been scruti-
nized for their perceived impact on honey bee population decline in the U.S. In Arkansas,
insecticides are essential to limit yield losses from insects in rice. Most notably, the
neonicotinoid seed treatments CruiserMaxx Rice and NipsItTM INSIDE are important
for rice water weevil and grape colaspis control. To date, all of the research focusing
on the fate of neonicotinoid insecticides has been done in other southern crops such as
corn, soybean, and cotton (Stewart et al., 2014)., No research has been conducted in rice
to this point. As environmental groups continue to challenge the use of neonicotinoids
in agriculture and pressure the U.S. Environmental Protection Agency to ban their use,
it will become more important to generate information to refute their claims.
1
Associate Department Head/Extension Entomologist, Program Associate I - Entomology, Graduate
Assistant, Program Associate - Entomology, and Program Technician, respectively, Department of
Entomology, Lonoke.
2
3
Program Associate - Entomology, Department of Entomology, Stuttgart.
4
Program Associate III, Rice Research and Extension Center, Stuttgart.
146
PROCEDURES
Objective 1. Measuring Levels of Neonicotinoid Insecticides
in Rice Plants at Flowering
Experiments were conducted at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center near Stuttgart Ark. The cultivar in
these studies was CL151, which was drill-seeded on 6 May 2015 and grown according
to standard agronomic practices for Arkansas. Plots were 9 rows on 7-inch drill spacing
and 15 ft in length arranged in a randomized complete block design with four replica-
tions. The treatments included: an untreated check, Cruiser Maxx Rice 7 oz/cwt seed
treatment; preflood Belay 4.5 oz/acre; postflood Belay 4.5 oz/acre; and NipsIt INSIDE
1.92 oz/cwt seed treatment. CruiserMaxx Rice contains the neonicotinoid insecticide
thiamethoxam while NipsIt INSIDE and Belay contain the neonicotinoid insecticide
clothianidin. The Belay preflood application was made on 10 June and the postflood ap-
plication was made 18 June. Flag leaf and panicle samples were taken 5 August, 60 days
after planting, 25 days after preflood foliar application, and 17 days after postflood foliar
treatment. Standard laboratory practices were conducted to ensure no contamination of
samples occurred. Flag leaves from each plot were removed at the collar, placed in a
labeled plastic bag, weighed, and stored on ice in a cooler. A sample size of 125 leaves
was taken from the center rows of each plot to ensure enough tissue for testing. Each
treatment was processed separately to lessen the possibility of contamination. Between
each treatment, hands were cleaned with a 5% bleach solution, rinsed with water, and
new gloves were used. The leaf tissue remained in a freezer until shipped. Panicles from
each plot were removed, placed in a paper bag, stored on ice in a cooler, and brought
to the laboratory for processing. A sample size of 50 panicles was removed to ensure
enough tissue for testing. From each of 30 panicles, 15 florets were removed, placed in
a labeled conical tube, and weighed to ensure 3 g of tissue were present. If the sample
weighed less than 3 g, more florets were removed from the remaining panicles and the
sample was weighed again. The tubes were placed in a freezer until shipped. To prepare
for processing; tables, scales, and forceps were cleaned with a 5% bleach solution and
wax paper was placed on each table to prevent contamination. Between each sample,
the wax paper was removed, tables, forceps, and scales were cleaned with the bleach
solution, and the tables were covered with a new piece of wax paper.
Samples were analyzed to determine the levels of neonicotinoid residues by the
USDA AMS Science and Technology Laboratory Approval and Testing Division of
the National Science Laboratories Gastonia Lab in Gastonia, N.C. This laboratory
is accredited to ISO/IEC 17025:2005 for specific tests in the fields of chemistry and
microbiology, including testing for pesticide residues. The samples were extracted for
analysis of agrochemicals using a refined methodology for the determination of neo-
nicotinoid pesticides and their metabolites using an approach of the official pesticide
extraction method (AOAC, 2007), also known as the QuEChERS method, and analyzed
by liquid chromatography coupled with tandem mass spectrometry detection (LC/MS/
MS). Samples were analyzed for the presence of 17 insecticides or their metabolites.
147
Quantification was performed using external calibration standards prepared from cer-
tified standard reference material. Only detections of clothianidin, imidacloprid, and
thiamethoxam were reported. The method detection limit for these compounds was 1
ng/g (1 ppb).
Objective 2. Survey Conducted to Determine the Frequency at Which Honey

Bees Visit Flowering Rice Plants
Beginning in late September, ten flowering rice fields in Arkansas and Jefferson
counties were monitored for the presence of honey bees. Between the hours of 8:30
AM and 11:00 AM, at least five transects of 300-ft sections were observed by slowly
walking and visually looking for honey bees visiting rice panicles. All observations
were recorded as well as the location, stage of rice, and crops surrounding each field.
Data was processed using the latest version of Agriculture Research Manager
(Gylling Data Management, Inc., Brookings, S.D.), analysis of variance, and Duncans
New Multiple Range Test (P = 0.05).

Objective 1. Measuring Levels of Neonicotinoid Insecticides
in Rice Plants at Flowering
In flag leaf samples taken at flowering, only the CruiserMaxx (thiamethoxam)
IST indicated a low level of detection at 7.93 ppb; NipsIt INSIDE and both Belay fo-
liar applications had no detection of clothianidin as seen in the untreated check (Table
1.) A similar trend was observed for pollen with an even lower level of CruiserMaxx
found in florets and pollen with 2.23 ppb and all other treatments having no detection
of clothianidin. This study correlates well with a previous study (Stewart et al., 2014)
on cotton, soybean, and corn where very low levels of detections were found in pollen.
Objective 2. Survey Conducted to Determine the Frequency at Which Honey

Bees Visit Flowering Rice Plants
A total of 57 transects were made. In those transects, only one bee was observed
(Tables 2 and 3). There was no difference in bee population between time, day, or stage
of rice. The crops surrounding each field had no impact on the appearance of bees in
rice fields. Rice, like most of our major row crops, is predominantly self-pollinated and
from these studies does not appear to be attractive to bees.
In previous studies we have demonstrated that insecticide seed treatments not
only provide protection of the rice plant from insects and reduce stress, but increase
148
yields and profitability and are vital for rice production in Arkansas and the mid-South.
Although neonicotinoid insecticide seed treatments have been under fire recently for
impact on honey bees, these and other studies continue to show it is largely unfounded
and focus should be placed on the real issues impacting pollinators.
ACKNOWLEDGMENTS
We would like to express our appreciation for funding and support for these studies
from the Arkansas rice producers through monies administered by the Rice Research
and Promotion Board and the University of Arkansas System Division of Agriculture.
LITERATURE CITED
AOAC. 2007. Official Methods of Analysis. Method 2007.01. Pesticide residues in
foods by acetonitrile extraction and partitioning with magnesium sulfate. AOAC
International, Gaithersburg, Md. Available at: http://www.weber.hu/PDFs/QuECh-
ERS/AOAC_2007_01.pdf
Stewart, S.D., G.M. Lorenz, A.L. Catchot, J. Gore, D. Cook, J. Skinner, T.C. Mueller,
D.R. Johnson, J. Zawislak, and J. Barber. 2014. Potential exposure of pollinators
to neonicotinoid insecticides seed treatments in the mid-southern United States.
dx.doi.org/10.1021/es501657w | Environ. Sci. Technol. 48:9762-9769.
Table 1. Levels of neonicotinoid insecticides (ppb) in the flag leaf and florets of rice from
plots treated with thiamethoxam and clothianidin insecticide seed treatments at planting
and clothianidin foliar applications made preflood or postflood on rice at bloom (60 days
after planting, 25 days after preflood application, and 17 days after postflood application).
Treatment Rate/acre Active ingredient Flag leaf Pollen
-----------------(ppb)---------------
Untreated check NA 0.0 b 0.0 b
CruiserMaxx rice 7 oz/ cwt thiamethoxam 7.93 a 2.23 a
NipsIt INSIDE 1.92 oz/ cwt clothianidin 0.0 b 0.0 b
Belay preflood 4.5 oz/ acre clothianidin 0.0 b 0.0 b
Belay postflood 4.5 oz/ acre clothianidin 0.0 b 0.0 b
Means followed by the same letter do not significantly differ at least significant difference
P = 0.05.
149
Table 2. Observations of the number of bees observed

in flowering rice fields at different times of the day at 300 ft transects
across the field in Jefferson and Arkansas Counties (Observations = 57).
Number of bees in transect
Field Growth stage Date Time 1 2 3 4 5 6
1 Flowering 9/21 8:30 AM 0 0 0 0 0 0
2 Flowering 9/24 9:15 AM 0 0 0 0 0 0
3 Flowering 9/24 10:00 AM 0 0 1 0 0 0
4 Flowering and milk 9/24 10:50 AM 0 0 0 0 0 0
5 Flowering and milk 9/25 9:10 AM 0 0 0 0 0 -
6 Flowering 9/25 9:35 AM 0 0 0 0 0 -
7 Flowering 9/25 10:00 AM 0 0 0 0 0 -
8 Flowering and milk 9/28 9:20 AM 0 0 0 0 0 0
9 Flowering 9/28 10:00 AM 0 0 0 0 0 0
10 Flowering 10/1 10:00 AM 0 0 0 0 0 0
150
Table 3. Field location of bee observations and surrounding crops or vegetation.
Field Field location North of field South of field East of field West of field
1 Arkansas Soybeans Mature rice Mature rice Soybeans
2 Jefferson Mature rice Soybeans Mature rice Soybeans
3 Jefferson Tree line Soybeans Tree line Soybeans
4 Arkansas Soybeans Soybeans Tree line Mature rice
5 Jefferson Flowering rice Tree line Soybeans Mature rice
6 Jefferson Fallow Flowering rice Soybeans Mature rice
7 Jefferson Mature corn Tree line Mature corn Cut milo
8 Arkansas Soybeans Soybeans and mature corn Soybeans Mature rice
9 Arkansas Soybeans Soybeans Flowering rice Soybeans
10 Arkansas Tree line Soybeans Tree line Flowering rice
151
Value of Insecticide Seed Treatments in Arkansas Rice

N.M. Taillon1, G.M. Lorenz1, W.A.Plummer1, H.M. Chaney2, and J. Black1
ABSTRACT
Rice insecticide seed treatment (IST) trials from 2008 through 2015 were analyzed to
determine impact on plant stand and yield to determine economic benefit to producers.
Data was summarized across all trials and results indicated a significant increase in
both stand and yield.
INTRODUCTION
Rice water weevil and grape colaspis are common pests in Arkansas rice that
have the potential to reduce stand counts and yield. In 2010, Cruiser (Syngenta Crop
Protection) and Dermacor X-100 (DuPont Crop Protection) received full labels and
in 2012 NipsIt INSIDE (Valent U.S.A. Co.) received a full label permit for use in rice
to provide effective control of both insects during the developmental stages of rice.
Throughout testing there was a general trend for seed treatments to improve stand count
and yield (Lorenz et al., 2013). We observed many times that under stressful conditions,
the seed treatment helped to moderate or buffer stress (Taillon et al., 2014).
PROCEDURES
Trials were conducted on numerous grower fields across the state, at the University
of Arkansas System Division of Agricultures Pine Tree Research Station near Colt,
Ark., and at the Rice Research and Extension Center near Stuttgart, Ark. The selection
of locations was based on fields with a history of problems with either grape colaspis or
rice water weevil. However, we did not experience insect problems in every field. These
trials consisted of a small plot randomized complete block design with 4 replications as
well as large plot demonstration trials in a randomized strip block design with a minimum
of 3 replications. Seed treatments included Cruiser 5FS 3 oz/cwt (thiamethoxam) or
CruiserMaxx Rice (thiamethoxam + fungicides premix), Dermacor X-100 1.5-6 oz/cwt
(chorantraniliprole), and NipsIt INSIDE 1.92 oz/cwt (clothianidin). All seed treatments,
1
Program Associate I - Entomology, Associate Department Head / Extension Entomologist, Program
Associate - Entomology, and Graduate Assistant, respectively, Department of Entomology, Lonoke.
2
Associate Director, Agriculture and Natural Resources, Cooperative Extension Service, Conway.
152
as well as the check, included a fungicide package of Apron XL (Mefenoxam), Maxim

4 FS (Fludioxonil), and Dynasty 83 FS (Azoxystrobin). Seed treatments were applied
at the University of Arkansas System Division of Agricultures Lonoke Extension Cen-
ter, Lonoke, Ark. using small and large batch seed treaters. Tests were conducted with
these seed treatments on conventional, Clearfield, and hybrid cultivars of rice using
standard seeding rates. Stand count data was collected 2 to 3 weeks post planting by
counting plants in 10 row feet per plot. Yields for small plot trials were taken with a
plot combine and large block tests were harvested with a standard combine. A metadata
analysis across these trials was conducted to determine the effect of insecticide seed
treatments on stand establishment and yield. Data was processed using the latest version
of Agriculture Research Manager (Gylling Data Management, Inc., Brookings, S.D.),
analysis of variance, and Duncans New Multiple Range Test (P = 0.10).

Results indicated that seed treatments increased stand counts in many trials 0%
to 58% above the untreated check (UTC) and averaged 14.5% across 201 trials (Fig.
1). Cruiser-treated plots had an increase in stand counts ranging from 0% to 58% above
the UTC with an average stand increase of 15.8% across 87 trials (Fig. 2). NipsIt seed
treatment had an increase in stand counts ranging from 1.6% to 53% above the UTC,
and averaged 15.1% across 51 trials (Fig. 3). Dermacor-treated plots had an increase
in stand counts ranging from 0.3% to 39.8% above the UTC with an average increase
of 12% across 61 trials (Fig. 4).
Seed treatments across 201 trials provided an average 8.33 bu/acre increase com-
pared to the untreated check (Fig. 5). Based on the yield results shown in the figures
below, Cruiser, NipsIt, and Dermacor provided a 76%, 75%, and 85% probability of a
net return, respectively (Figs. 6, 7, and 8).
Insecticide seed treatments not only improve stand counts, but also increase yields
80% of the time for Arkansas rice producers. This allows growers the flexibility of
choosing lower seeding rates to reduce input costs while still maintaining profitability.
Based on our findings, these seed treatments are recommended for use in Arkansas rice.
Research will continue to evaluate new chemistries.
ACKNOWLEDGMENTS
We would like to acknowledge funding for this project was provided by: Rice
Check-Off funds, administered by the Rice Research and Promotion Board, University
of Arkansas System Division of Agriculture, Syngenta, Valent, and Dupont. We also
thank Jarrod Hardke and his crew at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center, and Shawn Clark and his crew at
153
the University of Arkansas System Division of Agricultures Pine Tree Research Station
for their cooperation and support in these studies.
LITERATURE CITED
Lorenz, G.M., N. Taillon, J. Hardke, and G. Studebaker. 2013. Rice insecticide seed
treatments: Is there value to the Arkansas rice producer? In: University of Ar-
kansas Agricultural Research and Extension Agriculture and Natural Resources.
FSA7085.
Taillon, N., G.M. Lorenz, J. Black, W.A. Plummer, and H.M. Chaney. 2014. Insec-
ticide seed treatments in rice: Is there value to the grower? In: R.J. Norman and
144. Fayetteville, Ark.
Fig. 1. Percent stand increase of insecticide seed treatments

over untreated check (fungicide only) across 201 trials from 2008 to 2015.
154
Fig. 2. Percent stand increase of Cruiser/CruiserMaxx over

untreated check (fungicide only) across 87 trials from 2008 to 2015.
Fig. 3. Percent stand increase of NipsIt INSIDE over

155
Fig. 4. Percent stand increase of Dermacor over

Fig. 5. Percent yield increase of insecticide seed treatments

over untreated check (fungicide only) across 201 trials from 2008 to 2015.
156
Fig. 6. Percent yield increase of Cruiser/CruiserMaxx over

Fig. 7. Percent yield increase of NipsIt INSIDE over

157
Fig. 8. Percent yield increase of Dermacor over

158
Residual Weed Control and Crop

Response to Pethoxamid Systems in Rice
R.C. Doherty1, L.T. Barber2, J.K. Norsworthy3, and Z.T. Hill1
ABSTRACT
Rice weed control systems in Arkansas are complex and must contain multiple modes
of action to provide control of troublesome weed species. Many weed control programs
no longer provide adequate control of some of the most troublesome weeds such as
barnyardgrass and Amazon sprangletop. A trial was conducted in 2015 to evaluate weed
control and crop response to pethoxamid, a new potential residual herbicide in rice.
Pethoxamid controlled barnyardgrass and Amazon sprangletop 80% to 91% alone and
up to 96% and 98%, respectively, when tank-mixed with imazethapyr at 0.063 lb ai/acre
and applied delayed-preemergence to spiking rice. Eclipta control was also improved
with the addition of pethoxamid to clomazone at 0.3 lb ai/acre or imazethapyr at 0.063 lb
ai/acre. No rice injury was caused by any pethoxamid-containing treatment. Pethoxamid
does provide an additional mode of action which increases control of barnyardgrass,
Amazon sprangletop, and eclipta in Arkansas rice.
INTRODUCTION
Clomazone, quinclorac, and pendimethalin are relied upon heavily to provide
early-season residual grass control in Arkansas rice. The use of these herbicides is
crucial for early season systems to be successful (Scott et al., 2015). Due to increased
barnyardgrass resistance to multiple herbicide modes of action such as propanil, ima-
zethapyr, and quinclorac when applied post-emergence, pre-emergence systems are
becoming much more important to provide complete control of barnyardgrass prior to
flooding rice. The prevention of herbicide resistance is also a driving force in the use
of multiple modes of action in rice weed control. Pethoxamid, a new chloroacetamide
herbicide and a new potential mode of action for grass control in rice, has proven ben-
eficial when used in weed control programs in other crops. Jursik et al. (2013) found
that pethoxamid provided good control of redroot pigweed and barnyardgrass when
1
Program Associate - Weed Science and Weed Program Associate, respectively, Southeast Research and
Extension Center, Monticello.
2
Associate Professor - Weed Science, Department of Crop, Soil, and Environmental Science, Lonoke.
3
159
applied pre-emergence in sunflower. The purpose of this research was to evaluate the
effectiveness of pethoxamid herbicide on grass control in rice, and to determine if any
crop injury would occur.
PROCEDURES
A trial was conducted in 2015 at the University of Arkansas System Division of
Agricultures Southeast Research and Extension Center in Monticello, Ark., to evaluate
weed control and crop response to pethoxamid herbicide systems in rice. A random-
ized complete block design with four replications was used. The cultivar CL111 was
drill-seeded into Sharkey clay soil at 90 lb/acre, and weed seed were broadcast after
planting rice. Pethoxamid was applied alone at 0.375 or 0.5 lb ai/acre and in conjunction
with clomazone (0.3 lb ai/acre), imazethapyr (0.063 lb ai/acre), pendimethalin (1.0 lb
ai/acre), and quinclorac (0.375 lb ai/acre). Treatments were applied using a Mudmaster
sprayer equipped with a compressed air powered multi-boom, calibrated to deliver 12
gal/acre. Treatments were applied 6 days after planting to spiking rice. Weed control
and crop injury were evaluated on a scale from 0% to 100%, where 0% equals no weed
control or crop injury and 100% equals complete control. Data were subjected to an-
alysis of variance and means were separated using Fishers protected least significant
difference test (P = 0.05).

Pethoxamid at 0.375 or 0.5 lb ai/acre applied alone, or in combination with
clomazone, imazethapyr, pendimethalin, or quinclorac, provided equivalent control
(97% to 99%) of barnyardgrass, Amazon sprangletop, and eclipta 26 days after ap-
plication (DAA). No crop injury was caused by any treatment, pethoxamid alone or in
combination with any other herbicide used, 26 DAA (data not shown).
Sixty six DAA pethoxamid at 0.375 and 0.5 lb ai/acre provided 84% and 80%
control of barnyardgrass, respectively, which was similar to all other standard herbicides
including clomazone and pendimethalin (Table 1). However, when pethoxamid at 0.375
and 0.5 lb ai/acre was mixed with imazethapyr at 0.063 lb ai/acre, barnyardgrass control
increased to 93% and 96%, respectively. All other pethoxamid combination treatments
provided 90% or less barnyardgrass control. Pethoxamid at 0.375 lb ai/acre provided
91% control of Amazon sprangletop at 66 DAA. The addition of other herbicides such
as pendimethalin at 1.0 lb ai/acre, imazethapyr at 0.063 lb ai/acre, or the higher rate of
pethoxamid (0.50 lb ai/acre) did not improve control over pethoxamid alone at the 0.375
lb ai/acre rate. Control of Amazon sprangletop was the highest (98%) 66 DAA with the
combination of pethoxamid at 0.5 lb ai/acre plus imazethapyr at 0.063 lb ai/acre. All
other treatments provided 94% or less control of Amazon sprangletop. Pethoxamid at
0.375 lb ai/acre controlled eclipta 97% at 66 DAA and was not significantly different
than all other herbicides tested alone or in combination which controlled eclipta 95%
to 99%. Pethoxamid at 0.5 lb ai/acre provided 94% control of eclipta 66 DAA and was
160
significantly lower than five treatments that provided 99% control. No crop injury was
observed 66 DAA by any treatment, pethoxamid alone or in combination with any other
herbicide used in this trial.
Pethoxamid herbicide provided excellent control of both barnyardgrass and
Amazon sprangletop and appears to have potential to be a reliable rice herbicide for
long lasting residual grass control. The best overall treatment was pethoxamid at 0.5
lb ai/acre plus Imazethapyr at 0.063 lb ai/acre, which provided 96%, 98%, and 99%
control of barnyardgrass, sprangletop, and eclipta, respectively. No significant crop
injury was caused by any treatment, which further supports the use of pethoxamid
at the delay-preemergence or spiking stage in Arkansas rice weed control programs.
If labeled, pethoxamid can provide an alternative mode of action for improved grass
weed control in rice.
ACKNOWLEDGMENTS
Special appreciation is extended to the Arkansas Rice Research and Promotion
Board and the University of Arkansas System Division of Agriculture for providing
funding for this project.
LITERATURE CITED
Jursik, M., M. Kocarek, K. Hamouzova, J. Soukup, and V. Venclova. 2013. Effect of
precipitation on the dissipation, efficacy, and selectivity of three chloroacetamide
herbicides in sunflower. Plant Soil Environ. 59:175-182.
Scott, R.C., L.T. Barber, J.W. Boyd, G. Selden, J.K. Norsworthy, and N.R. Burgos.
2015. Recommended chemicals for weed and brush control. University of Arkan-
sas System Division of Agriculture, Cooperative Extension Service MP442015.
Little Rock, Ark.
161
Table 1. Weed control and crop response to pethoxamid 66 days after application.
Weed control
Crop injury
Barnyard- Amazon
Treatment Rate Timing grass sprangletop Eclipta Stunting
(lb ai/acre) ------------------------------(%)------------------------------
Pethoxamid 0.375 spike 84 91 97 0
Clomazone 0.3 spike 76 44 95 0
Clomazone 0.3
Clomazone 0.3
Imazethapyr 0.063 spike 84 74 97 0
Imazethapyr 0.063
Imazethapyr 0.063
Pendimethalin 1.0 spike 91 80 97 0
Pendimethalin
1.0
Pendimethalin
1.0
Quinclorac 0.375 spike 86 45 99 0
Quinclorac 0.375
Quinclorac 0.375
LSD0.05 11 12 4 NSa
a
NS = not significant.
162
Evaluation of Very Long-Chain Fatty

Acid-Inhibiting Herbicides in Arkansas Rice
J.A. Godwin Jr.1, J.K. Norsworthy1, R.C. Scott2,
L.T. Barber2, M.L. Young1, and M.W. Duren3
ABSTRACT
Two field experiments were conducted in 2015 to evaluate the use of very long-chain
fatty acid-inhibiting (WSSA Group 15) herbicides in Arkansas rice on silt loam and
silty clay soils. The experiments were designed as a three (application timings) by four
(herbicides) factorial in a randomized complete block design. The herbicides included
Warrant (acetochlor at 0.94 lb ai/acre), Zidua (proxasulfone at 0.133 lb ai/acre), Dual
Magnum (S-metolachlor at 0.955 lb ai/acre), and pethoxamid at 0.75 lb ai/acre ap-
plied delayed pre-emergence (DPRE), spiking, and at the 1- to 2-lf rice stages. Of the
herbicides evaluated, pyroxasulfone caused the most crop injury. On the silt loam soil,
pyroxasulfone caused 68% rice injury, averaged over all application timings. Unaccept-
able levels of injury to rice also occurred with S-metolachlor, although, rice appeared
more tolerant to S-metolachlor at the silty clay soil location. Averaged across applica-
tion timings, rice exhibited acceptable tolerance to pethoxamid and acetochlor at both
locations. Because of the tolerance of rice to acetochlor and pethoxamid observed in
this work, additional research is merited to understand the level of weed control that
can be obtained with these herbicides in Arkansas rice systems.
INTRODUCTION
Due to the repetitive use of the same herbicide mechanisms of action (MOA)
in Arkansas rice, the evolution of resistance has occurred in several common weeds.
Herbicide resistance in barnyardgrass (Echinochloa crus-galli) and red rice (Oryza
sativa) has resulted in increased weed management difficulties for growers across the
state. Many of the most commonly used herbicide MOA in rice have an extremely high
risk for resistance: ALS inhibitors (157 resistant species worldwide), PSII inhibitors (73
resistant species worldwide), ACCase inhibitors (47 resistant species worldwide), and
1
Graduate Assistant, Professor, and Graduate Student, respectively, Department of Crop, Soil, and Envi-
ronmental Science, Fayetteville.
2
Professor and Associate Professor, respectively, Department of Crop, Soil, and Environmental Science,
Lonoke.
3
163
synthetic auxins (32 resistant species worldwide) (Heap, 2015). Hence, it is important
to integrate new MOA into rice whenever possible in order to combat or prevent the
further development of resistant weeds. Group 15 herbicides are commonly used in
the U.S. in crops such as corn, soybean, and cotton for control of annual grasses and
small-seeded broadleaves. Currently, there are no Group 15 herbicides used in U.S.
rice production; however, Group 15 herbicides such as pretilachlor and butachlor are
commonly used with great success in Asian dry-seeded rice culture (Chauhan, 2012).
Group 15 herbicides are at a relatively low risk for resistance when compared to other
MOA considering that there are only four resistant weed species worldwide (Heap,
2015). With success in other U.S. crops, Asian rice, and a low risk for resistance, Group
15 herbicides may have a potential fit in Arkansas rice. The objectives of this research
were to evaluate four group 15 herbicides at three different application timings for
potential use in rice based on crop tolerance.
PROCEDURES
Two field experiments were conducted in 2015 at the University of Arkansas
System Division of Agricultures Rice Research and Extension Center (RREC) near
Stuttgart, Ark., and the Northeast Research and Extension Center (NEREC) in Keiser,
Ark., to evaluate the use of very long-chain fatty acid-inhibiting (WSSA Group 15)
herbicides in Arkansas rice. These locations represent two distinct soil types, a DeWitt
silt loam (RREC) and a Sharkey silty clay soil (NEREC). The experiments were de-
signed as a three (application timings) by four (herbicides) factorial in a randomized
complete block design. The herbicides: Warrant (acetochlor at 0.94 lb ai/acre), Zidua
(pyroxasulfone at 0.133 lb ai/acre), Dual Magnum (S-metolachlor at 0.955 lb ai/acre),
and pethoxamid at 0.75 lb ai/acre were applied delayed pre-emergence (DPRE), at
spiking, and to 1- to 2-lf rice. A control (check) was included for comparison. All plots
were maintained weed-free throughout the experiments using herbicides labeled for
rice and applied as needed.
Clearfield 111 rice was planted at 22 seed/ft of row in 7.5-inch-wide rows in
6 20 ft plots. All herbicides were applied with a CO2-pressurized backpack sprayer
at 15 gallons of spray solution per acre. Observations were taken on rice injury on a
scale of 0 to 100, with 0 being no injury and 100 being complete crop death. Rough
rice grain was harvested at crop maturity using a small-plot combine. All data were
subjected to analysis of variance and means were separated using Fishers protected
least significant difference test (0.05).

Silt Loam Soil
There was no interaction between herbicide and application timing for injury rated
2 to 3 weeks after treatment (WAT); however, both main effects were significant. Aver-
aged across application timings, rice exhibited the greatest tolerance to acetolachlor and
pethoxamid, with injury averaging no more than 5% for both herbicides (Fig. 1). As ap-
164
plication timing was delayed, rice tended to have greater tolerance to all herbicides (data
not shown). Due to the reactivation of herbicides after establishment of the permanent
flood, rice injury increased substantially for pyroxasulfone and S-metolachlor to unac-
ceptable levels (Fig. 2), but rice injury did not increase for acetochlor and pethoxamid.
Rough rice yield was statistically similar to the nontreated check for acetochlor
treatments applied at spiking and to 1- to 2-lf rice and pethoxamid at all application
timings (Table 1). Significant reduction in rice yields resulted for all pyroxasulfone and
S-metolachlor application timings compared to the nontreated checkand when compared
to most of the acetolachlor and pethoxamid timings.
Silty Clay Soil

There was only a herbicide main effect at 4 to 5 (Fig. 3) and at 7 to 8 WAT (data
not shown) for rice injury. At both evaluation dates, pyroxasulfone caused the greatest
amount of rice injury, whereas injury was minimal following S-metolachlor, pethoxamid,
and acetochlor applications. Similar to the silt loam soil, rice injury on the silty clay
soil increased after the permanent flood was established (data not shown).
Rough rice yields following acetochlor and pethoxamid DPRE and pethoxamid at
spiking were comparable to the nontreated control (Table 1). Similar to the silt loam site,
rice treated with pyroxasulfone often yielded the lowest among all treatments (Table 1).
On both soils, rice exhibited a high level of tolerance to pethoxamid and acetochlor.
Neither of these herbicides are currently labeled in rice, nor are there other products
labeled in rice having a similar MOA. Hence, either of these products could potentially
provide growers additional options in the battle against and to prevent the development
of herbicide-resistant barnyardgrass, especially considering that acetochlor is currently
labeled in other mid-South crops and has demonstrated a high level of barnyardgrass
control in other research (Riar et al., 2011). In the upcoming year, efforts will continue
around testing pethoxamid and acetochlor across soil textures and environments in
rice along with evaluating weed control programs that contain these two herbicides.
ACKNOWLEDGMENTS
We thank the staff at both the RREC and NEREC for help with this research, and
appreciate the Arkansas Rice Research and Promotion Board for funding this research
as well as the University of Arkansas System Division of Agriculture.
LITERATURE CITED
Chauhan, B.S. 2012. Weed ecology and weed management strategies for dry-seeded
rice in Asia. Weed Sci. 26:1-13.
165
Heap, I. 2015. The international survey of herbicide resistant weeds. Access date: 16
December 2015. Available at: www.weedscience.org
Riar, D.S., J.K. Norsworthy, D.B. Johnson, C.E. Starkey, and A. Lewis. 2011. Ef-
ficacy and cotton tolerance to Warrant herbicide. In: D.M. Oosterhuis (ed.). Sum-
maries of Arkansas Cotton Research 2011. University of Arkansas Agricultural
Experiment Station Research Series 602:30-35. Fayetteville.
Table 1. Rough rice yields following applications of four

Group 15 herbicides at three application timings on the DeWitt silt
loam soil at the Rice Research and Extension Center, near Stuttgart, Ark., and
the Sharkey silty clay soil at the Northeast Research and Extension Center, Keiser, Ark.
Rough rice yield (bu/acre)
Herbicide Timing Silt loam Silty clay
None (Check) ------ 175 ab 147 a
Acetochlor DPRE 157 cd 137 a-d
Spiking 162 b-d 127 d
1-to 2-lf rice 180 a 132 b-d
Pethoxamid DPRE 161 b-d 140 a-c
Spiking 162 a-d 14 ab
1-to 2-lf rice 168 a-c 130 cd
S-metolachlor DPRE 139 e 106 e
Spiking 134 e 97 ef
1-to 2-lf rice 147 de 140 a-c
Pyroxasulfone DPRE 80 g 88 fg
Spiking 100 f 83 g
1- to 2-lf rice 91 fg 89 fg
Abbreviations: delayed pre-emergence (DPRE) and leaf (lf).
Means within a column followed by the same letter are not significantly different based on
Fishers protected least significant difference test (P < 0.05).
166
Fig. 1. Rice injury observed following application of

four Group 15 herbicides 2 to 3 weeks after treatment at the Rice Research
and Extension Center. Means followed by the same letter are not significantly
different according to Fishers protected least significant difference test at P = 0.05.

four Group 15 herbicides 8 to 9 weeks after treatment at the Rice Research
and Extension Center. Means followed by the same letter are not significantly
167

four Group 15 herbicides 4 to 5 weeks after treatment at the Northeast
Research and Extension Center. Means followed by the same letter are not significantly
168
Sharpen Tank-Mixtures with Rice

Herbicides for Barnyardgrass Control in Provisia Rice
R.R. Hale1, J.K. Norsworthy1, J.A. Godwin Jr.1,
N.R. Steppig1, C.J. Meyer1, R.C. Scott2, L.T. Barber2, and J. Schultz3
ABSTRACT
Provisia rice is a new non-genetically modified organism (GMO) trait resistant to
quizalofop, an acetyl-coenzyme A carboxylase (ACCase)-inhibiting herbicide (Group
1) that only controls grass weed species. Often in rice, both broadleaf and grass weeds
are present in the field and a broadleaf herbicide is needed as a tank-mixture or as a
split application when applying a selective grass herbicide. A field experiment was
conducted in the summer of 2015 at the University of Arkansas System Division of
Agriculture Pine Tree Research Station near Colt, Ark, to evaluate the possible inter-
action of Sharpen herbicide when applied in tank-mixtures with other rice herbicides
typically used for controlling barnyardgrass (Echinochloa crus-galli). This experiment
was set up as a randomized complete block design with three factors: herbicide, rate,
and the addition of Sharpen. Herbicide treatments were 1/2 and 1 labeled rates of
Sharpen (saflufenacil), Clincher (cyhalofop), Ricestar HT (fenoxaprop), and Provisia
(quizalofop - soon to be labeled). Overall, injury did not exceed 5% for any treatment,
regardless of herbicide, rate, or the addition of Sharpen at 7 and 14 days after treat-
ment (DAT). Antagonism was seen at 7 DAT for 1/2 rates of Sharpen + Clincher; and
by 14 DAT, the 1/2 rate of Sharpen + Clincher and the 1 rate of Sharpen + Ricestar
HT were also deemed antagonistic. Based on these results, tank-mixing Sharpen with
Clincher or Ricestar HT may result in a decrease in barnyardgrass control; hence, split
applications may be considered when using these herbicides if broadleaf and grass
weeds are both present in the field.
INTRODUCTION
Barnyardgrass is one of the most problematic weeds in Arkansas rice produc-
tion. In 2011, a survey was issued to Arkansas crop consultants and 68% of responses
1
Graduate Assistant, Professor, Graduate Assistant, Graduate Assistant, and Graduate Assistant, respec-
tively, Department of Crop, Soil, and Environmental Science, Fayetteville.
2
Lonoke.
3
BASF Tech Service Representative (Arkansas and Missouri Bootheel), Sherwood.
169
listed barnyardgrass as the most problematic weed in rice (Norsworthy et al., 2013).
Barnyardgrass alone can cause rice grain yield reductions of 70% if allowed to interfere
with the crop throughout the growing season (Smith, 1988).
By the early 1990s, propanil-resistant barnyardgrass began to infest Arkansas
rice fields (Carey et al., 1995). During the 1990s, 98% of Arkansas rice received at
least one application of propanil. Continued repetition of propanil as the only means
of post-emergence (POST) weed control quickly resulted in barnyardgrass evolving
resistance to this herbicide (Carey et al., 1995). Today, barnyardgrass populations exist
with resistance to propanil, quinclorac, clomazone, and imazethapyr in Arkansas rice
(Talbert and Burgos, 2007; Norsworthy et al., 2013). With high levels of resistance
evolving, new technologies are needed to help minimize or decrease the evolution of
resistance within barnyardgrass and other troublesome weeds such as red rice (Oryza
sativa). A new herbicide-resistant rice technology is being developed by BASF that will
allow applications of Provisia (quizalofop), an ACCase-inhibiting herbicide, to be used.
Currently, Clincher and Ricestar HT are ACCase-inhibiting herbicides recom-
mended and registered for grass control in rice (Scott et al., 2014). Clincher and Ricestar
HT can be tank mixed with other contact rice herbicides to provide broad-spectrum
broadleaf and grass control in rice without negatively effecting yield (Talbert et al.,
2003). Acetyl-coenzyme A carboxylase-inhibiting herbicides like Provisia, Clincher,
and Ricestar HT are commonly used graminicides that are systemic whereas Sharpen,
a protoporphyrinogen oxidase (PPOase)-inhibiting herbicide, mainly controls broad-
leaf weeds. It has been reported that when mixing a PPOase-inhibiting herbicide like
Aim (carfentrazone), a current POST option for broadleaf control in Arkansas rice,
with imazethapyr provided an increase in weed control and resulted in a higher yield
than imazethapyr alone (Zhang et al., 2006; Montgomery et al., 2015). Also, Sharpen
is an emulsifiable concentrate (EC) herbicide formulation and an increase in crop or
weed response can be oberserved when compared with other formulations of the same
herbicide (Fish et al., 2014; Montgomery et al., 2015).
To achieve effective control of both grass and broadleaf weeds, the evaluation of
combinations between herbicides is often based on Colbys method using Equation 1:
E = A + B - (A - B) Eq. 1
100
where E is the expected response when herbicides A and B are mixed, A is the efficacy
obtained with one herbicide alone, and B is the efficacy of the other herbicide when
applied alone (Colby, 1967). When the observed response is statistically greater than the
calculated expected response, the particular herbicide combination is synergistic. The
inverse response shows the herbicide combination to be antagonistic. If the expected
value and observed values do not differ, the herbicide combination is deemed additive
(Colby, 1967). When tank-mixed with Newpath (imazethapyr), a systemic herbicide,
Sharpen can cause antagonism on grass weed control due to the rapid degradation of
plant cell membranes (Camargo et al., 2012).
170
PROCEDURES
A study to evaluate the interaction of Sharpen tank-mixed with other rice herbi-
cides on barnyardgrass control was conducted in the summer of 2015 at the University
of Arkansas System Division of Agriculture Pine Tree Research Station near Colt, Ark.
Provisia rice was drill-seeded at a rate of 20 seed/ft of row in 7-inch-wide rows into 6
17 ft plots. Treatments consisted of two rates of Sharpen at 0.5 and 1 fl oz/acre. Rice
herbicides were applied at 1/2 and 1 rates alone and tank-mixed with both rates of
Sharpen, which included Clincher (cyhalofop) at 7.5 oz/acre and 15 oz/acre, Ricestar
HT (fenoxaprop) 12 oz/acre and 24 oz/acre, and Provisia (quizalofop) at 10.34 oz/acre
and 20.68 oz/acre. All treatments included crop oil concentrate (COC) at 1% v/v. Ap-
plications were applied using a CO2-pressurized backpack sprayer calibrated to deliver
15 gal/acre.
Treatments were arranged in a randomized complete block design with 3 factors:
herbicide, rate, and the addition of Sharpen. Applications were made once barnyard-
grass reached the 3- to 4-lf growth stage. Data collection included visual assessments
of rice injury and weed control at 7, 14, and 21 days after treatment (DAT). Ratings
were made on a 0 to 100 scale, with 0 being no crop injury or no weed control and 100
being complete crop death or complete control. All data were analyzed using JMP Pro
12 (SAS Institute Inc., Cary, N.C.), and means were separated using Fishers protected
least significant difference test (P < 0.05).

For injury, there were no significant interactions between factors, but there was a
main effect with the addition of Sharpen at 7 DAT. With the addition of Sharpen, injury
averaged across all other treatments was 7%, whereas treatments without Sharpen aver-
aged 4% injury (data not shown). By 14 DAT, main effects of herbicide and rate were
significant. Both Ricestar HT and Provisia caused 4% injury while Clincher caused
<1% injury (data not shown). The main effect of rate was significant, which would be
expected because applying higher rates of herbicide can result in an increase in injury. By
21 DAT, injury had dissipated to 0% for all treatments (data not shown). The observed
injury is similar to that reported by Montgomery et al. (2015) where <10% injury was
observed when applying Sharpen POST in rice.
There were no significant interactions for the control of barnyardgrass, but the
main effects of herbicide and the addition of Sharpen were significant. Treatments
containing Sharpen alone averaged 94% and 89% control at 7 and 14 DAT, respectively
(data not shown). By 14 DAT, only the main effect of herbicide was significant, with
Ricestar HT and Provisia both providing 96% control and Clincher providing 89%
control (data not shown).
The interaction of herbicide tank-mixtures using Colbys method showed that
there was significance for certain tank-mixtures (Table 1). By 7 DAT, Sharpen alone
provided 94% control (data not shown). Clincher at 7.5 oz/acre + Sharpen at 0.5 oz/
171
acre provided 89% control, and the expected control obtained from Colbys method
was 99%, resulting in the combination being deemed antagonistic (Table 1). All other
treatments were additive. By 14 DAT, Clincher at 7.5 oz/acre + Sharpen 0.5 oz/acre was
antagonistic along with Ricestar HT at 24 oz/acre + Sharpen 1 oz/acre. As for other tank
mixtures, treatments showed only additive effects. The occurrence of antagonism was
not surprising because the response often occurs when systemic and contact herbicides
are tank-mixed (Myers and Coble 1992; Zhang et al. 2005).
The significance of this research is that growers may attempt to tank-mix Sharpen
with graminicides to achieve a high level of control of both broadleaves and grasses.
Unfortunately, antagonism was sometimes observed when mixing Sharpen with gra-
minicides based on Colbys method. However, it should be noted that barnyardgrass
control was never reduced by mixing a graminicide with Sharpen, but rather control
was not as high as expected based on Colbys method. Growers should remember that
a high level of control is needed for barnyardgrass or else the likelihood of this weed
to develop resistance to the ACCase-inhibiting herbicides is high.
ACKNOWLEDGMENTS
Funding for this research was provided by the Arkansas Rice Research and Pro-
motion Board, BASF Corporation and the University of Arkansas System Division of
Agriculture.
LITERATURE CITED
Camargo, E.R., S.A. Senseman, G.N. McCauley, and J.B. Guice. 2012. Rice (Oryza
sativa) response and weed control from tank-mix applications of saflufenacil and
imazethapyr. Crop Prot. 31:94-98.
Carey, V.F., R.E. Hoagland, and R.E. Talbert. 1995. Verification and distribution
of propanil-resistant barnyardgrass (Echinochloa crus-galli) in Arkansas. Weed
Technol. 9:366-372.
Colby, S.R. 1967. Calculating synergistic and antagonistic responses of herbicide
combinations. Weeds 15:20-22.
Fish, J.C., E.P. Webster, and B.M. McKnight. 2014. Potential of saflufenacil in
Clearfield rice. Rice Technical Working Group Annual Meeting. 18-21 February
2014. New Orleans, La.
Montgomery, G.B., J.A. Bond, B.R. Golden, J. Gore, H.M. Edwards, T.W. Eubank,
and T.W. Walker. 2015. Utilization of saflufenacil in a Clearfield rice (Oryza
sativa) system. Weed Technol. 29:255-262.
Myers, P.F. and H.D. Coble. 1992. Antagonism of graminicide activity on annual
grass species by imazethapyr. Weed Technol. 6:333-338.
172
Norsworthy, J.K., J.A. Bond, and R.C. Scott. 2013. Weed management practices and
needs in Arkansas and Mississippi rice. Weed Technol. 27:623-630.
Scott, R.C., L.T. Barber, J.W. Boyd, G. Selden, J.K. Norsworthy, and N.R. Burgos.
2014. Recommended chemicals for weed and brush control. pp 83-98. University
of Arkansas Ssytem Division of Agriculture, Cooperative Extension Service MP-
44. Available at: www.uaex.edu/publications/pdf/mp44/mp44.pdf.
Smith, R.J. 1988. Weed thresholds in southern U.S. rice, Oryza sativa. Weed Technol.
2:232-241.
Talbert, R.E. and N.R. Burgos. 2007. History and management of herbicide-resis-
tant barnyardgrass (Echinochloa crus-galli) in Arkansas rice. Weed Technol. 21:
324-331.
Talbert, R.E., B.V. Otis, M.S. Malik, and A.T. Ellis. 2003. Field Evaluation of Herbi-
cides on Rice. University of Arkansas Agricultural Experiment Station, University
of Arkansas Research Series 514:61-67. Fayetteville.
Zhang, W., E.P. Webster, D.C. Blouin, and C.T. Leon. 2005. Fenoxaprop interactions
for barnyardgrass. Weed Technol. 19:293-297.
Zhang, W., E.P. Webster, K.J. Pellerin, and D.C. Blouin. 2006. Weed control pro-
grams in drill-seeded imidazolinone-resistant rice (Oryza sativa). Weed Technol.
20:956-960.
Table 1. Barnyardgrass control 7 and 14 days after treatment (DAT), including the
observed values from the field and the expected value provided by Colbys method.
7 DAT 14 DAT
Herbicide tank-mixturesa Expected Observed Expected Observed
1/2 Rate
Clincher 7.5 oz/acrecre + 99 89b*b 99 84*
Sharpen 0.5 oz/acre
Ricestar HT 12 oz/acre + 91 97 100 93
Sharpen 0.5 oz/acre
Provisia 10.34 oz/acre + 100 94 99 98
Sharpen 0.5 oz/acre

1 Rate
Clincher 15 oz/acre + 84 90 99 92
Sharpen 1 oz/acre
Ricestar HT 24 oz/acre + 100 98 100 97*
Sharpen 1 oz/acre
Provisia 20.68 oz/acre + 100 99 100 97
Sharpen 1 oz/acre
a
All treatments contained COC 1 % v/v at 1 qt/acre.
b
For a given herbicide tank-mixture, an asterisk indicates a significant student t-test for compar-
ing Expected and Observed values according to Colbys method.
173
Obey and Command Herbicide Programs

Z.T. Hill1, L.T. Barber2, J.K. Norsworthy3, and R.C. Doherty1
ABSTRACT
Barnyardgrass and Amazon sprangletop are two of the most troublesome grass weeds
in rice production in Arkansas. With resistance to propanil and quinclorac becoming
more prevalent across Arkansas, especially in barnyardgrass, other herbicides with dif-
ferent mechanisms of action will have to be considered. An experiment was conducted
on a Sharkey clay soil at the University of Arkansas System Division of Agricultures
Southeast Research and Extension Center at Monticello, Ark., to determine if the use
of clomazone (Command) alone or premixed with quinclorac, commonly known as
Obey (clomazone + quinclorac), will improve control of barnyardgrass and Amazon
sprangletop in rice. Prior to the post-emergence (POST) applications, clomazone
applied pre-emergence (PRE) at 0.8 lb ai/acre provided greater control of both grass
weeds than most other programs at 7 days after the delayed pre-emergence (DPRE)
application. Throughout the remainder of the season, clomazone PRE at 0.8 lb ai/acre
followed by (fb) Prowl H2O (pendamethalin) + Ricebeaux (propanil+thiobencarb) at the
4- to 5-lf stage of rice continued to provide > 90% control of both barnyardgrass and
Amazon sprangletop; albeit, comparable to other programs that included clomazone +
quinclorac applied PRE. Due to the increasing occurrence of barnyardgrass resistance
to quinclorac POST, quinclorac appears to have a much better fit in rice when applied
in combination with clomazone as a PRE application.
INTRODUCTION
Generally, Arkansas rice producers prefer to plant rice between mid-March and
mid-May to increase the likelihood of achieving the highest percent of relative yield
as well as controlling common rice weeds early in the season (Hardke et al., 2013).
When producers fail to plant rice prior to 1 June, young rice has difficulty growing
efficiently due to competition with troublesome grass weeds, most notably barnyard-
grass and Amazon sprangletop. With resistance to propanil and quinclorac progressing
1
Weed Program Associate and Program Associate - Weed Science, respectively, Southeast Research and
Extension Center, Monticello.
2
3
174
throughout the state, other herbicide mechanisms of action must be relied upon. Since
the early 2000s, clomazone (Command 3ME) has been utilized on > 90% of rice acres
in Arkansas (Norsworthy et al., 2007). Although clomazone-resistant barnyardgrass
populations have been confirmed in Arkansas (Norsworthy et al., 2008), incorporat-
ing it as the foundation of a rice herbicide program with or without quinclorac, and in
combination with pendimethalin (Prowl H2O), could provide effective control of the
troublesome grass weeds.
PROCEDURES
An experiment was conducted in 2015 at the University of Arkansas System
Division of Agricultures Southeast Research and Extension Center at Monticello, Ark,
to evaluate Command (clomazone) and Obey (clomazone + quinclorac) programs for
controlling barnyardgrass and Amazon sprangletop in late planted rice. This experiment
was setup in a randomized complete block design with four replications. Clearfield
111 was planted on 15 June 2015 on a Sharkey clay soil at a planting rate of 90 lb/acre.
Barnyardgrass and Amazon sprangletop seed were broadcast over the test site immedi-
ately after planting rice. Herbicide treatments were applied using a Mudmaster sprayer
equipped with a compressed air pressurized multi-boom calibrated to deliver 12 gal/
acre. Herbicide programs included clomazone at 0.4 or 0.8 lb ai/acre (PRE and POST),
clomazone+quinclorac at 0.8 lb ai/acre (PRE and POST), Prowl H2O (pendimethalin) at
0.95 lb ai/acre (DPRE and POST), and Ricebeaux (propanil + thiobencarb) at 4.5 lb ai/
acre (POST, 4- to 5-lf rice). All clomazone + quinclorac POST treatments were applied
with a crop oil concentrate at 1% v/v, while all propanil + thiobencarb treatments were
applied with a non-ionic surfactant at 0.25% v/v. Weed control and crop injury were
evaluated on a scale of 0 to 100% control, where 0 equals no control and 100 equals
complete control. Data were subjected analysis of variance and means were separated
using Fishers protected least significant difference test (P = 0.05).

No crop injury was observed following any application (data not shown). Prior to
the POST applications, clomazone at 0.4 lb ai/acre PRE fb pendamethalin DPRE and
clomazone at 0.8 lb/acre PRE were the only treatments to provide > 92% control of both
grass species 7 days after the DPRE application (Table 1). At 19 days after the POST
application, clomazone at 0.8 lb ai/acre fb pendamethalin plus propanil + thiobencarb
provided 98% and 99% control of barnyardgrass and Amazon sprangletop, respectively
(Table 2). Additional programs that provided > 95% control of both species included
clomazone + quinclorac PRE fb clomazone at 0.4 lb ai/acre + propanil + thiobencarb
POST and clomazone + quinclorac PRE fb pendamethalin DPRE fb propanil + thio-
bencarb POST. By 34 days after the POST application, a slight reduction in control
was observed for most treatments; however, the same previous treatments continued to
provide > 90% control of both species (Table 3). Additionally, clomazone + quinclorac
175
at 0.8 lb ai/acre PRE fb clomazone at 0.4 lb ai/acre + pendamethalin + propanil + thio-

bencarb POST provided comparable control of barnyardgrass and Amazon sprangletop,
with 93% and 92% control, respectively.
Throughout the course of the season, clomazone at 0.8 lb/acre PRE fb pendameth-
alin + propanil + thiobencarb POST provided excellent control of both barnyardgrass
and Amazon sprangletop; albeit, control was comparable to most of the other herbicide
programs evaluated. It can be concluded from these data, that the use of clomazone
alone or in combination with quinclorac (Obey) as the foundation of a rice herbicide
program is capable of providing effective control of these problematic grass species in
late planted rice. The results also indicate that in order to obtain complete control of
barnyardgrass and Amazon sprangletop, producers should consider overlapping residual
herbicides through use of PRE, DPRE, and EPOST timings.
ACKNOWLEDGMENTS
Sincere gratitude is extended to the Rice Research and Promotion Board, FMC
Corporation, and the University of Arkansas System Division of Agriculture for fund-
ing this research.
LITERATURE CITED
Hardke, J.T., L.T. Barber, R.D. Cartwright, P. Counce, B.C. Griggs, C. Henry, F.N.
Lee, G. Lorenz, R. Mazzanti, K. Moldenhauer, R.J. Norman, J.K. Norsworthy,
T.L. Roberts, S. Sadaka, L. Schmidt, B. Scott, T. Siebenmorgen, N. Slaton, Y.
Wamishe, B. Watkins, and C.E. Wilson. 2013. Arkansas Rice Production Hand-
book. University of Arkansas System Division of Agriculture, Cooperative Exten-
sion Service MP192, Little Rock, Ark.
Norsworthy, J.K., R.C. Scott, and K.L Smith. 2008. Confirmation and management
of clomazone-resistant barnyardgrass. In: R.J. Norman, J.-F. Meullenet, and
K.A.K. Moldenhauers (eds.). B.R. Wells Rice Research Studies 2007. Univer-
sity of Arkansas Agricultural Experiment Station Research Series 560:113-116.
Fayetteville, Ark.
Norsworthy, J.K., N.R. Burgos, R.C. Scott, and K.L. Smith. 2007. Consultant per-
spectives on weed management needs in Arkansas rice. Weed Technol. 21:832-
839.
176
Table 1. Barnyardgrass and Amazon sprangletop

control at 7 days after the delayed pre-emergence application.
Application Barnyard- Amazon
Programa Rate(s) timing grass sprangletop
(lb ai/acre) ------------------------- (%)-------------------------
Nontreated control ---- ---- 0 0
Clomazone 0.4 PRE 46 49
Clomazone + quinclorac 0.8 PRE 83 87
Clomazone fb 0.4 PRE
pendamethalin 0.95 DPRE 98 98
Clomazone + quinclorac fb 0.8 PRE
LSD (P = 0.05) 23 24
a
PRE = pre-emergence; DPRE = delayed pre-emergence; fb = followed by.
177

control at 19 days after the post-emergence application.
(lb ai/acre) ------------------------- (%)-------------------------
clomazone + quinclorac 0.8 POST (4-5 lf) 0 0
clomazone 0.4 POST (4-5 lf) 25 0
pendamethalin fb 0.95 DPRE
propanil + thiobencarb 4.5 POST (4-5 lf) 89 94
clomazone + 0.4 POST (4-5 lf)
propanil + thiobencarb 4.5 98 96
pendamethalin + 0.95 POST (4-5 lf)
propanil+thiobencarb 4.5 98 99
pendamethalin + 0.95
LSD (P = 0.05) 20 7
a
DPRE = delayed pre-emergence; fb = followed by; PRE = pre-emergence;
POST = post-emergence.
178

control at 34 days after the post-emergence application.
(lb ai/acre) ------------------------- (%)-------------------------
clomazone 0.4 POST (4-5 lf) 44 6
pendamethalin + 0.95 POST (4-5 lf)
clomazone +0.4 POST (4-5 lf)
pendamethalin + 0.95
LSD (P = 0.05) 21 13
a
DPRE = delayed pre-emergence; fb = followed by; PRE = pre-emergence;
POST = post-emergence.
179
Optimizing Quizalofop Rate Structure

for Sequential Application in Provisia Rice
Z.D. Lancaster1, J.K. Norsworthy1, S.M. Martin1,
M.L. Young1, R.C. Scott2, and L.T. Barber2
ABSTRACT
The BASF Corporation is currently developing a new non-genetically modified organ-
ism (GMO) rice trait (Provisia rice) that will be resistant to quizalofop, an acetyl
coenzyme A carboxylase (ACCase)-inhibiting herbicide. An experiment was conducted
in the summer of 2014 and 2015 at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center (RREC) near Stuttgart, Ark., to
determine the best rate structure for sequential applications of quizalofop when the
first application was made at either the 2-lf or 6-lf stage of grasses. The experiment
was set up as a two factor, randomized complete block design with factor-A being the
growth stage at first application and factor-B being the rate structure of quizalofop.
Herbicide rate structures were 10.3, 15.5, or 20.6 fl oz/acre followed by 10.3, 15.5,
or 20.6 fl oz/acre sequential application 14 days after the initial application. The total
amount of quizalofop applied in a rate structure never exceeded 31 fl oz/acre; hence,
some combinations were excluded. In 2014, the greatest control of both barnyardgrass
and broadleaf signalgrass was achieved when quizalofop was sequentially applied at the
15.5 followed by 15.5 fl oz/acre rates, resulting in 99% and 98% control, respectively.
The sequential 10.3 followed by 10.3 fl oz/acre quizalofop treatment had significantly
less barnyardgrass and broadleaf signalgrass control. In 2015, there were no significant
differences among herbicide rates. Control for barnyardgrass and broadleaf signalgrass
was reduced by making the first quizalofop application on 6-lf grass compared to 2-lf
grass for 2014, and the same effect was observed for red rice in 2015. Based on these
results, the most likely recommended rate structure for quizalofop will be 15.5 fl oz/
acre on 2-lf grasses followed by a subsequent application at approximately 14 days
after the initial application.
1
Graduate Assistant, Professor, Graduate Assistant, and Graduate Student, respectively, Department of
Crop, Soil, and Environmental Science, Fayetteville.
2
Lonoke.
180
INTRODUCTION
Rice is one of the most important crops grown in Arkansas, with a major obstacle
to rice production being weed control. Weeds compete with rice for sunlight, water,
nutrients, and other growth requirements (Smith, 1988). In a 2011 survey, 63% of
Arkansas crop consultants listed barnyardgrass (Echninochloa crus-galli) as the most
problematic weed of rice, with red rice (Oryza sativa) ranking second (Norsworthy et
al., 2013). Red rice and barnyardgrass can potentially cause yield losses as high as 82%
and 70%, respectively (Smith, 1988).
Barnyardgrass has evolved resistance to multiple herbicides used in Arkansas
rice, the first of which was propanil in the early 1990s (Carey et al., 1995). Poor stew-
ardship of alternative herbicides led to continued herbicide resistance of barnyardgrass
to quinclorac, clomazone, and the imidazolinone herbicides used in ClearfieldTM rice
(Talbert and Burgos, 2007; Norsworthy et al., 2013). With the evolution of weeds that
have resistance to multiple herbicide modes of action, weed control has increasingly
become more challenging in Arkansas rice production systems. A new technology is
needed to control many of these troublesome weeds. The BASF Corporation is cur-
rently developing a new herbicide-resistant rice technology that will allow for topical
applications of quizalofop, an acetyl coenzyme A carboxylase (ACCase)-inhibiting
herbicide. Quizalofop will be primarily used in the ProvisiaTM rice system to control
barnyardgrass, red rice, and other annual grass weeds. With the anticipated launch of
Provisia rice within the next 2 to 3 years, research is needed to understand the best fit
for this technology in Arkansas rice production systems.
PROCEDURES
An experiment was conducted to determine the best rate structure of sequential
applications of quizalofop to Provisia rice, when applied initially to either 2- or 6-lf
grass. The field experiment was conducted in the summer of 2014 and 2015 at the
RREC near Stuttgart, Ark., on a Dewitt silt-loam soil. The experiment was set up as a
two factor factorial, randomized complete block design having four replications with
factor-A being the growth stage at first application and factor-B being the rate structure
of quizalofop. Provisia rice was planted on 7 inch rows, in 6 20 ft plots, with a typical
rate of 22 seeds/foot of row. Treatments of quizalofop were applied to a natural popu-
lation of weeds. Sequential applications of quizalofop were applied, with the second
application being made 14 days after the first application. All quizalofop applications
were made at 15 gallons of spray solution per acre with a crop oil concentrate added at
1% v/v to all treatments. Broadleaf weeds and sedges were controlled by over-spraying
the test site with 2,4-D and halosulfuron (Permit). The quizalofop rate structure for this
experiment was 10.3, 15.5, or 20.6 fl oz/acre (first application) followed by a sequential
application of 10.3, 15.5, or 20.6 fl oz/acre. All combinations were evaluated that al-
lowed for no more than the maximum anticipated rate of 31 fl oz/acre to be applied over
both applications. For instance, 20.6 fl oz/acre followed by 15.5 or 20.6 fl oz/acre was
not evaluated because the total amount of herbicide exceeded the anticipated allowable
181
maximum (John Harden, BASF corp). Observations were taken on weed control with
0 being no control and 100 being complete weed control. In 2014, weed control rat-
ings were taken for barnyardgrass and broadleaf signalgrass, and in 2015, ratings were
taken for barnyardgrass, broadleaf signalgrass, and red rice. All data were subjected to
analysis of variance using JMP 12.1, and means were separated using Fishers protected
least significant difference test (P = 0.05).

In 2014, there was not a significant rate structure by growth stage interaction,
but there were significant main effects for rate structure and growth stage for both
barnyardgrass and broadleaf signalgrass control. The highest numerical control for
barnyardgrass was produced with the 15.5/15.5 fl oz/acre rate structure of 98%, but
was only significantly different from the 80/80 g ai/ha structure which resulted in 89%
control (Fig. 1). Likewise, the highest numerical control for broadleaf signalgrass was
produced with the 15.5/15.5 fl oz/acre rate structure at 99%, but again was only signifi-
cantly different from the 10.3/10.3 fl oz/acre structure which resulted in 91% control.
For barnyardgrass, when the initial application of quizalofop was made at the 2-lf
growth stage it resulted in 98% control, averaged over rates; but when the application
was initiated at the 6-lf growth stage, control declined to 87% (Fig. 2). For broadleaf
signalgrass when the initial application was made at the 2-lf stage, control averaged
98% over application rates; whereas control averaged 95% when the initial application
was delayed until the 6-lf stage.
In 2015, weed density was less than in 2014 and results slightly differed from
the previous year. Only the main effect for growth stage of initial application was sig-
nificant for red rice control in 2015. When quizalofop was initially applied at the 2-lf
growth stage, red rice control averaged 99% (Fig. 3). Delaying the application to the
6-lf growth stage resulted in average red rice control of 97%. For barnyardgrass and
broadleaf signalgrass, comparable levels of control were obtained across all application
timings and rate structures (data not shown). The lower density of grasses in 2015 likely
contributed to the inability to separate experimental treatments because herbicides are
often more efficacious as weed density decreases.
The significance of this research is that the 15.5 followed by 15.5 fl oz/acre rate
provides excellent control of barnyardgrass, broadleaf signalgrass, and red rice (> 97%)
when applied to small, actively growing grasses. A residual herbicide is likely to be
needed at planting first to remove some of the selection pressure placed on quizalofop for
resistance, especially barnyardgrass. Secondly by lowering the density of grass weeds,
present growers should have more flexibility in timing the initial quizalofop application.
182
ACKNOWLEDGMENTS
Funding and support for this research provided by the Arkansas Rice Research
and Promotion Board, and the University of Arkansas System Division of Agriculture
is gratefully appreciated. Appreciation is also expressed to BASF for their cooperation
and support of this project.
LITERATURE CITED
Technol. 9:366-372.
Norsworthy, J.K., J. Bond, and R.C. Scott. 2013. Weed management practices and
Smith, R.J. 1988. Weed thresholds in southern U.S. rice, Oryza sativa. Weed Tech-
nol. 2:232-241.
Talbert, R.E. and N.R. Burgos. 2007 History and management of herbicide-resistant
barnyardgrass (Echinochloa crus-galli) in Arkansas rice. Weed Technol. 21:324-
331.
Fig. 1. Main effect of quizalofop rate structure

with first rate applied followed by (/) second rate applied on
barnyardgrass and broadleaf signalgrass control, averaged over application
timings in 2014. Letters used to separate means within grass weed species (P = 0.0092).
183
Fig. 2. Main effect of grass growth stage at initial application of quizalofop

on barnyardgrass and broadleaf signalgrass control, averaged over sequential
rates in 2014. Letters used to separate means within grass weed species (P = 0.0215).
Fig. 3. Main effect for growth stage on red rice control with
quizalofop, averaged over sequential rates in 2015 (P = 0.0403).
184
Effects of CruiserMaxx Rice on Rice

Tolerance to Pre-Emergence Herbicides
S.M. Martin1, J.K. Norsworthy1, R.C. Scott2, G.M. Lorenz3, and J.T. Hardke4
ABSTRACT
Historically, some herbicide and insecticide interactions have proven beneficial whereas
others have been detrimental to certain crops. In-furrow applications of insecticides have
proven to be beneficial in cotton production with clomazone (Command) receiving
a label in cotton under restrictions that an in-furrow application of insecticide such as
disulfoton or phorate be used. However, when some insecticides are used in conjunction
with propanil, significant rice injury can occur. With recent evidence that insecticide
seed treatments can reduce injury in rice from drift events, a field study was developed
to test different pre-emergence herbicides for safened use in rice with the addition of
insecticide seed treatments. This study was conducted at the University of Arkansas
System Division of Agricultures Rice Research and Extension Center (RREC) near
Stuttgart, Ark., and at the University of Arkansas Pine Bluff Farm (UAPB) near Lonoke,
Ark., in 2014 and 2015. CruiserMaxx Rice was used to evaluate different pre-emergence
herbicides for safened use in rice production. At 3 weeks after planting (WAP), the
clomazone standard had significantly more plants emerge than all other treatments.
By 4 WAP, fluridone- and thiobencarb-treated plots had similar injury amounts to the
clomazone standard. Clethodim- and thiobencarb-treated plots yielded similar to the
clomazone-treated plots with yields ranging from 163 to 180 bu/acre. When averaged
across herbicide treatments, the CruiserMaxx Rice treated seed provided 9% to 10%
less injury and a 9% yield increase over the fungicide-only treated rice. In conclusion,
clethodim and thiobencarb gained interest as possible options for pre-emergence use
in rice along with CruiserMaxx Rice which provided a proven benefit to the rice crop.
INTRODUCTION
One of the major obstacles in rice production is weed control which in recent years
has become a major problem due to resistance in some of the most problematic weeds
1
Graduate Assistant, and Professor, respectively, Department of Crop, Soil, and Environmental Science,
Fayetteville.
2
Professor, Department of Crop, Soil, and Environmental Science, Lonoke.
3
Associate Department Head / Extension Entomologist, Department of Entomology, Lonoke.
4
185
in rice. Some of these problematic weeds in Arkansas rice are barnyardgrass and red
rice, which can cause yield losses of up to 80% (Norsworthy et al., 2013; Smith, 1988).
Clomazone is commonly applied at or near planting to most of the Arkansas rice
acreage for control of barnyardgrass and other grass weeds. Unfortunately, populations
of barnyardgrass now exist in Arkansas with resistance to clomazone (Norsworthy et
al., 2013). Additionally, barnyardgrass exhibits resistance to propanil, quinclorac, and
acetolactate synthase (ALS)-inhibiting herbicides, making control options for this weed
somewhat limited (Heap, 2016).
In addition to barnyardgrass, red rice has evolved resistance to ALS-inhibiting
herbicides since the introduction of Clearfield rice (Talbert and Burgos, 2007). With
the two most problematic weeds in rice developing resistance to multiple herbicide
modes of action, new modes of action are needed to combat these ongoing resistance
problems. The last new mode of action to be commercialized in crops was discovered
in the early 1980s; therefore, the best method to combat resistance would be to label
an already existing herbicide that is not currently used in rice.
Previously, the use of insecticides has enabled the labeling of a herbicide in a
crop that was sensitive in the absence of the insecticide. York et al. (1991) found that
the addition of phorate or disulfoton used in-furrow in cotton would allow for cloma-
zone herbicide to be safely used. Most recently, CruiserMaxx Rice, a fungicide and
insecticide seed treatment, has been proven to provide multiple benefits to the rice crop,
including safening against low rates of glyphosate and imazethapyr (Scott et al., 2014).
Additionally, CruiserMaxx Rice provides the expected benefits of insect control and
overall better plant vigor (Wilf et al., 2010; Plummer et al., 2012).
PROCEDURES
An experiment was conducted to determine if the use of a CruiserMaxx Rice
seed treatment could safen the crop against herbicides not currently labeled for pre-
emergence use. Experiments were conducted at the University of Arkansas Pine Bluff
Farm (UAPB) near Lonoke, Ark., and at the Rice Research and Extension Center
(RREC) near Stuttgart, Ark., in the summers of 2014 and 2015. Both locations were
planted with CL152 rice at a seeding rate of 75 lb/acre (approximately 20 seed/ft of
row) in 7.5-inch-wide rows.
The study was organized using a randomized complete block design with four
replications and two factors. Factor A, insecticide seed treatment, consisted of Cruiser-
Maxx Rice (7 oz/cwt) and a fungicide-only check. Factor B consisted of pethoxamid
(0.5 lb ai/acre), fluridone (0.2 lb ai/acre), pyroxasulfone (0.1 lb ai/acre), S-metolachlor
(0.955 lb ai/acre), thiobencarb (6 lb ai/acre), clethodim (0.12 lb ai/acre), quizalofop
(0.106 lb ai/acre), and clomazone (0.6 lb ai/acre), the current standard, all applied im-
mediately after planting at 15 gal/acre. The plots were kept weed-free with standard
rice herbicides and all production practices followed University of Arkansas System
Division of Agricultures Cooperative Extension Service (CES) recommendations
throughout the season.
Data collection included stand counts, injury estimates, and rough-rice grain yield.
Data were subjected to analysis of variance using JMP Pro 11 (SAS Institute Inc., Cary,
186
N.C.) with site years and replications considered random effects. Seed treatments and
herbicides were considered fixed effects. Fishers protected least significant difference
test was used to separate means (P < 0.05).

