FLÁVIO ALVES DAMASCENO 

COMPOST BEDDED PACK BARNS SYSTEM AND

COMPUTATIONAL SIMULATION OF AIRFLOW THROUGH

NATURALLY VENTILATED REDUCED MODEL

Thesis presented to the Universidade

Federal de Viçosa, as part of the
requirements of the Agricultural
Engineering’s Graduate Program, for
the attainment of the title Doctor
Scientiae.
 

VIÇOSA
MINAS GERAIS – BRAZIL
2012
 BIOGRAPHY

Flávio Alves Damasceno, son of Abadia Alves Damasceno and José Lincoln de Oliveira
Damasceno, was born on August 13, 1980, in Luziânia, Goiás, Brazil. He graduated in
Agricultural Engineering at the Universidade Federal de Lavras, Minas Gerais, Brazil, in
November, 2006. In February 2007, he began his Master’s Degree in Agricultural
Engineering at the Universidade Federal de Lavras, obtaining the title of Master in July
2008. In August 2008, he began his Doctor’s Degree in Agricultural Engineering at the
Universidade Federal de Viçosa, Minas Gerais, Brazil. In August 2010, he began his
research as visiting researcher at the University of Kentucky, Lexington, USA. He obtained
the title of Doctor in August 2012.

ii 
 
 ACKNOWLEDGMENTS

I am ever grateful to God, who has blessed me with this opportunity.

My parents and family who have always supported all of my decisions.
Professor Dr. Fernando da Costa Baêta for his guidance during the Doctors program.
Professors Dr. Joseph L. Taraba, Dr. Rolf Jentzsch, Dr. Ilda de F. Ferreira Tinôco, and
Dr. Cecília de Fátima Souza for their wisdom and guidance as my orienting committee.
Professor Dr. Joseph L. Taraba, who contributed with innumerous instructive comments,
which allowed me to accomplish this high quality work.
Professor Dr. George Day for his excellent advices and thoughts, which I will carry with
me everywhere I go throughout my academic career and life.
Professor Dr. Tadayuki Yanagi Junior was the first person to motivate me to pursue the
academic career and I am deeply grateful for all his motivation, help and friendship.
My appreciation goes to Professors Dr. Scott Shearer and Dr. Doug Overhults, who
contributed with assistance and support, as well as the entire Biosystems and Agricultural
Engineering community, which welcomed me to University of Kentucky.
I would like to express my deep appreciation and gratitude to the following people for
helping me complete this work, specially Dr. Cecilia Souza, Dr. Jeffrey Bewley, Ms. Randi
Black, Dr. Celia Caruso, Mr. Ígor Lopes, Dr. Jairo Alexander, and Dr. Keller Sullivan.
My friends at the Ambiagro, especially Adriana, Akemi, Alexandre, Cinara, Conceição,
Déborah, Diogo, Elohanna, Fábio, Fatinha, Fernanda, Humberto, Irene, Jofran, Keles,
Luciano, Luis Gustavo, Maria Clara, Marilú, Maurílio Duarte, Múcio, Neiton, Olga,
Roque, Rui, Samuel, and Seu Pedro.
My friends at University of Kentucky, especially Carla, Chance, Christina, Enrique,
Ester, Gabriela, Guilherme, Jeniffer, Joe Luck, Lucas Melo, Lucas Revolti, Maíra, Maria,
Marina, Paulo, Pedro, Rafael, Rodrigo, Sam Mullins, Santosh, and Tatiana, who helped me
in Lexington.
My friends at the Viçosa, especially Anderson, Carlos, Evan, and Wagner.
My friends at the Brasilia, especially Marcelo, Bruno, and Marcos.
Leidimar Brandão, who shared wonderful moments with me, which will eternally
remain in my mind and heart.
This work was financed by the Brazilian Organizations (CNPq, FAPEMIG, and
CAPES) and University of Kentucky, whose support was greatly appreciated.
iii 
 
 TABLE OF CONTENTS

ABSTRACT................................................................................................................... viii
RESUMO....................................................................................................................... x
1 - INTRODUCTION................................................................................................... 1
1.1 Summary and justification............................................................................. 1
1.2 Dissertation objectives................................................................................... 3
2 - LITERATURE REVIEW............................................................................ 4
2.1. Compost dairy barn....................................................................................... 4
2.1.1. Facility design............................................................................................ 5
2.1.2. Waterers..................................................................................................... 6
2.1.3. Lighting...................................................................................................... 6
2.1.4. CBP barn ventilation and location............................................................. 7
2.1.4.1. Natural ventilation in CBP barns............................................................ 8
2.1.4.2. Mechanical ventilation in CBP barns...................................................... 10
2.1.5. Economic considerations........................................................................... 13
2.2. Compost bedding pack.................................................................................. 13
2.2.1. Bedding source........................................................................................... 13
2.2.2. Management of the bedded pack................................................................ 14
2.2.3. Composting process in bedded pack.......................................................... 15
2.2.3.1. Microbiological processes in composting bedded pack.......................... 18
2.2.4. Composting and moisture.......................................................................... 22
2.2.5. Manure and effluent management.............................................................. 23
2.3. Physical properties of the bedding material.................................................. 25
2.3.1. Porosity...................................................................................................... 25
2.3.2. Bulk density............................................................................................... 25
2.3.3. Particle density........................................................................................... 27
2.3.4. Moisture content......................................................................................... 27
2.3.5. Water Holding Capacity............................................................................. 28
2.3.6. Particle size................................................................................................ 29
2.4. Thermal property of the bedding material.................................................... 30
2.4.1. Thermal conductivity................................................................................. 31

iv 
 
 2.4.2. Specific heat capacity................................................................................. 31
2.4.3. Thermal diffusivity..................................................................................... 32
2.5. Chemical properties of bedding material...................................................... 32
2.6. Biological properties of compost bedding material...................................... 34
2.6.1. Predominant groups................................................................................... 34
2.6.2. Bacteria...................................................................................................... 35
2.7. Mathematical modeling................................................................................. 36
2.8. Computational Fluid Dynamics (CFD) ........................................................ 39
2.8.1. ANSYS....................................................................................................... 41
2.8.2. Governing equations.................................................................................. 42
2.8.3. Method solution......................................................................................... 43
2.8.4. Simulation in CFD for animal facilities..................................................... 45
2.9. Reduced models............................................................................................ 49
2.9.1. Similitude theory........................................................................................ 50
2.9.2. Buckingham Pi -Theorem.......................................................................... 52
2.9.3. Models versus prototype............................................................................ 53
2.10. Wind tunnel................................................................................................. 55
2.10.1. Wind tunnel design.................................................................................. 58
2.10.2. Flow visualization.................................................................................... 61
3 - MATERIAL AND METHODS.............................................................................. 62
3.1. Introduction................................................................................................... 62
3.2. Characterization and description of CBP barns............................................ 62
3.2.1. Barn facilities measurements..................................................................... 62
3.2.2. Environmental measurements.................................................................... 63
3.2.3. Bedding temperature measurements.......................................................... 64
3.2.4. Bedding moisture analysis......................................................................... 65
3.3. Physical, chemical, bacterial and thermal properties of compost
materials...................................................................................................... 67
3.3.1. Sample preparation and characteristics of materials.................................. 67
3.3.2. Physical properties of compost materials................................................... 67
3.3.2.1. Particle size distribution.......................................................................... 67
3.3.2.2. Bulk density, particle density, and porosity............................................ 69

v 
 
 3.3.2.3. Water holding capacity (WHC) ............................................................. 70
3.3.3. Thermal property measurement................................................................. 71
3.3.4. Chemical properties of the compost bedding material............................... 75
3.3.5. Bedding bacterial of the compost bedding material................................... 76
3.4. Study of natural ventilation in reduced model of CBP
barn............................................................................................................ 77
3.4.1. Wind tunnel................................................................................................ 77
3.5. Reduced model.............................................................................................. 80
3.5.1. Determination of pertinent variables.......................................................... 81
3.5.2 Reduced model procedure........................................................................... 83
3.5.2.1 Experimental reduced model.................................................................... 83
3.5.2.2 Heating system......................................................................................... 84
3.5.2.3 Static pressure, airspeed and temperature measurements........................ 85
3.6. The CFD model............................................................................................. 87
3.6.1. CFD representation of the reduced model geometry................................. 87
3.6.2. Computational modeling............................................................................ 89
3.6.3. Boundary conditions.................................................................................. 89
3.6.4. Statistical analysis methods........................................................................ 90
4 - RESULTS AND DISCUSSION.............................................................................. 92
4.1. Introduction................................................................................................... 92
4.2. Characterization and description of CBP barns............................................ 92
4.2.1. Barn structure and layout........................................................................... 92
4.2.2. Description of structure and layout in each barn....................................... 99
4.2.3. Compost bedding pack management......................................................... 129
4.2.3.1. Management practices in each barn........................................................ 130
4.2.4. Environment characteristics....................................................................... 158
4.2.5. Herd characteristics.................................................................................... 162
4.2.6. Producer Responses................................................................................... 162
4.3. Physical, chemical, and thermal properties of compost
materials................................................................................................... 165
4.3.1. Compost particle size distribution.............................................................. 165
4.3.2. Particle density, bulk density, and porosity............................................... 166

vi 
 
 4.3.3. Water holding capacity.............................................................................. 168
4.3.4. Chemical properties of composts............................................................... 171
4.3.5. Bacterial analysis of composts................................................................... 172
4.3.6. Thermal properties of composts................................................................. 173
4.3.6.1. Thermal properties of composts in different moisture content and
compaction degree.................................................................................. 173
4.3.6.2. Thermal properties of composts in different particle
size...................................................................................................... 194
4.4. Develop computational fluid dynamics (CFD) model of compost
barn.............................................................................................................. 198
4.4.1. Mesh details............................................................................................... 198
4.4.2. Validation of reduced model and simulations............................................ 200
4.4.3. Airflow patterns......................................................................................... 228
4.4.4. Computational simulation.......................................................................... 228
4.4.4.1. Applying computational simulation of the reduced model.................... 228
4.4.4.2. Evaluation of the types of ridge vents and wind direction...................... 255
5 - CONCLUSION........................................................................................................ 257
6 - REFERENCES........................................................................................................ 260
7 - APPENDICES......................................................................................................... 275

vii 
 
 ABSTRACT

DAMASCENO, Flávio Alves, D.Sc., Universidade Federal de Viçosa, August, 2012.

Compost bedded pack barns system and computational simulation of airflow through
naturally ventilated reduced model. Advisor: Fernando da Costa Baêta. Co-advisers: Ilda
de Fátima Ferreira Tinôco and Luiz Fernando Teixeira Albino.

Compost bedded pack (CBP) barns have been receiving increased attention as an
alternative housing system for dairy cattle due they consist of a roofed open loose resting
area that is bedded with sawdust where compost is actively stirred to aerate in order to
maintain an active composting process. Thus systematic investigation of the main
management practices of dairy operations utilizing green or kiln dried sawdust in compost
CBP barns at Kentucky state (USA) has proven to be of environmental and economic
relevance. The present work has as main objectives: a) Summarization of compost bed
data, barn dimension data and determination of the major interactive factors in the success
of bed composting from qualitative and quantitative methods; b) Summarization of
moisture contents and water holding capacities of the compost bed to determine the
influence of moisture on successful compost bed operation; c) Development of predictive
equations for thermal properties of compost as a function of various combinations of
moisture content, compaction, and particle size; and d) Development of a three-
dimensional computational fluid dynamic (CFD) model of compost barn with alternate
ridge designs and visual demonstration of its impact on air flow through structure. Data for
this study was collected from 42 CBP barns, distributed throughout the state of Kentucky
(USA) between October 2010 to March 2011. These data were used to describe the
building layouts and dimensions, identify barn management practices, and characterize the
compost bedding material concerning moisture content, WHC, particle size, thermal
properties, and others parameters. The average herd size found in the study was 90 cows
and the breeds consisted of Jersey (6.8%), Holstein (72.7%), and mix (20.5%).The average
CBP barn dimensions were 49.1 m (length) and 21.9 m (width). Majority of these barns
had feed alleys and driveways; overshot ridge with frequent orientation from NE to SW;
green sawdust, kiln-dried sawdust, and mixture of both as most common bedding material;
bedding process and aeration using mechanical tillage at depths lower than 0.25 m; and
loading of fresh material added every one to five weeks, varying by season, weather
conditions, barn size, and cow density. Temperatures of bedding compost tended to be
viii 
 
lower on barns which utilized green sawdust. Average bedding moisture content was found
to be 59.0% (w.b.) and ranged from 36.2 to 71.8%. The percentage distributions of mass
weight in the particle size fractions from the bedding compost materials were very different
when compared among them. The average finer index (p < 0.05) was the highest weight
percent (30.1%) in the samples studied. WHC increased with increasingly fine particle size.
The higher bulk density value was 3.6 times that of the lowest bulk density value. In
general, the thermal conductivity increased with increasing particle size, while thermal
resistivity decreased with increasing particle size. An increasing trend in the thermal
conductivity was also observed with the increase in both moisture content and static and
dynamic compaction degree for all compost material tested. Whereas, an increasing trend
in the thermal resistivity of bed compost was observed with the decrease in both moisture
content and static and dynamic compaction degree. Bedding compost always contained
bacteria, moreover Coliforms count was not present in barns that had a higher compost
temperature and Coliforms, E. coli, Bacillus and Streptococcus count were higher in the
barns that showed lower moisture content. The developed barn ventilation CFD model
provided good agreement with experimental measurements and it was able to identify the
effect on air flow through structure in a CBP barns with alternative roof ridge type. The
CFD models results showed that the best ridge vent and wind direction in the winter
weather was observed in the open ridge with chimney (ORC) and West to East,
respectively.

ix 
 
 RESUMO

DAMASCENO, Flávio Alves, D.Sc., Universidade Federal de Viçosa, Agosto, 2012.

Galpões com sistema de cama com compostagem e simulação computacional do fluxo
do ar através de modelo reduzido naturalmente ventilado. Orientador: Fernando da
Costa Baêta. Coorientadores: Ilda de Fátima Ferreira Tinôco e Luiz Fernando Teixeira
Albino.