Rice in all plots emerged 1 to 2 weeks after planting (WAP), and stand counts
were taken 3 WAP. There was no significant interaction between the effects of seed
treatment and herbicide for rice stands; but rather only main effects were significant.
The use of CruiserMaxx Rice did provide an increase of 17% in rice emergence when
averaged across all herbicides. Averaged over seed treatments, the standard clomazone
treatment had 11.5 plants/ft of row, which was significantly greater than stands in all
other treatments (Table 1). Fluridone- and thiobencarb-treated plots had the next highest
density, with stands of 1.5 to 2.0 plants/ft of row less than clomazone.
By 4 WAP, visual injury symptoms began to occur in all plots along with thinned
stands in some treatments. All treatments had some injury including the standard
clomazone treatment. Clomazone-, thiobencarb-, and fluridone-treated plots had the
least amount of injury at 17%, 23%, and 23%, respectively (Table 1). S-metolachlor
and pyroxasulfone had 75% or more injury at 4 WAP, mainly due to reduced rice stands.
Injury at 8 WAP began to show more differences. S-metolachlor and pyroxasulfone
remained the most injured with at least 90% visual injury. Fluridone-treated plots had
not recovered from injury while the clomazone-treated plots had only 9% injury at 8
WAP (Table 1). Also averaged across herbicides, the CruiserMaxx Rice plots were 9%
to 10% less injured than the fungicide-only treatments at both 4 and 8 WAP.
Clomazone remained the highest yielding treatment along with thiobencarb and
clethodim with yields ranging from 163 to 180 bu/acre (Table 1). All treatments with
the exception of S-metolachlor and pyroxasulfone had grain yields of over 140 bu/acre.
In addition, the CruiserMaxx Rice provided a 9% yield increase over the fungicide-only
treatments across all herbicides.
Thiobencarb is labeled in rice as a delayed pre-emergence treatment; however, it
is not labeled for pre-emergence use because rice stands can be reduced as observed in
this research. The level of safening from the CruiserMaxx Rice seed treatment could
possibly allow for thiobencarb to be applied as a pre-emergence application. Overall, the
CruiserMaxx Rice seed treatment did slightly safen the rice to non-labeled herbicides;
albeit, the safening was not sufficient for commercial use of most evaluated herbicides.
However, some herbicides identified in this research would warrant further evaluations.
Clethodim can be used in rice currently for burndown applications, but has a 30-
day plant back interval which possibly could be shortened if rice seed is treated with
CruiserMaxx Rice. Pethoxamid is also a potential candidate for further evaluation in
rice, especially timings other than at planting.
187
Although there were no significant interactions between herbicides and the

CruiserMaxx Rice for any of the evaluated parameters, CruiserMaxx Rice did provide
several noticeable benefits to the rice crop. CruiserMaxx Rice provided about a 10%
increase in yield and reduced overall injury at each rating. With these added benefits
along with those identified in previous research such as safening from herbicide drift,
increased plant vigor and insect control (Wilf et al., 2010; Plummer et al., 2012; Scott
et al., 2014), CruiserMaxx Rice should become a staple component of Arkansas rice
production.
ACKNOWLEDGMENTS
Special thanks to the Rice Research and Promotion Board, the University of
Arkansas System Division of Agriculture, and Syngenta Crop Protection for providing
funding and support for this project.
LITERATURE CITED
Heap, I. 2016. The International Survey of Herbicide Resistant Weeds. Access
date: 11 January 2016. Available at: http://weedscience.org/summary/species.
aspx?WeedID=79
Plummer, W.A., G.M. Lorenz III, N.M. Taillon, B.C. Thrash, J.W. Fortner, and C.K.
Colwell. 2012. Impact of insecticide seed treatments in large block field trials
in Arkansas 2009-2011. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R.
Wells Rice Research Studies 2011. University of Arkansas Agricultural Experi-
ment Station Research Series 600:128-132. Fayetteville.
Scott, R.C., G.M. Lorenz III, J.T. Hardke, B.M. Davis, and J.W. Dickson. 2014. Use
of CruiserMaxx Rice seed treatment to improve tolerance of conventional rice
to Newpath (imazethapyr) and Roundup (glyphosate) at reduced rates. In: R.J.
Norman and K.A.K. Moldenhauer (eds.). B.R. Wells Rice Research Studies 2013.
233. Fayetteville.
Smith, R.J. 1988. Weed thresholds in southern U.S. rice, Oryza sativa. Weed Tech-
nol. 2:232-241.
Talbert, R.E. and N.R. Burgos. 2007. History and management of herbicide-resistant
barnyardgrass (Echinochloa crus-galli) in Arkansas rice. Weed Technol. 21:324-
331.
Wilf, H., G. Lorenz III, K. Colwell, and N. Taillon. 2010. Comparing the efficacy
of insecticide seed treatments at three seeding rates. In: R.J. Norman and K.A.K.
Moldenhauer (eds.). B.R. Wells Rice Research Studies 2009. University of Arkan-
sas Agricultural Experiment Station Research Series 591:104-109. Fayetteville.
York, A.C., D.L. Jordan, and R.E. Frans. 1991. Insecticides modify cotton (Gossy-
pium hirsutum) response to clomazone. Weed Technol. 5:729-735.
188
Table 1. Effect of various herbicides applied

immediately after planting CL152 rice on plant density, visible
estimates of injury, and rough-rice grain yield averaged over seed treatments.
Injury Yield
Treatment
Densitya 4 WAPb 8 WAP (relative yield)c
(plants/10 row ft) ------------------(%)----------------- [bu,acre (%)]
Clomazone 115 17 9 180 (100)
Pethoxamid 65 60 44 144 (80)
Fluridone 97 23 25 143 (79)
Pyroxasulfone 53 75 90 43 (24)
S-metolachlor 44 78 93 43 (24)
Thiobencarb 94 23 17 163 (91)
Clethodim 73 39 29 164 (91)
Quizalofop 70 40 35 146 (81)
LSD 14 12 11 19
a
Density measured for 10 feet of row 3 WAP.
b
WAP = weeks after planting.
c
Relative yield compared to the current standard of clomazone.
189
Evaluation of ProvisiaTM Herbicide Tank Mixtures

M.S. McCown1, L.T. Barber2, J.K. Norsworthy1,
Z.D. Lancaster1, J. Green1, and M.R. Miller1
ABSTRACT
The BASF Corporation is currently developing a new non-genetically modified (GMO)
rice trait that will be resistant to quizalofop, an acetyl coenzyme A carboxylase (AC-
Case)-inhibiting herbicide. Along with this new trait, BASF will be marketing the
herbicide quizalofop under the tradename ProvisiaTM. An experiment was conducted
in 2015 at the University of Aransas System Division of Agricultures Southeast Re-
search and Extension Center (SEREC) in Monticello, Ark., and at the Rice Research
and Extension Center (RREC) near Stuttgart, Ark., to evaluate early post-emergence
(EPOST) tank mixtures containing Provisia herbicide in Provisia rice. In this study,
nine common rice herbicides were evaluated in combination with Provisia herbicide
for weed control and crop tolerance. Tank mixture candidates included: quinclorac
(Facet), pendimethalin (Prowl H20), saflufenacil (Sharpen), carfentrazone (Aim),
penoxsulam (Grasp), bispyribac-sodium (Regiment), halosulfuron (Permit), propanil
+ quinclorac (Duet), and propanil + thiobencarb (Ricebeaux). All treatments were
applied at the 1- to 3-lf stage of rice (EPOST) and followed by (fb) quizalofop (Provisia)
applied prior to flooding (PREFLD). As a result of some tank mixes (Aim or Sharpen),
slight injury was observed on Provisia rice at both locations; however, no more than
10% injury was observed with any tank mixture. At 22 days after EPOST applications,
adequate control of barnyardgrass was seen in those tank mixes that contained more
than one mode of action; however, these differences were no longer present 10 days
following the mid-POST application. Similarly, Amazon sprangletop control increased
when a tank mix was made with Provisia herbicide. At SEREC, antagonism was ob-
served when propanil + quinclorac was mixed with Provisia, with no more than 60%
control of barnyardgrass or Amazon sprangletop being observed. Red rice control at
SEREC was >89% with all tank mixtures and Provisia alone after the first application
and 99% control after the second application timing. Red rice control at RREC was
75% to 90% after the first application. The two application timings were sufficient to
provide 99% control of all off-type rice cultivars. We conclude that having a tank-mix
partner with Provisia herbicide is beneficial in controlling weedy grasses and red rice.
1
Graduate Assistant, Professor, Graduate Assistant, Graduate Assistant, and Graduate Research
Assistant, Department of Crop, Soil, and Environmental Science, Fayetteville.
2
190
INTRODUCTION
Weed control is continuing to be a major obstacle in rice production. In a 2011
survey, 63% of Arkansas crop consultants listed barnyardgrass as the most problem-
atic weed of rice with red rice ranking second (Norsworthy et al., 2013). Red rice and
barnyardgrass can potentially cause yield losses as high as 82% and 70%, respectively
(Smith, 1988).
Barnyardgrass has evolved resistance to multiple herbicides used commonly in
Arkansas rice, the first of which was propanil in the early 1990s (Carey et al., 1995).
Poor stewardship of alternative herbicides led to continued herbicide resistance of barn-
yardgrass to quinclorac, clomazone, and the imidazolinione herbicides used in Clearfield
rice (Talbert and Burgos, 2007; Norsworthy et al., 2013). With the evolution of weeds
that have resistance to multiple herbicide modes of action, a new technology is needed
to control many troublesome weeds. The BASF Corporation is currently developing
a new herbicide-resistant rice technology that will allow for topical applications of
quizalofop, an acetyl coenzyme A carboxylase (ACCase)-inhibiting herbicide. Quiza-
lofop will be marketed as ProvisiaTM herbicide, and primarily used in the Provisia rice
system to control barnyardgrass and red rice. With the anticipated launch of Provisia
rice within the next 2 to 3 years, research is needed to understand the best fit for this
technology in Arkansas rice production systems.
PROCEDURES
This experiment was designed to evaluate the potential of various tank mixtures
that could be applied in a Provisia rice production system. The field experiment was
conducted in 2015 at the Southeast Research and Extension Center in Monticello, Ark.,
and at the Rice Research and Extension Center near Stuttgart, Ark., to evaluate early post-
emergence (EPOST) tank mixtures containing Provisia herbicide in Provisia rice. The
experiment was set up as a randomized complete block. Nine common rice herbicides
were evaluated in combination with Provisia herbicide for weed control and crop toler-
ance. Tank mixture candidates included: quinclorac (Facet) at 32 oz/acre (0.42 kg ai/
ha), pendimethalin (Prowl H2O) at 15.4 oz/acre (1.12 kg ai/ha), saflufenacil (Sharpen)
at 0.75 oz/acre (0.0187 kg ai/ha), carfentrazone (Aim) at 3.2 oz/acre (0.056 kg ai/ha),
penoxsulam (Grasp) at 2.4 oz/acre (0.042 kg ai/ha), bispyribac-sodium (Regiment)
at 0.93 oz/acre (0.052 kg ai/ha), halosulfuron (Permit) at 1 oz/acre (0.052 kg ai/ha),
propanil + quinclorac (Duet) at 96 oz/acre (3.38 kg ai/ha), and propanil + thiobencarb
(Ricebeaux) at 96 oz/acre (5.04 kg ai/ha). All treatments contained crop oil concentrate
at 1% volume per volume (v/v). Treatments were applied at the 1- to 3-lf stage of rice
(EPOST) and followed by (fb) quizalofop (Provisia) applied prior to flooding (PRE-
FLD). Visual observations were taken for crop injury and weed control on a scale of 0
to 100, with 0 being no injury or weed control and 100 being complete crop death or
weed control. At both locations, weed control was evaluated on barnyardgrass and red
rice. In addition, Amazon sprangletop was evaluated at SEREC. All data were processed
using analysis of variance with JMP 12.1 (SAS Institute Inc., Cary, N.C.), and means
were separated using Fishers protected least significant difference test (P = 0.05).
191

As a result of some tank mixes (Aim or Sharpen), slight injury was observed on
Provisia rice at both locations; however, no more than 7% injury was observed with
any tank mixture. Weed control was evaluated on barnyardgrass and red rice at both
locations. In addition, Amazon sprangletop and some off-type rice cultivars were evalu-
ated at the SEREC. At 22 days after the EPOST applications, the greatest barnyardgrass
control was seen in those tank mixes that contained more than one mode of action,
such as halosulfuron, at both locations (Figs. 1 and 2); however, these differences were
no longer present 10 days following the mid-POST application. A similar response
was observed for Amazon sprangletop where increased control was seen when a tank
mix was made with Provisia (Fig. 3). At the SEREC, some antagonism was observed
when propanil + quinclorac was mixed with Provisia, resulting in only 60% control
of barnyardgrass or Amazon sprangletop (Figs. 2 and 3). The herbicide combinations
with Provisia herbicide were more efficacious at SEREC than at RREC. At SEREC,
>89% red rice control was observed with all tank mixtures and Provisia alone after
the first application and 99% control after the second application (data not shown). At
RREC, red rice control ranged from 75% to 90% after the first application (Fig. 4).
After two applications, 99% control of red rice was observed with all treatments. From
these results, we conclude that having a tank mixing partner with Provisia herbicide is
beneficial in controlling weedy grasses and off-type rice cultivars, including red rice.
The significance of this research is primarily to determine potential tank mix
partners with Provisia herbicide to achieve complete weed control and reduce crop
injury. The results from this experiment demonstrate that there is little or no risk of
injury to Provisia rice from the herbicides evaluated. In addition, we can conclude that
having a tank mix partner, such as Facet L + Prowl H2O, with Provisia is beneficial in
controlling broadleaf species, red rice, and troublesome grass weeds.
ACKNOWLEDGMENTS
This research was funded by the Arkansas Rice Research and Promotion Board
LITERATURE CITED
Technol. 9:366-372.
192
Smith, R.J. 1988. Weed thresholds in southern U.S. rice, Oryza sativa. Weed Technol.
2:232-241.
Talbert, R.E. and N.R. Burgos. 2007. History and management of herbicide-resistant
barnyardgrass (Echinochloa crus-galli) in Arkansas rice. Weed Technol. 21:324-334.
Fig. 1. Control of barnyardgrass visually rated 21 days

after early post-emergence (EPOST) and 21 days following the preflood
(PREFLD) application at the Rice Research and Extension Center near
Stuttgart, Ark. Means followed by same letter do not significantly differ at P = 0.05.
193
Fig. 2. Control of barnyardgrass visually rated 21 days

following the early post-emergence (EPOST) applications and 10 days following
the preflood (PREFLD) application at the Southeast Research and Extension Center
near Monticello, Ark. Means followed by same letter do not significantly differ at P = 0.05.
Fig. 3. Control of Amazon sprangletop visually rated 21 days

following the early post-emergence (EPOST) applications and 10 days following
the preflood (PREFLD) application at the Southeast Research and Extension Center
near Monticello, Ark. Means followed by same letter do not significantly differ at P = 0.05.
194
Fig. 4. Control of red rice visually rated 21 days

following early post-emergence (EPOST) and 21 days following the
preflood (PREFLD) application at the Rice Research and Extension Center near
Stuttgart, Ark. Means followed by same letter do not significantly differ P = 0.05.
195
Influence of Herbicide Rate, Application Volume,

and Adjuvant Use on Efficacy of Rinskor Active
M.R. Miller1 and J.K. Norsworthy1
ABSTRACT
As the evolution of herbicide resistance continues, the development of a herbicide with
a new active ingredient is needed. The introduction of Loyant herbicide with Rins-
kor Active brings a valuable new tool to weed control by providing an alternative
mode of action for use in rice that provides broad-spectrum post-emergence control of
broadleaf, grass, and sedge species. A field experiment was conducted in the summer of
2014 and repeated in 2015 to evaluate the influence of herbicide rate, spray volume, and
adjuvant use on the efficacy of Rinskor on problematic weeds in rice. Factors included
0.015 and 0.03 lb ai/acre of Rinskor formulated as a soluble concentrate (SC) applied
at three application volumes: 5, 10, and 20 gal/acre across four rates of methylated
seed oil (MSO): 0, 16, 32, and 48 fl oz/acre. Weeds evaluated included barnyardgrass,
hemp sesbania, yellow nutsedge, and Palmer amaranth planted in a non-flooded dryland
setting. No significant differences were observed between years; therefore, years were
combined. Rinskor at 0.03 lb/acre provided greater control than 0.015 lb/acre, regard-
less of spray volume or MSO rate, and control with 0.03 lb/acre improved as spray
volume and MSO rate increased.
INTRODUCTION
Today, rice producers in the mid-southern United States face many challenges.
Among these is achieving control of barnyardgrass which continues to be the most
problematic weed in Arkansas rice (Norsworthy et al., 2013), and is historically one
of the most problematic weeds in the world (Holm et al., 1997). The high level of
competiveness and ever-present risk for the evolution of herbicide resistance makes
barnyardgrass particularly concerning. Currently, barnyardgrass has evolved resistance
to at least 9 sites of herbicide action worldwide and at least 7 sites of action in the
United States (Heap, 2015). In Arkansas rice, barnyardgrass has evolved resistance to
propanil, quinclorac, clomazone, and acetolactate synthase (ALS)-inhibiting herbicides
(Lovelace et al., 2002; Norsworthy et al., 2009).
1
Graduate Research Assistant and Professor, respectively, Department of Crop, Soil, and Environmental
196
LoyantTM herbicide with Rinskor ActiveTM (no common name is yet approved) is
a new herbicide discovered by Dow AgroSciences and has potential for use in global
rice culture. The compound will provide an alternative mode of action (MOA) in rice
thereby providing effective control of propanil-, quinclorac-, clomazone-, and ALS-
resistant barnyardgrass, ALS-resistant rice flatsedge, smallflower umbrella sedge, yellow
nutsedge, and other troublesome weeds in rice.
PROCEDURES
A field experiment was conducted during 2014 and repeated in 2015 at the Uni-
versity of Arkansas System Division of Agricultures Agricultural Experiment Station
in Fayetteville, Ark., to determine efficacy of Rinskor. In both years, plots measured 10
ft 20 ft and consisted of two rows each of barnyardgrass, hemp sesbania, and yellow
nutsedge planted across the plots perpendicular to herbicide application. The site in-
cluded a population of glyphosate-resistant (GR) Palmer amaranth. The experiment was
arranged as a randomized complete block design with a three-factor factorial treatment
structure and four replications. The first factor consisted of herbicide rate either 0.015 or
0.03 lb ai/acre of Rinskor Active formulated as a soluble concentrate (SC). The second
factor consisted of three application volumes: 5, 10, and 20 gal/acre. The third factor
consisted of three rates of methylated seed oil (MSO): 0, 16, 32, and 48 fl oz/acre.
Treatments were applied at the 3- to 4-lf growth stage of all weed species. Herbicides
were applied using a CO2-pressurized backpack sprayer calibrated to deliver the ap-
propriate application volume. In order to achieve the correct spray volume and similar
droplet spectra, 11001 TT nozzles were used for 5 and 10 gal/acre, while 11002 TT were
used to achieve 20 gal/acre of spray volume. Visual estimates for barnyardgrass, hemp
sesbania, yellow nutsedge, and Palmer amaranth control were estimated on a scale of
0% to 100% with 0% representing no control and 100% representing complete control
at 28 days after application. Data were subjected to analysis of variance (ANOVA) in
JMP Pro 12 (JMP Pro 12, SAS Institute Inc., Cary, N.C.). Where the ANOVA indicated
significance, means were separated using Fishers protected least significant difference
test (P = 0.05).

No significant differences were observed between years; therefore, years were
combined. The ANOVA indicated a significant three-way interaction between rate of
Rinskor applied, application volume, and adjuvant use for all parameters evaluated
(Tables 1 and 2). Poor levels of control of all species were observed when no MSO was
added. Barnyardgrass control 28 days after application was the greatest when Rinskor
Active was applied at either 10 or 20 gal/acre of spray volume in conjunction with 48
fl oz/acre of MSO (Table 1). In contrast, low levels of barnyardgrass control were ob-
served when Rinskor was applied at 5 gal/acre of spray volume and only contained 0 or
16 fl oz/acre of MSO. A similar trend was also observed for yellow nutsedge (Table 2).
197
Hemp sesbania in general exhibited high levels of sensitivity to applications of Rinskor,

but control tended to improve as spray volume and MSO increased (Table 1). Similar
to hemp sesbania, Palmer amaranth, a problematic weed on rice levees, displayed sen-
sitivity to applications of Rinskor and control tended to improve as spray volume and
MSO increased. Behavior of all weeds species evaluated in this trial may have been
attributed to the dryland environment in which they were subjected and higher levels of
control should occur in a flooded environment typical of mid-South rice culture. Even
so, all species exhibited a high level of sensitivity to Rinskor especially in situations
where the application parameters were conducive, thereby indicating the potential for
this new herbicide to control problematic weeds in rice.
The significance of this research is primarily for the optimization of Loyant her-
bicide containing Rinskor Active on problematic weeds in mid-South rice. Increasing
rates of MSO tend to improve overall weed control, and excellent coverage was achieved
with spray volumes of 10 gal/acre. Higher spray volumes (e.g., 20 gal/acre) may be
needed in areas with dense weed populations. Loyant herbicide with Rinskor Active
will provide mid-South rice growers with an alternative herbicide mode of action that
is capable of achieving a high level of weed control.
ACKNOWLEDGMENTS
The support for this research provided by Dow AgroSciences, the Arkansas Rice
Research and Promotion Board, and the University of Arkansas System Division of
Agriculture is appreciated.
LITERATURE CITED
Heap, I. 2015. The International Survey of Herbicide Resistant Weeds. Access date: 18
December 2015. Available at: http://www.weedscience.org/summary/Species/aspx
Holm, L.G., D.L. Plucknett, J.W. Pancho, and J.P. Herberger. 1997. The worlds
worst weeds. Distribution and Biology. Univ. Press Hawaii, Honolulu, Hawaii.
609 pp.
Lovelace, M., R.E.Talbert , B.W. Skulman, and E.F. Scherder. 2002. Evaluation of
physiological responses in quinclorac-resistant and -susceptible barnyardgrass.
Proc. South. Weed Sci. Soc. 55:114.
Norsworthy, J.K., R. Scott, K. Smith, J. Still, L.E. Estorninos Jr., and S. Bagarwa.
2009. Confirmation and management of clomazone-resistant barnyardgrass in
rice. Proc. South. Weed Sci. Soc. 62:210.
198
Table 1. Influence of herbicide rate, application volume, and adjuvant use on

barnyardgrass and hemp sesbania control 28 days after application of RinskorTM Active.
Control
Herbicide rate Application volume MSO rate Barnyardgrass Hemp sesbania
(lb ai/acre) (gal/acre) (fl oz/acre) ------------------------ (%)----------------------
0.015 5 0 0 j 0m
16 48 i 78 j
32 49 i 80 i
48 59 h 80 h
10 0 0 j 10 l
16 74 f 85 e
32 84 e 85 e
48 84 e 90 d
20 0 0 j 10 l
16 89 d 95 b
32 95 bc 95 b
48 95 bc 97 a

0.030 5 0 0 j 10 l
16 69 g 84 d
32 74 f 85 d
48 84 e 89 c
10 0 0 j 11 l
16 81 e 89 c
32 93 c 93 b
48 97 ab 99 a
20 0 0 j 17 k
16 94 bc 98 a
32 95 bc 99 a
48 99 a 99 a
MSO = methylated seed oil.

Means followed by the same letter are not statistically different according to Fishers protected
199
Table 2. Influence of herbicide rate, application volume, and adjuvant use on yellow
nutsedge and Palmer amaranth control 28 days after application of RinskorTM Active.
Control
Herbicide rate Application volume MSO rate Yellow nutsedge Palmer amaranth
(lb ai/acre) (gal/acre) (fl oz/acre) ------------------------ (%)----------------------
0.015 5 0 0 j 0l
16 41 i 49 h
32 44 i 69 g
48 44 i 74 f
10 0 0 j 4l
16 64 h 84 e
32 74 f 89 c
48 84 f 94 b
20 0 0 j 9k
16 89 g 94 d
32 95 ab 97 c
48 97 ab 97 b

0.030 5 0 0 j 5l
16 61 e 79 b
32 74 c 89 b
48 95 ab 94 a
10 0 0 j 0j
16 84 b 94 b
32 93 bc 94 a
48 97 ab 98 a
20 0 0 j 14 i
16 97 ab 94 b
32 99 a 97 a
48 99 a 99 a
MSO = methylated seed oil.

Means followed by the same letter are not statistically different according to Fishers protected
200
Evaluation of Topramezone
(Armezon/Impact) for Rice Weed Control
R.C. Scott1, J.K. Norsworthy2, J.C. Moore1, and B.M. Davis1
ABSTRACT
A field trial was conducted in the late summer of 2015 to evaluate weed control and rice
tolerance to topramezone (Armezon/Impact) herbicide. Armezon 2.8 L was used as
the source of topramezone. Treatments included Armezon applied alone at 0.25, 0.50,
0.75, and 1.0 fl oz/acre or at 0.50 fl oz/acre in combination with 32 fl oz/acre of Facet
L, 20 fl oz/acre of RiceStar HT, or 0.67 oz/acre of Permit Plus. Armezon applied at
the 3- to 4-lf rice stage resulted in 30% crop injury or more at 14 days after treatment
(DAT) when applied alone at 1.0 fl oz/acre or at 0.5 fl oz/acre in combination with
RiceStar HT. All other rates and combinations resulted in 15% injury or less. Armezon
alone at 0.5 fl oz/acre injured rice only 3% at 14 DAT. Armezon applied alone at 0.5 fl
oz/acre or higher resulted in over 90% control of barnyardgrass, Amazon sprangletop,
and hemp sesbania when evaluated at 28 DAT. No antagonism was observed between
Armezon and its tank-mix partners.
INTRODUCTION
Approximately 50% of the rice grown in Arkansas is Cleafield rice and receives
applications of the herbicides Newpath (imazethapyr) or Beyond (imazamox) (Wilson
et al., 2010). The other 50% of rice grown in the state lacks the Clearfield tolerance trait
and is grown using conventional rice herbicide chemistries. Over the years, repeated
use of the same herbicides has resulted in the development of resistant barnyardgrass
biotypes to propanil (Weed Science Society of America; WSSA Group 7), the acetolac-
tate synthase (ALS) chemistry (WSSA Group 2) such as Newpath and Regiment, and
Facet (WSSA Group 4) herbicides (Carey et al., 1995; Norsworthy et al., 2013a, b;
Wilson et al., 2014). Hence, there has been a heavy reliance on Group 1 and Group 13
herbicide chemistries such as RiceStar HT and Command, respectively. Resistance
to these chemistries, while not currently widespread, has been documented in species
including barnyardgrass and Amazon sprangletop (Norsworthy et al., 2013a,b). Although
1
Professor, Program Technician - Weed Science, and Program Associate - Weed Science, respectively,
Department of Crop, Soil, and Environmental Science, Lonoke.
2
201
some new herbicide chemistries for rice are being developed, many are in the same
basic herbicide classes or do not effectively control all resistant weeds. Therefore, any
new herbicide mode of action for use in rice would be valuable in preventing further
development of resistant weeds, especially barnyardgrass.
Topramezone is sold under the trade names Armezon and Impact for use in corn.
It is a member of the HPPD class of herbicides (WSSA Group 27; WSSA Herbicide
Handbook, 2014). It controls weeds by inhibition of the enzyme 4-hydroxyphenyl-
pyruvate dioxygenase (4-HPPD) which effects carotenoid biosynthesis. This results in
bleaching and chlorosis of susceptible plant leaves and ultimately death. Currently, there
are no other herbicides labeled for use in rice from this family of chemistry, although
at least one other is currently under development (Davis et al., 2014). However, that
product (i.e., benzobicyclon) can only be applied post-flood, which will likely limit its
potential for widespread use.
The objective of this research was to evaluate the potential of topramezone for
weed control and crop tolerance in rice and assess the value of further study.
PROCEDURES
This experiment was conducted at the University of Arkansas System Division of
Agricultures Pine Bluff Research Farm located just north of Lonoke, Ark., in the sum-
mer of 2015. The soil texture is a silt loam with a pH of 6.3. Clearfield rice CL151 was
planted on 5 August 2015 using a Hege plot drill calibrated to deliver a seeding rate of
90 lb/acre on 7.5-in. wide rows. Plot size was 5 ft 25 feet. The study was conducted
with a randomized complete block design having four replications.
Treatments consisted of topramezone as the formulated product Armezon (2.8
lb ai/gal) applied alone at 0.25, 0.50, 0.75, and 1.0 fl oz/acre or at 0.50 fl oz/acre in
combination with 32 fl oz/acre of Facet L, 20 fl oz/acre of RiceStar HT, or 0.67 oz/acre
of Permit Plus. The treatments were applied early post-emergence when rice was at
the 3- to 4-lf growth stage with a CO2-backpack sprayer calibrated to deliver a spray
solution at 10 gal/acre. Also at the time of application, barnyardgrass (2- to 3-lf), hemp
sesbania (2- to 3-lf), and Amazon sprangletop (1- to 2-lf) were present. All treatments
included 1% v/v crop oil concentrate. The rice was grown according to University of
Arkansas System Division of Agriculture Cooperative Extension Service recommen-
dations with the exception of a late planting date (17 August 2015); therefore, plots
were not harvested.
Data collected included estimates of visible injury and control ratings for barn-
yardgrass, Amazon sprangletop, and hemp sesbania at 7, 14, and 28 days after treat-
ment (DAT). Data were subjected to analysis of variance and Fishers least significant
difference test used for mean separation at P = 0.05 level of significance using ARM
9.1.4 (Gylling Data Management, Inc., Brookings, S.D.).

Armezon applied at 0.25 to 0.50 fl oz/acre resulted in less than 5% crop injury at
all times evaluated (Table 1). At 7 DAT, the 0.5 fl oz/acre rate of Armezon applied in
202
combination with RiceStar HT (20 fl oz/acre) injured rice 16%, 12 percentage points
more than Armezon applied alone at the same rate. By 14 DAT, 1.0 fl oz/acre of Armezon
alone and Armezon at 0.5 fl oz/acre plus RiceStar HT resulted in 30% and 36% injury,
respectively. These data indicate a synergistic effect on rice injury between Armezon
and RiceStar HT; however, the research needs to be repeated to confirm this observation.
Armezon at 0.75 fl oz/acre injured rice 15% at 14 DAT. By 28 DAT, only Armezon
plus RiceStar HT cause rice injury greater than 12% (Table 1). Based on these data, the
rice variety evaluated does express tolerance to near labeled rates of Armezon for
corn. Further studies are needed to look at multiple varieties and include at least one
additional rate (0.625 fl oz/acre) of Armezon to fully evaluate rice tolerance.
Armezon applied alone at 0.5 fl oz/acre or more and in the various tank-mixtures
provided at least 90% control of barnyardgrass, hemp sesbania, and Amazon sprangletop
at 28 DAT (Table 1). The addition of Facet L, RiceStar HT, or Permit Plus did not im-
prove weed control at 28 DAT. However, at 14 DAT, the addition of Facet L increased
hemp sesbania control to 100%. No antagonism was observed from the three tank-mix
partners evaluated. These treatments should be repeated and future research should
include: Command, propanil, Provisia, Newpath, Rinskor Active, and Clincher to aid
in further understanding the fit of Armezon in various rice weed control programs.
The results of this exploratory study indicate that it may be possible to refine the
rate and timing of application of topramezone to fit a drill-seeded rice production system
in Arkansas. Good control of barnyardgrass was observed which could mean a new
mode of action for control of resistant biotypes of this weed in rice. This could have a
significant impact on the on-going development of resistant barnyardgrass in Arkansas.
Over the next 2 years, we will take an in-depth look at the use of topramezone in rice
and hopefully have a product that warrants labeling in mid-South rice.
ACKNOWLEDGMENTS
Special appreciation is extended to the Rice Research and Promotion Board and
the University of Arkansas System Division of Agriculture for providing funding and
support for this project. The authors also acknowledge Cletus Youmans with BASF for
bringing this project to our attention.
LITERATURE CITED
Technol. 9:366-372.
Davis, B.M., R.C. Scott., C.A. Sandoski, L.T. Barber, and J.K. Norsworthy. 2014.
Weed control demonstration of five rates of benzobicyclon applied at two main-
203
tained flood depths to rice weeds. In: R.J. Norman and K.A.K. Moldenhaur (eds.).
B.R. Wells Rice Research Studies 2013. University of Arkansas Agricultural
Experiment Station Research Series 617:201-205. Fayetteville, Ark.
Norsworthy, J.K., J. Bond, and R.C. Scott. 2013a. Weed management practices and
Norsworthy, J.K., R.C. Scott, and D. B. Johnson. 2013b. A six-year summary of the
herbicide-resistance weed screening program in rice at the University of Arkan-
sas: 2006-2012. In: R.J. Norman and K.A.K. Moldenhaur (eds.). B.R. Wells Rice
Research Series 609:153-158. Fayetteville, Ark.
Wilson, C.E., S.K. Runsick, and R. Mazzanti. 2010. Trends in Arkansas rice produc-
tion. In: R.J. Norman and K.A.K. Moldenhaur (eds.). B.R. Wells Rice Research
Wilson, J.M., J.K. Norsworthy, R.C. Scott, and E.E. Gbur. 2014. Program approaches
to control herbicide-resistant barnyardgrass in mid-southern United States rice.
Weed Technol. 28:39-46.
WSSA Herbicide Handbook, 10th edition, 2014. Weed Science Society of America.
D. Shaner (ed.). p. 15, 449 pp.
204
Table 1. Rice injury at 7, 14, and 28 days after treatment (DAT) and
barnyardgrass, Amazon sprangletop, and hemp sesbania control at 14 and 28 DAT.
Visible injury Barnyardgrass Hemp sesbania Amazon sprangletop
Herbicide
Treatment rate 7 DAT 14 DAT 28 DAT 14 DAT 28 DAT 14 DAT 28 DAT 14 DAT 28 DAT
(oz/acre)
-----------------------------------------------------------------------------(%)----------------------------------------------------------------
Armezon 0.25 0 3 0 63 68 88 60 70 82
Armezon 0.50 4 3 1 79 93 88 96 87 92
Armezon 0.75 7 15 1 83 94 99 98 88 96
Armezon 1.0 9 30 12 88 95 96 99 90 98
Armezon + 0.5
Facet L 32 3 2 0 90 98 100 100 90 100
Armezon + 0.5
RiceStar HT 20 16 36 19 90 99 100 100 95 98
Armezon + 0.5
Permit Plus 0.67 0 2 0 90 94 91 99 88 96
LSD0.05 2 4 5 12 7 7 6 4 4
205
Rice Flatsedge Resistance to

Acetolactate Synthase-Inhibiting Herbicides
P. Tehranchian1, J.K. Norsworthy1, R.C. Scott2, and L.T. Barber2
ABSTRACT
Rice flatsedge is a persistent weed in rice crops in the mid-southern United States.
In 2010, halosulfuron (Permit) was reported to be non-effective on a rice flatsedge
biotype in an eastern Arkansas rice field. The biotype was later confirmed resistant
to the labeled field rate of halosulfuron (0.047 lb ai/acre) in the greenhouse herbicide
screening program. The resistant biotype was not controlled at the highest halosulfuron
dose (3.0 lb ai/acre) tested in a dose-response study. Based on the resistance index (R/S)
ratio calculated from the lethal dose required for 50% plant mortality (LD50), the resis-
tant biotype was > 483-fold less responsive to halosulfuron compared to a susceptible
biotype. The resistant biotype was also cross resistant to other acetolactate synthase
(ALS)-inhibiting herbicides [bispyribac-sodium (Regiment), imazamox (Beyond),
imazethapyr (Newpath), and penoxulam (Grasp)] from four different chemical fami-
lies. Control of the resistant and susceptible biotypes was 93% using 2,4-D (Weedar
64), bentazon (Basagran), and propanil (SuperWham) at the labeled field rate. Based
on this research, a high level of resistance to halosulfuron has evolved in a biotype of
rice flatsedge in Arkansas; fortunately, alternative herbicides are available that provide
effective control.
INTRODUCTION
Rice flatsedge (Cyperus iria) is considered one of the predominant annual monocot
weeds in dry-seeded rice production systems in Arkansas (Norsworthy et al., 2013).
Over the past 10 years in the mid-southern U.S., ALS-inhibiting herbicides have been
the most popular class of herbicides used for post-emergence (POST) weed control
including sedge species in imidazolinone (IMI)-resistant rice production. Effective
control of herbicide-resistant (e.g. propanil, quinclorac, and clomazone) barnyardgrass
(Echinochloa crus-galli) biotypes using ALS-inhibiting herbicides has been achieved
along with control of rice flatsedge in IMI-resistant rice (Levy et al., 2006; Webster et
1
Post-doctoral Research Associate and Professor, respectively, Department of Crop, Soil, and Environ-
mental Science, Fayetteville.
2
Professor and Professor, respectively, Department of Crop, Soil, and Environmental Science, Lonoke.
206
al., 2012). However, overuse of the ALS-inhibiting herbicides for weed control and
lack of integrated weed management strategies has caused selection for resistance in
several weed species common in rice (Heap, 2015).
In 2010, halosulfuron at the labeled field rate failed to control a rice flatsedge
biotype in a rice field in eastern Arkansas. Resistance of the rice flatsedge biotype to
the labeled field rate of halosulfuron was confirmed in preliminary greenhouse studies.
The objectives of this research were to: 1) determine the magnitude of halosulfuron
resistance in a rice flatsedge biotype in Arkansas, 2) characterize cross-resistance to
other ALS-inhibiting herbicides from four different chemical families, and 3) evaluate
alternative herbicides available in Arkansas rice/soybean systems for control of the
resistant biotype.
PROCEDURES
Greenhouse experiments were conducted at the University of Arkansas System
Division of Agricultures Altheimer Laboratory located in Fayetteville, Ark. Seeds of the
resistant biotype (Res) were collected from the infested field and seeds of a halosulfuron-
susceptible biotype (Sus) were provided from a seed repository in Davis, Calif. Seed of
all biotypes were planted individually in plastic flats and seedlings were transplanted at
the 2-lf stage into plastic pots (4 in. diameter 6 in. height). Plants were maintained in
the greenhouse under 86 F/68 F 3 F day/night temperature and a 14-h photoperiod
provided by high pressure sodium lights. All herbicide treatments were applied to 3- to
4-lf rice flatsedge plants using a research spray chamber equipped with a boom mounted
with 800067 flat-fan tips (Teejet Technologies, Springfield, Ill.) calibrated to deliver
20 gal/acre of herbicide solution at 40 PSI.
The dose-response experiment was conducted in a randomized complete block
design (RCBD). There were 20 seedlings of each biotype treated with 6 doses of halo-
sulfuron. The herbicide doses for the Sus biotype were 0.0625, 0.125, 0.25, 0.5, 1, and
2 the labeled field rate of halosulfuron, whereas the Res biotype was treated with 1, 2,
4, 8, 16, and 32 the labeled field rate. All herbicide solutions contained 1% v/v crop
oil concentrate, 0.25% v/v nonionic surfactant, or 0.75% v/v Dyne-A-Pak. Mortality
data were recorded at 28 days after treatment (DAT), and data were subjected to probit
analysis using PROC PROBIT in SAS V. 9.3 (SAS Institute, Inc., Cary, N.C.).
The herbicide evaluation studies were conducted in a factorial RCBD with five
ALS-inhibiting herbicides from four different chemical families and five non-ALS-
inhibiting herbicides currently labeled in Arkansas rice/soybean rotations (Table 1).
Each treatment was replicated four times, and the experiment was repeated. Herbicide
treatments were conducted as mentioned in the dose-response study. Adjuvants were
added as shown in Table 1. At 28 DAT, visual assessments of plant response to each
herbicide treatment were taken on a scale of 0 (no injury) to 100 (plant death). Data
were subjected to analysis of variation (ANOVA) using PROC MIXED in SAS (SAS
Institute, Inc., Cary, N.C.). Means were separated using Fishers protected least sig-
nificant difference test at P = 0.05.
207

Greenhouse Dose-Response Experiment
Even the highest dose of halosulfuron (3.0 lb ai/acre) did not kill 50% of the Res
biotype plants. Based on LD50 values, the Res biotype was > 438-fold less sensitive to
halosulfuron in comparison to the Sus biotype. The occurrence of such a high level of
resistance usually corresponds to an altered target site. A high level of resistance to this
chemistry has been confirmed in other sedge species in Arkansas rice such as yellow
nutsedge (Cyperus esculentus) and smallflower umbrellasedge (Cyperus difformis;
Tehranchian et al., 2014; 2015).
Cross and Multiple Resistance

Control of the Res biotype was 6% with the labeled rate of imazethapyr, ima-
zamox, halosulfuron, bispyribac-sodium, and penoxulam (Fig. 1). This confirmed that
the Res biotype was cross-resistant to ALS-inhibiting herbicides from four chemical
families (imidazolinone, pyrimidinyl benzoate, sulfonylurea, and triazolopyrimidine).
On the contrary, all ALS-inhibiting herbicides tested in this experiment effectively
(98%) controlled the Sus biotype. Cross-resistance to ALS-inhibiting herbicides
has been reported in other sedge species. For example, smallflower umbrellasedge in
California has also evolved cross-resistance to five ALS-inhibiting herbicide families
(Merotto et al., 2009). Cross-resistance to herbicides in crop weeds can be due to al-
tered target protein structure or enhanced metabolic capacity to detoxify the herbicides
(Preston and Mallory-Smith, 2001; Yu et al., 2013; Yu and Powles, 2014). A labeled field
rate of 2,4-D, bentazon, and propanil effectively controlled (93%) both rice flatsedge
biotypes (Fig. 2). In contrast, thiobencarb and quinclorac resulted in 52% control of
the Res and Sus biotypes.
A rice flasedge biotype has evolved cross-resistance to ALS-inhibiting herbicides
from four different chemical families in Arkansas rice. Application of non-ALS-inhib-
iting herbicides currently labeled in Arkansas rice can be used to effectively control
the resistant rice flatsedge biotype. Even with propanil-resistant barnyardgrass being
common in Arkansas rice, prudent use of propanil can decrease selection pressure for
the evolution of ALS-inhibiting herbicide resistance in rice flatsedge. Integrated weed
management strategies must be employed if the use of herbicides for controlling this
weed is to be sustained. Further research needs to be conducted to fully understand the
resistance mechanism(s) within this biotype.
ACKNOWLEDGMENTS
The authors are thankful to the Arkansas Rice Research and Promotion Board
and the University of Arkansas System Division of Agriculture for funding and sup-
porting this research.
208
LITERATURE CITED
Heap I. 2015. The international survey of herbicide resistant weeds. Access date: 25
April 2015. Available from: http://www.weedscience.org
Levy R.J. Jr., J.A. Bond, E.P. Webster, J.L. Griffin, and S.D. Linscombe. 2006. Effect
of cultural practices on weed control and crop response in imidazolinone-tolerant
rice. Weed Technol. 20:249-254.
Merotto Jr. A., M. Jasieniuk, M.D. Osuna, F. Vidotto, A. Ferrero, and A.J. Fischer.
2009. Cross-resistance to herbicides of five ALS-inhibiting groups and sequencing
of the ALS gene in Cyperus difformis L. J. Agric. Food Chem. 57:1389-1398.
Norsworthy, J.K., J.A. Bond, and R.C. Scott. 2013. Weed management practices and
Preston, C. and C.A. Mallory-Smith. 2001. Biochemical mechanisms, inheri-
tance, and molecular genetics of herbicide resistance in weeds: herbicide resis-
tance and world grains. S.B. Powles and D.L. Shaner (eds.). CRC Press. doi:
10.1201/9781420039085.ch2
Tehranchian P., J.K. Norsworthy, V. Nandula, S. McElroy, S. Chen, and R.C. Scott.
2014. First report of resistance to aceolactate synthase-inhibiting herbicides in
yellow nutsedge (Cyperus esculentus): confirmation and characterization. Pest
Manag. Sci. 71:1274-1280.
Tehranchian P., D.S. Riar, J.K. Norsworthy, V. Nandula, S. McElroy, S. Chen, and
R.C. Scott. 2015. ALS-resistant smallflower umbrella sedge (Cyperus difformis)
in Arkansas rice: physiological and molecular basis of resistance. Weed Sci.
63:561-568.
Webster, E.P., T.P. Carlson, M.E. Salassi, J.B. Hensley, and D.C. Blouin. 2012. Ima-
zethapyr plus residual herbicide programs for imidazolinone-resistant rice. Weed
Technol. 26:410-416.
Yu, Q., H. Han, G.R. Cawthray, S.F. Wang, and S.B. Powles. 2013. Enhanced rates
of herbicide metabolism in low herbicide-dose selected resistant Lolium rigidum.
Plant Cell Environ. 36:818-827.
Yu, Q. and S.B. Powles, 2014. Metabolism-based herbicide resistance and cross-re-
sistance in crop weeds: a threat to herbicide sustainability and global crop produc-
tion. Plant Physiol 166:1106-1118.
209
Table 1. Acetolactate synthase (ALS)-inhibiting and alternative

herbicides available in Arkansas rice/soybean rotations used in this study.
Common name Trade name Labeled field rate Adjuvant
(lb ai/acre) (% v/v)
ALS-inhibiting herbicides
Imazamox Beyond 0.047 1% COCa
Imazethapyr Newpath 0.095 0.25% NISb
Bispyribac-sodium Regiment 0.032 0.75% Dynec
Penoxsulam Grasp

0.031 1% COC
Halosulfuron Permit 0.047 1% COC

Alternative rice herbicides
2,4-D Weedar 64 1.0 0.25% NIS
Bentazon Basagran 1.0 ---
Propanil SuperWham 4.0 ---
Quinclorac Facet L 0.5 ---
Thiobencarb Bolero 4.0 ---
a
COC = crop oil concentrate.
b
NIS = nonionic surfactant.
c
Dyne-A-Pak.
Fig. 1. Effectiveness of acetolactate synthase ALS-inhibiting

herbicides [bispyribac-sodium (Regiment), imazethapyr (Newpath), imazamox
(Beyond), halosulfuron (Permit), and penoxulam (Grasp) for controlling rice flatsedge
biotypes (Res and Sus) at 28 days after treatment. Sus = susceptible; Res = Resistant.
210
Fig. 2. Effectiveness of non-acetolactate synthase (ALS)-inhibiting herbicides

2,4-D (Weedar 64), bentazon (Basagran), propanil (SuperWham), quinclorac (Facet L),
and thiobencarb (Bolero)] labeled in Arkansas rice for controlling rice flatsedge
biotypes (Res and Sus) at 28 days after treatment. Sus = susceptible; Res = Resistant.
211
Evaluation of a Benzobicyclon Plus

Halosulfuron Premix for Weed Control in Arkansas Rice
M.L. Young1, J.K. Norsworthy1, C.A. Sandoski2, R.C. Scott3, and L.T. Barber3
ABSTRACT
Rogue , a new rice herbicide, is being developed by Gowan Company for postflood con-

trol of problematic weeds. Rogue Plus, a premix, will contain a mixture of halosulfuron
(Group 2) and benzobicyclon (Group 27) herbicides and will control a broad-spectrum of
grasses, aquatics, broadleaves, and sedges, including those currently resistant to Group
2 herbicides. If labeled as expected, this will be the first 4-hydroxyphenylpyruvate
dioxygenase (HPPD) herbicide commercially available in U.S. rice production. This
new mode of action in rice will enable producers to control a variety of weed species
that become increasingly more problematic as the season progresses. Field studies were
conducted in 2014 and 2015 at the University of Arkansas System Division of Agri-
cultures Rice Research and Extension Center near Stuttgart, Ark., to understand if the
addition of halosulfuron (Permit) to benzobicyclon would increase the level of weed
control compared with benzobicyclon alone for barnyardgrass, Amazon sprangletop,
ducksalad, California arrowhead, hemp sesbania, northern jointvetch, yellow nutsedge,
and smallflower umbrellasedge. Treatments in 2014 included: benzobicyclon at 0.22
and 0.33 lb ai/acre, and a mixture of both rates of benzobicyclon plus halosulfuron at
0.03 and 0.05 lb ai/acre, and a control. In 2015, there were two additional treatments
added to the treatment structure of halosulfuron at 0.03 and 0.05 lb ai/acre applied
alone for a total of seven experimental treatments. Benzobicyclon alone was effective
in controlling Amazon sprangletop, ducksalad, California arrowhead, and smallflower
umbrellasedge. The addition of halosulfuron to benzobicyclon generally improved
control of those weeds that were marginally controlled by benzobicyclon alone. The
low rate combination of benzobicyclon plus halosulfuron was often as effective as the
high rate of benzobicyclon alone. The results of this study suggest that benzobicyclon
premixed with halosulfuron has potential for control of problematic weeds in Arkansas
rice and could be used as an additional weed management tool for control of herbicide-
resistant weeds.
1
Graduate Assistant and Professor, respectively, Department of Crop, Soil, and Environmental Science,
Fayetteville.
2
Southern Region Development Representative, Gowan, USA, Collierville, Tenn.
3
Professor and Professor, respectively, Department of Crop, Soil, and Environmental Science, Lonoke.
212
INTRODUCTION
Due to the repetitive use of the same herbicide modes of action in rice, many
weeds have evolved herbicide resistance across the mid-South (Carey et al., 1995;
Norsworthy et al., 2013a,b; Wilson et al., 2014). A new mode of action is needed in
rice production with the increasing stress on our current herbicide chemistries. Gowan
Company is developing benzobicyclon, a new herbicide for post-flood applications in
rice. Benzobicyclon, a Group 27 herbicide, will represent a new option for rice produc-
ers that has activity against resistant weeds including barnyardgrass, ducksalad, and
sedges (Heap, 2015). In particular, acetolactate synthase (ALS)-resistant barnyardgrass
and sedges pose a significant threat to Arkansas rice production (Riar et al., 2012;
Bagavathiannan et al., 2014).
Gowan Company also produces halosulfuron or Permit 75DF, which is a Group
2 herbicide labeled in rice and several other crops for the control of annual broadleaf
weeds and sedges. McCallister et al. (2009) documented that halosulfuron is most effec-
tive when combined with other herbicides especially for the control of broadleaf weeds.
The increasing amount of herbicide-resistant species makes it important to mix multiple
modes of action; therefore, the combination of halosulfuron plus benzobicyclon may
provide a broader spectrum of weed control over the herbicides applied individually.
The objective of this research was to evaluate a premix of benzobicyclon and
halosulfuron for its potential as a rice herbicide in mid-South rice production.
PROCEDURES
A study was initiated in the summer of 2014 and repeated in 2015 at the University
of Arkansas System Division of Agricultures Rice Research and Extension Center near
Stuttgart, Ark., to determine if the addition of halosulfuron (Permit) to benzobicyclon
would increase the level or spectrum of weed control compared to benzobicyclon alone.
Clearfield 111 rice was drill-seeded into a silt loam soil. The design was a randomized
complete block with three replications. Bays were 150 ft 10 ft wide and the plots
planted into the bays were 75 ft 6 ft with levees separating bays. Weeds across the
entire test site were initially suppressed with a 0.5 labeled rate of clomazone at planting
followed by a 0.5 labeled rate of imazethapyr at 2 weeks after rice emergence. Treat-
ments in 2014 included: benzobicyclon at 0.22 and 0.33 lb ai/acre, a mixture of both
rates of benzobicyclon plus halosulfuron at 0.03 and 0.05 lb ai/acre, and a control. In
2015, there were two additional herbicide treatments added to the treatment structure,
which included halosulfuron at 0.03 and 0.05 lb ai/acre applied alone for a total of
seven experimental treatments. All treatments were applied 1 week after the permanent
flood was established. All applications were made using a CO2-pressurized backpack
sprayer calibrated to deliver 15 gal/acre. Visual weed control ratings were taken every 2
weeks and were estimated using a scale of 0% to 100%, where 0 is no control and 100
is complete control. Rough-rice yield data were collected upon maturity of the crop.
All data were subjected to analysis of variance using JMP Pro 12.1.0 (SAS Institute
213
Inc. Cary, N.C.), and means were separated using Fishers protected least significant
difference test (P = 0.05).

In 2014, as late as 7 weeks after treatment (WAT), significant differences were
detected between treatments for the control of northern jointvetch. Where benzobicyclon
was applied alone, both treatments exhibited significantly lower levels of control com-
pared to when benzobicyclon was combined with halosulfuron (Fig. 1). By 8 WAT, both
rates of benzobicyclon applied alone or in mixture with halosulfuron provided complete
control of rice flatsedge, yellow nutsedge, and Amazon sprangletop (data not shown).
In 2015, similar results to those in 2014 were seen with the addition of halo-
sulfuron to benzobicyclon. The addition of halosulfuron to benzobicyclon did not
improve barnyardgrass control over benzobicyclon alone (Fig. 2). Benzobicyclon plus
halosulfuron at the high rate provided 81% control of barnyardgrass. It does not appear
that the herbicide combination of halosulfuron plus benzobicyclon will provide a level
of control deemed effective by most growers. By 9 WAT, all treatments containing
benzobicyclon provided complete control of ducksalad, California arrowhead, and
ALS-resistant smallflower umbrellasedge (Fig. 3).
Rough rice yields were collected both years; however, no significant differences
were detected among treatments (data not shown).
Rogue Plus, the likely tradename of the premix of benzobicyclon plus halosul-
furon, will provide Arkansas rice growers suppression of barnyardgrass and effective
control of Amazon sprangletop, northern jointvetch, hemp sesbania, ducksalad, Cali-
fornia arrowhead, yellow nutsedge (resistant and susceptible biotypes), rice flatsedge
(resistant and susceptible biotypes), and smallflower umbrellasedge (resistant and
susceptible biotypes). Additionally, the benzobicyclon plus halosulfuron combination
brings an additional herbicide mode of action to Arkansas rice for broad-spectrum,
postflood management of many of the most common and problematic weeds. It will
also aid in the fight against resistant weeds, especially sedges like yellow nutsedge, rice
flatsedge, and smallflower umbrellasedge.
ACKNOWLEDGMENTS
The authors thank the Arkansas Rice Research and Promotion Board for their
continued support, the University of Arkansas System Division of Agriculture, and
Gowan Company for their support of this work.
214
LITERATURE CITED
Technol. 9:366-372.
Bagavathiannan, M.V., J.K. Norsworthy, D. Riar, Z.T. Hill, B.W. Schrage, C.J.
Meyer, and H.D. Bell. 2014. Growth characterization of an ALS-inhibitor resis-
tant yellow nutsedge population from eastern Arkansas. Proc. South. Weed Sci.
Soc. 67:30.
Heap, I. 2015. International Survey of Herbicide Resistant Weeds. Access date: 10
January 2015. Available at: www.weedscience.org
McCallister, E.K., J.K. Norsworthy, J.D. Devore, M.J. Wilson, S.K. Bangarwa, G.M.
Griffith, R.C. Scott, and K.L. Smith. 2009. Herbicide combinations with halosul-
furon for hemp sesbania control in rice. In: R.J. Norman and K.A.K. Moldenhaur
(eds.). B.R. Wells Rice Research Studies 2009. University of Arkansas Agricul-
tural Experiment Station Research Series 581:123-132. Fayetteville, Ark.
Norsworthy, J.K., J. Bond, and R.C. Scott. 2013a. Weed management practices and
Norsworthy, J.K., R.C. Scott, and D. B. Johnson. 2013b. A six-year summary of the
herbicide-resistance weed screening program in rice at the University of Arkan-
sas: 2006-2012. In: R.J. Norman and K.A.K. Moldenhaur (eds.). B.R. Wells Rice
Research Series 609:153-158. Fayetteville, Ark.
Riar, D.S., J.K. Norsworthy, M.T. Bararpour, M.J. Wilson, R.C. Scott, and V. Sriv-
astava. 2012. Acetolactate synthase-inhibiting herbicide resistance in two barn-
yardgrass (Echinochloa crus-galli) biotypes of Arkansas. In: R.J. Norman, J.-F.
Meullenet, and K.A.K. Moldenhauer (eds.). B.R. Wells Rice Research Studies
2012. University of Arkansas Agricultural Experiment Station Research Series
600:181-186. Fayetteville.
Wilson, J.M., J.K. Norsworthy, R.C. Scott, and E.E. Gbur. 2014. Program approaches
to control herbicide-resistant barnyardgrass in mid-southern United States rice.
Weed Technol. 28:39-46.
215
Fig. 1. Control of northern jointvetch 7 weeks after

treatment in 2014. Means having the same letter are not significantly
Fig. 2. Barnyardgrass control 9 weeks after treatment

in 2015. Means having the same letter are not significantly different
according to Fishers protected least significant difference test at P = 0.05.
216
Fig. 3. California arrowhead, ducksalad, and acetolactate

synthase (ALS)-resistant smallflower umbrellasedge control 9 weeks
after treatment in 2015. Means having the same letter are not significantly
217
RICE CULTURE
2015 Degree Day 50 Thermal Unit Thresholds

for New Rice Cultivars and Seeding Date Studies
E. Castaneda-Gonzalez1, J.T. Hardke2, D.L. Frizzell1,
G.J. Lee2, T.L. Clayton3, R.J. Norman4, K.A.K. Moldenhauer1, and X. Sha1
ABSTRACT
The Degree-Day 50 (DD50) computer program has been one of the most successful man-
agement aids developed by the University of Arkansas System Division of Agriculture.
The program predicts critical growth stages that assist in optimizing crop management
operations. In order to be effective, the computer program must be updated continually
as new rice cultivars become available. To fulfill this goal, studies are conducted in
a controlled research environment where developmental data and DD50 thermal unit
thresholds for current and new cultivars are determined. Throughout the 2015 season,
plant developmental data, DD50 thermal unit accumulation, and grain and milling yield
performance data for nineteen cultivars were evaluated over five seeding dates under
the dry-seeded, delayed-flood management system that is commonly used in southern
U.S. rice production.
INTRODUCTION
Developed in the 1970s to help farmers time midseason nitrogen (N) applica-
tions with precision, the Degree-Day 50 (DD50) computer program currently provides
predicted dates for timing twenty-six key management decisions including fertilization,
pesticide applications, permanent flood establishment, times for scouting insect and
disease, predicted draining date, and suggested harvest time.
The DD50 is a modification of the degree-day growing concept with daily high and
low air temperatures used to quantify a days thermal accumulations for plant growth.
The DD50 Program generates a predicted rice plant development file that is cultivar-
specific based on the accumulation of DD50 units beginning at emergence. The file is
1
Program Technician I, Program Associate III, Professor, and Associate Professor, respectively, Rice
Research and Extension Center, Stuttgart.
2
Rice Extension Agronomist and Program Technician - Rice Agronomy, respectively, Department of
Crop, Soil, and Environmental Science, Stuttgart.
3
4
218
created by calculating thermal unit accumulation using 30-year average weather data
collected by the National Weather Service weather stations closest to a rice producers
location in Arkansas. As the season progresses the program is updated with the current
years weather data.
The data used to predict plant development for a specific cultivar is obtained in
yearly studies where promising experimental lines and newly released conventional
and hybrid rice cultivars are evaluated in four to six seeding dates per season within
the recommended range of rice seeding dates for Arkansas. Once a new cultivar is re-
leased, the information obtained in these studies is utilized to provide threshold DD50
thermal units to the DD50 computer program that enables the prediction of dates of
plant developmental stages and dates when particular management practices could be
performed. Therefore, the objectives of this study were to develop a DD50 thermal unit
accumulation database for promising new cultivars, to verify and refine the existing
database of current cultivars, and to assess the effect of seeding date on DD50 thermal
unit accumulation. Also, determining the effects of seeding date on grain and milling
yields helps determine optimal seeding dates for a particular cultivar.
PROCEDURES
The 2015 DD50 study was conducted at the University of Arkansas System
Division of Agricultures Rice Research and Extension Center (RREC) near Stuttgart,
Ark., on a DeWitt silt loam soil. Fifteen pure-line cultivars (i.e., CL111, CL151, CL153,
CL163, CL172, CL271, CL272, Diamond, Jupiter, LaKast, Mermentau, Roy J, Titan,
Wells, and the experimental line MSX4077) were dry-seeded at a rate of 30 seed/ft2
in plots 9 rows wide (7-inch spacing) and 15 ft long. Four hybrids (i.e., CLXL729,
CLXL745, XL753, and XL760) were seeded into plots of the same dimensions using
a reduced seeding rate for hybrids of 13 seed/ft2. The seeding dates for 2015 were 3
April, 21 April, 5 May, 19 May, 3 June, and 16 June. However, the 16 June seeding date
was lost due to bird depredation. General agronomic information is shown in Table 1.
Established cultural practices for dry-seeded, delayed-flood rice production were fol-
lowed. A single preflood application of 120 lb N/acre as urea was applied to all plots
at the 4- to 5-lf growth stage and flooded within 2 days of preflood N fertilization. The
flood was maintained until maturity. The collected data for each of the seeding dates
included: maximum and minimum daily temperatures, date of seedling emergence,
and the number of days and DD50 units required to reach 50% heading. The number
of days and DD50 thermal units required to reach 0.5-inch internode elongation (IE)
was collected for the 3 April, 5 May, and 3 June seeding dates for selected cultivars. At
maturity, the five center rows in each plot were harvested, weight of grain and moisture
content were recorded, and a subsample of grain was taken for milling purposes. The
grain yield was adjusted to 12% moisture and reported in bushels/acre (bu/acre). The
dry rice was milled to obtain percent head rice (HR; whole kernels) and percent total
white rice (TR, whole + broken kernels) to provide the ratio of %HR - %TR. The ar-
rangement of each seeding date was a randomized complete block design with four
219
replications. Statistical analyses were conducted using PROC GLM (SAS version 9.4,
SAS Institute, Inc., Cary, N.C.) and mean separation was conducted using Fishers
protected least significant difference test (P = 0.05) where appropriate.

Time between seeding and emergence ranged from 6 to 15 days (Table 1). In
general, seeding date (SD) studies report a decrease in days between seeding and
emergence as the seeding date is delayed. The 2015 study followed this general trend
of decreasing days from seeding to emergence as SD was delayed from early April to
late June, except for the second SD (21 April) which increased by four days compared
to the 3 April SD. The time from seeding to establishment of permanent flood followed
a similar decreasing pattern as the SD was delayed; ranging from 47 days for the 3
April to 23 days for the 3 June seeding dates and slightly increased to 24 days for the
mid-June SD. The times from emergence to flooding in 2015 had a general decreasing
trend as SD was delayed also, with 47 days for the 3 April SD and then a 3 day decrease
for each subsequent SD until the 19 May SD, a decrease of 8 days between the 19 May
and 3 June seeding dates and an increase of 1 day as SD was delayed until 16 June.
The days required from emergence to 0.5-inch IE for the selected cultivars in the
three seeding dates sampled, averaged 53 days (Table 2). A decreasing trend in time
was observed to reach 0.5-inch IE as SD was delayed. Across cultivars, the average
number of days to reach 0.5-inch IE ranged from 67 days when seeded in early April
to 43 days when seeded in early June. Time required for vegetative growth averaged
across seeding rates ranged between 48 days for CL153 to 57 days for Jupiter. The
DD50 thermal unit accumulation for vegetative growth averaged across SD ranged
from a low of 1215 for CL153 to a high of 1476 for Jupiter.
During 2015, the time needed from emergence to reach the developmental stage
known as 50% heading averaged across SD and cultivars was 78 days (Table 3). The
average time for cultivars to reach 50% heading ranged from 89 days when seeded
in early April to 71 days when seeded in early June. For individual cultivars the time
required to reach 50% heading ranged from 66 days for CLXL729 and CLXL745 when
seeded in early June to 95 days for Jupiter when seeded in early April. The thermal unit
accumulation from emergence to 50% heading averaged 2131 across SD and cultivars.
The individual cultivar DD50 thermal unit accumulation from emergence to 50% head-
ing ranged from a low of 1968 for CLXL745 seeded 3 April to a high of 2369 for Roy J
seeded 19 May. The 19 May SD generally required the greatest number of DD50 units
from emergence to 50% heading.
Average grain yield for the 2015 study across SD and cultivars was 174 bu/acre
(Table 4). When averaged across cultivars, the 3 April SD had the highest grain yield
and the 3 June SD had the lowest grain yield. Mean grain yields were numerically higher
for the 3 April SD followed by similar yields for 5 May and 19 May. An uncharacteris-
tic drop-off in grain yield was measured when the cultivars were seeded 21 April with
similar reports of poor yields in production fields seeded during the same time. The
220
most consistent cultivars across SD were the hybrids XL760, XL753, and CLXL729
with grain yield averages of 221, 220, and 215 bu/acre, respectively. Pure-line cultivars
performing well across SD were Diamond, Titan, and Roy J with grain yield averages
of 192, 175, and 170 bu/acre, respectively.
During 2015, the milling yield (head rice, HD / total white rice, TR) averaged
across SD and all cultivars was 63% head rice and 70% total white rice (Table 5). There
was a noticeable trend for increased head rice and total white rice as SD was delayed.
With a very few exceptions, all cultivars averaged 60% head rice yield or better during
this study year regardless of SD.
The data obtained during 2015 will be used to refine the DD50 thermal unit
threshold for new pure-line cultivars and hybrids being grown. The grain and milling
yield data will contribute to the database of information used by University personnel
to help producers make decisions in regard to rice cultivar selection, in particular for
early- and late-seeding situations.
ACKNOWLEDGMENTS
This research was funded and supported by the Arkansas Rice Research and
Promotion Board and the University of Arkansas System Division of Agriculture.
Special thanks are also given to Emmett C. Chuck Pipkins for his dedication to the
DD50 Program.
Table 1. General seeding, seedling emergence, and flooding date information for the DD50
seeding date study in 2015 at the Rice Research and Extension Center near Stuttgart, Ark.
Seeding date
3 April 21 April 5 May 19 May 3 June 16 June
Emergence date 14 April 6 May 15 May 27 May 10 June 22 June
Flood date 28 May 4 June 13 June 26 June 8 July 24 July
Days from seeding to emergence 11 15 10 8 7 6
Days from seeding to flooding 47 44 36 31 23 24
Days from emergence to flooding 36 29 26 23 16 18
221
Table 2. Influence of seeding date on DD50 accumulations and days

from emergence to 0.5-in. internode elongation of selected rice cultivars in studies
conducted at the Rice Research and Extension Center, Stuttgart, Ark., during 2015.
Seeding date
3 April 5 May 3 June Average
DD50a DD50 DD50 DD50
Cultivar days units days units days units days units
CL153 62 1322 46 1237 37 1085 48 1215
CL163 67 1478 52 1396 45 1331 55 1402
CL172 68 1509 51 1370 43 1259 54 1379
CL271 68 1508 54 1433 46 1379 56 1440
CL272 67 1488 53 1411 43 1275 54 1391
Jupiter 71 1594 54 1448 47 1387 57 1476
Diamond 67 1472 50 1331 42 1228 53 1343
LaKast 65 1418 49 1313 40 1180 51 1304
Titan 66 1441 51 1350 44 1307 53 1366
Wells 67 1486 50 1344 43 1259 53 1363
XL760 63 1351 47 1267 39 1157 50 1258
MSX4077 68 1516 54 1447 46 1355 56 1439

Mean 67 1465 51 1362 43 1267 53 1365
LSD(0.05)b 1.9 55.6 1.5 39.8 1.8 58.1 NS 76.4
a
DD50 units calculated daily by equation [(daily max temperature + daily min tempera-
ture)/2]-50.
b
LSD = least significant difference.
222
Table 3. Influence of seeding date on DD50 accumulations and days from emergence to 50% heading of
selected rice cultivars in studies conducted at the Rice Research and Extension Center, Stuttgart, Ark., during 2015.
Seeding date
3 April 21 April 5 May 19 May 3 June Average
DD50a DD50 DD50 DD50 DD50 DD50
Cultivar days units days units days units days units days units days units
CL111 86 2032 76 2055 72 1994 72 2098 67 2029 75 2042
CL151 87 2047 77 2086 72 2010 73 2146 67 2036 75 2065
CL153 89 2116 79 2142 75 2098 74 2162 68 2051 77 2114
CL163 88 2095 82 2222 77 2153 76 2217 77 2291 80 2196
CL172 92 2193 82 2222 75 2098 74 2178 69 2081 78 2154
CL271 92 2194 81 2198 79 2215 77 2256 77 2291 81 2231
CL272 92 2220 80 2166 76 2114 76 2225 73 2190 79 2183
CLXL729 86 2026 76 2031 74 2058 72 2098 66 1998 75 2042
CLXL745 84 1968 74 1976 72 2002 70 2026 66 1998 73 1994
Diamond 89 2124 79 2126 75 2082 75 2194 73 2178 78 2141
Jupiter 95 2305 82 2238 78 2193 77 2272 74 2200 81 2241
LaKast 88 2098 76 2039 72 1994 73 2146 68 2058 75 2067
Mermentau 88 2089 79 2150 74 2066 73 2146 68 2051 76 2100
Roy J 92 2210 84 2309 79 2231 81 2369 77 2274 83 2278

Titan 86 2040 75 2023 72 2002 74 2154 68 2058 75 2055
Wells 91 2170 80 2158 75 2082 75 2209 72 2170 78 2158
XL753 85 2004 75 1999 72 1994 69 2002 67 2013 73 2002
XL760 89 2116 81 2198 78 2192 77 2256 71 2150 79 2182
MSX4077 91 2162 84 2301 78 2193 79 2315 77 2291 82 2252

Mean 89 2116 79 2139 75 2093 74 2182 71 2127 78 2131
LSD(0.05)b 1.8 51.1 1.7 53.0 1.1 36.4 2.1 63.2 1.0 26.2 4.1 34.7
a
DD50 units calculated daily by equation [(daily max temperature + daily min temperature)/2]-50.
b
223
Table 4. Influence of seeding date on grain yield of selected rice cultivars in studies
Grain yield by seeding date
Cultivar 3 April 21 April 5 May 19 May 3 June 16 Junea Average
------------------------------------------- (bu/acre)---------------------------------------------
CL111 173 152 171 143 104 . 149
CL151 182 141 178 178 135 139 163
CL153 179 157 170 174 121 133 160
CL163 170 141 172 177 137 . 159
CL172 163 139 167 162 140 136 154
CL271 176 126 174 174 140 136 158
CL272 178 146 183 175 130 . 162
CLXL729 226 212 227 229 181 169 215
CLXL745 180 198 201 194 157 . 186
Diamond 211 164 192 209 185 181 192
Jupiter 183 157 165 161 161 151 165
LaKast 197 174 183 192 150 . 179
Mermentau 178 131 170 183 156 144 163
Roy J 193 142 165 180 169 160 170
Titan 222 178 178 149 148 . 175
Wells 159 139 170 179 146 . 159
XL753 225 227 232 225 191 202 220
XL760 233 212 224 235 204 191 221
MSX4077 189 141 171 167 140 . 162

Mean 221 162 184 184 152 158 174
LSD(0.05)b 22.6 17.5 12.4 159 17.1 NA 13.4
a
The 16 June seeding date not included in average due to bird damage at trial emergence.
Values reported from plots with no bird damage only.
b
224
Table 5. Influence of seeding date on milling yield of selected rice cultivars in studies
Milling yield by seeding date
Cultivar 3 April 21 April 5 May 19 May 3 June Average
------------------------------------------- (%HR - %TRa)---------------------------------------
CL111 62-69 65-71 65-71 65-70 67-72 65-70
CL151 62-69 67-70 66-71 67-71 67-72 66-71
CL153 63-68 66-69 66-70 66-70 66-71 65-70
CL163 57-65 63-69 63-69 64-69 65-69 62-68
CL172 64-69 66-70 65-70 66-70 66-71 65-70
CL271 63-68 63-69 64-70 63-71 69-71 64-70
CL272 62-68 63-70 59-70 60-70 67-70 62-69
CLXL729 56-66 56-68 61-69 65-71 66-71 61-69
CLXL745 57-68 55-69 60-71 63-71 65-72 60-70
Diamond 55-67 60-69 61-71 66-71 65-70 61-70
Jupiter 63-66 66-68 65-69 66-69 66-68 65-68
LaKast 53-67 54-68 58-69 65-71 65-72 59-70
Mermentau 62-68 66-69 66-70 65-69 67-71 65-69
Roy J 56-68 63-70 65-71 65-71 67-72 63-70
Titan 63-67 63-70 58-69 53-68 67-69 61-69
Wells 53-68 60-70 56-70 60-70 66-72 59-70
XL753 55-68 51-70 56-70 61-71 65-72 58-70
XL760 56-67 62-70 62-70 65-71 66-71 62-70
MSX4077 58-67 63-69 60-69 64-69 65-69 62-69

Mean 59-68 62-69 62-70 64-70 66-71 63-70
%HR LSD(0.05)b 2.8 2.4 2.5 2.8 1.1 2.3
% TR LSD(0.05) 1.0 1.1 0.8 1.1 0.9 0.9
a
%HR - %TR = percent head rice - percent total white rice.
b
225
RICE CULTURE
Continued Validation of the Nitrogen

Soil Test for Rice on Clay Soils in Arkansas
J.T. Davidson1, T.L. Roberts1, R.J. Norman1,
C.E. Greub1, N.A. Slaton1, and J.T. Hardke2
ABSTRACT
The development of the Nitrogen Soil Test for Rice (N-STaR) allows a site-specific
nitrogen (N) recommendation for rice on silt loam soils in Arkansas. Expansion of the
site-specific N test into clay soils and its use in 27 counties in Arkansas necessitated the
validation of N-STaR for rice produced on clay soils. Davidson et al. (2014) initiated
the validation of this new soil test, but stated that additional data should be collected
to substantiate the findings. Therefore, seven additional sites across Arkansas were
selected in 2015 for their wide difference in native soil-N and seeded to one of three
rice cultivars. Stands were monitored for disease and pest pressure and yields were
measured at the end of the season. Soil samples were taken at a 12-in. depth, analyzed
using the N-STaR method, and the site-specific N rates were predicted using the cali-
bration curves for 95% and 100% relative grain yield (RGY). In the validation trial, six
treatments were compared: a control (0 lb N/acre); the N-STaR 95% and 100% RGY N
rates applied in a standard two-way split (2-WS) application with 45 lb N/acre applied
at beginning internode elongation and the remainder preflood; the N-STaR 95% and
100% RGY N rates applied in a single preflood (SPF) application; and the standard
N recommendation based on cultivar, soil texture, and previous crop. Nitrogen rates
predicted using N-STaR ranged from 0 lb N/acre to 180 lb N/acre. Rice yields obtained
with the 95% RGY recommendation were statistically similar or greater than the standard
N-rate recommendation for six of the seven sites while reducing the N rate between
50 to 200 lb N/acre compared to the standard N recommendation. Similarly, six of the
seven 100% RGY recommendation trials were equal to or greater than the standard N
recommendation. Overall, N-STaR is able to predict site-specific N fertilizer rates for
rice produced on clay soils over a wide range of environmental and production settings.
INTRODUCTION
Rice is a major crop in eastern Arkansas, particularly around the Arkansas Grand
Prairie and Mississippi Delta regions. In order to achieve optimum rice yields, N fertiliza-
1
Graduate Assistant, Assistant Professor, Professor, Senior Graduate Assistant, and Professor, respec-
tively, Department of Crop, Soil, and Environmental Science, Fayetteville.
2
226
tion is required at 150 lb N/acre for most cultivars and is adjusted based on soil texture,
rice cultivar, and previous crop. While rice has one of the highest N use efficiencies when
managed correctly (Norman et al., 1992), fields with high levels of native-N may not
respond or have little response to N fertilization which leads to poor N use efficiency.
The excess N fertilizer at these locations increases the likelihood of N losses into the
environment, and is an unnecessary expense for rice producers. Since the development
of the Illinois Soil Nitrogen Test (ISNT) by Mulvaney and Khan in 2001, interest has
reawakened for a N soil test that measures potentially mineralizable-N. The development
of N-STaR by Roberts et al. in 2011 enabled a site-specific N recommendation for rice
producers in Arkansas rather than the standard approach based on cultivar, soil texture,
and previous crop. Site-specific N recommendations take into account the native-N
of a soil and identify the sites where a decreased or increased rate of N, compared to
the standard approach, is needed for optimal rice production. Soils separated by tex-
tural classclays vs. silt loams and sandsyielded the strongest correlation between
alkaline hydrolyzable-N and N rate required for optimum yield. In addition, Fulford
et al. (2012) found the greatest predictability for clay soils to be a soil sample at the
0- to 12-in. depth. During the development of N-STaR for silt loam soils, the N-STaR
calibration curves for the 95% and 100% relative grain yield (RGY) were validated to
ensure predictability and to aid in the implementation of the N soil test (Roberts et al.,
2013). The validation of N-STaR 95% and 100% calibration curves on clay soils in
Arkansas was initiated by Davidson et al. in 2014 and this study is a continuation of that
research. Davidson et al. (2014) found that the N-STaR recommendations accurately
predicted the N needs of rice with the exception of the N-STaR 95% single preflood
(SPF) N recommendation which may need to be adjusted. The purpose of this research
is to increase the number of site-years in order to further evaluate the N-STaR N-rate
recommendations under a two-way split (2-WS) and SPF application. The N-STaR
method has the potential to decrease N loss into the environment and reduce the cost
of fertilizer inputs for farmers (Williamson et al., 2013). The validation of N-STaR for
clay soil will speed the adoption of this site-specific N soil test throughout Arkansas.
METHODS AND MATERIALS

In 2015, seven field experiments were conducted across producers fields (4 sites)
and at the University of Arkansas System Division of Agriculture experiment stations
(3 sites) in Arkansas. Clay soil locations were chosen to ensure a wide range of native
soil-N availability across sites. The plots were 9 rows wide (7-in. spacing) and 15 ft in
length, and arranged in a randomized complete block design. Rice was dry-seeded (100
lb seed/acre on station) and grown to the 3- to 5-lf stage before a permanent flood was
established (2- to 4-in. depth) and maintained until physiological maturity. Plots were
monitored for pest pressure throughout the season. Four soil samples were taken at each
location from the 0- to 12-in. depth, analyzed using N-STaR as outlined by Roberts
et al. (2009), and the average N-STaR soil test value was used to produce the 95% or
100% RGY N-rate recommendations for each location. Six treatments were conducted
227
at each location: a control (0 lb N/acre); N-STaR 95% and 100% RGY applications;
N-STaR 95% and 100% RGY SPF applications; and the standard N recommendation.
For the standard N recommendation and the N-STaR 95% and 100% RGY applications,
a 2-WS of 45 lb N/acre was applied at panicle initiation and the remainder of the N
recommendation was applied preflood at <5 days before permanent flood. The N-STaR
SPF N recommendations were all applied preflood at <5 days before permanent flood.
The N-STaR SPF N recommendations were 20 lb N/acre less than the N-STaR 2-WS
N recommendation calculated by the 95% and 100% RGY calibration curves. The N
fertilizer applied was urea treated with the urease inhibitor N-(n-butyl) thiophosphoric
triamide (NBPT); trade name Agrotain Ultra, (Koch Fertilizer LLC., Wichita, Kan.).
Grain was harvested from the middle four rows of each plot and weights were adjusted
to 12% grain moisture and expressed in bushels (bu)/acre. Rough rice grain yield was
compared across treatments within a location using JMP Pro 11.0 (SAS Institute, Inc.,
Cary, N.C.) using Fishers protected least significant difference test at the P = 0.05 level.

The standard N recommendation currently used by Arkansas rice producers is
altered according to cultivar, soil texture, and previous crop. The introduction of N-
STaR allowed producers to test for an index of potentially mineralizable N that was
calibrated to a site-specific N recommendation. Two N-STaR calibration curves, the
95% RGY and the 100% RGY, were developed for use in clay soil that would provide
site-specific N-rate recommendations needed to obtain the respective percentage (95%
or 100%) of yield on a field by field basis. Historically, the 95% and 100% RGY N
fertilizer recommendations have not been significantly different with regard to yield,
although, the 100% RGY is often numerically higher (Roberts et al., 2013; Davidson
et al., 2014). The substantially larger N-rate recommendation predicted by the 100%
RGY reflects the high cost to the producer in achieving the last 5% of rice yield.
In order to reduce variability across locations, the RGY of the five N fertilizer
recommendations was compared at each of the seven sites as shown in Table 1. The
2-WS N-STaR 100% RGY was statistically similar to the standard N recommendation
for five of the seven sites. At Site ES-3, the 2-WS N-STaR 100% RGY was significantly
lower than the standard N recommendation, however, the yield decrease was relatively
small at 15 bu/acre. At this location, the SPF N rate resulted in a greater yield compared
to the 2-WS N rate within the 95% and 100% RGY N-rate recommendations. This
likely resulted from adverse environmental conditions causing preflood N losses since
previous studies conducted at this location by Davidson et al. (2014) showed the 2-WS
to perform as well or better than the SPF. The 2-WS N-STaR 100% RGY treatment
yielded significantly higher than the standard N recommendation at Site P-7. Site P-7
was unique in that it had a N-STaR recommendation of 0 lb N/acre for both the 95%
and 100% RGY calibration curves and, therefore, will be discussed separately in detail.
With the 2-WS N-STaR 95% RGY, N rates ranged from 0 to 150 lb N/acre. Site
P-5 did not receive a 2-WS N-STaR 95% RGY treatment since the total fertilizer-N
228
rate was small and not economical to split apply in a production setting. Five of the six
sites receiving a 2-WS N-STaR 95% RGY yielded equal to (P-6, P-8, ES-4, and ES-5)
or significantly greater than (P-7) the standard N recommendation. The 2-WS N-STaR
95% RGY at Site ES-3 yielded significantly lower than the standard N recommendation
at a magnitude of 38 bu/acre.
When the 2-WS N-STaR 95% and 100% RGY treatments were compared, only
ES-3 showed a statistical difference with the 100% RGY yielding higher than the 95%
RGY. Overall, the 2-WS N-STaR 95% RGY is able to achieve similar yields compared
to the 100% RGY while applying less total fertilizer N and reflects results reported by
Roberts et al. (2013) and Davidson et al. (2014). The 95% and 100% RGY option al-
lows producers to select the N fertilizer strategy that best fits their nutrient management
philosophy and capabilities.
Research conducted by Wilson et al. (1989) has shown that N applied as a preflood
application can have a very high fertilizer N use efficiency. Under optimum conditions,
the SPF application translates into less N applied by the grower and potentially lower
application costs. The SPF N-STaR application rates for clay soil were calculated using
the 2-WS N-STaR 95% or 100% RGY N-rate calibration curves and then subtracting a
constant of 20 lb N/acre from the total N rate. For six of the seven sites, the SPF N-STaR
100% RGY was statistically similar or greater than the standard N recommendation while
the N rate difference ranged from 40 to 200 lb N/acre less than the standard N recom-
mendation. The SPF N-STaR 100% RGY at Site P-5 numerically yielded the lowest out
of all N treatments and was statistically lower than the standard N recommendation as
well as the 2-WS 100% RGY at a magnitude of 21 and 22 bu/acre, respectively. The
remaining six sites had no statistical differences between the SPF N-STaR 100% RGY
and the 2-WS N-STaR 100% RGY, although the SPF numerically averaged 4 bu/acre
less in yield than the 2-WS.
The SPF N-STaR 95% RGY recommended a N rate that ranged from 0 to 130 lb
N/acre. For the SPF N-STaR 95% RGY, six of the seven sites were statistically similar
(Sites P-5, P-6, P-8, ES-4, and ES-5) or greater than (Site P-7) the standard N recom-
mendation. At Site ES-3, the SPF N-STaR 95% RGY yielded 23 bu/acre less than the
standard N recommendation, although it did yield statistically greater than the 2-WS
N-STaR 95% RGY at a magnitude of 15 bu/acre. The SPF and 2-WS N-STaR 95%
RGY treatments were statistically similar at all other locations. When the 95% RGY
SPF and 2-WS are examined at sites that were statistically similar, the SPF application
tended to have a larger yield decrease than the 2-WS application when compared to the
standard N recommendation. The yield for the SPF 95% RGY was numerically lower
than the standard N recommendation at four of the five sites and yields ranged from
-17 to 16 bu/acre. Alternatively, the 2-WS 95% RGY yielded numerically lower than
the standard N recommendation at two of four sites with yields ranging from -6 to 14
bu/acre. In general, the SPF tended to produce comparable yields to the 2-WS for the
N-STaR 95% RGY.
229
Site P-7 was unique in that the N-STaR value was very high which resulted in
2-WS and SPF N-STaR 95% and 100% RGY N recommendations that were 0 lb N/
acre. In order to verify the N-STaR recommendation, alternate N rates of 25, 45, 75,
and 90 lb N/acre were included in the trial along with the check plot and the standard N
recommendation. For Site P-7, the check plot received 0 lb N/acre while the standard N
recommendation received 200 lb N/acre, yet the check plot yielded statistically higher
than the standard N recommendation at a magnitude of 72 bu/acre. Excess N has been
known to decrease yields through lodging or by increased disease or pest pressure. In
this case, yield differences between the check and the standard N recommendation were
caused by heavy lodging. When compared to the alternate N treatments, the check plot
yielded statistically higher than the 90 lb N/acre alternate rate and statistically similar
to the remaining alternate rates. Overall, the N-STaR recommendation of 0 lb N/acre
(check plots) yielded the highest with yields tending to decrease with increasing N
fertilizer rates.
Site P-7 was in a field adjacent to Site P-8 and a comparison of the N recom-
mendation predictions show the importance of sampling each field separately when
using N-STaR. The N-STaR N recommendations at Site P-8 ranged from 95 to 145 lb
N/acre, while the N-STaR N recommendations at Site P-7 were 0 lb N/acre. In both
cases, the N-STaR N recommendation maximized yield and profitability. Following
specified sampling protocols is vital in maintaining the effectiveness of the N-STaR N
recommendations, and therefore teaching correct procedures is of utmost importance.
The N-STaR is the first soil-based N test for rice in Arkansas and accurately in-
dexes the mineralization potential of clay soil during rice production. In this study, the
2-WS and SPF N-STaR 95% and 100% RGY calibration curves all produced statisti-
cally similar results with six of seven sites yielding equal to or greater than the standard
N recommendation. This is different from the previous study conducted by Davidson
et al. (2014) where the SPF N-STaR 95% RGY yielded statistically lower than the
standard N recommendation at half of the locations and where there were no statistical
differences between the 2-WS or SPF N-STaR 100% RGY treatments and the standard
N recommendation. Overall, N-STaR has the ability to optimize productivity through
maximum yields and typically lower N input costs as seen in this study. The agronomic
value of the N-STaR N recommendations will continue to grow as N fertilizer prices
and concerns of agricultural pollution continue to increase.
ACKNOWLEDGMENTS
and the University of Arkansas System Division of Agriculture. Appreciation is also
expressed to the N-STaR lab personnel for conducting the analysis of N-STaR samples.
230
LITERATURE CITED
Davidson, J.T., T.L. Roberts, C.E. Greub, A.M. Fulford, N.A. Slaton, J. Hardke, J.
Shafer, S. Williamson, and C. Scott. 2014. Validation of the Nitrogen Soil Test for
Rice (N-STaR) on clay soils in Arkansas. In: R.J. Norman and K.A.K. Mold-
enhauer (eds.). B.R. Wells Arkansas Rice Research Studies 2014. University of
Arkansas Agricultural Experiment Station Research Series 626:195-201. Fayette-
ville, Ark.
Fulford, A.M., T.L. Roberts, R.J. Norman, N.A. Slaton, C.E. Wilson Jr., T.W. Walker,
D.L. Frizzell, C.E. Greub, C.W. Rogers, S.M. Williamson, M.W. Duren, and J.
Shafer. 2012. Evaluation of the Illinois Soil Nitrogen Test and the Nitrogen-Soil
Test for Rice grown on clayey soils. In: R.J. Norman and K.A.K. Moldenhauer
Mulvaney, R.L. and S.A. Khan. 2001. Diffusion methods to determine different
forms of nitrogen in soil hydrolysates. Soil Sci. Soc. Am. J. 65:1284-1292.
Norman, R.J., D. Guindo, B.R. Wells, and C.E. Wilson Jr. 1992. Seasonal accumula-
tion and partitioning of nitrogen-15 in rice. Soil Sci. Soc. Am. J. 56:1521-1527.
Roberts, T.L., R.J. Norman, A.M. Fulford, and N.A. Slaton. 2013. Field validation
of N-STaR for rice produced on silt loam soils in Arkansas. Soil Sci. Soc. Am. J.
77:539-545.
Roberts, T.L., R.J. Norman, N.A. Slaton, C.E. Wilson Jr., W.J. Ross, and J.T. Bush-
ong. 2009. Direct steam distillation as an alternative to the Illinois soil nitrogen
test. Soil Sci. Soc. Am. J. 73:1268-1275.
Roberts, T.L., W.J. Ross, R.J. Norman, N.A. Slaton, and C.E. Wilson Jr. 2011. Pre-
dicting nitrogen fertilizer needs for rice in Arkansas using alkaline hydrolyzable-
nitrogen. Soil Sci. Soc. Am. J. 75:1-12.
Williamson, S.M., T.L. Roberts, C.L. Scott, R.J. Norman, N.A. Slaton, A.M. Fulford,
and C.E. Greub. 2013. Summary of the Nitrogen Soil Test for Rice (N-STaR)
nitrogen recommendations in Arkansas during 2013. In: R.J. Norman and K.A.K.
Moldenhauer (eds.). B.R. Wells Rice Research Studies 2013. University of Arkan-
sas Agricultural Experiment Station Research Series 617:339-344. Fayetteville,
Ark.
Wilson, C.E. Jr., R.J. Norman, and B.R. Wells. 1989. Seasonal uptake patterns of fer-
tilizer nitrogen applied in split applications to rice. Sci. Soc. Am. J. 53:1884-1887.
231
Table 1. Comparison of the 2-way split (2-WS) N-STaR 95% and 100% relative grain yield
232
(RGY), single preflood (SPF) N-STaR 95% and 100% RGY and standard (Std. Rec.) fertilizer N rate
recommendations and the resulting rice grain yields for the four producer (P) and three experiment station (ES) sites utilized in 2015.
P-5 P-6 P-7 P-8
Fertilizer N-rate
recommendation N Rate Yield N Rate Yield N Rate Yield N Rate Yield
(lb N/acre) (bu/acre) (lb N/acre) (bu/acre) (lb N/acre) (bu/acre) (lb N/acre) (bu/acre)
Check 0 129 c 0 149 b 0 221 a 0 172 b
SPF 95% RGY 65 157 ab 80 179 a 25 212 ab 95 217 a
2-WS 95% RGY - - 100 186 a 45 205 ab 115 215 a
SPF 100% RGY 75 153 b 110 183 a 75 214 ab 125 214 a
2-WS 100% RGY 95 175 a 130 195 a 90 200 b 145 216 a
Std. Rec. 170 174 a 170 190 a 200 149 c 200 201 a
ES-3 ES-4 ES-5

Fertilizer N-rate
recommendation N Rate Yield N Rate Yield N Rate Yield
(lb N/acre) (bu/acre) (lb N/acre) (bu/acre) (lb N/acre) (bu/acre)
Check 0 78 e 0 89 b 0 109 c
SPF 95% RGY 95 169 c 130 162 a 130 226 b
2-WS 95% RGY 115 154 d 150 169 a 150 238 ab
SPF 100% RGY 125 186 ab 160 166 a 160 244 ab
2-WS 100% RGY 145 177 bc 180 171 a 180 256 a
Std. Rec. 180 192 a 200 175 a 200 236 ab

RGY = Relative Grain Yield.

Means within a column followed by a different superscript letter are significantly different at the P = 0.05 level.

N-STaR recommendation was for no N fertilizer to be applied at this site. However, alternate SPF N rates of 25, 45, 75, and 90 lb N/acre were
applied to ensure that there would be no response to N fertilizer addition.
RICE CULTURE
Evaluation of Alternative Nitrogen Fertilizer

Application Timings in Four Water Management Regimes
D.L. Frizzell1, J.T. Hardke2, T.L. Roberts3,
R.J. Norman3, E. Castaneda-Gonzalez1, G.J. Lee2, and T.L. Clayton1
ABSTRACT
Approximately 96% of Arkansas rice is grown using the dry-seeded, delayed-flood
system. In this system, nitrogen (N) fertilizer is applied to the crop as a single preflood
(SPF) or two-way split (2WS) application. The large preflood N application is made
around the 4- to 6-lf growth stage and the second application, if needed, is applied
during early reproductive growth. Regardless of N fertilization strategy, the preflood N
application should be made onto dry soil and incorporated with the floodwater to obtain
maximum uptake of the preflood N fertilizer. In recent years, there has been increased
interest in using alternative water management practices as a possible means to save
water resources and lower input costs. The question has arisen of how water practices
might affect N management and whether currently recommended N application methods
are the best for water regimes other than the continuous flood system. Therefore, a study
was initiated in 2014 to determine the effect of alternative N fertilization practices and the
recommended optimum (single preflood) N application timing on rice grain yield within
each of four water management regimes using cultivars commonly grown in Arkansas.
Studies were conducted on a DeWitt silt loam soil using N-(n-butyl) thiophosphoric
triamide (NBPT)-coated urea as the N source. Rice cultivars evaluated included Roy J
and CLXL745 in 2014 and LaKast and XL753 in 2015. When the rice reached the 4- to
5-lf growth stage, four water management regimes were implemented: continuous flood
(CF), straighthead drain (SD), intermittent flood (IF), and flush (FL) irrigation. Plots
received either a SPF N application to dry soil prior to permanent flood establishment
or a split N application based on water regime. During 2014 and 2015, there was no
cultivar N timing interaction. Grain yield was optimized in both study years using a
SPF N application applied to dry soil and water-incorporated in the continuous flood
regime. Grain yield was comparable between N application timings in the straighthead
1
Program Associate III, Program Associate I, and Program Associate - Entomology, respectively, Rice
2
3
Assistant Professor and Professor, respectively, Department of Crop, Soil, and Environmental Science,
Fayetteville.
233
drain and flush irrigated water regimes. Grain yield was optimized in the intermittent
flood regime using a split fertilizer application consisting of 120 lb N/acre applied prior
to flooding followed by 30 lb N/acre applied early in reproductive growth.
INTRODUCTION
Approximately 96% of Arkansas rice is grown using the dry-seeded, delayed-
flood system (Hardke, 2015). In this system, nitrogen (N) is applied at the 4- to 6-lf
growth stage using urea or ammonium sulfate onto a dry soil surface prior to permanent
flood establishment (Norman et al., 2013). The preflood N can be applied as a single
application, termed single preflood (SPF) or may be split into two timings commonly
referred to as a 2-way split (2WS) with approximately 75% of the total N rate being
applied preflood and the remaining 45 lb N/acre applied after the rice has begun repro-
ductive growth. These two options are based on field conditions such as the timeliness
of flood establishment and the ability to maintain an adequate flood for a minimum of
3 weeks. Also, establishing a permanent flood serves two purposes: (1) the ammonium
fertilizer is pushed into the soil profile where it can be taken up by the rice roots, and
(2) the flood maintains an anaerobic environment where the fertilizer is not lost due to
ammonia volatilization or nitrification/denitrification processes.
In recent years, there has been increased interest in using alternative water man-
agement practices as a possible means to save water resources and lower input costs.
The question has arisen of how water practices might affect N management and whether
our currently recommended N application methods are the best for water regimes other
than the continuous flood system. Therefore, a study was initiated in 2014 to determine
the effect of alternative N fertilization practices and the recommended optimum (single
preflood) N application timing on rice grain yield within each of four water management
regimes using cultivars commonly grown in Arkansas.
PROCEDURES
Studies were conducted during 2014 and 2015 at the University of Arkansas
System Division of Agricultures Rice Research and Extension Center near Stuttgart,
Ark., on a DeWitt silt loam soil using N-(n-butyl) thiophosphoric triamide (NBPT)-
coated urea as the N source. The rice was grown using the dry-seeded, delayed-flood
system using cultural management practices common to this system. During 2014,
the rice cultivars Roy J and RiceTec CLXL745 were seeded on 24 April, emerged 10
May, and the permanent flood established on 6 June. During 2015, the cultivars LaKast
and RiceTec XL753 were seeded 6 May, emerged 15 May and the permanent flood
established 10 June. The pure-line varieties, Roy J and LaKast, were drill-seeded at a
rate of 75 lb seed/acre in plots 9 rows (7-inch spacing) wide and 15 ft in length. The
hybrid entries, CLXL745 and XL753 were sown into the same plot configuration using
a seeding rate of 30 lb seed/acre.
At the time of flood establishment when the rice reached the 4- to 5-lf growth stage,
four water management regimes were implemented: continuous flood (CF), straighthead
234
drain (SD), intermittent flood (IF), and flush (FL) irrigation. The CF bay flood depth
was maintained at approximately 4 inches until just prior to harvest. The SD bay was
flooded to the same depth for approximately 10 days then drained and allowed to dry
until the soil was cracked in keeping with recommended practices for management of
straighthead disorder. The bay was then re-flooded based on suggested dates from the
DD50 program and maintained at 4 inches until just prior to harvest. These dates in
the DD50 program for a drain/dry period are based on heat unit accumulation during
the growing season and are timed to reduce the potential for straighthead in susceptible
cultivars. The IF bay was also flooded to a depth of 4 inches but additional water was
added to this bay only when the soil moisture level dropped to a predetermined level of
approximately 20 centibars using soil moisture sensors. These sensors were also used
in the FL bay to determine water needs. A flood was applied and held on this bay for 12
to 24 hours each time the water moisture sensor reached 20 centibars. This period was
determined to be the time required for the sensor reading to drop back to 0 centibars.
Plots were arranged as a 2 (cultivar) 2 (N timing) factorial within each water
management regime. Plots received either a recommended SPF N application to dry
soil prior to permanent flood establishment or a split N application based on water
regime. The split N application timing made to the CF bay consisted of a series of five
applications of 45 lb N/acre into the floodwater beginning 1 day postflood at 7 day
intervals. During 2015, an additional application of 30 lb N/acre was applied at late
boot to the hybrid XL753. The SD bay split N timing plots received one application
of 60 lb N/acre preflood followed by 60 lb N/acre applied just prior to re-flooding.
Each plot in the IF bay received the same preflood N application but the split N timing
plots received an additional 30 lb N/acre applied to damp soil just prior to re-flooding
after reaching the low soil moisture level to reflood. The split N timing plots in the FL
water regime received 60 lb N/acre at the same preflood application time as the other
three water regimes but the flood was only held 12 to 24 hours during each irrigation
event, then released. A second N application was made 10 days after the preflood N
was applied, when the soil moisture was at 20 centibars, and an irrigation event was
used to incorporate the urea into the soil. Statistical analysis were conducted with SAS
9.4 (SAS Institute Inc., Cary, N.C.) and means were separated using Fishers protected
least significant difference test with P = 0.05.

During 2014 and 2015, there was no cultivar N timing interaction in any of
the four water management regimes. Only the main effect of N application timing was
significant. It should be noted that grain yield data was not combined across years to
be able to look at individual data sets from two dissimilar growing seasons.
In the CF water management regime, grain yield was optimized in both 2014
and 2015 using a SPF N application applied to dry soil and flood-incorporated (Tables
1 and 2). Average grain yield was similar regardless of N application timing in the SD
regime. These results suggest that a properly timed drain for management of straighthead
disorder can be achieved without yield loss and that N fertility may be more flexible in
235
this situation. However, it should be noted that this trial was not conducted in an area
prone to straighthead disorder; the results simply indicate that the prescribed drain can
be completed without sacrificing yield regardless of whether straighthead conditions
are present.
Interestingly, grain yield was greater during 2014 and 2015 using the split N ap-
plication timing in the IF water management regime. As noted above, all plots received
the same preflood N rate prior to flooding. In 2014 a standing flood was lost early but
regular rainfall prevented the soil from reaching the low soil moisture level needed to
re-flood, and the second portion of the split N application was not made until 3 July,
27 days after initial flood establishment. In 2015 the trial quickly reached the low soil
moisture level of 20 centibars and was re-flooded just 12 days after the initial flood,
and the second portion of the split N application wasnt made until 1 July, 21 days after
initial flood establishment. The timing of the second N application corresponds to stud-
ies by Wilson et al. (1989) showing peak fertilizer N uptake occurred within 21 days of
permanent flood and remained steady through the 28 day sampling period. However,
when the permanent flood is lost and re-established prior to peak fertilizer N uptake
at ~21 days, N loss can occur and the (split) N application used in this trial may have
acted as a supplement for N lost prior to the time of peak uptake. The results of the IF
water regime in this study helped to emphasize current recommendations to maintain
the initial permanent flood for at least 3 weeks before beginning the alternate wet and
dry periods used in the intermittent flood system (Wilson et al., 1989).
Grain yield was similar between the two N timings in the FL water regime dur-
ing both study years. Although comparisons between water management regimes were
not made for the scope of this report, numerical grain yields obtained in the FL water
regime were notably lower than those of the other three. In many row rice or furrow-
irrigated fields in Arkansas, producers have been able to obtain yields similar to those
seen in fields where a continuous flood is maintained throughout the growing season.
The low yields reflected in the FL bay are representative of only a small portion of fields
irrigated with a flush/furrow system as the middle and bottom of the fields typically
remain saturated or flooded while the upper portion alone is flushed. Adjustments in
experimental design are needed before pursuing further research in best management
practices for furrow-irrigated rice to better represent in-field soil moisture variation.
It should be noted there were no blast disease symptoms in the study during either
2014 or 2015 that would have had a negative impact on grain yield. Blast disease pres-
sure is greater in fields where a deep flood cannot be maintained and has the potential
to negatively impact grain yield in fields using water management practices of alternate
wetting and drying such as the IF and FL water regimes.
Results from the water management study will aid University of Arkansas System
Division of Agriculture personnel in answering grower questions concerning the best
way to fertilize rice grown using alternate water management regimes.
236
ACKNOWLEDGMENTS
This and other studies are supported by the University of Arkansas System Divi-
sion of Agriculture and through grower check-off funds administered by the Arkansas
Rice Research and Promotion Board. We wish to thank the following people for their
dedication to making these studies possible each year: Chuck Pipkins, Taylor Roush,
Logan Stokes, and Greg Wilson.
LITERATURE CITED
Hardke, J.T. 2015. Trends in Arkansas rice production. In: R.J. Norman and K.A.K.
Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies 2014. University
of Arkansas Agricultural Experiment Station Research Series 626:11-22. Fayette-
ville, Ark.
Norman, R.J., N.A. Slaton, and T.L. Roberts. 2013. Soil Fertility. pp. 69-101. In: J.T.
Hardke (ed.). Arkansas Rice production handbook. Misc. Publ. 192. University of
Arkansas Cooperative Extension Service, Little Rock, Ark.
Wilson, C.E. Jr., R.J. Norman, and B.R. Wells. 1989. Seasonal uptake patterns of fer-
tilizer nitrogen applied in split applications to rice. Soil Sci. Soc. Am. J. 53:1884-
1887.
237
Table 1. Influence of nitrogen (N) fertilizer application timing on rice grain yield
among four water management regimes at the University of Arkansas System Division
of Agricultures Rice Research and Extension Center near Stuttgart, Ark., during 2014.
Water management regime
N timing Continuous flood Straighthead drain Intermittent flood Flush
----------------------------------------------(bu/acre)--------------------------------------------
SPF Na 191 193 184 143
Split Nb 153 193 198 145
LSD0.05c 19.6 NSd 12.3 NS
a
Single preflood nitrogen application, 120 lb N/acre applied to dry soil prior to flooding.
b
Continuous flood: 5 applications of 46 lb N/acre applied once every 7 days for 5 weeks begin-
ning at flood; Straighthead drain: 60 lb N/acre preflood and 60 lb N/acre before reflood follow-
ing drain; Intermittent flood: 120 lb N/acre preflood plus 30 lb N/acre approximately 3 weeks
later before reflood; and Flush: 60 lb N/acre preflood and 60 lb N/acre 10 days later.
c
d
Table 2. Influence of nitrogen (N) fertilizer application timing on rice grain yield
among four water management regimes at the University of Arkansas System Division
of Agricultures Rice Research and Extension Center near Stuttgart, Ark., during 2015.
Water management regime
N timing Continuous flood Straighthead drain Intermittent flood Flush
----------------------------------------------(bu/acre)--------------------------------------------
SPF Na 239 235 219 168
Split Nb 210 236 235 162
LSD0.05c 14.5 NSd 12.3 NS
a
Single preflood N application of 120 lb N/acre applied to dry soil prior to flooding.
b
Continuous flood: 5 applications of 46 lb N/acre applied once every 7 days for 5 weeks begin-
ning at flood; Straighthead drain: 60 lb N/acre preflood and 60 lb N/acre before reflood follow-
ing drain; Intermittent flood: 120 lb N/acre preflood plus 30 lb N/acre approximately 3 weeks
later before reflood; and Flush: 60 lb N/acre preflood and 60 lb N/acre 10 days later.
c
d
238
RICE CULTURE
Effect of Delayed Nitrogen Fertilization into the Floodwater

D.L. Frizzell1, J.T. Hardke2, T.L. Roberts3,
R.J. Norman , E. Castaneda-Gonzalez1, G.J. Lee2, and T.L. Clayton1
3
ABSTRACT
Nitrogen (N) fertilizer (typically urea or ammonium sulfate) is applied in the dry-
seeded, delayed-flood system common to Arkansas rice production as a single preflood
(SPF) or two-way split (2WS) application. The large preflood N application is made
around the 4- to 6-lf growth stage onto dry soil and the second application, if needed,
is applied during early reproductive growth. Rainy weather in recent years during the
recommended window of application have made the likelihood of dry soil conditions
unlikely and questions have arisen concerning the best management strategy to preserve
rice grain yield and optimize N efficiency in those fields. Therefore, a study was initiated
in 2015 to address these concerns. Treatments to LaKast rice consisted of the recom-
mended SPF with all N applied preflood onto dry soil, a suboptimum 2WS option with
a reduced preflood N rate (to mimic wet soil conditions) applied to dry soil followed
by midseason N, variations of sequential fertilizer applications into the floodwater, and
a no-N fertilizer check. Treatments were initiated based on dates noted from a DD50
report generated for this study. The SPF N fertilizer application resulted in significantly
higher grain yield than the 2WS treatment. Both of these options produced grain yields
significantly higher than any of the treatments made only into the floodwater.
INTRODUCTION
Approximately 96% of Arkansas rice is grown using the dry-seeded, delayed-
flood system (Hardke, 2015). In this system, nitrogen (N) is applied at the 4- to 6-lf
growth stage using urea or ammonium sulfate onto a dry soil surface prior to perma-
nent flood establishment (Norman et al., 2013). The preflood N can be applied as a
single application, termed single preflood (SPF) or may be split into two timings
1
Program Associate III, Program Associate I, and Program Associate - Entomology, respectively, Rice
2
Crop, Soil, and Environmental Sciences], Stuttgart.
3
Assistant Professor and Professor, respectively, Department of Crop, Soil, and Environmental Science,
Fayetteville.
239
commonly referred to as a 2-way split (2WS) with approximately 75% of the total
N rate being applied preflood and the remaining 45 lb N/acre applied after the rice has
begun reproductive growth. These two options are based on field conditions such as
the timeliness of flood establishment and the ability to maintain an adequate flood for
a minimum of 3 weeks. Also, establishing a permanent flood serves two purposes: (1)
the ammonium or ammonium forming N fertilizer is moved into the soil profile with the
wetting front where it can be taken up by the rice roots, and (2) the flood maintains an
anaerobic environment where the N fertilizer is not lost due to ammonia volatilization
or nitrification/denitrification processes.
Occasionally, questions arise concerning the presence of muddy or flooded field
conditions in the recommended window for preflood N fertilizer application. Therefore,
a study was initiated in 2015 to determine the best management strategy for applying
N fertilizer in fields where weather conditions prevent dry soil conditions.
PROCEDURES
A study was initiated during 2015 utilizing N-(n-butyl) thiophosphoric triamide
(NBPT)-coated urea applied to LaKast rice at various application timings. The study
was conducted at the University of Arkansas System Division of Agricultures Rice
Research and Extension Center near Stuttgart, Ark., on a DeWitt silt loam soil. The rice
was drill-seeded at a rate of 75 lb seed/acre in plots 9 rows (7-inch spacing) wide 15
ft in length on 15 May, emerged 22 May, and the permanent flood established 11 June.
Cultural management practices used were standard to the dry-seeded, delayed-flood
production system. Nitrogen fertilizer treatments evaluated are listed in Table 1 and
consisted of the recommended SPF and 2WS options with the preflood N applied to
dry soil, variations of sequential fertilizer applications into the floodwater, and a no-N
fertilizer check. The 2WS treatment was modified so that a reduced preflood N rate was
used to mimic N loss due to application onto wet soil. Treatments were initiated based
on dates noted from a DD50 report generated for this study. One treatment made solely
into the floodwater was initiated 1 day following permanent flood establishment. The
remaining treatments made into the floodwater were initiated at the final recommended
date to apply preflood N as determined by heat unit accumulation in the DD50 program
which occurred approximately 10 days after flooding. The study was arranged as a
randomized complete block and means were separated using Fishers protected least
significant difference test with P = 0.05.

During this initial study year, the SPF N fertilizer treatment resulted in signifi-
cantly higher grain yield than the 2WS treatment (Table 1). The recommended 2WS
treatment would normally receive 25 lb N/acre more than the SPF treatment; however,
in this study the rate was reduced to simulate an ineffective preflood N application to
muddy soil. As a result, the 2WS treatment for this study was determined by subtracting
240
the standard midseason N rate of 45 lb N/acre from the SPF rate of 120 lb N/acre. The
remaining 75 lb N/acre was applied at preflood onto dry soil followed by the standard
midseason N rate of 45 lb N/acre into the floodwater at beginning reproductive growth.
Both the SPF and 2WS treatments produced grain yields higher than any of the treat-
ments made only into the floodwater. Of N applications made into the floodwater, all
of those treatments that were delayed and initiated at the final preflood application time
according to the DD50 report produced greater yields than the treatment initiated at 1
day after flood establishment. For the delayed in-flood treatments, application of 45 lb
N/acre every 7 d (total of 5 applications) resulted in the greatest grain yield compared
to all treatments except the SPF and 2WS treatments. These results suggest that the SPF
treatment applied to dry soil continues to be the best option for N fertilizer manage-
ment; and the 2WS treatment in this study (designed to mimic preflood N application
to muddy soil) is still a better option than the spoonfeed approach of N applications
into the standing flood.
Results from this study will aid University of Arkansas System Division of Ag-
riculture personnel in answering grower, agent and consultant questions concerning N
management decisions on rice fields where the preflood N fertilizer cannot be applied
according to University of Arkansas System Cooperative Extension Service recom-
mendations onto a dry soil surface.
ACKNOWLEDGMENTS
This and other studies are supported through grower check-off funds adminis-
tered by the Arkansas Rice Research and Promotion Board and are much appreciated.
Also, we appreciate the funding and support from the University of Arkansas System
Division of Agriculture. We wish to thank the following people for their dedication to
making these studies possible each year: Chuck Pipkins, Taylor Roush, Logan Stokes,
and Greg Wilson.
LITERATURE CITED
Hardke, J.T. 2015. Trends in Arkansas rice production. In: R.J. Norman and K.A.K.
Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies 2014. Univer-
sity of Arkansas Agricultural Experiment Station Research Series 626:11-22.
Fayetteville, Ark.
Hardke (ed.). Arkansas rice production handbook. Misc. Publ. 192. University of
Arkansas Cooperative Extension Service, Little Rock, Ark.
241
Table 1. Influence of nitrogen (N) fertilizer application timing on

the grain yield of LaKast rice at the University of Arkansas System Division of
Agricultures Rice Research and Extension Center near Stuttgart, Ark. during 2015.
Treatment N application Total N Grain

no. N timing Frequency Rate fertilizer applied yield
----------- (lb N/acre)--------- (bu/acre)
1 Control none 0 0 71 g
2 SPF PF 120 120 186 a
3 2WS PF fb MS 75+45 120 165 b
4 Flood initiation 7-8d intervals 5 @ 45 225 117 f
5 Final DD50# 4-5d intervals 5 @ 45 225 141 d
6 Final DD50 7-8d intervals 5 @ 45 225 155 c
7 Final DD50 4-5d intervals 4 @ 60 240 138 d
8 Final DD50 7-8d intervals 4 @ 60 240 140 d
9 Final DD50 4-5d intervals 45+60+60+45 210 127 e
LSD0.05 7.6
PF = preflood, fb = followed by, and MS = midseason.
Means followed by the same letter are not significantly different (P = 0.05).
One day postflood (12 June -- two days after initial recommended date to apply preflood N
fertilizer when rice has reached 4- to 5-lf growth stage based on DD50 Rice Management
Program).
5 @ 45 represents 5 applications of 45 lb N/acre at each application.

#
Final DD50 = final recommended date (i.e., 10 days postflood) to apply preflood N fertilizer (22
June for this trial) based on DD50 Rice Management Program.
242
RICE CULTURE
Validation of Soil-Test Based Fertilizer Recommendations for Rice

M. Fryer1, N.A. Slaton2, T.L. Roberts2, J. Hardke3, R.J. Norman2,
R.E. DeLong2, Y. Liyew4, D.L.Frizzell5, J. Pace6, and M.W. Duren7
ABSTRACT
Fertilizer recommendations for phosphorus (P) and potassium (K) are often based on
soil-test results, but recommended fertilizer nutrient rates may differ among labs or
consultants providing the recommendations. The research objective was to validate
the accuracy of the existing University of Arkansas System Division of Agriculture
(UASDA) soil-test based fertilizer-P and -K recommendations for predicting rice (Oryza
sativa L.) yield response to fertilization. Eight trials were established at four UASDA
research centers/stations in 2015. Statistical comparisons were evaluated at three sig-
nificance levels (P 0.05, 0.10, and 0.25) to validate the existing recommendations.
The treatments evaluated rice yield response to: i) with vs without fertilizer-P, ii) with
vs without fertilizer-K, and 3) the recommended fertilizer-P and -K rates compared
to no fertilizer-P and -K. The overall soil-test interpretation accuracy (P 0.25) of
predicting the correct rice yield response was 40% to fertilizer-P and 75% for -K. The
level of significance at which results were interpreted affected the accuracy only for
K (decreased to 50% at P 0.10). The most common error in recommendations was a
false positive meaning the soil-test interpretation indicated the need for fertilizer but
crop yield was not increased.
INTRODUCTION
Fertilizer recommendations for P and K are usually based on soil-test results, but
recommended fertilizer nutrient rates from the same soil analysis may differ among
the laboratories or consultants providing the recommendation (Kleinman et al., 2001).
Differences in nutrient management philosophy, poor understanding of the dynamics of
1
County Extension Agent - Agriculture, Crawford County, Van Buren.
2
Professor, Assistant Professor, Professor, and Program Associate II, respectively, Department of Crop,
Soil, and Environmental Science, Fayetteville.
3
4
Research Program Technician, Pine Tree Research Station, Colt.
5
6
Program Technician, Southeast Research and Extension Center, Hope.
7
243
soil nutrient availability and how fertilizer recommendations are developed, misconcep-
tions of the accuracy of the soil-test-based recommendation, or combinations of these
and other factors contribute to confusion among the end-users of soil-test information.
An ongoing project has been evaluating the accuracy of existing fertilizer-P and
-K recommendations of the University of Arkansas System Division of Agriculture
(UASDA) to identify where adjustments might be needed to make predictions regard-
ing what nutrients are needed and how much to apply more accurately (Fryer, 2015).
Information included in this report summarizes the third and final year of the research.
The overall goal of the research was to improve the accuracy of soil-test based P and K
fertilizer recommendations for flood-irrigated rice. The objective of research presented
in this report was to evaluate whether rice responded to the: i) currently recommended
fertilizer-P and -K rates, ii) recommended fertilizer-P rate alone, and iii) recommended
fertilizer-K rate alone, all compared to rice that received no fertilizer-P or -K. The yield
of rice was expected to benefit from fertilization when soil-test P and K levels were
Very Low, Low, or Medium and no yield increase was predicted when soil-test levels
were Optimum or Above optimum.
PROCEDURES
Eight fertilization trials were established at four UASDA research center/stations
during 2015 including the Northeast Research and Extension Center (NEREC), Pine Tree
Research Station (PTRS), Rice Research and Extension Center (RREC), and Rohwer
Research Station (RRS). Soil and agronomic information as well as the field name
used for each site in this report are listed in Table 1. Initial soil samples were collected
in the spring to define the UASDA fertilizer treatments at each site. Individual plots
of rice were 9 rows wide (~5.6-ft wide) 18- to 20-ft long. Plot borders were marked
and 0- to 4-in. deep samples were taken from each replicate (n = 6). Plant-available,
soil nutrients were extracted using the Mehlich-3 solution and determined analytically
by inductively coupled plasma spectroscopy. Selected soil chemical property means
are listed in Table 2.
Each trial contained a total of six treatments with four K2O rates (0, 60, 90 and 120
lb K2O/ace) and two P2O5 (0 and 60 lb P2O5/acre) rates including: 1) the recommended
P2O5 rate plus 0 lb K2O/acre, 2) the recommended P2O5 rate plus 60 lb K2O/acre, 3) the
recommended P2O5 rate plus 90 lb K2O/acre, 4) the recommended P2O5 rate plus 120
lb K2O/acre, 5) the recommended K rate plus the second P2O5 rate, and 6) no P2O5 or
K2O fertilizer (control). The six treatments in each trial were organized as a randomized
complete block design with six blocks. Crop management practices closely followed
recommendations from the UASDA Cooperative Extension Service. Urea-nitrogen (N)
fertilizer treated with a urease inhibitor was applied using the N-STaR recommended
rate near the 5-lf stage and the rice was flooded. Additional N was applied to some
sites at midseason based on a visual assessment of growth and knowledge of the soil
conditions when preflood-N was applied.
Whole aboveground plant samples were collected from selected plots at the midtil-
lering and early heading stages to assess plant P and K concentrations. A harvested seed
244
sample was collected from each plot at maturity. Plant and seed samples were weighed,
digested, and analyzed for nutrient concentrations, but the results will not be reported
here. A small plot combine harvested 5 to 8 of the 9 rows in each plot. Weights and
moistures were recorded, and rice grain moisture was adjusted to 12% in the final yield
calculations which are expressed in bushels (bu)/acre.
Three single-degree-of-freedom contrasts were performed using the MIXED pro-
cedure in SAS v. 9.2 (SAS Institute Inc., Cary, N.C.) to evaluate grain yield differences.
Specific comparisons were made by comparing rice plots receiving: 1) fertilizer-P (no
K), 2) fertilizer-K (no P), and 3) the recommended fertilizer-P and -K rates all compared
to the rice that received no fertilizer-P or -K. Three levels of significance (P 0.05,
0.10, and 0.25) were used to define yield differences. Fertilization effects on rice yield
were categorized as a yield increase, no yield change, or yield decrease. The hypothesis
was that soils with Very Low, Low, and Medium P and K levels would show a yield
increase to fertilizer, and soils with Optimum and Above Optimum P and K levels would
show no change in yield. For sites with Medium soil-test P and K levels, smaller and
less frequent yield increases were expected. A yield decrease was not expected from
P and K fertilization in any soil-test level, but it was included as a possible outcome.

Fertilization studies were conducted at six loamy soil sites and two clayey soil sites
in 2015 (Table 1). The two clayey soils (NEREC and RRS-CL) were the only two sites
where no yield increase was expected from fertilizer-P, fertilizer-K, or their combina-
tion (Table 2). At the six loamy sites (PTRS-I10, PTRS-L2, PTRS-MJC, PTRS-F18,
RRS-SL, and RREC), the soil-test level interpretations suggested that rice grain yield
would increase from the application of both P and K. Rice yield was not affected by
P and K fertilization at the two clayey sites which had Optimum (NEREC) or Above
Optimum (RRS-CL) soil-test P and Above Optimum soil-test K (NEREC and RRS-CL;
Table 3). Rice grain yield was not changed by fertilizer-P application at the six loamy
soil sites, which contained suboptimal soil-test P levels, when evaluated at P 0.05
and 0.10, but when evaluated at P 0.25, rice yield at PTRS-I10 was reduced by 4.6%
(9 bu/acre) compared to rice that received no fertilizer-P or -K.
Soil-test K levels were Very Low (PTRS-L2), Low (PTRS-I10), or Medium
(PTRS-MJC, PTRS-F18, RRS-SL, and RREC) for the six loamy soil sites (Table 2).
Grain yield responses to fertilizer-K occurred at one (PTRS-I10; P 0.05) or two (PTRS-
I10 and PTRS-L2; P 0.25) of the six loamy sites (Table 3). The average yield increase
to K fertilization at the two responsive sites was 11 bu/acre or 6.1%. The four remaining
sites with suboptimal soil-test K levels did not have significant yield changes from K
fertilization. Rice yield response to the recommended combination of fertilizer-P and -K
on the six loamy soils depended on significance level and showed that one (PTRS-L2;
P 0.05) or four (PTRS-L2, PTRS-I10, PTRS-MJC, and PTRS-F18; P 0.25) of the
sites responded positively to fertilization, with an average increase of 8 bu/acre or 4.6%.
245
Table 4 shows the overall accuracy of the soil-test interpretations. The accuracy
of soil-test P interpretation (40%) was not affected by the level of significance used in
the evaluation, with all of the error occurring in the suboptimal soil-test P levels where
a yield increase was expected (false positive error). Like soil-test P, all of the error in the
soil-test K interpretation occurred within the suboptimal level, but the amount of error
decreased when results were evaluated at the 0.25 significance level. Soil-test K was
50% (P 0.05 and 0.10) to 75% (P 0.25) accurate. The existing soil-test K recom-
mendations appear to work reasonably well on soils with Low and Very Low K levels
and serve to maintain K fertility on the soils that test Medium. Although the tissue K
results were not included in this report, the results showed that rice-K concentrations
have a strong correlation with soil-test K, which agrees with prior research (Slaton
et al., 2009; Fryer, 2015). When the overall (P and K) soil-test recommendation was
compared to the no-fertilizer control, accuracy was numerically greater (58% to 83%
at P 0.05 and 0.25, respectively) than the evaluations made to P or K only, but the
results tended to mimic the fertilizer-K response tendencies. Overall, when soil-test P
or K was Optimum or Above Optimum, soil-test interpretations were 100% accurate,
indicating false negative errors seldom occur for properly collected soil samples. Soil-test
P levels should be lowered to encompass the error that occurs in the suboptimal levels.
ACKNOWLEDGMENTS
Research was funded by a grant from the Rice Check-off Program administered
by the Arkansas Rice Research and Promotion Board, Fertilizer Tonnage Fees, and the
University of Arkansas System Division of Agriculture.
LITERATURE CITED
Fryer, M.S. 2015. Validation of soil-test-based phosphorus and potassium fertilizer
recommendations for rice and soybean. M.S. Thesis. University of Arkansas,
Fayetteville Available from ProQuest Dissertations and Theses database. No.
1605004.
Kleinman, P.J.A., K. Gartley, and W.M. Jarrell. 2001. Interlaboratory comparison of
soil phosphorus extracted by various soil test methods. Commun. Soil Sci. Plant
Anal., 32:2325-2345.
Slaton, N.A., B.R. Golden, R.J. Norman, C.E. Wilson Jr., and R.E. DeLong. 2009.
Correlation and calibration of soil potassium availability with rice yield and nutri-
tional status. Soil Sci. Soc. Am. J. 73:1192-1201.
246
Table 1. Selected soil and agronomic information for eight rice fertilization trials conducted at multiple sites during 2015.
Previous crop and fertilizerb
a
Site Soil series Cultivar Row width Planting date Crop P2O5 K2O
(in.) --------- (lb/acre)-------
PTRS-I10 Calloway Roy J 7.5 30 April Soybean 0 0
PTRS-L2 Calloway CL111 7.5 1 May Soybean 0 0
PTRS-MJC Calhoun CL172 7.5 8 April Fallow 0 0
PTRS-F18 Calhoun Roy J 7.5 30 April Soybean 0 0
RRS-SL Desha CL172 7.0 5 May Soybean 0 0
RREC Dewitt LaKast 7.0 1 May Soybean 40 60
NEREC Sharkey CL172 7.0 6 May Soybean 0 0
RRS-CL Sharkey/Desha CL172 7.0 5 May Rice 0 0
a
Abbreviations include: NEREC, Northeast Research and Extension Center; PTRS, Pine Tree Research Station; RREC, Rice Research and Ex-
tension Center; and RRS, Rohwer Research Station. The letter or letters after the site abbreviation represent the field name.
b
Crop grown and fertilizer applied during the 2014 growing season.
247
Table 2. Selected soil (0- to 4-in. depth) chemical property means (n = 6) from the
248
unfertilized control in eight phosphorus (P) and potassium (K) fertilization trials during 2015.
Soil test levelb 4-in. depth samplec
a
Site pH P K P K Ca Mg S Zn SOM
-------------------------------------------(ppm)------------------------------------ (%)
PTRS-I10 7.2 Low Low 25 (3) 66 (3) 1652 281 8 2.8 2.2
PTRS-L2 7.4 Very Low Very Low 7 (<1) 37 (2) 1660 302 7 0.6 2.5
PTRS-MJC 7.3 Low Medium 22 (2) 100 (8) 1636 314 7 2.7 2.3
PTRS-F18 7.2 Very Low Medium 9 (<1) 97 (9) 1650 303 7 2.6 2.9
RRS-SL 7.5 Low Medium 16 (4) 117 (9) 2203 592 6 1.8 2.1
RREC 6.7 Medium Medium 35 (4) 120 (12) 1299 157 8 7.9 1.7
NEREC 7.1 Optimum Above Opt 45 (3) 260 (7) 4211 903 8 4.0 3.5
RRS-CL 7.8 Above Opt. Above Opt 66 (2) 235 (11) 4213 747 10 3.2 2.7
a
tension Center; and RRS, Rohwer Research Station. The letter or letters after the site abbreviation represent the field name.
b
The soil-test levels are based on the interpretation when the project was initiated (Opt. = Optimum).
c
The value in parentheses following each P and K mean represents the standard deviation.
Table 3. Expected rice yield response to phosphorus (P), potassium (K), or P and K
fertilization compared to a no P and K control at eight research sites established during 2015.
Recommended fertilizer
Check Expected yield responseb Yield response tod
Sitea P2O5 K2O P K yieldc P only K only Recommendede
-------- (lb/acre)------ (bu/acre)
--- [bu/acre yield difference (P-value)]---
PTRS-I10 60 90 Increase Increase 197 -9 (0.25)f +17 (0.01) +9 (0.19)
PTRS-L2 60 120 Increase Increase 154 -4 (0.27) +5 (0.13) +8 (0.03)
PTRS-MJC 60 60 Increase Increase 135 0 (0.94) +6 (0.26)g +7 (0.16)
PTRS-F18 60 60 Increase Increase 216 -2 (0.81) +1 (0.83) +8 (0.20)
RRS-SL 60 60 Increase Increase 164 -3 (0.63) +4 (0.47) +1 (0.91)
RREC 60 60 Increase Increase 151 +1 (0.71)h +1 (0.85) +2 (0.54)
NEREC 0 0 No Change No Change 169 -3 (0.66) -3 (0.60) NA
RRS-CL 0 0 No Change No Change 133 +2 (0.79) 0 (0.99) NA
a
tension Center; RRS, Rohwer Research Station. The letter or letters after the site abbreviation represent the field name.
b
Expected Response: Increase, soil-test level is Very Low, Low, or Medium; and No Change, soil-test level is Optimum or Above Optimum.
c
Check yield, the mean yield of rice that received no fertilizer-P or -K.
d
Yield response: P only, single-degree-of-freedom contrast comparing the yield of rice receiving no fertilizer-P or -K to rice receiving fertilizer-P;
K only, single-degree-of-freedom contrast comparing the yield of rice receiving no fertilizer-P or -K to rice receiving the recommended fertilizer-K
rate; and Recommended, single-degree-of-freedom contrast comparing the yield of rice receiving no fertilizer-P or K to that of rice fertilized with
the recommended rates of both fertilizer-P and -K.
e
NA indicates that the comparison was not possible. The soil-test recommended no fertilizer-P and K.
f
Comparison was made using plots that received the same K rate (90 lb K2O/acre) but different P rates.
g
Comparison was made using plots that received the same P rate (60 lb P2O5/acre) but different K rates.
h
Comparison was made using plots that received the same K rate (60 lb K2O/acre) but different P rates.
249
Table 4. Site responses and accuracy of the soil-test prediction of rice yield
250
response (see Table 3) to phosphorus (P) and potassium (K) fertilization at eight research sites in 2015 as
defined by soil-test P (STP) and K (STK) level and the level of significance at which statistical comparisons were made.
Interpreted at P-value 0.05c Interpreted at P-value 0.10c Interpreted at P-value 0.25c
Nutrient Soil-test Soil-test Total
evaluationa levelb concentration sites I NC D I NC D I NC D
(ppm) ----------------------------------------- (number of sitesd)-------------------------------------------
P-only VL <16 2 0 2 0 0 2 0 0 2 0
P-only L 16-25 3 0 3 0 0 3 0 0 2 1
P-only M 26-35 1 0 1 0 0 1 0 0 1 0
P-only O 36-50 1 0 1 0 0 1 0 0 1 0
P-only AO >50 1 0 1 0 0 1 0 0 1 0
Overall STP Accuracy (%)e ------------ 40%----------- -----------40%--------- ----------- 40%---------
K-only VL <61 1 0 1 0 0 1 0 1 0 0
K-only L 61-90 1 1 0 0 1 0 0 1 0 0
K-only M 91-130 4 0 4 0 0 4 0 0 4 0
K-only O 131-175 0 -- -- -- -- -- -- -- -- --
K-only AO >175 2 0 2 0 0 2 0 0 2 0
Overall STK Accuracy (%)e ------------ 50%----------- -----------50%--------- ----------- 75%---------
P&K Recommendedf 6 1 5 0 1 5 0 4 2 0
None Recommendedf 2 0 2 0 0 2 0 0 2 0
Recommendation Accuracy (%)e ------------ 58%----------- -----------58%--------- ----------- 83%---------
a
Evaluation of the response of rice fertilized with P-only, K-only, and the Recommended P&K all compared to no fertilizer. None is compared to
P-only and K-only.
b
Soil-test level abbreviations: VL, Very Low; L, Low; M, Medium; O, Optimum; AO, Above Optimum.
c
Abbreviations: I, Increase (yield); NC, No Change in yield; D, Decrease (yield).
d
Values that are bold and italic represent the anticipated correct yield response to fertilizer applications. Note: small and less frequent yield in-
creases were expected in the M soil-test level compared to the VL and L soil-test levels for both P and K.
e
Accuracy calculated as the weighted average for the five soil-test levels where the number of sites with the correct outcome (see footnote b) is
divided by the number of sites. Overall STK Accuracy was calculated using only four soil-test levels because no site fell within the Optimum soil-
test level.
f
The evaluation of soil-test P and K recommendations regardless of the soil-test level and concentration of each site.
RICE CULTURE
Effects of Three Different Alternate Wetting

and Drying Regimes in Rice Cultivation on Yield, Water
Use, and Water Use Efficiency in a Clay Soil During a Wet Year
J.P. Gaspar1, C.G. Henry1, M.W. Duren2, A.P. Horton1, and H. James1
ABSTRACT
Water available for irrigation is declining in many rice-growing regions around the
world. Global populations continue to rise, increasing crop production demand. Rice
production systems must face the dilemma of maintaining or increasing yields with less
water available to irrigate. Alternate wetting and drying (AWD) has shown to be an ef-
fective tool for water conservation in irrigated rice systems. Research on AWD practices
is lacking and more information is needed to verify the success of AWD across varying
soil types. More work is needed to develop clear recommendations for AWD irrigation
practices in Arkansas. In this study we compared the effects of three different AWD
regimes and a continuous flood management on rice yields and water-use efficiency
(WUE) from a conventional, pure-line cultivar (Roy J) and a hybrid (XL753). The
study was located in the northeast corner of the Mississippi delta rice-growing region in
Arkansas and results were complicated by a high rainfall pattern in 2015, and unknown
factors contributing to low yields even in the conventionally flooded treatments. Even
with these complications, the trends in the data indicated that AWD is a feasible water
management practice for rice in Arkansas. For both cultivars, all AWD regimes tested
in this experiment were associated with a loss in yield, the hybrid cultivar had a higher
yield than the conventional cultivar in all treatments. Water-use efficiency for the wet-
test AWD treatment was higher than the conventional flood treatments and the dryer
AWD treatments. Difference in WUE between cultivars was significant and suggests
that the hybrid may have a higher WUE than the conventional.
INTRODUCTION
Water available for irrigation is declining in the main crop-growing regions.
Irrigation is the largest component of fresh water use (Haddeland et al., 2014). High
water use and drought are depleting water available for human use (Schewe et al.,
1
Program Associate, Assistant Professor, Program Technician, and Field Technician, respectively, De-
partment of Biological and Agricultural Engineering, Rice Research and Extension Center, Stuttgart.
2
251
2014). The alluvial aquifer in the east-central region of Arkansas is being depleted at
unsustainable rates (ANRC, 2012). It has been estimated that 1.8 billion people will
be living in regions with absolute water shortages and as much as two-thirds of the
global population may be under water stress conditions by 2025 (FAO, 2013). At the
same time global populations continue to rise, increasing crop production demand.
Ray et al. (2013) estimates that global crop production needs will double by 2050 with
an increase of 2.4% annually. Agricultural production systems must face the dilemma
of maintaining or increasing yields with less water available to irrigate. Globally, rice
production systems account for one-third of the total fresh water use (Bouman, 2009).
Although rice and other crops have similar transpiration rates, substantially more water
loss is associated with anaerobic rice cultivation practices than aerobic crop produc-
tion systems due to soil percolation losses and evapotranspiration (Bouman, 2009).
Water shortages coupled with the high costs associated with irrigation create the need
to research alternate production methods that minimize water use while maximizing/
maintaining yields. This can also be referred to as water-use efficiency (WUE) measured
as unit of grain per area divided by the volume of water applied per area. Such informa-
tion will help guide rice producers that face the dilemma of water shortages first hand
and provide viable alternative methods to minimize profit losses.
One such method that has been receiving increased attention in recent years is a
rice production method referred to as alternate wetting and drying (AWD). Alternate
wetting and drying combines the beneficial side effects of anaerobic rice cultivation
(nematode and weed control), and aerobic cultivation practices (reduction in water use,
grain toxin builds, and greenhouse gas emissions; Price et al., 2013). Alternate wetting
and drying has shown to be an effective tool for water conservation in rice-production
systems. Zhang et al. (2009) found that AWD can lower water use in rice production
by ~35%, while maintaining and even increasing rice yields relative to continual flood
methods. Not only does this method reduce water use, but also it has been shown to be
very effective in reducing greenhouse gas emissions that result from the brief aerobic
periods (Yan et al., 2005; Feng et al., 2013), and at reducing buildup of arsenic in rice
grains (Takahashi et al., 2004; Talukder et al., 2012).
In the literature, AWD methods in comparison to anaerobic rice cultivation have
a range of results: no difference in yields, yield increases, and yield decreases. Davies
et al. (2011) reviewed existing literature and found that mixed results on yield differ-
ences is likely dependent on severity of the soil moisture deficit during the dry-down
events. This implies that target deficits will vary with differences in soil characteristics.
An extensive study has been conducted in the Grand Prairie rice-growing region near
Stuttgart, Ark. Linquist et al. (2015) found that in Dewitt silt loam soils, although yields
were reduced less than 1% to 13%, the WUE was improved by 18% to 63% and AWD
(early season) followed by flooding practices (late season) reduced water use by 18%
while maintaining similar yields to that of flooded controls. Research on AWD practices
is lacking in other regions of the state and across varying soil types, more work is needed
in order to develop clear recommendations for AWD irrigation practices in the state of
Arkansas. In this study we compared the effects of three different AWD regimes and
252
a continuous flood management, on rice yields from a conventional, pure-line cultivar

(Roy J) and a hybrid cultivar (XL753) grown on Sharkey silty clay soils in the northeast
corner of the Mississippi delta rice-growing region in Arkansas.
PROCEDURE
This study was conducted at the University of Arkansas System Division of Agri-
cultures Northeast Research and Extension Center near Keiser, Ark., in 2014. The soil
type was a Sharkey silty clay with 3% sand, 33.1% silt, and 63.9 % clay (USDA-NRCS,
2013). Saturation, field capacity, and wilting point were calculated using Soil-Plant-
Atmosphere-Water (SPAW) softwares (USDA-ARS, Washington State University,
Pullman, Wash.) soil water characteristics (Saxton et al., 1986) using pesudotransfer
functions to determine saturation, field capacity and wilting points of 45.1%, 34.5%,
and 13% volumetric soil water content (VWC), respectively. Rice was drill-seeded at
a rate of 90 lb/acre for the conventional and 30 lb/acre for the hybrid on 12 June 2015
and plants emerged 19 June 2015. No irrigations were applied until the initial flood 20
July 2015, rainfall was sufficient for stand establishment. Plot sizes were 30 ft 52
ft (1560 sq ft), separated by dual packed levees to prevent water movement between
plots. Plots were planted half with a conventional, pure-line cultivar (Roy J) and half
with a hybrid (XL753) of which 260 sq ft of each cultivar in each plot was harvested
on 20 October 2015.
The study involved four water management treatments replicated four times in
a randomized complete block design. Treatments were: 1) flood (continuously flooded
control), 2) AWD/21.6% VWC, 3) AWD/25.4% VWC, and 4) AWD/30.2% VWC. The
AWD represents alternate wetting and drying followed by the volumetric soil water
content at which subsequent irrigations were triggered. The available water holding
capacity of this soil is 21.5% VWC (difference between field capacity and wilting point).
The actual deficits the trigger levels represent correspond to 20%, 42.3%, and 60%
managed allowable depletions (MAD). These deficits resulted in soil moisture trigger
points of 30.2% VWC, 25.4% VWC, and 21.6% VWC respectively.
All treatments were flooded to a 2- to 3-inch depth for 10 days (20 to 30 July) after
the preflood nitrogen (N; i.e., urea) fertilizer application of 120 lb N/acre (20 July). In
the flooded treatments, this flood depth was maintained throughout the growing season.
After the initial 10 day flood, the AWD treatments were allowed to dry until the soil
moisture reached the critical VWC triggers for each respective treatment (21.6%, 25.4%,
and 30.2% VWC at a soil depth of 2.5 inches) at which time the plots were re-flooded.
Critical VWC thresholds were determined using a Dynamax TH300 soil moisture
probe. Three measurements were collected from each replication in each treatment if
the overall average of all the reps in that treatment reached the threshold or lower; a
flood was applied to all plots of that treatment. Water inputs were also measured with
4-inch McCrometer propeller flowmeters in three out of the four replicates to determine
the average total water usage for each water management treatment. Weather data was
also obtained from the Northeast Research and Extension Center onsite weather sta-
253
tion. At harvest, grain was harvested, weighed, and moisture readings were obtained
for 260 sq ft from each cultivar within each plot using an Almaco SPC40 small plot
research combine with 5-foot header width. All yields in bushels per acre (bu/acre)
were corrected to 12% moisture.
Data Analysis
All data were analyzed using SYSTAT 13, the treatment and cultivar effects on
yield and water-use efficiency were evaluated with an analysis of variance. Normality
of all data was confirmed using a Kolmogorov-Smirnov normality test and both models
passed homogeneity of variances. Significant treatment effects were further analyzed
using a Tukey test method of mean comparison.
In order to compare the differences in yields and relative yields (% relative to the
flooded yield average of each respective cultivar) among treatments, an analysis of vari-
ance was used with a response variable, yield (bu/acre), and two factors, water treatment
(four factor levels: flood, AWD/21.6% VWC, AWD/25.4% VWC, and AWD/30.2%
VWC), cultivar (two factor levels: XL753 and Roy J), and a water treatment/cultivar
interaction term. Water-use efficiency, bushels per acre-inch of water applied (bu/acre-
inch), was calculated for all replicates in each treatment and each cultivar by dividing
yield per acre (bu/acre) by the average inches of water applied (acre-inch/acre) for each
respective treatment. In order to compare the differences in WUE between water treat-
ments and across cultivars, a balanced analysis of variances with a response variable of
WUE and two factors: water treatment (four factor levels: flood, AWD/21.6% VWC,
AWD/25.4% VWC, and AWD/30.2% VWC), cultivar (two factor levels, XL753 and
Roy J), and a water treatment/cultivar interaction term.

The yields regardless of treatment suffered greatly this season likely due to
several factors. First due to the prolonged rain, planting this year was delayed till mid-
June and also the plot combine is not a rice machine and the operator expressed that
a good percentage of grain was lost through the combine (estimated 20-25%). Onset
of irrigation and preflood N fertilizer applications was also delayed due to rain ren-
dering the plots inaccessible as well as warm minimum temperatures could also have
contributed to the lower yields this year (Fig. 1). It is likely that other factors (such as
possibility of drift) also played into the low yields obtained, however, we have little
to no evidence that can help us speculate on other possible factors contributing to the
low yields experienced across all water treatments including the conventional flooded
treatment. The highest deficit treatment of 60%, AWD/21.6%VWC was far too much
of a deficit for use in AWD studies or applications in Sharkey silt clay soils, experienc-
ing on average a 59.6% reduction in yield relative to the average of the conventional
flooded treatment, and the average 21.6% VWC treatment yield was 58.4% less than
the average flooded yield. This is similar to last years results (Gaspar et al., 2015), that
254
show the AWD 24%VWC treatment resulted in a 73.5% reduction in yield from the
conventional flooded treatment.
Yields
The interaction effect between water treatment and cultivar was not significant for
yield (P = 0.691) and relative yield (P = 0.504). This indicated that cultivar effects on
yield and water treatment effects on yield are consistent across all water treatments and
cultivars, respectively. Significant effects of water treatment (P < 0.001) and cultivar (P
< 0.001) on yield were observed. The mean comparison for water treatment indicated
that the flooded treatment and the AWD/30.2% VWC were significantly similar and
had the highest yields (Table 1). The flood treatment average yield (48.8 bu/acre) was
23.4%, 36.7%, and 58.4% greater than the average yield of the AWD/30.2% VWC,
AWD/25.4% VWC, and AWD/21.6% VWC treatments, respectively, independent of
cultivar. The mean comparison between cultivars indicated that XL753 yielded on
average 40.4% more yield than Roy J, irrespective of water treatments.
The relative yield analysis similarly shows that significant effects of water treat-
ment (P < 0.001) and cultivar (P = 0.030) on yield were observed. The AWD/30.2%
VWC, AWD/25.4% VWC, and AWD/21.6% VWC treatment replicates experienced
an average reduction in grain production of 24.5%, 38.9%, and 59.6% relative to the
flooded treatment average yield, respectively (Table 1).
Water Use Efficiency

The interaction effect between water treatment and cultivar was not significant
(P = 0.154). This indicated that cultivar effects on WUE and water treatment effects on
WUE are consistent across all water treatments and cultivars, respectively. Significant
effect of water treatment (P < 0.001) on WUE was also observed. The AWD/30.2%
VWC (1.77 bu/ac-in) and AWD/25.4% VWC (1.27 bu/ac-in) treatments had the highest
grain to water use ratio (Table 2). The data indicate that on average AWD/30.2% VWC
yielded 1.1 and 1.25 more bushels of grain/acre-inch of water applied, than AWD/21.6%
VWC and flood treatment, respectively. The cultivar difference in WUE was significant
(P < 0.001), indicating that XL753 on average yielded 0.63 bu of grain more per ac-in
of water used than Roy J across all irrigation treatments.
The deviation between cultivar WUE means can be explained from examining
the overall mean difference in in WUE between cultivars (Table 2) as well as the least
square mean for WUE and yield for each cultivar within each water treatment (data not
shown). Despite the fact that both cultivars were planted in each treatment replication
and they experienced the same amount of irrigation within each replication, Roy J had
consistent lower yields and WUE than XL753 which ultimately lowered the average
yield and WUE for each irrigation treatment considerably.
255
Observational Results
The average water used in each water treatment was greatest for the flood, fol-
lowed by AWD/21.6% VWC, AWD/25.4% VWC and AWD/30.2% VWC (Table 2). The
AWD/21.6% VWC reached trigger point once 54 days after termination of the initial
flood, AWD/25.4% VWC reached trigger point once 40 days after termination of the
initial flood, and the AWD/30.2% VWC trigger was met twice 32 days after termination
of the initial flood, then again 54 days later (Fig. 1). This year had substantial amounts
of rain totaling 14.47 inches during the growing season and 9.42 inches during the ir-
rigation period. The dates of the reflood for the treatments was a considerable length
of time and is likely due to the high amount of rain during and post initial flood (Fig.
1). Aside from the amount of rainfall this year, the water applied to all treatments was
extremely high (Table 2); due to the difficulties establishing the initial flood (Table 2)
and in pulling levees in this soil, the levee ditch depth ranged 8 to 9 inches, which could
have also contributed to the high water usage. It is also probable that seepage from the
levees can also explain the high water use, such was observed in the previous study in
2014 (Gaspar et al., 2015). Bouman and Tuong (2001) found that AWD methods may
lead to increased water use due to drying cycles leading to soil shrinkage and cracking.
Data like soil moistures and depth data across the levees after a flooding event would be
needed to determine if leakage across the levees was occurring. Similarly soil moisture
and depth measurement readings across the soil profile could indicate the amount of
deep percolation occurring in each plot.
The flood treatment yielded the most grain relative to the AWD treatments across
all cultivars. As the deficit increased so did yield reduction. Overall, XL753 yielded
significantly more grain than Roy J (Table 1). On average, WUE was greater for the
AWD/32% VWC/ 20% deficit treatment (Table 2) than all other treatments. Difference
in WUE averages between cultivars was significant, and suggests that XL753 on aver-
age had a 46% higher WUE than Roy J across all irrigation treatments. Although the
yields this year were extremely low, the yield reduction was expressed in all irrigation
treatments and the trends in the data are very similar to the trends observed in the 2014
season (Gaspar et al., 2015). In this study, AWD had considerable water savings and
thus additional research is needed to investigate the potential. Small plot research in
this soil type is problematic and further work may need larger plots so that the levee
seepage influence is reduced and results will be more relevant to what farmers may
experience. No significant difference was found in yield between the 20% and 42%
deficit thresholds, so more research is needed to better define allowable depletions
for re-flooding. More AWD research is needed to determine applicable thresholds for
AWD methods on a wide variety of soil types in order to establish useful guidelines
for farmers that wish to implement this water conservation practice.
256
ACKNOWLEDGMENTS
The authors wish to thank the Arkansas Rice Research and Promotion Board for
financial support, the staff at the Northeast Research and Extension Center in helping
to conduct this study, and support from the University of Arkansas System Division
of Agriculture.
LITERATURE CITED
ANRC. 2012. Arkansas Natural Resources Commission. Arkansas Groundwater
Protection and Management Report for 2011. Little Rock, Ark.
Bouman, B.A.M. and T.P. Tuong. 2001. Field water management to save water and
increase its productivity in irrigated lowland rice. Agricultural Water Manage-
ment. 49:11-30.
Bouman, B.A.M. 2009. How much water does rice use. Rice Today. 8:28-29.
Davies, W.J., J. Zhang, J. Yang, and I.C. Dodd. 2011. Novel crop science for water-
limited agriculture. J. Agric. Sci. 149:123-131.
FAO. 2013. Food and Agriculture Organization of the United Nations. Access date:
23 January 2015. Available at: http://www.fao.org/nr/water/issues/scarcity.html
Feng, J., C.Q. Chen, Y. Zhang, Z. Song, A. Deng, C. Zheng, and W. Zhang. 2013.
Impacts of cropping practices on yield-scaled greenhouse gas emissions from
rice fields in China: A meta-analysis. Agriculture Ecosystems and Environment.
164:220-228.
Gaspar, J.P., C.G. Henry, M.M. Anders, M. Duren, D. Hendrix, and A.P. Horton.
2015. Effects of water saving rice cultivation methods on yield, water use, and
water-use efficiency. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells
Arkansas Rice Research Studies 2014. University of Arkansas Agricultural Ex-
periment Station Research Series 626:236-246. Fayetteville.
Haddeland, I., J. Heinke, H. Bieman, S. Eisner, M. Flrke, N. Hanasaki, M.
Konzmann, F. Ludwig, Y. Masaki, J. Schewe, T. Stacke, Z.D. Tessler, Y. Wada,
and D. Wisser. 2014. Global water resources affected by human interventions
and climate change. Proc. Natl. Acad. Sciences of the United States of America,
111:3251-3256.
Linquist, B.A., M.M. Anders, M.A. Adviento-Borbe, R.L. Chaney, L.L. Nalley,
E.F.F. da Rosa, and C. van Kessel. 2015. Reducing greenhouse gas emissions,
water use, and grain arsenic levels in rice systems. Global change biology 21:407-
417.
Price, A.H., G.J. Norton, D.E. Salt, O. Ebenhoeh, A.A. Meharg, C. Meharg, M.
Rafiqul Islam, R. N. Sarma, T. Dasgupta, A.M. Ismail, K.L. McNally, H. Zhang,
I.C. Dodd, and W.J. Davies. 2013. Alternate wetting and drying irrigation for rice
in Bangladesh: Is it sustainable and has plant breeding something to offer? Food
and Energy Security. 2:120-129.
Ray, D.K., N.D. Mueller, P.C. West, and J.A. Foley. 2013. Yield trends are insuffi-
cient to double global crop production by 2050. PLoS ONE, 8:e66428.
257
Saxton, K.E., W.J. Rawls, J.S. Romberger, and R.I. Papendick. 1986. Estimating
gereralized soil water characteristics from texture. Trans. ASAE 50:1031-1035.
Schewe, J., J. Heinke, D. Gerten, I. Haddeland, N.W. Arnell, D.B. Clark, R. Dank-
ers, S. Eisner, B. M. Fekete, F.J. Coln-Gonzlez, S.N. Gosling, H. Kim, X. Liu,
Y. Masaki, F.T. Portmann, Y. Satoh, T. Stacke, Q. Tang, Y. Wada, D. Wisser, T.
Albrecht, K. Frieler, F. Piontek, L. Warszawski, and P. Kabat. 2014. Multimodel
assessment of water scarcity under climate change. Proc. Natl. Acad. Sciences of
the United States of America, 111:3245-3250.
Takahashi, Y., R. Minamikawa, K.H. Hattori, K. Kurishima, N. Kihou, and K. Yuita.
2004. Arsenic behavior in paddy fields during the cycle of flooded and non-flood-
ed periods. Environ. Sci. Technol. 38:1038-1044.
Talukder, A.S.M.H.M., C.A. Meisner, M.A.R. Sarkar, M.S. Islam, K.D. Sayre,
J.M. Duxbury, and J.G. Lauren. 2012. Effect of water management, arsenic and
phosphorus levels on rice in a high arsenic soilwater system: II. Arsenic uptake.
Ecotoxicol. Environ. Safety. 80:145-151.
USDA-NCRS. 2013. United States Department of Agriculture-Natural Resource
Conservation Service. Access date: January 2014. Accessible at: http://websoilsur-
vey.sc.egov.usda.gov/App/HomePage.htm.
Yan, X., K. Yagi, H. Akiyama, and H. Akimoto. 2005. Statistical analysis of the
major variables controlling methane emission from rice fields. Global Change
Biology. 11:1131-1141.
Zhang, H., Y. Xue, Z. Wang, J. Yang, and J. Zhang. 2009. Alternate wetting and mod-
erate soil drying improves root and shoot growth in rice. Crop Science. 49:2246-
2260.
258
Table 1. Yield differences between water treatment (P < 0.001)

and cultivar (P < 0.001) revealed by analysis of variance. Relative yield
differences between water treatment (P < 0.001) and cultivar (P = 0.030) revealed
by analysis of variance. Least square means for rice yields and relative yields for water
treatment and cultivar, with Tukey method for mean comparison of significant groupings.
Water treatment Average yield Relative yield
(bu/acre) (% of flooded yield)
Flood 48.8 a 100 a
20% Deficit/AWD/30.2% VWC 37.4 ab 75.5 ab
42% Deficit/AWD/25.4% VWC 30.9 bc 61.1 bc
60% Deficit/AWD/21.6% VWC 20.3 c 40.4 c
SEM 4.13 8.41
Cultivar
XL 753 43.1 a 79.0 a
RoyJ 25.7 b 59.6 b
SEM 2.92 5.95
Relative yield is actual yield divided by the average yields for the flooded treatment reps for
each respective cultivar 100.
Means within a column followed by different letters are significantly different at the
P = 0.05 level.
AWD = alternate wetting and drying; VWC = volumetric soil water content.
SEM = standard error of the mean.
Table 2. Summary of the water usage (applied) and number of irrigations

after the initial 10 day flood cycle. Water use efficiency (WUE) differences revealed
by analysis of variance for the factor level differences in WUE for water treatment
(P < 0.001) and cultivar (P < 0.001). Least square means for WUE in bushels/acre-inch,
for water treatment and cultivar, with Tukey method for mean comparison groupings.
No of Average Average Water
refloods post total water use use
Water treatment post initial flood water use post initial flood efficiency
---------(acre-in./acre)--------- (bu/acre-in.)
20% Deficit AWD/30.2% VWC 2 21.1 10.2 1.77 ab
42% Deficit AWD/25.4% VWC 1 24.3 3.9 1.27 a
60% Deficit AWD/21.6% VWC 1 30.3 4.3 0.67 b
Flood NA 94.1 68.2 0.52
SEM = 0.147
Cultivar
XL753 1.37 a
RoyJ 0.74 b
SEM = 0.104
AWD = alternate wetting and drying; VWC = volumetric soil water content.
Means within a column followed by different letters are significantly different at the P = 0.05 level.
SEM = standard error of the mean.
259
Fig. 1. Minimum and maximum temperature, rainfall,

and irrigation dates in the three irrigation treatments [21.6%, 25.4%, and
30.2% volumetric soil water content (VWC)]. AWD = alternate wetting and drying.
260
RICE CULTURE
Grain Yield Response of Four New Rice Cultivars to Seeding Rate

J.T. Hardke1, D.L. Frizzell2, E. Castaneda-Gonzalez2,
T.L. Clayton3, G.J. Lee1, and R.J. Norman4
ABSTRACT
The cultivar seeding rate studies determine the proper seeding rates for new rice (Oryza
sativa L.) cultivars over a range of production/growing conditions in Arkansas. The four
rice cultivars evaluated in 2015 were CL172, Diamond, LaKast, and Roy J. Each cultivar
was seeded at 20, 40, 60, 80, and 100 lb/acre. In accordance with current recommenda-
tions and predominate grower practice, all seed received insecticide and fungicide seed
treatments. Trials were seeded on research centers/stations at three locations in eastern
Arkansas. Stand density and grain yield results were consistent with current seeding
rate recommendations of 65 to 70 lb/acre (30 seed/ft2) under optimum conditions and
seeding dates on silt loam soils. Adverse conditions such as late seeding date or clay
soil types currently recommend a 20% seeding rate increase (~80 lb/acre; 36 seed/ft2)
compared to a loamy soil and optimum seeding date. Stand density and grain yield at
study locations with these conditions also agreed with current recommendations. Care
should be taken that without the use of an insecticide seed treatment, stand density and
grain yield may be reduced compared to results in this study. Reduced milling yields
were only consistently observed at the lowest (20 lb/acre) seeding rate.
INTRODUCTION
The cultivar seeding rate studies measure the grain yield performance of the
new rice (Oryza sativa L.) cultivars over a range of seeding rates on representative
silt loam and clay soils and determines the proper seeding rate to maximize yield on
these soils under climatic conditions that exist in Arkansas. Optimal stand density for
cultivars is considered to be 10 to 20 plants/ft2 (Wilson et al., 2013). The release of new
cultivars, combined with changes in production practices including the use of insecticide
1
2
Program Associate III and Program Associate I, respectively, Rice Research and Extension Center,
Stuttgart.
3
4
261
and fungicide seed treatments, requires the continued evaluation of seeding rates for
new cultivars to ensure recommendations maximize profit potential for rice growers.
The objective of this study was to determine the optimal seeding rate for four new rice
cultivars in environments and growing conditions common to Arkansas rice production.
PROCEDURES
The three locations for the 2015 cultivar seeding rate studies included the
University of Arkansas System Division of Agricultures Rice Research and Extension
Center (RREC) near Stuttgart, Ark., on a DeWitt silt loam; the Pine Tree Research Sta-
tion (PTRS) near Colt, Ark., on a Calloway silt loam; and the Northeast Research and
Extension Center (NEREC) near Keiser, Ark., on a Sharkey clay. Studies were seeded
at RREC, PTRS, and NEREC on 6 May, 5 June, and 4 May, respectively. All seed was
treated with CruiserMaxx Rice seed treatment. Clearfield CL172, Diamond, LaKast,
and Roy J were seeded at each location. Seeding rates evaluated for each cultivar were
20, 40, 60, 80, and 100 lb seed/acre. Actual seeds sown varied according to cultivar
with the 60 lb/acre seeding rate equivalent to 27 seed/ft2 for CL172, 28 seed/ft2 for
Diamond, 25 seed/ft2 for LaKast, and 27 seed/ft2 for Roy J. Plots were 9 rows (7-inch
spacing) wide and 15 ft in length. Cultural practices otherwise followed recommended
practices for maximum yield. The experimental design for all trials and cultivars was
a randomized complete block design with 6 replications.
Stand density was determined approximately 3 weeks after rice emergence by
counting the number of seedlings emerged in 10 row ft of a single row. Nitrogen (N)
was applied to studies at the 4- to 5-lf growth stage in a single preflood application
of 120 lb N/acre on silt loam soils and 160 lb N/acre on clay soils using urea as the N
source. The permanent flood was applied within 2 days of preflood N application and
remained flooded until rice reached maturity. At maturity, the center 5 rows of each
plot were harvested, the moisture content and weight of grain were determined, and a
subsample of harvested grain removed for milling yield determinations. Grain yields
were adjusted to 12% moisture and reported on a bushels/acre (bu/acre) basis. The
dried rice was milled to obtain percent head rice (%HR, whole kernels) and percent
total white rice (%TR) to provide a milling yield expressed as %HR - %TR. Data were
analyzed using analysis of variance, PROC GLM, SAS v. 9.4 (SAS Institute, Inc., Cary,
N.C.) with means separated using Fishers least significant difference test (P = 0.05).

Stand density increased as seeding rate increased at all locations (Table 1). A
seeding rate of 60 lb/acre reached optimal stand density (~15 plants/ft2) at RREC which
corresponds with current seeding rate recommendations when seeding in the optimal
window on a silt loam soil. At PTRS, which was also a silt loam soil but seeded later
than the optimal window, an 80 lb/acre seeding rate was needed to achieve optimal stand
density. Again, the results at PTRS agree with current recommendations to increase
262
seeding rate by 20% when seeding after 1 June. On the clay soil at NEREC, optimal
stand density was reached at 80 lb/acre, in agreement with current recommendations to
increase seeding rate by 20% on a clay soil. Stand densities greater than optimal have
the potential to increase the risk of loss rather than profitability. Lodging and disease
pressure typically increase when greater than optimal stand densities are reached. It
should be noted that the use of an insecticide seed treatment, as in this trial, has been
shown to increase stand density by over 10% and increase grain yield by an average of
8 bu/acre (Taillon et al., 2015). Lower stand densities and grain yields may be expected
when seeding without the use of insecticide seed treatments.
Seeding rates with above optimum stand densities did not result in greater grain
yields compared to the optimal stand density (Table 2). No interaction was observed
between cultivar and seeding rate. Grain yields for the seeding rate resulting in optimal
stand density were not statistically different than the highest yielding seeding rate for
any cultivar or location.
Milling yields were evaluated at the NEREC and PTRS locations (RREC samples
could not be evaluated; Table 3). At both locations, the lowest seeding rates resulted
in significantly lower head rice and total milled rice yields compared to seeding rates
that resulted in optimal stand density.
Comparison of grain yields by converting to percent of optimal yield at each
location is provided in Fig. 1. At NEREC, the 80 and 100 lb/acre seeding rates resulted
in 96% and 100% optimal grain yields, respectively. At PTRS, the 80 and 100 lb/acre
seeding rates produced 99% and 100% of optimal grain yields, respectively. At RREC,
the 40, 60, 80, and 100 lb seeding rates produced 96%, 98%, 99%, and 100% of optimal
grain yields, respectively. These were the only seeding rates at each location to achieve
greater than 95% of optimal grain yield with no adjustment in management practices.
When using lower seeding rates or when environmental or soil conditions result in less
than desired stand density, other inputs such as N can be managed to recover some lost
yield potential (Counce et al., 1992; Wells and Faw 1978).
The cultivar seeding rate studies in 2015 agree with previous research that
an optimum seeding rate for new rice cultivars is approximately 30 seed/ft2 (65 to
70 lb/acre). Lower seeding rates risk insufficient stand densities that will be unable
to maximize grain yield potential. Currently recommended seeding rate adjustments
based on soil type, seeding date, and environmental conditions are in agreement with
the findings of this study.
ACKNOWLEDGMENTS
The authors wish to thank all Arkansas rice growers for financial support through
the Rice Check-Off funds administered by the Arkansas Rice Research and Promotion
Board, the support from the University of Arkansas System Division of Agriculture,
and Horizon Ag.
263
LITERATURE CITED
Counce, P.A., B.R. Wells, and K.A. Gravois. 1992. Yield and harvest-index respons-
es to preflood nitrogen fertilization at low rice plant populations. J. Prod. Agric.
5:492-497.
Taillon, N.M., G.M. Lorenz, J. Black, W.A. Plummer, and H.M. Chaney. 2015.
Insecticide seed treatments in rice: Is there value to the grower? In: R.J. Norman
and K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies
626:137-144. Fayetteville, Ark.
Wells, B.R. and W.F. Faw. 1978. Short-statured rice response to seeding and N rates.
Agron. J. 70: 477-480.
Wilson, C.E. Jr., Y. Wamishe, G. Lorenz, and J. Hardke. 2013. Rice stand establish-
ment. In: J.T. Hardke (ed.). Arkansas Rice Production Handbook. University of
Arkansas Division of Agriculture Cooperative Extension Service MP192: 31-40.
Little Rock, Ark.
Table 1. Influence of seeding rate on

stand density at three locations during 2015.
Stand density
Seeding rate NEREC
PTRS RREC
(lb seed/acre) ---------------------------(bu/acre)--------------------------
20 5.1 e 5.6 e 6.1 e
40 10.3 d 10.2 d 11.6 d
60 14.2 c 13.4 c 17.4 c
80 18.1 b 18.5 b 22.2 b
100 22.3 a 23.7 a 26.1 a
LSD0.05 1.9 2.3 1.4

Averaged across CL172, Diamond, LaKast, and Roy J cultivars.

NEREC = Northeast Research and Extension Center, Keiser, Ark.; PTRS =
Pine Tree Research Station, near Colt, Ark.; and RREC = Rice Research and
Extension Center, Stuttgart, Ark.

Means within a column followed by the same letter are not significantly differ-
ent (P > 0.05).
264
Table 2. Influence of seeding rate on

stand density at three locations during 2015.
Stand density
Seeding rate NEREC
PTRS RREC
(lb seed/acre) ---------------------------(bu/acre)--------------------------
20 143.1 d 102.5 d 147.7 c
40 161.6 c 133.5 c 161.8 b
60 168.9 b 143.4 b 165.4 ab
80 177.2 a 151.7 a 165.9 ab
100 183.7 a 153.8 a 168.2 a
LSD0.05 7.1 5.9 5.5


NEREC = Northeast Research and Extension Center, Keiser, Ark.; PTRS =
Pine Tree Research Station, near Colt, Ark.; and RREC = Rice Research and
Extension Center, Stuttgart, Ark.

Means within a column followed by the same letter are not significantly differ-
ent (P > 0.05).
.
Table 3. Influence of seeding rate on rice

milling yield at three locations during 2015.
Milling yield
Seeding rate NEREC
PTRS RREC
(lb seed/acre) --------------------------------------(bu/acre)----------------------------------
20 63.9 b 69.7 b 58.3 b 68.3 c ----
40 64.7 a 70.2 a 58.6 b 68.9 ab ----
60 64.7 a 70.2 a 58.4 b 68.7 bc ----
80 64.9 a 70.2 a 59.5 ab 69.3 a ----
100 65.2 a 70.5 a 60.4 a 69.0 ab ----
LSD0.05 0.6 0.4 1.2 0.5 ----


NEREC = Northeast Research and Extension Center, Keiser, Ark.; PTRS = Pine Tree
Research Station, near Colt, Ark.; and RREC = Rice Research and Extension Center,
Stuttgart, Ark.