Galpões com sistema cama em compostagem (CBP) têm recebido maior atenção como um
sistema de criação alternativo para gado leiteiro por ser constituído de uma área de
descanso aberta e coberta utilizada como cama serragem onde o composto é ativado e
aerado por agitação a fim de manter um processo de compostagem ativo. Sendo assim, a
investigação sistemática das principais práticas de manejo das operações de produção de
leite empregando-se serragem verde ou seca nos compostos de instalações do tipo CBP no
estado de Kentucky (USA) tem demonstrado ser de grande interesse econômico e
ambiental. O presente trabalho teve como objetivos principais: a) Sumarizar os dados do
composto da cama, dimensão das instalações e determinar os principais fatores interativos
para o sucesso do composto da cama a partir de métodos qualitativos e quantitativos; b)
Sumarizar a capacidade de retenção de água do composto da cama (WHC) e comparar dos
teores de umidade encontrados com a WHC, a fim de determinar a influência da umidade
sobre o composto da cama; c) Desenvolvimento de equações de previsão para propriedades
térmicas em composto de cama em função de varias combinações de teor de umidade,
compactação e tamanho de partícula; e d) Desenvolvimento de um modelo em três
dimensões em dinâmica de fluídos computacional (CFD) de instalações em compostagem
com projetos alternativos de cumeeira e demonstração visual do impacto do fluxo do ar
sobre a estrutura. Dados para realização deste estudo foram coletados em 42 instalações
CBP, distribuídos através do estado de Kentucky, entre outubro de 2010 e março de 2011.
Estas informações foram usadas para descrever o esboço e as dimensões das instalações,
identificar as práticas de manejo e caracterizar o material de compostagem da cama quanto
ao teor de umidade, capacidade de retenção de água (WHC), tamanho da partícula,
propriedades térmicas, entre outros parâmetros. A média de tamanho do rebanho
encontrado nesse estudo foi de 90 vacas com proporção de 6,8% de Jersey, 72,7% de
Holstein e 20,5% de mistura destas raças. A média das dimensões das camas de
compostagem foram 49,1 m (comprimento) e 21,9 m (largura). A maioria destas
x 
 
instalações tinham corredores de alimentação e acesso de veículos, cumeeiras do tipo
“overshot” (com um dos lados ultrapassado) e frequência de orientação de NE para SO,
serragem verde, seca ou mista como materiais mais comuns da cama, preparo da cama por
aeração mecânica a profundidade menores do que 0,25 m e carga de material fresco
adicionado a cada uma ou cinco semanas, variando por estação do ano, condições
climática, tamanho da instalação e densidade animal. As temperaturas dos compostos das
camas tenderam ser menores em instalações que utilizaram serragem verde. A umidade
média do material da cama foi de 59,0% (b.u.), variando de 36,3% a 71,8%. A comparação
da distribuição percentual do peso (massa) nas fracões de tamanho de partícula revelou que
esses materiais de compostagem da cama encontrados nas instalações CBP foi muito
diferentes. Assim, a média do índice de finura (p < 0.05) foi a porcentagem de peso mais
alta (30.1%) nas amostras estudadas. WHC aumentou com o aumento de partículas finas. O
maior valor de densidade aparente da cama foi 3,6 vezes maior que o menor valor de
densidade aparente encontrado. Em geral, a condutividade térmica aumentou com aumento
do tamanho da partícula, enquanto a resistividade térmica diminuiu com o aumento do
tamanho da partícula. Uma tendência de aumento na condutividade térmica foi ainda
observada com o aumento no conteúdo de umidade e grau de compactação estática e
dinâmica para todos os materiais de compostos testados. Por outro lado, uma tendência de
aumento na resistividade térmica do composto da cama foi observada com a diminuição do
conteúdo de umidade e do grau de compactação estática e dinâmica. Os materiais da cama
sempre mostraram presença de bactérias, sendo que as contagens de Coliformes não
estavam presentes em instalações que possuíam compostos com temperaturas mais
elevadas e as contagens Coliformes, E. coli, Bacilo e Estreptococos foram maiores nas
instalações que apresentaram menores teores de umidade. O desenvolvimento do modelo
em CFD permitiu uma boa concordância com medições experimentais e foi capaz de
identificar os efeitos do fluxo de ar através da estrutura em instalações CBP com diferentes
tipos de cumeeiras. Os resultados dos modelos CFD mostraram que a melhor tipo de
telhado e direção do vento no durante o periodo de inverno foi observado em cumeeira
aberta com chaminé (ORC) e oeste para leste, respectivamente.

xi 
 
 1 - INTRODUCTION

1.1 Summary and justification

In the late 1980`s, innovative dairy producers in the state of Virginia introduced a new
variation on the loose-housing dairy system generally referred to as a compost bedded pack
(CBP) barn. Since then, CBP barns are being used in many states in the United States,
especially in the Midwest and Northeast, and in other countries such as Japan, China,
Germany, Italy, Netherlands, Israel, and recently in Brazil. The first CBP barns of
Kentucky were constructed in 2002. By the end of 2010, CBP barn number had been
increased to 58 barns. This significant increase was an indicative that a CBP barn may be a
reasonable, economically feasible alternative type of dairy housing facility for Kentucky
dairy producers wanting to upgrade or modernize their milking herd facilities. As
consequence, factors based on dairy producer experiences such as current design,
management recommendations of manure compost packs, animal behavior, herd health,
and milk production has been described in some publications (Russelle et al., 2009).
Additional field observations have also been reported in several U.S. states (Shane et al.,
2010; Barberg et al., 2007; Janni et al., 2007) and specially in Kentucky (Black et al.,
2011a; Black et al., 2011b; Bewley and Taraba, 2009).
CBP barn structures generally have a retaining wall that typically surrounds the bed
and also separates the feed alley from the composted manure pack. The most commonly
employed bedding material in the composting process is green and dry sawdust or kiln
dried shavings to maintain a dry bedding surface and to absorb water. However, producers
often ask whether green sawdust (high moisture sawdust from sawing green wood) is
harmful or beneficial to the composting bed. The primary difference between green and
kiln dried sawdust is that green sawdust will not absorb as much water as kiln-dried
sawdust. All that this implies is that to achieve the same compost bed moisture content,
more green sawdust will be used than kiln dried sawdust.
Compost packs produced by composting are tilled or stirred twice daily to incorporate
the manure, provide a fresh, soft surface, and aerate the pack to encourage aerobic
microbial activity in the pack (Janni et al., 2007). Compared to windrow composting, CBP
has a larger surface area to heat generating volume, thus more heat losses. Since the
temperature is one of the most important CBP operational parameters, the understanding of

1 
 
the thermal properties and heat balance in compost systems is required to design and
properly control the composting processes (Ahn et al., 2009; Janni et al., 2007).
Additionally, the key to maintaining a good working CBP barn include, a structure
designed with an open ridge and high open side walls, enhanced ventilation, frequent bed
stirring, optimal moisture in the bedded pack, appropriate biodegradability of the
bedding/manure bed mixture, and optimal animal density.
Therefore, the effectiveness of the composting process is dependent upon the
environmental conditions as oxygen content, moisture, temperature, amount of organic
matter, and the size and activity of microbial populations presents within the CBP (Janni et
al., 2005). Composting increases the bedding temperature which reduces pathogenic
microbial populations, and decreases bedding moisture by increased drying rate (Black et
al., 2011a; Barberg at al., 2007). The large numbers of microorganisms and high microbial
activity in the compost bedding are largely a reflection of manure deposition, bedding
chemical properties (available C and N), and water contents. Facultative anaerobic bacteria
also are more numerous in the surface of CBP material and constitute a larger proportion of
the total microbial population in tilled or stirred compost (NRCS, 2010). Some research
shows the direct influence of poor environment conditions that could be created within the
CBP as one of the factors that predispose the development of respiratory diseases in birds
and pigs (Reece et al., 1980; Sainsbury, 1981; Takai et al., 1983; Menegali et al., 2009), but
a thorough search of the literature found few studies related to air quality in facilities for
cattle.
A more in-depth study is needed to assess and correlate detailed information regarding
the building structure, bed management, and compost parameters of these housing systems
and to determine the most significant factors that achieve CBP success. Thus, the findings
of this research may be used by current and future compost bedded pack barn managers.

2 
 
1.2 Dissertation objectives
The present work will provide a systematic investigation of the main management
practices of dairy operations utilizing green or kiln dried sawdust in their compost bedded
pack barns. Four major objectives were established:

1) Summarization of compost bed data, barn dimension data and determination of the
major interactive factors in the success of bed composting from qualitative and quantitative
methods;

2) Summarization of moisture contents and water holding capacities of the compost bed
to determine the influence of moisture on successful compost bed operation;

3) Development of predictive equations for thermal properties of compost as a function

of various combinations of moisture content, compaction, and particle size;

4) Development of a three-dimensional computational fluid dynamic (CFD) model of

compost barn with alternate ridge designs and visual demonstration of its impact on air
flow through structure.

 
 

3 
 
 2 - LITERATURE REVIEW

2.1. Compost dairy barn system

For the past several decades, loose housing systems of dairy cattle has given way to
freestall barns in light of less labor, cleaner cows and reduced bedding costs when
compared to the conventional straw bedded-pack. Nevertheless, loose housing with
freestalls can still lead to a high risk of lameness and hock lesion. Since the concrete
flooring systems are constantly covered with urine and manure, this type system may cause
digital dermatitis and other hoof disorders (Klaas et al., 2010), especially with inadequate
positioning of stall hardware, limited area resting space, hard lying surface or insufficient
bedding (Weary and Taszkun, 2000; Tucker et al., 2004; Dippel et al., 2009). In view of
this, a new concept in loose housing called compost bedded dairy barn system, commonly
called compost barns or compost bedded pack barns (CBP), has been gaining attention
from dairy producers and the scientific community.
CBP barns are an alternative loose housing system for dairy cows. In CBP barns, cows
have more freedom of movement and are able to lie down in a more natural manner. In
theory, this should lead to cows spending more time lying down. However, even as cows
are standing up, they also spend more time on a softer surface than in concrete alleys
within freestall barns. Thus, cow welfare and health are important indications of the
compost bed. Bedding quality can also be decreased as the moisture content increase once
the cows are directly exposed to the elements constituting of bedding, feces and urine
resulting in an increase in the bacterial cells counts of both bedding and teats (Zdanowicz
et al., 2004). But, when CBP barns are properly managed, drier bedding material is present
and may result in decrease of the milk somatic cell counts (SCC), cleaner cows, increased
milk production, and reduced lameness. A published example of this was found in
Minnesota where the percentage of mastitis infection rates in CBP barns was 12% lower
than in freestall herds (Barberg et al., 2007).
CBP barns are generating a lot of interest among Kentucky dairy producers who are
adopting CBP barns as dairy cattle housing at a rapid rate. Kentucky has at least fifty one
compost barns in operation and some more under construction. Experience with well-
managed CBP barns in Kentucky has generally been positive. Cows are relatively clean,
very comfortable, have fewer lameness issues, and in most cases has increased milk

4 
 
production and lower SCC after moving from tie-stall or freestall barns, or pasture based
systems to a compost barn.
However, since this is a new system, many questions still remain regarding best
management practices and key factors for their success. More research is needed on these
housing systems after assessing the effect of current design and management
recommendations based on dairy producer experiences. As result, CBP barn
recommendations can be made for proper design, location, and exceptional management.
Thus, the keys to maintaining a good working CBP barn should include a structure
designed with an open ridge and high open side walls, enhanced ventilation, frequent bed
stirring, optimal moisture in the bedded pack, and optimal animal density (Wagner, 2002)
in order to provide a healthy, comfortable surface on which cows may be kept to result in
with a higher quality of life and milk production.

2.1.1. Facility design

CBP barns are similar in many respects to typical freestall dairy barns (NRAES-76,
1995; MWPS-7, 2000). Both have feed mangers, feed alleys and waterers. One of the key
differences is that the freestall and freestall alleys are replaced with a bedded pack area in
CBP barns which it is aerated at least twice daily.
CBP barns have a concrete feed alley, a composting bedded pack, a 1.20 m high wall
surrounds the pack on three sides and one side that separates the pack from the feed alley.
The 1.20 m wall the bedded pack and the feed alley has a fence for safety to prevent cows
from walking over the wall (Endres and Janni, 2009). The wall separating the pack and the
feed alley has one to four walkways for cows and equipment to access the pack. This wall
is typically of poured concrete construction.
A feed alley is located along one long side of the bedded pack. CBP barns can be laid
out for drive-by feeding, covered feeding, or drive-through feeding with pens on both sides.
They can also be used with outside feed bunks under roof. The recommended feed alley
width is 4.25 m because cows eat off one side of the alley and drink from waterers on the
other side (Gay, 2009).
The bedded pack area can have a concrete or clay base. The sidewall height for a
bedded pack is recommended to be higher than that for a freestall barn to accommodate the
sidewall opening lost due to the manure pack walls. A size 0.91 m or more of eave

5 
 
overhangs is recommended to minimize the chance of roof runoff and rain being blown
onto the bedded pack (Janni et al., 2005). Roof gutters will help reduce roof runoff from
being blown onto the pack. The ground surrounding a compost barn should be sloped to
minimize rain and snow runoff from entering the barn and wetting the bedded pack (Endres
and Janni, 2009). Dairy producers in Minnesota have used 4.25 m sidewalls but some
producers have used 4.90 m for their next barn to provide better access for bedding trucks.
Walkways, at least 3.05 m to 3.65 m wide, are minimum recommended at each end of the
wall separating the bedded pack and the feed alley for cow and equipment access to the
pack area. Long barns will require additional walkways (Endres and Janni, 2009).

2.1.2. Waterers
Water intake increases during heat stress, and one of the critical factors in managing
heat stress is to provide adequate access to water. Economy and ruggedness are two criteria
for selection of the waterers. Two choices for materials are concrete or plastic. Concrete is
more rugged for the long term but requires more skill and equipment for forming and
placing. Plastic can be easy to modify or repaired and economically viable.
Waterers located closer to the feed alley are easier to inspect from the alley and
cleaning can be performed more easily. Many CBP barns have the waterers along the
retaining wall with access from, the feed alley, usually within an indented space . Waterers
are not located in the bedded pack area to minimize wetting of the pack, to keep cleaner
waterers, and to avoid having to adjust waterer height as pack depth increases. Ideally, 0.60
meters of tank perimeter should be provided for each 10 to 20 cows (Brouk et al., 2001).

2.1.3. Lighting
Lighting in compost barns is similar to that used in freestall barns (MWPS-7, 2000).
Lighting in CBP barns is desired mainly for the convenience of the operator, for inspecting
animals and handling dairy cows. Dairy producers, veterinarians, and other animal care
workers often report easier and better cow observation and care. Cows move more easily
through uniformly lit entrances and exits.
Enhanced lighting for the milking herd is profitable (Dahl et al., 1997). The increased
milk production generates a payback of less than one year considering initial, installation,

6 
 
operating, and replacement costs of the lights, and the increased feed intake (Chastain and
Hiatt, 1998).
Light levels are measured in lux. The power and type of the lamp determines the
brightness. The height of the lamp above the ground and the lamp spacing determines the
uniformity and distribution of light (Winchell et al., 1996). Yard lights should therefore be
located over feed alleys and/or waterers.

2.1.4. CBP barn ventilation and location

Determining the best CBP barn ventilation system to employ is one of the biggest
facility related decisions to make it. The ventilation in the CBP barns helps in maintaining
a healthy and comfortable environment for cows and reduces the effect of environment on
the structure of the building. The renewal of the air removes cow heat and moisture as well
as the odor, gases, heat and moisture that the composting pack generates. It is necessary to
constantly exchange warm, humid air inside the barn for cooler, drier air outside of the
barn. This exchange must occur regardless of outside temperature or weather conditions.
A continuous supply of fresh air is necessary whenever animals are closely confined, as
in CBP barn, in order to maintain animal health and integrity of the building. Evidence of
inadequate ventilation includes condensation or condensation stains on walls, ceilings and
structural members, and mainly the presence of foul odors in the barn. Perennial respiratory
diseases may be another symptom of poor ventilation. Winter barn temperatures should not
exceed 10°C. If this temperature is often exceeded, inadequate ventilation may be the
reason (Wells, 2004).
The rate of ventilation is the air volume changed in a given period of time. The
ventilation system in dairy barns may be natural or mechanical ventilation (positive or
negative), or a combination of both (Teye and Hautala, 2008), as seen in Figure 2.1.
A mechanical ventilating system uses fans to provide air exchange. Both naturally and
mechanically ventilated buildings need properly sized openings for fresh air to enter and
exhaust air to exit. Ventilating system control can be accomplished either automatically or
manually. Mechanical ventilating systems almost always use thermostatic control while
natural ventilating systems are commonly adjusted manually (MWPS-7, 2000).

7 
 
 (a) (b)

(c)
Figure 2.1 - Ventilation system in dairy barns: (a) negative pressure, (b) positive and (c)
natural (Adapted from Teye, 2008).