Means within a column followed by the same letter are not significantly different
(P > 0.05).
265
Fig. 1. Influence of seeding rate on rice grain yield at the Northeast Research and
Extension Center (NEREC), Pine Tree Research Station (PTRS), and Rice Research
and Extension Center (RREC) during 2015. Percent of optimal grain yield calculated
based on the highest grain yield at each location equivalent to 100% optimal grain yield.
266
RICE CULTURE
Arkansas Rice Performance Trials, 2013-2015

J.T. Hardke1, D.L. Frizzell2, E. Castaneda-Gonzalez2,
G.J. Lee1, K.A.K. Moldenhauer2, X. Sha2,
Y.A. Wamishe3, R.J. Norman4, M.M. Blocker2, J.A. Bulloch2,
B.A. Beaty2, R.S. Mazzanti1, R. Baker5, W. Kirkpatrick6, M. Duren7, and Y. Liyew8
ABSTRACT
The Arkansas Rice Performance Trials (ARPT) are conducted each year to evaluate
promising experimental lines from the Arkansas rice breeding program and commercially
available cultivars from public and private breeding programs. The ARPTs are seeded
on experiment stations and cooperating producers fields in a diverse range of environ-
ments, soil types, and agronomic and pest conditions. The ARPTs were conducted at
five locations during 2015. Averaged across locations, grain yields were highest during
2015 for the commercial cultivars XL753, XL760, CLXL729, CLXL745, and Diamond.
Cultivars with the highest overall milling yields during 2015 included: Mermentau,
Antonio, CL111, CL153, and CL163
INTRODUCTION
Cultivar selection is likely the most important management decision made each
year by rice producers. This choice is generally based upon past experience, seed
availability, agronomic traits, and yield potential. When choosing a rice cultivar, grain
yield, milling yield, lodging potential, maturity, disease susceptibility, seeding date,
field characteristics, the potential for quality reductions due to pecky rice, and market
strategy should all be considered. Data averaged over years and locations are more
reliable than a single year of data for evaluating rice performance for such important
1
Rice Extension Agronomist, Program Technician - Rice Agronomy, and Program Associate - Rice
Verification Coordinator, respectively, Department of Crop, Soil, and Environmental Science, Stuttgart.
2
Program Associate III, Program Associate I, Professor, Associate Professor, Program Technician I, Pro-
gram Associate I, and Program Associate I, respectively, Rice Research and Extension Center, Stuttgart.
3
4
5
Rice Verification Program Coordinator, Department of Crop, Soil, and Environmental Science, Piggott.
6
Former County Agent, Cooperative Extension Service, McGehee.
7
8
Research Program Technician, Pine Tree Research Station, near Colt.
267
factors as grain and milling yields, kernel size, maturity, lodging resistance, plant height,
and disease susceptibility.
The Arkansas Rice Performance Trials (ARPT) are conducted each year to com-
pare promising new experimental lines and newly released cultivars from the breeding
programs in Arkansas, Louisiana, Texas, Mississippi and Missouri with established
cultivars currently grown in Arkansas. Multiple locations each year allow for continued
reassessment of the performance and adaptability of advanced breeding lines and com-
mercially available cultivars to such factors as environmental conditions, soil properties,
and management practices.
PROCEDURES
The five locations for the 2015 ARPTs included the University of Arkansas
System Division of Aricultures Rice Research and Extension Center (RREC) near
Stuttgart, Ark.; the Pine Tree Research Station (PTRS) near Colt, Ark.; the Northeast
Research and Extension Center (NEREC) near Keiser, Ark.; the Trey Bowers farm in
Clay County (CLAY); and the Brandon Truax farm in Desha County (DESHA). Ninety
entries, including established cultivars and promising breeding lines, were grown across
a range of maturities.
The studies were seeded at RREC, PTRS, NEREC, CLAY, and DESHA on 22
April, 5 June, 4 May, 30 April, and 5 May, respectively. Pure-line cultivars (varieties)
were drill-seeded at a rate of 30 seed/ft2 in plots 8 rows (7-inch spacing) wide and 15
ft in length. Hybrid cultivars were drill-seeded into the same plot configuration using a
seeding rate of 13 seed/ft2. Cultural practices varied somewhat among the ARPT loca-
tions but overall were grown under conditions for high yield. Phosphorus and potassium
fertilizers were applied before seeding at the RREC and PTRS locations. Nitrogen was
applied to ARPT studies located on experiment stations at the 4- to 5-lf growth stage in
a single preflood application of 120 lb N/acre on silt loam soils and 150 lb N/acre on
clay soils using urea as the N source. The permanent flood was applied within 2 days
of preflood N application and maintained throughout the growing season. At maturity,
the center five rows of each plot were harvested, the moisture content and weight of the
grain were determined, and a subsample of harvested grain removed for grain quality
and milling determinations. Grain yields were adjusted to 12% moisture and reported
on a bushels/acre (bu/acre) basis. The dried rice was milled to obtain percent head rice
(%HR, whole kernels) and percent total white rice (%TR) to provide a milling yield
expressed as %HR - %TR. Each location of the study was arranged in a randomized
complete block design with four replications.

The 3-year average of agronomic traits, grain yields, and milling yields of selected
cultivars evaluated during 2013-2015 are listed in Table 1. The top yielding entries,
averaged across three study years, include: XL753, Diamond, Caffey, and Titan with
268
grain yields of 239, 210, 204, and 204 bu/acre, respectively. In regard to milling yield
(%HR - %TR), Antonio, CL151, Mermentau, CL111, CL163, and Roy J had the highest
overall average milling yields from 2013-2015.
Selected agronomic traits, grain yield, and milling yields from the 2015 ARPT
are shown in Table 2. Mean grain yield across all locations and cultivars was 167 bu/
acre. Cultivar XL753 was the only commercial cultivar to maintain a grain yield above
200 bu/acre at all locations, although XL760 did have a mean grain yield >200 bu/acre
over the five locations. Other notable cultivars with high mean grain yields over the
five locations in 2015 included CLXL729, CLXL745, Diamond, Caffey, and Jupiter.
Milling yield, averaged across locations and cultivars, was 59-69 (%HR - %TR) during
2015. The long-grain cultivars Mermentau, Antonio, CL111, CL153, and CL163 had
the highest milling yields of all commercial entries, averaging 63-69, 62-70, 62-70,
62-70, and 62-69, respectively, across all locations.
The most recent disease ratings for each cultivar are listed in Table 3. Ratings
for disease susceptibility should be evaluated critically to optimize cultivar selection.
These ratings should not be used as an absolute predictor of cultivar performance with
respect to a particular disease in all situations. Ratings are a general guide based on
expectations of cultivar reaction under conditions that strongly favor disease; however,
environment will modify the actual reaction in different fields.
Growers are encouraged to seed newly released cultivars on a small acreage to
evaluate performance under their specific management practices, soils, and environ-
ment. Growers are also encouraged to seed rice acreage in several cultivars to reduce
the risk of disease epidemics and environmental effects. Cultivars that have been tested
under Arkansas growing conditions are more likely to reduce potential risks associated
with crop failure.
Data from this study will assist rice producers in selecting cultivars suitable to the
wide range of growing conditions, yield goals, and disease pressure found throughout
Arkansas.
ACKNOWLEDGMENTS
The Arkansas Rice Performance Trials are supported through grower check-off
funds administered by the Arkansas Rice Research and Promotion Board and the Uni-
versity of Arkansas System Division of Agriculture. We wish to thank the following
people for their dedication to making the ARPT possible each year: Chuck Pipkins,
Tara Clayton, and Shawn Clark.
269
Table 1. Results of the Arkansas Rice Performance Trials

Straw Milled
Grain strength 50% Plant Test kernel Chalky
Cultivar lengtha ratingb headingc height weight weightd kernelsd
(days) (in.) (lb/bu) (mg) (%)
Antonio L 2.0 80 37 43.6 20.3 1.94
Caffey M 2.2 82 38 43.1 23.5 1.47
CL111 L 1.6 80 38 43.1 19.6 2.88
CL151 L 1.6 80 38 43.1 19.6 2.88
CL163 L 1.8 84 38 44.2 -- --
CL271 M 1.6 84 38 44.0 -- --
CLXL729 L 3.9 80 43 43.7 20.5 2.19
CLXL745 L 4.1 77 44 44.0 22.2 1.93
Diamond L 1.8 81 40 43.0 21.5 1.40
Jupiter M 2.9 83 37 42.2 21.0 2.38
LaKast L 2.5 80 41 43.4 21.6 1.04
Mermentau L 1.4 80 37 43.3 19.6 1.88
MM14 M -- 83 35 44.5 -- --
Roy J L 1.0 85 41 42.7 21.0 1.13
Taggart L 1.8 84 43 43.0 22.5 1.02
Titan M 2.3 77 38 43.0 22.8 2.47
Wells L 2.1 82 40 43.2 21.4 1.37
XL753 L 2.5 78 43 43.8 21.0 2.13

Mean 2.2 81 39.3 43.4 21.3 1.76
a
Grain length: L = long-grain; M = medium-grain.
b
Relative straw strength based on field tests using the scale: 1 = very strong straw, 5 = very
weak straw; based on percent lodging (2012-2014 data due to no lodging in 2015).
c
Number of days from plant emergence until 50% of the panicles are visibly emerging
from the boot.
d
Data from Riceland Grain Quality Lab, 2012-2014. Based on weight of 1000 kernels.
270
averaged across the three-year period of 2013-2015.
Milling yield by year Grain yield by year

2013 2014 2015 Mean 2013 2014 2015 Mean
-------- (% head rice - % total rice)------- ---------------------- (bu/acre)-------------------

65-70 66-72 62-70 64-71 191 174 141 169

58-67 57-69 56-68 57-68 217 216 179 204

65-70 65-71 61-70 64-70 189 202 166 186

65-70 65-71 61-70 64-70 189 202 166 186
-- 63-70 61-70 62-70 -- 186 151 168
-- 58-70 52-68 55-69 -- 190 166 178

62-69 61-70 59-69 61-69 205 202 194 200

61-69 61-71 58-69 60-70 179 203 187 189

62-68 61-69 60-69 61-69 226 218 186 210

61-66 59-68 61-68 60-67 200 213 176 196

63-70 62-71 56-68 60-70 203 202 162 189

65-69 66-71 63-69 65-70 190 181 161 177
-- 52-69 61-69 56-69 -- 196 155 176

63-70 62-70 61-70 62-70 210 207 169 195

62-69 60-70 58-70 60-70 205 200 167 191

58-67 55-69 56-68 56-68 212 235 165 204

62-70 57-70 57-70 59-70 196 192 161 183

60-70 57-71 54-69 57-70 245 259 212 239

60-71 63-69 62-70 62-70 203 203 169 190
271
Table 2. Results of the Arkansas Rice Performance Trials at five locations during 2015.
272
Grain yield by location and seeding date
Grain Straw 50% Plant Test Milling CLAY DESHA NEREC PTRS RREC
Cultivar lengtha strengthb headingc height weight yieldd 30 April 5 May 4 May 5 June 22 April Mean
(rating) (days) (in.) (lb/bu) (%HR-%TR) -------------------------------------(bu/acre)---------------------------------
Antonio L -- 75 37 46.9 62-70 162 139 166 122 118 141
Caffey M -- 79 37 46.0 57-68 194 139 206 176 182 179
CL111 L -- 74 39 47.6 62-70 160 141 147 139 133 144
CL151 L -- 74 38 46.5 61-70 182 164 188 161 136 166
CL153 L -- 77 37 46.8 62-70 166 144 185 150 124 154
CL163 L -- 79 37 46.3 62-69 133 149 180 148 145 151
CL172 L -- 77 35 46.1 61-70 156 130 165 128 133 142
CL271 M -- 79 38 45.9 58-69 177 138 198 177 140 166
CL272 M -- 77 37 46.5 52-68 171 132 206 158 141 162
CLXL729 L -- 74 45 47.7 59-69 194 165 207 211 195 194
CLXL745 L -- 72 45 47.6 58-69 194 167 185 189 198 187
Diamond L -- 77 39 46.1 60-69 189 169 232 171 172 186
Jupiter M -- 79 36 45.0 61-68 183 137 201 164 194 176
LaKast L -- 74 41 47.4 56-68 178 150 184 148 149 162
Mermentau L -- 76 37 46.5 63-69 174 167 187 140 137 161
MM14 M -- 77 33 46.5 61-69 173 135 169 134 166 155
Roy J L -- 81 39 45.4 61-70 182 149 195 170 146 169
Taggart L -- 80 41 46.0 59-70 182 157 194 159 143 167
Titan M -- 73 36 46.1 56-68 180 129 200 142 173 165
Wells L -- 78 39 46.5 57-70 179 149 182 165 131 161
XL753 L -- 73 46 47.7 54-69 219 202 218 209 210 212
XL760 L -- 77 45 46.7 59-69 211 187 226 210 204 207
MSX4077 L -- 77 39 46.1 60-69 189 169 232 171 172 186
Mean -- 77 39 46.5 59-69 177 151 191 160 157 167
a
b
Relative straw strength based on field tests using the scale: 1 = very strong straw, 5 = very weak straw; based on percent lodging
(no lodging in 2015).
c
Number of days from plant emergence until 50% of the panicles are visibly emerging from the boot.
d
% HR - % TR = percent head rice - percent total rice.
Table 3. Rice cultivar reactionsa to diseases (2015).
Bacterial Narrow Black
Sheath Straight- panicle brown Stem Kernel False sheath
Cultivar blight Blast head blight leaf spot rot smut smut Lodging rot
Antonio S S -- MS MS S S MS MS --
Caffey MS MR -- S R -- -- MS -- --
Cheniere S VS VS VS S S S S MR MS
CL111 VS MS S VS VS VS S S MS S
CL151 S VS VS VS S VS S S MS S
CL152 S VS S S MR -- VS S -- --
CL153 S MS -- -- -- -- S S -- --
CL163 S S -- MS -- -- MS -- -- --
CL172 MS MS -- MS -- -- MS S -- --
CL271 S MR -- MS MR -- MS -- -- S
CL272 S S -- VS -- -- MS -- -- --
CLXL729 MS R MS MR MS S MS S S S
CLXL745 S R R MR MS S MS S S S
CLXP756 MS -- -- -- -- -- -- S -- S
Cocodrie S S VS S S VS S S MR S
Della-2 S R -- S MS -- -- -- -- --
Francis MS VS MR VS S S VS S MS S
Jazzman MS S S S S S MS S MS MS
Jazzman-2 VS S -- VS MR -- S S -- --
Jupiter S S S MR MS VS MS MS MS MR
LaKast S S MS S MS S S S MS MS
Mermentau S S VS MS MS -- S S MS --
MM14 -- -- -- S -- -- -- S -- --
Rex S S S S MS S S S MR S
Roy J MS S S S MR S VS S MR MS
XL723 MS R S MR MS S MS S MS S
XL753 MS R MS MR -- -- MS S -- S
XL760 MS -- -- -- -- -- MS VS -- --
Taggart MS MS R MS MS S S S MS MS
continued
273
Table 3. Continued.
274
Sheath Straight- panicle brown Stem Kernel False sheath
Cultivar blight Blast head blight leaf spot rot smut smut Lodging rot
Titan S MS -- S -- -- S MS -- --
Wells S S S S S VS S S MS MS
a
Reaction: R = resistant; MR = moderately resistant; MS = moderately susceptible; S = susceptible; VS = very susceptible. Cells with no values
indicate no definitive Arkansas disease rating information is available at this time. Reactions were determined based on historical and recent
observations from test plots and in grower fields across Arkansas and other rice states in southern U.S.A. In general, these ratings represent
expected cultivar reactions to disease under conditions that most favor severe disease development. Table prepared by Y. Wamishe, Assistant
Professor/Extension Plant Pathologist
RICE CULTURE
Methane Emissions from Direct-Seeded, Delayed-Flood Rice

Production as Influenced by Cultivar and Water Management
J.J. Humphreys1, K.R. Brye1, A.D. Smartt1, J.T. Hardke2,
D.L. Frizzell3, E. Castaneda-Gonzalez3, G.J. Lee2, and R.J. Norman1
ABSTRACT
Methane (CH4) emissions from direct-seeded, delayed-flood rice (Oryza sativa L.)
production are a source of concern in the environmental and agricultural communities.
Addressing both communities needs requires the use of new cultivars and appropri-
ately timed production practices to achieve the desired goal of determining methods
to reduced CH4 emissions without decreasing yields or milling quality. The objective
of this study was to evaluate the effects of rice cultivar (i.e., conventional, pure-line
rice cultivar LaKast and hybrid rice cultivar RiceTec XL753) and water management
(i.e., full-season flood and midseason drain) on CH4 emissions from a silt-loam soil
at the University of Arkansas System Division of Agricultures Rice Research and
Extension Center, near Stuttgart, Ark. Vented, non-flow-through, non-steady-state
chambers were used to collect gas samples for measurement of CH4 fluxes. Methane
fluxes were calculated according to changes in concentration in the chamber headspace
over the 60-min sampling interval. The hybrid with the full-season flood produced the
numerically largest yield (249 bu/acre, 12,560 kg/ha) and the numerically largest CH4
emissions (79.1 kg CH4-C/ha/season). The numerically lowest CH4 emissions (28.9 kg
CH4-C/ha/season) came from the hybrid with the midseason drain, which had a yield
of 227 bu/acre (11,451 kg/ha). The proper combination of cultivar selection and water
management can help reduce CH4 emissions from rice production on silt-loam soils.
INTRODUCTION
Total United States greenhouse gas (GHG) emissions increased by 8.4% from
1990 to 2011, with a 1.6% decrease from 2010 to 2011, followed by a 2% increase in
2012 to a total 2013 U.S. GHG emissions of 6673 million metric tons of carbon diox-
1
Graduate Student, Professor, Program Associate, and Professor, respectively, Department of Crop, Soil,
and Environmental Science, Fayetteville.
2
3
Program Associate III and Program Technician I, respectively, Rice Research and Extension Center,
Stuttgart.
275
ide CO2 equivalent [United States Environmental Protection Agency (USEPA), 2014].
However, over the last 250 years, the concentration of methane (CH4) in the atmosphere
has increased by 158% (IPCC, 2007). The overall amount of CO2 emissions from 1990
to 2011 increased by 504 Tg of CO2 equivalent, while emissions of CH4 have decreased
by 57.2 Tg of CO2 equivalent (IPCC, 2014). Methane is a potent GHG that is produced
under anoxic conditions when organic matter is converted to CH4 by a class of micro-
organisms known as methanogens. Several biochemical processes exist where carbon
(C) is reduced to CH4, thus releasing energy for metabolic processes. As of 2011, CH4
emissions from rice (Oryza sativa L.) cultivation represented 1.1 % of overall U.S. CH4
production (IPCC, 2014). The main source of CH4 in the soil column is in the topsoil,
where >99% of the total soil-produced CH4 is emitted (Mitra et al., 2002).
Management practices under which rice is cultivated are one of the most important
factors affecting CH4 emissions. Hybrid cultivars have shown a decreased amount of
CH4 emissions compared to conventional cultivars (Rogers et al., 2014). Rice in the
U.S. is generally grown under continuously flooded conditions throughout the growing
season. Midseason drainage does not occur except by accident or when controlling for
straighthead, which is a disorder that causes sterility of the spikelets and reduces yield
(IPCC, 2014). To reduce CH4 emissions from flooded rice, field management practices
must be developed that will reduce CH4 emissions without decreasing yields or milling
quality (Lindau et al., 1993). The objective of this study was to evaluate the effects of
rice cultivar (i.e., conventional, pure-line rice cultivar LaKast and hybrid rice cultivar
RiceTec XL753) and water management (i.e., full-season flood and midseason drain)
on CH4 emissions from a silt-loam soil at the Rice Research and Extension Center,
near Stuttgart, Ark.
MATERIAL AND METHODS

Field research conducted in 2015 was similar to that conducted by Rogers et al.
(2014) at the University of Arkansas System Division of Agricultures Rice Research
and Extension Center near Stuttgart, Ark. (342754.5 N, 91258.6 W). The study
area consisted of a total of 16 field plots, 5-ft 5-in (1.6-m) wide 16-ft 5-in (5-m) long,
with 9 rice of rows and 7-in (18-cm) row spacing. Plots were arranged in 2 blocks with 8
plots per block. Rice was seeded on 6 May 2015. The conventional rice cultivar LaKast
and the hybrid rice cultivar RiceTec XL753 were seeded. The flood was established on
15 June 2015. Optimal N fertilization was used for both cultivars. The conventional
cultivar received 104 lb N/acre (117 kg/ha) 24 hr before the flood was established, and
45 lb N/acre (50 kg/ha) at 0.5-in. internode elongation. The hybrid cultivar XL753 re-
ceived 120 lb N/acre (134 kg/ha) preflood, and 30 lb N/acre (33 kg/ha) at the boot stage,
after planting, and before the flood was established. After planting and before flooding,
a boardwalk system was constructed throughout the plots to reduce disturbances to the
plants and allow easier access to the plots during the growing season.
Prior to flood establishment, two soil cores, 2-in. (4.8-cm) in diameter were col-
lected from the top 4-in. (10-cm) in each plot for a total of 32 cores collected from the
entire study area. All soil samples were dried at 158 F (70 C) for 72 h, crushed, and
sieved through a 2-mm metal mesh screen for soil property determinations. Sixteen soil
276
samples were used for particle-size analyses, where the average particle-size distribu-
tion in the top 10 cm was 21% sand, 72% silt, and 7% clay, which results in a silt-loam
soil texture (Table 1). The other 16 soil samples collected were analyzed by inductively
coupled, atomic emission spectrometry (Spectro Arcos, Spectro Analytical Instruments,
Kleve, Germany) using a 1:10 soil mass to solution extractant ratio as used by Tucker
(1992) for Mehlich-3 extractable nutrients (P, K, Ca, Mg, S, Na, Fe, Mn, Zn, Cu, and B;
Mehlich, 1984). Total carbon (TC) and total nitrogen (TN) concentrations were measured
by high-temperature combustion with a VarioMaxCN analyzer (Elementar Americas,
Inc., Mt. Laurel, N.J.). Soil organic matter (SOM) concentration was determined by
weight-loss-on-ignition after 2 h at 360 C. Soil pH and electrical conductivity (EC)
were analyzed potentiometrically in a 1:2 (m/v) soil/water suspension.
Schedule 40 polyvinyl chloride (PVC) was used in the construction of cylindri-
cal base collars that measured 12-in. (30-cm) in diameter 12-in. (30-cm) tall, and
were inserted to a depth of approximately 4-in. (10-cm). Vented, non-flow-through,
non-steady-state chambers (Livingston and Hutchinson, 1995) were used for the col-
lection of gas samples for the determination of CH4 fluxes. Seven days after the last gas
sampling, all aboveground dry matter was collected from the interior of the chamber.
Using a flame ionization detector (250 C) equipped with a gas chromatograph (Model
6890-N; Agilent Technologies, Santa Clara, Calif.) with a 0.53-mm-diameter by 30-m-
HP-Plot-Q capillary column (Agilent Technologies, Santa Clara, Calif.), gas samples
were analyzed for CH4 concentrations within 48 h of collection. Methane fluxes were
calculated according to changes in concentrations in the chamber headspace over a
60-min sampling interval following procedures outlined by (Rogers et al., 2013). To
determine the change in concentration over time, measured concentrations (mL L-1; y
axis) were regressed against time (min; x axis) of sample extraction (i.e., 0, 20, 40, and
60 min). Means of CH4 emissions, rice yields, and emissions per unit grain yield were
calculated and presented.

Initial Soil Properties
Initial soil properties were relatively uniform among plots throughout the study
area, as indicated by the relatively low standard errors associated with soil property
means reported in Table 1. Most initial soil properties were within recommended
ranges for optimal rice production on a silt loam soil (Table 1). Extractable soil K and
Zn concentrations were within recommended optimum levels at 167 and 7.7 ppm (mg/
kg), respectively, and extractable soil P was above optimum levels at 71 ppm (mg/kg)
(Table 1) for rice production on a silt loam soil according to the University of Arkansas
System Division of Agriculture Cooperative Extension Service (Norman et al., 2013).
Methane Fluxes and Emission, and Rice Yields

Methane fluxes from the hybrid XL753 with the midseason drain were compa-
rable to that from the full-season flood before the drain occurred and the flood was
277
reestablished, while fluxes from the drained area were reduced and remained lower than
that from the full-season flood after flood reestablishment (Fig. 1). Regardless of water
management, the conventional cultivar had a similar temporal trend of CH4 fluxes, but
overall had a numerically lower yield than the hybrid. Methane fluxes peaked just after
50% heading and exhibited a small increase just after the end-of-season drain, which
was consistent with previous reports (Rogers et al., 2014).
The numerically lowest CH4 emissions of 28.9 kg CH4-C/ha/season came from
the hybrid with the midseason drain, which had a yield of 227 bu/acre (11,451 kg/ha)
(Table 2). Methane fluxes for the conventional rice cultivar LaKast had a yield of 205
bu/acre (10,341 kg/ha) and emissions of 76.4 kg CH4-C/ha/season for the full-season
flood, and a yield of 198 bu/acre (9988 kg/ha) and emissions of 56.6 kg CH4-C/ha/
season for the midseason drain.
The numerically largest-yielding cultivar-water-management treatment combina-
tion was the hybrid with the full-season flood at 249 bu/acre (12,560 kg/ha), but this
treatment combination also had numerically the largest CH4 release at 79.1 kg CH4-C/
ha/season (Table 2). The hybrid with the midseason drain had numerically the greatest
C efficiency at 0.003 kg CH4-C/kg grain, while the lowest numerical efficiency was
from the conventional cultivar full-season flood treatment combination at 0.007 kg
CH4-C/kg grain. The hybrid midseason drain produced 65.7% less CH4 emissions, but
only a 9.8% reduction in grain yield compared to the hybrid full-season flood treatment
combination. The conventional cultivar midseason flood had 23.2% lower CH4 emis-
sions, but only 3.5% lower grain yield compared to the conventional cultivar full-season
flood treatment combination.
The midseason drain numerically decreased the amount of CH4 produced in both
conventional and hybrid rice cultivars with only slight yield reductions compared to the
full-season flood. Methane production and mitigation in direct-seeded, delayed-flood
rice production are important not only because of the climate-change impact, but also
as a way to potentially conserve organic matter in the soil if CH4 emissions are reduced.
This study showed that, numerically, midseason draining combined with a hybrid rice
cultivar can reduce CH4 emissions compared to a conventional cultivar grown with a
full-season flood. Continued investigation is needed to better understand the relation-
ships among rice cultivars (i.e., conventional vs hybrid), water management strategies,
and CH4 emissions from silt-loam soils.
ACKNOWLEDGMENTS
and the University of Arkansas System Division of Agriculture. I would also like to
thank the agronomic crews, and special thanks to Emmett C. Chuck Pipkins for his
hard work and help throughout the year.
278
LITERATURE CITED
IPCC. 2007. Intergovernmental Panel on Climate Change. Climate Change 2007:
Synthesis Report. Contribution of Working Groups I, II and III to the Fourth As-
sessment Report of the Intergovernmental Panel on Climate Change [Core Writing
Team, Pachauri, R.K and Reisinger, A. (eds.)]. IPCC, Geneva, Switzerland, p. 104.
IPCC. 2014. Intergovernmental Panel on Climate Change. United States Greenhouse
Gas Inventory Chapter 6. IPCC, Geneva, Switzerland. pp. 15-21.
Lindau, C.W., P.K. Bollich, R.D. DeLaune, A.R. Mosier, and K.E Bronson. 1993.
Methane mitigation in flooded Louisiana rice fields. Biol. Fertility Soils:174-178.
Livingston, G. and G. Hutchinson. 1995. Enclosure-based measurement of trace gas
exchange: Applications and sources of error. pp. 14-51. In: P.A. Matson, and R.C.
Harriss (eds.). Biogenic trace gases measuring emissions from soil and water.
Blackwell Science, Oxford [England]; Cambridge, Mass., USA.
Mehlich, A. 1984. Mehlich 3 soil test extractant: a modification of Mehlich 2 extract-
ant. Commun. Soil Sci. Plant Anal. 15:1409-1416.
Mitra, S., R. Wassman, M. Jain, and H. Patak. 2002. Properties of rice soils affecting
methane production protentials: 2. Differences in topsoil and subsoil. Nutr. Cycl.
Agro. 64:183-191.
Harke (ed.). Arkansas Rice Production Handbook Misc. Publ. 192. University of
Arkansas Division of Agriculture Cooperative Extension Service, Little Rock, Ark.
Rogers, C.W., K.R. Brye, R.J. Norman, E.E. Gbur, J.D. Mattice, T.B. Parkin, and
T.L. Roberts. 2013. Methane emissions from drill-seeded, delayed-flood rice pro-
duction on a silt-loam soil in arkansas. J. Environ. Qual. 42:1059-1069.
Rogers, C.W., K.R. Brye, A.D. Smartt, R.J. Norman, E.E. Gbur, and M.A. Evans-
White. 2014. Cultivar and previous crop effects on methane emissions from drill-
seeded, delayed-flood rice production on a silt-loam soil. Soil Sci. 179:28-36.
Tucker, M.R. 1992. Determination of phosphorous by Mehlich 3 extraction. In: S.
J. Donohue (ed.). Soil and media diagnostic procedures for the Southern Region
of the United States. Virginia Agriculture Experiment Station Service Bulletin,
Blacksburg, Va. pp. 6-8.
USEPA. 2014. United States Environmental Protection Agency. Inventory of U.S.
greenhouse gas emissions and sinks: 1990-2013. Accessed date: 11 Jan. 2016.
Available at: www.epa.gov/climatechange/emissions/usinventoryreport.html
279
Table 1. Mean soil physical and chemical

properties (n = 16) before flood establishment from the
top 10 cm of a Dewitt silt loam located at the Rice Research and
Extension Center near Stuttgart, Ark., during the 2015 growing season.
Soil property Mean
( Standard Error)
Sand (%) 21.1 (0.17)
Silt (%) 71.7 (0.21)
Clay (%) 7.2 (0.21)
pH 6.6 (0.05)
Electrical conductivity (mhos/cm) 229 (9.4)
Mehlich-3 Extractable Nutrients (ppm, mg/kg)
P 71 (2.5)
K 167 (5.5)
Ca 1159 (15.1)
Mg 116 (1.2)
S 10.5 (0.4)
Na 108 (4.1)
Fe 468 (7.4)
Mn 212 (4.7)
Zn 7.7 (0.9)
Cu 0.97 (0.04)
B 0.73 (0.01)
Total N (%) 0.09 (0.00)
Total C (%) 0.76 (0.01)
Soil organic matter (%) 1.53 (0.01)
Table 2. Summary of average methane (CH4) emissions, rice yield, and methane
emissions per unit grain yield by rice cultivar/water management treatment combination.
Cultivar/water
management Methane Rice Rice Emissions:yield
combination emission yield yield ratio
(kg CH4-C/ha/season) (bu/acre) (kg/ha) (kg CH4-C/kg grain)
LaKast/midseason drain 56.6 198 9988 0.006
XL753/midseason drain 28.9 227 11451 0.003
LaKast/full-season flood 76.4 205 10341 0.007
XL753/full-season flood 79.1 249 12560 0.006
280
Fig. 1. Growing-season methane (CH4) fluxes

or the treatment combinations of cultivar (conventional LaKast
or hybrid XL753) and water management (full-season flood or midseason drain).
The thick vertical lines indicate the timing of (1) flood release for the midseason drain,
(2) flood reestablishment after the midseason drain, and (3) flood release from
all plots prior to harvest. Standard error bars accompany treatment means (n = 4).
281
RICE CULTURE
Utilization of On-Farm Testing to Evaluate Rice Cultivars, 2015

G.J. Lee1, E. Castaneda-Gonzalez2, D.L. Frizzell2,
J.T. Hardke1, Y.A. Wamishe3, and R.J. Norman4
ABSTRACT
On-farm testing provides researchers with the best opportunity to evaluate the perfor-
mance of cultivars under diverse production conditions. The Producer Rice Evalua-
tion Program (PREP) utilizes studies in commercial rice fields throughout the state to
evaluate experimental lines and commercial cultivars for disease, lodging, grain yield
potential, and milling quality in diverse growing conditions, soil types and farming
practices. For producers, knowing the optimum cultivar for each field is their biggest
and most important tool. On-farm testing can indicate which cultivars are suited for a
particular growing situation. Field studies were located in Chicot, Craighead, Greene,
Lincoln, Lonoke, Prairie, and Woodruff counties during the 2015 growing season.
Nineteen cultivars were selected for evaluation in the on-farm tests. The average grain
yield across all locations was 185 bu/acre and the mean milling yield (%HR-%TR)
was 59-69. The cultivars with the highest grain yields averaged across locations were
XL753 and CLXL729 followed by XL760, CLXL745, and LaKast.
INTRODUCTION
One goal of the University of Arkansas System Division of Agriculture is to
offer a complete production package to producers when southern U.S. rice cultivars
are released, including grain and milling yield potential, disease reactions, fertilizer
recommendations, and DD50 Program thresholds. Many factors can influence grain
yield potential including: seeding date, soil fertility, water quality and management,
disease pressure, weather events, and cultural management practices.
Rice disease can be a major factor in the profitability of any rice field in Arkansas.
Host-plant resistance, optimum farming practices, and fungicides (when necessary based
1
Program Technician - Rice Agronomy and Rice Extension Agronomist, respectively, Department of
2
Program Technician I and Program Associate III, respectively, Rice Research and Extension Center,
Stuttgart.
3
4
282
on integrated pest management practices) are the best line of defense we have against
these profit robbing diseases. The use of resistant cultivars, combined with optimum
cultural practices, provide growers with the opportunity to maximize profit at the lowest
disease control expense by avoiding the use of costly fungicide applications.
New rice cultivars are developed and evaluated each year at the University of
Arkansas under controlled experiment station conditions. A large set of data on grain
yield, grain quality, plant growth habit, and major disease resistance is collected during
this process. Unfortunately, the dataset under these conditions is not complete for many
of the environments where rice is grown in Arkansas because potential problems may
not be evident in nurseries grown on experiment stations. With information obtained
from field research coupled with knowledge of a particular field history, growers can
select the cultivar that offers the highest yield potential for their particular situation. The
Producer Rice Evaluation Program (PREP) was designed to better address the many
risks faced by newly released cultivars across the rice-growing regions of Arkansas.
The on-farm evaluation of new and commercial cultivars provides better information
on disease development, lodging, grain yield potential, and milling yield under different
environmental conditions and crop management practices. These studies also provide a
hands-on educational opportunity for county agents, consultants, and producers.
The objectives of the PREP include: 1) to compare the yield potential of commer-
cially available cultivars and advanced experimental lines under commercial production
field conditions, 2) to monitor disease pressure in the different regions of Arkansas,
and 3) to evaluate the performance of rice cultivars under conditions not commonly
observed on experiment stations.
PROCEDURES
Field studies were located in Chicot, Craighead, Greene, Lincoln, Lonoke, Prai-
rie, and Woodruff counties during the 2015 growing season. Nineteen cultivars were
selected for evaluation in the on-farm tests. Non-Clearfield entries evaluated during
2015 included Diamond, Jupiter, LaKast, Mermentau, Roy J, Taggart, Titan, a Missis-
sippi State University experimental line (MSX4077), and the RiceTec hybrids XL753
and XL760. Clearfield entries included CL111, CL151, CL153, CL163, CL172, CL271,
CL272, and the RiceTec hybrids CLXL729 and CLXL745.
Plots were 8 rows (7-in. spacing) wide and 15-ft in length arranged in a random-
ized complete block design with four replications. Pure-line cultivars (varieties) were
seeded at a rate of approximately 30 seed/ft2 while hybrids were seeded at a rate of
approximately 14 seed/ft2. Trials were seeded on 8 April (Lincoln and Woodruff), 9
April (Prairie), 21 April (Lonoke), 27 April (Craighead), 30 April (Greene), and 5 May
(Chicot). Since these experiments contain both Clearfield and non-Clearfield entries,
all plots were managed as non-Clearfield cultivars.
Plots were managed by the grower with the rest of the field in regard to fertilization,
irrigation, and weed and insect control, but in most cases did not receive a fungicide
application. If a fungicide was applied, it was considered in the disease ratings. Plots
283
were inspected periodically and rated for disease. Percent lodging notes were taken im-
mediately prior to harvest. At maturity, the center five rows of each plot were harvested,
the moisture content and weight of the grain were determined, and a subsample of
harvested grain was removed for milling purposes. Grain yields were adjusted to 12%
moisture and reported on a bushels/acre (bu/acre) basis. The dried rice was milled to
obtain percent head rice (%HR, whole kernels) and percent total white rice (%TR) to
provide a milling yield expressed as %HR-%TR. Data were analyzed using analysis of
variance, PROC GLM, SAS v. 9.4 (SAS Institute, Inc., Cary, N.C.) with means separated
using Fishers least significant difference test (P = 0.05).

All cultivars were represented at all locations during the 2015 growing season; a
summary of the results by county and corresponding date of seeding is presented in Table
1. The average grain yield across all locations was 185 bu/acre. The cultivars with the
highest grain yields averaged across locations were XL753 and CLXL729 followed by
XL760, CLXL745, and LaKast. In the Chicot Co. trial grain yield averaged 185 bu/acre
and the highest-yielding entries were XL753, XL760, CLXL729, and CLXL745 (Table
2). The mean milling yield for Chicot Co. was 67-72. In the Craighead Co. trial, grain
yield for the location averaged 164 bu/acre with a mean milling yield of 66-71 (Table
3). The highest yielding cultivars in Craighead Co. were CLXL729, XL753, XL760,
and CLXL745. In Greene Co., XL753 and CLXL729 were the highest-yielding cultivars
and the location averaged 180 bu/acre with a mean milling yield of 67-71 (Table 4). The
Lincoln Co. trial was the highest-yielding location during 2015 with an average yield of
219 bu/acre and a mean milling yield of 61-66 (Table 5). The highest yielding cultivars
were CLXL729 and CLXL745 in Lincoln Co. In Lonoke Co., XL760 and XL753 were
the highest yielding cultivars (Table 6). The average grain yield in Lonoke Co. was
204 bu/acre with a mean milling yield of 57-67 (Table 6). Rice Tec XL753, Roy J, and
Diamond were the highest yielding cultivars in Prairie Co. The cultivars in Prairie Co
had an average yield of 184 bu/acre and a mean milling yield of 44-67 (Table 7). The
highest yielding cultivars in the Woodruff Co. trial were XL753, CLXL745, CL111,
and LaKast (Table 8). Woodruff Co. had the lowest overall yield of all the counties at
158 bu/acre with a mean milling yield of 50-68. Unfortunately, the Woodruff Co. loca-
tion had issues with hydrogen sulfide toxicity affecting plant development and grain
yield. Monitoring cultivar response to disease presence and the severity of reactions is
a significant part of this program. The observations obtained from these plots are often
the basis for disease ratings developed by University of Arkansas System Division of
Agriculture for use by growers (Table 9). This is particularly true for minor diseases
that may not be encountered frequently, such as narrow brown leaf spot, false smut,
and kernel smut.
Yield variability among the study sites represents differences in environments and
management practices, but also susceptibility to lodging and disease pressure present
at individual locations.
284
The 2015 Producer Rice Evaluation Program (PREP) provided additional data to
the rice breeding and disease resistance programs. The program also provided supple-
mental performance and disease reaction data on new cultivars that will be more widely
grown in Arkansas during 2016.
ACKNOWLEDGMENTS
The authors appreciate the cooperation of all participating rice producers and
thank all Arkansas rice growers for financial support through the Rice Check-Off funds
administered by the Arkansas Rice Research and Promotion Board and the support by
the University of Arkansas System Division of Agriculture. The authors especially
thank the following county agents who made this work possible: Dave Freeze, Brent
Griffin, Richard Klerk, Keith Perkins, Steven Stone, Brandon Thiesse, and Gus Wilson.
285
Table 1. Results of the Producer Rice Evaluation Program at seven locations during 2015.
286
Grain yield by location and planting date
Grain Test Milling Chicot Craighead Greene Lincoln Lonoke Prairie Woodruff
Cultivar lengtha Lodging Moisture weight yieldb 5/5 4/27 4/30 4/8 4/21 4/9 4/8 Mean

---------- (%)---------- (lb/bu) (%HR-%TR) ---------------------------------------------(bu/acre)---------------------------------------------
CL111 L 0.0 14.3 43.4 60-70 170 141 178 223 194 145 172 175
CL151 L 2.7 14.9 43.2 60-70 165 169 192 225 185 153 147 177
CL153 L 0.0 15.1 42.9 61-70 172 166 173 234 195 190 143 182
CL163 L 0.9 15.5 41.7 62-69 166 103 137 216 190 196 161 167
CL172 L 0.2 15.9 43.2 61-69 151 150 133 197 170 184 151 162
CL271 M 0.5 15.5 43.6 57-69 163 165 157 189 175 176 162 169
CL272 M 0.2 15.5 44.8 54-69 174 164 170 210 193 177 161 178
CLXL729 L 3.0 14.3 40.2 58-69 231 206 243 287 248 192 161 224
CLXL745 L 9.6 14.1 40.3 60-70 212 190 229 268 243 152 174 210
Diamond L 2.1 16.1 43.8 58-69 206 187 195 205 219 204 165 197
Jupiter M 0.0 18.0 46.9 61-68 152 141 142 196 181 175 139 161
LaKast L 0.2 14.5 43.8 57-69 208 182 193 230 218 193 169 199
Mermentau L 0.0 16.0 42.8 62-69 178 155 163 196 183 170 141 169
Roy J L 0.9 17.5 43.2 59-70 171 177 151 192 205 209 142 180
Taggart L 0.0 16.4 44.7 57-69 181 175 160 202 186 193 138 176
Titan M 0.7 16.2 46.4 52-68 157 149 197 222 199 191 167 183
XL753 L 3.8 14.0 41.4 57-70 245 198 244 257 250 227 193 231
XL760 L 2.1 15.9 40.4 59-69 239 197 202 252 261 202 154 215
MSX4077 L 0.5 15.2 44.7 60-69 170 95 164 180 188 176 154 161
Mean -- 1.4 15.5 43.2 59-69 185 164 180 220 204 184 158 185
LSD0.05c -- 3.3 0.4 0.7 1.9-0.5 11.8 12.9 14.8 10.9 12.4 22.3 NS 7.9
a
b
c
Table 2. Results of Chicot Co. Producer Rice Evaluation Program Trial during 2015.
Grain Test Grain Milling
Cultivar lengtha Lodging Moisture weight yield yieldb

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 14.8 42.3 170 67-72
CL151 L 0.0 15.9 42.0 165 67-72
CL153 L 0.0 17.2 43.6 172 66-71
CL163 L 0.0 20.2 43.1 166 68-73
CL172 L 0.0 18.5 44.3 151 67-72
CL271 M 0.0 17.3 44.6 163 67-72
CL272 M 0.0 16.2 45.5 174 67-72
CLXL729 L 15.0 15.8 39.5 231 69-73
CLXL745 L 50.0 16.2 39.9 212 68-73
Diamond L 0.0 20.4 44.7 206 68-72
Jupiter M 0.0 19.4 47.5 152 66-71
LaKast L 0.0 15.8 44.3 208 69-73
Mermentau L 0.0 18.0 42.9 178 67-72
Roy J L 0.0 22.0 43.6 171 68-72
Taggart L 0.0 18.9 45.9 181 67-71
Titan M 0.0 17.9 47.1 157 67-70
XL753 L 0.0 15.4 41.8 245 67-72
XL760 L 0.0 18.8 41.0 239 66-71
MSX4077 L 0.0 18.1 46.0 170 66-70
Mean -- 3.4 17.7 43.7 185 67-72
LSD0.05c -- 12.8 1.3 0.9 11.8 NS-1.5
a
b
c
287
Table 3. Results of Craighead Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 14.1 43.9 141 66-72
CL151 L 2.5 14.4 43.0 169 66-71
CL153 L 0.0 14.6 43.5 166 67-72
CL163 L 5.0 15.2 39.3 103 67-72
CL172 L 0.0 15.3 43.2 150 65-71
CL271 M 0.0 15.3 44.3 165 67-70
CL272 M 0.0 14.8 45.0 164 67-72
CLXL729 L 2.5 14.5 40.8 206 65-71
CLXL745 L 0.0 14.2 40.6 190 67-72
Diamond L 0.0 15.9 45.5 187 64-71
Jupiter M 0.0 18.0 47.7 141 61-68
LaKast L 0.0 14.9 44.3 182 68-72
Mermentau L 0.0 15.4 43.1 155 67-71
Roy J L 0.0 17.2 44.8 177 66-71
Taggart L 0.0 17.2 45.9 175 61-68
Titan M 0.0 15.2 46.2 149 66-70
XL753 L 0.0 13.4 41.2 198 68-72
XL760 L 2.5 16.8 40.0 197 68-73
MSX4077 L 0.0 15.6 44.2 95 67-73
Mean -- 0.7 15.3 43.5 164 66-71
LSD0.05c -- NS 1.0 1.4 12.9 1.0-0.7
a
b
c
288
Table 4. Results of Greene Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 14.0 42.1 178 67-71
CL151 L 16.3 14.1 42.6 192 67-72
CL153 L 0.0 15.1 41.7 173 67-71
CL163 L 0.0 13.5 40.4 137 64-69
CL172 L 0.0 18.4 43.0 133 64-70
CL271 M 0.0 14.9 42.7 157 70-72
CL272 M 0.0 15.7 43.0 170 68-71
CLXL729 L 0.0 13.3 39.1 243 66-71
CLXL745 L 13.3 12.7 39.5 229 68-73
Diamond L 0.0 16.6 44.3 195 66-71
Jupiter M 0.0 18.6 46.0 142 68-71
LaKast L 0.0 14.8 43.7 193 66-72
Mermentau L 0.0 15.7 42.7 163 67-71
Roy J L 0.0 20.1 42.6 151 62-70
Taggart L 0.0 17.7 44.5 160 66-72
Titan M 0.0 16.2 45.8 197 69-71
XL753 L 25.0 13.1 40.9 244 69-74
XL760 L 0.0 16.6 39.0 202 65-71
MSX4077 L 0.0 13.2 45.0 164 66-71
Mean -- 2.9 15.5 42.5 180 67-71
LSD0.05c -- NS 1.7 1.5 14.8 1.5-1.0
a
b
c
289
Table 5. Results of Lincoln Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 16.6 43.6 223 63-68
CL151 L 0.0 17.0 43.9 225 63-68
CL153 L 0.0 17.7 43.9 234 64-67
CL163 L 0.0 17.8 43.4 216 60-64
CL172 L 0.0 17.9 44.0 197 63-67
CL271 M 0.0 18.3 44.8 189 62-67
CL272 M 0.0 18.4 45.9 210 60-66
CLXL729 L 0.0 16.4 39.8 287 60-66
CLXL745 L 0.0 16.6 40.1 268 63-69
Diamond L 0.0 18.8 44.1 205 55-65
Jupiter M 0.0 21.7 49.1 196 61-64
LaKast L 0.0 15.8 43.5 230 53-65
Mermentau L 0.0 19.3 43.1 196 61-65
Roy J L 0.0 19.8 44.0 192 60-68
Taggart L 0.0 18.3 45.0 202 60-67
Titan M 0.0 19.4 48.1 222 62-66
XL753 L 0.0 16.9 41.1 257 59-67
XL760 L 0.0 18.7 39.9 252 58-66
MSX4077 L 0.0 18.9 46.9 180 64-67
Mean -- 0.0 18.1 43.9 220 61-66
LSD0.05c -- NA 1.1 0.8 10.9 1.5-0.8
a
b
c
290
Table 6. Results of Lonoke Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 14.6 44.6 194 61-69
CL151 L 0.0 15.5 44.1 185 64-68
CL153 L 0.0 14.5 43.5 195 65-69
CL163 L 0.0 15.4 43.0 190 63-67
CL172 L 0.0 15.5 44.2 170 64-69
CL271 M 0.0 16.3 44.1 175 51-68
CL272 M 0.0 16.4 45.5 193 43-67
CLXL729 L 0.0 13.7 40.6 248 52-65
CLXL745 L 0.0 13.4 40.8 243 55-68
Diamond L 0.0 14.3 44.2 219 60-68
Jupiter M 0.0 18.8 47.7 181 62-67
LaKast L 0.0 13.7 43.3 218 56-66
Mermentau L 0.0 15.6 43.6 183 63-68
Roy J L 0.0 16.4 43.7 205 65-70
Taggart L 0.0 15.5 45.4 186 61-68
Titan M 0.0 16.1 46.9 199 34-65
XL753 L 0.0 13.2 41.7 250 46-67
XL760 L 0.0 14.7 40.4 261 61-66
MSX4077 L 0.0 14.0 45.4 188 57-67
Mean -- 0.0 15.1 43.8 204 57-67
LSD0.05c -- NA 1.2 1.2 12.4 3.9-1.6
a
b
c
291
Table 7. Results of Prairie Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 13.5 42.1 145 46-67
CL151 L 0.0 14.3 43.3 153 42-68
CL153 L 0.0 13.7 40.0 190 46-67
CL163 L 0.0 13.2 40.7 196 55-67
CL172 L 0.0 13.3 40.1 184 53-68
CL271 M 0.0 13.8 42.0 176 38-67
CL272 M 0.0 13.9 44.3 177 34-67
CLXL729 L 0.0 12.8 38.1 192 44-66
CLXL745 L 0.0 13.1 39.1 152 45-68
Diamond L 0.0 13.7 41.7 204 44-67
Jupiter M 0.0 15.9 46.0 175 57-68
LaKast L 0.0 13.8 43.4 193 39-66
Mermentau L 0.0 14.3 39.9 170 57-68
Roy J L 0.0 13.9 41.0 209 46-68
Taggart L 0.0 13.9 41.8 193 40-67
Titan M 0.0 13.9 45.6 191 22-65
XL753 L 0.0 13.3 40.3 227 39-67
XL760 L 0.0 12.9 39.5 202 45-66
MSX4077 L 0.0 13.6 41.6 176 50-68
Mean -- 0.0 13.7 41.6 184 44-67
LSD0.05c -- NA 0.9 2.5 22.3 3.7-1.0
a
b
c
292
Table 8. Results of Woodruff Co. Producer Rice Evaluation Program Trial during 2015.

------------ (%)------------ (lb/bu) (bu/acre) (%HR-%TR)
CL111 L 0.0 12.8 45.3 172 52-69
CL151 L 0.0 13.2 43.9 147 53-69
CL153 L 0.0 13.0 44.3 143 53-70
CL163 L 1.3 13.0 42.0 161 56-69
CL172 L 1.3 12.8 44.0 151 50-68
CL271 M 3.8 12.7 43.1 162 43-67
CL272 M 1.3 12.8 44.3 161 42-68
CLXL729 L 3.8 13.2 43.4 161 50-68
CLXL745 L 3.8 12.8 41.8 174 53-69
Diamond L 15.0 13.2 42.6 165 49-67
Jupiter M 0.0 13.7 44.6 139 54-68
LaKast L 1.3 12.5 44.3 169 48-69
Mermentau L 0.0 13.9 44.5 141 52-68
Roy J L 6.3 12.9 42.8 152 49-67
Taggart L 0.0 13.3 44.6 138 44-67
Titan M 5.0 14.7 45.4 167 48-68
XL753 L 1.3 13.1 42.9 193 52-69
XL760 L 12.5 13.1 42.7 154 54-68
MSX4077 L 3.8 13.0 43.6 154 51-68
Mean -- 3.2 13.1 43.7 158 50-68
LSD0.05c -- NS NS NS NS NS-NS
a
b
c
293
Table 9. Rice cultivar reactionsa to diseases (2015).
294
Sheath Straight- panicle brown Stem Kernel False sheath Sheath
Cultivar blight Blast head blight leaf Spot rot smut smut Lodging rot spot
Caffey MS MR -- S R -- -- MS -- -- --
Cheniere S VS VS VS S S S S MR MS --
CL111 VS MS S VS VS VS S S MS S --
CL151 S VS VS VS S VS S S MS S --
CL152 S VS S S MR -- VS S -- -- --
CL163 S S -- MS -- -- MS -- -- -- --
CL172 MS MS -- MS -- -- MS S -- -- --
CL271 S MR -- MS MR -- MS -- -- S --
CL272 S S -- VS -- -- MS -- -- -- --
CLXL729 MS R MS MR MS S MS S S S --
CLXL745 S R R MR MS S MS S S S --
Cocodrie S S VS S S VS S S MR S --
Della-2 S R -- S MS -- -- -- -- -- --
Francis MS VS MR VS S S VS S MS S --
Jazzman-2 VS S -- VS MR -- S S -- -- --
Jupiter S S S MR MS VS MS MS S MR --
LaKast S S MS S MS S S S MS MS S
Mermentau S S VS MS MS -- S S MS -- --
Roy J MS S S S MR S VS S MR MS --
Taggart MS MS R MS MS S S S MS MS --
Titan S MS -- S -- -- S MS -- -- --
Wells S S S S S VS S S MS MS --
XL723 MS R S MR MS S MS S MS S --
XL753 MS R MS MR -- -- MS S MS S --
XL760 MS -- -- -- -- -- MS VS -- -- --
a
Reaction: R = resistant; MR = moderately resistant; MS = moderately susceptible; S = susceptible; VS = very susceptible (cells with no values
indicate no definitive Arkansas disease rating information is available at this time). Reactions were determined based on historical and recent
observations from test plots and in grower fields across Arkansas. In general, these ratings represent expected cultivar reactions to disease under
conditions that most favor severe disease development. Table prepared by Y. Wamishe, Assistant Professor/Extension Plant Pathologist
RICE CULTURE
Grain Yield Response of Six New

Rice Cultivars to Nitrogen Fertilization
R.J. Norman1, T.L Roberts1, J.T. Hardke2, N.A. Slaton1, K.A.K. Moldenhauer3,
X. Sha3, D.L. Frizzell3, M.W. Duren4, E. Castaneda-Gonzalez3, and G.J. Lee2
ABSTRACT
The cultivar nitrogen (N) fertilizer rate studies determine the proper N fertilizer rates
for the new rice (Oryza sativa L.) cultivars across the array of soil and climatic condi-
tions which exist in the Arkansas rice-growing region. The six rice cultivars studied in
2015 were: Diamond, LaKast, Titan, and Horizon Ags Clearfield CL163, CL172, and
CL271. Grain yields in 2015 at the University of Arkansas System Division of Agri-
cultures Rice Research and Extension Center (RREC), Stuttgart, Ark., were typical
of most years. In contrast, grain yields were lower at the Pine Tree Research Station
(PTRS) near Colt, Ark., due to late planting and a zinc deficiency that appeared to harm
some cultivars more than others, and at the Northeast Research and Extension Center
(NEREC), Keiser, Ark., where bird damage decreased yields and increased variability
of the data for most of the cultivars. This was the first year Diamond and Titan were in
the cultivar N rate study and thus there is not enough data to make a recommendation
at this time. The four years of results collected for LaKast and two years of results for
CL163, CL172, and CL271 indicate the cultivars should do well with minimal lodging
if 150 lb N/acre is applied in a two-way split of 105 lb N/acre at preflood and 45 lb N/
acre at midseason when grown on silt loam soils and 180 lb N/acre in a two-way split
of 135 lb N/acre at preflood and 45 lb N/acre at midseason when grown on clay soils.
INTRODUCTION
The cultivar N fertilizer rate studies measure the grain yield performance of
the new rice cultivars over a range of N fertilizer rates on representative clay and silt
loam soils and determines the proper N fertilizer rates to maximize yield on these soils
1
Professor, Assistant Professor, and Professor, respectively, Department of Crop, Soil, and Environmen-
tal Science, Fayetteville.
2
3
Professor, Associate Professor, Program Associate III, and Program Associate I, respectively, Rice
4
295
under the climatic conditions that exist in Arkansas. Promising new rice selections from
breeding programs in Arkansas, Louisiana, Mississippi, and Texas as well as those
from private industry are evaluated in this study. Six new rice cultivars were entered
and studied in 2015 at three locations as follows: Arkansas entered the short stature,
long-grains Diamond and LaKast, and the semidwarf, medium-grain Titan; and Horizon
AG entered the Clearfield short stature, long-grain cultivar CL163 (which has higher
amylose content for processing quality) in cooperation with Mississippi; the short stat-
ure, long-grain CL172 in cooperation with Arkansas; and the semidwarf, medium-grain
CL271 in cooperation with Louisiana. Clearfield rice cultivars are tolerant to the broad
spectrum herbicide imazethapyr (Newpath).
PROCEDURES
University of Arkansas System Division of Agriculture locations where the cultivar
N fertilizer rate studies were conducted and corresponding soil series are as follows:
Northeast Research and Extension Center (NEREC), Keiser, Ark., on a Sharkey clay
(Vertic Haplaquepts); Pine Tree Research Station (PTRS), near Colt, Ark., on a Calloway
silt loam (Glossaquic Fragiudalfs); and the Rice Research and Extension Center (RREC),
near Stuttgart, Ark., on a DeWitt silt loam (Typic Albaqualfs). The experimental design
utilized at all locations for each of the rice cultivars studied was a randomized complete
block with four replications. A single preflood N fertilizer application was utilized for
all cultivars and was applied as urea on to a dry soil surface at the 4- to 5-lf stage. The
preflood N rates were: 0, 60, 90, 120, 150, 180, and 210 lb N/acre. The studies on the
two silt loam soils at the PTRS and the RREC received the 0 to 180 lb N/acre fertilizer
rates and the studies on the clay soil at the NEREC received the 0 to 210 lb N/acre N
rates with the 60 lb N/acre rate omitted. Rice usually requires about 20 to 30 lb N/acre
more N fertilizer to maximize grain yield when grown on clay soils compared to the silt
loams. All of the rice cultivars were drill-seeded on the silt loams and clay soil at rates
of 73 and 91 lb/acre, respectively, in plots 9 rows wide (row spacing of 7 in.), 15 ft. in
length. Pertinent agronomic dates and practices at each location are shown in Table 1.
The studies were flooded at each location when the rice was at the 4- to 5-lf stage and
within 2 days of preflood N fertilization. The studies remained flooded until the rice
was mature. At maturity, the center 5 rows of each plot were harvested, the moisture
content and weight of the grain were determined, and yields were calculated as bu/acre
at 12% moisture. A bushel (bu) of rice weighs 45 pounds (lb). Statistical analyses were
conducted with SAS (SAS Institute Inc., Cary, N.C.) and mean separations were based
upon Fishers protected least significant difference test (P = 0.05) where appropriate.

A single, optimum preflood N application method was adopted in 2008 in all
cultivar N fertilizer rate studies due to the rising cost of N fertilizer and the prefer-
ence of the short stature and semidwarf rice plant types currently being grown. The
296
currently grown rice cultivars typically reach a maximum yield with less N when the
N is applied in a single preflood application compared to a two-way split application.
Usually the rice cultivars require 20 to 30 lb N/acre less when the N is applied in a
single preflood application compared to a two-split application where the second split
is applied between beginning internode elongation and 0.5-in. internode elongation.
Thus, if 150 lb N/acre is recommended for a two-way split application, then 120 to 130
lb N/acre is recommended for a single preflood N application. Conditions critical for
use of the single, optimum preflood N application method are: the field can be flooded
timely, the urea is treated with the urease inhibitor NBPT or ammonium sulfate used,
unless the field can be flooded in 2 days or less for silt loam soils and 7 days or less
for clay soils, and a 2- to 4-in. flood depth is maintained for at least 3 weeks following
flood establishment.
In most years, the silt loam soil at the RREC has the largest amount of plant-
available/readily available native N, followed by the silt loam soil at the PTRS and then
the clay soil at the NEREC. Thus, most rice cultivars typically require a lower N fertil-
izer rate to maximize grain yield at the RREC compared to at the PTRS or NEREC, and
usually a little less at the PTRS than at the NEREC. Pertinent agronomic information
such as planting, herbicide, fertilization, and flood dates are shown in Table 1. Grain
yields in the 2015 cultivar N rate studies at the RREC were typical of most years.
However at the PTRS, yields were lower due to late planting and a zinc deficiency that
appeared to harm some cultivars more than others; and yields were also lower at the
NEREC where there was bird damage; thus the variability of the data was increased for
most of the cultivars.
Diamond achieved a grain yield of 191 bu/acre on the clay soil at NEREC when
120 lb N/acre were applied preflood and maintained this grain yield when up to the
maximum N rate of 210 lb N/acre was applied with no lodging (Table 2). Diamond
did experience some bird damage at NEREC that probably decreased grain yields
and definitely caused variability in the grain yield data, but Diamond still was able to
achieve a maximum numerical yield of 197 bu/acre. The grain yield of Diamond did
not significantly increase on the silt loam soils at PTRS and RREC when more than
120 lb N/acre was applied preflood and was able to maintain this yield with no lodg-
ing when up to 180 lb/acre was applied. The maximum numerical grain yield of 187
bu/acre by Diamond at the PTRS is remarkable considering the late planting and the
zinc deficiency that limited the yields of some of the other cultivars in the study. Thus,
over the three locations Diamond achieved a maximum grain yield of around 190 bu/
acre with no lodging and appeared to have a stable yield over a wide range of N rates.
This was the first year Diamond was in the N-rate study and one to two more years of
research will be required before an N-rate recommendation can be made.
LaKast achieved a grain yield of 203 bu/acre when 210 lb N/acre was applied
preflood, but did not significantly increase in grain yield on the clay soil at NEREC
when more than 150 lb N/acre was applied preflood (Table 3). Like Diamond, LaKast
did experience some bird damage at NEREC that probably decreased grain yields and
definitely caused variability in the grain yield data. Somewhat similarly, previous studies
(Norman et al., 2013; 2014) of the effect of N rate on the grain yield of LaKast on the
297
clay soil at NEREC reported maximum yields of over 200 bu/acre when 120 to 180 lb
N/acre were applied preflood. The grain yield of LaKast did not significantly increase
on the silt loam soil at PTRS when more than 150 lb N/acre was applied preflood and
LaKast did not seem that harmed by the zinc deficiency that plagued some of the other
cultivars achieving a maximum numerical yield of 186 bu/acre. Maximum grain yields
of LaKast over the previous 3 years were achieved at the PTRS when 120 to 180 lb N/
acre were applied preflood (Norman et al., 2013; 2014; 2015). LaKast achieved a grain
yield of 189 bu/acre on the silt loam soil at the RREC when 120 lb N/acre was applied
preflood and did not significantly increase in yield when up to 180 lb N/acre (195 bu/
acre) was applied preflood. In the previous 3 years, LaKast had a maximum grain yield
of over 200 bu/acre at RREC when 120 lb N/acre was applied preflood; although in
all of those years, the grain yield never significantly increased when more than 90 lb
N/acre was applied at the RREC. LaKast experienced no lodging at any of the three
locations in 2015. After four years of study, it appears LaKast has a stable yield over
a wide range of N rates and should do well with minimal lodging if 150 lb N/acre is
applied in a two-way split of 105 lb N/acre at preflood and 45 lb N/acre at midseason
when grown on silt loam soils and 180 lb N/acre in a two-way split of 135 lb N/acre at
preflood and 45 lb N/acre at midseason when grown on clay soils.
Titan achieved a grain yield of 177 bu/acre when 180 lb N/acre was applied pre-
flood on the clay soil at the NEREC and did not significantly increase in yield when the
N rate was increased to 210 lb N/acre (Table 4). Bird damage was minimal on Titan at
the NEREC and probably had only a minor effect on grain yield and yield variability.
Similarly, the grain yield of Titan maximized when 180 lb N/acre was applied to the
silt loam soil at the PTRS. Titan displayed zinc deficiency at PTRS and was slow to
recover when zinc was applied. The zinc deficiency definitely had a negative impact
on the grain yield of Titan at the PTRS and the late planting probably aggravated the
problem. Titan achieved a grain yield of 200 bu/acre on the silt loam soil at the RREC
when 150 lb N/acre was applied preflood and did not significantly increase or decrease
in yield when up to 180 lb Nacre was applied. Lodging was not an issue for Titan at
any of the three locations in 2015. This was the first year Titan was in the cultivar
N rate study and one to two more years of research will be required before an N-rate
recommendation can be made.
The Clearfield cultivar CL163 did not significantly increase in grain yield when
more than 120 lb N/acre was applied preflood on the clay soil at the NEREC and the
grain and resulting yield did not appear to have been damaged by the birds (Table 5).
The zinc deficiency definitely and late planting probably negatively impacted the grain
yield of CL163 on the silt loam soil at the PTRS and caused the data to be more variable
and the yield to not significantly increase when greater than 90 lb N/acre was applied
preflood. In 2014, when the planting was timely and there was no zinc deficiency,
CL163 had a maximum grain yield of 197 bu/acre at the PTRS when 120 lb N/acre
was applied preflood (Norman et al., 2015). Clearfield CL163 achieved a maximum
grain yield on the silt loam soil at the RREC when 150 lb N/acre was applied preflood
which is higher than the 120 lb N/acre required in 2014 (Norman et al., 2015). After
two years of study, it appears CL163 should do well with minimal lodging if 150 lb
298
N/acre is applied in a two-way split of 105 lb N/acre at preflood and 45 lb N/acre at

midseason when grown on silt loam soils; and when grown on clay soils, the preflood
N rate should be increased by 30 lb N/acre.
The grain yield of CL172 did not significantly increase when more than 120 lb
N/acre was applied preflood on the clay soil at the NEREC and there was some bird
damage that negatively affected yield and increased yield data variability (Table 6). The
cultivar CL172 maintained a grain yield of about 165 bu/acre when the N rate increased
to as high as 210 lb N/acre with no lodging at the NEREC. The zinc deficiency at the
PTRS was particularly harmful to CL172 and caused it to reach a maximum yield of
only 146 bu/acre on the silt loam soil. Grain yields did not significantly increase at the
PTRS when more than 120 lb N/acre was applied. When grown on the silt loam soil
at the PTRS in 2014, CL172 obtained a grain yield of 192 bu/acre when 150 lb N/acre
was applied preflood (Norman et al., 2015). The yield of CL172 did not significantly
increase at the RREC when more than 150 lb N/acre was applied preflood in 2015 and
had a maximum numerical yield of 169 bu/acre. In 2014, CL172 obtained a maximum
yield of 194 bu/acre when 120 lb N/acre was applied preflood on the silt loam soil at
RREC (Norman et al., 2015). After two years of study, CL172 appears to have a stable
yield over a wide range of N rates once the N rate to achieve maximum yield is ap-
proached and exceeded. Cultivar CL172 should do well with minimal lodging if 150
lb N/acre is applied in a two-way split of 105 lb N/acre at preflood and 45 lb N/acre
at midseason when grown on silt loam soils and a two-way split of 135 lb N/acre at
preflood and 45 lb N/acre at midseason when grown on clay soils.
The medium grain CL271 achieved a grain yield of 183 bu/acre when 180 lb N/
acre was applied preflood on the clay soil at the NEREC and maintained this grain yield
when the N rate increased to 210 lb N/acre (Table 7). The cultivar CL271 experienced
minimal bird damage at the NEREC. The grain yield of CL271 did not significantly
increase above 164 bu/acre on the silt loam soil at the PTRS when more than 120 lb N/
acre was applied preflood. The zinc deficiency coupled with the late planting negatively
affected the yield and the yield variability of CL271 at the PTRS in 2015. For instance
in 2014 at the PTRS, CL271 obtained a grain yield of 203 bu/acre when 120 lb N/acre
was applied preflood and did not significantly increase or decrease in yield (209 bu/
acre) when up to 180 lb N/acre was applied preflood (Norman et al., 2015). On the silt
loam soil at the RREC, CL271 did not significantly increase in yield when more than
150 lb N/acre was applied preflood and reached a maximum yield of only 162 bu/acre.
This is somewhat in contrast to the results in 2014 when CL271 obtained a maximum
yield of 196 bu/acre at the RREC when 150 lb N/acre was applied preflood; although it
did not significantly increase in yield above the 189 bu/acre obtained when 90 lb N/acre
was applied preflood (Norman et al., 2015). After two years of study, CL271 appears to
have a stable yield over a wide range of N rates once the N rate to achieve maximum
yield is approached and exceeded. This cultivar should do well with minimal lodging
if 150 lb N/acre is applied in a two-way split of 105 lb N/acre at preflood and 45 lb N/
acre at midseason when grown on silt loam soils and a two-way split of 135 lb N/acre
at preflood and 45 lb N/acre at midseason when grown on clay soils.
The Wells rice cultivar was included in the study as a control and to give a frame
of reference for comparing the grain yield performance and lodging percentage of the
299
new cultivars over the N fertilizer rates applied at the three locations (Table 8). The
N-rate recommendation for Wells is 150 lb N/acre applied in a two-way split of 105 lb
N/acre at preflood and 45 lb N/acre at midseason when grown on silt loam soils and a
two-way split of 135 lb N/acre at preflood and 45 lb N/acre at midseason when grown
on clay soils.
The cultivar N fertilizer rate study examines the grain yield performance of a
new rice cultivar across a range of N fertilizer rates on representative soils and under cli-
matic conditions that exist in the Arkansas rice-growing region. Thus, this study enables
the estimation of the proper N fertilizer rate for a cultivar to achieve maximum grain
yield when grown commercially in the Arkansas rice-growing region. The six cultivars
studied in 2015 were: Diamond, LaKast, Titan, CL163, CL172, and CL271. The data
generated from multiple years of testing of each cultivar will be used to determine the
proper N fertilizer rate to achieve maximum yield when grown commercially on most
silt loam and clay soils in Arkansas.
ACKNOWLEDGMENTS
This research was supported by the Arkansas Rice Research and Promotion Board,
the University of Arkansas System Division of Agriculture, and Horizon AG.
LITERATURE CITED
Norman, R.J., T.L. Roberts, C.E. Wilson Jr., N.A. Slaton, K.A.K. Moldenhauer, D.L.
Frizzell, J.D. Branson, and M.W. Duren. 2013. Grain yield response of ten new
rice cultivars to nitrogen fertilization. In: R.J. Norman and K.A.K. Moldenhauer
Norman, R.J., T.L. Roberts, C.E. Wilson Jr., N.A. Slaton, K.A.K. Moldenhauer, D.L.
Frizzell, J.D. Branson, and M.W. Duren. 2014. Grain yield response of eight new
rice cultivars to nitrogen fertilization. In: R.J. Norman and K.A.K. Moldenhauer
(eds.). B.R. Wells Arkansas Rice Research Studies 2013. University of Arkansas
Agricultural Experiment Station Research Series 617:293-302. Fayetteville, Ark.
Norman, R.J., T.L Roberts, J.T. Hardke, N.A. Slaton, K.A.K. Moldenhauer, D.L.
Frizzell, M.W. Duren, E. Castaneda-Gonzalez, and A.M. Fulford. 2015. Grain
yield response of five new rice varieties to nitrogen fertilization. In: R.J. Nor-
man and K.A.K. Moldenhauer (eds.). B.R. Wells Arkansas Rice Research Studies
626: 256-271. Fayetteville, Ark.
300
Table 1. Pertinent agronomic information for the Northeast Research and Extension Center (NEREC), Keiser, Ark.; the Pine
Tree Research Station (PTRS), near Colt, Ark.; and the Rice Research and Extension Center (RREC), Stuttgart, Ark., during 2015.
Practices NEREC PTRS RREC
Preplant fertilizers ---- ---- 90 lb P2O5/acre,
90 lb K2O/acre +
10 lb Zn/acre
Planting date 4 May 5 June 1 May
Emergence date 20 May 11 June 10 May
Herbicide spray date 6 May 5 May 1 May
and procedures 1.3 pt Command/acre + 1.0 pt/acre Command + 20 oz Obey/acre
40 oz Facet L/acre + 0.75 oz/acre Permit Plus
0.75 oz/acre Permit Plus
Herbicide spray date 11 June 16 June 1 June
and procedures 4 qt /acre Stam + 3 qt/acre Riceshot + 2 qt/acre Prowl +
1.0 pt/acre Grandstand 32 oz/acre Facet L 0.75 oz/acre Permit Plus
Herbicide spray date ----- 23 June ----
and procedures 32 oz/acre Facet L +
1.0 pt/acre Bolero
Preflood N date 18 June 24 June 3 June
Flood date 19 June 25 June 4 June

Drain date 9 September 1 October 28 August
Harvest date 23 September 14 October 2 September
301
Table 2. Influence of nitrogen (N) fertilizer rate on the

grain yield of Diamond rice at three locations during 2015.
Grain yield
N fertilizer rate NEREC a
PTRS RREC
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 62 70 85
60 ---- 127 148
90 148 153 169
120 191 177 183
150 191 187 192
180 197 180 194
210 189 ---- ----
LSD0.05b 24.4 12.0 11.6
a
NEREC = Northeast Research and Extension Center, Keiser, Ark.;
PTRS = Pine Tree Research Station, near Colt, Ark.; and RREC = Rice
Research and Extension Center, Stuttgart, Ark.
b

grain yield of LaKast rice at three locations during 2015.
Grain yield
N fertilizer rate NERECa PTRS RREC
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 76 81 89
60 ---- 133 144
90 162 149 173
120 178 168 189
150 192 176 188
180 194 186 195
210 203 ---- ----
LSD0.05b 21.1 10.6 15.3
a
b
302

grain yield of Titan rice at three locations during 2015.
Grain yield
PTRS RREC
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 74 56 93
60 ---- 92 150
90 145 108 163
120 158 127 180
150 164 137 200
180 177 150 191
210 179 ---- ----
LSD0.05b 10.0 10.4 10.5
a
b
Table 5. Influence of nitrogen (N) fertilizer rate on the grain

yield of Clearfield CL163 rice at three locations during 2015.
Grain yield
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 67 77 81
60 ---- 127 135
90 152 152 152
120 153 158 155
150 161 143 171
180 165 150 161
210 161 ---- ----
LSD0.05b 12.7 14.6 7.3
a
b
303

Grain yield
PTRS RREC
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 64 60 67
60 ---- 104 123
90 135 126 146
120 154 141 154
150 164 146 166
180 165 146 169
210 167 ---- ----
LSD0.05b 16.2 7.3 14.5
a
b

Grain yield
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 77 66 67
60 ---- 125 110
90 155 144 123
120 167 164 149
150 174 170 158
180 183 175 162
210 183 ---- ----
LSD0.05b 8.3 14.8 9.8
a
b
304

grain yield of Wells rice at three locations during 2015.
Grain yield
PTRS RREC
(lb N/acre) -------------------------- (bu/acre)-------------------------
0 63 55 57
60 ---- 102 106
90 134 126 127
120 162 148 149
150 181 162 165
180 191 173 162
210 192 ---- ----
LSD0.05b 11.0 7.9 9.5
a
b
305
RICE CULTURE
Response of Two Rice Cultivars to Midseason Nitrogen

Fertilizer Application Timing in the Presence of a Zinc Deficiency
R.J. Norman1, J.T. Hardke2, T.L. Roberts1, N.A. Slaton1,
D.L. Frizzell3, A.D. Smartt1, E. Castaneda-Gonzalez3, and G.J. Lee2
ABSTRACT
A midseason nitrogen (N) application timing rice (Oryza sativa L.) study was being
conducted at the University of Arkansas System Division of Agricultures Pine Tree
Research Station (PTRS) in 2015 when a zinc (Zn) deficiency was noticed on 8 July;
around beginning internode elongation (BIE) for Mermentau and about a week before
BIE for Roy J. The grain yield results clearly reflected the more severe damage the Zn
deficiency inflicted on the semidwarf Mermentau compared to the standard stature Roy
J that appeared to have yields typical or greater for a June seeding date. One reason
may be the stage of growth when the Zn deficiency was first observed. Mermentau
was already at BIE and the end of vegetative growth whereas Roy J was about a week
away from BIE. Another reason may be Mermentau is a semidwarf and therefore more
susceptible to Zn deficiency compared to a short stature cultivar like Roy J. Visual
observation indicated that Mermentau had greater stunting, greater loss of stand, and
less tillering than Roy J. The yield results suggested that when a Zn deficiency occurs
in a semidwarf like Mermentau a two-way split may result in a greater grain yield
compared to a single preflood application and it might be better to delay the midseason
application until BIE + 14 to 21 days.
INTRODUCTION
Zinc deficiency typically occurs in rice during early vegetative growth and
shortly after application of the permanent flood. It is most often observed when rice is
grown on silt and sandy loam soils with a pH above 7. Causes of Zn deficiency in rice
in Arkansas are typically from the use of irrigation water high in CaHCO3, excessive
1
Professor, Assistant Professor, Professor, and Program Associate, respectively, Department of Crop,
2
3
Stuttgart.
306
application of lime and/or non-uniform lime distribution. Visual symptoms associated

with Zn deficiency during the early vegetative growth stage of rice are: i) basal chlo-
rosis of newest/youngest leaves; ii) midrib of lower/oldest leaves becomes yellow to
white; iii) loss of leaf turgidity causing floating leaves; iv) bronzing of older leaves;
v) inhibition of tillering; vi) eventual stand loss under flooded conditions; vii) stacked
leaf collars; and vii) delayed maturity (Norman et al., 2003).
A midseason N application timing study was being conducted at the University
of Arkansas System Division of Agricultures Pine Tree Research Station (PTRS),
near Colt, Ark., in 2015 when a Zn deficiency was noticed on 8 July: around beginning
internode elongation (BIE) for Mermentau and about a week before BIE for Roy J.
Typical Zn deficiency symptomology was observed such as bronzing, stunting, loss of
some stand, and atypical tillering; especially for the semidwarf Mermentau and not so
much for the short stature Roy J. It was decided the severity of the Zn deficiency war-
ranted the application of Zn, but not the removal of the flood. We thought this would be
a good opportunity to measure the influence of a Zn deficiency on the grain yield of a
semidwarf and a short stature rice cultivar when N was applied in a two-way split with
different midseason N application timings and a single preflood application.
PROCEDURES
The study was conducted in 2013 at the PTRS, near Colt, Ark., on a Calhoun
silt loam that had a pH of 7.8 and a Zn concentration of 2.1 parts per million, which
is considered low. The two conventional rice cultivars in the study were the Louisiana
semidwarf, long-grain Mermentau and the Arkansas short stature, long-grain Roy J.
Two preflood N rates of 85 and 105 lb N/acre were utilized along with four midseason
N application timings. The midseason N rate of 45 lb N/acre was applied at BIE, BIE+7
days, BIE+14 days, or BIE+21 days. There was a check or no midseason N application
and a single preflood N application of 130 lb N/acre. The preflood N was applied onto dry
soil just prior to flooding and the midseason N was applied directly into the floodwater.
The permanent flood was established the day after the preflood N was applied when
the rice was at the 4- to 6-lf stage and the flood maintained until the rice was mature.
The rice was drill-seeded at a rate of 73 lb/acre in plots 9 rows wide (row spac-
ing of 7 in.), 15 ft in length. The rice was seeded at the PTRS on 5 June, emerged 11
June, the preflood N applied 24 June, and the BIE application was applied on 15 July
for Mermentau and 21 July for Roy J. Zinc EDTA at 1 lb Zn/acre was applied on 10
July. At maturity, the center 5 rows of each plot were harvested, the moisture content
and weight of the grain were determined, and yields were calculated as bu/acre at 12%
moisture. A bushel (bu) of rice weighs 45 pounds (lb).
The treatments were arranged as a randomized complete block, 2 (cultivar) 2
(preflood N rate) 4 (midseason N application time), factorial design with four rep-
lications, a no midseason N application (control) with four replications, and a single
preflood N application with four replications was included. Analysis of variance was
performed on the grain yield data utilizing SAS v. 9.1 (SAS Institute, Inc., Cary, N.C.).
307
Differences among means were compared using Fishers protected least significant
difference (LSD) procedure at a P = 0.05 probability level.

Analysis of variance P values for the studies indicated there was no significant
(P = 0.05) three-way interactions of cultivar preflood N rate midseason N timing
(P = 0.9467) on rice grain yield (Table 1). However, there were significant two-way
interactions on rice grain yield of cultivar preflood N rate (P < 0.0274) and cultivar
midseason N timing (P = 0.0480), but not preflood N rate midseason N timing
(P = 0.5354).
Roy J had a substantially greater response to N fertilizer application than did
Mermentau (Table 2). Roy J obtained substantially greater grain yields than Mermentau
when 85 or 105 lb N/acre was applied preflood and 45 lb N/acre was applied at mid-
season as well as when 130 lb N/acre was applied in a single preflood N application.
Mermentau had a greater yield when the N fertilizer was applied in a two-way split
application regardless of the preflood N rate compared to a single preflood applica-
tion. Somewhat similarly, Roy J had a greater yield when the N fertilizer was applied
in a two-way split application of 105 lb N/acre preflood and 45 lb N/acre at midseason
compared to when the N was applied in a single preflood application of 130 lb N/acre
or when the N was applied in a two-way split of 85 lb N/acre preflood and 45 lb N/acre
at midseason with the latter two being similar. The grain yield results clearly reflected
the more severe damage the Zn deficiency inflicted on Mermentau compared to Roy
J. One reason may be the stage of growth when the Zn deficiency was first observed
and corrected with an application of Zn. Mermentau was already at BIE and the end
of vegetative growth whereas Roy J was about a week away from BIE and still had
a week of vegetative growth to recover. Another reason may be that Mermentau is a
semidwarf and more susceptible to Zn deficiency compared to a short stature cultivar
like Roy J. Visual observation indicated that Mermentau had greater stunting, greater
loss of stand, and less tillering than Roy J.
Mermentau did not significantly increase in grain yield from the midseason N
application of 45 lb N/acre until the midseason N was applied at BIE + 14 or 21 days
(Table 3). The greatest yield increase Mermentau displayed from the midseason N ap-
plication was only 12 bu/acre. Mermentau obtained a maximum yield of 145 bu/acre
when midseason N was applied which was substantially greater than the 119 bu/acre
with the single preflood application. Additionally, Mermentau produced a significantly
greater yield of 133 bu/acre when no midseason N was applied (preflood N rates of
85 and 105 lb N/acre averaged) compared to when 130 lb N/acre was applied in a
single preflood N application. These results, along with those in Table 2, indicate that
preflood N had a significant impact on the severity of Zn deficiency in Mermentau
with severity apparently increasing as the preflood N rate increased. By contrast, when
the midseason N was delayed until 14 or 21 days past BIE, the yield of Mermentau
significantly increased compared to when no midseason N was applied or when it was
308
applied at BIE or BIE + 7 days. Perhaps Mermentau responded better to the later mid-
season N applications compared to the earlier applications because it had more time to
recover from the Zn deficiency; then again, previous studies (Norman et al., 2014) on
midseason N timing has shown a tendency at times for grain yield to increase as the
midseason N application time was delayed. Mermentau recovered somewhat from the
Zn deficiency as the season went along; however, remnants of the Zn deficiency were
observable even at maturity.
Application of midseason N to Roy J did not significantly increase grain yield
compared to no midseason N until it was delayed until BIE + 7 to 21 days and the yield
was greater when the midseason N was delayed until BIE + 21 days compared to at BIE
(Table 3). The yield of Roy J was greater with the single preflood N application than
when no midseason N was applied and similar to when midseason N was applied at
all of the application times. Although it should be mentioned that when the midseason
N was delayed until BIE + 21 days, the yield was close to being significantly greater
than the yield with the single preflood N application. As with Mermentau, Roy J may
have responded better to the later midseason N applications compare to the earlier
applications because it had more time to recover from the Zn deficiency. However, as
mentioned earlier, previous studies (Norman et al., 2014) on midseason N timing has
shown a tendency at times for grain yield to increase as the midseason N was delayed.
Roy J displayed only mild Zn deficiency symptoms and as the season went along the
deficiency was less observable with no noticeable deficiency symptoms displayed by
heading or ostensibly in the grain yield results, especially for a June seeding date.
The grain yield results clearly reflected the more severe damage the Zn deficiency
inflicted on the semidwarf Mermentau compared to the standard stature Roy J that ap-
peared to have yields typical or greater than typical for a June seeding date. One reason
may be the stage of growth when the Zn deficiency was first observed with Mermentau
already at the end of vegetative growth, BIE; whereas Roy J was about a week away
from BIE. Another reason may be that Mermentau is a semidwarf and more susceptible
to Zn deficiency compared to a short stature cultivar like Roy J. Visual observation
indicated that Mermentau had greater stunting, greater loss of stand, and less tillering
than Roy J. Although Mermentau recovered from the Zn deficiency as the season went
along, remnants of the Zn deficiency were observable even at maturity. By contrast,
Roy J displayed only mild Zn deficiency symptoms; and as the season went along the
deficiency was less observable with no noticeable deficiency symptoms displayed by
heading or apparently in the grain yield results. The yield results suggested that when
a Zn deficiency occurs in a semidwarf like Mermentau, a two-way split may result in a
greater grain yield compared to a single preflood application and it might be better to
delay the midseason application until at least BIE + 14 to 21 days.
309
ACKNOWLEDGMENTS
LITERATURE CITED
Norman, R.J., J.T. Hardke, T.L. Roberts, N.A. Slaton, D.L. Frizzell, M.W. Duren,
and E. Castaneda-Gonzalez. 2014. Response of two rice varieties to midseason
nitrogen fertilizer application timing. In: R.J. Norman and K.A.K. Moldenhauer
(eds.). B.R. Wells Arkansas Rice Research Studies 2013. University of Arkansas
Agricultural Experiment Station Research Series 617:303-310. Fayetteville, Ark.
Norman, R.J., C.E. Wilson Jr., and N.A. Slaton. 2003. Soil fertilization and rice
nutrition in U.S. mechanized rice culture. pp. 331-411. In: C.W. Smith and R.H.
Dilday (eds.). Rice: Origin, History, Technology, and Production. Wiley Sciences.
Table 1. Analysis of variance P values

for rice grain yield as affected by rice cultivar,
preflood N rate, and midseason N timing at
the Pine Tree Research Station during 2015.
Source P values
Cultivar (Cul) <0.0001
Preflood N rate (Pfn) 0.0011
Midseason N timing (Msn) <0.0001
Cul Pfn 0.0274
Cul Msn 0.0480
Pfn Msn 0.5354
Cul Pfn Msn 0.9467
310
Table 2. Influence of preflood N

rate and cultivar, averaged over
midseason N timing, on rice grain yield at the
Pine Tree Research Station, near Colt, Ark., during 2015.
Grain yield
Preflood N rate Mermentau Roy J
(lb N/acre) -------------- (bu/acre)------------
85 135 c 168 b
105 138 c 180 a
130 119 d 172 b
LSD0.05 5.9

An additional 45 lb N/acre was applied at midseason to
these preflood N rates.

Single preflood N fertilizer application with no midseason N.

Table 3. Influence of midseason (MS)

N application timing and cultivar,
averaged over preflood N rate, on rice grain
yield at the Pine Tree Research Station during 2015.
Grain yield
MS N timing Mermentau Roy J
------------- (bu/acre)--------------
No MS N 133 e 161 c
BIE 129 e 170 bc
BIE + 7days 131 e 179 ab
BIE + 14days 145 d 178 ab
BIE + 21days 145 d 181 a
SPF 119 f 172 ab
LSD0.05 9.3

45 lb N/acre was applied at midseason to all BIE treat-
ments.

BIE = beginning internode elongation.

SPF = single preflood fertilizer application of 130 lbs N/A
with no midseason N.

311
RICE CULTURE
Rice Cultivar Yield Response to Delayed Preflood

Nitrogen Application and Flood Establishment Time
N.A. Slaton1, T. Richmond1, J.T. Hardke2,
T.L. Roberts1, R.J. Norman1, Y. Liyew3, and D.L. Frizzell4
ABSTRACT
The preflood nitrogen (N) application must sometimes be delayed to obtain the dry soil
conditions that allow for maximal rice uptake of the fertilizer-N. Our research objective
was to determine the influence of preflood urea-N/flooding application time on the grain
yield of selected rice cultivars. Three trials were conducted in 2015 with each involving
five urea-N rates, two to five rice cultivars and five or six fertilizer-N application and
flood establishment times that ranged from 296 to 1640 cumulative Degree-Day 50
(DD50) units following emergence. For this report, only yield results from rice fertilized
with 160 lb N/acre will be presented since this treatment consistently maximized grain
yield. For each site, grain yield data were regressed across the cumulative DD50 units
at the time of fertilizer-N application. The rice yield results suggest that the preflood-N
can be delayed longer than is currently recommended without harming rice yield, even
when no other fertilizer-N was applied during the seedling stage. For most cultivars
and sites, application of the preflood-N and flooding at 650 to 1000 DD50s after emer-
gence appears to maximize yield. The yield of some cultivars was not reduced when
fertilizer-N was delayed beyond this range of DD50 accumulation. Results show that
rice grain yields can be maximized by a wide range of preflood-N application timings
provided other aspects of rice management including timely flood establishment and
weed control can be managed appropriately.
INTRODUCTION
The application of the preflood urea-N to rice must sometimes be delayed to obtain
the dry soil condition that is desired to slow the transformation of urea-N to NH3-N or
NO3-N and obtain maximal N uptake. Research on this topic is limited to work pub-
1
Professor, Graduate Assistant, Assistant Professor, and Professor, respectively, Department of Crop,
2
3
Research Program Technician, Pine Tree Research Station, near Colt.
4
312
lished by Norman et al. (1992) who reported that preflood-applied N could be delayed
up to 3 weeks past the 5-lf stage without loss of rice yield potential. Interest in how
long preflood-N can be delayed without yield loss has been stimulated by the need to
reduce water use, the availability of shorter-season cultivars and hybrids, and several
consecutive years of wet field conditions when rice is ready for preflood-N application.
Knowledge of how rice responds to preflood-N timing under different native soil-N
availabilities and among cultivars and hybrids is needed to aid growers in making the
proper decisions on fertilization.
The current recommendations suggest the optimum time to apply preflood-N is
350 to 550 Degree-Day 50 (DD50) units after rice emergence provided the soil is dry.
If the soil is wet during this recommended period, growers should delay preflood-N
application until a time that is about 3 weeks before the predicted 0.5-in. internode
elongation. Based on this recommendation, early-season cultivars have a shorter window
for preflood-N application than longer-season cultivars. Upon reviewing the available
literature, we could find no information stating how the current recommendation for
preflood-N timing was developed and presume that the practice of flooding fields near
the 4- to 5-lf stage became accepted primarily to reduce weed competition and control
practices (Johnston and Miller, 1973).
Our research objective was to determine the influence of preflood-N/flooding
application time on the grain yield of selected rice cultivars. The goal of this project
is to provide growers with the information they need to manage preflood fertilizer-N
when field conditions deviate from the standard recommended practice and to establish
the best time to apply preflood-N and the permanent flood to rice grown in the direct-
seeded, delayed-flood production system.
PROCEDURES
Trials were established on a Calhoun silt loam in two different fields (A and B) at
the University of Arkansas System Division of Agricultures Pine Tree Research Sta-
tion (PTRS) and on a Dewitt silt loam at the Rice Research Extension Center (RREC).
Rice was seeded on 8 April at PTRS-A, 1 May at PTRS-B, and 1 May at the RREC.
Rice management differed from the standard recommended practices only in regard to
the time of fertilizer-N application and establishing the permanent flood, which was
the primary factor being examined in these experiments. The research area at each
site was managed using standard practices for stand establishment, phosphorus and
potassium fertilization and pest control to ensure these factors were not yield limiting.
Weed control varied among the fertilization time treatments within each site (data not
shown) since we did not want weed infestations to influence rice yield potential. Thus,
we assume that the additional weed control measures required in the delayed flooding
treatments had no beneficial or detrimental effect on rice yield potential beyond the
control measures that were performed across all fertilization times.
The PTRS-A site contained four cultivars including CL111, LaKast, Roy J, and
Jupiter. Roy J and RiceTec XL753 were seeded at PTRS-B. The RREC trial included
313
CL111, LaKast, XL753, Roy J, and Jupiter. Within each fertilization time, each cultivar
was represented within a single bay and the order of the cultivars was the same within
each bay. The bays were flooded sequentially (not in random order) to facilitate man-
agement including establishing the flood, draining floodwater at maturity, and harvest.
At each site, urea-N treated with 3 qt Agrotain Ultra/ton urea was applied at rates
of 0 to 160 lb N/acre in 40 lb N/acre increments on five or six different dates after rice
had reached the 3-lf stage (Tables 1-3). At each preflood-N fertilization time, urea-N
was applied to a dry soil surface and flooded 2 or 3 days after preflood-N application
at the RREC and PTRS, respectively. The intended time interval between preflood-N
fertilization/flood times was 7 days but times were adjusted as needed to apply urea-N
to a dry soil surface. No midseason-N was applied at any site. Heading progress notes
and grain yield were measured on all cultivars at each site. Harvest at each site was
performed with the same small-plot combine, but at different times since rice heading
and maturation was different among N fertilization times (Tables 1-3). Grain yield data
from each plot were adjusted to 12% moisture for statistical analysis.
At each site, each fertilization time was a randomized complete block design
(RCB) with each preflood-N rate represented within each of four blocks. For simplifica-
tion, only grain yield data from the greatest preflood-N rate (160 lb N/acre) was used
in the analysis of variance as an RCB because it produced among the highest yields of
all N rates at each fertilization time. When appropriate, means were separated using
Fishers least significant difference test. Yield differences were interpreted as significant
at the P 0.10 level.

The estimated time of 50% heading was affected by the time of preflood-N/flood
application with heading being delayed as preflood-N/flood application was delayed
(Fig. 1). The date of 50% heading was delayed by nearly 3 weeks when preflood-N
application and flooding was delayed from 28 May to 25 June. Rice in the two trials
located at the PTRS showed a similar response in delayed maturation (data not shown).
The four rice cultivars grown at PTRS-A exhibited different responses to preflood-
N fertilization/flood application time (Table 1). The initial N fertilization time in this
field was delayed for more than a week due to wet soil conditions once seedling rice had
accumulated >250 DD50s. Grain yield of the two earliest maturing cultivars, CL111 and
LaKast, was not affected by N fertilization time. The yield of Jupiter rice was lowest
when the preflood-N was applied at the first two application times (578 and 872 DD50s)
and greatest when the preflood-N was applied 1244 or 1401 DD50s after emergence.
Preflood-N applied at the other two times produced intermediate yields for Jupiter. Roy
J yields were constant for the first four N fertilization times and then decreased when
urea-N was applied after 17 June (1244 DD50 units).
The grain yield of Roy J and XL753 rice seeded at the PTRS-B were both affected
by preflood-N/flood application time (Table 2). Maximal yield for both genotypes was
produced when the preflood-N was applied 552 to 941 cumulative DD50s after rice
314
emergence. The lowest yields were produced when preflood-N was applied before 300
and after 1100 DD50s had accumulated. The lower yields of the XL753 fertilized 1357
DD50s after rice emergence were partially attributed to substantial lodging (45%) that
occurred before harvest. Roy J rice did not lodge. Overall, the results from these two
cultivars at PTRS-B suggest that applying preflood-N/flood too early or too late can
be detrimental to grain yield.
Like the cultivar-specific yield results reported for the rice at the PTRS, the
effect of N fertilization/flood application time differed among the five rice cultivars
seeded at the RREC (Table 3). The yields of LaKast and XL753 were not influenced by
preflood-N fertilization time. The yield of CL111 was lowest for rice fertilized before
610 DD50s had accumulated and greatest when rice was fertilized when 821 DD50s
had accumulated before decreasing slightly on the two final fertilization times (1001 and
1253 DD50s). For Jupiter, there was no clear pattern in the yield response to preflood-N/
flood application time. Yields were greatest when fertilizer-N was applied on the first
(453 DD50s) and last (1253 DD50s) times, intermediate when applied at 821 DD50s,
and lowest on the second (610 DD50s) and fourth (1001 DD50s) times. Roy J grain
yields were greatest when preflood-N/flood was applied at 1001 DD50s compared to
all other application times, which were equal but 14 to 22 bu/acre lower than the yields
produced when preflood-N/flood was applied at 1001 DD50s.
The rice yield results from three trials and multiple cultivars seeded in 2015 sug-
gest that the preflood urea-N/flood can probably be delayed longer than is currently
recommended without harming rice yield, even when no other fertilizer-N is applied
during the seedling stage. The current recommendation indicates preflood-N should be
applied no later than 510 DD50s before the predicted date of 0.5-in. internode elonga-
tion and assumes that rice development occurs consistently regardless of N availability
and flooding. For most cultivars and hybrids, it appears that the optimal time to ap-
ply preflood-N/flood may be 650 to 1000 DD50s after emergence. The yield of some
cultivars was not reduced when fertilizer-N was delayed beyond this range of DD50
accumulation. While the results clearly show that preflood-N ferftilizer application can
be substantially delayed, delaying the flood influences other aspects of rice management
with the most obvious effects experienced in 2015 being the need for additional weed
control and delayed maturity when the fertilizer-N and flood were delayed several weeks
beyond the 4- to 5-lf stage. The results from 2015 suggest that cultivars may respond to
delayed preflood-N/flood differently. The different responses for rice cultivars seeded
at the same site could be due to differences in weather (e.g., rainfall and temperatures)
at critical times (e.g., flowering) that influenced grain yield. Thus, conclusions from the
2015 trial results should be evaluated with caution. When this study is concluded, results
from all preflood N rates and measurements will need to be examined to consider the
advantages and disadvantages in relation to rice management, use of natural resources,
and the economic implications for rice growers.
315
ACKNOWLEDGMENTS
Research was funded by a grant from the Rice Check-off Program administered
by the Arkansas Rice Research and Promotion Board, Fertilizer Tonnage Fees, and the
University of Arkansas System Division of Agriculture.
LITERATURE CITED
Johnston, T.H. and M.D. Miller. 1973. Culture. pp. 88-134 In: Rice in the United
States: Varieties and Production. USDA-ARS Handbook 289. U.S. Department of
Agriculture, Washington, D.C.
Norman, R.J., R.S. Helms, and B.R. Wells. 1992. Influence of delaying flood and
preflood nitrogen application on dry-seeded rice. Fertilizer Research J. 32:55-59.
Table 1. The date, cumulative Degree-Day 50 (DD50) units, and grain

yield of four rice cultivars as affected by preflood-N fertilization/flood date
for rice seeded on 8 April 2016 at the Pine Tree Research Station, near Colt, Ark.
Fertilization time Cultivar
Date Cumulative DD50 units CL111 Jupiter LaKast Roy J
------------------------------- (bu/acre)-----------------
23 May 578 167 a 174 d 174 a 175 ab
4 June 834 159 a 167 d 165 a 174 ab
8 June 944 161 a 196 c 160 a 178 ab
17 June 1244 154 a 218 ab 177 a 187 a
23 June 1401 152 a 222 a 178 a 172 b
1 July 1640 151 a 209 b 174 a 140 c
LSD0.10 NS 12 NS 14
P-value 0.4497 <0.0001 0.3178 0.0005
316
Table 2. The date, cumulative Degree-Day 50 (DD50) units, and grain yield
of Roy J and RiceTec XL753 as affected by preflood-N fertilization/flood date
for rice seeded on 1 May 2016 at the Pine Tree Research Station, near Colt, Ark.
Date Cumulative DD50 units Roy J XL753
------------(bu/acre)------------
23 May 296 154 d 211 c
4 June 552 187 ab 236 a
8 June 659 197 a 228 ab
17 June 941 183 b 214 bc
23 June 1118 168 c 207 c
1 July 1357 169 c 190 d
LSD0.10 11 15
P-value 0.0001 0.011
Table 3. The date, cumulative Degree-Day 50 (DD50) units, and grain yield
of five rice cultivars as affected by preflood-N fertilization/flood date for rice
seeded on 1 May 2016 at the Rice Research Extension Center, near Stuttgart, Ark.
Date Cumulative DD50 units CL111 Jupiter LaKast Roy J XL753
----------------------------(bu/acre)-----------------------------
28 May 453 126 d 196 a 181 a 183 b 241 a
4 June 610 132 cd 144 c 179 a 191 b 229 a
11 June 821 152 a 179 b 179 a 191 b 231 a
17 June 1001 139 bc 148 c 171 a 205 a 240 a
25 June 1253 141 b 207 a 186 a 183 b 228 a
LSD0.10 8 13 NS 11 NS
P-value 0.0057 <0.0001 0.1586 0.0187 0.5368
317
Fig. 1. Heading progress of Roy J rice receiving 160 lb urea-N/acre preflood

as affected by application and flood establishment timing in an experiment
conducted at the Rice Research Extension Center, near Stuttgart, Ark., in 2015.
318
RICE CULTURE
Response of Two Rice Cultivars to

Midseason Nitrogen Fertilizer Application Timing
A.D. Smartt1, R.J. Norman1, J.T. Hardke2, T.L. Roberts1, N.A. Slaton1,
D.L. Frizzell3, M.W. Duren4, E. Castaneda-Gonzalez3, and G.J. Lee2
ABSTRACT
A study was conducted at two locations in 2015 to examine the influence of midseason
nitrogen (N) application and its timing on the grain yield of conventional, pure-line rice
(Oryza sativa L.) cultivars from Louisiana and Arkansas. The conventional rice cultivars
chosen for the study at the University of Arkansas System Division of Agricultures
Northeast Research and Extension Center (NEREC) and Rice Research and Extension
Center (RREC) were the Louisiana semidwarf, long-grain Mermentau and the Arkansas
short stature, long-grain Roy J. There were two preflood N rates and four midseason
N application timings at beginning internode elongation (BIE), BIE + 7 days, BIE +
14 days, and BIE + 21 days. There was also a control, or no midseason N application,
and a single preflood N application. It should be noted that issues at the NEREC such
as difficulty maintaining a flood and bird damage increased the variability of the data
and probably affected the mean grain yields. Roy J produced a greater grain yield
than Mermentau at the NEREC and RREC. A single preflood N application produced
a similar or greater yield than the two-way split application at both the NEREC and
RREC. Application of midseason N increased grain yield when applied at BIE + 21 days
at the NEREC and at BIE + 7, BIE + 14, or BIE + 21 days at the RREC, while other
earlier midseason N application times did not significantly increase grain yield. Grain
yield increased as the midseason N was delayed at the RREC, but not at the NEREC.
Results from this and previous studies have led to the new recommendation that the
midseason N application should be applied no earlier than BIE and at least 3 weeks
after the preflood N application; both of these conditions have to be met to obtain the
full grain yield benefit from the midseason N application.
1
Program Associate, Professor, Assistant Professor, and Professor, respectively, Department of Crop,
2
3
Stuttgart.
4
319
INTRODUCTION
Nitrogen fertilizer typically is applied in a two-way split application for conven-
tional, pure-line rice cultivars in dry-seeded, delayed-flood systems (Norman et al.,
2013b).The first N application is applied preflood, onto dry soil, at beginning tillering
and the second N application occurs into the floodwater at midseason between begin-
ning internode elongation (BIE) and BIE + 7 days, or approximately 0.5-inch internode
elongation (IE; Norman et al., 2013b). The preflood N application is the larger of the
two and ranges, for pure-line cultivars, from 75 to 105 lb N/acre depending on the
cultivar (Roberts and Hardke, 2015). The preflood N rate is increased by 30 lb N/acre
for rice grown on clay soils, but the midseason N application rate of 45 lb N/acre is
consistent among all conventional, pure-line cultivars and soil textural classes (Roberts
and Hardke, 2015). The current recommendation for midseason N application to occur
from BIE to 0.5-inch IE has not been updated for nearly 20 years (Wilson et al., 1998).
Due to the introduction of several new rice cultivars since the last midseason N timing
study was conducted, new studies have been initiated in order to determine how recently
released conventional, pure-line rice cultivars respond to midseason N application and
the optimal application timing window.
Recent research has indicated some of the new cultivars do not consistently
respond to midseason N application, particularly when an adequate rate of preflood N
has been applied. Furthermore, the results of recent studies indicate, when midseason N
application produces a grain yield response, the midseason N application time window
may be wider and/or later than the week between BIE and 0.5-inch IE as suggested
by Wilson et al. (1998). Results of a 2011 midseason N application study indicated a
positive influence on rice grain yield when midseason N was applied from BIE to BIE +
14, while BIE + 21 days was not tested (Norman et al., 2012). The 2012 study indicated
midseason N applied from BIE to BIE + 21 days significantly increased rice grain yield
at two locations, while none of the midseason N application timings resulted in a yield
increase at the third location (Norman et al., 2013a). Similarly, the 2013 study showed
midseason N applications from BIE to BIE + 21 days generally increased grain yield
for both preflood N rates at the University of Arkansas System Division of Agricultures
Northeast Research and Extension Center (NEREC) and Rice Research and Extension
Center (RREC), while no midseason N application timings produced a yield response
at the Pine Tree Research Station (PTRS) with the greater preflood N rate (Norman et
al., 2014). Consequently, the midseason N application timing study was continued in
2015 to further clarify the impact of midseason N applied at four times from BIE to BIE
+ 21 days on the grain yield of rice based on two preflood N application rates. The two
conventional, pure-line rice cultivars selected for the 2015 study were the semidwarf,
long-grain cultivar Mermentau from Louisiana and the short stature, long-grain cultivar
Roy J from Arkansas.
PROCEDURES
The study was conducted in 2015 at the RREC, near Stuttgart, Ark., on a DeWitt
silt loam and the NEREC, Keiser, Ark., on a Sharkey clay. The two conventional,
320
pure-line rice cultivars chosen for the study were the Louisiana long-grain, semidwarf
Mermentau and the Arkansas long-grain, short stature cultivar Roy J. Two preflood N
rates were utilized at each location along with four midseason N application timings.
Preflood N application rates of 85 and 105 lb N/acre were used at the RREC, while larger
rates of 115 and 135 lb N/acre were used on the clay soil at the NEREC. The midseason
N rate was 45 lb N/acre at both locations and was applied at BIE, BIE + 7 days, BIE
+ 14 days, or BIE + 21 days. Additional treatments were a control, or no midseason N
application, and a single preflood N application of 130 and 160 lb N/acre at the RREC
and NEREC, respectively. All treatments were replicated four times at each location.
Preflood N was applied onto dry soil within 24 hours prior to flood establishment and
midseason N applications occurred directly into the floodwater.
The rice was drill-seeded in plots 9 rows wide and 15 ft in length with row spacing
of 7 in. at a rate of 73 lb/acre on the silt-loam soil at the RREC and 91 lb/acre on the
clay soil at the NEREC. Rice was seeded at the NEREC on 4 May and emerged on 20
May, the preflood N was applied on 18 June, and the BIE application occurred on 8 July
and 16 July for the cultivars Mermentau and Roy J, respectively. Rice was seeded at the
RREC on 1 May and emerged on 10 May, the preflood N was applied on 3 June, and
the BIE application occurred on 22 June and 26 June for the cultivars Mermentau and
Roy J, respectively. A permanent flood was established at both locations the day after
preflood N application when the rice was at the 5- to 7-leaf stage and maintained until
the rice reached maturity. The center 5 rows of each plot were harvested at maturity,
the moisture content and weight of grain were determined, and yields were calculated
based on 12% moisture and a 45-lb bushel weight.
Treatments were arranged in a four replicate randomized complete block factorial
design with 2 cultivars 2 preflood N rates 4 midseason N application timings. A
control with no midseason N application and a single preflood N application treatment
were included, each with four replications at both locations. Analysis of variance was
performed on the grain yield data utilizing SAS v. 9.4 (SAS Institute Inc., Cary, N.C.).
When necessary, differences among means were compared using Fishers protected
least significant difference (LSD) procedure at a P = 0.05 probability level.

It should be noted that there were some issues at the NEREC that impacted the
results. We had difficulty maintaining the flood due to seepage through the levees and
there was bird damage to the plots that increased the variability of the data and probably
affected the mean grain yields.
Analysis of variance P values for the studies indicated there were no significant
(P = 0.05) three-way interactions of cultivar preflood N rate midseason N timing
or two-way interactions of the treatments on grain yield at either of the two locations
(Table 1). There were, however, significant (P < 0.05) main effects of cultivar, preflood
N rate, and midseason N timing on rice grain yield at both the NEREC and RREC.
At both locations, averaged across preflood N rate and midseason N application
timing, the cultivar Roy J outyielded Mermentau by 21 bu/acre at the NEREC and the
321
RREC (Table 2). Averaged across cultivar and midseason N application timing, the
single preflood N application at both locations produced greater grain yields than the
two-way split application at the lower preflood N rates and similar yields to the two-way
split application at the larger preflood N rates (Table 3). Additionally, the two-way split
with the larger preflood N rate increased grain yields compared to the two-way split
with the smaller preflood N rate at the RREC, but not at the NEREC.
Averaged over cultivar and preflood N rate at the NEREC, the single preflood N
application and BIE + 21 days midseason N application resulted in greater grain yields
than the control (no midseason N application); however the three earlier midseason N
application timings (BIE, BIE + 7, and BIE + 14 days) did not produce different grain
yields than the control, the single preflood application, or the BIE + 21 days midsea-
son application timing (Table 4). At the RREC, averaged over cultivar and preflood N
rate, midseason N application increased grain yield over the control when applied at
BIE + 7, BIE + 14, or BIE + 21 days, but not when applied at BIE. Rice grain yields
were greatest at the RREC and did not differ among the single preflood N application
and the two-split when the midseason N was applied at BIE + 14, or at BIE + 21 days.
A single preflood N application resulted in equal to or greater grain yields than
when N was applied in a two-way split application at both locations. On the clay soil
at the NEREC, midseason N application only resulted in a grain yield increase over the
control (no midseason N) when applied at BIE + 21 days; whereas, on the silt-loam soil
at the RREC, midseason N application produced a yield increase over the control when
applied at BIE + 7, BIE + 14, or BIE + 21 days. Grain yield increased as the midseason
N was delayed at the RREC, but not at the NEREC. The general trend of increasing
grain yield as midseason N application is delayed past BIE, up to 21 days, observed at
the RREC has been reported in previous studies (Norman et al., 2012; 2013a; 2014).
The 2015 results coupled with results from previous studies has led to a change in the
recommendation for midseason N application timing. The new recommendation is the
midseason N application should be applied no earlier than BIE and at least 3 weeks
after the preflood N application; both of these conditions have to be met to obtain the
full grain yield benefit from the midseason N application. Future research will help
determine how late midseason applications should occur in new cultivars and how wide
the application time window is in order to optimize midseason N application timing to
ensure maximum grain yield benefits.
ACKNOWLEDGMENTS
and University of Arkansas System Division of Agriculture.
322
LITERATURE CITED
Norman, R.J., T.L. Roberts, D.L Frizzell, N.A. Slaton, C.E. Wilson Jr., and J.D.
Branson. 2012. Response of two rice varieties to midseason nitrogen fertilizer ap-
plication time. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells Rice
Norman, R.J., J.T. Hardke, T.L. Roberts, N.A. Slaton, D.L. Frizzell, J.M. Wiggins,
M.W. Duren and J.D. Branson. 2013a. Response of two rice varieties to midsea-
son nitrogen fertilizer application time. In: R.J. Norman and K.A.K. Moldenhauer
Norman, R.J., N.A. Slaton, and T.L. Roberts. 2013b. Soil Fertility. In: J.T. Hardke
(ed.). Rice Production Handbook. University of Arkansas Division of Agriculture
Cooperative Extension Service. MP192. Pp. 69-102. Little Rock.
Norman, R.J., J.T. Hardke, T.L. Roberts, N.A. Slaton, D.L. Frizzell, M.W. Duren,
and E. Castaneda-Gonzalez. 2014. Response of two rice varieties to midseason
nitrogen fertilizer application timing. In: R.J. Norman and K.A.K. Moldenhauer
Roberts, T.L. and J.T. Hardke. 2015. University of Arkansas Cooperative Extension
Service. 2015. Recommended nitrogen rates and distribution for rice varieties in
Arkansas. Fact Sheet. University of Arkansas Division of Agriculture Cooperative
Extension Service.
Wilson, C.E., P.K. Bollich, and R.J. Norman. 1998. Nitrogen application timing ef-
fects on nitrogen efficiency of dry seeded rice. Soil Sci. Soc. Am. J. 62:959-964.
Table 1. Analysis of variance P values for

rice grain yield as affected by rice cultivar, preflood
N rate, and midseason N timing at the Northeast
Research and Extension Center (NEREC) and Rice
Research and Extension Center (RREC) during 2015.
Source NEREC RREC
Cultivar (cult) <0.0001 <0.0001
Preflood N rate (pfn) 0.0164 <0.0001
Midseason N timing (msn timing) 0.0344 <0.0001
Cult pfn rate 0.2081 0.8711
Cult msn timing 0.6047 0.4343
Pfn rate msn timing 0.0729 0.4095
Cult pfn rate msn timing 0.1105 0.5884
323
Table 2. Influence of rice cultivar,

averaged across preflood N rate and midseason
N timing, on rice grain yield at the Northeast
Research and Extension Center (NEREC) and the Rice
Grain yield
Cultivar NEREC RREC
------------- (bu/acre)--------------
No MS N 133 e 161 c
Mermentau 151 b 149 b
Roy J 172 a 170 a
LSD0.05 6.6 3.7

Values in the same column followed by different letters are
significantly different (P < 0.05).

Table 3. Influence of preflood N

rate, averaged across midseason N timing
and cultivar, on rice grain yield at the Northeast
Research and Extension Center (NEREC) and the Rice
Grain yield
Preflood N rate NEREC RREC
------------- (bu/acre)--------------
85 ----- 154 b
105 ----- 163 a
130 ----- 166 a
115 156 b -----
135 166 ab -----
160 168 a -----
LSD0.05 10.9 5.9

45 lb N/acre applied at midseason.

Single preflood N fertilizer application with no midseason N.


324
Table 4. Influence of midseason (MS) N

application timing, averaged across preflood N
rate and cultivar, on rice grain yield at the Northeast Re-
search and Extension Center (NEREC) and the Rice
Grain yield
MS N timing NEREC RREC
------------- (bu/acre)--------------
No MS N 152 b 149 c
BIE 162 ab 154 bc
BIE + 7d 161 ab 158 b
BIE + 14d 163 ab 164 a
BIE + 21d 166 a 170 a
SPF 168 a 166 a
LSD0.05 12.0 6.7


BIE = beginning internode elongation.

SPF = single preflood fertilizer application of 130 lb N/acre
at RREC; 160 lb N/acre at NEREC.

325
RICE CULTURE
Summary of Nitrogen Soil Test for Rice (N-STaR)

Nitrogen Recommendations in Arkansas During 2015
S.M. Williamson1, T.L. Roberts1, C.L. Scott1,
R.J. Norman1, N.A. Slaton1, and J. Shafer2
ABSTRACT
Seeking to fine-tune nitrogen (N) fertilizer application, increase economic returns, and
decrease environmental N loss, some Arkansas farmers are turning away from blanket
N recommendations based on soil texture and cultivar and using the Nitrogen Soil Test
for Rice (N-STaR) to determine their field specific N rates. First introduced in 2010,
Roberts et al. (2010) correlated direct steam distillation (DSD) results from 18-inch
depth silt loam soil samples to plot-scale N response trials and subsequently performed
field-scale validation. The N-STaR has since been correlated for use on clay soils, using
a 12-inch depth soil sample, both at small-plot and field scale validation, and has been
offered to the public since 2013. To summarize the samples submitted to the University
of Arkansas System Division of Agricultures N-STaR Soil Testing Lab during 2015,
samples were categorized by county and soil texture. Samples were received from
19 Arkansas counties, with Mississippi county and Arkansas county submitting the
largest number of fields, with 41 and 31 fields respectively. The total samples received
were from 68 silt loam fields and 49 clay fields. The N-STaR N-rate recommendations
for these samples were then compared to the producers estimated N rate, the 2015
Recommended Nitrogen Rates and Distribution for Rice Cultivars in Arkansas, and
the standard Arkansas N-rate recommendation of 150 lb N/acre for silt loam soils and
180 lb N/acre for clay soils and divided into three categories and categorized based
on a decrease in recommendation, no change in recommended N rate, or an increase
in the N-rate recommendation. County, much like in 2013 and 2014, was found to be
a significant factor (P < 0.0001) in all three comparisons when N-STaR called for a
decrease in N rate suggesting that some areas of the state may have higher residual N
not accounted for in the other N-rate recommendation strategies when compared to
N-STaR. Soil texture was a significant factor in fields where N-STaR proposed a de-
creased N rate in the cultivar recommendation comparison (P < 0.05) and the standard
N rate comparison (P < 0.0005), and was also a significant factor (P < 0.05) in fields
for which N-STaR revealed an N rate increase in the producers estimate comparison.
1
Program Associate II, Assistant Professor, Program Technician, Professor, and Professor, respectively,
Department of Crop, Soil, and Environmental Science, Fayetteville.
2
Program Associate, Pine Tree Research and Extension Center, near Colt.
326
INTRODUCTION
In the past, nitrogen (N) fertilizer recommendations for rice in Arkansas were
based on soil texture, cultivar, and previous crop; often resulting in over-fertilization
which can decrease possible economic returns and increase environmental N loss (Khan
et al., 2001). For years researchers tried to develop a soil-based N test that would allow
them to better predict the actual N feretilizer needs for a particular field, and finally in
2010 researchers at the University of Arkansas System Division of Agriculture expanded
upon previous research at the University of Illinois which used organic-N content in the
form of amino sugars to predict corn response to N fertilizer (Mulvaney et al., 2006).
University scientists correlated several years of plant-available N estimates from direct
steam distillation (DSD) results obtained from 18-inch depth soil samples, equivalent
to rice rooting depth on a silt loam soil (Roberts et al., 2009), to plot-scale N fertilizer
response trials across the state and developed a site-specific soil-based N test for Ar-
kansas rice (Roberts et al., 2011). Direct-seeded, delayed-flood rice production, with
proper flood management and the use of ammonium-based fertilizers and best manage-
ment practices, has a consistent soil N mineralization rate and one of the highest and
consistent N use efficiencies of any cropping system, therefore lending itself to a high
correlation of soil mineralizable N to grain yield response (Roberts et al., 2011). After
extensive field testing, the Nitrogen Soil Test for Rice (N-STaR), became available to
the public for silt loam soils in 2012 with the initiation of the University of Arkansas
N-STaR Soil Testing Lab in Fayetteville, Ark. Later, researchers correlated DSD results
from 12-inch depth soil samples to N fertilizer response trials on clay soils (Fulford,
2013), and N-STaR rate recommendations became available for clay soils in 2013.
PROCEDURES
In an effort to summarize the effect of the N-STaR program in Arkansas, samples
submitted to the University of Arkansas System Division of Agricultures N-STaR Soil
Testing Lab during 2015 were categorized by county and soil texture. The N-STaR N-rate
recommendations for these samples were then compared to the producers estimated N
rate if supplied on the N-STaR Soil Test Laboratory Soil Sample Information Sheet, the
2015 Recommended Nitrogen Rates, and Distribution for Rice Cultivars in Arkansas
(Roberts and Hardke, 2015), or to the standard Arkansas N-rate recommendation of
150 lb N/acre for silt loam soils and 180 lb N/acre for clay soils, and divided into three
categoriesthose with a decrease in N fertilizer rate recommendation, no change in
recommended N rate, or an increase in the N-rate recommendation. The resulting data
was analyzed using JMP 12 (SAS Institute, Inc., Cary, N.C.).

Samples were received from 118 fields which represented 30 farmers across 19
Arkansas counties. Mississippi County and Arkansas County, ranked 15th and 3rd in
327
planted acres (USDA-FSA, 2015), and evaluated the largest number of fields, with 41
and 31 fields, respectively. The samples received were from 68 silt loam fields and 49
clay fields (Table 1). There were only four farmers that sent samples for more than five
fields, the average number of fields submitted per farmer was 3.93, with 16 farmers only
submitting samples from one field. There were eight farmers who submitted samples
in 2015 that also submitted samples in 2014.
Planted rice acreage was down from 1.486 million acres in 2014 (Hardke, 2015)
to approximately 1.280 million in 2015 with close to 270,000 prevented acres (USDA-
FSA, 2015) most likely due to excessive rains at critical planting times. The N-STaR
sample submissions for 2015 mirrored the same trend and were significantly down from
the 233 fields submitted in 2014 most likely because of the very wet spring and the
inability to sample wet fields. Sample submission by county did not reflect the planted
acre estimates for 2015 with Poinsett and Lawrence counties having the highest estimates
(USDA-FSA, 2015) yet only submitting samples for seven and one field, respectively.
County (P < 0.0001) and soil texture (P < 0.0005) were found to be significant
factors in the fields with a decrease in N fertilizer rate when the N-STaR recommen-
dation was compared to Arkansas standard N-rate recommendation of 150 lb N/acre
for silt loam soils and 180 lb N/acre for clay soils. This suggests that areas of the state
may be prone to N savings potential due to cropping systems and soil series (Fig. 1).
County and soil texture were not significant in the fields where an increase in N rate
was recommended by N-STaR, however it should be noted that there were no clay
fields that resulted in an increased N rate in this comparison (Table 1). Of the fields in
this comparison, there was a decrease in the N recommendation for 105 fields (89%
of the 118 fields submitted) with an average decrease of 38.6 lb N/acre. No change in
N recommendation was found for four fields, while nine fields had an increase in N
recommendation (8%), with an average increase of 9.3 lb N/acre.
Eleven of the submitted fields had no estimated N fertilizer rate specified on
the N-STaR Sample Submission Sheet and were excluded from the comparison of the
N-STaR recommendation to the farmers estimated N rate. Of those compared, there
was a decrease in the N recommendation for 59 fields (~55% of the remaining 107
fields submitted) with an average decrease of 29.7 lb N/acre (Table 1). No change in
N recommendation was found for three fields, while 45 fields had an increase in N
recommendation (42%), with an average increase of 17.9 lb N/acre. Soil texture was
found to be a significant factor (P < 0.05) for the fields that resulted in an increase
from the producers estimate to the N-STaR recommendation but was not significant
in the fields that resulted in a decrease in N rate. The difference in significance may
be due to soil texture variability, soil texture classification errors, and the differences
in sample depth and the N-STaR calculations for the two soil textures. The N-STaR
recommendations continue to be largely dependent on proper sampling depth for the
respective soil texture and the farmers classification of his field. County was found to
be a significant factor in fields that showed a decrease in N-rate recommendation (P <
0.0001), but was not a significant factor for fields that called for an increase in N-rate
recommendation (Table 2).
328
When the N-STaR recommendation was compared to the 2015 Recommended

Nitrogen Rates and Distribution for Rice Cultivars in Arkansas (Roberts and Hardke,
2015), cultivar recommendations were adjusted for soil texture as recommended by
adding 30 lb N/acre for rice grown on clay soils and then compared to the N fertilizer
rates determined by N-STaR. The 21 fields that did not list a cultivar on the N-STaR
Sample Submission Sheet were excluded from this comparison. There was a decrease
in the N fertilizer recommendation for 79 fields (81% of the 97 fields) with an average
decrease of 33 lb N/acre (Table 3). No change in N recommendation was found for two
fields, while 16 fields had an increase in N recommendation (16%), with an average
increase of 15 lb N/acre. County (P < 0.0001), soil texture (P < 0.05), and cultivar
(P < 0.0005) were all significant factors in the fields exhibiting a decrease in N rate
suggesting that N rates for some cultivars may be overestimated for certain areas of
the state or soil textures. However, no factors were found to be significant in the fields
which resulted in an increase in N rate.
These results continue to show the importance of the N-STaR program to Ar-
kansas producers and can help target areas of the state that would most likely benefit
from its incorporation. Standard recommendations and cultivar recommendations will
continue to be good ballpark estimations for N rates, but field-specific N rates continue
to offer the best estimate of the N fertilizer needed for each particular field no matter
soil texture or cultivar selection. Farmers are encouraged to consider taking N-STaR
samples at the harvest of the previous crop when fields are typically in optimal condi-
tions for soil sampling.
ACKNOWLEDGMENTS
This research was funded by Arkansas Rice Check-Off funds administered by the
Division of Agriculture. Sincere gratitude is extended to the employees of the University
of Arkansas N-STaR Soil Testing Lab and the hourly personnel who spent many hours
conducting the analysis of N-STaR samples.
LITERATURE CITED
Fulford, A.M., T.L Roberts, R.J. Norman, N.A. Slaton, C.E. Wilson, T.W. Walker,
D.L. Frizzell, C.E. Greub, C.W. Rogers, M.W. Duren, and J. Shafer. 2013. Evalua-
tion of the Illinois nitrogen soil test and the nitrogen-soil test for rice grown on
clay soils. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells Rice
329
Hardke, J.T. 2015. Trends in Arkansas Rice Production, 2015. In: R.J. Norman and
22. Fayetteville, Ark.
Khan, S.A., R.L. Mulvaney, and R.G. Hoeft. 2001. A simple soil test for detecting sites
that are nonresponsive to nitrogen fertilization. Soil Sci. Soc. Am. J. 65:1751-1760.
Mulvaney, R.L., S.A. Khan, and T.R. Ellsworth. 2006. Need for a soil-based ap-
proach in managing nitrogen fertilizers for profitable corn production. Soil Sci.
Soc. Am. J. 70:172-182.
Roberts, T.L., R.J. Norman, N.A. Slaton, and C.E. Wilson Jr. 2009. Changes in alka-
line hydrolyzable nitrogen distribution with soil depth: Implications for fertilizer
correlation and calibration. Soil Sci. Soc. Am. J. 73:2143-2150.
Roberts, T.L. and J.T. Hardke. 2015. 2015 Recommended Nitrogen Rates and
Distribution for Rice Cultivars in Arkansas. University of Arkansas Cooperative
Extension Service. Access date: 3 September 2015. Available at: http://www.uaex.
edu/farm-ranch/crops-commercial-horticulture/rice/
Roberts, T.L., R.J. Norman, N.A. Slaton, C.E. Wilson Jr., and W.J. Ross. 2010. N-
ST*R- A Soil-Based Nitrogen Test for Fertilizer Recommendations in Arkansas
Rice Production. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R. Wells
Rice Research Studies 2009. University of Arkansas Agricultural Experiment Sta-
tion Research Series 581:190-196. Fayetteville.
Roberts, T.L., W.J. Ross, R.J. Norman, N.A. Slaton and C.E. Wilson Jr. 2011. Pre-
dicting nitrogen fertilizer needs for rice in Arkansas using alkaline hydrolyzable-
nitrogen. Soil Sci. Soc. Am. J. 75:1-12.
USDA-FSA. 2015. United States Department of Agriculture Farm Service Agency.
2015 acreage data as of 1 November 2015. Access date: 13 November 2015.
Available at: http://www.fsa.usda.gov/FSA/webapp?area=newsroom&subject=lan
ding&topic=foi-er-fri-cad
330
Table 1. Distribution and change in nitrogen (N) fertilizer rate

compared to the standard recommendation, producers estimated N rate, and
the 2015 recommended N rates for rice cultivars in Arkansas based on soil texture.
Recommendation
Number Decreased N-STaR Increased N-STaR

Soil of fields Mean N Mean N No
texture submitted Fields decrease Fields increase change
(no.) (lb N/acre) (no.) (lb N/acre)
Standard soil texture
Clay 49 48 43.8 - - 1
Silt Loam 68 57 34.2 9 9.3 3
Total 118 105 38.6 9 9.3 4
Producer estimate
Clay 44 19 27.6 23 21.1 2
Silt Loam 63 40 30.8 22 14.5 1
Total 107 59 29.7 45 17.9 3
Cultivar
Clay 40 38 37.8 - - 2
Silt Loam 57 41 28.3 16 14.9 -
Total 97 79 32.8 16 14.9 2
Table 2. Distribution and change in nitrogen (N) fertilizer

rate compared to the producers estimated N rate by countya.
Recommendation
Arkansas 30 18 -27.8 11 15.8 1
Ashley 1 - - 1 30.0 -
Chicot 1 1 -70.0 - - -
Clay 4 2 -25.0 2 22.5 -
Cross 1 1 -10.0 - - -
Desha 2 2 -12.5 - - -
Independence 1 1 -45.0 - - -
Lawrence 1 - - 1 10.0 -
Lee 1 - - 1 5.0 -
Lonoke 5 5 -62.0 - - -
Mississippi 40 16 -20.3 22 19.1 2
Phillips 2 2 -30.0 - - -
Poinsett 7 7 -40.0 - - -
Prairie 3 1 -15.0 2 25.0 -
Randolph 2 - - 2 10.0 -
St. Francis 2 2 -12.6 - - -
White 4 1 -40.0 3 16.7 -
Total 107 59 -29.7 45 17.9 3
a
Eleven fields did not list an estimated N rate on their N-STaR Sample Submission Sheet and
were excluded from the analysis.
331
Table 3. Distribution and change in nitrogen (N) fertilizer rate compared

to the 2015 recommended N rates for rice cultivars in Arkansas by cultivara.
Recommendation
Caffey 8 6 -22.5 2 12.0 -
CL 111 5 5 -42.0 - - -
CL 151 10 7 -32.9 3 21.7 -
CLXL 729 1 1 -20.0 - - -
CLXL 745 4 1 -100.0 2 12.5 1
Francis 1 1 -50.0 - - -
Jupiter 6 5 -46.0 1 5.0 -
Lakast 17 17 -20.0 - - -
Mermentau 1 - - 1 10.0 -
Roy J 15 8 -23.8 6 15.8 1
XL 753 29 28 -38.9 1 15.0 -
Total 97 79 -32.8 16 14.9 2
a
Twenty-one fields did not list a cultivar on their N-STaR Sample Submission Sheet and were
excluded from the analysis.
332
Fig. 1. Percent and mean decrease and increase in N-STaR nitrogen (N) fertilizer rate
recommendation by county compared to the standard N fertilizer rate recommendation.
333
Impact of Storage Duration, Temperature, and

Moisture Content on Mold Growth on Hybrid Rice
G.G. Atungulu1 and S. Thote1
ABSTRACT
There is a need to determine kinetics of mold growth on rough rice during storage at
various temperatures and moisture contents (MCs) and to delineate conditions that ef-
fectively mitigate mycotoxin contamination of the grain. The objective for this study
was to simulate conditions of delayed drying and prolonged storage of rough rice and to
characterize kinetics of mold growth for hybrid rice at different temperatures and MCs.
Long-grain hybrid rice cultivars (XL753 and CL XL745) conditioned to four different
MCs (12.5%, 16.0%, 19.0%, and 21.0% wet basis) were stored in sealed containers at
temperatures ranging from 45 F (10 C) to 104 F (40 C) for a period of 16 weeks.
For both cultivars, a direct relationship between mold counts and MC was observed;
as the MC level increased, the mold counts increased; whereas more complex trends
were observed for the effect of temperature and the duration of storage on mold growth.
The study concludes that long-grain hybrid rough rice could be stored at low MC levels
and moderate temperatures for up to 6 weeks without any change in the mold growth
profile. However, storing the rice at high MC (>17%) for more than 8 weeks, especially
at higher temperatures should be avoided.
INTRODUCTION
The main factor contributing to spoilage of rice is microbial development
(Skyrme et al., 1998; Ranalli and Howell, 2002; Atungulu et al., 2014). In particular,
mold contamination poses the greatest problem to rice producers, processors, and
consumers. Depending on stress conditions, which may include rice kernel physical
characteristics, moisture content (MC), storage temperature, and relative humidity,
mold harbored on rice may produce mycotoxins. Mycotoxins, especially aflatoxin, are
known carcinogens that pose health hazards to consumers of rice and rice co-products.
Microbial contamination of rice may lead to kernel discoloration and undesirable quality,
appearance, flavor changes, and other problems such as weakened kernels, resulting in
breakage and economic loss (Sahay and Gangopadhyay 1985; Singaravadivel and Raj,
1
Assistant Professor and Graduate Assistant, respectively, Department of Food Science, Fayetteville.
334
1983; Phillips et al., 1988; Misra and Vir, 1991). According to Frazier and Westhoff
(1967), the spores of storage molds have optimum temperature and relative humidity
ranges for development. When the conditions for growth of toxigenic mold are met,
some toxigenic molds such as Aspergillus flavus may produce aflatoxin (Frazier and
Westhoff, 1967; Atungulu et al., 2014; Richard et al., 2003; Lacey and Magan, 1991).
Previous research work focused on characterizing microflora on rice in general
(Skyrme et al., 1998; Ranalli and Howell, 2002). However, hybrid rice has a pubescent
characteristic which may predispose it to harbor and support mold growth in a different
manner when compared to non-pubescent rice. With increased concerns for food secu-
rity, there is a crucial need to conduct specific research to address the kinetics of fungal
mold growth in hybrid rice. Results from such research are vital for implementation of
specific postharvest management practices that target the increasingly popular hybrid
rice cultivars. This study endeavors to provide useful information regarding prevalence
of mold during different storage conditions that can help determine the optimum time-
frame and conditions for storage of hybrid rice to not only reduce the economic impact
of spoilage due to mold contamination, but also to mitigate the health risks posed by
toxigenic molds. The specific objectives for this study were to (1) determine the kinetics
of mold growth on long-grain hybrid rice at different MCs and storage temperatures,
and (2) determine the optimum conditions for mold growth.
PROCEDURES
Samples
Two long-grain, hybrid rice cultivars planted in Arkansas in 2014 were chosen
for this study, CLXL745 from Running Lake Farms near Pocahontas, Ark., and XL753,
from the University of Arkansas System Division of Agricultures Northeast Research
and Extension Center near Keiser, Ark. The rice samples were cleaned to remove chaff
and foreign matter and then conditioned to the set MC levels (12.5%, 16%, 19%, and
21%). After conditioning the rice to the set MC levels, the rice samples were immediately
stored in individual glass jars to prevent significant alterations of their initial MCs and
then transported to five separate environments with temperatures set at 45 F, 59 F,
68 F, 80.6 F, and 104 F (7.2 C, 15 C, 20 C, 27 C, and 40 C). The equipment to
generate environments used consisted of a combination of two parameter generation
and control (PG&C) units, one refrigerator, one equilibrium moisture content (EMC)
control chamber, and one incubator. The samples were stored for a period of up to 16
weeks with samples retrieved at an interval of every 2 weeks. The experiment design
is illustrated in Table 1.
The temperature was monitored by placing two temperature sensors (HOBOTM,
Onset Computer Corporation, Bourne, Mass.) in each environment. The MC of each
sample was measured by placing a 0.53 oz (15 g) sample into a 266 F (130 C) con-
vective oven (Shellblue, Sheldon Mfg., Inc., Cornelius, Ore.) for 24 hours followed
by cooling in a desiccator for at least one half hour (Jindal and Siebenmorgen, 1987).
Table 2 provides the MC values obtained using the method described above.
335
Microbial Analysis
The AOAC method 997.02 (2002) for the 3M Petrifilm Mold Count Plates (3M
Microbiology Product, Minneapolis, Minn.) was used to determine the rough rice total
(ground sample) microbial counts which were expressed in terms of colony forming units
(CFU) per gram of the sample. The suspensions were prepared by masticating the rice
samples at two different settings (Silver Panoramic, iUL, S.A., Barcelona, Spain). The
masticator was set at 240 s and 0.5 stroke/s, allowing the rice samples to be pulverized
into powder for total microbial load analysis. The successive dilutions of 10-4 to 10-7
concentration were made by mixing 1 mL of the original mixture with 9 mL of phosphate-
buffered dilution water and plated. After the recommended incubation period of 120 h,
the CFUs on each plate were calculated using the following formula:
PCFU
Tcfu = Eq. 1
Dr
where, Tcfu is total CFUs per gram of rice (CFUs/g), PCFU is CFUs counted on plate per
gram of rice (CFUs/g), and Dr is dilution factor.
For the studied rice samples, preliminary results showed that yeast counts were
very low with nearly none detected even with 10-10 dilution. Therefore, yeast count was
not reported in this research.
Statistical Analysis
Linear regression, analysis of variance, Students t test (least significant difference
test), and the Tukey honest significant difference tests were performed with statistical
software (JMP v. 12.0.0, SAS Institute, Inc., Cary, N.C.). Level of significance (P) was
set at 5% for comparing means. Table 3 gives the P values and degrees of freedom
associated with the three factors studied.

Interaction - Week Number, Temperature, Moisture Content (MC%)
Since the three-way interaction between duration of storage (week number),
temperature, and MC was significant (Table 3), the analysis for all three terms was
performed together for each cultivar. Figure 1a gives five separate contour plots for
the cultivar, CLXL745, showing the distribution of mold counts (log10 CFU/g) over
the weeks (y-axis) for various MC levels (x-axis) for each of the storage temperature
conditions. For all five temperatures, the contour plots show that less than (or equal to)
6.0 log10 CFU/g of mold were detected for most of the samples. The other distinguishing
characteristic revealed by observing all five of the plots was that below 17% MC, the
mold growth remained consistently at or below 6.0 log10 CFU/g. The trends observed
in these plots suggest that as storage temperature rises, the mold contamination peaks at
336
about 8 weeks for higher MC samples, and begins dropping after week 12. The steady
fall in CFUs detected for increasing temperatures suggests that at lower MC, higher
temperature conditions are not conducive to the proliferation of mold.
Similarly, figure 1b shows contour plots for the cultivar XL753. The trends ob-
served are very similar to those observed for CLXL745, except the density and area
occupied by the dark contours (greater than or equal to 7.0 log10 CFU/g) were smaller
for temperatures below 80.6 F (27 C). However, the same trend of increased mold
growth was observed for samples above 17% MC and between weeks 6 and 12.
Moisture Content (MC %)

The difference between the molds detected for the different MC levels for the
two cultivars was also significant (Table 3). Figure 2 plots the average mean (log10
CFU/g) versus the MC level for cultivars CLXL745 and XL753. The trend observed
from these two curves supports the trends observed from the contour plots. For both
cultivars, the mold counts for the two lower MC levels are significantly lower than the
two higher MC levels (P < 0.0001 for both). Additionally, for cultivar CLXL745, the
mold counts for MC = 21.1% are significantly higher than MC = 18.8%. However, no
difference was detected between the mold counts for MC = 20.8% and MC = 20.3%
for cultivar XL753. For both cultivars, a direct relationship between mold counts and
MC was observed; as the MC level increased, the mold counts increased.
The findings from this study provide baseline information which is very helpful
in modeling kinetics of microbial growth during rice storage. The information may be
useful to guide decisions on drying and storage conditions, especially of hybrid rice
cultivars, to avoid mold growth leading to mycotoxin contamination.
ACKNOWLEDGMENTS
The authors acknowledge the support of the University of Arkansas System Di-
vision of Agriculture. The authors greatly appreciate the Arkansas Rice Research and
Promotion Board for financially supporting this research and thank Kathryn Haydon, B.
Bluhm, Andy Mauromastakos, and Terry Siebenmorgen (of the University of Arkansas
System Division of Agriculture) for support during the research.
LITERATURE CITED
AOAC International. 2002. Official Methods of Analysis, 18th Ed. Method 997.02.
Yeast and mold counts in foods. AOAC International, Gaithersburg, Md., USA.
337
Atungulu, G.G., H. Zhong, and T.J. Siebenmorgen. 2014. Microbial population

on freshly-harvested rice and factors affecting prevalence. ASABE Paper No.
141914312. St. Joseph, Mich.: ASABE.
Brooker, D.B., F.W. Bakker-Arkema, and C.W. Hall. 1974. Drying cereal grains. AVI
Publishing Company.
Frazier, W.C. and D.C. Westhoff. 1967. Food Microbiology. Mc Graw-Hill Book
Company. New York, N.Y., pp. 252-282.
Jindal, V.K. and T.J. Siebenmorgen. 1987. Effects of oven drying temperature
and drying time on rough rice moisture content determination. Trans. ASAE
30(4):1185-1192.
Lacey, J. and N. Magan. 1991. Fungi in cereal grains: Their occurrence and water
and temperature relationships. pp. 77-117 In: J. Chelkowski (ed.). Cereal Grain.
Mycotoxins, Fungi and Quality in Drying and Storage. Elsevier, Amsterdam.
Misra, A.K. and D. Vir. 1991. Assessment of losses due to discoloration of paddy
grains. Indian J. Mycol. Plant Pathol. (India) 21(3):277-278.
Phillips, S., S. Widjaja, A. Wallbridge, and R. Cooke. 1988. Rice yellowing dur-
ing post-harvest drying by aeration and during storage. J. Stored Prod. Res.,
24(3):173-181. Available at: http://dx.doi.org/10.1016/0022-474X(88)90015-X
Ranalli, R.P. and T.A. Howell. 2002. Effects of storage conditions on the total aero-
bic and yeast/mold bacterial counts of rough rice during on-farm storage. J. of
Food Sci. 67(2):807-810.
Richard, J.L., G.E. Payne, A.E. Desjardins, C. Maragos, W.P. Norred, and J.J. Pestka.
2003. Mycotoxins: risks in plant, animal and human systems. CAST Task Force
Report, 139:101-103.
Sahay, M.N. and S. Gangopadhyay. 1985. Effect of wet harvesting on biodeteriora-
tion of rice. Cereal Chem. 62(2):80-83.
Singaravadivel, K. and S.A. Raj. 1983. Changes in parboiled rough rice caused by
improper drying and microbial infection. J. Food Biochem. 7(1):15-21.
Skyrme, D.S., B.P. Marks, M.G. Johnson, and T.J. Siebenmorgen. 1998. Distribution
of total aerobic and coliform bacterial counts among rice kernel components. J.
Food Sci, 63(1):154-156.
338
Table 1. Conceptual illustration of the

experiment design. The experiment was set up as
a full-factorial design with Week 0 and moisture content
= 12.5 % (wet basis) serving as the control conditions.
Moisture Storage
Cultivar content Temperature duration
(% wet-basis) [F (C)]
CLXL745 12.5 45 (10) Week 0
XL753 16 59 (15) Week 2
19 68 (20) Week 4
21 80.6 (27) Week 6
104 (40) Week 8
Week 10
Week 12
Week 16
Table 2. Difference between percent moisture

content (MC%) values set during the design stage
and the actual, experimentally measured MC% values.
Actual moisture content
Set value XL745 XL753
--------------(MC %)-------------
12.5 12.7 12.3
16 16.1 17.4
19 18.8 20.3
21 21.1 20.8
Table 3. Summary of the P values from the

analysis of variance for the effect of week number, temperature,
and percent moisture content (MC%) by cultivar on mold counts.
Prob > F
Factor dfa CLXL 745 XL753
Week number 7 0.7124 0.0006
Temperature 4 0.1387 0.4442
Week number temperature 28 0.1465 0.754
MC % 3 <.0001 <.0001
Week number MC % 21 <.0001 <.0001
Temperature MC % 12 0.0001 <.0001
Week number temperature MC % 84 <.0001 <.0001
a
df = degrees of freedom.
339
Fig. 1. Contour plots of mold counts [colony forming units (log10 CFU)] as
influenced by the interaction between storage duration (week number), percent moisture
content (MC%) and storage temperature for cultivar XL745 (a) and XL753 (b), respectively.
340
Fig. 2. Relationship between mold counts (log10 CFU/g) and

percent moisture (MC%) content for the two cultivars, CLXL745 and XL753.
Means followed by different letters are significantly different (P 0.05) from each other.
341
Simulation and Validation of On-Farm

In-Bin Drying and Storage of Rough Rice
G. G. Atungulu1 and H. Zhong1
ABSTRACT
The objectives for this study were to: (1) determine accurate models for predicting equi-
librium moisture content (EMC) of rough rice at set conditions of air temperature and
relative humidity (RH); and (2) validate developed mathematical model for predicting
rough rice moisture content (MC) and temperature profiles during natural air (NA), in-
bin drying and storage. Adsorption and desorption isotherms of long-grain hybrid rough
rice cultivar, Clearfield (CL) XL745, at temperatures ranging from 15 C to 35 C and
RHs of 10% to 90% were determined by using Dynamic Vapor Sorption analysis equip-
ment. Non-linear models were used to determine constants of models for predicting the
rough rice adsorption and desorption EMCs. It was determined that the best model to
describe the studied rough rice adsorption isotherms was the modified Halsey equation
(RMSE = 0.54% MC dry basis), while the modified Chung-Pfost equation (RMSE =
0.91% MC dry basis) was best to describe desorption isotherms. The updated rough rice
EMC prediction equations were incorporated into an equilibrium-based finite difference
model used to simulate temperature and MC of in-bin rough rice drying at two selected
rice growing locations in Arkansas. The model was validated using field experiments
that used modern, on-farm bins equipped with sensors for in-bin RH and rough rice
temperature measurements. The rough rice MC was calculated using the equilibrium
models with inputs of measured RH and temperature data. Analyses were conducted
to compare sensor-determined rough rice MC and temperature data to that determined
from both the simulation model and from laboratory moisture meter-measurements.
The simulation results described the general trends of rough rice sensor-determined
MC and temperature profiles well (for MC, mean RMSE = 0.56% MC on a wet basis;
for temperature, mean RMSE = 1.77 C). The rough rice MC data determined by the
sensors overpredicted the meter-measured MCs by 4.54% and 7.60%, and the RMSEs
were 1.48% MC and 0.73% MC for drying bins located at Burdette and Dermott, Ar-
kansas, respectively. The study generated useful information for predicting rough rice
MC and temperature during NA in-bin field drying and storage.
1
Assistant Professor and Graduate Student, respectively, Department of Food Science, Fayetteville.
342
INTRODUCTION
In-bin drying and storage of rough rice using natural air (NA), if not managed
properly, is prone to contamination of rough rice with mycotoxins (e.g., aflatoxin),
posing significant public health risks and reducing overall rice quality. The NA drying
method involves use of a fan (often more than one) to mechanically push ambient air
through a rough rice column, from the bottom to the top of the bin. As the air moves
vertically through the rough rice inside the bin, the air quality determines whether
the rough rice gains or loses moisture. The air quality, also referred to as the equilib-
rium moisture content (EMC), defines the capability of the rough rice to hold moisture
at set conditions of air temperature and relative humidity (RH). Depending on local
weather conditions, the duration required for NA, in-bin drying may not be conducive
for timely and complete drying, especially for upper layers of rough rice (Atungulu et
al., 2015). It is important to know the extent to which drying with NA in a particular
location influences rough rice drying duration to mitigate rice quality reduction, mold
growth, and development of mycotoxins.
A field study to obtain NA, in-bin drying kinetics for rough rice would require
extensive, time-consuming, and costly experimentation. However, an accurate math-
ematical model may simplify prediction of the NA, in-bin drying process of rough
rice. There is, therefore, a critical need to develop and validate an accurate model that
could be used to simulate NA, in-bin drying of rough rice and the effects on rice qual-
ity. The model could be used for simulations to provide suitable drying conditions for
rough rice at different geographical regions. In the absence of such a model to predict
suitable drying strategies, in-bin rough rice drying and storage with NA will be more
susceptible to grain quality loss and contamination with toxigenic fungi and their as-
sociated mycotoxins, many of which are carcinogenic to humans.
The objectives for this research were to use developed mathematical models to
simulate NA, in-bin drying and storage of rice and to validate the results. Specifically,
the study: (1) determined accurate models for predicting EMC of rough rice at set
conditions of air temperature and RH; (2) incorporated the new EMC models into the
developed mathematical models for NA, in-bin drying of rice; and (3) validated the ac-
curacy of the new models to predict moisture content (MC) and temperature profiles of
rough rice during NA, in-bin drying at different Arkansas locations by field experiments.
PROCEDURES
Moisture Content Determination
Long-grain rice cultivar Clearfield (CL) XL745, grown at Newport, Ark., was
harvested in fall of 2014 at 16% MC. Immediately after harvest, rough rice samples
were placed in sterile polyethylene bags, sealed, and stored in a cooler set at 4 C. The
samples were stored in the cooler and retrieved later for experiments. To determine
the initial moisture content (IMC) of rice, samples from the bags were retrieved and
allowed to equilibrate at room conditions, and then 15 g of the sample was placed into
a conductive oven set at 130 C (Shellblue, Sheldon Mfg., Inc., Cornelius, Ore.). The
343
sample was kept in the oven for 24 hours, after which it was removed, placed in the
desiccator, and allowed to cool for at least one half hour (Jindal and Siebenmorgen,
1987). The adsorption and desorption isotherms of the rough rice samples were de-
termined using Dynamic Vapor Sorption (DVS) analysis equipment [AquaLab Vapor
Sorption Analyzer (VSA), Decagon Devices, Inc., Pullman, Wash.].
Rough rice MC during desorption and adsorption versus temperature and water
activity (equilibrium RH in decimal) were analyzed using nonlinear regression in JMP
Pro v. 12.1.0 (SAS Institute, Inc., Cary, N.C.) to estimate the empirical constants (A, B,
and C) of the four rough rice EMC models in Table 1. The main effects of all variables
on EMC and statistical parameters were determined using Fit Model JMP Pro v. 12.1.0
(SAS Institute, Inc., Cary, N.C.), and statistical significance (P) was set to 0.05. Excel
software (Microsoft Office 2013, Microsoft Corp., Redmond, Wash.) was used to calcu-
late the Root Means Square error (RMSE), Nash-Sutcliffe efficiency (NSE), and Percent
bias (PBIAS) which were used to evaluate the relationships among meter-measured
sample MCs, the sensor-determined MCs, and MCs obtained from simulation output.
Development of Computer Simulation Platform and Mathematical Simulations

A computer simulation program based on the Post-harvest Aeration Simulation
Tool - Finite Difference Model (PHAST-FDM), written in Visual Basic.NET (Micro-
soft Corp., Redmond, Wash.), was modified for use with rough rice (Lawrence et al.,
2015). Modifications of the program included development of an interactive Graphical
User Interface (GUI), incorporation of updated EMC models of rough rice, addition of
functions to calculate percent overdrying and rice dry matter loss (DML), and addition
of multiple options for controlling when to end the simulations (Fig. 1). The modified
program was used to simulate NA, in-bin drying of rough rice at representative rice-
growing locations.
The energy balance applied to a thin rice layer was determined as described by
Jindal and Siebenmorgen (1994):
caTo + Ho (hv + cvTo ) + cgGor + cwGo (hf Ho ) = caTf + Hf (hv + cvTf ) + cgTf r Eq. 1
where, ca is specific heat of dry air (J/kg of dry air/K); cg is specific heat of grain (J/kg
of wet grain/K); cv is specific heat of water vapor (J/kg of water vapor/K); cw is specific
heat of water (J/kg of water in grain/K); Ho is absolute humidity of air entering the
control volume (kg of water/kg of dry air); Hf is absolute humidity of air leaving the
control volume (kg of water/kg of dry air); To is initial air temperature (C); Tf is final
air and grain temperature (C); hv is latent heat of vaporization of water (J/kg of water
vapor); Go is initial grain temperature (C); and r is grain mass to dry air ratio (kg of
wet grain/kg of dry air). On the left side of Eq. 1, the first term represents the energy of
the dry air with respect to 1 kg of dry air before air entering the rough rice layer. The
second term represents the energy of the water vapor in the air with respect to 1 kg of
dry air before air entering the rough rice layer. The third term represents the energy of
the wet rough rice layer (percent MC) with respect to 1 kg of dry air before air entering
the rough rice layer. It is assumed that the change of the amount of water in the rough
344
rice is equal to the change of humidity in the air; therefore, the fourth term represents
the energy difference of the water which is desorbed or adsorbed by the rough rice with
respect to 1 kg of dry air. The three terms on the right side of Eq. 1 correspond to the
first three terms of the left side of the equation, and they are for conditions after the air
exits the rough rice layer.
The moisture balance applied to a thin rice layer, with the assumption that the
mass of water evaporated from the rough rice in the layer is equal to the change in
mass of water vapor in the air passing through the layer, was determined as described
by Jindal and Siebenmorgen (1994):
Hf Ho = (MCo MCf ) r / 100 Eq. 2
r = (g dx ) / a va t Eq. 3
where, MCo is initial MC of grain in percentage wet basis; MCf is final MC of grain in
percentage wet basis; t is time interval (s); a is density of air (kg of air/m3); g is density
of grain (kg of wet grain/m3); va is velocity of air (m/s); and dx is layer thickness (m).
Using the exiting air RH and temperature, the rough rice MC in each layer at
the end of a time step was determined. The modified Chung-Pfost equation Eq. 4 with
constants specified for long-grain hybrid rough rice in adsorption and desorption con-
ditions was used:
1 (T + C) ln (RH)
MCe ln Eq. 4
B A
where, MCe is equilibrium moisture content of rough rice in percentage dry basis; RH
is relative humidity in decimal; T is temperature (C); and A, B, and C are empirical
constants.
Adsorption and desorption constants (two sets of A, B, and C) for Eq. 4 were
determined in this study for long-grain hybrid rice cultivar (CLXL745). Relative hu-
midity, temperature, and partial and saturated vapor pressures of air were determined
using (ASABE Standards, 2014):
Pv
RH = Eq. 5
Ps
101325 H
Pv = Eq. 6
0.6219 + H
A + BT + CT2 + DT3 + ET4
Ps = K exp ( ) Eq. 7
FT GT2
where, RH is air relative humidity in decimal; Pv is partial vapor pressure of air (Pa); Ps
is saturated vapor pressure of air (Pa); H is absolute humidity of air (kg of water/kg of
dry air); T is air temperature (C); K = 22105649.25; A = -27405.526; B= 97.5413; C =
-0.146244; D = 0.12558 10-3; E = -0.48502 10-7; F = 4.34903; G = 0.39381 10-2.
345
Using Eqs. 1 to 7, the temperature and RH of air exiting each rough rice layer
and the rough rice temperatures at the end of each time step were determined. A finite
difference method was used to determine the time-step solutions of rough rice MC and
temperature as well as the intake and exit air conditions in successive layers of the rough
rice inside the bin. The rough rice inside the bin was divided into N thin layers (N = 20).