2.1.4.1. Natural ventilation in CBP barns

In a natural ventilating system, the wind and the difference between inside and outside
temperature moves air through the building. Natural air exchange or building ventilation
rates are influenced sidewall opening, eave opening, building width, ridge opening, and
wind speed (Brouk et al., 2001).
Wind blowing across the open ridge and sidewall and endwall openings, and
indoor/outdoor temperature differences moves air through the building to provide the air
exchange. Fresh air enters through eave or sidewall openings, see Figure 2.1c.
Heat, moisture, and air contaminants exits through the open ridge and downwind
sidewall opening. Some ventilation occurs even on calm days because warm, moist air rise
up causing a chimney effect. Large, adjustable or full sidewall openings allow increased air
movement due to wind in summer. Continuous openings help provide good fresh air
distribution.

a) Location and Sitting

The most CBP barns are naturally ventilated, which makes location very important.
Naturally ventilated barns are located in an open area where summer winds can blow
through open sidewalls and the ridge in warm weather (Teye, 2008). Building location is
8 
 
critical for natural ventilation to work well. Buildings on high ground are located where
trees or structures do not disturb airflow around or through the building.
Trees, silos, tall growing crops, and other structures can disrupt airflow for a distance
of 5 to 10 times their height downwind. Thus, naturally ventilated buildings should be
located at least 25 meters, in any direction, from such obstructions (MWPS-7, 2000).

b) Building Orientation
The first dairy barn design criteria to be considered are the orientation of the structure.
Barns with a north-south orientation have a greater internal solar radiation exposure than
barns with a east-west orientation (Brouk et al., 2001). Moreover, building orientation
affects natural ventilating system performance. Orient naturally ventilated buildings to
provide maximum wind exposure, especially for summer winds. Greater and more uniform
wind driven air exchange occurs when the wind enters through the continuous sidewall or
eave opening rather than through the smaller end wall. Cross ventilation, from side to side,
is especially important during warm weather. Orient buildings so prevailing summer winds
are perpendicular to the ridge openings (MWPS-7, 2000).
The ideal orientation of the barn allows prevailing summertime winds to intersect the
barn perpendicular to the sidewalls. When the farmer builds the barn using this orientation,
the air entering the barn though the sidewall travels the shortest distance possible to exit the
barn through the opposite sidewall. This improves the rate of air exchange in the barn and
consequently enhances the environment of the cows. Since winds change direction, winds
striking the sidewall within 45 degrees of the perpendicular will still result in adequate air
exchange in most cases. Other factors that can affect barn orientation include future
expansion, topography, cow flow, and manure flow. Producers should consider all of these
factors when sitting and orientating their barns.

c) Ventilation Openings
The primary types of openings used in a naturally ventilated dairy barn are eave, ridge,
and wall openings (sidewall and endwall). Controlling the size of these openings can
control the ventilation rate to a great extent. The amount of inlet control needed depends on
the season of the year and the climate (Chastain, 2000).
In addition to building width and sidewall height, ridge openings are required.
Armstrong et al. (1999) observed greater increases in afternoon respiration rates relative to
9 
 
morning rates when cows were housed in barns with ridge coverings as compared to
opening ridges. A ridge opening of at least 0.15 m (measured horizontally) should be
provided for each 9.0 m of building width (Chastain, 2000) and the bedded pack area or
other less critical areas should also located under the ridge (Brouk et al., 2001).
The sidewall opening should be at least 2.50 meters high, continuous the entire length
of the building. Open sidewalls fully to provide maximum summer ventilation. With large
sidewall openings, provide a 0.90 to 1.20 meters overhang at the eave to shade and partially
shield the opening from blowing rain and snow piles and drifts. Ventilation through
existing buildings can be improved by increasing the amount of open area, but do not
compromise the structural integrity of the building. Endwall openings can provide
additional ventilation to supplement sidewall openings in the summer (MWPS-7, 2000).
The vertical separation between the eave and ridge openings has a significant impact on
the pressure differences generated by the chimney effect (Chastain, 1987). Therefore, the
roof slope is an important consideration in the design of a naturally ventilated building
(Chastain, 2000). Roof slopes of 4/12 to 6/12 works very well in naturally ventilated dairy
barns. Condensation and high interior summer temperatures are more likely with roof
slopes less than 4/12 because of reduced air movement. Buildings with roof slopes steeper
than 6/12 have greater chimney effect but they are generally not needed. Slopes of 3/12 are
acceptable for open front buildings less than 9.0 m wide (MWPS-7, 2000).
Determine ventilation rates in naturally ventilated facilities is a task even more difficult
(Albright, 1990; Zhang et al., 2005). The air flow in CBP barn naturally ventilated is
irregular and multidirectional, and generally has a very small value to measure accurately.
Ventilation is also driven by temperature differences between indoor and outdoor
installation. Therefore, to accurately estimate the natural ventilation is very difficult,
because the areas of ventilation openings are large and along the entire length of the
installation or through the ridge vent. Most CBP barns which has ventilation problems are
resulted from poor design, construction or ventilation operational control system (Seedorf
et al., 1998).

2.1.4.2. Mechanical ventilation in CBP barns

There are endless approaches to the mechanical ventilation design in dairy farming. For
example, the precise number and location of fans required in a building would depend upon
factors such as the density, construction, utilization and local climatic conditions (Wells,
10 
 
2004). However, the components of the ventilation system should accommodate winter and
summer conditions, such as:
i) A continuous minimum air exchange: This is certainly required to remove the
moisture constantly created by dairy cows, particularly during the colder, sub-freezing
months of the year.
ii) Higher air exchange rates and air velocity across the cows during the hotter summer
weather to remove excessive heat from the cows.

Mechanical ventilating systems have fans, controls, and air inlets or outlets. A well-
designed system provides greater control over barn temperature and air movement than
natural ventilation (MWPS-32, 2000). Mechanical ventilation is used to ventilate barns or
rooms with flat ceilings and buildings that are blocked from good natural ventilation by
existing buildings and obstructions.
Fans are the driving force in providing the required air exchange and it must be
combined with properly sized and distributed fresh air inlets in order to work properly. The
main criteria for selecting ventilation fans are capacity to provide a good air throw, fan
size, drive type, and purchase and operating costs. In barns with tight construction and
continuous, fan location has little effect on air distribution. Fans placed in groups or banks
at a single location are preferred rather than spacing them out singly at different locations
(Wells, 2004). Thus, fans should be selected to provide the required air volume for
different weather conditions based on the number of cows in the barn.
The minimum ventilation rates for dairy cows are given in Table 2.1. Recommended
ventilation rates depend primarily upon the total weight of livestock housed in the space to
be ventilated.

Table 2.1 - Minimum ventilation rates for dairy barns.

Animals Winter Mild weather Summer
3 -1
Air flow (m s ) per animal
Calves (0 - 2 months) 0.007 0.024 0.047
Heifers
(2 - 12 months) 0.009 0.028 0.061
(12 - 24 months) 0.014 0.038 0.085
Cows 0.024 0.080 0.236
Source: Dairy freestall housing and equipment handbook (MWPS-7, 2000).

11 
 
 A recent innovation for hot weather cooling in CBP barns is the High Volume Low
Speed (HVLS) ceiling fan (Figure 2.3). These fan units may be up to 7.2 m in diameter and
they are operated by a 1 to 2 hp motor (Clarke and House, 2010). While the units are
expensive to purchase, the payback period falls into the 4 to 5 year range, depending on the
electricity prices and installation costs at the time. HVLS fans create a higher percentage of
floor area with fewer dead or low velocity spaces.

(a)

(b) (c)
Figure 2.2 - High Volume Low Speed (HVLS) ceiling fan (a) and differences before (b)
and (c) after the installation HVLS in CBP barns.

The Figure 2.2 illustrates a typical dairy herd enclosure in the height of the summer
months without mechanical circulation. Thereby, before ceiling fans (HVLS) are installed
(Figure 2.2b), the warm, muggy air sits in the building, which creates uncomfortable
conditions for the cattle, creating heat stress for the animals, attracting flies, whilst also
creating uncomfortable working conditions for staff. By installing ceiling fans within the
enclosure (Figure 2.2c), this provides positive air movement, equalizing the temperature
levels within the building during the summer for cattle and workers, with better cooling to
minimize heat stress with the constant air circulation.
12 
 
2.1.5. Economic considerations
Numerous economic factors must be considered before deciding to build a CBP barn.
CBP barns have lower capital costs compared to freestall barns because CBP barns require
less concrete and stall equipment. In other words, the bedded pack can be built up on clay.
Furthermore, there is less investment in manure storage structures because the bedded pack
provides storage (Gay, 2009).
Bedding costs and availability of bedding materials was by far the major concern
expressed by the producers. Dairies with more than 200 cows may want to consider an
alternative bedding to a CBP barn if they plan to use a compost barn for all their cows. A
large number of cows will produce enough manure to require a significant amount of
bedding material. In CBP barns, approximately three to four times more bedding is used in
a CBP barn as in a freestall barn. Alternative bedding options are currently being tested.
Preliminary results indicate that sawdust is the best option for these types of barn, but
combinations of sawdust with other materials, such as finely ground soybean or flax straw,
finely processed corn cobs, or wood chip fines can work relatively well (Endres and Janni,
2009).
The building costs ranged widely depending on the amount of on-farm labor utilized
and amenities added. Labor requirements are similar for both types of barns. Properly
maintained CBP barns require more daily labor (packs must be stirred twice daily and must
have dry bedding added). However, less daily labor is required for manure handling.
Producers should also consider the cost of handling manure. CBP barns reduce the
amount of manure storage, but producers may need equipment to handle both liquid and
dry manure. Keep in mind that dry manure can be easier to handle and may not have to be
applied as frequently (Bewley and Taraba, 2009).

2.2. Compost bedding pack

2.2.1. Bedding source
There are two driving factors behind good bedding choices. One is cow comfort, and
the other is farmer comfort. Both of these objectives have some common and diverging
areas. Cow comfort is critical because of the importance to both cow and farmer for the
cow to spend time to lie down and performing the process of rumination. Therefore
bedding must be comfortable to lie on. Nonabrasive bedding promotes both comfort and

13 
 
injury reduction. A dry bedding surface is critical at all times both for cow comfort,
reduction of infection from environmental mastitis organisms and for reduction in pathogen
growth. Farmer comfort requires low bedding cost, labor efficiency, and the bedding drain
or dry well to keep cows dry and limit growth of pathogens (Hashemi et al., 2001).
Besides, drainage of bedding can be considered a negative factor, because regulatory
agencies requires protection from groundwater contamination. Thus, operation of the bed
should be below the water holding capacity (WHC).
Fine wood shavings or sawdust is most commonly used in CBP barns as a bedding
material (Janni et al., 2005). Wood shavings are preferred over sawdust because the small
size of sawdust particles can enhance bacterial growth and irritate teat ends. Sawdust
should be kiln-dried if used as bedding. However, competition for wood products has
stiffened, driving up the price for wood residue. This calls into question the future cost and
availability of this bedding. Faced with this uncertainty, producers and researchers have
explored alternative materials that could work for bedding in CBP barns (Vokey, 2008).
Some experience indicates that green or wet sawdust is not recommended. Other types of
bedding may not work satisfactorily, increasing the mastitis risk for dairy cows.

2.2.2. Management of the bedded pack

Management of the composting bedded-pack is critical to the success of compost barns.
The minimum space per cow for a CBP barn is recommended to be 6.0 m2 per animal for
Jersey cows and 7.9 m2 for large breeds (Janni et al., 2007). However, 9.3 m2 or more per
cow is recommended to prevent overcrowding and compaction of the bedding (Gay, 2009).
Overcrowding may lead to increased SCC levels, increased bedding needs and potential for
increased teat injury. The compost pack area is sized to allow all cows to lie down at the
same time and still have space for a cow to get up to go and eat or drink. It is hypothesized
that compost pack area per animal depends on the amount of manure and urine added to the
pack daily. Moreover, more space per animal will reduce the frequency of bedding addition
as a result of water evaporation rate, as affected by weather conditions and bed
temperature, that may be comparable to deposition rate from manure and urine. The
microbial activity and drying need to balance the amount of manure and urine added daily.
CBP barns have bedding covering the loose housing floor to an initial depth of 0.30 to
0.60 m. Cows deposit manure and urine onto the bedding when utilizing the CBP barn.
Sometimes confinement in the barn is continuous between milking sessions. When cows
14 
 
are in the barn, they have free access to a concrete alley where waterers and the feed bunk
are installed. While the cows are being milked (two or three times per day), the bedded
pack is tilled or stirred to incorporate the urine and manure that accumulate between
milkings and to aerate the bedding material. Tillage is done with a skid steer, small tractor,
or even an ATV equipped with a variety of tools (field cultivators, harrows, and rototillers
have been employed), to a depth of 0.20 to 0.40 m. It is proposed that aeration is essential
to incorporate oxygen for aerobic decomposition and to provide a fresh surface without
accumulated manure for cows to lie down on after returning from the milking parlor or
feeding. Some producers are aerating the pack deeper using a chisel plow with 0.40 to 0.45
m deep, and have observed a reduction in bedding needs and increased pack temperatures
(Janni et al., 2007). Manure handling on a daily basis takes about the same time or less as
maintaining a freestall barn.
Fresh bedding is added when there is moisture build up in the bedding and the cows
become dirty. Typically this occurs every few weeks. The CBP barn may have increased to
a depth of 1.0 to 1.2 m when removed after one year.

2.2.3. Composting process in bedded pack

Composting is the biological decomposition process that occurs under controlled
aerobic conditions where microorganisms reduces heterogeneous organic matter into more
stable chemical compounds with the release of heat, water vapour, and carbon dioxide. The
description of the microorganisms participating in the composting process is complex,
because the populations and communities change continuously as a function of the
evolution of temperature, nutrient availability, oxygen concentration, water content and pH
in the course of composting. The composting processes occur in nature, but systems can be
designed and managed to enhance and accelerate the process. The by-product of
composting consists of the biomass of dead and living microorganisms, and undegradable
raw material besides stable by-products of decomposition (NRCS, 2010).
Compost, in turn, is the nutrient amendment produced by an aerobic composting
process as it is viewed in a compost bedded pack which is sufficiently stabilized to
minimize or eliminate unpleasant odors, substantially reduce viable pathogens, facilitate
storage, and handling without attracting insects and vectors. However, its complete
stabilization is not considered practical or desirable because it would consume all the
slowly degradable organic matter. The degree of stabilization and pathogen destruction
15 
 
desired in compost depends on the purpose for composting and intended use of the compost
by-product with a low risk of pathogen contamination. Compost is more stable (less
microbially degradable), lighter, smaller in volume than raw manure. The finished product
will need to meet the customer’s or operator’s needs (Goyal et al., 2005; Gajalakshmi and
Abbasi, 2008).
In addition to the presence of the needed microorganisms, major factors as nutrients,
carbon to nitrogen ratio, moisture content, temperature, and aeration are important factors
to a successful composting process. Characteristics as appearance, texture, and even
maturity may not be an issue if the compost from a bedded pack is used for land
application on the farm.
In a CBP barn, the bed depth from surface that is tilled is called an oxic zone and acts
like a frequently turned aerobic compost windrow with temperatures rising to the range of
45 to 60°C when good composting conditions occur. The bottom of the CBP is called an
anoxic zone and is under an anaerobic decomposition process with little oxygen available.
No agitation occurs and it becomes compacted with the added overburden of compost, by
cows and tractor wheels. As would be typical for compacted manure piles, temperature
rises above ambient, but it is only in the 25 to 40°C range. Figure 2.3 illustrates a CBP
zones with possible gas pathways and temperature profile.

Figure 2.3 - Scheme of CBP zones with possible gas pathways and temperature profile.

As previously mentioned, composting is an aerobic process, meaning it requires large

amounts of oxygen for the microorganisms to grow. It can be supplied to the compost bed
by diffusion, wind, natural convection or forced aeration. A minimum oxygen
concentration of 5% within the pore spaces of the compost is necessary for composting

16 
 
occurs. If the supply of oxygen is limited, the composting process slows and may turn
anaerobic, which is a much slower and odorous process with the potential for large N
nutrient losses through denirification. During the initial stages of this aerobic process,
oxygen and the easily degradable nutritional components of the raw materials are rapidly
consumed by the microorganisms (NRCS, 2010).
Carbon (C), nitrogen (N), phosphorous (P), and potassium (K) are the essential and
primary nutrients required by the microorganisms involved in composting. Microorganisms
use carbon for both energy and growth, while nitrogen is essential for protein production
and reproduction. Thus, an appropriate ratio of carbon to nitrogen (C:N ratio) usually
ensures that the others required nutrients for their survival are present in adequate amounts.
In this process, the temperature is directly related to the microorganism activity of
composting and is a good indicator of how the process is going inside the bed (Goyal et al.,
2005; Gajalakshmi and Abbasi, 2008).
Composting process becomes inhibited when the moisture content is below 40%. Water
displaces much of the air in the pore spaces of the composting materials when the moisture
content is above 65%. This limits air movement and leads to anaerobic conditions. Once
moisture, is over 65%, the rate of decomposition in the anaerobic zone is likely not to be
high enough to generate mature compost because it is too wet to support rapid bacterial
activity. Some activity certainly does occurs, however, the material is not uniformly heated
to the extent necessary to be sold commercially as compost and it would not meet USDA
Organic Certification requirements for pathogen reduction. In addition, the volume of the
pack does not increase much during the summer months, as it does in winter, indicating
that composting is reducing the volume of the bedded pack. If the operator wants to market
the material as compost after removing it from the barn, it will require further processing
(NRCS, 2010).
Oxic zone can be modeled as a semi-continuously stirred tank reactor (CSTR) with
moisture and fresh organic matter (manure and urine) added by the confined cows and
intermittently stirred with gases continuously exchanged between atmosphere and CBP
barn. As the depth of the bed deepens with organic matter, solid material is exiting the base
of the oxic zone since tillage or mixing is only in the top 0.20 to 0.40 m. Anoxic layer can
be modeled as a batch reactor with organic matter decomposition occurring producing
gases that move up to the oxic layer. A transition oxic/anoxic zone lies between these

17 
 
layers where the oxygen in the gas phase decreases with depth from the bottom of the oxic
layer to zero concentration at the anoxic layer. The material in this zone could be mixed
during tillage (Goyal et al., 2005; Gajalakshmi and Abbasi, 2008).

2.2.3.1. Microbiological processes in composting bedded pack

In a model system, aeration decreases as depth of CBP bed increases in depth and
causes an oxygen concentration gradient between aerobic surface and anaerobic subsurface
environments. The CBP system represents a dynamic environment in which spatial
distribution of microbes and microbial metabolic processes develops. The composting
process results from the breakdown and stabilization of organic matter, with CO2 evolution
and H2O evaporation. It is impacted by relatively simple management practices such as
barn ventilation and circulation and mechanical mixing of the bed.
Manure deposition provides a steady input of unstable C, organic N (protein and urea),
and liquid fraction of the animal waste, which starts a cyclical series of microbial
transformations over time transforming a fraction of this organic matter into inorganic C or
N (Warner, 1942; Kristensen, 2000; Karthikeyan and Bhandari, 2001; Beffa, 2002;
Megonigal, 2004; Peigne and Girardin, 2004; Goyal, et al., 2005). Figure 2.4 shows a
spatial distribution of various microbiological processes in composting bedding.

Figure 2.4 - Spatial distribution of various processes in composting bedding that lead to
trace gas formation.
18 
 
a) Oxic zone:
Gaseous emissions from the oxic zone during aerobic composting would contain
emissions of ammonia (NH3) and moisture and greenhouse gas emissions (GHGs): carbon
dioxide (CO2), methane (CH4), and nitrous oxide (N2O). In aerobic respiration, bacteria and
fungi use molecular oxygen (O2) to liberate the bulk of the energy (ΔH) from the organic
matter that is being decomposed, producing carbon dioxide (CO2) and water (H2O) into the
CBP oxic zone.
CxHyOz   +   O2      CO2   +   H2O   +   ΔH 

This is the principle fate of mineralized or oxidized organic carbon, occurring first by
oxygen (O2). This process generates more energy, increasing the temperature of the tilled
top zone, and consequently inactivates the disease organisms and increases the evaporation
of moisture (Warner, 1942; Kristensen, 2000; Karthikeyan and Bhandari, 2001; Beffa,
2002; Megonigal, 2004; Peigne and Girardin, 2004; Goyal, et al., 2005).
In the manure substrate, nitrogen (N) is present in organic matter as protein or in urine
in the form of urea (CO(NH2)2). Aerobically, CO(NH2)2 undergoes hydrolysis and it is
rapidly decomposed to NH4+ and CO2 by the intracellular and extracellular urease enzyme
produced by bacteria present in both feces and soil. NH4+ and CO2 is also produced from
the hydrolysis of protein in the organic matter. This process is known as ammonification.
Controlled by thermodynamic equilibrium, the NH4+ form is in equilibrium with NH3. This
equilibrium is pH and temperature dependent. Due to concentration gradients, a fraction of
NH4+/NH3 is diffuse in the air-water interface of the manure, and it is partially emitted into
the air. However, the dominant part of NH4+/NH3 is either assimilated into microbial matter
or nitrified to NO3-.

Organic-N       NH3        NO2- (+NO, N2O)       NO3- 

        
NH4+

At the first stage of nitrification, the NH4+/NH3 fraction is oxidized to NO2-. In the
second stage, NO2- is further oxidized to NO3-. NO3- can subsequently be converted into
nitrous oxide (N2O), nitric oxide (NO) and dinitrogen (N2) through denitrification
(Blackburn, 1988; Robertson et al., 1989; Davidson, 1991; Henry et al., 1999; Maier et al.,
2000; Beffa, 2002; Senga, 2002; Megonigal, 2004). The process by which NO3- is reduced

19 
 
to nitrogen gases (N2 or N2O) is normally anaerobic but it can also occur in aerobic
conditions where anaerobic microzones exist, such as internal spaces of porous organic
particles.

b) Anoxic Zone:
The amount of denitrification depends on the availability of NO3-, on moisture
saturation, on temperature, and on availability of easily decomposable organic matter.
Thus, if the bed becomes depleted of O2, some microorganisms can use the oxygen in NO3-
as an alternative. During this process, NO3- is used as electron acceptor by fermentative
microorganisms able to grow without free oxygen. The metabolic processes occurring
while NO3- is used as electron acceptor to form N2O and N2 are equal to those occurring
with the presence of O2. In addition to N2 formation, N2O, NO and NO2 may be produced
under not completely anaerobic conditions.
The lower CBP anoxic layer also shows typical anaerobic gaseous emissions as
methane (CH4), CO2, and other major gaseous emissions such as hydrogen sulfide (H2S)
and NH3, that moves upward or are absorbed in the oxic zone. Anaerobically,
carbohydrates and organic acids originating from protein degradation are converted by
mixed heterotrophic and fermentative bacteria and fungi to acetate (CH3COO-/CH3COOH)
and CO2, which in turn are transformed into CH4 by two different microbial pathways in a
process called methanogenesis.
CO2   +   H2      CH4 
CH3COOH       CH4   +  CO2
 

Methanogenesis represents a significant portion of the carbon gas flow in the anaerobic
zone. CH4 diffusing upward reaches the anoxic/oxic zone and it is consumed by oxidizing
bacteria in a process known as methanotrophy (Megonigal, 2004). Thus, CH4 is oxidized in
the presence of O2 so that it diffuses upwards to the oxic zone.

CH4  +  O2      H2O   +  CO2 

The same microorganisms that oxidize NH3 can oxidize CH4. This implies that CH4
could inhibit NH3 oxidation and NH3 could inhibit CH4 oxidation, depending on the
concentrations. Acidity of the CBP can affect this balance (Maier et al., 2000).

20 
 
c) Oxic/Anoxic Zone:
The oxic/anoxic interface of these two layers becomes a zone where organisms can
potentially produce or decompose similar emission gases. Denitrifiers convert NO3- to
gaseous end products (NO, N2O, N2). The final gaseous phase products depend on
available C and O2 supply. Oxygen level is reduced by microbial oxidation of available
organic C, which affects the reductive steps controlled by distinct enzymes affected.

NO3-      NO2-    NO      N2O     N2

N2O is formed as an intermediate and it is available for release to the atmosphere

through diffusion or during tillage (Warner, 1942; Kristensen, 2000; Karthikeyan and
Bhandari, 2001; Beffa, 2002; Megonigal, 2004; Peigne and Girardin, 2004; Goyal et al.,
2005; Coyne, 2008). Within the oxic/anoxic environment, an alternative route for the
conversion and removal of NH3 is possible for which substantially less information is
known - anaerobic oxidation of NH3 (ANAMOX).

NH3       NO2- 
NH3  +  NO2-      N2 

This process has been found at the oxic/anoxic interface, both in an aquatic
microenvironment (Warner, 1942; Kristensen, 2000; Karthikeyan and Bhandari, 2001;
Beffa, 2002; Megonigal, 2004; Peigne, and Girardin, 2004; Goyal, et al., 2005) and in
wastewater treatment (Maier et al., 2000). In natural oxic/anoxic interface of aquatic
sediment microenvironments where CH4, NO2- and NO3- are present, the following
anaerobic methane oxidations are found (Raghoebarsing et al., 2006).

CH4    +   NO3-         N2   +   H2O   +  CO2 

CH4    +   NO2-         N2   +   H2O   +  CO2 

d) Entire Bed:
There are implications that gases, such as H2S, CO2, and CH4 that are formed in the
anoxic zone and move upward through the CBP zones, can affect the microbial processes
shown above. H2S emission can block the enzymes of microorganisms that oxidize NH3 to
NO2- at the oxic zone and the enzyme that reduces N2O to N2 at the oxic/anoxic interface of

21 
 
both zone (Joye and Hollibaugh, 1995; Kristensen, 2000; Karthikeyan and Bhandari, 2001;
Beffa, 2002; Megonigal, 2004; Peigne and Girardin, 2004; Goyal, et al., 2005).
The enzymatic system for anaerobic NH3 oxidation can also be blocked by H2S effect.
Because of these blockages, the N2O concentration at the oxic/anoxic interface rises if the
organisms producing the needed enzyme for its production are present. The NH3
concentration also rise. The CH4 produced in the anoxic zone it infiltrates the oxic zone and
compete for the enzyme that oxidizes NH3. It is because the same microorganisms that
oxidize NH3 can oxidize CH4 (Peigne and Girardin, 2004; Goyal et al., 2005).
Factors, such as, aeration, moisture, mixing, and substrate supply alter the dynamic
equilibrium that exists between all these processes. Aeration can increase methanotrophy
and nitrification process. Poor ventilation or moisture saturation can enhance the capacity
for menthanogenesis and denitrification. GHG evolution depends in large part on how the
systems are managed to generate the starting substrates, the rates of the microbial
processes, and the subsequent transfer of by-products formed between aerobic and
anaerobic zones (Warner, 1942; Kristensen, 2000; Karthikeyan and Bhandari, 2001; Beffa,
2002; Megonigal, 2004; Peigne and Girardin, 2004; Goyal et al., 2005).
Within the oxic layer, it is still possible to have the formation of anoxic zones where
dense clumps exist (Bewley and Taraba, 2009) and the air diffusion is restricted. The
tillage sweeps do not produce homogenous aggregates which leads to a second mode by
which aerobic and anaerobic microbial processes can produce GHGs. Spatial distribution
of aerobic and anaerobic zones is not simply by depth, but also by location within large
aggregates that are created in the CBP.
 
2.2.4. Composting and moisture
Composting requires sufficient moisture for active microbial activity but not so much
moisture to hinder aeration. Moisture levels between 50% and 60% are generally
recommended for the most of composting materials. Urine, wet manure, and moisture from
microbial activity are the moisture sources in a composting bedded-pack. When moisture is
too low, the microbes won’t have enough water, and the compost will be too cool, resulting
in a compost rate that it is too slow. If the moisture level is too high, the pack becomes
anaerobic (lacking in oxygen). In another words, the rate of microbial decomposition will
be slow and again, composting and heat generation will also be too slow (Bewley and
Taraba, 2009).
22 
 
 By stirring the pack, both moisture and manure can be mixed into the composting
bedded-pack and will aerate the tilled layer. Good ventilation helps dry the freshly turned
bedded-pack surface to retard bacterial growth on the surface and keep cows cleaner since
dry bedding does not stick to the teat or leg surfaces. Compost barn owners from
Minnesota recommend adding fresh bedding when the bedded pack becomes moist enough
to stick to cows (Janni et al., 2005).
A wet bedded pack is more vulnerable to compaction. The compaction may be caused
by the machinery that is used for stirring. Lower ground pressure tillage machinery should
help reduce media compaction. Compaction in turn reduces air flow and oxygen in the
bedded pack and thereby the pack can become anaerobic more rapidly. An anaerobic pack
shows lower temperatures which leads to reduced pathogen kill, and more pest and odor
problems (Janni et al., 2005; Graves, 1999).

2.2.5. Manure and effluent management

Management of animal manure and effluent is an integral part of a total dairy producer
management plan. Changes in agricultural production have led to fewer producers, larger
average herd sizes, purchase of feed from distant areas, and a much higher concentration of
animals and production facilities on less land. In some areas livestock operations have
become concentrated, and non-farm residents have located amongst the farms. These
features increase the challenge for implementing good manure management plans (MWPS-
7, 2000).
A manure management system includes: collection, transfer, storage, possible
treatment, land application, nutrient utilization, and sampling.
Intensive, more specialized, concentrated animal production systems have altered the
nutrient cycle and balance between crop production and animal production. Independent of
operation size, a proper manure management system and plan are essential. A complete
manure management system has the following goals: a) maintain good animal health and
milk quality by providing sanitary facilities; b) avoid pollution of soil, groundwater or
surface water; c) reduce odors and dust; d) control insect and other pest reproduction; e)
and others (SWIM, 2011).
Failure to provide adequate manure collection, handling, and storage facilities in
conjunction with adequate land area for proper application and utilization of manure
nutrients could adversely affect air, water, and land resources.
23 
 
 CBP barns owners have some strategies for managing the effluent and manure. Some
producers have added a little of the composted pack to the feed alley to absorb moisture
(Barberg et al., 2007). The concrete feed alley is scraped at least twice a day. The collected
manure and bedding is stored in an approved manure storage unit until land applied
according to a manure management plan.
Bedded packs should be completely cleaned out each fall and applied to cropland as
part of a manure management plan (Gay, 2009). CBP barns owners in Minnesota have been
removing a portion of the pack for land application in Spring, and doing a complete
cleanout in the Fall. While this works well for manure management, it presents concerns
for nutrient management, since manure applied in Fall may lose nitrogen over the winter.
The rationale for Fall cleanout is that the CBP barns owners want maximum manure
storage over the winter (Barberg et al., 2007). When the material is removed from the barn,
it can either be land spread as is, or composted further to produce a product suitable for
horticulture, organic agriculture, and home gardens. As it comes out of the barn, the bedded
pack material has a high C:N ratio, is high in soluble salts, and has pathogens in it, but it
works well for meeting the nutrient requirements of field crops (Petzen et al., 2009).
Compost used as a soil conditioner has the potential to result in higher crop yields and
reduced erosion (NRAES-54, 1992).
Care must be taken to avoid disturbing the clay base during clean out and pack stirring
when the bedded pack is less than 0.30 m deep. Some users recommend starting with 0.45
to 0.60 m of sawdust to avoid problems. The composting bedded pack is typically cleaned
out and land applied in the fall after corn silage is harvested. Fall clean out allows time for
a new pack to accumulate and begin composting before cold weather sets in. Some
operators remove some of the bedded pack in the spring before fields are planted to make
sure there is sufficient space for pack accumulation during the summer when cropland is
not available for manure application. However, if a compost bed is active, particularly
during the non-winter months, bed depth is reduced by microbial activity, but it can also
occur due to less bedding used. Going into winter, there may not be a great depth and
therefore adequate storage is present to get through winter. CBP barns require equipment to
handle manure as both a solid (the pack) and a slurry (the feed alley manure). This
increases a producer's investment in equipment (Janni et al., 2007).