Adsorption and desorption isotherms of long-grain hybrid rough rice cultivar
CLXL745 were determined (Table 2). The result showed the average rough rice
desorption EMC was higher than the average adsorption EMC, and rough rice EMC
significantly increased when air temperature decreased, or air RH increased. Four em-
pirical EMC models (modified Chung-Pfost, modified Halsey, modified Henderson,
and modified Oswin) were fitted with actual data, and constants for predicting both
adsorption and desorption EMCs were established (Table 3). The modified Halsey and
modified Chung-Pfost equations best predicted EMCs of long-grain hybrid rough rice
for adsorption and desorption conditions, respectively (Table 3). The modified Chung-
Pfost equation and associated adsorption and desorption constants were selected for
NA, in-bin drying simulations and model validation performed in this study.
Experiments using on-farm, in-bin drying systems, equipped with sensors for
automatic monitoring of grain temperature and RH in the bin, were performed to vali-
date the developed model for NA, in-bin drying of rough rice in Arkansas locations
(Table 4). Rough rice MC data determined by field sensors overpredicted the meter-
measured MCs by 4.54% and 7.60%, and the root mean square errors (RMSEs) were
1.48% MC and 0.73% MC for drying bins located at Burdette and Dermott, Arkansas,
respectively (Table 5).
Simulation results of rough rice MC and temperature from the developed model
were compared with field sensor-determined profiles for rough rice drying with equi-
librium moisture content natural air (EMC-NA) fan control strategy; the mean RMSEs
of rough rice MC and temperature were less than 0.57 0.10% MC and 1.91 0.21 C,
respectively (Table 6); mean Nash-Sutcliffe efficiencies (NSE) of rough rice MC and
temperature were greater than 0.68 0.50 and 0.49 0.04, respectively; percent biases
(PBIAS) range of rough rice MC were between -3.53% and 5.76%; PBIAS range of
rough rice temperature was between -3.96% and 1.78%. These statistical parameters
indicated that the simulated rough rice MC and temperature profiles reasonably predicted
the rough rice temperature and MC data, determined using field sensors.
The study validated models for NA, in-bin drying of rough rice, thereby providing
new tools for understanding the kinetics of rough rice MC and temperature during on-
farm, in-bin drying and storage using natural air; such information is critical to further
prediction of rice quality and safety.
346
ACKNOWLEDGMENTS
The authors wish to acknowledge Chandra Singh of OPISystems Inc., (Calgary,
Canada) and Mike Shook of Agri Process Innovations Inc. (Stuttgart, Ark.) for very
helpful collaboration during field experiments. The authors also wish to thank rice
producers in the Arkansas Delta and White River ecological zones, especially Mike
Sullivan, Steven Henderson, Sam Nicholson, Paul McKinney, Donnie Jacobs, and
Benny Petrus for collaborations during these research efforts. The authors also thank
the Grain Processing Engineering group at the University of Arkansas, Food Science
Department, and G.S. Osborn, A. Mauromoustakos, and T.J. Siebenmorgen of the
University of Arkansas System Division of Agriculture for support and collaboration.
Lastly, the authors greatly appreciate the Arkansas Rice Research and Promotion Board
and the University of Arkansas System Division of Agriculture for support and partially
funding this research project.
LITERATURE CITED
ASABE Standards. 2014. D271.2 Psychrometric Data. St. Joseph, Mich.: ASABE.
Atungulu, G.G., H. Zhong, S. Thote, A. Okeyo, and A. Couch. 2015. Microbial prev-
alence on freshly-harvested long-grain pureline, hybrid and medium-grain rice
cultivars. ASABE, Appl. Engin. J. 31(6):949-956, DOI 10.13031/aea.31.11216.
Jindal, V.K. and T.J. Siebenmorgen. 1987. Effects of oven drying temperature
and drying time on rough rice moisture content determination. Trans. ASAE,
30(4):1185-1192.
Jindal, V. K. and T.J. Siebenmorgen. 1994. Simulation of low temperature rough rice
drying and rewetting in shallow beds. Trans. ASAE, 37(3):863-871.
Lawrence, J., G. Atungulu, and T. Siebenmorgen. 2015. Modeling in-bin rice drying
using natural air and controlled air drying strategies. Trans. ASABE, 58(4):1103-
1111.
347
Table 1. Moisture sorption prediction models according to American

Society of Biological and agricultural engineers standard D245.6 (2012).
Name of model Equation
1 (T + C) ln (RH)
Modified Chung-Pfost MC 100 = ln
B A
1
exp(A + B T) C
Modified Halsey MC 100 =
ln (RH)
1
Modified Henderson ln (1 RH) B
MC 100 =
A (T + C)
1
Modified Oswin
MC 100 = (A + B T)
1 RH
RH
( ) C
MC is air equilibrium moisture content in decimal dry basis, RH is relative humidity expressed
as a decimal and T is temperature (C). The terms A, B, and C are empirical constants.
Table 2. Equilibrium moisture contents of long-grain hybrid

rice cultivar CLXL745 at temperatures ranging from 15 C to 35 C and
relative humidities ranging from 10% to 90% in desorption and adsorption conditions.
Temperature (C)
Relative Adsorption EMC Desorption EMC

humidity 15 25 35 15 25 35
(%) --------------------------------------- (% w.b.)----------------------------------------
10 11.9 a B 10.0 ab D 9.0 b D 12.4 a D 9.9 b E 9.1 b E
30 12.3 a B 10.5 ab D 9.5 b D 13.9 a C 12.3 b D 10.8 c D
50 13.3 a B 12.5 a C 12.0 a C 16.3 a B 15.0 ab C 14.0 b C
70 16.1 a A 15.8 a B 15.6 a B 18.8 a A 18.5 a B 17.3 a B
90 22.6 a A 22.3 a A 22.6 a A 22.3 a A
EMC = equilibration moisture content.
% w.b.= percent wet basis.
For each sorption type and temperature combination, values within individual columns followed
by the same lowercase letter are not significantly different.
For each sorption type and relative humidity combination, values within individual rows fol-
lowed by the same uppercase letter are not significantly different.
348
Table 3. Constants, root mean square errors (RMSEs), and sums of RMSE (SRMSEs) of Modified Chung-Pfost, Modified Halsey,
Modified Henderson, and Modified Oswin equations used for predicting desorption and adsorption isotherms of long-grain rough rice.
Model constants
Model Sorption type A B C RMSE SRMSE
---------- (% MC d.b.)---------
Modified Chung-Pfost Desorption 861.4998 0.2382 9.6002 0.54 1.86
ASABE 579.9479 0.1894 52.1784 0.61
ASABE 412.02 0.17528 39.016 0.26
Adsorption 1086.5411 0.2454 29.8549 1.32
ASABE 590.3090 0.1936 55.2229 0.78
Modified Halsey Desorption 10.9302 -0.0320 3.9053 0.86 1.77
ASABE 8.1726 -0.006112 3.3578 1.66
Adsorption 9.0526 -0.0183 3.4783 0.91
ASABE 8.3821 -0.01029 3.3874 1.45
Modified Henderson Desorption 8.0518 10-7 3.5691 21.4187 2.34 4.34
ASABE 2.9747 10-5 2.1177 79.1993 0.51
ASABE 4.1276 10-5 2.1191 49.828 0.29
Adsorption 8.9734 10-7 3.3119 93.1951 2.00
ASABE 3.3900 10-5 2.1226 73.0724 0.69
Modified Oswin Desorption 17.9920 -0.1104 5.2319 0.79 2.11

ASABE 13.1323 -0.02873 3.9156 1.20
ASABE 14.431 -0.07886 3.137 0.36
Adsorption 15.0560 -0.0560 4.7226 1.32
ASABE 13.4692 -0.04410 3.9542 1.10

% MC d.b. = percent moisture content on a dry basis.

Constants generated using JMP 12.0.0 based on the experimental result of this study for predicting desorption and adsorption isotherms of long-
grain hybrid rough rice (cv. CLXL745).

Constants generated using JMP 12.0.0 based on the equilibrium moisture content data listed in American Society of Agricultural and Biological
Engineers (ASABE) standard D245.6 (2012) for long-grain rough rice (cv. Inga) in desorption and adsorption conditions.

Constants listed in American Society of Agricultural and Biological Engineers (ASABE) standard D245.6 (2012) for long-grain rough rice.
349
Table 4. Configurations and drying strategies of bins A, B, C, and D.
350
Bin A Bin B Bin C Bin D
Bin location Burdette, Ark. Dermott, Ark. Dermott, Ark. Dermott, Ark.
Sensor locations Center and side Center and side Center and side Center and side
Bin diameter 14.63 m (48 ft) 14.63 m (48 ft) 14.63 m (48 ft) 14.63 m (48 ft)
The top-most sensor depth Side: 6.1 m (20 ft) Side: 6.1 m (20 ft) Side: 6.1 m (20 ft) Side: 6.1 m (20 ft)
Center: 7.32 m (24 ft) Center: 7.32 m (24 ft) Center: 7.32 m (24 ft) Center: 7.32 m (24 ft)
Fan configuration 40 HP 1750 RPM centrifugal 40 HP 1750 RPM centrifugal 30 HP 1750 RPM centrifugal 30 HP 1750 RPM centrifugal
fan; 2 fans in parallel fan; 2 fans in parallel fan; 2 fans in parallel fan; 2 fans in parallel
Fan control strategy EMC-H EMC-NA EMC-NA EMC-NA
Target EMC 12.5% (w.b.) 12.5% (w.b.) 12.5% (w.b.) 12.5% (w.b.)
Targeted EMC range N/A 14.5% - 10.5% (w.b.) 13.5% - 10.5% (w.b.) 14.5% - 10.5% (w.b.)
Issues for simulation Targeted EMC limits changed One fan was off during the first N/A N/A
and validation during the drying period half of drying period
Simulation-start date and time N/A N/A 26 Sept. 2014 at 12:00 AM 03 Oct. 2014 at 4:00 AM
Simulation-end date and time N/A N/A 21 Oct. 2014 at 11:59 PM 27 Oct. 2014 at 11:59 PM
Initial rough rice N/A N/A Sensor6=20.44, Sensor6=26.33,
Temperature (C) Sensor5=20.39, Sensor5=26.17,
Sensor4=21.11, Sensor4=26.11,
Sensor1=22.83. Sensor1=24.94.
Initial rough rice N/A N/A Sensor1=15.32, Sensor1=14.74,
moisture content Sensor2=16.95, Sensor2=16.84,
(%.w.b.) Sensor3=16.67, Sensor3=16.38,
Sensor6=16.01. Sensor6=19.53.
Plenum pressure N/A N/A 1673.03 Pa (6.71 in H2O) 1637.07 Pa (6.57 in H2O)

EMC-H represents equilibrium moisture content controlled air with supplemental heat fan control strategy; EMC-NA represents equilibrium mois-
ture content controlled natural air fan control strategy.

w.b.=wet basis.

Sensor1 represented the first (bottom) layer, and Sensor2 represented the second layer MC which was 1.22 m (4 ft) above the first layer. There-
fore, Sensor6 was 7.32 m (24 ft) from the bottom of the bin.
Table 5. Sensor-determined moisture content (MC)

and meter-measured MC data for rough rice in bins A and B;
comparison of root mean square errors (RMSEs) and percent biases (PBIASs).
Rice in the bin was sampled 1.22 m (4 ft) below the top-most layer surface at the center.
In-bin sensor- Meter-
Sampling determined measured
Bin location date MC data MC data RMSE PBIAS
----------------------(%w.b.)--------------------- (%)
Bin A
Dermott, Ark. 12/10/2014 20.33 19.22 1.48 7.60
26/10/2014 20.62 19.22
9/11/2014 19.64 17.35
21/11/2014 14.76 14.24
Bin B
Burdette, Ark. 13/9/2014 16.33 15.53 0.73 4.54
27/9/2014 16.36 16.13
11/10/2014 13.83 12.72
25/10/2014 13.07 12.62
w.b. = wet basis.
Table 6. Statistical parameters [Root mean square

errors (RMSEs), Nash-Sutcliffe efficiencies (NSEs) and percent biases
(PBIASs)] used to evaluate model performance for bins C and D. (a) comparison
of moisture content calculated from sensor data of temperature and relative humidity
and simulation, (b) comparison of temperature determined by sensor and simulations.
Layer 1 2 3 4 5 6 Mean
(a) Moisture content comparison
Bin C
RMSE (% w.b. MC) 0.41 0.32 0.64 0.48 0.30 1.07 0.540.29
NSE 0.60 0.95 0.85 0.94 0.97 0.42 0.790.23
PBIAS (%) 0.46 -1.43 -3.31 -1.66 -0.73 5.76 2.232.00
Bin D
RMSE (% w.b. MC) 0.60 0.41 0.53 0.53 0.67 0.67 0.570.10
NSE -0.33 0.83 0.83 0.93 0.93 0.91 0.680.50
PBIAS (%) 3.22 -0.87 -1.98 -1.99 -3.53 -2.65 2.370.97
(b) Temperature comparison

Bin C
RMSE (C) 1.77 1.47 1.55 1.60 1.62 1.47 1.580.11
NSE 0.81 0.87 0.85 0.83 0.80 0.82 0.830.03
PBIAS (%) -1.46 0.06 1.78 0.12 0.17 -3.07 1.111.21
Bin D
RMSE (C) 2.28 2.04 1.83 1.82 1.82 1.69 1.910.21
NSE 0.56 0.48 0.46 0.50 0.50 0.45 0.490.04
PBIAS (%) -3.54 -3.96 -2.48 -2.85 -2.83 -2.07 2.960.69
%w.b. MC = percent wet basis moisture content.
Numbers represent the mean and the standard deviation.
Numbers represent the mean and the standard deviation of the absolute value of the PBIAS
(mean absolute PBIAS = Ni = 1 (Yi ) / N).
351
Fig. 1. User interface of a computer program, written in

Visual Basic.NET, for simulation of natural air drying of rough rice.
352
Optimization of Process Parameters in

Rough Rice Drying Using Industrial Microwave
G.G. Atungulu1, G.A. Olatunde1, D.L. Smith1, S. Sadaka2, and S. Rogers3
ABSTRACT
The use of industrial microwave (MW) at 915 MHz may have the potential to achieve
one-pass rough rice drying. However, it is vital to determine the optimal processing
parameters for continuous-flow drying operations that maximize drying throughput
and minimize milled rice quality reduction. Therefore, the objective for this study was
to quantify the effect of MW energy applied per unit mass of rough rice on moisture,
and milled rice quality reduction for a continuous one-pass MW drying operation of
rough rice. Freshly harvested, medium-grain rough rice with moisture content of 25%
wet basis (w.b.) was used for this study. A portion of the sample was used as control
and the remaining was dried in a pilot scale 915 MHz MW for up to 8 minutes drying
duration. The rough rice was heated at specific energies of 194, 258, and 322 Btu/lb
(450, 600, and 750 kJ/kg) at various rice bed thicknesses of 0.4, 1.2, and 2.0 inches
(0.01, 0.03, and 0.05 m). The results showed that moisture removed varied between 6
and 15 percentage points as specific energy increased. Also, increase in bed thickness
with increase in specific energy from 194 to 258 Btu/lb (450 to 600 kJ/kg) resulted in
reduction in milling yield from 73% to 62%. Head rice yield (HYR) was reduced from
66% to 37% as specific energy increased from 194 to 322 Btu/lb; there were marginal
reductions in measured rice fat (0.50% to 0.34%) and protein (6.13% to 5.60%) contents.
In conclusion, drying of medium-grain rough rice using a continuous-flow, one-pass
MW heating at a specific energy of 194 Btu/lb and bed thickness of 0.4 inches appeared
to be most the promising treatment of all the studied conditions.
INTRODUCTION
A convective air drying method is typically used to dry freshly harvested (rough)
rice to a safe moisture content (MC) usually 12% to 13% wet basis (w.b.) prior to stor-
age. However, formation of temperature gradient between the core and the surface of
1
Assistant Professor, Post Doctoral Associate, and Graduate Assistant, respectively, Department of Food
2
Assistant Professor / Extension Engineer Biosystems, Department of Bio and Agricultural Engineering,
Little Rock.
3
Vice President, AMTek Applied Microwave Technology, Inc., Cedar Rapids, Iowa.
353
the kernel causes moisture migration towards the surface of the kernel and could induce
tensile stress at the surface and compressive stress at the interior of the kernel (Fan et
al., 2000). Stress in the rice kernel causes rice fissuring and degradation of mechanical
properties, which ultimately is responsible for the kernels inability to withstand the
processes of hulling and bran removal without breaking. The milled rice is quantified
by the head rice yield (HRY) as percentage by mass of those kernels retaining three-
fourths or more of their original length, to the original rough rice. Reduction in HRY
during milling operation results in economic loss to the farmer. Hence, there is need to
explore methods that can be used for drying rough rice without causing reduction in
HRY. Microwave (MW) heating may reduce stresses caused by temperature and MC
gradients within the rice kernel and potentially improve the rice milling yield because
the technology uses volumetric heating method.
Microwave is a form of electromagnetic energy with the frequency range of 300
MHz to 300 GHz and the corresponding wavelengths of 1 mm and 1000 mm (Oghbaei
and Mirzaee, 2010). Microwave drying has a volumetric heating characteristics which is
unlike the convective heating method; this leads to accelerated increase in temperature
at the interior of the kernel (Datta and Davidson, 2000; Gowen et al., 2006; Vadivam-
bal and Jayas, 2007). Microwave drying may offer many advantages over convective
drying methods under similar conditions such as high thermal efficiency, and shorter
drying time (Ren and Chen, 1998). Heat losses to the surrounding air are much lower
since heat generated occurs primarily inside the product. Since energy is not consumed
in heating the walls of the apparatus or the environment, operational cost is lower and
floor space requirement compared to conventional driers with similar plant capacity
is significantly lower (Bouraoui et al., 1994; Chandrasekaran et al., 2013; Jiao et al.,
2014; Zhang et al., 2006).
The most commonly used MW energy frequencies for drying/heating purposes
are 915 MHz and 2.45 GHz (Oghbaei and Mirzaee, 2010). Domestic MW appliances
operate at a frequency of 2.45 GHz, while industrial MW systems typically operate
at frequencies of 915 MHz. Industrial MW produces a longer wavelength with more
penetrating power than domestic MW (Chandrasekaran et al., 2013). Despite the ad-
vantages of MW heating technology, there is no commercial use of MW technology for
rice drying in the United States. Few published work on MW rice drying that use 2450
MHz have been reported (Kaasova et al., 2002; Le et al., 2014). Scaling up MW system
that operate at 2450 MHz has been a challenge due to associated low penetration depth,
non-uniformity of heating, and energy inefficiencies. On a commercial drying level,
there is a need to investigate the impact of using industrial MW on rice milling yield,
and functional quality indices. The objective for this research was to determine the fea-
sibility of using an industrial-type MW heating system operated at 915 MHz frequency
to achieve one-pass rough rice drying with minimum implications on the rice quality.
PROCEDURES
Materials
Freshly harvested medium-grain rough rice (cv. Jupiter), grown in the 2014 rice
crop season at Cash, Ark., was procured at initial MC of 23% to 24% w.b. and used for
354
this study. The samples were cleaned using a dockage equipment (MCi Kicker Dock-
age Tester, Mid-Continent Industries Inc., Newton, Kan.), and placed in an airtight tub
and stored in a laboratory cold room set at 39.2 F (4 C) (Koolco, Hialeah, Fla.). At
the beginning of the experiments, the samples were retrieved from the cold room, and
allowed to equilibrate at room conditions (77 F) for 24 h before the time of use. The
MCs were determined using grain moisture tester (AM 5200, Perten, Kurva, Sweden),
which was calibrated according to the American Society of Biological Engineers (AS-
ABE) standard (Jindal and Siebenmorgen, 1987). All reported MCs were on wet basis.
Microwave Heating Treatment

A 915 MHz industrial MW dryer (AMTek Microwaves, Cedar Rapids, Iowa) was
used for the experiment (Fig. 1). The system consists of a transmitter, a wave guide
and microwave heating zone (oven). The system transmitter is a high-powered vacuum
tube that works as a self-excited microwave oscillator converting high-voltage electric
energy to microwave radiation. The system waveguide consists of a rectangular pipe
through which the electromagnetic field propagated lengthwise; it was used to couple the
microwave power from the magnetron into the lab oven. The oven is the internal cavity
of the microwave unit where the product was placed to provide uniform temperatures
throughout and while in use.
Drying Test and Procedure

A total of 27 drying runs were conducted at three specific energy levels of 194,
258, and 322 Btu/lb, bed thickness of 0.4, 1.2, and 2.0 in., and 480 sec drying duration.
The choice of the specific energy levels was adopted based on our previous work on
microwave batch drying of rice where we reported that 258 Btu/lb was found to be
optimum in a batch process that also incorporated tempering of rice after MW heating
as an additional step. For a particular experimental run, the appropriate sample size
was weighted and fed into the hopper in which the outlet was pre-set at the expected
bed thickness and the corresponding power (using Eq. 1 and Table 1).
7.5ME
P= Eq. 1
t
Where: P is the power applied (Btu/h), M is the mass (lb), E is the specific energy (Btu/
lb) and t is the duration of heating (h).
As samples were conveyed out of the heating chamber to the collection point, three
different subsamples weighing 4.4 lb each were collected at the beginning, middle, and
toward the end of the bed. Samples collected were transferred into different glass jars
immediately and the jars were sealed and placed inside an incubator (VWR, Radnor,
Pa.) that was pre-set at a temperature of 104 F for 4 h. After tempering, the samples
were removed from the jars and spread on a flat wire mesh and allowed to gradually cool
down until surface temperatures dropped to ambient condition inside a controlled envi-
355
ronment with temperature and relative humidity of 79 F and 56%, respectively (5580A,
Parameter Generation and Control, Black Mountain, N.C.) (Jindal and Siebenmorgen,
1994). The MC of the cooled sample was determined. The percentage point moisture
removed before MW heating and after cooling was calculated as the difference between
the original and final MCs after cooling.
Rice Dehulling and Milling

A weight of 0.33 lb was measured from each sample and was de-hulled using a
laboratory Sheller (THU 35B, Satake Engineering Co., Tokyo, Japan) with a clearance
of 0.02 in. between the rollers. The resultant brown rice was milled (McGill No.2. RAP-
SCO, Brookshire, Texas) for 30 sec. The milled rice was quantified and then the brokens
were separated from the lots with a sizing device (61 Grain Machinery manufacturing
Corp., Miami, Fla.) and the resulting mass of head rice was quantified. The white rice
(both broken and rough remaining) was clearly marked for subsequent analysis.
Milling Yield and Head Rice Properties

The milled rice from the treatment described above was then separated into the
head rice out of the milled fraction using a double-tray shaker table (Grain, Grain Ma-
chinery Corp., Miami, Fla.) with both trays having indented holes (3.96 for medium-
grain). The holes are used to separate the broken kernels from the head rice. Hence,
HRY was calculated as the weight percentage of rough rice that remained as head rice.
All HRY determinations were replicated and reported as average values.
Statistical Analysis
All experiments were conducted in triplicate. The results are presented in relevant
sections as mean values and standard deviation. Analysis of variance was conducted
using a generalized linear model and surface response methodology from SAS statisti-
cal software (SAS Institute Inc., Cary, N.C.). In addition, a pairwise t-test using least
square mean with Tukey multiple comparison test was computed for each effect and
the interactions. Significant difference for interaction effects was separated by letter
reporting. All tests were considered to be significant when P 0.05.

Effect of Microwave Treatment on Moisture Reduction
The moisture loss rates were 0.018, 0.033, and 0.027 g/g/sec for 0.4 in., 1.2 in.,
and 2.0 in. bed thickness, respectively. The moisture loss first increased then reduced
as bed thickness increased from 0.4 in. to 1.2 in. The moisture loss at 1.2 in. bed thick-
ness was highest with about 88% and 22% higher at 0.4 in. and 2.0 in. bed thickness,
356
respectively. Similarly, moisture loss increased from 6% to 15% as specific energy

increased from 194 to 322 Btu/lb. However, moisture loss at 258 and 322 Btu/lb were
not significantly different. The surface response plot of the effect of thickness and
specific energy on moisture loss is shown in Fig. 2.
Effect of Microwave Treatment on HRY Reduction

The HRY of controlled samples dried at 79 F, 56% relative humidity was found
to be 67.0%. Although, the HRY of the sample treated at 0.4 in. bed thickness with
specific energy of 194 Btu/lb was almost the same as the control, the difference was not
statistically significant. Generally, HRY reduced from 66% to 37% as specific energy
increased from 194 to 322 Btu/lb. From the response surface plot (Fig. 3), it can be
seen that the linear relationship occurred between HRY and specific energy applied;
this was in agreement with the findings of Wadsworth (1993). An increase in specific
energy increased the intensity of MW power and increased the rate of moisture removal
and thermal stress in the kernel. This resulted in rice fissuring and reduction in HRY.
Effect of Microwave Treatment on Milling Yield

Increasing bed thickness and specific energy resulted in a reduction in milling
yield from 73% to 62%. Milling yield reduced because of the brittleness of kernel as
a result of moisture removal and thermal stress. Both MRY and HRY are highly de-
pendent upon the physical condition of rough rice kernels after drying. The effect of
specific energy and bed thickness were significant on the milling rice yield (R2 = 0.95,
RMSE = 2.5, Table 2).
The findings from this study show that freshly harvested rough rice can be dried
using microwave with minimal reduction in HRY and MRY. It was observed that treat-
ment at 194 Btu/lb with bed thickness of 0.4 in. resulted in dried rice with optimum
MRY and HRY.
ACKNOWLEDGMENTS
The authors acknowledge the support of the University of Arkansas System Divi-
sion of Agriculture, Grain Processing Engineering and Rice Processing Program at the
University of Arkansas, Food Science Department, and Applied Microwave Technology
Inc., for supporting this research.
357
LITERATURE CITED
Bouraoui, M., P. Richard, and T. Durance. 1994. Microwave and convective drying
of potato slices. J. Food Process. Engineering 17(3):353-363.
Chandrasekaran, S., S. Ramanathan, and T. Basak. 2013. Microwave food process-
ing-A review. Food Res. Int. 52(1):243-261.
Datta, A.K. and P.M. Davidson. 2000. Microwave and radio frequency processing. J.
Food Sci. 65(s8):32-41.
Fan, J., T. Siebenmorgen, and W. Yang. 2000. A study of head rice yield reduction
of long- and medium-grain rice varieties in relation to various harvest and drying
conditions. Trans. ASAE 43(6):1709.
Gowen, A., N. Abu-Ghannam, J. Frias, and J. Oliveira. 2006. Optimisation of de-
hydration and rehydration properties of cooked chickpeas (Cicer arietinum L.)
undergoing microwave-hot air combination drying. Trends Food Sci. Technol.
17(4):177-183.
Jiao, A., X. Xu, and Z. Jin. 2014. Modelling of dehydration-rehydration of instant rice
in combined microwave-hot air drying. Food Bioproducts Proc. 92(3):259-265.
Jindal, V., and T. Siebenmorgen. 1987. Effects of oven drying temperature and dry-
ing time on rough rice moisture content determination. Trans. ASAE 30(4):1185-
1192.
Jindal, V., and T. Siebenmorgen. 1994. Effects of rice kernel thickness on head rice
yield reduction due to moisture adsorption. Trans. ASAE 37(2):487-490.
Kaasova, J., P. Kadlec, Z. Bubnik, B. Hubackova, and J. Prihoda. 2002. Physical
and chemical changes during microwave drying of rice. Chemical Papers-Slovak
Academy of Sciences 56(1):32-35.
Le, T.Q., S. Songsermpong, P. Rumpagaporn, A. Suwanagul, and S. Wallapa. 2014.
Microwave heating for accelerated aging of paddy and white rice. Aust. J. Crop
Science 8(9):1348.
Oghbaei, M. and O. Mirzaee. 2010. Microwave versus conventional sintering: a
review of fundamentals, advantages and applications. J. Alloys Compounds
494(1):175-189.
Ren, G. and F. Chen. 1998. Drying of American ginseng (Panax quinquefolium)
roots by microwave-hot air combination. J. Food Engineering 35(4):433-443.
Vadivambal, R. and D.S. Jayas. 2007. Changes in quality of microwave-treated agri-
cultural products-a review. Biosystems Engineering 98(1):1-16.
Wadsworth, J.I. 1993. Microwave-vacuum drying. Food Sci. Technol. Marcel Dekker
Inc., New York.
Zhang, M., J. Tang, A.S. Mujumdar, and S. Wang. 2006. Trends in microwave-
related drying of fruits and vegetables. Trends Food Sci. Technol.17(10):524-534.
358
Table 1. Experimental design.

Specific energy Bed thickness Total mass Power
[Btu/lb (kJ/kg)] [In. (m 10-3)] [lb (kg)] [Btu/h (kW)]
194 (450) 0.4 (1) 211.6 (96.08) 10236.426 (3)
1.2 (3) 635.5 (288.26) 30709.278 (9)
2.0 (5) 1059.1 (480.43) 51182.13 (15)
258 (600) 0.4 (1) 211.6 (96.08) 13648.568 (4)
1.2 (3) 635.5 (288.26) 40945.704 (12)
2.0 (5) 1059.1 (480.43) 68242.84 (20)
322 (750) 0.4 (1) 211.6 (96.08) 17060.71 (5)
1.2 (3) 635.5 (288.26) 51182.13 (15)
2.0 (5) 1059.1 (480.43) 85303.55 (25)
Table 2. Statistical analysis of the effect of

microwave treatments on the milled rice yield of rough rice.
Mean square
Factors DF Type I SS error F Value Pr > F
Specific energy (Btu/lb) 2 146.91 73.45 11.32 0.0012
Thickness (inches) 2 1827.63 913.81 140.88 <0.0001
Power (Btu/h) 2 110.17 55.08 8.49 0.0038
Fig. 1. Industrial-type microwave system used in the study showing the

transmitter (1), wave guide (2), heating zone (3), conveyor belt (4), and control panel (5).
359
Fig. 2. Surface response of the effect of

microwave treatments on percentage point moisture content (MC)
reduction of rough rice with initial moisture content at 24% to 25% wet basis.
360
Fig. 3. Surface response for head rice

yield (HRY) at different thickness and specific energy.
361
Impacts of Rough Rice Temperature and

Moisture Content on Laboratory Milling Yields
B.C. Grigg1, C.D. Shook1, and T.J. Siebenmorgen1
ABSTRACT
Milling yields of a rice lot are often estimated using a laboratory-scale mill and a set
milling duration. Temperature and moisture content of the rice sample can affect the
milling results from such mills. A study was conducted to illustrate the impacts and
potential interactions of rough rice temperature (40 F, 55 F, 70 F, and 85 F) and
moisture content (MC; 10%, 12%, and 14% MC; wet basis, w.b.) at the time of milling;
measurements comprised milled rice yield (MRY), head rice yield (HRY), and surface
lipid content (SLC, representing the degree of milling), for lots of the long-grain XL753
and the medium-grain Jupiter rice. All samples were milled using a McGill No. 2 mill
for 30 s. For the XL753 lot, increasing rough rice temperature and/or MC resulted in
decreased MRYs, HRYs, and SLCs. For the Jupiter lot, increasing rough rice temperature
and/or MC decreased SLCs; however, only increasing rough rice MC levels resulted in
significantly decreased MRYs and HRYs. No significant interactions between rough rice
temperature and MC were observed with respect to MRYs or HRYs of either rice lot.
Compared to rough rice temperature, rough rice MC had a greater impact on milling
yields within each rice lot. These data support reporting of milling yields at a consistent
SLC to equitably valuate rice lots of different temperatures, MCs, and cultivars.
INTRODUCTION
The economic value of rough rice (Oryza sativa L.) is largely determined by milled
rice yield (MRY)the mass fraction of unprocessed, rough rice that remains as milled
rice, including both head rice and broken kernelsand head rice yield (HRY)the
mass fraction of rough rice that remains as head rice, defined as well-milled rice kernels
three-fourths or more of the original kernel length (USDA-FGIS, 2009). Well-milled
rice refers to the degree of milling (DOM), the extent of bran removal from brown rice
during milling. Increased DOM invariably increases the mass removed from rice kernels,
thus decreasing MRY (Wadsworth, 1994) and HRY (Lanning and Siebenmorgen, 2011).
1
Rice Quality Laboratory Manager, Student, and Distinguished Professor, respectively, Department of
Food Science, Fayetteville.
362
Surface lipid content (SLC, the amount of oil/lipid remaining on milled kernels) is often
used to represent DOM (decreasing SLC corresponds to increasing DOM), both in labo-
ratory operations (Cooper and Siebenmorgen, 2007) and commercial milling systems.
Many receivers of rough rice estimate milling yields of a rice lot using a labo-
ratory-scale mill and a set milling duration (e.g. 30 s), regardless of the rice lot being
evaluated. However, factors such as cultivar (Siebenmorgen et al., 2006), temperature
(Archer and Siebenmorgen, 1995), and moisture content (MC) (Reid et al., 1998) of
rough rice at the time of milling can affect the milling duration required to achieve
a desired SLC. Decreasing SLC (i.e. increasing DOM) results in decreased HRY of
long-grain rice cultivars (Cooper and Siebenmorgen, 2007; Lanning and Siebenmorgen,
2011). Inconsistent SLCs resulting from either over- or under-milling rice when milling
for a set duration, can in turn impact milling yields (Cooper and Siebenmorgen, 2007).
While studied independently, the potential interactions between rough rice temperature
and MC have not been reported. Thus, a study was conducted to illustrate the impacts
and potential interactions of rough rice temperature and MC at the time of milling on
milling yields and SLC.
PROCEDURES
One long-grain (XL753) lot and one medium-grain (Jupiter) lot were evaluated.
Both lots were cleaned with a dockage tester (XT4, Carter-Day, Minneapolis, Minn.),
and conditioned in a climate-controlled chamber (5580A, Parameter Generation &
Control, Black Mountain, N.C.) to one of three moisture contents (MCs), 10%, 12%,
or 14% (all 0.5% MC, wet basis). A moisture meter (AM5200, Perten Instruments,
Hgersten, Sweden) was used to measure MC. For each MC, sixteen 150-g samples
of rough rice from each lot were placed in zippered plastic bags, with four bags stored
at each of four temperatures (40 F, 55 F, 70 F, or 85 F) for 24 h. As such, the study
comprised 96 samples (2 cultivars 3 MCs 4 temperatures 4 replications).
Samples of rough rice were individually removed from their controlled-tem-
perature environments immediately before dehulling and milling. Each sample was
first dehulled using a laboratory sheller (THU 35B, Satake Corp., Hiroshima, Japan)
with a clearance of 0.019 in. between the rollers. Then each sample was milled for a
30-s duration in a laboratory-scale mill (McGill No. 2, RAPSCO, Brookshire, Texas),
equipped with a 3.3-lb weight on the lever arm, situated 6 in. from the milling cham-
ber centerline. Both MRY and HRY were determined, with head rice being separated
from broken kernels using a sizing device (61, Grain Machinery Manufacturing Corp.,
Miami, Fla.). Head rice DOM was quantified in terms of surface lipid content (SLC)
using a near-infrared-reflectance (NIR) spectrophotometer (DA7200, Perten Instruments,
Hgersten, Sweden) (Saleh et al., 2008).
Analysis of variance (P = 0.05), and means separation (Tukey-Kramer Honestly
Significant Difference), were conducted using statistical software (JMP v. 12.0, SAS
Institute, Inc., Cary, N.C.).
363

Increasing the rough rice temperature of the long-grain, XL753 lot from 40 F
to 85F decreased MRYs by 0.5 percentage points (pp) (Fig. 1a), and HRYs by 1.6
pp (Fig. 1c). Increasing the rough rice MC of the long-grain lot from 10% to 14% de-
creased MRYs by 2.2 pp (Fig. 1b), and HRYs by 6.3 pp (Fig. 1d). Similar to the trends
described for milling yields (Fig. 1a-d), SLCs of the long-grain lot decreased by 0.06 pp
in response to increasing rough rice temperature (Fig. 1e), and by 0.14 pp in response
to increasing rough rice MC (Fig. 1f). Thus, for the ranges of rough rice temperature
and MC levels evaluated, rough rice MC at the time of milling had a greater impact on
the magnitudes of MRYs, HRYs, and SLCs of the long-grain lot (Fig. 1a-f).
For the medium-grain, Jupiter lot, increasing rough rice temperature resulted in
significantly decreased SLCs (Fig. 2e). Increasing the rough rice temperature of the
medium-grain lot produced a slight trend of decreasing MRYs and HRYs (Fig. 2a and
c); however, there were no statistical differences in either MRYs or HRYs. As observed
for the long-grain lot (Fig. 1), increasing rough rice MCs resulted in significant, cor-
responding decreases in MRYs, HRYs, and SLCs of the medium-grain lot (Fig. 2b,
2d, and 2f). Thus for the medium-grain lot, rough rice MC again had a greater impact
on the magnitudes of MRYs when compared to rough rice temperature. Differences in
the responses of the two lots to changing rough rice temperature/MC levels illustrate
the effects of cultivar/type on milling characteristics, as previously shown between
long-grain cultivars (Siebenmorgen et al., 2006; Lanning and Siebenmorgen, 2011),
and between long-grain and medium-grain types (Pereira et al., 2008).
Only a single statistically significant interaction between rough rice temperature
and MC was observed; this for SLCs of the long-grain lot, decreasing greatly when
increasing rough rice temperature from 70 F to 85 F at the 10% MC level (Fig. 3a).
While not statistically significant, a similar trend for decreased SLC was observed with
the same temperature/MC combinations for the medium-grain lot (Fig. 3b). However,
for the rough rice temperatures and MCs evaluated, there were no significant interac-
tions with respect to MRYs of either rice lot.
At a set 30-s milling duration, increasing rough rice temperature and/or MC
decreased SLCs (i.e. increased DOM) of both long-grain and medium-grain lots, with
rough rice MC levels having a greater impact than rough rice temperature. While the
trends of MRY, HRY, and SLC were similar for the two lots, the rice cultivar/type affected
the overall magnitude of the response. Increasing either rough rice temperature or MC
decreased MRYs of the long-grain lot, agreeing with previous reports for long-grain
rice (Archer and Siebenmorgen, 1995; Reid et al., 1998). Similar to the long-grain lot,
increased rough rice temperature and/or MC resulted in trends for decreased MRYs of the
medium-grain lot; however, only increasing rough rice MC levels resulted in statistically
significant differences. These data illustrate cultivar-dependent milling characteristics,
agreeing with the report of Siebenmorgen et al. (2006), which recommended milling
364
to a consistent DOM for equitable comparison of HRYs across cultivar lots. Moreover,
these data suggest the need for milling to consistent DOM levels in response to different
levels of rough rice temperature and MC, within and across rice lots. The SLC values of
head rice samples can be rapidly measured using NIR instruments (Saleh et al., 2008),
allowing for subsequent mathematical adjustment of HRYs to an equitable DOM level
(Pereira et al., 2008). This readily accounts for different sample SLCs that may result
from variations in MC, temperature, or cultivar of rough rice samples.
ACKNOWLEDGMENTS
The authors recognize the support of the staff and corporate sponsors of the
University of Arkansas Rice Processing Program, the Arkansas Rice Research and
Promotion Board, and the University of Arkansas System Division of Agriculture.
LITERATURE CITED
Archer, T.R. and T.J. Siebenmorgen. 1995. Milling quality as affected by brown rice
temperature. Cereal Chem. 72(3):304-307.
Cooper, N.T.W. and T.J. Siebenmorgen. 2007. Correcting head rice yield for surface
lipid content (degree of milling) variation. Cereal Chem. 84(1):88-91.
Lanning, S.B. and T.J. Siebenmorgen. 2011. Comparison of milling characteristics of
hybrid and pureline rice cultivars. Appl. Eng. Agric. 27(5):787-795.
Pereira, T., N. Cooper, and T.J. Siebenmorgen. 2008. Effects of storage temperature
and duration on the milling properties of rice. Discovery. 9:64-74.
Reid, J.D., T.J. Siebenmorgen, and A. Mauromoustakos. 1998. Factors affecting the
head rice yield versus degree of milling relationship. Cereal Chem. 75(5):738-
741.
Saleh, M.I., J.-F. Meullenet, and T.J. Siebenmorgen. 2008. Development and valida-
tion of prediction models for rice surface lipid content and color parameters using
near-infrared spectroscopy: A basis for predicting rice degree of milling. Cereal
Chem. 85(6):787-791.
Siebenmorgen, T.J., A.L. Matsler, and C.F. Earp. 2006. Milling characteristics of rice
cultivars and hybrids. Cereal Chem. 83(2):169-172.
USDA-FGIS. 2009. United States Department of Agriculture Federal Grain Inspec-
tion Service. United States standards for rice. Washington, D.C.
Wadsworth, J.I. 1994. Degree of milling. pp. 139-176. In: W.E. Marshall and J.I.
Wadsworth, (eds.). Rice Science and Technology. New York, N.Y.: Marcel
Dekker, Inc.
365
Fig. 1. Impacts of rough rice temperature or moisture

content (MC) at the time of milling on milled rice yield (MRY) (a-b),
head rice yield (HRY) (cd)], and surface lipid content (SLC) of head
rice (e-f) of the long-grain (XL753) rice lot. Samples were milled with a McGill No. 2 mill
for 30 s. When considering temperature impacts, data are averaged across MCs. When
considering MC impacts, data are averaged across temperatures. Means with the same
letter were not statistically different; comparisons are valid only within a subgraph.
366
Fig. 2. Impacts of rough rice temperature or moisture

content (MC) at the time of milling on milled rice yield (MRY) (a-b),
head rice yield (HRY) (cd), and surface lipid content (SLC) of head rice (e-f) of
the medium-grain (Jupiter) rice lot. Samples were milled with a McGill No. 2 mill for
30 s. When considering temperature impacts, data are averaged across MCs. When
considering MC impacts, data are averaged across temperatures. Means with the same
letter were not statistically different; comparisons are valid only within a subgraph.
367
Fig. 3. Surface lipid content (SLC) of head rice of (a) long-grain (XL753) and
(b) medium-grain (Jupiter) rice lots as impacted by rough rice temperature and moisture
content (MC) at the time of milling. Samples were milled with a McGill No. 2 mill for 30 s.
368
Effects of Temperature, Moisture Content, and Rough Rice

Storage Duration on Milling Properties and Discoloration
K.N. Haydon1 and T.J. Siebenmorgen1
ABSTRACT
Two long-grain rice cultivars, XL753 and CL XL745, both grown in Arkansas in 2014,
were stored in rough rice form at moisture contents (MCs) of 12.5%, 16%, 19%, and
21% (wet basis) at temperatures of 50 F (10 C), 59 F (15 C), 68 F (20 C), 81 F
(27 C), and 104 F (40 C) for a total of 16 weeks, with samples taken at 2, 4, 6, 8, 10,
12, and 16 weeks. After drying, dehulling, and milling to an approximate surface lipid
content (SLC) of 0.4%, head rice was separated from broken kernels and discoloration
of the head rice was measured by a pixel-by-pixel assessment of kernel area with color
values established by a set of discolored kernels of interest chosen from this study.
Head rice yield declined only when discoloration had already developed on milled rice.
Two divergent patterns of discoloration appeared in 21%-MC rice stored at 81 F (27
C) and 104 F (40 C) for at least 8 weeks. The head rice from samples stored at 50
F (10 C) and 59 F (15 C) at MCs up to 21% and 19%, respectively, did not show
any significant color degradation throughout the storage duration, indicating that grain
chilling may be a viable option for maintaining quality in undried rough rice.
INTRODUCTION
At high moisture contents (MCs), rough rice respires rapidly, especially in the hot
temperatures typical of Arkansas in August and September, promoting fungal growth
with the potential for mycotoxin development, and discoloration. Respiration contrib-
utes to many deleterious effects on rice quality, including dry matter loss and kernel
discoloration (Smith and Dilday, 2003).With a fixed storage temperature of 77 F (25
C), Trigo-Stockli and Pederson (1994) found that increasing rough rice MC leads to
increased kernel discoloration, reduced milling yields, and reduced seed germination
rates over 30 days; fungal growth also generally increased with increasing MC. Drying
within a short period of time after harvest is therefore recommended for maintaining
rice quality.
Graduate Assistant and Distinguished Professor, respectively, Department of Food Science, Fayetteville.
1
369
Theoretically, however, the quality of high-MC rough rice could be maintained by

short-term storage with a grain cooling system. According to Brunner (1986), cooling
was used in the 1960s to preserve fresh rice in the interim between harvest and drying,
yet this is not widely practiced now. Utilizing cooling systems could enable farmers
to safely hold rice before accessing commercial dryers or before parboiling. Industry
research suggests multiple benefits of grain cooling, including cool-temperature retention
without additional cooling for several months, reduction of dry-matter loss, prevention
of insect infestation, mold growth, and discoloration, a small reduction in grain MC
thereby reducing drying costs, and preservation of head rice yield (HRY; Kolb, 2008;
Kolb and Braunbeck, 2013).
This study was designed to both evaluate the viability of grain cooling systems
and to determine the inter-related impacts of MC, temperature, and storage duration
on milling properties and discoloration of rice. This research also developed a novel
system of quantifying discoloration that uses computer software, rather than human
judgment, to gauge whether rice is discolored or not.
PROCEDURES
Two long-grain hybrid rice cultivars, XL753 and CL XL745, grown at the Uni-
versity of Arkansas System Division of Agricultures Northeast Research and Exten-
sion Center (NEREC) in Keiser, Ark., and Running Lakes Farm near Pocahontas, Ark.,
respectively, were harvested in 2014 at approximately 22% MC (wet basis). The rice
was cleaned with a dockage tester (Model XT4, Carter-Day, Minneapolis, Minn.) and
conditioned to MCs of 21%, 19%, 16%, and 12.5% as measured by a moisture tester
(AM 5200, Perten Instruments, Hgersten, Sweden). Rice was placed in quart glass
Mason jars and distributed among temperature-controlled storage units maintained at
50 F (10 C), 59 F (15 C), 68 F (20 C), 81 F (27 C), and 104 F (40 C). One jar
of each cultivar/MC/temperature combination was removed after each of seven storage
durations: 2, 4, 6, 8, 10, 12, and 16 weeks. The storage conditions are summarized in
Table 1. After storage, the rough rice was dried to 12.5% MC.
Duplicate 5.20 oz (150 g) rough rice samples from each jar were dehulled with
an impeller husker (Model FC2K, Yamamoto, Yamagata, Japan), then milled using a
laboratory mill (McGill No. 2, RAPSCO, Brookshire, Texas), having a 3 lb 5 oz (1.5
kg) mass placed on the lever arm, 15 cm from the centerline of the milling compart-
ment. Milling durations were 17 s and 22 s for XL753 and CL XL745, respectively,
which resulted in a head rice surface lipid content (SLC) of 0.4%. After milling, head
rice was separated from brokens using a sizing device (Model 61, Grain Machinery
Manufacturing Corp., Miami, Fla.), and head rice SLC was verified by a diode array
near-infrared reflectance (NIR) analyzer (DA 7200, Perten instruments, SE-141 05
Huddinge, Sweden).
Head rice color was measured with an image analysis system (WinSEEDLE Pro
2005aTM, Regent Instruments Inc., Sainte-Foy, Quebec, Canada). Approximately 100
kernels were arranged on a clear, acrylic tray, placed on a flatbed scanner, and imaged
370
with a blue background. The software analyzed the area of the kernels and quantified
what percent of the kernel area was occupied by pixels of pre-set color values. These
values were established by a set of discolored kernels chosen from samples in this study.
Eight different colors were measured: white, three shades of yellow, red/brown, brown/
black, pink/red, and salmon/light pink. Two trays of kernels were analyzed from one
of the two sub-millings from each jar of rice. The total discoloration of the kernels was
calculated as the sum of all of the non-white color percentages.
Head rice yield from two sub-millings from each jar of rice and total discoloration
from two trays analyzed from a single sub-milling from each jar were averaged and
plotted against storage duration for each temperature and moisture content, and the
statistical significance of these correlations were determined by analysis of variance
(ANOVA) at P = 0.05 using linear regression analysis (JMP Pro v. 12.0.1, SAS Institute,
Inc., Cary, N.C.).

At storage temperatures of 50 F (10 C) and 59 F (15 C), HRY was maintained
in CL XL745 stored at MCs up to 21% at 50 F or 19% at 59 F for the entirety of the
16 week storage period (Fig. 1, Table 2). The data from XL753 is not shown, but this
cultivar exhibited very similar trends to CL XL745. Though the HRY reductions in
12.5%-MC rice stored at or above 68 F were slight, they were statistically significant.
Head rice yield losses of a larger magnitude only occurred in rice stored at 21% MC,
and primarily at temperatures of 81 F (27 C) and 104 F (40 C) after 8 weeks, though
HRY also significantly decreased after 16 weeks of storage in rice stored at 59 F (15
C). There was typically visible mold present on these samples that may have impacted
the physical integrity of the kernels, leading to greater breakage during milling. These
reductions in HRY only occurred after discoloration had already exceeded acceptable
levels (Fig. 2). A rough rice storage study conducted by Houston et al. (1957) also found
that HRY losses only occurred after other extensive quality reductions.
Because XL753 and CL XL745 were very similar with respect to total discolor-
ation over the storage duration, the data from both cultivars were pooled (Fig. 2, Table
2). Storage at 50 F (10 C) maintained a low level of discoloration in rice of all MCs
for the entire 16-week storage duration. At 21% MC, a small increase in discoloration
occurred after 10 weeks of storage at 59 F (15 C) and 68 F (20 C). Storage at 81
F (27 C) caused a significant increase in discoloration over 16 weeks in rice stored at
19% and 21%. Discoloration increased significantly over time in rice at all four MCs
stored at 104 F (40 C), with the magnitude of changes in discoloration also increas-
ing with increasing MC.
Two patterns of discoloration appeared at 81 F (27 C) and 104 F (40 C) in
rice stored at 21% MC for at least 8 weeks (Fig. 3, presented for illustrative purposes
only). Data from 16 weeks were excluded due to extreme kernel integrity degradation
that skewed individual color measurements. Though yellow was the predominant color
at both temperatures (Fig. 3d), at 104 F the milled rice appeared uniformly yellow
371
with other colors only appearing at barely noticeable levels. At 81 F, a mottled pat-
tern appeared, with a combination of white, yellow (Fig. 3d), pink/red (Fig. 3a), black/
brown (Fig. 3b), and red/brown (Fig. 3c) kernels. Yellowing at 104 F (40 C) occurred
in rice of all MCs, even 12.5% MC (Fig. 3d). The mechanism for the divergent color
patterns is unknown, but these results corroborate findings from discolored rice samples
obtained from on-farm bins.
Though some researchers favor fungi as the true causes of discoloration due to
their coincident appearance, Belefant-Miller et al. (2005) argue this is not the case. This
study observed that discoloration penetrates the endosperm, but fungal hyphae were
not found within the endosperm of yellowed samples. Also, yellowing was induced in
fungus-free endosperm from rice plants grown from fungicide-treated seed and sprayed
with fungicide during development. However, Belefant-Miller et al. used 5-day incuba-
tions at the longest and temperatures of 142.7 F to 178.9 F (61.5 C to 81.6 C), so
fungi cannot be ruled out in the present study given the possibility that discoloration
occurs by a different mechanism at temperatures of 68 F (20 C) to 104 F (40 C)
maintained throughout 16 weeks of storage. Until the mechanism for post-harvest kernel
discoloration is found and understood, interventions such as cooling may be used to
mitigate these discoloration effects.
For current storage or in-bin drying systems that do not utilize cooling, these
results demonstrate the importance of thorough drying, as well as proper temperature
control through aeration. Even fully dried rice was susceptible to discoloration after only
6 weeks at 104 F (40 C). Because the top layer of a drying bin requires the longest
duration to dry, it may therefore maintain a MC at or above 21% for several weeks.
Hot, early-autumn temperatures typical of Arkansas rice harvest season may induce
discoloration in this high-MC layer.
With respect to color and HRY, cooling to temperatures of 50 F (10 C) may
offer a viable solution for short-term storage of rice at MCs up to 21%. Though milling
properties did not change over the 16-week storage period at cool temperatures, it is not
known how these conditions, which also slow aging effects on starch structure (Swamy
et al., 1978), would affect processing characteristics in operations such as parboiling.
This study is being repeated with rice grown and harvested in 2015 to confirm the
findings from 2014. Further study is needed to understand the causes of discoloration,
especially as they relate to the temperature-dependent patterns observed in this study.
ACKNOWLEDGMENTS
The authors acknowledge the Rice Processing Program staff who assisted with
data collection, the financial support of the Arkansas Rice Research and Promotion
Board, the University of Arkansas System Division of Agriculture, and the corporate
sponsors of the University of Arkansas Rice Processing Program.
372
LITERATURE CITED
Belefant-Miller, H., M.G. Kay, and F.N. Lee. 2005. Small-scale induction of posthar-
vest yellowing of rice endosperm. Cereal Chem. 82:721-726.
Brunner, H. 1986. Cold preservation of grain. Proc. 4th Int. Work. Conf. Stored-
product Prot. 219-229.
Houston, D., R. Straka, I. Hunter, R. Roberts, and E. Kester. 1957. Changes in rough
rice of different moisture content during storage at controlled temperatures. Cereal
Chem. 34:444-456.
Kolb, R.E. 2008. Treatment of Grains through Conservation Cooling with the
GRANIFRIGORTM. FrigorTec GmbH, Amtzell, Germany.
Kolb, R.E. and C.M. Braunbeck. 2013. Quality Assurance of Paddy & Rice with
GRANIFRIGORTM Cooling Conservation. FrigorTec GmbH, Amtzell, Germany.
Smith, C.W. and R.H. Dilday. 2003. Rice: Origin, History, Technology, and Produc-
tion; John Wiley & Sons: Hoboken, N.J. 642 pp.
Swamy, Y.M.I., C.M. Sowbhagya, and K.R. Bhattacharya. 1978. Changes in the
physicochemical properties of rice with aging. J. Sci. Food Agric. 29:627-639.
Trigo-Stockli, D.M. and J.R. Pedersen. 1994. Effect of rice storage conditions on the
quality of milled rice. Proc. 6th Int. Work. Conf. Stored-product Prot. 1:706-711.
Table 1. Overview of the experimental design.

Cultivars Initial moisture content Temperature Duration
(%) [F (C)] (weeks)
XL753 12.5 50 (10) 2
CL XL745 16 59 (15) 4
19 68 (20) 6
21 81 (27) 8
104 (40) 10
12
16
373
Table 2. Correlation coefficients of total discolored kernel area

(%) with storage duration for cultivars XL753 and CLXL745 (data pooled
and averaged), and of head rice yield (%) with storage duration for cultivar CLXL745.
Moisture content
Temperature 12.5% 16.0% 19.0% 21.0%
[F (C)] ----------------------------(%)----------------------------
Head rice 50 (10) NSa NS NS NS
yield (%) 59 (15) NS NS NS -0.67
68 (20) -0.7 -0.5 NS NS
81 (27) -0.65 NS NS -0.7
104 (40) -0.71 NS NS -0.66
Discolored 50 (10) NS NS NS NS
kernel area 59 (15) NS NS NS 0.68
(%) 68 (20) NS NS NS 0.69
81 (27) NS NS 0.52 0.7
104 (40) 0.72 0.87 0.95 0.91
a
374
Fig. 1. Head rice yield (HRY) of CL XL745 over 16 weeks of storage as affected by rough
rice moisture content (MC) and storage temperature. These figures include a baseline, pre-storage HRY at
0 weeks that is the same for all MCs and temperatures, representing rice that was dried immediately after harvest, then milled.
375
376
Fig. 2. Total percentage of kernel area that was discolored (any non-white color) of XL753 and
CLXL745 over 16 weeks of storage as affected by rough rice moisture content (MC) and storage temperature.
Fig. 3. Percentage of kernel area occupied by a) pink/red, b) black/brown,

c) red/brown, and d) yellow of 21% moisture content XL753 over 12 weeks of storage as affected by storage
temperature. Y-axes only extend to 1% for pink/red, 5% for black/brown, and 10% for red/brown to allow for trendline visibility.
377
Variable Impact of Nighttime Air Temperatures

on Rice Chalk and Milling Properties Due to Heading Date
K.N. Haydon1, T.J. Siebenmorgen1, and P.A. Counce2
ABSTRACT
High nighttime air temperatures (NTATs) during reproductive growth stages (R-stages)
are known to cause increased chalkiness and reduced head rice yields (HRYs) in suscep-
tible rice cultivars. These effects have been clearly demonstrated in field trials conducted
from 2007-2010 across the eastern Arkansas rice-growing region (Ambardekar et al.,
2011; Lanning et al., 2011). Field trials continued from 2011-2014 in order to further
verify these NTAT impacts on an expanded set of cultivars. Analysis of the 2012 and
2014 growing years indeed confirmed the trends of increasing chalk and decreasing
HRY with increasing NTATs during critical R-stages. However, in 2011 and 2013 the
trends were quite different among many cultivars. This was attributed to the abnormal
growing environment in these years, when spring rains forced late planting, leading to
significantly later heading dates and markedly reduced NTATs during later R-stages.
These results demonstrate that during years of late planting with late heading of rice,
NTAT is not an entirely accurate predictor of rice quality, as temperatures below a
certain threshold may counterintuitively reduce quality.
INTRODUCTION
The reproductive growth stages (R-stages) of rice are highly sensitive and critical
times for plant development, especially the grain-filling stages (R6 to R8). Ambardekar
et al. (2011) and Lanning et al. (2011) documented correlations between high nighttime
air temperatures (NTATs) during grain-filling, reduced milling quality, and elevated rice
chalkiness during the 2007-2009 and 2007-2010 harvest years, respectively. However,
because rice is a warm-season crop, it is highly susceptible to the effects of cold stress.
In Arkansas, late planting can be prompted by wet spring weather, postponing drill-
seeding until fields are acceptably dried. Consequently, rice may begin maturing as
summer temperatures cool and day lengths shorten.
Graduate Assistant and Distinguished Professor, respectively, Department of Food Science, Fayetteville.
1
Professor, Rice Research and Extension Center, Stuttgart.

2
378
Resurreccion et al. (1977) showed that plants exposed to either day or night
temperatures below 68 F (20 C) or above 86 F (30 C) from 3 days before flowering
through full maturity both produced a greater proportion of chalky kernels. The effects of
cold-temperature exposure are being considered here due to the phenomenon of late 50%
heading dates, a condition prompted by late planting dates that increase the likelihood
of cooler temperatures during critical maturation stages (Siebenmorgen et al., 2013).
PROCEDURES
Samples were obtained from plots cultivated at multiple locations across Arkansas,
spanning from northern to southern latitudes, as part of the Arkansas Rice Performance
Trials. Table 1 summarizes the cultivars and locations. At every location, cultivars were
assigned in a randomized block design to sub-plots within larger plots, such that each
cultivar was replicated three (2011) or four times (2012-2014). Plots were harvested at
moisture contents (MCs) expected to give close-to-optimum head rice yields (HRYs):
approximately 19% to 22% for long-grain cultivars and 22% to 24% for medium-grain
cultivars. Roughly 120 panicles from each plot were randomly selected and cut by hand,
then mechanically threshed in a portable thresher (SBT, Almaco, Nevada, Iowa). All
lots were cleaned using a dockage tester (Model XT4, Carter-Day Co., Minneapolis,
Minn.) and conditioned to 12 0.5% (wet basis) moisture content.
One 5.20 oz (150 g) rough rice sample from each lot was dehulled in a labora-
tory sheller (THU 35B, Satake, Hiroshima, Japan), then milled using a laboratory mill
(McGill No. 2, RAPSCO, Brookshire, Texas), having a 3 lb 5 oz. (1.5 kg) mass placed
on the lever arm, 5.90625 in. (15 cm) from the centerline of the milling compartment.
Head rice was then separated from brokens using a sizing device (Model 61, Grain
Machinery Manufacturing Corp., Miami, Fla.). Surface lipid content (SLC) of head rice
was measured using a lipid extraction system in 2011 (Soxtec Avanti 2055, Foss North
America, Eden Prairie, Minn.) and with a diode array near-infrared reflectance (NIR)
analyzer (DA 7200, Perten instruments, SE-141 05 Huddinge, Sweden), calibrated to
match extraction values, in 2012 to 2014. Head rice yield was adjusted to an arbitrary
0.4% SLC according to the finding by Pereira et al. (2008) that for every 0.10 percent-
age point (pp) difference in SLC from 0.4%, HRY changes in long-grain cultivars by
1.13 pp and in medium-grain cultivars by 0.85 pp.
Chalk measurements were performed on brown rice for each lot. In 2011 and
2012 this was done with an image analysis system (WinSEEDLE Pro 2005aTM, Regent
Instruments Inc., Sainte-Foy, Quebec, Canada), configured to color-classify chalk by a
completely chalky brown rice kernel scanned as a reference. Chalkiness was expressed
as a percentage of the total kernel area. In 2013 and 2014, chalk was quantified by a
similar image analysis system (SeedCount SC5000TR, Next Instruments Pty Ltd.,
Condell Park, NSW, Australia). The softwares calibration was adjusted to correlate
closely with values obtained from the WinSEEDLE system (r = 0.985).
Each year at the University of Arkansas System Division of Agricultures Rice
Research and Extension Center (RREC) near Stuttgart, Ark., growth stages of rice
379
development for each cultivar were visually identified and dates of each stages initia-
tion from R3, the heading date, to R9 were recorded according to the staging system
developed by Counce et al. (2000). Only the heading date, R3, was recorded for each
plot at all other growing locations. The staging data from Stuttgart was used to estimate
when each plot at all other locations would have initiated each R-stage after R3 accord-
ing to the procedure described in detail by Ambardekar et al. (2011).
There was no complete data set of R-stage measurements taken at RREC in 2011,
so a 4-year average of required thermal units for each cultivar, as published by Counce
et al. (2015), was utilized to estimate that years staging data. Ambient-air temperatures
were logged every 30 min with two temperature sensors (HOBO Pro/Temp Data Log-
ger, Onset Computer Co., Bourne, Mass.) at each growing location. Data from the two
sensors were averaged.
The 95th percentiles of NTATs (NT95s) during each R-stage were calculated
according to the protocol used by Ambardekar et al. (2011), using nighttime ambient
air temperature data, considered to begin at 8:00 PM and end at 6:00 AM. Mean chalk
levels and SLC-adjusted HRYs were plotted against the NT95s during each R-stage.
A multivariate analysis platform was used to determine pair-wise correlation coeffi-
cients, and the statistical significance of these correlations was determined by analysis
of variance (ANOVA) at P = 0.05 using linear regression analysis. Tukeys Honestly
Significance Test was used to compare mean HRY and chalk levels among years (JMP
Pro v. 12.0.1, SAS Institute, Inc., Cary, N.C.).

Of special consideration during these 4 years of research were uncharacteristically
late 50% heading dates in 2011 and 2013 (Fig. 1). This was especially noticeable in
2011, when the rice reached R3 around and after mid-August rather than in mid-July to
early August. This was most likely caused by late planting dates. During critical grain-
filling stages in these years, the rice plants generally experienced cooler temperatures
than in years when rice reached 50% heading by late July to early August. It is reasoned
that these conditions may alter the trend of increasing chalk and decreasing HRYs with
increasing NTATs, as warmer temperatures during relatively cooler development times
may confer a benefit to the rice plants.
With this in mind, the data was partitioned into early/normal heading years
(2012 and 2014) and late/atypical heading years (2011 and 2013). In normal heading
years, the expected trends of increasing chalk and decreasing HRY emerged among
susceptible cultivars (Fig. 2, Table 2). Atypical heading years did not demonstrate any
clear trends that could clearly describe the effects of increasing temperature under such
conditions. However, overall temperature climate may play a direct role in the overall
HRY and chalk levels.
Rice that reached 50% heading in late July to early August experienced warmer
temperatures during critical R-stages, with NT95s ranging from 73 F (23 C) to 82 F
(28 C) during R8. Later-heading rice was shifted into a cooler climate during grain-
filling, with NT95s as low as 64 F (18 C) during R8, indicating 95% of temperatures
380
during R8 were below 64 F, and likely capable of producing more chalky and immature
kernels, as described by Resureccion et al. (1977).
These cooler temperatures may have caused the significantly reduced HRYs and
increased chalk among some cultivars in 2011 and 2013 as compared to 2012 and 2014
(Fig. 3). Previous research demonstrated variable susceptibility to high NTAT effects, and
likewise these results show that cultivars respond differently to relatively low NTATs.
Medium-grain cultivar Caffey and long-grain cultivars Roy J, Taggart, and XL753 ex-
hibited significantly elevated chalk levels in 2011 and 2013 as compared to 2012 and
2014 (Fig. 3a). Caffey, Taggart, and the long-grain Wells also showed significant HRY
reductions in late years (Fig. 3b). LaGrue, however, which was shown in 2007-2010 to
be extremely susceptible to NTAT-induced quality reductions, was significantly chalkier
in early heading years, and HRY was not significantly affected by late heading. This
cultivars tendency towards high chalk levels at high temperatures may either mask
any effects due to low temperature stress, or it may indicate a degree of cold tolerance
not seen in other susceptible cultivars, such as XL753 and Wells.
The trends of increasing chalk and decreasing HRY with increasing NTATs, as
reported previously, may only be considered reliable predictors when rice is planted at
recommended times, early in the year, as in 2012 and 2014. New models for understand-
ing temperature effects on agronomic grain yields, HRY, and chalk should consider the
effect of planting and heading dates on the growing environment.
Agronomic grain yields tend to decrease with late planting and subsequent late
heading dates (Siebenmorgen et al., 2013). When weather forces delayed planting, it
appears to be beneficial for some cultivars with respect to chalk and milling quality to
experience relatively warmer NTATs during critical grain-filling stages. Though early
seeding promotes greater agronomic yields, it also creates the conditions that maximize
susceptibility to elevated NTAT effects. However, late planting and low temperatures
during grain-filling may also contribute to quality and yield reductions.
ACKNOWLEDGMENTS
The authors acknowledge the financial support of the Arkansas Rice Research and
Promotion Board, the corporate sponsors of the University of Arkansas Rice Process-
ing Program, the University of Arkansas System Division of Agriculture, and the Rice
Processing Program staff who assisted with data collection.
LITERATURE CITED
Ambardekar, A.A., T.J. Siebenmorgen, P.A. Counce, S.B. Lanning, and A. Mauro-
moustakos. 2011. Impact of field-scale nighttime air temperatures during kernel
development on rice milling Quality. F. Crop. Res. 122:179-185.
381
Counce, P.A., T.C. Keisling, and A.J. Mitchell. 2000. A uniform, objective, and adap-
tive system for expressing rice development. Crop Sci. 40:436-443.
Counce, P.A., T.J. Siebenmorgen, and A.A. Ambardekar. 2015. Rice reproductive de-
velopment stage thermal time and calendar day intervals for six U.S. rice cultivars
in the Grand Prairie, Arkansas, over 4 years. Ann. Appl. Biol. ISSN 0003-:1-15.
Lanning, S.B., T.J. Siebenmorgen, P.A. Counce, A.A. Ambardekar, and A. Mauro-
moustakos. 2011. Extreme nighttime air temperatures in 2010 impact rice chalki-
ness and milling quality. F. Crop. Res. 124:132-136.
Pereira, T., N.T.W. Cooper, and T.J. Siebenmorgen. 2008. Effect of storage tempera-
ture and milling duration on milling properties of rice. Discovery J. 9:64-74.
Resurreccion, A.P., T. Hara, B.O. Juliano, and S. Yoshida. 1977. Effect of tempera-
ture during ripening on grain quality of rice. Soil Sci. Plant Nutr. 23:109-112.
Siebenmorgen, T.J., B.C. Grigg, and S.B. Lanning. 2013. Impacts of preharvest fac-
tors during kernel development on rice quality and functionality. Annu. Rev. Food
Sci. Technol. 4:101-115.
Table 1. Cultivars and growing locations by harvest year.

Year
2011 2012 2013 2014
Medium-grain Bengal Bengal Caffey Caffey
Jupiter Jupiter Jupiter Jupiter
Long-grain LaGrue CL151 Roy J CL151
Wells LaGrue Taggart LaGrue
Roy J XL753 Roy J
Taggart Taggart
XL753 Wells
XL753
Locationa NEREC NEREC PTRS NEREC
NEC Knobel SEREC Knobel
RREC PTRS RREC PTRS
RREC SEREC
RREC
a
NEREC = Northeast Research and Extension Center near Keiser, Ark.; NEC = Newport Exten-
sion Center near Newport, Ark.; Goodman farm (2012) and Turner farm (2014) near Knobel,
Ark.; PTRS = Pine Tree Research Station near Colt, Ark.; SEREC = Southeast Research and
Extension Center near Rohwer, Ark.; and RREC = Rice Research and Extension Center near
Stuttgart, Ark.
382
Table 2. Correlation coefficients of chalk and head rice yield (HRY)

with the 95th percentiles of nighttime air temperature frequencies (NT95s)
during the R5 to R8 reproductive stages in 2012 and 2014. The data from these two
years were pooled and only cultivars planted in both 2012 and 2014 were considered.
Cultivars
Medium-
grain Long-grain
Quality R-stage Jupiter CL151 LaGrue Roy J Taggart Wells XL753
Chalk R5 NSa NS NS NS NS NS -0.42
R6 0.36 0.60 0.42 NS NS -0.64 -0.54
R7 NS 0.37 0.53 0.43 NS NS NS
R8 NS NS 0.77 0.46 0.50 NS NS
HRY R5 NS NS NS NS NS NS NS
R6 NS -0.49 NS NS -0.49 NS 0.37
R7 NS NS NS NS -0.52 NS NS
R8 NS -0.42 -0.71 NS -0.52 NS NS
a
Fig. 1. Average day of year for all cultivars and locations in which each
reproductive stage initiated from R3 through R9 across the 2011-2014 growing
years. Letters indicate significant differences between years as determined by the Tukey-
Kramer honest significant difference test (P = 0.05). Bars denote standard error. For
reference, days 200, 225, and 250 are 18 July, 13 August, and 7 September, respectively.
383
Fig. 2. Relationships of CL151, LaGrue, Jupiter,

Taggart, Roy J, Wells, and XL753 in the early/normal
heading years of 2012 and 2014 for a) chalk and b) head rice yield with
the 95th percentile of nighttime air temperature frequencies during the R8 stage.
384
Fig. 3. Impact of early (2012 and 2014) and late (2011 and 2013) heading dates on a) mean
chalk and b) mean head rice yield across all growing locations as determined by the
Tukey-Kramer honest significant difference test (P = 0.05). Bars denote standard error.
385
Characterization of Broken Rice Kernels

Caused by Moisture-Adsorption Fissuring
S. Mukhopadhyay1 and T.J. Siebenmorgen1
ABSTRACT
Fissuring caused by rapid moisture adsorption generates appreciable amounts of broken
kernels upon milling, thereby reducing the economic value of rice. This study inves-
tigated the extent of kernel fissuring and resultant milling yield reduction in rice lots
that had incurred various levels of moisture adsorption-induced fissuring, as well as the
physical and functional characteristics of broken kernels that resulted from milling such
lots. Roy J, CLXL745, and Jupiter cultivar lots were conditioned to 9%, 11%, 13%,
15%, and 17% initial moisture content (IMC) levels, rewetted in water at 86 F (30 C)
for 2 h, gently re-conditioned to 12% moisture content (MC), and then milled. Results
showed that as IMC prior to rewetting decreased, the extent of fissuring increased, and
hence, milling yield decreased. The mass distribution of broken kernels was different
between long-grain (LG) and medium-grain (MG) cultivar lots and also between the
two LG cultivar lots. Peak and final viscosities were greatest for head rice, and de-
creased significantly with decreasing size of brokens. Although further investigation of
the physical and functional characteristics of broken-kernel fractions is needed before
conclusions on practical significance can be drawn, the paste viscosity trends suggest
that brokens of different sizes may have different functional properties and hence, may
be best suited for different end-use applications.
INTRODUCTION
Moisture-adsorbing environments such as those created by rainfall or high humid-
ity conditions in fields before harvest may induce fissuring in rice kernels of low mois-
ture content (MC). The resultant fissuring causes breakage upon milling, thus reducing
milling yield. This can be a common problem faced by rice producers, primarily due
to logistical harvesting considerations. Moisture-adsorption fissuring can also occur in
post-harvest operations due to inadvertent over-drying and subsequent rewetting of rice.
The recent growth of the pet food industry, wherein brokens are used as an in-
gredient, as well as the increasing demand for rice flour, which is typically produced
1
Senior Graduate Assistant and Distinguished Professor, respectively, Department of Food Science,
Fayetteville.
386
from brokens, has led to a steady increase in the demand for brokens. While several
studies have addressed the impact of rapid moisture adsorption on milling yields, little
research was found that investigated the impact on the physical and functional char-
acteristics of broken kernels. Mukhopadhyay and Siebenmorgen (2015) reported the
impact of moisture adsorption on the extent of fissuring, as well as the particle-size
distribution and functionality of the resultant brokens using one pure-line, long-grain
(LG) cultivar, Roy J. The present study was undertaken to expand the findings of
Mukhopadhyay and Siebenmorgen (2015). Thus, the objectives were to evaluate the
impacts of rapid moisture adsorption on the extent of kernel fissuring and resultant
milling yield reduction, as well as to study the physical and functional characteristics
of broken kernels generated from multiple rice lots that had incurred various levels of
moisture adsorption-induced fissuring.
PROCEDURES
Sample Procurement and Preparation
Figure 1 shows the flowchart for the experiment. Roy J (pure-line, LG), CLXL745
(hybrid, LG), and Jupiter (pure-line, MG) cultivar lots were combine-harvested at Ar-
kansas locations in the fall of 2014 at 17.1%, 19.1%, and 20.6% MC (wet basis, w.b.),
respectively. The lots were cleaned and stored in sealed containers at 39 F (4 C) for
6 months prior to testing.
For each of the three replicates of this study, 31 lb (14 kg) of clean, rough rice
from each cultivar lot was equilibrated to room temperature (~72 F or 22 C) for 24
h. Each 31-lb cultivar/replicate lot was divided into six sublots; five, 4.4-lb sublots
were conditioned to initial moisture contents (IMCs) of 9%, 11%, 13%, 15%, or 17%,
respectively, and a 8.8-lb (4-kg) sublot to 12% IMC (all, 0.5 percentage points) in a
climate-controlled chamber (5580A, Parameter Generation & Control Inc., Black Moun-
tain, N.C.). For all sublots, the IMC of the conditioned lots was determined by drying
duplicate, 15-g subsamples in a convection oven (1370FM, Shellblue, Sheldon Mfg.
Inc., Cornelius, Ore.) maintained at 266 F (130 C) for 24 h (Jindal and Siebenmorgen,
1987). The 12%-IMC sublots were used as controls.
Rewetting of Samples
To create fissures due to rapid moisture adsorption, each of the 45, 4.4-lb (2-kg)
sublots (3 cultivar lots 5 IMCs 3 replications) were wrapped in vinyl screen cloth
bags and rewetted for 2 h in a water bath (Precision 280, Precision Scientific, Win-
chester, Va.) at 86 F (30 C), then drained for 0.5 h, air-dried at ~72 F (22 C) for 1
h, and then re-conditioned to ~12% MC at 79 F/52% RH (26 C/52% RH) inside the
climate-controlled chamber. The 12%-IMC (control) sublots were not rewetted.
Brown Rice Fissure Enumeration and Determination of Milling Yield

After gently re-conditioning to 12% MC, 300 rough rice kernels were randomly
selected from each of the 45 sublots and manually dehulled. Resultant brown rice
kernels were visually examined for fissures using a grainscope (TX-200, Kett Electric
387
Laboratory, Tokyo). Fissured kernels (FK) were enumerated and expressed as a number
percentage of the 300 rough rice kernels.
For measurement of milling yield, one, 150-g subsample from each of the 45 re-
wetted sublots was dehulled using a laboratory dehuller (THU-35A, Satake Engineering
Co., Ltd., Tokyo, Japan) and then milled for the required duration (mentioned below)
using a laboratory mill (McGill No.2, Rapsco, Brookshire, Texas) with a 3.3-lb (1.5-kg)
mass placed on the lever arm 5.9 in. (15 cm) from the center of the milling chamber.
By adjusting the milling duration according to cultivar lot (19 s for Roy J, 25 s for
CLXL745, and 24 s for Jupiter), samples were milled to a consistent degree of milling
as indicated by a target surface lipid content (SLC) of 0.4%. Milled rice contains head
rice (kernels retaining three-fourths or more of their original length) (USDA-FGIS,
2010) and brokens. Milled rice yield (MRY) was quantified as the mass of milled rice,
expressed as a percentage of the original, dried rough rice mass. After brokens were
removed using a sizing device (61, Grain Machinery Manufacturing Co., Miami, Fla.),
head rice yield (HRY) was quantified as the mass of head rice, expressed as a percentage
of the original, dried rough rice mass (USDA-FGIS, 2010).
Physical and Functional Characteristics of Broken Kernels

The size distribution of broken kernels was determined using a sieve analysis per
ANSI/ASAE Standard S319.6 (1997). A sieve shaker (RO-TAP, RX-29, Mentor, Ohio)
with US sieve numbers 10 and 12, having square openings of 0.079 in. (2.00 mm) and
0.066 in. (1.68 mm), respectively, was operated for 15 min, distributing brokens into
large (retained on the 0.079-in. sieve), medium (passed through the 0.079-in. sieve
but retained on the 0.066-in. sieve), and small (passed through the 0.066-in. sieve)
fractions. The 12%-IMC (control) sublots generated negligible amounts of brokens,
hence, particle-size distribution and functionality analyses (described below) were not
conducted for these sublots.
Peak (PV) and final viscosities (FV) of flour from head rice and the small, me-
dium, and large broken-kernel fractions were determined according to AACCI Method
61-02.01 (1997). From each fraction, 7 g of head rice/broken kernels were ground into
flour using a cyclone sample mill (3010-30, UDY Corporation, Fort Collins, Colo.)
equipped with a 0.020 in. (0.5-mm) screen. Duplicate 2-g subsamples of flour were
dried in the convection oven at 266 F (130 C) for 1 h to determine MCs, per AACCI
Method 44-15.02 (1975). Adjusted for MC, viscosities were determined on a paste of
3 g of rice flour in 0.007 liquid gal (25 ml) of distilled water using a viscometer (RVA-
Super 4, Newport Scientific Pvt. Ltd., Warriewood, NSW, Australia). The flour paste was
held at 122 F (50 C) for 1.5 min, heated to 203 F (95 C) at 12.2 C/min, held at 203
F for 2 min, cooled to 122 F at 12.2 C/min, and finally held at 122 F for 1.5 min.
Data Analyses
Statistical analyses were performed (JMP Pro software, Ver. 12.0.1, SAS Institute,
Inc., Cary, N.C.). Analysis of variance (ANOVA, P = 0.05) was conducted and means
separated using Fishers least significant difference procedure (LSD, P = 0.05).
388

Enumeration of Fissures and Determination of Milling Yield
Figure 2 shows FK, MRY, and HRY for the rewetted sublots. Across all cultivars,
IMC had a profound effect on the extent of fissuring and resultant milling yields; as
IMC prior to rewetting decreased, the extent of fissuring increased, and HRY corre-
spondingly decreased. At 9% IMC, HRY reached a value of 0%, similar to results of
Mukhopadhyay and Siebenmorgen (2013). Thus, the lesser the IMC of the rice when
it is rewetted, particularly when IMC decreases below 15%, the greater the extent of
fissuring and consequent breakage of kernels when milled, and hence, the more the
reduction in HRY.
For all cultivars, MRY decreased with decreasing IMC, although clear statistical
differences did not occur until 9%-IMC rice was rewetted. This suggests that with se-
vere fissuring and consequent severe breakage during milling, some endosperm leaves
with the bran stream, thus decreasing the total mass of rice produced through milling.
These results corroborated the findings of Mukhopadhyay and Siebenmorgen (2013) and
Mukhopadhyay and Siebenmorgen (2015). Reduced MRY has economic implications
in that both head rice and broken kernels have economic value and thus contribute to
the total value of a rice lot.
Physical and Functional Characteristics of Broken Kernels

Figure 3 shows the mass percentages of the broken-kernel fractions. Across all
IMCs, the mass percentage of the medium broken-kernel fraction was the greatest for
LGs Roy J and CLXL745, whereas the mass percentage of the large broken-kernel frac-
tion was the greatest for MG, Jupiter. The two LGs differed in the mass-distribution of
broken-kernel size fractions; for Roy J, it was medium > small > large [corroborating
results of Mukhopadhyay and Siebenmorgen (2015)]; whereas for CLXL745, it was
medium > large > small. These results indicate that the size-distribution of broken ker-
nels may differ even among LG cultivar lots. For Jupiter, the mass percentages of the
broken-kernel fractions were large > medium > small. This knowledge may be useful
if processors need to fractionate broken kernels of different sizes for specific purposes.
Figure 4 shows the PV and FV of head rice and the broken-kernel fractions. Some
data could not be obtained (indicated by missing bars in Fig. 4) owing to insufficient
quantities of samples for analysis using the viscometer. In general, for all cultivars and
across all IMCs, both PV and FV were greatest for head rice and decreased significantly
with decreasing size of brokens. Thus, regardless of cultivar and IMC prior to rewetting,
kernels that broke during milling had different viscosity properties as compared to the
kernels that did not break (i.e., head rice).
The paste viscosity trends suggest that brokens of different sizes may have dif-
ferent functional properties. Although statistically significant, further investigation is
required to confirm if the differences in the functional properties of the flour obtained
from the different broken-kernel fractions are practically significant from a processing/
product-formulation/sensory perception standpoint.
389
The size-distribution of broken kernels was different between LG and MG cultivar
lots and also between the two LG cultivar lots. In general, PV and FV were greatest for
head rice, and both decreased significantly with decreasing size of brokens. The paste
viscosity trends suggest that brokens of different sizes may differ in their functional
properties. This information can be used to ascertain if broken kernels with different
physical and functional characteristics should be directed to different end-use applications.
ACKNOWLEDGMENTS
The authors thank the University of Arkansas System Division of Agriculture,
the Arkansas Rice Research and Promotion Board, and the University of Arkansas Rice
Processing Program Industry Alliance Group for the financial support of this project.
LITERATURE CITED
AACC International. Approved Methods of Analysis, 11th Ed. Method 61-02.01.
Determination of the pasting properties of rice with the Rapid Visco Analyser.
Approved October 15, 1997. AACC International: St. Paul, Minn. http://dx.doi.
org/10.1094/AACCIntlMethod-61-02.01
AACC International. Approved Methods of Analysis, 11th Ed. Method 44-15.02.
Moisture - air-oven methods. Approved October 30, 1975. AACC International:
St. Paul, Minn. http://dx.doi.org/10.1094/AACCIntlMethod-44-15.02
ANSI/ASAE S319.6. Method of determining and expressing fineness of feed materi-
als by sieving. Approved July 1997. Reaffirmed by American National Standards
Institute February 2003. American Society of Agricultural Engineers: St. Joseph,
Mich.
Jindal, V.K. and T.J. Siebenmorgen. 1987. Effects of oven drying temperature and
drying time on rough rice moisture-content determination. Trans. of the ASAE.
30(4):1185-1192.
Mukhopadhyay, S. and T.J. Siebenmorgen. 2013. Impact of rapid moisture adsorp-
tion on milling yields. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B R
Wells Rice Research Studies. University of Arkansas Agricultural Experiment
Station Research Series 609:338-344. Fayetteville.
Mukhopadhyay, S. and T.J. Siebenmorgen. 2015. Physical and functional character-
istics of broken rice kernels created by rapid moisture adsorption. In: R.J. Norman
and K.A.K. Moldenhauer (eds.). B R Wells Arkansas Rice Research Studies.
339. Fayetteville.
USDA-FGIS. United States Department of Agriculture, Federal Grain Inspection
Service. 2010. United States Standards for Rice. Grain Inspection, Packers, and
Stockyards Administration. Washington D.C., USA. Access date: 15 Febraury
2016. Available at: https://www.gipsa.usda.gov/fgis/standards/ricestandards.pdf
390
Fig. 1. Process flowchart for the experiment.
391
Fig. 2. Fissured kernel percentage (FK) (A), milled

rice yield (MRY) (B), and head rice yield (HRY) (C), for the indicated
sublots after being conditioned to five initial moisture contents (MCs), rewetted,
and conditioned to 12% MC prior to milling. The control (12% IMC) sublots were not
rewetted, but rather conditioned from harvest MC to 12% MC prior to milling. Within each
cultivar/initial MC set, values followed by the same letter are not significantly different (P
> 0.05). Bars are based on the mean values of three experimental treatment replications.
392
Fig. 3. Mass percentages of small, medium and large broken-kernel fractions for the
indicated cultivar sublots after being conditioned to five initial moisture contents (MCs),
rewetted, and conditioned to 12% MC prior to milling. Within each IMC/broken-kernel
fraction set, values followed by the same letter are not significantly different (P > 0.05).
393
394
Fig. 4. Peak (PV) and final viscosities (FV) of head rice (HR), small (S), medium (M) and large (L) broken-kernel fractions
for the indicated cultivar sublots after being conditioned to five initial moisture contents (MCs), rewetted, and conditioned to 12%
MC prior to milling. Within each initial MC/kernel fraction set, values followed by the same letter are not significantly different (P > 0.05).
Observing Fissures in Rough Rice Kernels

Using X-Ray Imaging: Preliminary Observations
Z.R. Odek1, B. Prakash1, and T.J. Siebenmorgen1
ABSTRACT
Fissured rice kernels generally break during milling, leading to head rice yield reductions
and consequently, large reductions in commercial value. Owing to the fact that there is
no rapid method of visualizing fissures in rough rice kernels, the exact mechanism(s)
and occurrence of fissuring remains unresolved. While laboratory systems are available
to observe fissures in brown and milled rice kernels, these instruments, unfortunately,
cannot detect fissures in rough rice, the state in which rice is normally dried. In this
manuscript, an instrument for viewing fissures in rough rice kernels using X-rays is
presented. The successes, challenges, and other observations experienced during pre-
liminary experimentation are discussed.
INTRODUCTION
Fissured kernels are a major concern in the rice industry due to their susceptibility
to breakage during milling, leading to a decrease in head rice yield (HRY; Siebenmor-
gen et al., 2005). Two general types of fissures are observed in rice kernels: moisture
desorption (drying) fissures (Kunze and Prasad, 1978; Schluterman and Siebenmorgen,
2007) and moisture adsorption (rewetting) fissures (Bansazek and Siebenmorgen, 1990;
Kunze, 2008). Cnossen and Siebenmorgen (2000) proposed material state differences
between the inner core and the endosperm periphery, which cause differential intra-kernel
stress, as the main cause of moisture desorption fissures. These fissures are associated
with rapid drying and resemble a turtle back pattern appearing in no specific alignment
(Stermer, 1968; Bautista et al., 2000). On the other hand, moisture adsorption fissures
occur when dry kernels are rapidly rewetted. (Kunze and Hall, 1965). These fissures
originate from the center portion of the kernel and progress outwards across the kernel,
parallel to the minor axis; these are referred to as cross-wise or straight fissures
(Stermer, 1968; Bautista et al., 2000).
Observation and quantification of fissures in dehulled rice kernels has been pos-
sible using instruments such as a grainscope (TX-200, Kett Electric Laboratory, Tokyo,
1
Graduate Assistant, Post Doctoral Associate, and Distinguished Professor, respectively, Department of
Food Science, Fayetteville.
395
Japan; Siebenmorgen et al., 2005) or a video microscopy system (Bautista et al., 2000).
A grainscope is a portable instrument that allows for fissure assessment in rice kernels
placed on a tray through which light is passed, enabling fissures to be easily observed
(Siebenmorgen et al., 2005). On the other hand, a video microscopy system consists
of a camera with a 50 magnification lens, luminance controller, and a video recorder
allowing for continuous visual monitoring of kernels during drying. Both instruments
(grainscope and video microscopy system) are only applicable in brown and milled
rice kernels, hence cannot be used to observe fissures in actual drying and tempering,
which occurs in the rough rice state with the hulls intact. Therefore, to fully understand
fissuring during the drying process, there is need for an alternative method.
X-ray imaging has shown potential for use in achieving this goal. Henderson
(1954) used X-ray imaging to develop a drying procedure that would yield maximum
HRYs. Menezes et al. (2012), using X-ray imaging, showed a relationship between the
degree of fissuring and the germination potential of rice kernels.
Due to the rare use of X-ray imaging as a technique for fissure visualization in
rough rice kernels, there are no standard procedures for its application. In addition,
with increasing technological advancements, better X-ray imaging systems are being
manufactured, which have advanced capabilities. Some of these include the capability
of producing digital images, eliminating time that would have been used for film de-
velopment, being fully shielded to eliminate need for additional X-ray shielding, being
easy to operate without specialized X-ray knowledge, and laser centering capability
to aid in proper sample positioning. However, depending on the research objectives
and the type of X-ray equipment being used, any research involving X-ray imaging
involves preliminary experiments aimed at optimizing a particular process. The goal of
this overall study is to investigate the fissure occurrence process in rough rice kernels
during drying, tempering, and cooling stages. To achieve this goal, there is a need to
design an auxiliary system that would allow simulation of the drying, tempering, and
cooling processes of rough rice kernels inside an X-ray cabinet while periodically
acquiring images. The objectives of the preliminary testing include:
1) Determine a suitable magnification level that would allow for high-resolution
visualization of fissures in rough rice kernels.
2) Determine an appropriate orientation of rough rice kernels during X-ray imaging
that would allow for visualization of fissures present.
3) Determine a suitable material/holder for use inside the X-ray cabinet to secure and
position rough rice kernels without reducing the visual quality of images pro-
duced.
PROCEDURES
An X-ray system (UltraFocus 60, Faxitron Bioptics LLC, Tucson., Ariz.) with a
maximum sample area dimensions of 10 cm 15 cm (4 in. 6 in.) and up to 6 geometric
magnification was used for this study. The system produces X-rays with an energy range
of 10 to 60 kV, a maximum tube current of 0.3 mA and requires 100 to 200 V, 50/60
Hz to operate. Increasing magnification is achieved by raising an acrylic sample shelf
396
within the chamber. While increasing the magnification level is often advantageous to
improve the image resolution, raising the sample shelf reduces the available field of
view, thereby limiting the number of kernels that can be viewed. Therefore, there is a
tradeoff between image resolution and available field of view. Figure 1 illustrates the
relationship between the magnification level and the available field of view in the X-
ray cabinet. As the sample shelf is raised in the X-ray cabinet, the distance between the
samples and the X-ray emitter decreases, leading to a decreased field of view.
To determine a suitable magnification for high-resolution fissure observation,
long-grain CLXL745 rough rice kernels dried from 18.5% (wet basis) initial moisture
content (MC) to a final MC of 12.1% at 140 F (60 C) and 12% RH, were placed on
the sample shelf at the least magnification (1) and then an X-ray image was taken. The
sample shelf was then raised one level and another X-ray image was taken. This was
repeated for all possible magnifications. The images were then visually analyzed and
a suitable magnification selected based on the image resolution and number of kernels
that could be visualized at each magnification.
In order to determine an appropriate orientation of rough rice kernels that would
allow fissure observation, fissured rice kernels were randomly selected. X-ray images
of the kernels were taken at 5 magnification with the kernels placed on the sample
shelf using two orientations. First, images were taken with the kernels placed on the
width side (Fig. 2a), then on the thickness side (Fig. 2b). Using Microsoft Paint (Mi-
crosoft Corporation, Redmond, Wash.), the two images were then placed side by side
into one viewing frame for ease of comparison. The image was then visually analyzed
to determine the appropriate orientation for fissure visualization in rough rice kernels.
In designing a mechanism to hold kernels inside the X-ray cabinet, two design
constraints were addressed. Firstly, kernels were to be positioned so as to allow uniform
airflow across the kernels, a feature needed for subsequent drying research. Secondly,
the material used for the mechanism had to be transparent to X-rays to avoid interfer-
ence with fissure visualization. Three designs of kernel-positioning mechanism were
evaluated. Acrylic was the material of choice for the mechanism since it is transparent
to X-rays. The three designs considered for the experiment were, design one which
comprised an acrylic plate with cut-out slots slightly larger than the size of a rice
kernel, together with a nylon mesh placed below the slots to prevent the kernels from
falling over. Design two comprised two acrylic plates one glued on top of the other,
with the top plate having cut-out slots and the bottom having drilled holes. Finally,
design three comprised an acrylic plate with a 2-mm wide double-sided tape used to
secure the kernels. Rough rice kernels were attached to one side of the tape while the
other was attached to the acrylic plate. X-ray images using each of the three designs
were compared to determine a suitable method for securing and positioning rough rice
kernels during image acquisition.

Figure 3 shows a set of X-ray images of rough rice kernels at 1 to 6 magni-
fication. Figure 3 clearly indicates that, as magnification increased, image resolution
397
increased. However, as magnification increased, the available field of view decreased

(Table 1), hence fewer kernels could be visualized.
Of most critical importance to the goal of this work is the need to clearly observe
fissures in rough rice kernels. It was observed that 3 to 6 magnification produced
high-resolution images in which fissures could be visualized (Fig. 3). Magnification
at 4 was deemed to yield an optimal high-resolution X-ray images that clearly show
fissures with an adequate field of view, allowing approximately 20 to 30 rough rice
kernels to be observed.
Fissures in rough rice kernels were found to be visible in both the width and thick-
ness orientations. Figure 4 shows an X-ray image of five rough rice kernels positioned
in both orientations. All fissures visible in the thickness orientation were also visible in
the width orientation. However, not all fissures visible in the width orientation were also
visible in the thickness orientation. Therefore, placing kernels so as to allow exposure
to the width side was deemed appropriate for visualizing fissures in rough rice kernels.
The material used for holding and positioning kernels in the X-ray cabinet influ-
enced the visual quality of the X-ray images. It was observed that density differences
of materials within the field of view were greatly amplified by the X-ray images, lead-
ing to a reduction in visual quality. Figure 5 shows the effect of density differences of
materials within the field of view on the visual quality of X-ray images of rough rice
kernels on the cut-out slots and a nylon mesh underneath the kernel-holder, and the
cut-out slots with drilled holes underneath the kernel-holder. It was observed in Fig. 5a
that the density differences between the acrylic plate, the cut-out slots, and the nylon
mesh led to a reduction in the quality of the X-ray image of this kernel-holder design.
Fissures that were aligned with the grid lines of the nylon mesh were not clearly visible.
As shown in Fig. 5b, density differences caused by the cut-out slots and the drilled holes
led to a reduction in the quality of X-ray images of that kernel holder. Figure 6 shows
a photograph and an X-ray image of the double-sided tape kernel holder. Using a thin,
double-sided tape resulted in minimum density differences and no interferences in the
X-ray image quality (Fig. 6b). This approach was deemed to provide a highly suitable
method for holding the rough rice kernels both for ease of fissure identification and for
allowing uniform air flow around the kernels.
A 4 magnification with rough rice kernels oriented on their width sides and
attached to an acrylic plate using a double-sided tape was considered an appropriate
magnification level, kernel orientation, and method of securing/positioning kernels for
fissure observation; this combination produced high-resolution images with visibility
of fissures in rough rice kernels. This preliminary study serves to provide techniques
for visualizing fissures in rough rice kernels. The findings will be applied in designing
an auxiliary system that simulates actual rough rice drying, and simultaneously allows
visualizing fissures during the drying, tempering, and cooling processes, which will
ultimately provide a better understanding of the dynamics of rice fissuring.
398
ACKNOWLEDGMENTS
The authors acknowledge the financial support by the corporate sponsors of the
University of Arkansas Rice Processing Program, the Arkansas Rice Research and
Promotion Board, and the University of Arkansas System Division of Agriculture. Ad-
ditionally, of special acknowledgment is the monetary support provided by the Kellogg
Company for purchase of the X-ray system.
LITERATURE CITED
Banaszek, M. M. and T.J. Siebenmorgen. 1990. Head rice yield reduction rates
caused by moisture adsorption. Trans. ASAE 33(4):1263-1269.
Bautista, R.C., T.J. Siebenmorgen, and J.W. Fendley. 2000. Fissure formation charac-
terization in rice kernels during drying using video microscopy. In: R.J. Norman and
C.A. Beyrouty (eds.). B.R. Wells Rice Research Studies 1999. University of Arkan-
sas Agricultural Experiment Station Research Series 476:325-330. Fayetteville.
Cnossen, A.G. and T.J. Siebenmorgen. 2000. The glass transition temperature
concept in rice drying and tempering: Effect on milling quality. Trans. ASAE
43(6):1661-1667.
Henderson, S.M. 1954. The causes and characteristics of rice checking. Rice J.
57(5):16-18.
Kunze, O.R. 2008. Effects of drying on grain quality - Moisture readsorption causes
fissured rice grains. Agricultural engineering international: CIGR J.
Kunze, O.R. and C.W. Hall. 1965. Relative humidity changes that cause brown rice
to crack. Trans. ASAE 73(2):448-450.
Kunze, O.R. and S. Prasad. 1978. Grain fissuring potential in harvesting and drying
of rice. Trans. ASAE 21(2):361-366.
Menezes, N.L., S.M. Cicero, F.A. Villela, and R.P. Bortolotto. 2012. Using x-rays to
evaluate fissures in rice seeds dried artificially. Revista brasileira de sementes 34
(1):70-77.
Schluterman, D.A. and T.J. Siebenmorgen. 2007. Relating rough rice moisture con-
tent reduction and tempering duration to head rice yield reduction Trans. ASABE
50(1):137-142.
Siebenmorgen, T.J., G. Quin, and C. Jia. 2005. Influence of drying on rice fissure for-
mation rates and mechanical strength distributions. Trans. ASAE 48(5):1835-1841.
Stermer, R.A. 1968. Environmental conditions and stress cracks in milled rice. Ce-
real Chemistry 45(4):365-373.
399
Table 1. Field of view for each level of

magnification in the Faxitron UltraFocus 60 X-ray system.
Field of view
Magnification level Metric English
(cm cm) (in. in.)
6 1.7 2.5 0.7 1
5 2 3 0.8 1.2
4 2.5 3.75 1 1.5
3 3.3 2 1.3 2
2 5 7.5 2 3
1.5 6.7 10 2.7 4
1 10 15 46
Fig. 1. Illustration of the Faxitron Ultrafocus 60 X-ray

cabinet with a decreased field of view as magnification is increased.
400
Fig. 2. Illustration of (a) the width-side and

(b) thickness-side views of a rough rice kernel.
401
Fig. 3. X-ray images of rough rice kernels at 1, 2, 3, 4, 5, and 6 magnifications.
402
Fig. 4. X-ray images of five rough rice kernels aligned

to be viewed in the width (left) and thickness (right) orientations (Fig. 2).
All fissures observed in the thickness orientation were also visible in the width
orientation but not all in the width orientation were visible in the thickness orientation.
Fig. 5. X-ray images of rough rice kernels on (a) an acrylic plate

with cut-out slots and (b) a nylon mesh underneath and in between two
acrylic plates, one with cut-out slots and the other with drilled holes. Both show a
reduction in visual quality of the images due to density differences of the materials used.
403
Fig. 6. A (a) photograph and (b) X-ray image of rough rice kernels
on a 2-mm wide, double-sided tape secured on an acrylic plate. This method of
positioning the kernels caused the least reduction in visual quality of the X-ray images.
404
Kernel and Starch Properties of United States-Grown

and Imported Medium- and Short-Grain Rice Cultivars
J. Patindol1, J.-R. Jinn1, Y.-J. Wang1, and T.J. Siebenmorgen1
ABSTRACT
The acreage for medium- and short-grain rice in the mid-southern United States rice-
growing region is increasing. This work aimed to identify the quality traits of impor-
tance to the markets of these grain types. Twenty-five medium- and short-grain milled
rice samples were analyzed for physical, gelatinization, pasting, and starch structural
properties. Six samples were from Arkansas (AR), 5 were from California (CA), and
14 were imported (IM) from various countries (Bangladesh, Bhutan, China, India,
Italy, Mexico, and Taiwan). Compared to the CA samples, the AR samples had greater
kernel yellowness, gelatinization temperature, mineral content, and percentages of B2
and B3 amylopectin chains; but less whiteness, setback viscosity, and percentage of
amylopectin A chains. The IM samples had a wide range of physical, functional, and
structural characteristics as the samples may be imported to meet the requirements of
specific ethnic groups and/or culinary applications.
INTRODUCTION
Combined medium- and short-grain rice production in the United States has been
projected to be 2.8 million metric tons for 2016, and accounts for more than one-fourth
of the U.S. rice production (Childs, 2015). Since 2014, the mid-South rice-growing
regions have attained a larger-than-normal share of U.S. medium- and short-grain
rice production due to reduced acreage in California and consequent water shortages.
Export and import forecast for these grain types are 1.7 and 0.2 million metric tons,
respectively, and imports account for ~11% of the U.S. medium- and short-grain rice
domestic consumption (Childs, 2015). Northeast Asia, North Africa, and the Middle
East are significant export markets for U.S. medium- and short-grain rice (Childs,
2015). Domestically, U.S. medium- and short-grain cultivars are preferred for breakfast
cereals, pudding, baby food, and brewing. Calrose-type cultivars are generally used
in making California rolls and localized sushi products (Bryant et al., 2013; Marton,
1
Professional Assistant I, Program Technician, Professor, and Distinguished Professor, respectively,
Department of Food Science, Fayetteville.
405
2014). Medium-grain rice is the primary staple of various consumers in Asia, Australia,
Africa, Italy, Latin America, and the Caribbean (Calingacion et al., 2014). Some Asian
countries (e.g., Bangladesh, Sri Lanka, and China) grow aromatic, short-grain variet-
ies with very small round grains that are highly prized and bring premiums on local
markets (Efferson, 1985).
This work compared the kernel physicochemical properties and starch structural
features of medium- and short-grain rice cultivars grown in Arkansas with those grown
in California, and with samples imported from different regions of the world. The pri-
mary goal is to gain a better understanding of the fundamental grain quality traits of
importance to the medium- and short-grain rice markets.
PROCEDURES
Rice Samples
The milled rice sample set consisted of 6 samples from Arkansas (AR), 5 samples
from California (CA), and 14 samples imported (IM) from various countries (Bangla-
desh = 1, Bhutan = 1, China = 5, India = 1, Italy = 1, Mexico = 1, and Taiwan = 4).
All AR samples were obtained as rough rice from the 2014 crop (3 foundation seeds
and 3 breeding lines) and milled in the laboratory. Milled rice samples from China and
Taiwan were gifts from a research colleague; all other samples were purchased from
grocery or specialty stores in the northwest Arkansas area during the first quarter of
2015. The Arkansas samples were milled to a target surface lipid content of 0.4% and
whole kernels (head rice) were separated from broken kernels using a double-tray shaker.
Physical Qualities
Chalk measurements were performed on duplicate, 100-kernel milled rice samples
using an image analysis system (WinSEEDLETM Pro 2005a, Regent Instruments, Que-
bec, Canada). Head rice color was measured using a colorimeter (ColorFlex, Hunter
Associates Laboratory, Reston, Va.). Kernel dimensions (length, width, and thickness)
were measured on ~1000 kernels using a SeedCount 5000 digital image analysis system
(Next Instruments, New South Wales, Australia). Surface lipid content was determined
according to AACC Method 30-20 (AACC International, 2000) with modifications by
Matsler and Siebenmorgen (2005).
Chemical Components
Milled rice flour samples were obtained by grinding in a laboratory mill (cyclone
sample mill, Udy Corp., Ft. Collins, Colo.) to pass through a 0.5-mm sieve. Apparent
amylose content was determined by iodine colorimetry; moisture content by the oven-
drying method; total protein by the micro-kjeldahl method; and mineral content by the
dry-ashing method. Starch was prepared from milled rice flour by extraction with dilute
406
alkali (0.1% NaOH) followed by lipid removal with water-saturated n-butyl alcohol
(Patindol and Wang, 2002). Amylopectin chain-length distribution was characterized by
high-performance anion-exchange chromatography with pulsed amperometric detection
(HPAEC-PAD) using isoamylase-debranched starch samples (Patindol et al., 2014).
Gelatinization and Pasting Characteristics

Milled rice flour gelatinization properties were assessed with a differential scan-
ning calorimeter (DSC, Pyris Diamond, Perkin Elmer Instruments, Shelton, Conn.).
Flour pasting properties were determined with a Rapid Visco-Analyser (RVA model 4,
Perten Instruments, Springfield, Ill.). Rice flour slurry was prepared by mixing 0.11 oz
(3.0 g) of rice starch (12% moisture content basis) with 0.85 oz (25.0 mL) of deionized
water and heated from 122 F to 203 F (50 C to 95 C) at 52 F (11.2 C)/min, held
at 203 F (95 C) for 2.5 min, cooled to 122 F (50 C) at 11.2 C/min, and held at 50
C for 1.0 min.

Table 1 presents the ranges and means for the kernel and starch properties of
the 25 milled rice samples. The kernel images of 8 representative samples are shown
in Fig. 1. Kernel dimension measurements showed that the AR and CA samples were
comparable in length, width, and thickness (Table 1). Those of the IM samples varied
widely, with length-to-width ratios that ranged from 1.6 to 2.8. Based on Federal Grain
Inspection Service (FGIS) standards, milled rice may be classified according to kernel
length-to-width ratio as long (3.0), medium (2.0-2.9), or short (1.9) (USDA, 2014).
Only 12 of the 25 samples were medium-grain type 4 AR, 4 CA, and 4 IM, and the rest
were short-grain. The AR samples tended to have less whiteness and greater yellowness
than the CA samples; the IM samples had whiteness and yellowness values that varied
widely. Percent chalk ranged from 0.6% to 59.2%; it was highest for Arborio, a sample
imported from Italy and least for a sample (no cultivar name) imported from China. The
percent chalk of the United States-grown samples was less than 6%, except for one CA
sample (Haitai brand), in which the percent chalk was 13.7%.
The onset gelatinization temperature (GT) of most samples was below 70 C
(Table 1), with exception of Taichung 11 (from Taiwan, onset GT = 165.9 F; 74.4 C)
and Sona Masoori (from India, onset GT = 161.1 F; 71.7 C). The CA samples had
the narrowest range of gelatinization properties, whereas the AR samples had slightly
greater GT parameters (onset, peak, end, and range) than the CA samples by at least 1 C.
A wider range of pasting characteristics was observed in the IM samples than the
locally grown cultivars (Table 1). Figure 2 shows the pasting profiles of 8 representa-
tive samples, including AR samples of Jupiter and Caffey. Some samples with unique
pasting properties were noted. Morelos (from Mexico) had the highest peak (3838
cP) and final (5613 cP) viscosities. Sona Masoori had a very low peak (1832 cP) and
breakdown (16 cP) viscosities, and a very high setback (2411 cP). Tainong-71 (from
407
Taiwan) had a very high breakdown viscosity (1965 cP). All AR and CA samples had
negative setback viscosities. Between AR and CA samples, the AR samples tended to
have higher peak and lower setback and total setback viscosities.
Amylose content was comparable among AR and CA samples, whereas that of
the IM samples varied widely (Table 1). Five samples had an amylose content of 20%
(Taichung 1, Morelos, Bhutan, Sona Masoori, and Kalijira), and the rest had an amylose
content of 18%. Protein content varied widely among IM samples, but was comparable
among AR and CA samples. Total ash (mineral) content was below 1.0%, and tended
to be greater for the AR samples. Amylose content correlated positively with onset GT,
peak viscosity, setback, and total setback; and negatively with gelatinization range and
breakdown (Table 2). Protein correlated positively with yellowness, and mineral content
correlated positively with percent chalk only.
Amylopectin chain-length distribution analysis revealed a wide range of variation
in the percentages of A, B1, B2, and B3 chains among IM samples (Table 1). The AR
samples had lesser percentages of A chain and greater percentages of B2 and B3 chains,
in comparison with the CA samples. Correlation analysis showed that A chain correlated
negatively with GT, final viscosity and setback; and positively with GT range and paste
breakdown (Table 2). The B1 chain correlated positively GT, final viscosity, setback,
and total setback; and negatively with GT range, peak viscosity, and breakdown. The
B2 chain poorly correlated with any quality attribute. The B3 chain positively correlated
with GT and peak viscosity.
Producing rice that can satisfy domestic and/or export market demands is the
key to a successful and competitive Arkansas rice industry. Findings from this research
are useful to farmers in choosing specific cultivars to plant, to rice breeders in know-
ing the traits/markers to include in varietal improvement efforts, and to processors in
optimizing processing operations to consistently produce high-quality milled rice and/
or derived products.
ACKNOWLEDGMENTS
The authors wish to acknowledge the financial support from Arkansas Rice
Research and Promotion Board and the University of Arkansas System Division of
Agriculture.
LITERATURE CITED
AACC International. 2000. Approved Methods of Analysis, 10th Edition. Method
30-20.01. Crude fat in grain and stock feeds. Approved 3 November 1999. AACC
International. St. Paul, Minn.
408
Bryant, R.J., K.M. Yeater, and H. Belefant-Miller. 2013. The effect of induced yel-
lowing on the physicochemical properties of specialty rice. J. Sci. Food Agric.
93:271-275.
Calingacion M., A. Laborte, A. Nelson, A. Resurreccion, J.C. Concepcion, V.D. Day-
gon, R. Mumm, R. Reinke, S. Dipti, P.Z. Bassinello, J. Manful, S. Sophany, K.C.
Lara, J. Bao, L. Xie, K. Loaiza, A. El-hissewy, J. Gayin, N. Sharma, S. Rajeswari,
S. Manonmani, N.S. Rani, S. Kota, S.D. Indrasari, F. Habibi, M. Hosseini, F.
Tavasoli, K. Suzuki, T. Umemoto, C. Boualaphanh, H.H. Lee, Y.P. Hung, A.
Ramli, P.P. Aung, R. Ahmad, J.I. Wattoo, E. Bandonill, M. Romero, C.M. Brites,
R. Hafeel, H.S. Lur, K. Cheaupun, S. Jongdee, P. Blanco, R. Bryant, N. Thi Lang,
R.D. Hall, M. Fitzgerald. 2014. Diversity of global rice markets and the sci-
ence required for consumer-targeted rice breeding. PloS One 9:e85106. pp. 1-13.
doi:10.1371/journal.pone.0085106
Childs, N. 2015. U.S. rice production forecast lowered to 205.0 million cwt. In: Rice
Outlook/RCS-15h/August 14, 2015. USDA Economic Research Service. Avail-
able at: http://www.ers.usda.gov/media/1881770/rcs-outlook-august-final.pdf
Efferson, J.N. 1985. Rice quality in world markets. In: Rice grain quality and mar-
keting. International Rice Research Institute, Manila, Philippines. pp. 1-13.
Marton, R. 2014. Rice: A global history. Reaktion Books Ltd., London, UK. 128 pp.
Matsler, A.L. and T.J. Siebenmorgen. 2005. Evaluation of operating conditions for
surface lipid extraction from rice using a Soxtec system. Cereal Chem. 82:282-
286.
Patindol, J. and Y.-J. Wang. 2002. Fine structures of starches from long-grain rice
cultivars with different functionality. Cereal Chem. 79:465-469.
Patindol, J.A., T.J. Siebenmorgen, Y.-J. Wang, S.B. Lanning, and P.A. Counce. 2014.
Impact of elevated nighttime air temperatures during kernel development on
starch properties of field-grown rice. Cereal Chem. 91:350-357.
USDA. 2014. United States Department of Agriculture. Inspection of Milled Rice.
In: USDA Rice Inspection Handbook. Access date: 26 January 2015. Available at:
http://www.gipsa.usda.gov/fgis/handbook/rice_inspec.aspx
409
Table 1. Ranges and means for the kernel properties of United

States-grown and imported medium- and short-grain rice cultivars.
Kernel Range [Mean] according to sample origin
property Arkansas (n = 6)a California (n = 5)a Imported (n = 14)a
Appearance
Length (mm) 5.4-5.8 [5.6] 5.0-5.8 [5.6] 4.1-6.6 [5.2]
Width (mm) 2.7-3.0[2.9] 2.9-3.0 [2.9] 1.9-3.6 [2.9]
Thickness (mm) 2.1-2.2 [2.1] 2.0-2.2 [2.1] 1.4-2.5 [2.1]
Length/Width 1.9-2.1 [2.0] 1.7-2.0 [1.9] 1.6-2.8 [1.9]
Whiteness (L*) 64.9-71.8 [69.7] 70.2-74.2 [72.2] 70.5-76.1 [73.5]
Yellowness (b*) 14.4-20.1 [16.2] 14.0-16.7 [15.7] 10.3-18.1 [14.4]
Chalk (% by area) 1.9-5.1 [3.2] 1.5-13.7 [4.7] 0.6-59.2 [14.3]
Gelatinization property
Onset GT (F) 149-153 [150] 145-147 [146] 125-166 [148]
Peak GT (F) 161-165 [162] 156-157 [157] 149-172 [159]
Range (F) 55-58 [56] 52-57 [55] 46-64 [55]
Enthalpy (J/g) 10.3-14.2 [11.7] 8.9-11.6 [10.9] 8.4-11.8 [10.4]
Pasting propertyb
Peak viscosity (cP) 3163-3727 [3414] 3031-3275 [3168] 1832-3838 [2984]
Final viscosity (cP) 2726-2946 [2855] 2760-2942 [2854] 2651-5613 [3631]
Breakdown (cP) 1329-1965 [1588] 1465-1710 [1582] 16-1947 [1152]
Setback (cP) (-811)-(-416) [-599] (-415)-(-197) [-315] (-760)-2411 [647]
Total setback (cP) 913-1154 [1029] 1194-1352 [1267] 1187-2757 [1799]
Gross composition
Amylose (%) 9.5-14.4 [12.4] 12.7-15.1 [13.4] 11.6-25.5 [17.4]
Protein (%) 5.6-6.9 [6.4] 5.7-6.7 [6.2] 4.4-8.0 [6.4]
Mineral (%) 0.33-0.58 [0.44] 0.25-0.55 [0.38] 0.22-0.51 [0.38]
Amylopectin structurec
A Chain (%) 26.6-27.3 [26.8] 27.5-28.0 [27.7] 22.4-28.8 [26.5]
B1 Chain (%) 45.1-46.5 [46.0] 46.1-46.6 [46.3] 46.0-51.3 [47.6]
B2 Chain (%) 13.7-14.3 [14.0] 13.2-13.6 [13.4] 12.8-14.1 [13.4]
B3 Chain (%) 13.0-13.4 [13.2] 12.3-12.8 [12.6] 12.0-13.2 [12.5]
a
n = sample size; GT = gelatinization temperature.
b
Calculation: Breakdown = Peak viscosity minus - Trough; Setback = Final viscosity - Peak
viscosity; Total setback = Final viscosity - Trough.
c
Amylopectin chains: A chain, 6-12 glucose units; B1 chain, 13-24 glucose units; B2 chain, 25-
36 glucose units; B3 chain, 37-65 glucose units.
410
Table 2. Correlation coefficients for the relationship of appearance/functional
properties with kernel chemical components and amylopectin fine structure (n = 25)a.
Amylose Proteinb Mineral A Chainc B1 Chain B2 Chain B3 Chain
Whiteness 0.43*d -0.17ns -0.16ns -0.08ns 0.25ns -0.37ns -0.29ns
Yellowness -0.10ns 0.53** 0.02ns 0.09ns -0.10ns -0.08ns 0.10ns
Chalkiness 0.27ns 0.23ns 0.40* -0.13ns 0.19ns -0.21ns -0.02ns
Onset GTe 0.44* -0.04ns -0.05ns -0.86** 0.59** 0.31ns 0.60**
Peak GT 0.38ns -0.02ns -0.02ns -0.84** 0.57** 0.32ns 0.60**
Range -0.52** 0.10ns 0.12ns 0.66** -0.54** 0.07ns -0.33ns
Enthalpy -0.07ns -0.35ns 0.29ns -0.15ns 0.06ns 0.21 ns 0.14ns
Peak viscosity -0.34ns -0.16ns 0.30ns 0.12ns -0.41* 0.53** 0.54**
Final viscosity 0.73** 0.06ns 0.02ns -0.63** 0.60** -0.11ns 0.17ns
Breakdown -0.70** -0.12ns 0.24ns 0.54** -0.65** 0.32ns 0.12ns
Setback 0.83** 0.13ns -0.12ns -0.64** 0.74** -0.36ns -0.10ns
Total setback 0.83** 0.12ns -0.01ns -0.63** 0.73** -0.35ns -0.08ns
a
n = sample size.
b
ns = not significant at 95% probability (P < 0.05; n = 25).
c
Amylopectin chains: A chain, 6-12 glucose units; B1 chain, 13-24 glucose units; B2 chain, 25-36 glucose units; B3 chain, 37-65 glucose units.
d
* = Significant at 95% probability (P < 0.05; n = 25); ** = Significant at 99% probability (P < 0.01; n = 25).
e
GT = gelatinization temperature.
411
Fig. 1. Milled rice kernels of some locally grown and imported medium- and short-grain
rice cultivars: Arborio (Italy), Bhutan (Bhutan), Caffey (Arkansas), Calrose (California),
Jupiter (Arkansas), Kalijira (Bangladesh), Morelos (Mexico), and Taigen-9 (Taiwan).
412
Fig. 2. Flour pasting profiles of Jupiter (from Arkansas)

and seven other medium- and short-grain milled rice samples.
413
ECONOMICS
Trans-Pacific Partnership:
What Can it Mean for the U.S. Rice Sector?
A. Durand-Morat1 and E.J. Wailes1
ABSTRACT
We assess the impact of the negotiated rice market access outcome of the Trans-Pacific
Partnership (TPP) and the initial request by the U.S. regarding market access to the
Japanese rice market. The results suggest the impact of TPP will be limited, resulting
in expansion of intra-TPP trade to the benefit of the U.S. and Vietnam.
INTRODUCTION
Negotiations on the Trans-Pacific Partnership (TPP) concluded on October 2015
after more than 5 years of intensive negotiations primarily on a few key sectors, includ-
ing rice (Honma, 2015). As a sensitive sector, the motivation for this study is to better
understand the economics negotiated outcome on rice between Japan and the U.S.
The objective is to assess the impact of TPP on the global and regional rice markets.
It complements current literature (e.g., Burfisher et al., 2014; Government of Japan,
2013) in two ways. First, our modeling framework disaggregates rice markets by type
and milling degrees, allows for imperfect substitution of imports and domestic produc-
tion, and incorporates key trade policies affecting rice markets. Finally, we calibrate
the model for 2013, to represent the current global rice economy.
Rice trade among TPP members in 2013 was 41.8 million cwt (1.9 million metric
tons, mmt), milled basis, 5.3% of global rice trade. The effective import tariff of TPP
member import policies in 2013 is 72%, primarily affecting trade of medium-grain
rice. Japan has the highest level of protection among TPP members, trailed by Malaysia
and Chile.
Most TPP partners grant low most-favored-nation (MFN) protection to all export-
ers into their rice markets, thereby generating minimal trade distortions (WTO-TAO,
2015). For instance, Australia, Brunei, Canada, New Zealand, and Singapore maintain
zero MFN import tariffs on rice. Chile, the U.S., and Vietnam apply low import tariffs
(below 10%). Mexico reinstated a 9% import tariff on paddy rice and 20% import tar-
iff on brown and milled rice since December 2014 (USDA/FAS, 2014) after years of
1
Research Scientist and Distinguished Professor, respectively, Department of Agricultural Economics
and Agribusiness, Fayetteville.
414
duty-free imports from all sources. Imports of U.S. rice into Mexico remain duty free
under the North American Free Trade Agreement (NAFTA). Peru maintains a price
band system intended to stabilize domestic prices within set bounds.
Japan has historically maintained high market prices and rice farm incomes
through trade protection, income support programs, and domestic diversion programs
(Takahashi and Honma, 2015), and currently keeps a tariff-rate-quota (TRQ) of 15.0
million cwt (682 thousand metric tons, tmt), milled basis with an over-quota tariff set
at the prohibitive level of US$2.90/Kg ( 341/kg). The TPP deal reached with Japan,
which includes country-specific TRQs with Australia and the U.S. is the most conten-
tious aspect of the negotiated outcome for rice (Table 1).
PROCEDURES
We use the RiceFlow model (Durand-Morat and Wailes, 2010) to assess the impact
of TPP on the global rice economy. RiceFlow is a spatial, supply-chain model of the
global rice economy. The model disaggregates the rice economy into 73 regional markets
and 9 rice commodities derived from the combination of rice type (long, medium, and
fragrant rice) and milling degree (paddy, brown, and milled rice).
We assume imperfect substitution between imports and domestic rice consumption
(Hertel et al., 2003). Previous studies show that Japanese consumers prefer domestic
over imported rice (Peterson et al, 2013; Aizaki, 2015). For more details of the model
specification see Durand-Morat and Wailes (2016). The model is simulated dynamically
through 2029, with each subsequent year dependent on the previous years outcome.
By 2029 we assume TPP will reach full implementation.
Scenarios
A benchmark scenario projects the international rice market up to year 2029 based
on assumptions on key exogenous variables including rice yields, population, gross
domestic product, and policies. Projected rice yields are obtained from FAPRI (2015).
Macroeconomic projections are obtained from IHS Global Insight (2015).
In 2013 the Government of Japan introduced the Plan to Create Vitality for Agri-
cultural, Forestry, and Fishery Industries and Local Communities (The Plan), aimed at
increasing the competitiveness of Japanese agriculture (OECD, 2014). We implement
in the model changes introduced by The Plan to define some of the following scenarios:
TPP1: assumes the TPP Agreement is implemented as negotiated starting
in 2017.
TPP2: assumes the TPP Agreement is implemented as negotiated starting in
2017 by all countries, except that Japan grants a 4.4 million cwt (200 tmt)
TRQ to the U.S. as follows: 1.1 million cwt (50 tmt) from year 1 to 3, and
annual increases of 0.33 million cwt (15 tmt) up to 4.4 million cwt (200 tmt)
by year 2029. This scenario approximates the initial U.S. negotiation request
of Japan rice market access.
415
TPP1R: in addition to assumptions of scenario TPP1, we assume that domestic

rice policy reforms in Japan advance as expected but result in no improved
rice farm productivity.
TPP2R: in addition to assumptions of scenario TPP2, we assume that domestic
rice policy reforms in Japan advance as expected but result in no improved
rice farm productivity.
TPP1RE: in addition to assumptions of scenario TPP1, we assume that do-
mestic rice policy reforms in Japan generate a positive impact on rice farm
productivity. We assume that policy reforms result in larger farms (Takahashi
and Honma, 2015) and decrease rice production costs by 8.5% over 5 years
(from 2018 to 2023).
TPP2RE: in addition to assumptions of scenario TPP2, we assume that the
rice reforms in Japan generate the same effect on productivity as that esti-
mated in scenario TPP1RE.

Impacts of TPP on the global rice market compared to the benchmark are projected
to be limited for three reasons: the small amount of trade among TPP members relative
to global trade, the already high level of trade integration among many of its members,
and the limited trade concessions granted by Japan (Table 2).
The TPP is projected to have an impact on rice supply and demand aggregates
across the TPP region, increasing intra-regional rice trade above benchmark growth
by 0.93% and 1.14% a year as a result of TPP1 and TPP2, respectively (Table 2). Total
exports and imports by TPP members increase as a result of the agreement. In nominal
terms, TPP1 and TPP2 project an increase of rice exports from the U.S. by 0.77 thousand
cwt (35 tmt) and 2.09 million cwt (95 tmt) a year, respectively, when fully implemented.
The Japanese rice policy reforms have negligible spillover effects onto the region.
The U.S. gains from the TPP agreement on rice, primarily in the medium-grain
segment. Results project a steady increase in U.S. rice production of 1.46% annually in
the benchmark from 2014 to 2029, which strengthens slightly up to 1.56% as a result
of TPP2 (Table 3). The TPP promotes the production of medium-grain rice at the ex-
pense of long-grain rice (Fig. 1). United States rice consumption is expected to remain
unchanged across the scenarios analyzed, which means that the increased production
will go entirely for export.
Medium- and short-grain rice production is increasing in the mid-South, primar-
ily in Arkansas where it reached 16.5 million cwt (750 tmt) in 2014. United States
medium- and short-grain production increases in response to increasing productivity.
We assume that medium-grain from the mid-South will have the same acceptance in
markets overseas as that from California. Based on the 2015 crop budgets for rice from
the University of California and the University of Arkansas Cooperative Extension,
and transportation cost data from private industry sources, we estimate the medium-
grain rice cost advantage of mid-South compared to California into Japan to be around
$200/ton. This leads us to conclude that, conditional on similar consumer acceptance,
416
switching medium-grain production to the mid-South results in significant gains in price

competitiveness for U.S. medium-grain rice.
The TPP initiative stands as a great opportunity to advance trade integration and
generate sizable benefits to its members. However, the impact of the negotiated TPP
outcome for rice is expected to be small for the U.S. rice sector as a whole. The gains
accrue primarily to the medium- and short-grain segment, conditional on the ability to
expand exports of high quality medium-grain with acceptance into the Japanese mar-
ket. We estimate no sizable gains for the U.S. long-grain sector as a result of TPP. We
acknowledge the concerns generated by increasing price competitiveness of Vietnam-
ese long-grain rice in several Western Hemisphere markets such as Mexico, although
currently poor rice quality is likely the main constraint facing Vietnamese rice in the
Western Hemisphere markets participating in TPP.
ACKNOWLEDGMENTS
We acknowledge support from the University of Arkansas System Division of
Agriculture, the Arkansas Rice Research and Promotion Board and the U.S. Department
of Agriculture under award No. 58-0111-15-001.
LITERATURE CITED
Aizaki, H. 2015. Examining substitution patterns between domestic and imported ag-
ricultural products for broccoli, kiwifruit, rice and apples in Japan. Japan Agricul-
tural Research Quarterly 49(2):143-48. Available at: http://www.jircas.affrc.go.jp
Burfisher, M.E., J. Dyck, B. Meade, L. Mitchell, J. Wainio, S. Zahniser, S. Arita, and
J. Beckman. 2014. Agriculture in the Trans-Pacific Partnership. Washington DC:
U.S. Department of Agriculture, Economic Research Service, Pub. No. ERR-176.
Available at: http://www.ers.usda.gov/media/1692509/err176.pdf
Durand-Morat, A. and E. Wailes. 2010. RiceFlow: A Multi-Region, Multi-Product,
Spatial Partial Equilibrium Model of the World Rice Economy. Staff Report SP-
03-2010. Department of Agricultural Economics and Agribusiness, University of
Arkansas, Fayetteville.
Durand-Morat, A. and E. Wailes. 2016. Assessment of the TPP Agreement for Rice.
Staff Report WP-01-2016. Department of Agricultural Economics and Agribusi-
ness, University of Arkansas, Fayetteville.
FAPRI. 2015. Food and Agriculture Policy Research Institute. FAPRI Elasticity Da-
tabase. Access date: August 2015. Available at: http://www.fapri.iastate.edu/tools/
elasticity.aspx
Government of Japan, Cabinet Secretariat of Japan. 2013. Unified Estimates by the
Government of Japan of the Economic Effects of Tariff Elimination. Available at:
http://www.cas.go.jp/jp/tpp/pdf/2013/3/130315_nourinsuisan-2.pdf
417
Hertel, T.W., D. Hummels, M. Ivanic, and R. Keeney. 2003. How confident can we
be in CGE-based assessments of free trade agreements? West Lafayette, Ind.:
Center for Global Trade Analysis, Purdue University, GTAP Working Paper No.
26. Available at: https://www.gtap.agecon.purdue.edu/resources/download/2883.
pdf
Honma, M. 2015. The TPP and Agricultural Reform in Japan. The political economy
of Japanese trade policy. A.G. Mulgan and M. Honma, eds. London, UK: Pal-
grame Macmillan.
IHS Global Insight. 2015. Global Economic Data. Available at: https://www.ihs.com/
products/global-economic-data.html
OECD. Organization for Economic Cooperation and Development. 2014. Agricul-
tural Policy Monitoring and Evaluation 2014: OECD Countries. Paris: OECD
Publishing. Available at: http://dx.doi.org/10.1787/agr_pol-2014-en
Peterson, H.H., J.C. Bernard, J.A. Fox, and J.M. Peterson. 2013. Japanese Consum-
ers Valuation of Rice and Pork from Domestic, U.S., and other Origins. J. Agric.
Res. Econ. 38(1):93-106.
Takahashi, D. and M. Honma. 2015. A Reexamination of the Agricultural Adjustment
Problem in Japan. Paper presented at the International Conference of Agricultural
Economists, Milan, Italy, August 8-14, 2015.
USDA/FAS. U.S. Department of Agriculture, Foreign Agricultural Service. 2014.
Mexico Announces Rice Import Tariffs. Washington DC: GAIN Report No.
MX4085, 2014. Available at: http://gain.fas.usda.gov/Recent%20GAIN%20Publi-
cations/Mexico%20Announces%20Rice%20Import%20Tariffs%20%20_Mexico_
Mexico_12-16-2014.pdf
WTO-TAO. World Trade Organization, Tariff Analysis Online. Access date: 6 No-
vember 2015. Available at: http://tariffdata.wto.org/
418
Table 1. Country-specific Tariff Rate

Quota granted by Japan to Australia and
the U.S. in the Trans-Pacific Partnership (TPP).
Quota Country-specific Tariff Rate Quota
Yeara U.S. Australia
--------- (million cwt)----------
1 1.10 0.13
2 1.10 0.13
3 1.10 0.13
4 1.14 0.14
5 1.19 0.14
6 1.23 0.15
7 1.28 0.15
8 1.32 0.16
9 1.36 0.16
10 1.41 0.17
11 1.45 0.17
12 1.50 0.18
13 1.54 0.18
a
For Year 14 and for each subsequent year, the aggre-
gate quota quantity shall remain at the level reached in
Year 13. Source: TPP Full Text available at https://ustr.
gov/trade-agreements/free-trade-agreements/trans-
pacific-partnership/tpp-full-text
419
Table 2. Average 2015-2029 annual change in global and regional rice production,
consumption, and trade by various Trans-Pacific Partnership (TPP) scenarios.
Bench TPP1a TPP1Rb TPP1REc TPP2d TPP2Re TPP2REf
Global 0.67% 0.67% 0.67% 0.67% 0.67% 0.67% 0.67%
production
LGg 0.73% 0.73% 0.73% 0.73% 0.73% 0.73% 0.73%
MG 0.13% 0.13% 0.13% 0.13% 0.13% 0.13% 0.13%
FR 1.21% 1.21% 1.21% 1.21% 1.21% 1.21% 1.21%
Global 0.68% 0.68% 0.68% 0.69% 0.69% 0.69% 0.69%
consumption
LG 0.75% 0.75% 0.75% 0.75% 0.75% 0.75% 0.75%
MG 0.14% 0.13% 0.13% 0.14% 0.13% 0.13% 0.14%
FR 1.03% 1.03% 1.03% 1.03% 1.03% 1.03% 1.03%
Global 2.09% 2.10% 2.10% 2.10% 2.12% 2.12% 2.12%
trade
LG 2.22% 2.22% 2.22% 2.22% 2.22% 2.22% 2.22%
MG 0.86% 1.10% 1.10% 1.10% 1.48% 1.48% 1.48%
FR 1.61% 1.62% 1.62% 1.62% 1.62% 1.62% 1.62%
TPP 0.4% 0.4% 0.4% 0.4% 0.4% 0.4% 0.4%
production
TPP 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%
consumption
TPP intra 1.9% 3.0% 3.0% 3.0% 3.3% 3.3% 3.3%
trade
TPP 2.2% 2.4% 2.4% 2.4% 2.4% 2.4% 2.4%
member exports
TPP 1.7% 1.9% 1.9% 1.9% 2.1% 2.1% 2.1%
member imports
a
Scenario TPP1: it assumes the TPP Agreement is implemented as negotiated starting in 2017.
b
Scenario TPP2: it assumes the TPP Agreement is implemented as negotiated starting in
2017 by all countries, except that Japan grants a 4.4 million cwt (200 tmt) TRQ to the U.S. as
follows: 1.1 million cwt (50 tmt) from year 1 to 3, and annual increases of 0.33 million cwt (15
tmt) up to 4.4 million cwt (200 tmt) by year 2029. This scenario approximates the initial U.S.
negotiation request of Japan rice market access.
c
Scenario TPP1R: in addition to assumptions of scenario TPP1, we assume that domestic rice
policy reforms in Japan advance as expected but result in no improved rice farm productivity.
d
Scenario TPP2R: in addition to assumptions of scenario TPP2, we assume that domestic rice
policy reforms in Japan advance as expected but result in no improved rice farm productivity.
e
Scenario TPP1RE: in addition to assumptions of scenario TPP1, we assume that domestic rice
policy reforms in Japan generate a positive impact on rice farm productivity. We assume that
policy reforms result in larger farms (Takahashi and Honma, 2015) and decrease rice produc-
tion costs by 8.5% over 5 years (from 2018 to 2023).
f
Scenario TPP2RE: in addition to assumptions of scenario TPP2, we assume that the rice
reforms in Japan generate the same effect on productivity as that estimated in scenario TP-
P1RE.
g
LG: long-grain rice. MG: medium- and short-grain rice. FR: fragrant rice.
420
Table 3. United States supply and demand situation at the

end of the implementation of Trans-Pacific Partnership (TPP)a.
Baseline TPP1 TPP2
Production
2029 Level (million cwt) 174.6 175.4 176.8
Av.An.Chgb 1.46% 1.50% 1.56%
Consumption
Av.An.Chg 1.30% 1.30% 1.30%
Net exports
Av.An.Chg 1.46% 1.56% 1.73%
a
The scenarios incorporating the impact of the reforms to the Japanese rice policies yield the
same results as the TPP scenarios (Table 2) and therefore are not included.
b
Average annual change estimated from the following 2015 figures: 142.4 million cwt of produc-
tion; 86.9 million cwt of consumption; 66.4 million cwt of net exports.
Fig. 1. Change in U.S. rice production by type as a result of the

Trans-Pacific Partnership (TPP). TPP1: this scenario assumes the TPP Agreement is
implemented as negotiated starting in 2017. TPP2: this scenario assumes the TPP
Agreement is implemented as negotiated starting in 2017 by all countries, except that
Japan grants a 4.4 million cwt (200 tmt) TRQ to the U.S. as follows: 1.1 million cwt (50 tmt)
from year 1 to 3, and annual increases of 0.33 million cwt (15 tmt) up to 4.4 million cwt
(200 tmt) by year 2029. This scenario approximates the initial U.S. negotiation request of
Japan rice market access. MG: medium- and short-grain rice. LG: long-grain rice.
421
ECONOMICS
Rice Enterprise Budgets and Production Economic Analysis

W.A. Flanders1
ABSTRACT
Crop enterprise budgets are developed that are flexible for representing alternative
production practices of Arkansas producers. Interactive budget programs apply methods
that are consistent over all field crops. Production practices for base budgets represent
University of Arkansas System Division of Agriculture Cooperative Extension rec-
ommendations from the Rice Research Verification Program. Unique budgets can be
customized by users based on either Extension recommendations or information from
producers for their production practices. The budget program is utilized to conduct
economic analysis of field data in the Rice Research Verification Program.
INTRODUCTION
Technologies are continually changing for rice production. Simultaneously, vola-
tile commodity prices and input prices present challenges for producers to maintain
profitability. Producers need a means to calculate costs and returns of production alter-
natives to estimate potential profitability. The objective of this research is to develop
an interactive computational program that will enable stakeholders of the Arkansas rice
industry to evaluate production methods for comparative costs and returns.
PROCEDURES
Methods employed for developing crop enterprise budgets include input prices
that are estimated directly from information available from suppliers and other sources,
as well as costs estimated from engineering formulas developed by the American Soci-
ety of Agricultural and Biological Engineers. Input costs for fertilizers and chemicals
are estimated by applying prices to typical input rates. Input prices, custom hire rates,
and fees are estimated with information from industry contacts. Methods of estimating
these operating expenses presented in crop enterprise budgets are identical to producers
obtaining costs information for their specific farms.
1
Associate Professor, Agricultural Economics and Agribusiness, Northeast Research and Extension
Center, Keiser.
422
Ownership costs and repair expenses for machinery are estimated by applying
engineering formulas to representative prices of new equipment (Givan, 1991; Lazarus
and Selly, 2002). Repair expenses in crop enterprise budgets should be regarded as value
estimates of full service repairs. Repairs and maintenance performed by hired farm
labor will be partially realized as wages paid to employees. Machinery performance
rates of field activities utilized for machinery costs are used to estimate time require-
ments of an activity which is applied to an hourly wage rate for determining labor costs
(USDA-NASS, 2015). Labor costs in crop enterprise budgets represent time devoted
to specified field activities.
Ownership costs of machinery are determined by the capital recovery method
which determines the amount of money that should be set aside each year to replace
the value of equipment used in production (Kay and Edwards, 1999). This measure dif-
fers from typical depreciation methods, as well as actual cash expenses for machinery.
Amortization factors applied for capital recovery estimation coincide with prevailing
long-term interest rates (Edwards, 2005). Interest rates in this report are from Arkansas
lenders as reported in November 2015. Representative prices for machinery and equip-
ment are based on contacts with Arkansas dealers, industry list prices, and reference
sources (Deere and Company, 2015; MSU, 2015). Revenue in crop enterprise budgets
is the product of expected yields from following Extension practices under optimal
growing conditions and projected commodity prices.

The University of Arkansas System Division of Agriculture Department of Agri-
cultural Economics and Agribusiness (AEAB) develops annual crop enterprise budgets
to assist Arkansas producers and other agricultural stakeholders in evaluating expected
costs and returns for the upcoming field crop production year. Production methods
analyzed represent typical field activities as determined by consultations with farmers,
county agents, and information from Crop Research Verification Program Coordina-
tors in the Department of Crop, Soil, and Environmental Sciences. Actual production
practices vary greatly among individual farms due to management preferences and
between production years due to climactic conditions. Analyses are for generalized
circumstances with a focus on consistent and coordinated application of budget methods
for all field crops. This approach results in meaningful costs and returns comparisons
for decision making related to acreage allocations among field crops. Results should
be regarded only as a guide and a basis for individual farmers developing budgets for
their production practices, soil types, and other unique circumstances.
Table 1 presents a summary of 2016 costs and returns for Arkansas dry-seeded,
delayed-flood conventional rice. Costs are presented on a per acre basis and with an
assumed 1000 acres. Program flexibility allows users to change total acres, as well as
other variables to represent unique farm situations. Returns to total specified expenses
are $330.51/acre. The budget program includes similar capabilities for Clearfield pure-
line, hybrid, and Clearfield hybrid cultivars, as well as water-seeded rice production.
423
Crop insurance information in Table 1 associates input costs with alternative

coverage levels for insurance. For example, with an actual production history (APH)
yield of 162.0 bu/acre and an assumed projected price of $5.50/bu, input costs could be
insured at selected coverage levels greater than 42%. Production expenses represent what
is commonly termed as out-of-pocket costs, and could be insured at coverage levels
greater than 47%. Total specified expenses could be insured at coverage levels of 74%.
The crop enterprise budget program has a state level component that develops
base budgets. County extension faculty can utilize base budgets as a guide to develop-
ing budgets that are specific to their respective counties, as well as customized bud-
gets for individual producers. A county delivery system for crop enterprise budgets is
consistent with the mission and organizational structure of the Arkansas Cooperative
Extension Service.
The benefits provided by the economic analysis of alternative rice production
methods provide a significant reduction in financial risk faced by producers. Arkansas
producers have the capability with the budget program to develop economic analyses
of their individual production activities. Unique crop enterprise budgets developed for
individual farms are useful for determining credit requirements. Flexible crop enterprise
budgets are useful for planning that determines production methods with the greatest
potential for financial success. Flexible budgets enable farm financial outlooks to be
revised during the production season as inputs, input prices, yields, and commodity
prices change. Incorporating changing information and circumstances into budget
analysis assists producers and lenders in making decisions that manage financial risks
inherent in agricultural production.
ACKNOWLEDGMENTS
This research is made possible by funding from the Arkansas Rice Research
and Promotion Board and the University of Arkansas System Division of Agriculture.
LITERATURE CITED
Deere and Company. 2015. Products, Agriculture. Moline, Ill. Access date: October
2015. Available at: http://www.deere.com/en_US/industry/agriculture/agriculture
Edwards, W. 2005. Estimating farm machinery costs. Iowa State University, Univer-
sity Extension, Ames, Iowa.
Givan, W. 1991. Using machinery costs to make decisions, AG ECON 91-003, Co-
operative Extension Service, Extension Agricultural Economics, The University
of Georgia, Athens, Ga.
Kay, R.D. and W.M. Edwards. 1999. Farm Management. Fourth Edition. WCB/
McGraw-Hill, Boston, Mass.
424
Lazarus, W.F. and R.A. Selly. 2002. Suggested procedures for estimating farm
machinery. Staff Paper P02-16, Department of Applied Economics, College
of Agricultural, Food, and Environmental Sciences, University of Minnesota,
Minneapolis-St. Paul, Minn. December 2002.
MSU. 2015. Mississippi State University. Planning Budgets. Department of Agricul-
tural Economics, Mississippi State University, Starkville, MS. Access date: Octo-
ber 2015. Available at: http://www.agecon.msstate.edu/whatwedo/budgets.asp
USDA-NASS. 2015. U.S. Department of Agriculture, National Agricultural Statistics
Service. Farm Labor. Washington, D.C., May 2015.
Table 1. Summary of 2016 revenue and expenses,

conventional rice, per acre and 1000 acres.
Summary of revenue and expenses Crop insurance information
Revenue Per acre Farm Per acre
Acres 1 1000
Yield (bu) 180.0 180,000 APHa Yield 162.0
Price ($/bu) 5.50 5.50 Projected price 5.50
Grower share 100% 100%
Total crop revenue 990.00 990,000 Revenue 891.00
Expenses Percent of revenue

Seed 33.84 33,840 4%
Fertilizers and nutrients 107.78 107,777 12%
Chemicals 99.53 99,531 11%
Custom applications 44.10 44,100 5%
Diesel fuel, field activities 14.71 14,713 2%
Irrigation energy costs 67.32 67,324 8%
Other inputs 5.15 5150 1%
Input costs 372.44 372,435 42%
Fees 0.00 0 0%
Crop insurance 6.00 6000 1%
Repairs and maintenance, 25.02 25,016 3%
includes employee labor
Labor, field activities 12.27 12,273 1%
Production expenses 415.72 415,724 47%
Interest 9.87 9873 1%
Post-harvest expenses 119.43 119,430 13%
Custom harvest 0.00 0 0%
Total operating expenses 545.03 545,028
Returns to operating expenses 444.97 444,972
Cash land rent 0.00 0 0%
Capital recovery and fixed costs 114.46 114,457 13%
Total specified expenses 659.49 659,485
Returns to specified expenses 330.51 330,515
a
APH = actual production history.
425
ECONOMICS
Cost-Effective Use of Water? Factors

Influencing the Use of Irrigation Technologies and
Water Management Practices by Arkansas Producers
Q. Huang1, K. Kovacs1, and Y. Xu1
ABSTRACT
This study examines factors that influence the decisions of rice and soybean producers
to use more efficient irrigation technology and/or Water Management Practices (WMPs).
Data from the Farm and Ranch Irrigation Survey (FRIS) are used. Most producers rely
on gravity irrigation, and about half also use one or more water management practices
(WMPs) to improve existing gravity systems. Producers growing more diverse crops
or those with smaller shares of land irrigated by groundwater are more likely to use
WMPs. Sprinklers are more likely to be used on larger farms and by producers with
less experience on farms. When facing the risk of more frequent droughts, producers
tend to use WMPs combined with gravity system as a response, not sprinklers.
INTRODUCTION
Arkansas is the largest producer of rice in the nation. Its agriculture is heavily
irrigated with irrigated acreage ranking fourth nationwide (Schaible and Aillery, 2012).
More than 60% of the states water supply comes from groundwater in the Mississippi
River Valley alluvial aquifer (USGS, 2008). An annual gap in groundwater as large as
7 million acre-feet is projected for 2050 (ANRC, 2015). In the focus groups conducted
by the authors in November 2014 with producers from east Arkansas, the decline in
groundwater supply was ranked among the top concerns.
Switching to more efficient irrigation technologies such as center pivot has often
been proposed as a solution to water shortage problems. Water management practices
(WMPs) could also conserve water because many could improve the performance of
existing irrigation systems (Schaible and Aillery, 2012). Without switching to more
efficient irrigation technology, a rice producer can augment water supply by building
a tailwater pit to capture and store irrigation runoff from flooded fields or rainfall and
reuse it for future irrigation. The producer can further reduce water use by improving
1
Associate Professor, Assistant Professor, and Research Associate, respectively, Department of Agricul-
tural Economics and Agribusiness, Fayetteville.
426
irrigation uniformity with practices such as laser leveling and alternate row irrigation.
This study investigates what factors affect producers use of irrigation technologies
and/or WMPs.
PROCEDURES
The main data sets used are the Farm and Ranch Irrigation Survey (FRIS), argu-
ably the most comprehensive data on irrigation, and Census of Agriculture collected
by the U.S. Department of Agriculture. It is a set of repeated cross-sectional data. We
used 1988, 1994, 1998, 2003, and 2008. County-level climate data such as daily pre-
cipitation and temperature are obtained from National Climatic Data Center (NCDC,
2014). All temperature and precipitation measures are calculated as the average of the
previous 30 years and only within growing seasons. To measure soil quality, the average
saturated hydraulic conductivity (Ksat) is extracted from the Soil Survey Geographic
database (SSURGO).
Statistical methods, in particular regression analysis methods such as Logit
regression and Multinomial Logit regression, are used to analyze the data. Since grav-
ity irrigation is the dominant irrigation system in rice production, rice producers only
decide on whether to use any WMPs. The FRIS survey only asked about WMPs used
under gravity system. So we have grouped all WMPs together. The Logit model is used
to estimate the choice of rice producers: Pr(yi = 1) = exp(X)/[1 + exp(X)], where yi
is a dummy variable that equals 1 if producer i uses one or more WMPs. The objective
is to estimate the vector of parameters that link X, which contains a set of relevant
factors, to yi. Soybean producers face one of three choices: gravity irrigation without
WMPs, gravity irrigation with WMPs, and sprinkler irrigation. Their choices are esti-
mated using the method of Multinomial Logit: Pr(yi = j) = exp(Xj)/lexp(Xl), where
j and l are both indices for the choices and is the vector of parameters to be estimated.

In Arkansas, the main irrigation method for all major crops is still gravity irrigation,
which includes both furrow and flood irrigation. In 2008, 96.5% of rice acreage was
under gravity system (Table 1). For all other major crops, more than 69% of irrigated
acres use gravity irrigation. Producers that utilize sprinklers are still in the minority.
Within sprinkler system, center pivot is the dominant technology. The percent of irrigated
area using center pivots ranged from 1.6% for rice to 33.1% for cotton. About half of
the producers also use one or more WMPs to improve existing gravity systems. Laser
leveling and tailwater pit are the most commonly used practices (Table 2). Among the
farms with gravity irrigation systems, 24.4% used laser leveling in 2008 and 24.0% had
tailwater pits on farm. Other popular WMPs include alternative row irrigations (15.0%)
and restricting runoff by diking the end of field (13.9%).
Farm size has a positive and statistically significant effect on the probability of
observing sprinkler systems among soybean producers (Table 3). Given the require-
427
ment of large capital investments, larger farms are more likely to enjoy economies of
scale. They are also more likely to have access to more credit for capital investment.
It usually requires more labor to irrigate larger farms, which may also push the switch
to sprinkler irrigation which is less labor intensive. Experience on farm has a negative
and statistically significant impact on the probability of using sprinklers, probably
because sprinklers require knowledge of new irrigation techniques. In contrast, since
most WMPs such as laser leveling and tailwater pits involve structural change to the
farm fields, knowledge of the farm, which increases with years of experience on farm,
facilitates the adoption of WMPs. So more experienced producers may prefer WMPs
over sprinkler irrigation. Farms with more diverse crop mixes tend to move away from
gravity only systems or sprinkler systems to the combination of gravity and WMPs.
Crop diversity is measured by the number of crop categories produced on a farm (e.g.,
grain crops, cash crops, fruits and vegetables, fodder crops). Producers with a more
diverse crop mix usually grow cash crops such as cotton in addition to grain crops. It
is likely those producers are using one or more WMPs to satisfy the irrigation demands
of different crops both in terms of timing and quantities. For example, because different
crops are grown in different times of the year, farms demand irrigation water for longer
periods during the year. Water management practices such as tailwater pits can meet
this demand by increasing water stored on farm.
The cost of water has a positive and significant effect on the likelihood of using
WMPs among soybean producers, but not among rice producers (Table 3). The cost of
water does not seem to affect the decision to use sprinkler irrigation. This is different
from the positive relationship found in most previous studies (e.g., Caswell and Zil-
berman, 1986; Negri and Brooks, 1990). The small or insignificant effect in our study
may be because the estimation is done conditional on crop choice. It may also because
Arkansas soybean producers have chosen to use WMPs as the response to the higher
cost of water. Greater reliance on groundwater increases the likelihood of using gravity
irrigation but discourages the use of WMPs. Strong reliance on groundwater suggests
this source of water is abundant. Groundwater is a preferred source of water because
the quantity of groundwater varies much less seasonally than surface water. A greater
reliance on groundwater, as measured by percentage of acres with groundwater irriga-
tion, reduces the need for WMPs such as tailwater pits when water is scarce.
None of the temperature or precipitation related factors seem to affect a produc-
ers decision to use WMPs (Table 3). Since rice is almost 100% irrigated, its growth
may be much less sensitive to climatic factors. Among soybean producers, the percent
of years with severe droughts, constructed using the Palmer Drought Severity Index
(Palmer, 1965; NIDIS, 2014), has a negative and statistically significant impact on the
likelihood of choosing the sprinkler system. Both sprinklers and WMPs can improve
control over irrigation applications. Arkansas soybean producers have chosen WMPs
as their responses to more frequent droughts. In contrast, producers in other regions
such as California have relied on sprinkler irrigation (Zilberman et al., 1995). This may
be because Arkansas crop production is less capital intensive than in other states such
as California. It may also be because policy makers in Arkansas have been promoting
428
WMPs such as tailwater pits as a way to increase surface water use (ANRC, 2015).
The policy nudge toward surface water use may also reduce the popularity of sprinklers
since modern irrigation technologies are more likely to be used on fields with ground-
water supplies because groundwater is usually delivered at higher pressure (Caswell
and Zilberman, 1986).
Conditional on crop choice, cost of water does not seem to have a large impact
on the use of sprinklers. Increasing water price is not an effective tool to promote the
use of sprinklers in Arkansas. Farm experience does not seem to increase the chance of
using sprinklers. More Extension efforts should be made to reduce producers learning
curve of practices that require new knowledge. Examples include modern irrigation
technologies, such as WMPs that can improve gravity systems such as computerized hole
selection and sophisticated irrigation scheduling practices like soil or plant moisture-
sensing devices and computer simulation model.
Although both more efficient irrigation technologies and WMPs can alleviate
climate risk, WMPs are used to mitigate the impact of more frequent droughts. Given
Arkansass recent drought experience in 2010/11 and the expectation of higher frequen-
cies of droughts in the future (Arkansas Governors Commission on Global Warming,
2008), it is important for agricultural extension specialists to provide producers with
climate information to manage farm risk and help them with both modern irrigation
technology and WMPs as tools to deal with a more volatile climate and extreme events
such as drought stress or frost. Since most modern irrigation technologies are capital
intensive and most WMPs require more labor, the large farms tend to prefer irrigation
technologies and the smaller farms tend toward WMPs.
ACKNOWLEDGMENTS
This project was supported by the Arkansas Rice Research and Promotion Board,
the Arkansas Water Resource Center, and the U.S. Geologic Survey of the 104b research
grant program, and the University of Arkansas System Division of Agriculture. We thank
Becky Cross, Regional Director, and Jill Bishop, Deputy Regional Director of USDAs
National Agricultural Statistics Service (NAAS) Delta Regional Field Office, Jim Burt
from NAAS Data Laboratory, for providing access to the data used in this research.
LITERATURE CITED
Arkansas Governors Commission on Global Warming. 2008. Arkansas Governors
Commission on Global Warming, Final Report. Little Rock, Ark.: Arkansas Gov-
ernors Commission on Global Warming.
ANRC. Arkansas Natural Resources Commission. 2015. 2014 Arkansas Water Plan
(AWP) Update.
429
Caswell, M. and D. Zilberman. 1986. The effects of well depth and land quality on
the choice of irrigation technology. Amer. J. Agric. Econ. 68:798-811.
NCDC. 2014. National Climatic Data Center. NOAA Global Summary of the Day
(GSOD) Climate Data. Available at: http://www.climate.gov/daily-observational-
data-global-summary-day-gsod-%E2%80%93-gis-data-locator
Negri, D.H. and D.H. Brooks. 1990. Determinants of irrigation technology choices.
West J. Agric. Econ. 15:213-23.
NIDIS. 2014. National Integrated Drought Information System. http://www.drought.
gov/drought/content/products-current-drought-and-monitoring-drought-indicators/
palmer-drought-severity-index
Palmer, W.C. 1965. Meteorologic drought. U.S. Dept. of Commerce Weather Bureau,
Washington, D.C., Res. Pap. No. 45.
Schaible, G.D. and M.P. Aillery. 2012. Water conservation in irrigated agriculture:
Trends and challenges in the face of emerging demands. Economic Information
Bulletin Number 99. Economic Research Service, United States Department of
Agriculture, Washington, D.C. (September). Available at: http://www.ers.usda.
gov/media/884158/eib99.pdf
USGS. United States Geological Survey. 2008. Water levels and selected water-qual-
ity conditions in the Mississippi River Valley alluvial aquifer in eastern Arkansas,
2006. Scientific Investigations Report 2008-5092.
Zilberman, D., A. Dinar, N. MacDougall, M. Khanna, C. Brown, and F. Castillo.
1995. Individual and institutional responses to drought: The case of California
agriculture. Working Paper, Department of Agricultural and Resource Economics,
University of California, Berkeley.
Table 1. Percent of irrigated area under each irrigation system in 2008.

Sprinkler system Gravity
Center pivot Other sprinkler system system
-----------------------------------(%)-------------------------------------
Corn 28.3 2.4 69.3
Soybean 20.3 3.1 76.6
Rice 1.6 1.9 96.5
Cotton 33.1 0.8 66.1
430
Table 2. Percent of farms with gravity systems that have used a practice in 2008.
System Farms
(%)
Laser leveling 24.4
Tailwater pits 24.0
Alternate row irrigations 15.0
Restricting runoff by diking end of field 13.9
Reducing irrigation set times or number of irrigations 4.57
Shorten furrow length 2.83
Surge flow or cablegation irrigation 0.65
431
Table 3. Factors that influence the choice of irrigation system and water management practices (WMPs).
432
Ricea Soybeanb
Gravity Gravity
and WMP Gravity and WMP Sprinkler
Experience on farm 0.00172 0.000863 0.000667 -0.00153*
(0.00557) (0.000862) (0.000809) (0.000645)
Farm size in acre 0.00123 -0.300 -0.179 0.479*
(0.00146) (0.337) (0.331) (0.226)
Crop diversity 0.36580*** -0.0437* 0.0738*** -0.0302
(0.12672) (0.0218) (0.0195) (0.0185)
% of land that is rented in -0.22969 0.0528 -0.00817 -0.0446
(0.19132) (0.0351) (0.0328) (0.0247)
Cost of water in $/acre-foot -0.00019 -0.0128 0.0247** -0.0118
(0.00174) (0.00975) (0.00892) (0.00649)
% of acres irrigated by groundwater -1.13546*** 0.271*** -0.312*** 0.0409
(0.35458) (0.0478) (0.0422) (0.0351)
Soil permeability -0.00545 0.000302 -0.000890 0.000588
*Farm size (0.00356) (0.00120) (0.00112) (0.000698)
Mean daily temperature -0.01923 0.00240 0.00452 -0.00692
*Farm sizec (0.02701) (0.00550) (0.00546) (0.00381)
Total precipitation 0.00718 0.00449 -0.00732 0.00282
*Farm sizec (0.03083) (0.00597) (0.00600) (0.00391)
% of years with severe droughts 0.02304 -0.0847 0.752*** -0.668***
*Farm sizec (2.01045) (0.245) (0.225) (0.192)
Share of days with intensive precipitation 0.31185 -0.00389 -0.196 0.200
*Farm sizec (0.74268) (0.209) (0.204) (0.123)
Year dummies YES YES
County dummies NO NO
Observations 1,298 1,704
Observations 1,298 1,704
Standard errors not reported for the sake of brevity; *significant at 10%; **significant at 5%, ***significant at 1%.
a
The average marginal effects estimated with the Logit regression are reported for rice.
b
The average marginal effects estimated with the Multinomial Logit regression are reported for soybean.
c
All temperature and precipitation measures are calculated as the average of previous 30 years and only within growing seasons.
ECONOMICS
Economic Simulation Analysis of Margin

Protection Crop Insurance in Arkansas Rice Production
R.U. Mane1 and K.B. Watkins1
ABSTRACT
Rice is an irrigated crop, and irrigated crops are more insulated against yield risk than
non-irrigated crops. However, rice is largely dependent on energy related inputs like
fuel and fertilizer and suffers from systemic risks caused by increasing energy related
input costs. The USDA Risk Management Agency (RMA) is making available a new
insurance product to rice producers in 2016 called Margin Protection (MP). Margin
Protection provides coverage against an unexpected decrease in operating margin result-
ing from increased input costs. This study used simulation to evaluate the likelihood of
indemnities generated by MP at various coverage levels ranging from 70% to 90% for
three major rice counties in Arkansas. The likelihood of receiving indemnities under
MP was small for 70% and 75% coverage levels. Indemnity probabilities were 0.8%,
5.2%, and 18% for Arkansas County at MP coverage levels of 80%, 85%, and 95%,
respectively. Indemnity probabilities for Poinsett and Desha Counties were higher at the
80%, 85%, and 90% coverage levels (6%, 18.8%, and 31.9%, respectively, for Poinsett;
6.6%, 16.8%, and 34.6%, respectively, for Desha). The higher indemnity probabilities
in Poinsett and Desha may be due to higher yield variability for those counties.
INTRODUCTION
Rice production is an expensive enterprise when compared to production of other
row crops such as corn, soybean, grain sorghum, and wheat. According to Arkansas
Crop Enterprise Budgets reports (2011 - 2015) the net operating cost per acre for rice
is the highest ($641.08) when compared to other row crops like cotton ($523.90),
corn ($571.80), sorghum ($317.12), and soybean ($313.74). However, rice is more
profitable ($436.40) when we compare the per acre returns to operating expenses with
other crops like cotton ($378.50), corn ($397.80), soybean ($376.66), and sorghum
($211.10). A typical farm involved in rice production requires physical assets (e.g.,
equipment, irrigation infrastructure) worth $2.5 million; whereas a soybean farm would
require assets less than a $1 million (Fahr, 2015). Therefore, it is important for the rice
1
Program Associate and Professor, respectively, Rice Research and Extension Center, Stuttgart.
433
producers to be profitable or have certain margins over their production costs for their
economic sustainability.
In May 2015, the Risk Management Agency (RMA) of the U.S Department of
Agriculture (USDA-RMA; 2015a,b) announced a new crop insurance product called
Margin Protection (MP) would be available for selected crops in selected counties. In
Arkansas, MP Policies are available for 26 rice-growing counties (Arkansas, Ashley,
Chicot, Clay, Craighead, Crittenden, Cross, Desha, Drew, Greene, Independence, Jack-
son, Jefferson, Lawrence, Lee, Lincoln, Lonoke, Mississippi, Monroe, Phillips, Poin-
sett, Prairie, Randolph, Saint Francis, White, Woodruff) starting in the 2016 crop year
(USDA-RMA, 2015a). Margin Protection is an area-based product, using county-level
estimates of average revenue and input costs to establish the amount of coverage and
indemnity payments. The objective of MP is to provide coverage against an unexpected
decrease in operating margin (revenue less input costs). Allowed MP inputs subject to
price change in rice production include diesel, urea, DAP (diammonium phosphate),
potash, and operating interest. Margin Protection can be purchased as a stand-alone
product or in combination with other crop insurance products such as Revenue Protec-
tion (RP) and Yield Protection (YP). Coverage levels for MP range from 70% to 90%
of expected margins and are purchased in 5% increments (USDA-FCIC, 2015).
PROCEDURES
The procedure to estimate or calculate indemnities for rice MP is based on
guidelines set forth by the USDAs Federal Crop Insurance Corporation (FCIC) Mar-
gin Protection Plan of Insurance Standards Handbook (USDA-FCIC, 2015). The
indemnity for rice MP Insurance (MPI) is calculated as MPI = TM - HM, where TM
equals the trigger margin and HM is the harvest margin. The TM is calculated as TM =
EM CL, where EM is the expected margin and CL is the percent of EM covered by
MP (70%, 75%, 80%, 85%, and 90%). The EM is calculated as EM = ER - EC, where
ER is Expected Revenue (projected rice price expected county rice yield) and EC
= Expected Costs, which is the sum of variable inputs (allowable input quantities of
diesel, urea, diammonium phosphate (DAP) and potash multiplied by their respective
projected prices), plus fixed input costs (for maintenance, chemicals, and application),
plus projected interest applied to variable and fixed input costs for a 6 month period.
The Harvest Margin (HM) is calculated as HM = HR - HC, where HR = Harvest
Revenue and HC = Harvest Costs. Both HR and HC are calculated the same as ER and
EC above with the exception that harvest prices are used in place of projected prices,
a harvest interest rate is used instead of a projected interest rate, and final county yield
is used in place of the expected county yield. Finally, net indemnities are calculated as
NMPI = MPI - PREMIUM, where NMPI is Net MP Indemnity, and PREMIUM is the
producers MP subsidized premium for the coverage level purchased. Margin protec-
tion premiums by county and coverage level for this study were obtained from the MP
price discovery prompt of the Margin Protection for Corn, Rice, Soybeans and Wheat
website (USDA-RMA 2015b).
434
Allowable input quantities and fixed input costs for maintenance, chemicals, and
application are obtained from Watts and Associates, Inc. on behalf of the RMA (Watts
and Associates, 2015). Expected county yields for each county (Arkansas, Poinsett, and
Desha) were obtained from the MP price discovery prompt of the USDA-RMA Margin
Protection for Corn, Rice, Soybeans and Wheat website (USDA-RMA 2015b). Final
county yields and projected and harvest prices for rice, urea, DAP, potash, and interest
were simulated as empirical distributions using SIMETAR (SIMulation of Economet-
rics To Analyze Risk; Richardson et al., 2008). All price simulations were based on
historical price data for the period 2006-2015, while all final county yield simulations
were based on rice county yield data for Arkansas, Poinsett, and Desha Counties for the
period 2005-2014. Five hundred iterations were simulated for each price and county
rice yield. For more information about the price and yield data used in the simulations,
see Mane and Watkins (2016).