24 
 
2.3. Physical properties of the bedding material
The physical properties of compost bedding materials play an important role in every
stage of compost production as well as in the handling and utilization of the end product.
From the mixing of various feedstock and process monitoring and maintenance, to the
packaging and shipping of the final product, parameters such as porosity, bulk density,
particle density, moisture content, water holding capacity and particle size affect the
optimum composting environment and the design of machinery used in the system to
promote the aeration of the compost bedding material.
Although compost can vary and must be carefully tested to determine their exact
nature, numerous laboratory trials with different products have successfully quantified the
beneficial effects (CUC, 2003).

2.3.1. Porosity
Transport of mass and heat in a compost layers is significantly affected by the geometry
and size distribution of the pores and the composition of the composting material.
Therefore it is at least necessary to have information about the spatial distribution of
porosity. Porosity can be calculated from dry matter content, true density of dry matter and
total bulk density. Dry matter content and true density of dry matter are easily measured in
the laboratory (Ginkel et al., 1999).
Porosity is the ratio of the pore space to the total volume occupied by the material and
the ideal bed porosity is one that provides the greatest surface area for microbial activity
and the least resistance to airflow through the compost layer. It is determined by the
particle size, the size gradation of the materials, and the continuity of the air spaces. Larger
particles and more uniform particles increase porosity (NRAES-54, 1992).
Porosity greater than 50% causes the compost layer to remain at a low temperature
because energy loss exceeds heat produced. Too little porosity leads to anaerobic
conditions and odour generation from too little oxygen occupying the reduced void space.
The percentage air-filled pore space of composting layer should be in the range of 35% to
50% (Bernal et al., 2009).

2.3.2. Bulk density

The bulk density of compost is a measure of the mass of material within a given
volume and it is important in the determination of initial compost mixtures. The bulk
25 
 
density determines how much material can be placed at a certain site (Agnew et al., 2003).
The density of compost also influences the mechanical properties such as strength,
porosity, and ease of compaction (Agnew and Leonard, 2003). Therefore, knowing the bulk
density of the material throughout the compost layer is important for aeration, handling,
and storage requirements.
However, distinction is made between time dependent and independent processes to
describe bulk density distribution. The time independent process results from the
compressibility of composting material. Compressibility is defined as the reciprocal of the
resistance of the material against mechanical deformation, or more precisely, as the ratio
between deformation and stress. Due to this property the bulk density is a function of the
local pressure and therefore depends on the position within in the compost bed (Ginkel et
al., 1999).
The time dependent process is defined as the increase of mechanical deformation as a
function of time under constant pressure conditions (Ginkel et al., 1999). When a
composting material is spread over the pack area, the initial bulk density distribution is
only determined by the compressibility of the composting material. However, during the
process, bulk density and porosity are also influenced by the combined effects of
subsidence, loss of organic matter due to biological degradation processes and change of
water content due to transport processes.
The effect of compaction pressure on bulk density is also studied. It is observed that the
effect of compaction pressure on bulk density for larger particles is higher in comparison
with smaller particles, because smaller particles are closely packed leaving less space
among the particles (Mozammel et al., 2006).
Parameters that influence the bedding material performance are moisture content, bulk
density and porosity (Wright and Inglis, 2002). This is relative to green materials that are
high in water, which would not be suitable for bedding. Dried straw wood and other
forages would be needed. Thus, increasing bulk density, moisture content and depth of the
pack can result in decreased permeability, especially for oxygen. If the moisture content in
the pack increases, particles get wet resulting in reduced strength of the media (Ahn et al.,
2009). This can cause anaerobic condition in the pack which leads to odor, low
temperatures and emission of nitrogen compounds as well as methane. Materials that
absorb too much water or urine are not suitable such as bedding material as this high

26 
 
moisture content can rupture the cell walls which results in additional free water. In
addition wet bedding also contributes to dirtier cows.
Soybean straw and chipped pallet material worked well in combination with sawdust in
the compost bedded pack system. Both materials provided the absorbency and bulk density
necessary to meet cow comfort and the porosity required to promote an aerobic compost
process (Petzen et al., 2009).

2.3.3. Particle density

The accurate quantification of particle density, as a component of bulk density, is
vitally important for gravimetric compost measurements. For other media such as soil,
particle density is often assumed to be 2.65 g cm-3, the density of quartz (Brady and Weil,
2002). However, compost particle density is more inconsistent due to variation in source
materials.
Concisely stated, the method employs the submersion of a known mass of compost in
methanol to determine the volume of the compost solids. While methanol is used because
of its low density (0.66 g cm-3), the premise of the method remains valid for many other
low density liquids. The use of a low density liquid causes particles that are normally
buoyant in water to sink, facilitating an accurate measurement of the compost solids
volume (Weindorf and Wittie, 2003).

2.3.4. Moisture content

It is the fraction or percentage of a substance comprised of water. The moisture content
on a dry basis is the wet weight minus dry weight divided by the dry weight minus the
container weight. To obtain the moisture content in percent, multiply this ratio by 100
(Mukhtar et al., 2004). The general procedure involves weighing the wet sample and then
drying the sample for 24 hours at temperatures of 105°C. The moisture content on a wet
basis can be calculate using equation 2.1.

Wet weight - Dry weight

Moisture content (%) = 100 (2.1)
Wet weight - Container weight

Methods for determining moisture content on the farm differ in the way that the sample
is dried. Three common methods include air drying, conventional oven drying and
27 
 
microwave oven drying. Although the results produced by these methods are less accurate
than laboratory procedures, they are satisfactory for almost all composting situations
(NRAES-54, 1992).
The moisture content of compost affects the bulk density. Moisture content is also
relevant because it affects compost handling. Compost which is dry can be dusty and
irritating to the cows and workers, while compost which is wet can become heavy, making
its application more difficult and transport more expensive.
The ideal moisture content for composting must therefore be a compromise between
achieving adequate moisture for the microorganisms to survive and adequate oxygen flow
to maintain aerobic conditions. The moisture content for composting is generally
recommended to be in the range of 40 to 60% (Shane et al., 2010; NRAES-54, 1992), and
can be tested accurately using analytical equipment or approximated using a hand-squeeze
method. In the hand-squeeze method, a handful of compost material is squeezed firmly
several times to form a ball. If the ball crumbles or breaks into fragments, the moisture
content is much less than 40%. If it remains intact after being gently bounced 3-4 times, the
moisture content is nearly 50%. If the ball texture is slimy water droplets drip from or
appear on the surface of squeezed bedding, the pack is too wet, the moisture content is
much higher than 60% (Mukhtar et al., 2004). When the pack is working well, the bedding
material will appear loose and fluffy, not compacted and chunky (Bewley and Taraba,
2009). Pack moisture content in the top layer should ideally be monitored weekly to decide
when clean bedding needs to be added to the pack (Shane et al., 2010) as opposed to when
the cows are dirty.

2.3.5. Water Holding Capacity

Water holding capacity (WHC) is the ability of a given volume of compost to hold
water under one atmosphere of pressure. In another words, WHC is the ability of compost
to retain moisture against drainage due to gravity. Moisture (manure and urine) is held in
the spaces, or pores, between compost material particles and thin films around these
particles. Organic matter and particle size affect the compost ability to retain moisture.
WHC is measured as a percent of dry weight.
The WHC provides information on the amount of water that can be absorbed by the
bedding material. High WHC allows the media to absorb water into the bedding. However,

28 
 
depending on the cell structure of the media, high water content in the cell can sometimes
lead to rupture of the cell which results in the media releasing free water (Vokey, 2008).
This limits the media’s ability to function as a bedding material for longer periods of time.
Some research have observed that WHC increased with decrease in the particle size of the
compost media (Changirath et al., 2011). For this reason, one of the functions of the
bedding material in the CBP barns is to retain moisture. Therefore, if the compost material
has a high WHC, the ability of dairy producer to dry in the pack can be lower.

2.3.6. Particle size

Particle size and distribution are critical for balancing the surface area for growth of
microorganisms and the maintenance of adequate porosity for aeration. The larger the
particle size, the lower the surface area to mass ratio. So, compost with large particles does
not decompose adequately because the interior of the particles has difficult accessibility for
oxygen for aerobic microorganisms (Bernal et al., 1993). However, particles which are too
small can compact the mass, reducing the porosity and airflow within the compost layer
will be restricted. A mix of particle sizes creates the most porous layer, see Figure 2.5.

Figure 2.5 - Porosity affecting the aeration: (a) Loosely packed - well structured, (b)
Loosely packed - uniform particle size, (c) Tightly packed - uniform particle
size, and (d) Tightly packed - mixed particle sizes. (CUC, 2003).

Media particle size distribution is an important factor influencing the bedding moisture
content, compaction and bulk density. Sawdust serves as a good carbon source and when
fine materials like sawdust are used as a bedding media. They need to be turned frequently
in the pack. It has also been suggested that the amount of fine particles in the bedding has

29 
 
an effect on the bacterial population on the teat ends. Factors such as particle size may be
more important than simply bacterial counts in the used bedding (CWMI, 2006). The finer
the media, the more time is required to clean the udder before milking.
Microbial activity generally occurs on the surface of the organic particles. Therefore,
decreasing particle size, through its effect of increasing surface area, will encourage
microbial activity and increase the rate of decomposition. On the other hand, when
particles are too small and compact, air circulation through the compost layer is inhibited.
This decreases O2 available to microorganisms within the compost and ultimately
decreases the rate of microbial activity (NRAES-54, 1992).
Particle size also affects the availability of carbon and nitrogen. Large wood chips in
bedding, for example, provide a good bulking agent that helps to ensure aeration through
the compost layer, but they provide less available carbon per mass than they would have in
the form of wood shavings or sawdust.
Handreck (1983) studied the particle size and physical properties of compost media and
concluded that the fraction smaller than 0.5 mm, and in particular between 0.1 and 0.25
mm, has the highest influence on porosity and water retention.

2.4. Thermal property of the bedding material

Compost thermal properties are required in many areas of agriculture engineering,
agronomy, and animal science. In recent years, considerable effort has gone into
developing techniques to determine these properties (Abu-Hamdeh, 2003).
Thermal properties of compost bulking materials affect temperature and biodegradation
during the composting process. Well determined thermal properties of compost feedstock
will therefore contribute to practical thermodynamic approaches (Ahn et al., 2006).
Thermal conductivity, specific heat capacity, and thermal diffusivity are the three
important thermal properties regarding heat transfer analysis. These three thermal
properties can be measured by several methods (Ahn et al., 2009).
The thermal conductivity is the property that describes the heat transfer capacity within
the compost material. The specific heat capacity is the amount of energy required to raise
the compost material one degree Celsius. Thermal diffusivity has been used as an indicator
for the utility of compost materials as an insulator or heat conductor. In other words,
thermal diffusivity is the measure of thermal inertia.

30 
 
 Although thermal properties are very important in composting, information on their
values for various compost materials is lacking. In general, it is difficult to determine the
thermal properties of moist materials because forced heating during the measurements
causes internal liquid and/or gas convection. This often results in overestimating the
thermal properties. The thermal probe method is the most attractive method, because it uses
relatively simple equipment to determine the thermal properties for moist materials
(Iwabuchi et al., 1999).

2.4.1. Thermal conductivity

The thermal conductivity coefficient of a compound medium is often a very complex
quantity. It is dependent not only on the components themselves, but also on structure,
composition, density, porosity, water content and temperature (Van Ginkel et al., 1999).
Certainly in a compost bedding layer, where all these parameters are changing during the
compost management and process, the only way to obtain reliable values is to measure this
quantity. Thermal conductivity can be measured by the steady and non-steady state
(transient heat dissipation) methods (Mohsenin, 1980).
Transient line heat source methods have been used several years to measure thermal
conductivity of porous materials, like bedding compost. Typically a probe for this
measurement consists of a needle with a heater and temperature sensor inside. A current is
passed through the heater and the temperature of the probe is monitored over time. An
analysis of the probe temperature is used to determine thermal conductivity. More recently
the heater and temperature sensor have been placed in separate needles. In the dual probe
the analysis of the temperature versus time relationship for the separated probes yields
information on diffusivity and heat capacity as well as conductivity (Bristow et al., 1994).

2.4.2. Specific heat capacity

The specific heat capacity can be defined as the amount of heat energy required raising
the temperature by one degree Celsius per unit mass (Incropera and DeWitt, 2001). Several
researchers studied the behavior of thermal properties with the variation of water content of
various biological materials and observed that there is a relationship in which the increase
of water content causes high values of specific heat (Ahn et al., 2009).
Factors influencing specific heat capacity that can be managed externally include
moisture content and compost density. Moisture content plays a major role in specific heat
31 
 
capacity but is the most difficult to manage (Abu-Hamdeh, 2003). Compost management
affects specific heat capacity by the machinery that is used for stirring (Janni et al., 2005).
Because of this, the bulk density increases and the porosity of a compost bedding
decreases. This in turn have a significant effect on specific heat capacity.

2.4.3. Thermal diffusivity

Several studies have reported two of the thermal properties and obtained the thermal
diffusivity using the following relationship  (Haug, 1980; Briaud and Chaouch, 1997;
Bristow, 1998; Abu-Hamdeh, 2001):

k
α= (2.2)
ρ  Cp

where: α is the thermal diffusivity (m2/s), k is the thermal conductivity (W/m °C), ρ is the
bulk density (kg m-3) and CP is the specific heat capacity (J/kg °C).

Thermal diffusivity, like thermal conductivity, is a property of the substance. The term
k · ρ-1 · Cp-1 appears in many heat conduction problems and is related to the diffusion
coefficient used in mass transfer analysis or modeling.

2.5. Chemical properties of bedding material

Since a variety of organic materials can be used to make bedding composts (i.e., rice
husks, coffee husks, bagasse, paper, peanut shell), it becomes very important to know their
chemical characteristics, in order to obtain the required chemical characteristics for growth
media (Zoes et al., 2001).
Carbon (C), nitrogen (N), phosphorus (P) and potassium (K) are the primary nutrients
required by the microorganisms involved in composting. Nitrogen, phosphorus, and
potassium are also the primary nutrients for plants; so their concentrations also influence
the value of the compost.
Many organic materials, including manures and organic materials, contain ample
quantities of nutrients (Table 2.2). Composts that are derived primarily from wood by-
products have high carbon to nitrogen ratios unless additional nitrogen is added during the
composting process. Excessive or insufficient carbon or nitrogen is most likely to affect the
32 
 
composting process. Microorganisms use carbon for both energy and growth while
nitrogen is essential for essential protein production to the metabolism and reproduction. In
general, biological organisms need about 25 times more carbon than nitrogen (NRAES-54,
1992). Therefore, it is important to provide carbon and nitrogen in appropriate proportions.
The ratio of carbon to nitrogen is referred to as the C:N ratio.

Table 2.2 - Common feedstock and their characteristics

Feedstock Moisture content (%) C:N Bulk density
High in Carbon
Hay 8 - 10 15 - 30 -
Corn stalks 12 60 - 70 32
Straw 5 - 20 40 - 150 50 - 400
Sawdust 20 - 60 200 - 700 350 - 450
Wood chips - 100 - 500 -
High in Nitrogen
Dairy manure 80 5 - 25 1400
Grass clippings - 15 - 25
Source: Adapted from NRAES-54 (1992).