Margin Protection Premiums by County and Coverage Level
Rice MP premiums are presented by county and coverage level in Table 1. Ar-
kansas County has the lowest premiums followed by Poinsett and Desha Counties. The
premium data reported in Table 1 are total premiums and producer premiums (total
premiums less producer subsidies). The amount of premium paid by the producer
increases and the producer subsidy decreases as coverage levels increase. In general,
premiums are based on yield performance of counties. Based on National Agricultural
Statistics Service historic data, Arkansas County has the highest yield on average for
the last 10 years when compared to Poinsett and Desha Counties (USDA-NASS, 2016).
Likewise, Poinsett County has higher yields on average for 6 years when compared with
Desha County. The premium difference is less between Poinsett and Desha Counties
at different coverage levels when compared to Arkansas County. Based on analysis by
Makki and Somwaru (2001) of RMA data, it was concluded that producers choice of
an insurance product is based on cost of premium, subsidy associated with the premium,
and level of coverage. We can assume the preference to purchase a MP policy will be
based on the same parameters as listed above.
Simulated Rice Margin Protection Indemnity Statistics by

County and Coverage Level
Simulated rice MP indemnities are presented by county and coverage level in
Table 2. Average simulated indemnities are largest for the higher coverage levels (80%,
85%, and 90%). However, both minimum and median indemnities for all coverage
levels and counties equal zero, implying the likelihood of receiving an indemnity for
MP as well as for other traditional insurance products such and YP and MP is small for
any given year. Probabilities of receiving and indemnity over 500 simulated outcomes
are presented in the right column of Table 2. The probability of receiving an indemnity
increases as coverage levels increase from 70% to 90%.
435
Probabilities of receiving an indemnity vary by county. Indemnity probabilities

are lowest for Arkansas County (Table 2). The probability of receiving an MP indem-
nity in Arkansas County at 80%, 85%, and 90% coverage levels is 0.8%, 5.2%, and
18% respectively. There is no indemnity for MP at 70% and 75% coverage levels for
Arkansas County. Poinsett and Desha counties have a probability of approximately 1%
and 3% of receiving an indemnity at 70% and 75% coverage levels and have higher
probabilities of receiving an indemnity than Arkansas County at the 80%, 85%, and
90% coverage levels. Based on the simulated results we can argue that MP indemnities
are more likely to be triggered for higher coverage levels and are larger in magnitude
for counties with relatively lower and more variable rice yields.
Simulated Rice Margin Protection Net Indemnity Statistics by

County and Coverage Level
Simulated rice MP net indemnities are presented by county and coverage level
in Table 3. The net indemnities on MP is the difference between total indemnity (Table
2) and the producer premium (Table 1) at each coverage level. In most instances, net
indemnities are negative on average, reflecting the fact that in most years producers will
receive no indemnity and will incur a premium cost (Table 3). This is also borne out by
the fact that both the minimum and median net indemnities reported for all coverage
levels equal the premiums reported in Table 1 by coverage level. However, average net
indemnities are positive for Poinsett and Desha Counties at the 90% coverage level,
implying that positive net indemnities occur more frequently at the highest coverage
level for these two counties (Table 3).
The likelihood of receiving positive net indemnities is greater at all coverage
levels for Poinsett and Desha Counties relative to Arkansas County (Table 3). Higher
net positive indemnities in Poinsett and Desha Counties reflects higher yield variability
and lower yields in these counties which are most likely to trigger an indemnity. On the
contrary, Arkansas County has lower relative rice yield variability and higher rice yields
and is less likely to trigger indemnities relative to the other two counties.
Input and Output Factors Affecting the Likelihood of Triggering

Rice Margin Protection Indemnities
This section describes some insight obtained by looking at individual simulated
iterations with triggered indemnities. We found in most instances that escalating fertil-
izer prices (urea, DAP, and Potash) coupled with low yields and low harvested prices
increased the likelihood of triggering an indemnity for MP, particularly at the 85% and
90% coverage levels for all selected counties. It would be difficult to rank which of
the following factors; price of urea, DAP, potash, Yield of rice or Harvest Price of rice
is the most important factor contributing to indemnity payment for margin protection
policy. It is interesting to note that upward movement in the price of diesel as an input
by itself or in combination with other inputs like urea, DAP or potash would least likely
436
trigger an indemnity payment. Almost all indemnity payments on the input side were
associated with higher input prices for urea, DAP, and potash. Likewise, lower yields
and lower harvest prices can equally contribute to trigger an indemnity.
Margin Protection is an insurance product instituted to protect against unexpected
decreases in operating margin resulting from unexpected increases in input prices, reduc-
tions in output prices, and lower than expected county yields. Based on our simulated
results, MP is more effective in addressing risk at higher coverage levels (80%, 85%,
and 90%) when prices of urea, DAP, and potash are higher and yield and harvest prices
are lower. We also found the likelihood of indemnities increases for counties with rela-
tively more yield variability and/or lower yields.
ACKNOWLEDGMENTS
This research was funded by the Arkansas Rice Research and Promotion Board
LITERATURE CITED
Fahr, R. 2015. More rice, fewer farms. Arkansas Ag Business, Vol 2, No. 1, March
2015.
Makki, S.S. and A. Somwaru. 2001. Evidence of adverse selection in crop insurance
markets. J, Risk Insur. 685-708.
Mane, R. and B. Watkins. 2016. Stochastic analysis of Margin Protection (MP)
crop insurance in Arkansas rice production. Selected Paper Presented at the
2016 Southern Agricultural Economics Association Annual Meeting, 6-9 Febru-
ary 2016, San Antonio, Texas. Available at: http://ageconsearch.umn.edu/bit-
stream/229877/2/Mane%20and%20Watkins%202016.pdf
Richardson, J.W., K.D. Schumann, and P.A. Feldman. (2008). SIMETAR.Simulation
& Econometrics to Analyze Risk.Simetar, Inc. College Station, Texas. (Manual)
USDA-FCIC. 2015. U.S Department of Agriculture, Federal Crop Insurance Corpo-
ration. Margin Protection Plan of Insurance Standards Handbook, FCIC 20260U
(July 2015). Access date: 27 August 2015. Available at: http://www.rma.usda.gov/
handbooks/20000/2016/20260u.pdf
USDA-NASS. 2016. U.S. Department of Agriculture, National Agricultural Statistics
Service. Quick States Searchable Database. Access date: 31 May 2016. Available
at: https://quickstats.nass.usda.gov/
USDA-RMA. 2015a. U.S Department of Agriculture, Risk Management Agency.
Crop year 2016 margin protection for corn, rice, soybeans and wheat county
availability. Access date 8 August 2015. Available at: http://www.rma.usda.gov/
policies/2016/16mpcountiestext.pdf
437
USDA-RMA. 2015b. U.S Department of Agriculture, Risk Management Agency.

Margin Protection for corn, rice, soybeans and wheat. Access date: 8 August
2015. Available at: http://www.rma.usda.gov/policies/mp/index.html
Watts and Associates, Inc. 2015. MP fixed input costs and allowable input quantities
subject to price change. Access date: 2 September 2015. Available at: http://mar-
ginprotection.com/Content/Files/Rice%20MP%20Costs%20and%20Inputs.pdf
Table 1. Rice Margin Protection premiums at

different coverage levels for selected counties in Arkansas.a
Coverage levels
County and premium type 70% 75% 80% 85% 90%
----------------------------------($/acre)----------------------------------
Arkansas
Total premium 2.16 4.58 8.77 15.37 25.22
Producer premium 0.89 2.06 3.95 7.84 14.12
Poinsett
Total premium 2.92 5.69 10.07 16.82 26.21
Desha
Total premium 3.51 6.67 11.61 18.90 28.90
a
Source: Risk Management Agency (RMA) MP Actuarial Data as of 8 December 2015. (USDA-
RMA. 2015b).
438
Table 2. Simulated indemnities by coverage level summary
statistics ($/acre), for Arkansas, Poinsett, and Desha Counties in Arkansas.
Coverage level Meana SDb CVc Minimum Median Maximum Indemnity
(%) ------------------------------------------------------------($/acre)----------------------------------------------------------- (%)
Arkansas County
70 0.00 0.00 --- 0.00 0.00 0.00 0.0
75 0.00 0.00 --- 0.00 0.00 0.00 0.0
80 0.10 1.28 13.16 0.00 0.00 23.96 0.8
85 0.79 4.55 5.79 0.00 0.00 57.44 5.2
90 3.86 11.14 2.88 0.00 0.00 90.92 18.0
Poinsett County
70 0.16 2.33 14.43 0.00 0.00 41.38 0.8
75 0.54 3.97 7.41 0.00 0.00 55.88 3.0
80 1.58 7.54 4.76 0.00 0.00 70.38 6.0
85 4.39 13.20 3.00 0.00 0.00 84.88 18.8
90 10.85 21.40 1.97 0.00 0.00 115.81 31.6
Desha County
70 0.09 1.22 14.32 0.00 0.00 23.22 0.6
75 0.44 3.14 7.12 0.00 0.00 37.86 3.2
80 1.45 6.90 4.76 0.00 0.00 67.59 6.6

85 4.37 12.91 2.96 0.00 0.00 98.17 16.8
90 11.46 21.79 1.90 0.00 0.00 128.75 34.6
a
Indemnity statistics are based on 500 simulated iterations.
b
SD = Standard deviation.
c
CV = Coefficient of variation, a unitless measure of relative risk equal to the ratio of the standard deviation (SD) divided by the mean multiplied by 100.
439
Table 3. Simulated net indemnities by coverage level summary
440
statistics ($/acre), for Arkansas, Poinsett, and Desha counties in Arkansas.
Positive net
Coverage level Meana SDb CVc Minimum Median Maximum indemnity
(%) ---------------------------------------------------------($/acre)-------------------------------------------------------------- (%)
Arkansas County
70 -0.89 0.00 0.00 -0.89 -0.89 -0.89 0.00
75 -2.06 0.00 0.00 -2.06 -2.06 -2.06 0.00
80 -3.85 1.28 -0.33 -3.95 -3.95 20.01 0.60
85 -7.05 4.55 -0.65 -7.84 -7.84 49.60 3.20
90 -10.26 11.14 -1.09 -14.12 -14.12 76.80 9.40
Poinsett County
70 -1.04 2.33 -2.25 -1.20 -1.20 40.18 0.80
75 -1.98 4.16 -2.10 -2.56 -2.56 54.68 3.00
80 -2.75 8.21 -2.99 -4.53 -4.53 69.18 6.00
85 -2.80 15.34 -5.48 -8.58 -8.58 83.68 17.80
90 0.46 26.41 57.28 -14.68 -14.68 114.61 31.20
Desha County
70 -1.35 1.22 -0.90 -1.44 -1.44 21.78 0.60
75 -2.51 3.36 -1.34 -3.00 -3.00 36.42 3.00
80 -3.52 7.67 -2.18 -5.22 -5.22 66.15 6.40
85 -3.89 15.36 -3.94 -9.64 -9.64 96.73 16.20
90 0.38 27.30 71.76 -16.18 -16.18 127.31 33.20
a
Indemnity statistics are based on 500 simulated iterations.
b
SD = Standard deviation.
c
CV = Coefficient of variation, a unitless measure of relative risk equal to the ratio of the standard deviation (SD) divided by the mean multiplied by 100.
ECONOMICS
World and U.S. Rice Outlook:

Deterministic and Stochastic Baseline Projections, 2015-2025
E.J. Wailes1 and E.C. Chavez1
ABSTRACT
Global rice trade and prices are projected to grow steadily over the next decade as exports
expand in Cambodia, Thailand, Myanmar, and Vietnam, and imports continue to grow
strongly in the Western African countries and the Middle East. The U.S remains a top
five rice exporter in the world with exports accounting for nearly half of total U.S. rice
output. Growth in global rice production comes mainly from yield improvements as
limited area for expansion remains constrained. Abundant rice supplies are projected
due to current surpluses, productivity growth using new technologies, and increases
in production subsidies in many countries. Population remains the driver of global
consumption growth as per capita use declines slightly. Inherent risks in the global rice
economy due to the uncertainties of weather, domestic policies, and political develop-
ments, generate unexpected prices and trade. Hence historic statistical distributions of
yields are used for stochastic analysis, generating probable upper and lower bounds
(confidence intervals) of future distribution of prices and trade.
INTRODUCTION
This document contains baseline rice projections from the Arkansas Global Rice
Economics Program (AGREP) under the University of Arkansas System Division of
Agriculture Department of Agricultural Economics and Agribusiness, Fayetteville.
The purpose of this outlook is not to predict, but to present the current state and the
expected directions of the rice economies in the world over the next decade by assess-
ing potential supply and demand paths as well as the degree of variability on some of
the key variables.
Until 2014, the global rice market was dominated by Thailands controversial
and costly paddy pledging program (PPP), a politically motivated price-floor support
policy for Thai farmers (Wailes and Chavez, 2013). With the end of this costly program,
Thailand has re-captured its leadership position in global rice trade (USDA-FAS, 2014).
Thai rice is now priced competitively and is used in this study as the world long-grain
reference price.
1
Distinguished Professor and Program Associate II, respectively, Department of Agricultural Economics
and Agribusiness, Fayetteville.
441
PROCEDURES
The deterministic and stochastic baseline estimates presented in this report are
generated using the Arkansas Global Rice Model (AGRM), a partial equilibrium, non-
spatial, multi-country/regional statistical simulation and econometric framework that
covers 61 rice-producing and -consuming countries/regions developed and maintained
by the AGREP.
Most of the details and the theoretical structure of the Arkansas Global Rice Model,
with the exception of the newly added countries, can be found in online documentation
by Wailes and Chavez (2011). Historical rice data are sourced from USDA-FAS (2016)
and USDA-ERS (2016); and the macroeconomic data are from IHS Global Insight
through Food and Agricultural Policy Research Institute-Missouri (FAPRI-MU, 2016).
The baseline projections are grounded in a series of assumptions as of January 2016
about the general global economy, agricultural policies, weather, and technological
change. The basic assumptions include the following: a continuation of existing poli-
cies, IHS (IHS, Inc.) macroeconomic projections, no new World Trade Organization
(WTO) trade reforms, and average weather conditions. In light of the historical volatility
of the global rice economy, a stochastic analysis is included in this report to provide a
better understanding of the probable distribution of future price and trade outcomes.

Deterministic Analysis
Over the next decade, growth in global domestic use is projected to exceed domes-
tic supply causing increasing total rice trade and steadily increasing nominal long-grain
international prices. Major rice-deficit countries continue to import as domestic produc-
tion falls short of domestic demand despite subsidies and expressed desires to achieve
self-sufficiency. The average long-grain rice international reference price (Thai 100%B)
increases from $405/metric ton (mt) (2013-2015 average) to $545/mt in 2025 (Wailes
and Chavez, 2016).2 Over the same period, international medium-grain rice prices are
projected to remain at a relatively higher level, ranging from $828-845/mt (Table 1), as
segmentation remains in trade flows and prices of long- and medium-grain rice markets.
While Thai prices and other exporter prices have recently converged, Western
Hemisphere (U.S. and Mercosur) prices remain substantially higherwith margins to
Asian prices reaching as high as nearly $200/mt in September 2015, an unsustainable
level. Margins have narrowed since then and are projected to further decline to a more
historically consistent level of about $65/mt by 2025, the end of the projection period
(Table 1). Rigidity in moving the margin narrower is based on quality differences and
negotiated tariff preferences for U.S. rice in bilateral and regional trade agreements.
Over the projection period, India and the Peoples Republic of China (PRC) will
continue to account for the bulk of the global rice economy. In combination, these two
Although complete baseline projections for supply and demand variables are generated for all 61 countries/
2
regions covered by the AGRM, only selected variables are included in this report due to space consideration.
442
countries are projected to account for 35.3% of the world population from 2015-2025.
Over the same period, they will have an average combined share of 45.3% of world rice
area harvested, 51.2% of total milled rice production, 50.2% of total rice consumption,
and 66.2% of world rice ending stocks.
Rice output is projected to expand over the next decade, driven by the use of
higher-yielding varieties and hybrids and other improved production technologiesin
line with more subsidies for self-sufficiency programs of major consuming countries.
World production is projected to expand by 50.2 million metric tons (mmt) over the
next decade, equivalent to a growth of 1.0% per year; and reaching a total global output
of 525.8 mmt in 2025 with 90% of the growth from yield improvement and 10% from
increases in area harvested (Table 2). By volume, 34.1% of the expected output growth
will come from India; 40.9% from the seven countries of Bangladesh, Indonesia, Thai-
land, Cambodia, Myanmar, Philippines, and Vietnam combined; and 9.0% from the
15-member Economic Community of West African States (ECOWAS). Rice output of
the China however, declines by a total of 1.2 mmt, and those of Japan and South Korea
decline by a combined 1.1 mmt over the same period. Total U.S. rice production, on
the other hand, is projected to increase by a total of about 1.1 mmt (or 36.6 mil. cwt)
over the same period, equivalent to 1.6% annual growthwith annual growth of 1.0%
in area harvested and 0.6% annual yield growth (Table 3).
Factors driving rice consumption are income, population, and other demographic
variables. Rising incomes dampen rice demand in some Asian countries where rice is
considered an inferior good. These countries include Japan, Taiwan, PRC, Vietnam,
and Thailand. Demographic trends also weaken rice demand as aging populations and
increasing health consciousness shift preferences away from carbohydrates and towards
protein-based diets.
Over the baseline, global rice consumption is projected to increase by 43.6 mmt
reaching 524.9 mmt in 2025a growth of 0.8% per year, with population growth of
1.1% per year projected to be offset partly by a 0.3% decline in average world rice per
capita use (Table 2).
About 31.3% of the total growth is accounted for by India; 24.6% by the four
countries of Bangladesh, Indonesia, the Philippines, and Myanmar combined; and 17.8%
by ECOWAS. United States rice total consumption increases by nearly 280 thousand
metric tons (tmt) or 9.4 million hundred weight (cwt) over the same period, reaching
4.3 mmt (135.0 million cwt) in 2025 or an annual growth of 0.7%; which is solely
coming from population growth as per capita use declines slightly. Global rice stocks
have tightened recently, with stocks-to-use at 18.4% in 2015, the lowest on record over
the last 30 years. The reason for this situation is that consumption exceeded produc-
tion over the last 3 years, coupled with the disposal of excess rice stockpiles by India
and Thailand amounting to nearly 22 mmt from 2013-2015. Global stocks-to-use is
projected to decline further to 16.0% in 2019 before recovering gradually to 17.6% by
the end of the projection period, which to a certain extent, provides underlying support
to the projected steady increase in international prices.
Total global rice trade expands 1.4%/year, reaching 48 mmt in 2025 up from 42
mmt in 2015 (Table 1). On the exporters side, the significant investment in production
443
and processing capacity in Mekong Delta in Vietnam, Cambodia and Myanmar bodes
well for these countries increasing their role as major rice suppliers in the coming
years. As low-cost producers, these countries are well-poised geographically to supply
the steadily growing China market. The productivity gains from hybrids and Global
Rice Science Partnership (GRiSP) research are expected to have positive impacts on
Asian and African rice economies. As a result of Thailands termination of its costly,
unpopular, and controversial PPP, the country has resumed its leadership position in
global trade since 2015 and is projected to maintain a strong presence in the interna-
tional rice market over the next decadegiven its good infrastructural resources and
concerted focus on developing and maintaining a strong presence in the branded high
quality rice market. Based on USDA-FAS data (2016), the country has liquidated about
7.6 mmt of its excess stocks over the last 3 years. For the U.S., total rice exports expand
by 622 tmt or 20.3 million cwt over the next decade, reaching 3.8 mmt (or 119 million
cwt) in 2025; and total imports grow by 64 tmt (or 2.0 million cwt), reaching 828 tmt
(or 26 million cwt) in 2025. For reference purposes, a detailed U.S. rice supply and use
in English units is presented in Table 3. Cambodias exports are projected to expand at
11.6%/year, reaching 2.9 mmt in 2025 as both area and yield growth cause production
to exceed consumption consistently (Table 1). Myanmars exports, on the other hand,
are projected to expand from 1.8 mmt in 2015 to 2.9 mmt in 2025, supported by yield-
based growth in production.
On the demand side, while China remains an important major rice importer over
the next decade, its imports are relatively flat as it maintains a reasonable stock level.
Nearly 72% of the growth in global imports will come from Africa, with the ECOWAS
accounting for 79% of the growth in African imports. In general, expansion in imports
is associated with a combination of lagging production relative to consumption and
population growth.
Stochastic Analysis
The detailed results of the stochastic analyses for selected prices and trade are
presented in Figs. 1 through 5. In order to show the direction and dispersion of the
stochastic outcome distribution, four selected outcome items (stochastic average, 10th
percentile, 90th percentile, and the coefficient of variation) for selected variables are
presented. Intuitively, the gap between the two percentiles (10th and 90th) indicates
volatility or risk. Widening indicates increased volatility and narrowing indicates de-
creased volatility. Another measure of dispersion used is the coefficient of variation
(C.V.) which shows the extent of variability of data points in relation to the mean. Lower
C.V. indicates more stability, i.e., less risk. The information projected in each one of the
charts is similar in principle. Hence, for space consideration only one representative
chart (Fig. 1) will be discussedwhich can then be used as a basis to understanding the
remaining charts. Figure 1 shows the long-grain rice international reference price. For
2016, while the deterministic mean price is $456/mt (Table 1), the stochastic distribu-
tion indicates that 10% of the time the average price will be higher than $569/mt and
10% of the time lower than $426/mt (Fig. 1). The computed C.V. for 2016 is 11.0%,
444
which declines steadily throughout the projection period, reaching 8.1% by 2025. This
feature of the stochastic analysis provides an advantage as it indicates how the possible
outcomes are distributed, thus providing a better understanding of the dynamics of the
global rice market.
Understanding the market and policy forces that drive the global rice market are
beneficial for Arkansas rice stakeholders. This is especially true because Arkansas is
the top rice-producing state in the U.S. accounting for 45% of the countrys rice output;
and nearly half of Arkansas annual rice crop is exported. Market prices received by
Arkansas rice producers are primarily determined by the factors that affect interna-
tional trade. These include changes in rice production and consumption patterns, the
economics of alternative crops, domestic and international rice trade policies, as well
as the general macroeconomic environment in which global rice trade is transacted.
While the results presented in this outlook are not predictions, they can be considered
as a synthesis of the impacts of these factors, and serve to indicate what could happen
over the next decadeand could serve as baseline reference for further analysis. The
estimates are intended for use by government agencies and officials, farmers, consumers,
agribusinesses and other stakeholders who conduct medium- and long-term planning.
ACKNOWLEDGMENTS
The authors thank the University of Arkansas System Division of Agriculture
for its support; as well as the Arkansas rice farmers who provided support through the
Rice Research and Promotion Board to develop, update, and maintain the Arkansas
Global Rice Model. The model was updated twice in 2015 and once in January 2016
in collaboration with FAPRI-Missouri (FAPRI-MU, 2016) which provided the bulk
of the funds for the global rice modeling program. New country submodels for ten
Latin American countries were added in 2015, expanding AGRM coverage from 51 to
61countries/regions.
LITERATURE CITED
FAPRI-MU. 2016. U.S. Baseline Briefing Book. FAPRI-MU Report #02-16. Univer-
sity of Missouri-Columbia. Access date: 1 February, 2016. Available at: http://www.
fapri.missouri.edu/wp-content/uploads/2016/03/FAPRI-MU-Report-02-16.pdf
USDA-ERS. 2016. United States Department of Agriculture Economic Research
Service. Rice Outlook. RCS-16A (by Nathan Childs), 14 January 2016. Access
date: 14 January 2016. Available at: http://usda.mannlib.cornell.edu/usda/current/
RCS/RCS-01-14-2016.pdf
445
USDA-FAS. 2016. United States Department of Agriculture Foreign Agriculture Ser-

vice. Production, Supply, and Distribution (PSD) Online. Access date: 12 January
2016. Available at: http://apps.fas.usda.gov/psdonline/psdquery.aspx
Wailes, E.J. and E.C. Chavez. 2011. 2011 Updated Arkansas Global Rice Model.
University of Arkansas Department of Agricultural Economics and Agribusiness,
Division of Agriculture Staff Paper 2011-01. Access date: 1 February 2016. Avail-
able at: http://ageconsearch.umn.edu/handle/102650
Wailes, E.J. and E.C. Chavez. 2013. World rice outlook: International rice baseline
projections, 2012-2022. In: R.J. Norman and K.A.K. Moldenhauer (eds.). B.R.
Wells Rice Research Studies 2013. University of Arkansas Agricultural Experi-
ment Station Research Series 609:401-417. Access date 1 February 2016. Avail-
able at: http://arkansasagnews.uark.edu/609-52.pdf
Wailes, E.J. and E.C. Chavez. 2016. International Rice Outlook. International Rice
Baseline Projections, 2015-2015. University of Arkansas Department of Agri-
cultural Economics and Agribusiness, Division of Agriculture Staff Paper SP 01
2016. Available at: http://ageconsearch.umn.edu/handle/235158
446
Table 1. World rice total trade by country and international reference prices, 2014-2025.
Country Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
(mt)
Exporters
United States (1000 mt) 3207 3273 3412 3490 3576 3655 3695 3714 3723 3734 3749 3783
Thailand (1000 mt) 9200 10197 9482 9489 9551 9564 9604 9905 10405 10762 11295 11723
Pakistan (1000 mt) 4000 4584 4125 4166 4212 4276 4250 4340 4314 4328 4273 4235
Myanmar (1000 mt) 1750 1763 1741 1949 2045 2180 2251 2388 2495 2644 2748 2889
Vietnam (1000 mt) 6500 7177 7006 6967 7114 7215 7455 7428 7389 7555 7539 7504
Peoples (1000 mt) 426 450 523 596 650 686 730 776 824 873 905 929
Republic of China
India (1000 mt) 11871 8524 9132 8902 8666 8530 8579 8423 8322 8174 8142 8237
Cambodia (1000 mt) 1100 798 1116 1374 1588 1819 2035 2203 2439 2627 2736 2894
Lao PDR (1000 mt) -125 -113 -33 25 54 71 107 125 175 219 264 274
Australia (1000 mt) 370 97 221 329 420 484 525 549 562 567 571 573
Egypt (1000 mt) 250 397 425 464 489 506 515 516 513 508 503 493
Turkey (1000 mt) 27 27 27 27 27 27 27 27 27 27 27 27
European (1000 mt) 272 329 329 330 332 334 333 333 333 333 333 333
Union 28
Brazil (1000 mt) 1000 886 902 929 906 912 916 911 913 913 913 913
Cote dIvoire (1000 mt) 30 30 30 30 30 30 30 30 30 30 30 30
Senegal (1000 mt) 10 10 10 10 10 10 10 10 10 10 10 10
Guinea (1000 mt) 50 50 50 50 50 50 50 50 50 50 50 50
Tanzania (1000 mt) 30 30 30 30 30 30 30 30 30 30 30 30
Japan (1000 mt) 80 80 80 80 80 80 80 80 80 80 80 80
Argentina (1000 mt) 400 514 538 539 540 540 538 540 540 540 541 544
Uruguay (1000 mt) 800 948 897 940 965 990 1012 1025 1028 1037 1035 1034
Paraguay (1000 mt) 410 517 553 580 607 634 661 680 709 737 767 795
Peru (1000 mt) 20 68 75 80 83 87 88 90 91 92 93 94
ROW and (1000 mt) 1415 1296 1225 1184 1141 1128 1094 1073 1024 980 935 925
Residual
Total Exports (1000 mt) 43093 41932 41895 42563 43165 43839 44614 45246 46027 46851 47568 48402
continued
447
Table 1. Continued.
448
Country Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
(mt)
Importers
United States (1000 mt) 783 776 769 764 771 791 815 821 820 824 828 828
Thailand (1000 mt) 300 400 333 344 359 346 350 352 349 350 350 350
Pakistan (1000 mt) 30 35 32 32 33 33 33 33 33 33 33 33
Vietnam (1000 mt) 400 400 400 400 400 400 400 400 400 400 400 400
Peoples (1000 mt) 4315 4712 4081 3982 3985 3984 3995 4013 4030 4051 4076 4097
Republic of China
China- (1000 mt) 340 364 373 379 391 393 397 400 404 407 411 414
Hong Kong
Egypt (1000 mt) 34 25 25 25 25 25 25 25 25 25 25 25
Japan (1000 mt) 650 682 682 682 682 682 682 682 682 682 682 682
Bangladesh (1000 mt) 1224 646 784 695 790 873 817 798 817 848 879 957
Indonesia (1000 mt) 1100 2066 2075 2109 2075 2019 1950 1942 1867 1759 1628 1525
Iraq (1000 mt) 1100 1201 1395 1488 1562 1618 1681 1742 1795 1910 2009 2116
Iran (1000 mt) 1400 1681 1640 1651 1729 1768 1847 1871 1918 1971 2066 2097
Malaysia (1000 mt) 1000 1001 1062 1044 1060 1047 1045 1045 1049 1062 1076 1083
Philippines (1000 mt) 1800 2077 1404 1602 1709 1819 1809 1713 1645 1670 1633 1618
Saudi Arabia (1000 mt) 1460 1556 1511 1562 1593 1625 1660 1695 1731 1765 1799 1832
European (1000 mt) 1703 1512 1520 1537 1555 1574 1597 1621 1639 1660 1680 1700
Union 28
Singapore (1000 mt) 300 303 307 313 315 316 319 320 322 323 324 325
Brunei (1000 mt) 40 41 41 43 44 45 46 47 48 50 50 52
Darussalam
Turkey (1000 mt) 317 291 288 263 248 239 231 231 236 239 236 234
South Korea (1000 mt) 450 471 409 409 409 409 409 409 409 409 409 409
Taiwan (1000 mt) 125 126 126 126 126 126 126 126 126 126 126 126
Australia (1000 mt) 150 181 171 168 165 162 159 156 153 150 147 144
Brazil (1000 mt) 450 742 812 807 779 679 640 500 482 449 444 412
Mexico (1000 mt) 698 720 762 767 780 794 813 822 834 847 862 873
Canada (1000 mt) 350 361 366 383 396 404 414 422 430 438 446 454
continued
Table 1. Continued.
Country Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
(mt)
Importers, continued
Cote dIvoire (1000 mt) 1300 872 1049 1085 1160 1223 1276 1319 1348 1394 1449 1473
Nigeria (1000 mt) 3500 2522 3137 3147 3116 3208 3374 3530 3660 3813 3915 4026
South Africa (1000 mt) 980 989 990 1026 1033 1043 1062 1080 1086 1106 1097 1128
Senegal (1000 mt) 1200 1003 1110 1138 1158 1185 1201 1224 1256 1292 1327 1364
Ghana (1000 mt) 550 606 651 693 734 745 780 809 832 867 901 931
Cameroon (1000 mt) 525 546 572 631 665 712 745 765 795 822 855 888
Mozambique (1000 mt) 480 533 523 554 574 595 643 667 703 732 760 782
Guinea (1000 mt) 300 359 390 427 439 454 469 488 499 517 531 543
Kenya (1000 mt) 420 522 517 538 572 577 594 608 632 650 668 682
Tanzania (1000 mt) 100 169 168 176 172 189 208 196 189 185 176 176
Sierra Leone (1000 mt) 220 274 168 149 124 115 113 114 103 99 95 101
Mali (1000 mt) 180 166 130 104 73 63 74 69 67 80 73 84
Liberia (1000 mt) 340 361 383 405 425 444 468 490 518 540 561 589
Colombia (1000 mt) 300 348 336 351 355 358 356 358 363 369 378 383
ECOWAS 7a (1000 mt) 1436 1490 1585 1709 1815 1905 1998 2090 2186 2284 2389 2487
Cuba (1000 mt) 483 505 517 523 525 526 535 542 546 552 562 566
Costa Rica (1000 mt) 150 103 108 110 110 108 111 114 117 120 122 126
Dominican (1000 mt) 15 18 18 18 19 16 17 17 19 19 20 21
Republic
Guatemala (1000 mt) 88 78 79 80 83 86 89 92 95 99 102 105
Honduras (1000 mt) 134 150 144 145 145 144 147 149 152 155 159 163
Nicaragua (1000 mt) 70 74 80 79 78 77 79 81 83 86 89 92
Panama (1000 mt) 76 81 86 89 91 94 98 99 100 102 102 102
Chile (1000 mt) 127 122 146 144 143 144 145 147 150 153 156 159
Paraguay (1000 mt) 2 2 2 2 2 2 2 2 2 2 2 2
Peru (1000 mt) 205 221 242 276 283 289 284 275 272 275 283 294
ROW and (1000 mt) 9393 7447 7394 7391 7290 7369 7484 7733 8009 8089 8178 8352
Residual
Total Imports (1000 mt) 43093 41932 41895 42563 43165 43839 44614 45246 46027 46851 47568 48402
continued
449
Table 1. Continued.
450
Country Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
(mt)
Prices
International Rice (US$/mt) 420 369 458 472 496 518 522 530 534 539 541 545
Reference Price
U.S. FOBb (US$/mt) 518 508 540 552 559 575 581 583 585 590 603 610
Gulf Ports
U.S. No. 2 (US$/mt) 877 799 828 831 832 834 835 837 839 840 842 845
Medium FOB CA
a
ECOWAS 7 = Economic Community of West African States.
b
FOB = Free On Board.
Table 2. World rice supply and utilization, and macro data, 2014-2025.
Variable Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
a
Area harvested (1000 ha ) 160,011 158,935 159,387 160,394 160,681 161,031 161,446 161,505 161,748 161,925 161,819 161,641
Yield (mt/ha) 2.99 2.96 3.02 3.05 3.07 3.10 3.13 3.15 3.18 3.20 3.23 3.25
Production (1000 mt) 478,261 470,129 481,422 489,095 493,853 499,173 505,508 509,360 514,436 518,848 522,549 525,841
Beginning stocks (1000 mt) 107,487 103,866 89,351 83,755 81,167 79,257 79,281 81,488 83,611 86,413 89,108 91,097
Domestic supply (1000 mt) 585,748 573,995 570,773 572,850 575,020 578,430 584,789 590,848 598,048 605,261 611,658 616,938
Consumption (1000 mt) 480,121 484,374 486,893 491,672 495,844 499,298 503,493 507,456 511,868 516,394 520,808 524,864
Ending stocks (1000 mt) 103,866 89,351 83,755 81,167 79,257 79,281 81,488 83,611 86,413 89,108 91,097 92,325
Domestic use (1000 mt) 583,987 573,725 570,648 572,839 575,101 578,579 584,981 591,067 598,281 605,503 611,905 617,189
Total trade (1000 mt) 43,093 41,932 41,895 42,563 43,165 43,839 44,614 45,246 46,027 46,851 47,568 48,402
Per capita use (kilogram) 65.6 65.4 65.0 64.9 64.8 64.5 64.4 64.2 64.2 64.1 64.0 64.0
Percent stocks-to-use (%) 21.6 18.4 17.2 16.5 16.0 15.9 16.2 16.5 16.9 17.3 17.5 17.6
Population growth (%) 1.2 1.2 1.2 1.1 1.1 1.1 1.0 1.0 1.0 1.0 0.9 0.9
Real GDP growth (%) 2.7 2.6 2.9 3.2 3.3 3.3 3.4 3.3 3.4 3.4 3.3 3.2
a
ha = hectares; mt = metric tons.
Table 3. Detailed U.S. rice supply and utilization, and macro data, 2014-2025.
Variable Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
Yield (rough basis) (lb/acre) 7576.5 7469.5 7701.3 7798.4 7893.1 7984.6 8070.2 8146.6 8223.5 8301.7 8377.7 8444.5
Total harvested (1000 acre) 2933.0 2575.0 2662.5 2718.9 2742.6 2747.8 2747.8 2736.8 2722.4 2704.3 2716.9 2819.2
area
Supply (rough basis) (mil. cwta) 278.7 265.3 269.2 271.4 274.5 275.9 275.5 274.5 273.8 273.0 274.7 286.2
Production (mil. cwt) 222.2 192.3 205.0 212.0 216.5 219.4 221.8 223.0 223.9 224.5 227.6 238.1
Beginning stocks (mil. cwt) 31.8 48.5 39.9 35.3 33.8 31.6 28.1 25.7 24.1 22.6 21.0 22.1
Imports (mil. cwt) 24.7 24.4 24.2 24.1 24.3 24.9 25.7 25.9 25.8 26.0 26.1 26.1
Domestic use (mil. cwt) 129.9 122.4 126.4 127.8 130.3 132.7 133.5 133.5 134.0 134.4 134.6 135.0
(rough basis)
Food (mil. cwt) 105.3 106.3 107.3 108.5 109.8 110.9 112.1 113.8 115.8 117.8 119.8 121.7
Seed (mil. cwt) 3.2 3.2 3.3 3.4 3.4 3.4 3.4 3.4 3.4 3.3 3.4 3.5
Brewing (mil. cwt) 18.8 18.9 18.9 19.1 19.2 19.3 19.4 19.5 19.6 19.6 19.4 19.5
Residual (mil. cwt) 2.6 -6.0 -3.2 -3.2 -2.1 -1.0 -1.4 -3.3 -4.8 -6.3 -8.0 -9.7
Exports (mil. cwt) 100.3 103.1 107.5 109.9 112.6 115.1 116.4 117.0 117.3 117.6 118.1 119.2
Total use (mil. cwt) 230.2 225.4 233.8 237.7 242.9 247.8 249.9 250.4 251.2 252.0 252.7 254.1
Ending stocks (mil. cwt) 48.5 39.9 35.3 33.8 31.6 28.1 25.7 24.1 22.6 21.0 22.1 32.1
Prices
Loan rate (US$/cwt) 6.50 6.50 6.50 6.50 6.50 6.50 6.50 6.50 6.50 6.50 6.50 6.50
Season ave. (US$/cwt) 13.30 13.20 13.69 13.84 13.93 14.15 14.22 14.09 14.16 15.23 17.17 17.77
farm price
Long-grain (US$/cwt) 11.90 11.53 12.60 12.86 13.12 13.40 13.51 13.38 13.48 14.45 16.26 16.85
farm price
Medium-grain (US$/cwt) 18.20 17.10 17.20 17.24 16.97 17.13 17.16 17.03 17.08 18.42 20.79 21.50
farm price
(Average)
Japonica (US$/cwt) 21.40 20.46 20.16 20.18 19.76 19.85 19.83 19.65 19.64 21.02 23.46 24.19
farm price
Southern (US$/cwt) 14.40 12.05 12.61 12.71 12.41 12.53 12.55 12.41 12.52 13.62 15.57 16.15
medium-grain
farm price
continued
451
Table 3. Continued.
452
Variable Units 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
Prices, continued
Reference prices:
Long-grain (US$/cwt) 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00
farm price
Southern (US$/cwt) 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00 14.00
medium-grain
farm price
Japonica (US$/cwt) 16.10 16.10 16.10 16.10 16.10 16.10 16.10 16.10 16.10 16.10 16.10 16.10
Export price, (US$/cwt) 23.50 23.03 24.49 25.04 25.36 26.06 26.38 26.44 26.54 26.76 27.34 27.67
b
FOB Houston
(U.S. No. 2)
Medium-grain (US$/cwt) 39.78 36.25 37.56 37.68 37.72 37.82 37.87 37.98 38.06 38.11 38.17 38.31
price, FOB CA
(U.S. No. 2)
Program payment (US$/cwt) 1.52 2.23 1.21 0.96 0.76 0.54 0.46 0.55 0.47 0.02 0.00 0.00
Average (US$/cwt) 10.61 9.40 10.10 10.45 10.83 11.26 11.46 11.66 11.80 11.98 12.18 12.27
world price
(US$/cwt)
Income factors:
Production market (mil. US$) 3021 2547 2858 2992 3075 3166 3216 3204 3235 3483 3974 4300
value
Program payment (mil. US$) 338 428 247 203 165 119 101 123 105 6 0 0
Total income (mil. US$) 3359 2976 3106 3195 3240 3285 3317 3327 3340 3488 3974 4300
Market returns (US$/acre) 488 473 574 596 598 611 494 479 481 567 728 777
above variable cost
Total returns (US$/acre) 604 639 667 670 658 654 494 479 481 567 728 777
above variable cost
Per capita use (lb) 28.7 13.25 13.35 13.54 13.56 13.61 13.56 13.51 13.44 13.43 13.38 13.31
Stocks-to-use ratio (%) 0.21 17.15 19.51 17.84 16.47 15.68 15.37 14.88 14.63 14.24 14.12 14.57
Population growth (%) 0.74 0.75 0.82 0.81 0.80 0.80 0.79 0.78 0.77 0.75 0.74 0.73
Real GDP growth (%) 2.43 2.46 2.74 2.96 2.69 2.55 2.53 2.18 2.14 2.17 2.20 2.12
a
mil. cwt = million hundred weight.
b
FOB = Free On Board.
Fig. 1. Stochastic projections of long-grain rice

international reference price (U.S. dollars; $/metric ton; mt), 2015-2025.
Fig. 2. Stochastic projections of U.S. long-grain rice free

on board export price (U.S. dollars; $/metric ton; mt), 2015-2025.
453
Fig. 3. Stochastic projections of medium-grain rice mill price,

free on board California (U.S. dollars; $/metric ton; mt), 2015 -2025.
Fig. 4. Stochastic projections of world rice

total trade (million metric tons; mil. mt), 2015-2025.
454
Fig. 5. Stochastic projections of U.S. rice total

exports [million (mil.) hundredweight (cwt)], 2015-2025.
455
ECONOMICS
The Economics of Methane

Emissions in Arkansas Rice Production
F. Tisboe, L. Nalley, K. Brye, B. Dixon, and A. Shew
ABSTRACT
This study sets out to estimate average county/parish and state level methane (CH4)
emissions per hectare as well as CH4 efficiency levels (kg rice/kg CH4 ha-1) in rice
production for Arkansas, Louisiana, and Mississippi in order to highlight spatial dif-
ferences in CH4 emission in rice production. Our findings suggest that on average
Mississippi was the most efficient at converting CH4 into rice (267.46 kg rice/kg CH4
ha-1), followed by Arkansas (189.92) and Louisiana (178.80). Specifically, Louisiana
was negatively impacted by its large ratoon crop in terms of CH4 use efficiency, with
38% of its primary rice crop being ratooned. Moreover, the Mississippi results should
be interpreted with caution because seeding data, specifically the area seeded to hybrids
and county level yields, are not as robust as those for Arkansas and Louisiana. Overall,
these results provide rice buyers, producers, and consumers with important information
about the spatial aspects of sustainability in rice production.
INTRODUCTION
Increased consumer demand for food products with lower greenhouse gas (GHG)
emissions have prompted row crop producers to reduce GHG emissions associated with
crop production. More importantly, agricultural producers face increasing demand and,
in some cases, requirements from the private industry to reduce GHG emissions associ-
ated with crop production. To demonstrate, Wal-Mart recently announced a potential
plan to label each of its products with a sustainability rating and subsequently requested
that every Wal-Mart supplier provide its products GHG footprint, a direct measure of
climate impact (Wal-Mart Corporate-Sustainability Index, 2011). In response to these
commercial pressures, agricultural producers and processors have sought to increase
production efficiency with respect to GHG emissions. Particularly, rice production (from
seed to farm gate) has been identified as a significant source of atmospheric methane
(CH4) emissions from U.S. agricultural production (U.S. Environmental Protection
Agency, 2011). As a result, producers and large-scale purchasers of U.S. rice have at-
tempted to increase the efficiency of GHG emissions in rice production. As such, this
study estimates county/parish and state averages of CH4 use efficiency (kg rice/kg CH4
456
ha-1) in Arkansas, Louisiana, and Mississippi in 2014 in order to provide consumers

and large-scale buyers of rice more information about the spatial components related
to sustainability in rice production.
PROCEDURES
Methane emissions from rice production were estimated in each of the 96 rice-
growing counties/parishes in Arkansas, Louisiana, and Mississippi using a two-step
approach, consistent with the Intergovernmental Panel on Climate Change (IPCC, 2006).
In the first step, data from field research (Rogers et al., 2013; Brye et al., 2013) were
used in a regression model to estimate representative CH4 emission factors based on
rice cultivar type (hybrid and conventional pure-line), soil texture (loamy and clayey),
and crop rotation (rice-rice and soybean-rice). The field research was conducted in
2012 and 2013 at the University of Arkansas System Division of Agricultures Rice
Research and Extension Center (RREC) near Stuttgart, Ark., on a DeWitt silt loam
and at the Northeast Research and Extension Center (NEREC) at Keiser, Ark., on a
Sharkey clay soil. At both locations, the study areas had previously been managed in a
rice-soybean rotation for at least 15 years. Subsequently, in 2012 and 2013 at RREC,
four replications of the conventional pure-line cultivars Taggart and Cheniere and the
hybrid cultivar CLXL745 were sown following the previous crops of rice or soybean.
In 2012, at NEREC, four replications of Taggart were sown following soybean as the
previous crop; while in 2013, four replications of Taggart, Cheniere, and CLXL745
were sown following the previous crops of soybean or rice. Each year at each location,
plots were outfitted with a 30-cm diameter, enclosed-headspace gas sampling chamber
assemblage to measure CH4 emissions.
Unlike previous studies, this experimental design allowed for a direct compari-
son of the effects of soil texture, cultivar type, and crop rotation on CH4 emissions.
Research efforts have identified a diel CH4 emission pattern with soil texture, air and
water temperature, soil organic carbon, and cultivar as contributing factors to overall
CH4 emissions. Furthermore, previous studies have quantified the proportion of CH4
emissions by independently altering each factor mentioned above while holding the
other two constant, leading to potentially biased estimates. The goal of this study was
to holistically identify those factors contributing to CH4 emissions as a result of three
sources of variationsoil type, cultivar, and preceding crop.
The first step in this estimation process is to use a regression model specified for
the parameter estimates given by:
Yipst = 0 + ipsXipst + uispt Eq. (1)
where Yipst is the CH4 emissions (kg CH4-C ha ) for cultivar i, under crop rotation p,
-1
grown on soil with texture s, in year t. The variable Xipst is a categorical variable formed
by the combination of soil texture (two types), cultivar type (two types: due to lack of
degrees of freedom, both conventional cultivars were analyzed as conventional and
not specific cultivar lines), and crop rotation. The eight categories formed are: (1)
457
conventional cultivar grown on clay soil following a rice-rice rotation, (2) conven-
tional cultivar grown on clay soil following a soybean-rice rotation, (3) conventional
cultivar grown on loamy soil following a rice-rice rotation, (4) conventional cultivar
grown on loamy soil following a soybean-rice rotation, (5) hybrid cultivar grown on
clay soil following a rice-rice rotation, (6) hybrid cultivar grown on clay soil following
a soybean-rice rotation, (7) hybrid cultivar grown on loamy soil following a rice-rice
rotation, and (8) hybrid cultivar grown on loamy soil following a soybean-rice rotation.
The remaining influences on CH4 emissions are captured by the error term and are as-
sumed to be independent of the soil, the cultivar, and the rotation effects. By omitting
one of the categories in Xipst in Eq. (1), the parameter 0 serves as the emission factor
for the omitted category. Further, the seven ips parameters are the marginal differences
between the omitted category and the category for which ips is estimated. The omit-
ted category in this case is category 6hybrid cultivar grown on clay soil following a
soybean-rice rotation. Thus, the emission factor for a category not omitted is given by
the summation of 0 and the appropriate component of ips.
In the second step of the estimation process, the estimates derived from Eq. (1)
are applied to all 96 rice-growing counties/parishes in Arkansas, Louisiana, and Mis-
sissippi to simulate their respective annual CH4 emissions for the period 2002-2014,
which were based on a 1000 simulations based on categorical variables in Eq. (1).
This simulation is done by first synthesizing each countys (1) soil texture maps into
percentages of loamy versus clayey soils, (2) historical seeding area between hybrid
and conventional rice cultivars, and (3) percentage of rice area under different crop rota-
tions. Ideally, a CH4 measurement from each combination (crop rotation/cultivar type/
soil texture) would be obtained from field plots in each county/parish. However, given
the high cost of measuring CH4 emissions, extrapolation is currently the most feasible
alternative. Notably, data on the historical seeding area between hybrid and conventional
rice cultivars at the county level were collected from various annual publications of the
Proceedings of the Rice Technical Working Group. Soil texture data were collected from
the Web Soil Survey (WSS) provided by the USDA Natural Resources Conservation
Service (NRCS) (USDA-NRCS, 2015), and data on crop rotation were sourced from
over twenty extension agents throughout the growing region since observed data on
county/parish crop rotations are nonexistent.
The simulations recognized two sources of uncertainty and were modeled by
random draws according to the hypothesized distributions. The first source is a result
of uncertainty about the regression parameters from Eq. (1) since they are based on
sample data. In each of the 1000 simulations, a vector of emission factors was drawn,
assuming normality and using the covariance matrix of the estimated emission factors
from Eq. (1). The second source of uncertainty is the randomness due to the additive
error term of the regression model. This randomness is dealt with in two steps. First,
for a given combination of soil texture, cultivar type, and crop rotation, the standard
deviations (SDuipst) for the error terms (uipst) in Eq. (1) were estimated, accounting for
heteroscedasticity. A distribution of the error term (ips) was estimated for each of the
eight categories (differing by the variance). Then, 1000 draws from a standard, normal
458
distribution were generated by a random number generator. This draw was multiplied
by each of the eight SDuipst and added to the appropriate simulated sample mean to
give the simulated CH4 emissions (kg CH4-C ha1) for each unique combination of ips.
This procedure recognizes that the error term in Eq. (1) is heteroscedastic across the
eight categories in Xipst. Using these simulated emissions, the CH4 emitted from each
county was computed using the observed proportions of cultivars, soil types, rotations
and acres (hectares) seeded to rice observed for each county in 2014. This computa-
tion generates an extrapolation to all 96 rice-growing counties/parishes in Arkansas,
Louisiana, and Mississippi in order to simulate their respective annual CH4 emissions.
The total simulated CH4 emissions (CH

1 ) for the primary rice crop in the jh county/

parish in 2014 is represented as:
1 = [ ( + 0 + )]
CH Eq. (2)
where Hajips is the total area of rice seeded to cultivar i, under crop rotation p, and grown
on a soil with texture s, and and 0 are estimated from Eq. (1). In the summation
(2), the referring to a hybrid cultivar grown on clay soil following a soybean-rice
rotation is zero.
Previous literature suggests that ratoon rice crops generate CH4 at a considerably
higher rate than primary crops. This happens because the amount of organic carbon
available for anaerobic decomposition (from the crop residue of the primary crop) is
considerably higher during the ratoon crop production relative to the primary crop
(Wang et al., 2013). Currently, the climatic conditions of Texas, Louisiana, Arkansas,
and Florida typically allow for ratoon crop production. Previous studies in these areas
estimate the seasonal emission factor for primary and ratoon crops to be 237 kg CH4-C
ha-1 and 780 kg CH4-C ha-1, respectively (USEPA, 2015). Thus, CH4 emissions from
ratoon crops are about 3.29 (780 kg /237 kg) times that of the primary crop. The total
emissions from ratoon crop (CH 2) in the jh county/parish is therefore calculated as:
2 = 3.29 (CH
CH 1 1 )2 Eq. (3)
where 1 and 2 are the total area seeded to primary and ratoon rice in the jh county/
parish. Thus, the total emissions of the jh county/parish are:
= CH
CH
1
+ CH 2
Eq. (4)
where total emissions for the j county/parish ( CH
h ) is the sum the emissions from the
primary crop (CH
1
) and the ratoon crop ( CH 2 ). Finally, using the 2014 NASS-reported

(USDA-NASS, 2015a) rice production quantities (Qjt) in kilograms for each of the 96
counties/parishes, we also calculated the amount (kg) of rice produced per CH4 emis-
sions for the jh county/parish in year 2014 as:
QCH = Q CH
Eq. (5)
From these calculations we identify: 1) those counties/parishes with the largest net CH4
emissions, 2) the counties/parishes with the best efficiency in terms of kg rice/kg CH4,
459
and 3) state averages of kg of rice/kg CH4-C ha-1 to provide state aggregates. Given
the fact that rice mills often source from various counties/parishes from one state, state
averages eliminate the need for a potential buyer of rice to determine what county/parish
it originated in, but rather which state.

The regression estimates from Eq. (1) are displayed in Table 1. The R2 value
indicates that 91% of the variation in CH4 emissions is explained by the change across
the eight categories formed by the combination of soil texture, cultivar type, and crop
rotation. The results from this regression indicate that the driving factor for CH4 emis-
sions appears to be soil texture with loamy soils emitting 987% more than clayey soils.
The next largest factor in CH4 emissions is cultivar type with the conventional emitting
40% more CH4 than hybrids, and crop rotation with rice-rice rotations emitting 39%
more CH4 than rice-soybean. Thus, hybrids grown on clayey soils after soybeans had
the lowest season-long CH4 emissions (5.82 kg CH4-ha-1); while conventionals grown
on loamy soil after rice had the largest CH4 emissions (181.95 kg CH4-ha-1).
County-level rice production per CH4 emissions for Arkansas, Louisiana, and
Mississippi in 2014 are presented in Tables 2, 3, and 4. From an average efficiency
standpoint (kg rice/kg CH4- ha-1), Mississippi was the most efficient (267.46), followed
by Arkansas (189.92), and Louisiana (178.80). Accordingly, Louisiana was negatively
affected in terms of the efficiency of CH4 use by its large ratoon crop (38% ratooned area
as the percent of primary growth area, USEPA, 2015). To ensure unbiased comparisons,
we also estimated a weighted average by rice production (volume by location). That
is, several low-yielding counties/parishes could be outliers and thus skew the means.
Tables 2, 3, and 4 show that Arkansas again had the second-highest CH4 use efficiency
(200.06 kg rice/kg CH4 ha-1), behind Mississippi and ahead of Louisiana.
The results from Mississippi in this study should be observed with caution given
the poor data availability in 2014 (Table 4). In 2014, USDA-NASS (2015b) reports that
50% of counties in Mississippi have the same rice yield. This observation is a result
of the low number of rice producers in a county, and the fact that NASS must keep
the data of individual producers anonymous. Furthermore, Mississippi did not report
hybrid adoption at the county level, but only reported it at the state level for 2014.
While the county-level results are listed in the aforementioned tables, the state-level
results may be more robust because rice mills may not know the county of origin for
the rice they are selling in bulk, but may be surer of the state of origin. Thus, the ag-
gregate results of state comparisons is necessary to mimic the decision-making of rice
buyers and millers. Even more, on the average unweighted results, Arkansas was 25%
more efficient in terms of kg rice/kg CH4-ha-1 than Louisiana, and 108% less efficient
than Mississippi (Tables 2, 3, and 4). Notably, the model used for the extrapolation is
based on data solely from Arkansas locations. Similar experiments in Mississippi and
Louisiana would likely generate somewhat different parameter estimates and therefore
different CH4 estimates.
460
Understanding the CH4 use efficiency of rice across states is important for a number
of reasons: (1) buyers of rice may start paying premiums for rice from those states that
are more efficient in terms of how much rice is produced per unit of CH4 released, in
order to meet the growing consumer demand for environmentally friendly foods; (2) if
a carbon policy, i.e., tax/offset program, is instituted, the data and the results generated
from this study will provide producers and policy-makers at the state level with the
information they need to respond accordingly; (3) this study illustrates the largely nega-
tive effect of ratooning rice on the overall efficiency of CH4 production per kilogram of
rice produced, specifically in large ratooning states like Louisiana; and (4) these results
can provide a spatial component that gives rice producers information on where they
stand in the increasingly demanding world of sustainability.
ACKNOWLEDGMENTS
The authors would like to acknowledge the Arkansas Rice Research Promotion
Board and the University of Arkansas System Division of Agriculture for funding this study.
LITERATURE CITED
Brye, K., C. Rogers, A. Smartt, and R. Norman. 2013. Soil texture effects on methane
emissions from direct-seeded, delayed-flood rice production in Arkansas. Soil Sci.
178:519-529.
IPCC. Intergovernmental Panel on Climate Change. 2006. 2006 IPCC guidelines for
national greenhouse gas inventories. The National Greenhouse Gas Inventories
Programme.
Rogers, C.W., K.R. Brye, R.J. Norman, A.D. Smartt, A.M. Fulford, and D. Frizzell.
2013. Seasonal pattern of methane fluxes from a silt-loam soil as affected by pre-
vious crop and cultivar. In: B.R. Wells Rice Research Studies, 2012. R.J. Norman
and K.A.K. Moldenhauer (eds.). University of Arkansas Agricultural Experiment
Station Research Series 609:272-279. Fayetteville.
USDA-NASS. 2015a. United States Department of Agriculture National Agricultural
Statistics Service. Access date: 9 February 2016. Available at: http://www.nass.
usda.gov/Data_and_Statistics/
USDA-NASS. 2015b. United States Department of Agriculture, National Agricultur-
al Statistics Service (NASS). Mississippi County Estimates. Access date: 9 Febru-
ary 2016. Available at: https://www.nass.usda.gov/Statistics_by_State/Mississippi/
USDA-NRCS. 2015. United States Department of Agriculture Natural Resources
Conservation Service. Web Soil Survey. Access date: 9 February 2016. Available
at: http://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm
USEPA. 2011. United States Environmental Protection Agency. EPA Preliminary
Analysis of the Waxman Markey Discussion Draft: The Clean Energy and Se-
curity Act of 2009 in the 111th Congress. Washington, DC: U.S. Environmental
Protection Agency.
461
USEPA. 2015. United States Environmental Protection Agency. Inventory of U.S.

greenhouse gas emissions and sinks 1990-2013. Access date: 9 February 2016.
Available at: http://www3.epa.gov/climatechange/ghgemissions/usinventoryreport
Wal-Mart Corporate-Sustainability Index. 2011. Access date 5 January 2016. Avail-
able at: www.walmartstores.com/Sustainability/9292.aspx
Wang, J.J., S.K. Dodla, S. Viator, M. Kongchum, S. Harrison, S.D. Mud, and Z. Tian.
2013. Agriculture field management practices and greenhouse gas emissions from
Louisiana soils. Louisiana Agriculture, 56(2):8-9. Access date: 9 February 2016.
Available at http://www.lsuagcenter.com/NR/rdonlyres/78D8B61A-96A8-49E1-
B2EF-BA1D4CE4E698/93016/v56no2Spring2013.pdf
Table 1. Regression estimates.

Rice type/soil texture/ Estimated
crop rotation coefficienta
(number of observations)
Conventional clay rice 16.32***
(n = 8) (-2.53)
Conventional clay soybean 11.07**
(n = 12) (-4.44)
Conventional loamy rice 176.13***
(n = 16) (-7.36)
Conventional loamy soybean 147.92***
(n = 16) (-8.75)
Hybrid clay rice 8.77***
(n = 4) (-3.12)
Hybrid loamy rice 151.30***
(n = 8) (-7.35)
Hybrid loamy soybean 66.90***
(n = 8) (-8.58)
Constant hybrid clay soybean 5.82***
(n = 4) (-1.07)
R2 0.912
a
Significance levels: * = P < 0.10, ** = P < 0.05, *** = P <
0.01. Standard errors of the estimates are in parentheses.
462
Table 2. 2014 Arkansas county level methane (CH4) efficiency (kg rice/ kg CH4-C ha-1) estimatesa.
Soil texture CH4 efficiency
County Clayey Loamy Seeded areab Hybrid adoption Rice yieldb Meanc SDd
(%) (acre) (%) (lb/acre) (kg rice/kg CH4-C ha-1)
Pope 1.23 98.77 2204 (892) 54.10 8731 (9,786) 65.47 0.39
Faulkner 4.40 95.60 2582 (1,045) 51.00 9359 (10,490) 71.05 0.42
Lafayette 24.36 75.64 4433 (1,794) 40.00 9005 (10,093) 78.77 0.62
Pulaski 15.54 84.46 4169 (1,687) 81.92 9396 (10,531) 94.26 0.69
Lawrence 12.70 87.30 99,922 (40,437) 15.80 18,543 (20,784) 128.89 9.05
Craighead 9.82 90.18 71,510 (28,939) 15.19 20,699 (23,200) 139.67 6.01
Independence 4.39 95.61 12,746 (5,158) 26.74 20,955 (23,487) 141.68 1.02
Randolph 5.99 94.01 35,657 (14,430) 36.49 20,369 (22,831) 146.12 2.71
Jackson 13.29 86.71 104,194 (42,166) 24.77 20,613 (23,104) 149.92 7.83
Drew 10.83 89.17 11,312 (4,578) 41.67 20,421 (22,889) 157.06 0.80
White 10.91 89.09 13,193 (5,339) 70.03 19,167 (21,483) 171.19 1.04
Clay 15.74 84.26 81,505 (32,984) 34.79 22,185 (24,866) 173.09 5.19
Monroe 28.75 71.25 59,493 (24,076) 25.97 20,164 (22,601) 173.12 3.82
Ashley 12.32 87.68 1,1182 (4,525) 67.79 19,671 (22,048) 175.88 0.83
Arkansas 6.94 93.06 91,155 (36,889) 56.68 22,140 (24,,816) 177.13 6.11
St. Francis 37.02 62.98 38,445 (15,558) 18.29 19,973 (22,387) 182.45 2.41
Greene 23.66 76.34 78,404 (31,729) 42.62 20,878 (23,401) 183.65 4.70
Prairie 18.17 81.83 63,639 (25,754) 49.39 21,386 (23,970) 183.75 3.91
Poinsett 32.54 67.46 121,568 (49,197) 13.31 22,020 (24,681) 186.42 7.60
Lee 27.71 72.29 29,919 (12,108) 37.62 20,977 (23,512) 188.20 1.74
Woodruff 34.05 65.95 61,925 (25,060) 36.29 20,080 (22,507) 192.70 3.52
Cross 31.98 68.02 88,036 (35,627) 26.41 21,936 (24,587) 195.86 4.99
Phillips 40.04 59.96 32,643 (13,210) 39.51 19,436 (21,,785) 204.35 1.76
Lonoke 18.17 81.83 89,731 (36,313) 69.81 21,804 (24,439) 209.24 5.72
Lincoln 34.15 65.85 21,515 (8,707) 72.49 20,974 (23,509) 245.44 1.19
Jefferson 34.15 65.85 72,464 (29,325) 80.21 20,757 (23,265) 254.73 4.29
Desha 59.06 40.94 25,266 (10,225) 35.41 20,730 (23,235) 283.71 1.05
Mississippi 61.12 38.88 53,540 (21,667) 26.30 22,574 (25,302) 305.50 2.13
Crittenden 74.60 25.40 51,037 (20,654) 53.91 19,842 (22,240) 415.54 1.71
continued
463
Table 2. Continued.
464
County Clayey Loamy Seeded areab Hybrid adoption Rice yieldb Meanc SDd
(%) (acre) (%) (lb/acre) (kg rice/kg CH4-C ha-1)
Chicot 70.77 29.23 34,839 (14,099) 65.04 20,538 (23,020) 422.69 1.13
Average 25.81 74.19 48,941 (198056) 43.65 19,177 (21,495) 189.92 3.15
Weighted avg. by lb - - - - - 200.06 -
Weighted avg. by acre - - - - - 199.12 -
a
77.05% of rice in Arkansas is assumed to be grown under a rice-soybean rotation in 2014.
b
Equivalent metric units (ha and kg/ha) are in parentheses.
c
Ordered from lowest efficiency level to highest.
d
Standard deviations (SD) based on 1000 simulations.
Table 3. 2014 Louisiana parish level methane (CH4) efficiency (kg rice/ kg CH4-C ha-1) estimatesa.
Soil texture Rice-soybean Seeded Ratoon Hybrid Rice CH4 efficiency
Parish Clayey Loamy rotation areab cropb adoption yielda Meanc SDd
------------------------ (%)--------------------- ---------------(acres)------------- (%) (lb/acre) (kg rice/kg CH4-C ha-1)
Lafayette 9.53 0.00 71.86 534 (216) 502 (203) 10.69 8407 (9,423) 28.37 0.21
La Salle 12.52 33.74 71.86 554 (224) 0 0.00 8236 (9,231) 53.17 0.13
Rapides 20.33 35.76 71.86 11,001 (4,452) 0 0.00 7683 (8,612) 53.53 2.45
Ouachita 19.84 33.47 71.86 9175 (3713) 0 0.00 8189 (9,179) 56.77 1.93
Natchitoches 26.49 45.50 71.86 3813 (1,543) 0 21.56 8742 (9,798) 71.02 0.57
Allen 0.00 16.38 71.86 15,437 (6,247) 2454 (993) 35.99 18,157 (20,351) 104.94 1.55
Acadia 9.10 0.00 60.61 85,444 (34,578) 24,832 (10,049) 18.37 20,682 (23,181) 105.71 7.95
Jefferson Davis 14.71 0.71 60.61 83,484 (33,785) 23,001 (9,308) 32.84 19,082 (21,388) 110.04 7.07
Franklin 22.67 0.77 100.00 3005 (1,216) 0 82.66 9016 (10,106) 112.32 0.60
Beauregard 3.89 30.91 71.86 1248 (505) 613 (248) 58.77 21,772 (24,403) 112.64 0.13
Evangeline 6.97 4.78 60.61 45,905 (18,577) 5548 (2,245) 40.41 19,872 (22,274) 124.91 3.42
continued
Table 3. Continued.
Soil texture Rice-soybean Seeded Ratoon Hybrid Rice CH4 efficiency
Parish Clayey Loamy rotation areab cropb adoption yielda Meanc SDd
------------------------ (%)--------------------- ---------------(acres)------------- (%) (lb/acre) (kg rice/kg CH4-C ha-1)
Calcasieu 8.27 2.32 71.86 15,212 (6,156) 1789 (724) 33.62 19,726 (22,110) 125.90 1.26
Vermilion 31.25 0.89 60.61 53,427 (21,621) 12,291 (4,974) 9.79 19,404 (21,749) 126.64 4.28
Catahoula 61.06 9.99 71.86 1300 (526) 0 43.85 8684 (9,734) 127.06 0.12
St. Landry 45.76 2.81 60.61 26,062 (10,547) 4008 (1,622) 10.63 19,219 (21,542) 161.18 1.66
Morehouse 32.61 4.04 71.86 37,617 (15,223) 0 22.33 20,138 (22,572) 175.97 2.27
West Carroll 24.46 0.00 71.86 2155 (872) 0 74.74 18,286 (20,496) 188.50 0.15
East Carroll 86.39 4.25 71.86 1589 (643) 0 36.71 8324 (9,330) 215.59 0.13
Richland 32.96 1.19 100.00 5614 (2,272) 0 48.64 20,645 (23,140) 224.83 0.37
Cameron 71.36 6.87 71.86 11,834 (4,789) 3032 (1,227) 11.05 19,254 (21,581) 233.07 0.70
West Baton Rouge 69.82 1.11 71.86 576 (233) 0 100.00 9818 (11,005) 247.98 0.04
Tensas 88.34 3.77 71.86 3141 (1,271) 0 40.86 9001 (10,089) 255.90 0.24
Iberia 57.94 0.00 71.86 1127 (456) 0 18.00 21,134 (23,688) 257.85 0.05
Avoyelles 64.24 6.21 71.86 11,987 (4,851) 114 (46) 24.19 19,953 (22,364) 277.61 0.52
St. Martin 65.79 0.00 71.86 3,437 (1,391) 771 (312) 62.35 21,998 (24,657) 319.65 0.13
Madison 85.37 0.25 71.86 7319 (2,962) 0 0.00 18,044 (20,225) 369.82 0.40
Concordia 84.72 1.18 71.86 9790 (3,962) 230 (93) 58.94 21,190 (23,751) 586.67 0.28
Average 29.05 11.21 71.86 19,761 (7,997) 3717 (1,504) 32.98 14,921 (16,724) 178.80 1.43
Weighted avg. by lb - - - - - - - 149.34 -
Weighted avg. by acre - - - - - - - 146.14 -
a
A five-year average (2010-2014) was used for the area under ratoon.
b
c
d
Standard deviations (SD) based on 1000 simulations.
465
Table 4. 2014 Mississippi county level methane

(CH4) efficiency (kg rice/ kg CH4-C ha-1) estimatesa.
County Clayey Loamy Seeded area b
Rice yield Meanc SDd
b
--------- (%)--------- (acre) (lb/acre) (kg rice/ kg CH4-C ha-1)

Grenada 1.89 25.19 282 (114) 11,166 (12,515) 83.01 0.05
Tate 3.56 10.01 934 (378) 11,166 (12,515) 84.23 0.16
Panola 6.07 14.24 5523 (2,235) 11,166 (12,515) 86.13 0.95
Holmes 13.37 12.85 121 (49) 11,166 (12,515) 92.17 0.02
Desoto 26.66 4.34 1191 (482) 11,166 (12,515) 105.68 0.16
Tallahatchie 35.53 6.14 6963 (2,818) 20,544 (23,027) 215.50 0.47
Humphreys 76.22 4.72 1475 (597) 11,166 (12,515) 232.82 0.12
Sharkey 87.55 7.39 432 (175) 11,166 (12,,515) 321.24 0.03
Sunflower 65.33 8.60 13,635 (5,518) 20,096 (22,525) 331.43 0.65
Leflore 67.21 2.05 3904 (1,580) 19,511 (21,869) 333.88 0.19
Issaquena 92.19 4.75 1114 (451) 11,166 (12,515) 380.32 0.09
Coahoma 70.93 19.26 8110 (3,282) 21,087 (23,635) 389.67 0.35
Quitman 74.05 11.20 8765 (3,547) 19,917 (22,324) 394.58 0.39
Bolivar 78.90 6.13 33,735 (13,652) 20,876 (23,399) 465.64 1.41
Tunica 80.45 8.00 24,602 (9,956) 21,326 (23,903) 495.61 1.01
Washington 81.96 14.10 11,480 (4,646) 21,148 (23,704) 512.41 0.47
Average 51.99 9.66 7386 (2,988) 15,512 (17,386) 267.46 0.40
Weighted avg. - - - - 416.18 -
by kg
Weighted avg. - - - - 403.60 -
by ha
a
The Rice Technical Working Group Mississippi seeding report denoted only state aggregate
level of hybrid adoption of 38% and not county level. Thus, this study assumes each county
had 38% adoption. All rice in Mississippi is assumed to be under a soybean-rice rotation.
b
c
d
Standard deviations (SD) based on 1000 simulations. NASS uses yield estimates of 11,166
lb/acre for several counties given the low number of rice producers in each county, to prevent
a specific rice producer being identified. The average for the crop reporting district is used in
these cases.
466
Fig. 1. Methane emission efficiency in rice production by county, 2014.
467
B.R. Wells Arkansas Rice Research Studies 2015 Norman and Moldenhauer AAES

Arkansas Rice Research Study 2015

Uploaded by

Copyright:

Available Formats

Arkansas Rice Research Study 2015

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Arkansas Rice Research Study 2015

Uploaded by

Copyright:

Available Formats

25th Anniversary Edition

R.J. Norman and K.A.K. Moldenhauer, editors

ARKANSAS AGRICULTURAL EXPERIMENT STATION

Layout and editing by Marci A. Milus

R.J. Norman and K.A.K. Moldenhauer, editors

University of Arkansas System

Bobby R. Wells was born July 30, 1934, at Wickliffe,

Research reports contained in this publication may represent preliminary or only

The Arkansas Rice Research and Promotion Board

OVERVIEW AND VERIFICATION

Trends in Arkansas Rice Production, 2015

2015 Rice Research Verification Program

BREEDING, GENETICS, AND PHYSIOLOGY

Evaluation of Advanced Semi-Dwarf Medium-Grain and Long-Grain

Kompetitive Allele-Specific Polymerase Chain Reaction (KASPTM)

Titan, a Very Early-Maturing, High-Yielding, and High-Quality Conventional

Testing Uniform Regional Rice Nursery Varieties for Resistance to

Diamond, a High Yielding, Very Short Season, Long-Grain Rice Variety

Development of Superior Medium-Grain and Long-Grain Rice Varieties

Development of Male Sterile Line for Hybrid Rice Production

Development of Aromatic Rice Varieties

PEST MANAGEMENT: DISEASES

Rice Breeding and Pathology Technical Support

Studies on Rice Bacterial Panicle Blight Diseases Relative

Monitoring Bacterial Panicle Blight Disease of Rice and Germplasm

Understanding Autumn Decline, Evaluating Rice Varieties for

Evaluation of Rice Kernel Damage and Yields Due to Rice

Potential Exposure of Honey Bees to Neonicotinoid Insecticides in Rice

Value of Insecticide Seed Treatments in Arkansas Rice

PEST MANAGEMENT: WEEDS

Residual Weed Control and Crop Response to Pethoxamid Systems in Rice

Evaluation of Very Long-Chain Fatty Acid-Inhibiting

Sharpen Tank-Mixtures with Rice Herbicides for Barnyardgrass

Obey and Command Herbicide Programs

Optimizing Quizalofop Rate Structure for Sequential

Evaluation of ProvisiaTM Herbicide Tank Mixtures

Influence of Herbicide Rate, Application Volume, and Adjuvant

Evaluation of Topramezone (Armezon/Impact) for Rice Weed Control

Rice Flatsedge Resistance to Acetolactate Synthase-Inhibiting Herbicides

Evaluation of a Benzobicyclon Plus Halosulfuron Premix

Continued Validation of the Nitrogen Soil Test for Rice

Evaluation of Alternative Nitrogen Fertilizer Application

Effect of Delayed Nitrogen Fertilization into the Floodwater

Effects of Three Different Alternate Wetting and

Grain Yield Response of Four New Rice Cultivars to Seeding Rate

Arkansas Rice Performance Trials, 2013-2015

Methane Emissions from Direct-Seeded, Delayed-Flood Rice

Utilization of On-Farm Testing to Evaluate Rice Cultivars, 2015

Grain Yield Response of Six New Rice Cultivars to Nitrogen Fertilization

Response of Two Rice Cultivars to Midseason Nitrogen Fertilizer

Rice Cultivar Yield Response to Delayed Preflood Nitrogen

Summary of Nitrogen Soil Test for Rice (N-STaR)

RICE QUALITY AND PROCESSING

Impact of Storage Duration, Temperature, and Moisture Content

Simulation and Validation of On-Farm In-Bin Drying

Optimization of Process Parameters in Rough Rice Drying

Impacts of Rough Rice Temperature and Moisture Content

Effects of Temperature, Moisture Content, and Rough Rice Storage

Variable Impact of Nighttime Air Temperatures on Rice Chalk

Characterization of Broken Rice Kernels Caused by