A balanced C:N ratio usually ensures that the other required nutrients are present in
adequate amounts. To calculate the carbon content given C:N and percent nitrogen, solve:

C(%) = N(%) · C:N (2.3)

Raw materials blended to provide a C:N ratio of 25:1 to 30:1 are ideal for active
composting, although initial C:N ratios from 20:l up to 40:1 consistently give good
composting results. For many applications, C:N ratios of even 50:1 and higher are
acceptable. With C:N ratios below 20:1 the available carbon is fully utilized without
stabilizing all of the nitrogen. The excess nitrogen may then be lost to the atmosphere as
ammonia or nitrous oxide and odor can become a problem. Mixes of materials with C:N
ratios higher than 40:1 require longer composting times for the microorganisms to use the
excess carbon (NRAES-54, 1992).
Compost bedded pack manure is an excellent nutrient source to use with zone-till and
no-till cropping systems. This is because more of the nitrogen is in its stable organic form,
compared to liquid or fresh (Petzen et al., 2009).

33 
 
2.6. Biological properties of compost bedding material
One of the aspects of composting that makes it an attractive alternative to the direct
application of untreated manure is the high degree of pathogen destruction that is possible
with a well-managed composting operation. The pathogen content of the compost is
important because improperly treated compost can be a source of pathogens to the
environment and, as such, a threat to humans and animals.

2.6.1. Predominant groups

Pathogenic microorganisms that may be in compost include bacteria, viruses, fungi,
and parasites. Although parasites and viruses cannot reproduce apart from their host, they
can often survive for extended periods. If they are not killed during the composting
process, they can survive until the compost is land applied. At that time, they may infect a
new host (NRCS, 2010). In the compost process, these species will thrive or fail according
to their abilities to find a niche in the compost ecosystem (Sales, 2008). Typical species are
shown in Table 2.3.

Table 2.3 - Some species of microorganisms found in composting.

Actinomycetes Algae
Actinoplanes sp. Chlorococcum humicola
M. vulgaris Dactylococcus (bicandatus)
Micromonospora parva Euglena mutabilis
Nocardia brasiliensis Hormidium (nitens)
Pseudomocardia Kentrosphoera sp.
S. recius Microcoleus vaginatus
Bacteria Fungi
Aerobacter (aerogenes) Abridia (ramosa)
B. cereus Aspergillus flavus
B. mycoides Geotrichum candidum
B. stearothermophilus Hanisenuia sp.
Bacillus megatherium Mucor pusillus
Cellumonas folio Mucor racemosus
Chondrococcus exiguus Penicillium digitatum
Protozoans
Cercomonas (crassicauda)
Chilomonas (paramecium)
Cyathomonas (truncate)
Lycogala epidendrum
Adapted from Epstein (1997).
34 
 
2.6.2. Bacteria
Bacteria are the smallest living organisms and the most numerous in compost. They
make up 80 to 90% of the billions of microorganisms typically found in a gram of compost.
Bacteria are responsible for most of the decomposition and heat generation in compost.
They are the most nutritionally diverse group of compost organisms, using a broad range of
enzymes to chemically break down a variety of organic materials (CWMI, 2006).
Among the bacteria, the actinomycetes, and the bacteria (gram negative and positive)
are most important in the context of composting (Table 2.4), the former as allergens and
opportunistic pathogens, and the latter as indicators of an insufficient thermohygienization
(fecal coliforms) and as producers of endotoxins (Beffa, 2002).

Table 2.4 - The following groups of human and animal pathogens as having the most
importance in composting environments.
Bacteria Fungi Viruses
Coliform Aspergillus fumigatus Coxsackie-B-Virus
Enterobacteraceae Penicillium Echo-Virus
E. coli
Pseudomonas
Staphylococci
Streptococci
Staphylococcal species
Streptococcal species
Source: Beffa (2002).

Bacterial growth over time can be graphed as cell number versus time. This is called a
growth curve. The cell number is plotted as the log of the cell number, since it is an
exponential function. Regardless of the generation time, in a growing culture, the plot of
the log of cell number versus time gives a characteristic curve (MBS, 1999). This curve
typically has four distinct phases: lag phase, exponential phase, stationary phase, and death
phase, see Figure 2.6.
Many factors affect the generation time of the organism: temperature, pH, oxygen, salt
concentration and nutrients are some of the common factors that may change in the normal
environment of bacteria. Due to difficulties in tracking and quantifying the relationships
between physical factors and bacterial populations, it has not been possible to fully
understand the ecological processes important in composting. Identifying these

35 
 
relationships has been complicated by the presence of an uncontrolled bacterial inoculum
that varies between operations due to variations in material handling, air conditioning, and
substrate composition (Schloss et al., 2003). Several previous composting studies, though,
have attempted to quantify bacterial populations using different techniques (Nakasaki et al.,
1985; Strom, 1985).

Figure 2.6 - Bacterial growth curve and the four phases of growth versus time (Adapted
from MBS, 1999).

2.7. Mathematical modeling

Mathematical modeling is a method of research that investigates, through
mathematical problems, phenomena or situations from different areas such as physics,
chemistry, biology, engineering, and others, consisting in an area of research that is
essentially interdisciplinary, to make predictions about the situation or problem
investigated. This set of relations is called a mathematical model of the problem studied.
The dynamic process used to create and validate such models is called mathematical
modeling (Bassanezzi, 2002).
Mathematical modeling has a role to evaluate the interaction between several factors
allowing a more detailed study in order to predict the processes of interest, and the ideal is
to formulate a model capable of describing the original study, without the difficulties
encountered in observations and measurements made during examination of the problem.
Moreover, the mathematical model is a simplification of the system being studied, and
cannot predict all the variables present. The resulting prediction may not be accurate or
reliable. When using f mathematical models, it is necessary to determine the parameters

36 
 
through empirical experimentation implying the need for several samples (Pereira and
Machado, 1987).
Over many years agricultural buildings, such as dairy barns, have become tailored for
specialized production The need to maintain control over the indoor climate has become
ever more important so that the requirements of the production system can be continuously
met. In essence it is necessary to understand the interaction of all climatic variables,
alongside their contribution to the occupants heat exchange with its surroundings in order
to predict the micro-climate (Norton et al., 2007).
Therefore, the thermal environment of a agricultural building is intricately linked with
ambient conditions. Optimizing and controlling ventilation systems are not elementary
problems and may require extensive experimentation (Boulard et al., 2002a; Zhang et al.,
2000). Sophisticated techniques that allow efficient and accurate description of the
dominant phenomena are often required. The mathematical modeling of the interaction
between the indoor production system, the ventilation system and the outdoor environment
is thus a necessary requirement. The clear advantages of developing mathematical models
for agricultural buildings are as follows (Schauberger et al., 2000):
a) They can quantify the essential needs of the occupant in all parts of ventilated
buildings and consequently contribute to the development of optimization strategies aimed
at increasing production performance.
b) They can be incorporated as part of indoor climate control, as well as to check the
design values of the ventilation system.
c) They can combine the spatial and temporal dependent parameters that affect release
of all pollutants within the building to efficiently calculate time-varying concentrations
within, and emissions from production systems.
d) They can quantify different housing systems with alternative manure management
techniques to achieve a fully integrated control of pollution.

Therefore, agricultural production systems, which are amenable to description through

the physical laws of fluid motion, heat and mass transfer processes, can be effectively
modeled so that the benefits of pre-design evaluation, design and optimization strategies
can be realized (Norton et al., 2007).
The modeling of ventilation flow in agriculture building can be can be performed
using the Navier-Stokes equations (Norton et al., 2007). The Navier–Stokes equations
37 
 
represent the dynamic behavior of a flow regime, both spatially and temporally. These
equations constitute a micro-model of fluid motion, and require time consuming iterative
techniques to solve them. Thus, the efficient quantification of ventilation in agricultural
buildings can be carried with a macro-model based on Bernoulli’s equation. This requires
some qualifying assumptions so that the model matches the spatial domain of interest, such
as the flow is not viscous; steady, not rotational; with hydro-static and gravity as the only
acting external force (Roy et al., 2002). Consequently, much information on the flow is
lost, including details on the spatial distribution of the airflow patterns, which can lead to
inaccuracies in the calculated airflow variables (Boulard et al., 2002b).
According to Mueller and Krause (2007), when sizing ventilation systems there are
different methods which may be used to obtain an overall view of the heat transfer and
mass transport phenomena inside the animal production facility, including numerical
simulation of flow, physical models and measurements on a real scale.
a) Numerical simulation of flow: from this simulation it is possible to study the
behavior of fluids in motion. Therefore, it has become very interesting to study these
phenomena with the advance of faster computer processors and more efficient software.
Fluid mechanics include, for example, study of either laminar or turbulent fluid and gas
flow in animal installations. This type of study may present important practical applications
in the diverse areas of engineering.
b) Physical models: investigations with physical models offer greater support to the
numerical simulation of mass flux. The ventilation system as well as the flux of air volume
may be altered by a simple formula, i.e., the reconstruction of entire elements may be
performed quite simply in such models.
c) Another form of evaluating the dispersion behavior of gases coming from animal
installations is by the utilization of the wind tunnel boundary limit, where in many cases
these are equipped with air velocity sensors and gas analyzers to measure the variation in
concentration.
d) Real scale measurements: are necessary to confirm the numerical and physical
models in real conditions. A specific problem is the determination of volumetric air flux in
naturally ventilated animal installations (Mueller and Krause, 2007).

38 
 
 Various studies involving mathematical modeling and computational simulations have
been developed during experimentation to improve understanding of the processes
involved in the transfer of heat and mass between animals and their thermal environment.

2.8. Computational Fluid Dynamics (CFD)

Computational fluid dynamics (CFD) can be defined as a numerical resolution based
on computational loops and the equations of mass, momentum and energy conservation for
physical systems (Miranda, 2009). The method works by solving the equations of
transporting mass and energy and momentum through a process number called difference
numerical methods. Figure 2.7 illustrates the initial work of the numerical method, which is
discretize the domain in a number of non-coincident sub-domains forming a grid of small
elements (or volume control) to which a discrete solution can be obtained. For this, some
transformations are performed in differential transport equations, defined continuously in
the domain, in a system of algebraic equations for the discretized Domain (Maliska, 1995).

Figure 2.7 - The initial work of the numerical method: discretize the continuous domain
(Maliska, 1995).

The CFD methodology was initially and widely used in aerospace industry,
automotive, naval, and nuclear. Recently it has been applied in agricultural engineering.
The availability of high-performance, low cost computers and with the emergence of CFD
programs with friendly interfaces, has allowed great advances in the use of this technique
since the 90's. Moreover, the versatility and generality of the numerical methods for
simulation of engineering problems, and the relative simplicity of application of these
techniques are other factors motivating the use of CFD.
This technique provides precise solutions as well as visualizations of velocity,
pressures, temperature, turbulence and gas concentrations in the entire space of the animal

39 
 
installation without various physical scale test models. Results allow that engineers and
operators to develop new ventilation strategies to meet the standards for air quality (Lee
and Short, 2000).
Modeling in CFD is important to predict mass transport in different types of animal
structures, especially when field measurement of the variables related to the model is
difficult. In the case of modeling gas emission that eliminate from inside the installations,
for example NH3, the computational models are based on the NH3 transport process and the
transport mechanism to the atmosphere (Bjerg et al., 2002).
Currently, sophisticated commercial CFD programs like CFX (ANSYS CFX 13.0,
2011), can run on personal computers in the workplace and even at home. Also the costs of
the required equipment acquisition are getting smaller. Thus, it becomes easier to use
numerical techniques for solving engineering problems.,
Undertaking a CFD study demands the use of four predefined environments within the
software, with each environment representing an equally important section of the modeling
process. The following provides an introduction to the different modules comprising a CFD
software package: a) generating geometry and mesh; b) pre-processor; c) solver manager;
and d) post-processor.

a) Geometry and mesh generator:

It consist of insertion of problem into CFD code. The embedded information involves:
i) Definition of geometry and solution domain: This step can be performed in a
secondary program; CAD (Computer Assisted Drawing) as SOLIDWORKS® or
Rhinoceros® and then imported by the CFD code, operation commonly performed when
working with CFX.
ii) Meshing: consists of discretization of a continuous domain into a set of discrete
sub-domains, usually called elements.
Approximately 50% of the time used in a CFD project is to define the geometry and
mesh generation (Malalasekera and Versteeg, 1995). In CFX, DESIGNMODELER module
is used to generate or import geometry and CFX-Mesh is used to generate the mesh.

b) Pre-processing:
The pre-processor of CFD software holds all the mathematical statements attributable
to the potential success of a modeling exercise and therefore embodies the most important
40 
 
phase of model definition. The main tasks facing a user in the pre-processing environment
include problem consideration, geometry creation or import, mesh development, physical
property set-up, and the implementation of solving techniques and parameters.

c) Solver manager:
The solver environment, within all commercial CFD software packages, organizes the
mathematical input from the pre-processor into numerical arrays and solves them by
iterative methods (Norton et al., 2007). Iterative methods are commonly used to solve a
whole set of discretized equations so that they may be applied to a single dependent
variable. Segregated solving techniques, such as Semi-Implicit Method for Pressure-Linked
Equations (SIMPLE) used by most commercial CFD software packages, determine the
pressure field indirectly by closing the discretized momentum equations with the continuity
equations in a sequential manner (Patankar and Spalding, 1972). However, as the number
of cells increase the elliptic nature of the pressure field becomes more profound and the
convergence rate is inhibited. This has led to the development of multigrid techniques that
compute velocity and pressure corrections in a simultaneous fashion, thus enhancing
convergence rates (Ferry, 2002).

d) Post-processor:
The post-processing environment allows the user to visualize and investigates the
resulting field solution. Contour, vector and line plots enhance interpretation of results and
are being progressively fortified in commercial software packages. Some packages also
allow the export of field data to external modeling programs so that it can be processed
further (Norton et al., 2007). In CFX, the CFX-POST module is used for post-processing.
Thus, the complete analysis in CFD is composed of pre-processing, processing and
post-processing, which includes the procedures and phenomena to adequately analyze the
configuration of the problem with the solution, monitoring and analysis of results at the end
of simulation (Rocha, 2012).

2.8.1. ANSYS
ANSYS is a set of commercial software for numerical solutions in fluid dynamics
modeling and other physical phenomena with a multidisciplinary approach and the
possibility of integration with other types of physical simulation technologies, which the

41 
 
main tools for simulation and analysis of fluids are known by FLUENT and CFX (Xia and
Sun, 2002).

2.8.2. Governing equations

The governing equations of fluid flow and heat transfer can be considered as
mathematical formulations of the laws of conservation which regulate all fluid flows, heat
transfer and associated phenomena. These laws of conservation describe the rate of
variation of a desired fluid property as a function of external forces.

a) Conservation of mass:
The law of conservation of mass states that mass can neither be created or destroyed.
This is the basis for the continuity equation. It can be represented in Cartesian coordinates
by the following equation (Rocha, 2012; Hirsch, 2007).

ρ   ρ  u x    ρ  u y    ρ  u z 
= + + =0 (2.4)
t x y z

where,
t is the time (s);
ρ is the density (kg m-3);
x, y, z are the coordinate direction;
ux, uy, uz are the velocity in the directions x, y, z respectively.

b) Conservation of momentum:
The sum of the external forces acting on the fluid particle is equal to its rate of change
of linear momentum (Rocha, 2012).

u ∙ u ∙ ∙ ν ∙ ν∙ (2.5)

where,
ν is the kinematic viscosity (m2 s-1);
P is the pressure (Pa).
42 
 
 The equation for the moment can be described by the following expression.

∂u ∂u ∂P ∂2 u ∂2 u
ρ ∙ ui ρ ∙ uj ‐ ∙ μ ν∙ (2.6)
∂i ∂j ∂i ∂i2 ∂j2
Inertia Pressure Friction

c) Conservation of energy:
The equation for the conservation of the energy is derived from considerations of the
first law of thermodynamics. The rate of change of energy of a fluid particle is equal to the
heat addition and the work done on the particle (Rocha, 2012).

∂T ∂T ∂2 T ∂2 T
ui uj α ν∙ Ψ (2.7)
∂i ∂j ∂i2 ∂j2

where,
T is the temperature (K);
Ψ is the viscous dissipation (K s-1);
α is the thermal diffusivity (m2 s-1).

2.8.3. Method solution

Two methodologies are used to solve the system of equations: segregated solver
solution and coupled solution. The method employs a numerical strategy solution, where
the equations of conservation of momentum are solved with a estimated pressure, and then
a correction equation for the pressure is applied. Because of the nature of this methodology
of correction estimation, a large number of iterations, and a very careful selection of the
significant parameters affecting the convergence of a numerical parameters are necessary,
in the case of steady state (Santos, 2008).
The coupled method, used by CFX, solves the hydrodynamic equations as a single
system. Figure 2.8 illustrates the general procedure of solution. The solution of equations
for each set consists of two intensive numerical operations, for each step in time.

43 
 
Figure 2.8 - Overall solution strategy of CFX (ANSYS CFX 13.0, 2011).

The iteration time is controlled by a step of physical time, global, or a false time step,
locally configured to advance the solution for a simulation time in continuous operation. If
the model is steady state, there is only one interaction of linearization step of virtual time.
The solution's performance is improved with the use of a technique called Multigrid
(Brandt, 1977), a process to increase the size of the mesh elements for later iterations. This
results in a reduction of virtual quality of the spacing of the mesh during the iteration, and
return to the original mesh obtaining an accurate solution. The use of this technique
significantly improves the convergence rate.
The CFX uses a particular implementation of the algebraic Multigrid called correction
additive (Hutchinson and Raithby, 1986). This approach is ideally suited to the ANSYS
44 
 
CFX-Solver implementation, because it takes advantage of the fact that the discrete
equations are representative of the balance of conserved quantities over a control volume.
The coarse mesh equations can be created by merging the original control volumes to
create larger ones as shown in Figure 2.9. Figure 2.9 shows the merged coarse control
volume meshes to be regular, but in general their shape becomes very irregular.

Figure 2.9 - Scheme of an algebraic Multigrid method.

2.8.4. Simulation in CFD for animal facilities

The first studies using livestock installations in CFD were performed by Choi et al.
(1988), in which a simple geometry was used to represent the animal installation, but
results showed little precision. Small advances in the development of real solutions were
made in the next year. Hoff et al. (1992) utilized CFD to model air flow in a rectangular
shaped installation, in which heat transfer was modeled using a heated floor. Harral and
Boon (1997) later simulated isothermal flow patterns in a 3D geometric representation of a
cattle confinement. In following years, more studies were developed for validation of
animal installations with a focus on the creation of more precise CFD models (Zhang et al.,
2000). Only in recent years have CFD models used for simulation of heat and mass transfer

45 
 
in a geometric representation become reliable in animal production installations (Van
Wagenberg et al., 2004). Therefore, more detailed conditions are provided of all
phenomena that occur when CFD is employed, and dynamic models, together with
estimates, are provided.

a) Birds:
Among environmental factors, thermal factors represented by air temperature, relative
humidity, thermal radiation, and air movement are those which most directly affect the bird
since they are warm-blooded animals. These environmental variables may have either
positive of negative effects on bird production. High temperature and high relative
humidities inside aviculture installations are factors which limit the achievement of optimal
productivity (Damasceno et al., 2010a). Therefore, the need to maintain the thermal
environment within an optimal production scale is evident. However, there are few studies
with CFD which utilize a mass balance and heat transfer inside aviculture facilities.
Worley and Manbeck (1995) utilized a CFD model to analyze the distribution of
contaminants in a highly turbulent flow inside an aviculture installation under different
roofing configurations, whose objective was to obtain a better configuration of air
circulation to eliminate particulate material.
Pawar et al. (2007) analyzed the distribution of virus particles which generate diseases
in aviculture installations, and for this considered an analogy between the diffusion of
ammonia gas (NH3) and dispersion of virus particles.
Damasceno et al. (2010b) developed and validated a CFD model to analyze the
temperature and air velocity distribution inside a common commercial heater used in
Brazilian poultry production. These authors concluded that the CFD models may be used
for real time prediction of thermal behavior and velocity distribution (inside the
combustion chamber for heating), aiding in project improvements.
Tinôco et al. (2010) used CFD to simulate an aviculture installation with mechanical
ventilation associated with an evaporative cooling system. In this study the authors
developed a CFD model to obtain a better understanding of the temperature and air velocity
distribution inside the aviculture installation. The obtained results showed a good statistical
correlation with experimental data, which permits its use to address future modifications to
the installation.

46 
 
b) Cattle:
The air quality conditions in a cattle confinement are dynamic and change
continuously. Pollutant gases and excesses of humidity, microorganisms and particulates
are produced from various sources, such as the floors, walls, feeds, water and air which
infiltrates the installation. Therefore gas emissions directly depend on atmospheric
influences which permanently change the conditions.
Various approaches to solve this problem have been studied. The quantification of
carbon dioxide as well as mass and heat balances were performed by Van Ouwerkerk and
Pedersen (1994) according to a method for calculating the concentration difference
between interior and exterior air to determine the concentration levels.
Gebremedhin and Wu (2003) utilized predictions based on CFD simulations to aid in
modeling of the thermal environment in dairy cattle installations. In their study, the Fortran
computational program was used to calculate all models which were necessary for
determination of heat and mass transfer on the body of these animals. These authors thus
used this method together with CFD simulations for heat and mass transfer between
animals and the environment.
Gebremedhin and Wu (2005) studied the utilization of CFD simulations for air flux in
a simple geometric cattle installation with forced ventilation to investigate the heat and
mass transfer phenomena. They discovered that the total heat loss of an animal is highly
dependent on its position and orientation for the flow field. For example, a cow positioned
near the air entrance may lose a total of 710 W in heat, while an animal positioned at the
other extremity of the installation may lose only 214 W of heat. Figure 2.10 illustrates a
CFD predicted flow field around cows in a ventilated room.

 
(a) (b)

Figure 2.10 - Difference in the air velocity around cows in a ventilated room.
(Gebremedhin and Wu, 2005).

47 
 
 Norton et al. (2010a) used a CFD model to provide insight into the air mixing
characteristics in a calf building during wind-driven and wind and buoyancy-influenced
ventilation. This study found the porosity of the eave opening cladding system heavily
influenced the mixing of indoor air during wind-driven ventilation, with the low porosity
cladding significantly enhancing the building’s ventilation efficiency. Figure 2.11 shows
the simulated air distribution patterns in the calves building.

Figure 2.11 - The predicted flow patterns on a vertical plane in the calves building with a
wind speed of 0.5 m s-1 (Norton et al., 2010a).

c) Swine:
Production and reproductive performance of swine depend on the employed
management system, which involves the selected system of breeding, nutrition, and the
installations themselves. These installations, which demand a large initial investment, are
constructed in function of the costs and facilities of the caretaker, often neglecting the
comfort of the animals. Thus, many times the application of CFD models on the
environment of these animals are directed towards the optimization of environmental
conditions (Bjerg et al., 2000; Sun et al., 2002; Sun, 2004). These studies may be
efficiently grouped for modeling of pollutant dispersion and the thermal environment
inside swine production installations.
Sun (2004), utilizing CFD models for two and three dimensions, obtained good results
for the determination of ammonia concentration distributions during summer and winter
conditions in a commercial High-Rise™ Hog Building (HRHB).

48 
 
2.9. Reduced models
The use of reduced models is a widely used tool in engineering, but its use requires a
clear understanding of the principles governing the relationship between the model and
prototype. For the behavior of a prototype can be derived from a model is necessary that
both behave in a qualitatively similar and that a quantitative relationship can be established
among them (Jentzsch, 2002).
Reduced models, when examined with certain criteria of similarity, can be substitutes
for valid systems that, for some reason cannot be studied in full-size prototypes. The scale
utilization of physical phenomena can be advantageous for four reasons (Schuring, 1977):
a) When the problem is treated as very complex or unfamiliar, and empirical
information is necessary for an analytical approach;
b) Reduced models allow the reduction ratio of the system to facilitate the handling;
c) Allows a reduction in time required to perform the research; and
d) Provides a greater understanding of the phenomenon investigated.

The similarity theory is developed by dimensional analysis. Objectives establish the

necessary relations for the behavior of a prototype and can be determined from
observations in a model and the relationships between the variables involved in the
phenomenon, so that the data can be systematized safely (Költzsch and Walden, 1990).
The high cost involved in building a structure for agricultural research of new
technologies, such as dairy housing, often do not include the execution of a study due to
limited resources. In this way, the use of reduced models to study various types of roof
ridge and effects on natural ventilation in a dairy housing can contribute significantly to
reducing the cost of research, and allows testing various settings by changing the model at
a lower cost than in a full-scale prototype.
The use of reduced building models for animal production are mainly limited by the
difficulty of performing the tests under production conditions, thus simulating the heat and
moisture dissipation by the animals must be added to the system.
The use of scale models to study the behavior of physical phenomena of a system,
expanded with the development of the theory of similarity from the 19th century. This
development was due to the need to understand the behavior of physical phenomena
involved in new technologies generated mainly in the aerospace industry and shipbuilding.

49 
 
In this way, several models have been proposed to solve field problems and to facilitate the
understanding of various physical processes, such as air-flow characteristics of a scale
model chamber (Weller et al., 1970), reduced model to study of alternative ventilation
systems for a broiler house (Wilson and Bishop, 1974), ridge vent and wind direction
effects on ventilation characteristics of a model livestock building (Egan and Hellickson,
1977), algorithm for determining opening effectiveness in natural ventilation by wind
(Nääs et al., 1998), and others.
An important factor in making reduced models is the relationship between cost and
accuracy or quality. If the cost increases with the requirement of accuracy of the model, it
is necessary to establish the standard that meets the proposed objectives, see Figure 2.12.

Figure 2.12 - Relationship between cost and accuracy of the model. Adapted from Költzsch
and Walden (1990).

2.9.1. Similitude theory

Many problems in practical engineering, involving heat and mass transport over time,
can be modeled using solved equations and procedures to describe processes. But a large
number of problems can be solved only if the mathematical models are related to the
experimental data (Rocha, 2012).
A mathematical model describes the behavior of a real-life system in terms of
mathematical equations. These equations represent the relations between the relevant
properties of the system under consideration. Parameters that can be influenced by the
observer are referred to as adjustable. The other parameters act as constants in the model.
For example, in atmospheric models used in weather forecasting, one is interested in
properties like temperature and humidity (the dependent variables) as functions of position
50 
 
and time (the independent variables). Important parameters are then the gravity field and
the rotational speed of the earth, and these clearly belong to the class of nonadjustable
parameters. The solution of a mathematical model is known if we can determine the
relations between dependent and independent variables. Since the solution depends on the
values of the adjustable parameters, mathematical models are a powerful tool with which to
determine which values of the adjustable parameters yields specific required behavior.
The basic considerations of dimensional analysis and similarity theory are also used to
assist the engineer to understand and correlate the data that were obtained in their study or
by other researchers. One objective of experimental research is to analyze the systems to
make them applicable. Thus, the concept of similarity is often used (Vekariya et al., 2011).
In the process of construction of scientific knowledge, the models have been used as
the link between research and theory. In this way, it means that the model productions and
physical reproductions of the processes, structures, and objects can be accomplished using
the homology and analogy. In homology, the original and model have the same basic
physical qualities, and the difference between them is mainly quantitative (Jentzsch, 2002).
The use of homology results in physically similar models. However, in analogy, the
original and model have different basic physical qualities, but have in common structural
and functional characteristics. The use of models results in similar physical analogy
(Murphy,1950).
So the prototype and the model are similar when all quantities homologous features
present a constant relationship between them, in other words, similarity between them are
constants (Jentzsch, 2002). These constants of similarity (cs) can be set for all magnitudes a
basic measurement system, such as length, mass, time, temperature (index M = model and
P = prototype):

Length → cs (2.8)

Mass → cs (2.9)

Time → cs (2.10)

θ
Temperature → csθ (2.11)
θ

51 
 
 By combining these constants of similarity, one can obtain a variety of products, such
as widely used dimensionless physical quantities (terms Pi), the Reynolds, Grashof, and
Prandtl numbers (Jentzsch, 2002).

2.9.2. Buckingham Pi -Theorem

The Buckingham theorem provides a method for computing sets of dimensionless
parameters from the given variables, even if the form of the equation is still unknown.
Thereby, according Murphy (1950), consider a system with variables x1, x2, x3,. . . , xn and
a function between them:

ƒ(x1, x2, x3, ... , xn) = 0 (2.12)

If k is the number of fundamental dimensions required to describe the n variables, then

there will be k primary variables and the remaining s = k – n variables can be expressed as
n – k dimensionless and independent quantities or Pi groups. Thus, the function f can be
replaced by the function:

ƒ(π1, π 2, π 3, ... , π s) = 0 (2.13)

The previous function can be determined if the term π1 had a intended dependent
variable, then:
π1 = ƒ(π2, π3, ... , πs) (2.14)

Thus, Equation 2.14 is general and can be applied to any other system that is a
function of these variables, and consequently a particular system called model (index M).

π1M = ƒ(π2M, π3M, ... , πsM) (2.15)

Thus, can one can obtain a prediction equation for π1 from π1M through division
Equation 2.14 by Equation 2.15.

52 
 
 π1 f  π 2 , π3 ,..., πs 
= (2.16)
π1M f  π 2M , π3M ,..., π sM 

If the model is designed and operated as follows below:

π2M = π2 (2.17)
π3M = π3
πsM = πs

Then,

ƒ(π2, π3, ... , πs) = ƒ(π2M, π3M, ... , πsM) (2.18)

And consequently:

π1M = π1 (2.19)

Equation 2.19 is the prediction equation, which will be valid only if the operating
conditions and project demands are fulfilled (Murphy, 1950).
On the other hand, if it is not found that all design criteria and operating conditions
affecting the prediction equation are not met, then this may change the behavior of the
model, (Timmons et al, 1978). Thus, the correction of the prediction equation can be made
by determining predictive factor δ, so (Murphy, 1950):

π1 = δπ1M (2.20)

2.9.3. Models versus prototype

Performing scale analysis is mathematically straightforward. However, there are
several issues to consider (Bahrami, 2012):
a) List of variables that are involved in the phenomenon under investigation: Selection
of important variables is not a trivial task and requires considerable judgment and
experience.

53 
 
 b) Reynolds number: generally the prototype has a very large Reynolds number (Re),
in which case slight variation in Re causes little effect on the behavior of the problem.
Unfortunately, models are sometimes so small and the Reynolds numbers are large and the
viscous effects take effect. This situation should be avoided to achieve correct results.
Solutions to this problem would be to increase the size of the model, or more difficult, to
change the fluid (change the viscosity of the fluid) to reduce the Re.
c) After selecting important parameters and determining Pi groups, one should seek to
achieve similarity between the model tested and the prototype to be designed. In general,
flow conditions for a model test are completely similar if all relevant dimensionless
parameters have the same corresponding values for the model and the prototype. Therefore,
three particular types of similarity can be determined.
i) Geometric similarity: concerns the length dimension (L) and it must be ensured
before any model testing can proceed. In general geometrical similarity is established when
a model and prototype are geometrically similar when all body dimensions in all 3
coordinates have the same linear scale ratio.
Figure 2.13 is showing a general example of geometric similarity in model testing.

Figure 2.13 - Geometric similarity in model testing (Bahrami, 2012).

ii) Kinematic similarity: requires that the model and prototype have the same length
scale ratio and the same time scale ratio; thus the velocity scale ratio will be the same for
both.
Length scale equivalence simply implies geometric similarity, but time scale
equivalence may require additional dynamic considerations such as equivalence of the
Reynolds and Mach numbers.

54 
 
 iii) Dynamic similarity: exists when the model and the prototype have the same length
scale ratio, time scale ratio, and the force scale (mass scale) ratio.
Dynamic similarity exists, simultaneously with kinematic similarity, if the model and
prototype force and pressure coefficients are identical. This is established if:
- For compressible flow: the model and prototype Reynolds number and Mach number
and specific heat ratio are correspondingly equal.
- For incompressible flow:
With no free surface: model and prototype Reynolds numbers are equal;
With a free surface: model and prototype Reynolds number, Froude number, and (if
necessary) Weber number and cavitation number are correspondingly equal.

2.10. Wind tunnel

Roberts (2001) describes hydrodynamics as the study of how fluids and gases move
around an object. He further states that the study of hydrodynamics is important in that
fluid movements help determine the shape and function of many vehicles and building
structures.
Konstantin Tsiolkovsky conducted experiments in the study of fluid movement in 1892.
He developed prototypes of wind tunnels to study and measure aerodynamic concepts. His
contributions helped lead to the understanding of aerodynamics employed by race car
drivers, pilots and engineers today (Fitzgerald, 2005). Although CFD have added another
dimension in testing recently, wind tunnels are still used to familiarize and appreciate
aerodynamic forces at work. It is essential to recognize fluid properties and changes that
affect the accuracy throughout the test section (Husainie and Qamar, 2012).
In general, Agriculture Engineers have used wind tunnels to study mockups and
prototypes of animal building and food storage structures. They collect data so they can
design a better structure. They may use a wind tunnel to study the ventilation
characteristics of a mockup before spending money on a prototype building. They also
investigate the characteristics of the building, how the control surfaces react at various air
velocity and drag characteristics.
Wind tunnels represent a useful tool to investigate various flow phenomena (Lindgren
and Johansson, 2002). In another words, the wind tunnel is an installation producing a
regular and controlled air stream to determine experimentally the conditions of flow and to
measure distribution of pressures, monitoring the temperature, and other relevant aspects of
55 
 
the study (Gorecki, 1988). With various features provided by wind tunnels, these results
can be classified according to the position of construction (vertical or horizontal), the type
of camera (open or closed) and the fluid velocity (from subsonic to hypersonic).
Concerning buildings, horizontal models are the most used, due to convenience and
lower cost of construction. The use of an open or closed chamber will depend on the type
of study to be performed, as well as the fluid velocity and body size needed to meet the test
objectives (Barlow et al., 1999).
There are two basic types of wind tunnels and two basic test-section configurations
(Rae and Pope, 1984):
1) Open circuit tunnel: In this type of tunnel the air follows along a straight path from
the entrance through a contraction to the test section, followed by a diffuser, a fan section,
and an exhaust of the air. The tunnel may have a test section with or without solid
boundaries, see Figure 2.14.

Figure 2.14 - Schematic of open return tunnel (Adapted from Mueller, 1989).

An open circuit tunnel has the following advantages and disadvantages (Rae and Pope,
1984).
a) Advantages:
i) Construction cost is less;
ii) If one intends to run internal combustion engines or if there is flow visualization via
smoke, there is no purging problem if both inlet and exhaust are open to the atmosphere.
b) Disadvantages:
i) If located in a room, depending on the size of the tunnel to the room size, it may
require extensive screening at the inlet to get high-quality flow. The same may be true if

56 
 
the inlet and/or exhaust is open to the atmosphere, when wind and cold weather can affect
operation;
ii) For a given size and speed the tunnel will require more energy to run. This is usually
a factor only if used for developmental testing where the tunnel has a high utilization rate;
iii) In general, a tunnel is nosy. For larger tunnels noise may cause environmental
problems and limits on hours of operations.

Because of the low initial cost, an open circuit is often ideal for schools and universities
where a tunnel is required for classroom work and research and high utilization is not
required.

2) Closed return wind tunnel: This tunnel has a continuous path for the air. The great
majority of the closed circuit tunnels have a single return (Figure 2.16), although tunnels
with both double and annular returns have been built. Again, the closed circuit tunnel may
have either a closed or open test section, and some have been built where it can be run with
either an open or closed test section.

Figure 2.15 - Schematic of closed return wind tunnel (Adapted from Tatman, 2006).

A closed return tunnel has the following advantages and disadvantages (Rae and Pope,
1984).
a) Advantages:
i) Through the use of corner turning vanes and possibly screens, the quality of the flow
can be easily controlled;
57 
 
 ii) Less energy is required for a given test-section size and velocity;
iii) Less noise when operating.

b) Disadvantages:
i) Higher initial cost due to return ducts and corner vanes;
ii) If used extensively for smoke tests or running of internal combustion engines, there
must be a way to purge tunnel;
iii) If tunnel has high utilization, it will be necessary to install an air exchanger or some
other method of cooling during hot summer months.

The Reynolds number is the basis for the behavior of real systems when using reduced
physical models (Fox and McDonald, 1998). A common example is the wind tunnel that is
used to measure aerodynamic forces in models of wings of airplanes, automobiles, and
buildings. Thereby, in the ideal case, the model should be tested at the same Reynolds
number as the full-scale vehicle or building. Typically, for experimental values, the flow is
laminar when Re < 2100, and it is turbulent when Re > 4000 (Fox and McDonald, 1998;
Incropera and Dewitt, 2001).
In most wind tunnels, air at atmospheric pressure is used as the fluid. The gases are
compressible and density varies with temperature and pressure. But in many cases the
density of air can be considered constant, to assist engineer in calculating accurate results
(Barlow et al., 1999).

2.10.1. Wind tunnel design

Wind tunnel design is a complex field involving many fluid mechanics and engineering
aspects and it is impossible to cover them all in just a few pages. Some books and articles
have been written about this topic and some references are useful references when
designing and constructing low-speed wind-tunnels (Bradshaw and Pankhurst, 1964; Rae
and Pope, 1984).
There are many types of wind-tunnels and they can be classified by flow speed dividing
them into four groups (Lindgren and Johansson, 2002): a) Subsonic or low-speed wind-
tunnels; b) Transonic wind-tunnels; c) Supersonic wind-tunnels; d) Hypersonic wind-
tunnels.

58 
 
 There are special-purpose subsonic tunnels, such as spin tunnels, icing tunnels,
meteorological tunnels, and tunnels designed for the testing of buildings. In this study, a
subsonic tunnel was built for the evaluation of the air flow in reduced models of CBP
barns.
There are some large wind-tunnels used in the auto and aircraft industries. These
tunnels have test sections that can accommodate full scale vehicles and small aircraft
(Lindgren and Johansson, 2002). In the medium size tunnels with a test-section area around
9.3 m2, large amounts of both research and aircraft developmental testing is accomplished.
In a smaller size test section, the tunnels are used for research and instructional purposes.
The size of smaller tunnels is usually determined in the final analysis by the size of the
room that will house the tunnel. Ideally, the size of a tunnel is determined by its purpose
(Rae and Pope, 1984).

a) Test section:
This is the starting point in the design of a wind tunnel. The purpose of a wind tunnel
is to provide a uniform and controllable air flow in the test section that passes over the
model (Rae and Pope, 1984). The test section is the chamber in which measurements and
observations are made and its shape and size are largely determined by the testing
requirements.
The test section should be long enough that flow disturbances resulting from a
contraction or screens are sufficiently damped before the reaching the test object. However,
care should be taken not to make this section too long as this will lead to detrimental
boundary layer growth which can separate when it enters the exit diffuser and create a
power loss. This can be prevented by slightly enlarging the tunnel or by partially
obstructing the exit end of the tunnel to create an overpressure which allows the use of
small vents to control boundary layer growth (Tatman, 2006).

b) Diffuser:
Diffusers are chambers that slowly expand along their length, allowing fluid pressure to
increase with decreasing fluid velocity. Angles slightly larger than 5 degrees do increase
pressure recovery, but can also lead to boundary layer separation and thus flow
unsteadiness. Exit diffusers are located downstream of the test section and are used to

59 
 
recover pressure from kinetic motion of the fluid thereby reduced the power required to
drive the tunnel (Tatman, 2006).
Wide angle diffusers are located between the fan and the settling chamber and are
necessary in order to facilitate the use of a beneficial contraction section, but the wide
angle leads to boundary layer separation which must be controlled with the use of screens.

c) Contractions sections:
Contractions sections are located between the settling chamber and the test sections and
serve to both increase mean velocities at the test section inlet and moderate inconsistencies
in the uniformity of the flow. Large contraction ratios and short contraction lengths are
generally more desirable as they reduce the power loss across the screens and the thickness
of boundary layers. Small tunnels typically have contraction ratios between 6 and 9.

d) Fan section:
Axial fans are popular in open return tunnels, and are almost always found in closed
return tunnels. In larger tunnels, pre-rotation vanes called stators are commonly positioned
upstream of the fan, substantially decreasing swirl in the exit flow. Axial fans have a
relatively limited effective operating range as the reduction it pressure increase through the
fan as the blades approach stall speeds is far more abrupt than in centrifugal blowers. Care
must also be given to choosing the proper blade size, shape and spacing in order to prevent
shock wave production, stalling, and backflow (Wallis, 1983).
Centrifugal blowers are most often in blower type open return tunnels, though they can
be used in closed return tunnels if mounted in a corner. Centrifugal blowers have a much
larger operating range than axial fans with acceptable levels of unsteadiness (Johnson and
Hancock, 1997).

e) Settling chamber:
The settling chamber is located between the fan or wide angle diffuser and the
contraction and contains the honeycombs and screens used to moderate longitudinal
variations in the flow. Screens in the chamber should be spaced at 0.2 chamber diameters
apart so that flow disturbed by the first screen can settle before it encounters the second
(Tatman, 2006). Turbulence in the test section is reduced by the installation of honeycombs
and screens upstream of the contraction.

60 
 
 Screens reduce the axial turbulence more than lateral turbulence. Screens have a
relatively large pressure drop in the flow direction, which reduces the higher velocities
more than the lower, and thus promote a more uniform axial velocity (Rae and Pope,
1984).
Honeycombs have a small pressure drops and thus have less effect on axial velocities,
but owing to their length, they reduce the lateral velocities. The minimum length of a
honeycomb should be 6 - 8 times the diameter of the hole (Rae and Pope, 1984).
Both screens and honeycombs reduce lateral and axial turbulence, probably due to an
exchange in energy along the axis as the turbulence tends toward isotropic turbulence
downstream.

2.10.2. Flow visualization

In this field of hydrodynamics science, the flow visualization has been a nebulous
observation simply because air is invisible. Flow visualization has led to the discovery of
flow phenomena and has helped in the development of mathematical models for complex
flow problems. It is also useful in the verification of existing theories, and has been an
important tool in the development of complicated engineering systems (Mueller, 1989).
The use of saw dust, tufts, fluorescent dye in oil with black light, and laser beam, are
some of the methods and materials used to attempt flow visualization. The use of saw dust
increases the difficulty of maintenance in the wind tunnel and its associated equipment.
The tuft and oil methods show flow on the surface of the model. The laser beam has not yet
been perfected as a usable tool for this purpose (Shindo and Brask, 1969).
Mitchell and Ross (1977), and Ogilvie and Boyd (1984) used a water flume to visualize
the airflow patterns in naturally ventilated barns while Aynsley et al. (1977) and Simango
and Schulte (1983) used a neutral buoyancy bubble technique in a wind tunnel. Whatever
the method used, the substance introduced into the tunnel should be non-corrosive, non-
toxic, and the equipment must be safe to handle. Because of their high initial costs, neither
of these two techniques was used in this study. Rather, a simple wind tunnel with smoke
was used to allow qualitative determination of air patterns.
The smoke tunnel was an important research tool since 1959. The system consists of a
low-turbulence wind tunnel, smoke-generation and injection apparatuses, and lighting and
photographic equipment. The arrangement and use of these system components must be
coordinated to obtain good results (Mueller, 1989).
61 
 
 3 - MATERIAL AND METHODS

3.1. Introduction
This chapter is divided into four main sections based upon the chronology of the
investigation: i) Summarize compost bed data, barn dimension data and determine from
qualitative and quantitative methods major interactive factors in the success of bed
composting; ii) Summarize bed compost water holding capacities (WHC)  and compare
moisture contents found with water holding capacity to determine the influence of moisture
on bed compost; iii) Develop predictive equations for thermal properties of compost as a
function of at various combinations of moisture content, pressure, and particle size; and iv)
Develop 3-D CFD model of compost barn with alternate ridge designs and visually
demonstrate effect on air flow through structure.

3.2. Characterization and description of CBP barns

3.2.1. Barn facilities measurements
Data to realization of this study were collected from different CBP barns, distributed
throughout the state of Kentucky - U.S. (Figure 3.1), between October 2010 to March
2011. In this way, if a farmer described the barn as a compost bedded pack barn but the
pack was rarely or never stirred, the barn was not included in any data analysis. Thus,
forty-two farms participated in the study.

Figure 3.1 - CBP barns utilized of this study in Kentucky State.

62 
 
 The barn dimensions were measured using a steel tape measure, surveyor`s tape, level
rod, and a measuring wheel (Figure 3.2). Roof pitch was determined using a level and
framing square placed directly on the roof. Two fiberglass ladders were used to allow
access to roof structures and ventilation fans (Figure 3.3). Digital photography was used
extensively to document interior and exterior structures and systems to assist in accurate
creation of the database of barn structure, layout and ventilation details. All information
was recorded in a notebook.

(a) (b)
Figure 3.2 - Equipments used to measure CBP barn dimension: (a) surveyor`s tape and (b)
measuring wheel.

(a) (b)
Figure 3.3 - Equipments used to measured roof pitch and fans dimension: (a) Level and
framing square and (b) fiberglass ladders.

3.2.2. Environmental measurements

Air temperature, relative humidity, and air velocity and wind direction were measured
inside and outside of each barn. At nine locations within a barn, measurements were taken

63 
 
in the compost pack area at two different heights (0.05 and 1.2 m) one time on the day of
the site visit (Figure 3.4). Air temperature and relative humidity were measured using a
weather meter (accuracy of ±1°C; Kestrel®, model 4000, Sylvan Lake, MI, USA). Air
velocity was measured using a hot wire anemometer (accuracy of ±0.01 m s-1; Testo®,
model 425, Sparta, NJ, USA). The air direction was measurement with a weather vane.
Sampling points (A1 to A9) represent the approximate center of nine grid spaces within the
manure pack. The grids were established based on post spacing along the pack and
generally along the center of the pack (Figure 3.5).

Weather meter

Weather meter
Hot wire
anemometer

(a) (b)
Figure 3.4 - Equipments used to measured environment: (a) hot wire anemometer and (b)
weather meter.

Figure 3.5 - Nine grid spaces (A1 - A9) and one outside point (Poutside) of environmental
measurements and samples collections inside of the barns.

3.2.3. Bedding temperature measurements

Bedding temperature was measured at nine locations as previously described (Figure
3.5) across each pack at the surface and at two different depths (0.10 and 0.20 m). Bedding

64 
 
surface temperature was measured using two types of sensors. An infrared thermometer
(accuracy of ± 1°C; Fluke®, model 62, Everett, WA, USA) and a compact thermal camera
(accuracy of ± 0.1°C; Extech®, Flir i5, Waltham, MA, USA) were used (Figure 3.6).
Bedding temperature was measured using a thermocouple-based thermometer (0.22 m
length, accuracy of ± 2.2°C; Fluke Inc., model 87, Everett, WA, USA), see Figure 3.7.

Infrared thermometer
(a) (b)
Figure 3.6 - Equipments used to measured bedding surface temperature: (a) infrared
thermometer and (b) a picture of compact thermal camera.

(a) (b)
Figure 3.7 - Thermocouple-based thermometer used to measure bedding temperature.

3.2.4. Bedding moisture analysis

Samples of pack bedding were collected from a mixture of compost of the top (0.10 m)
in nine different locations inside the barns using plastic bags (Ziploc®, Double Zipper,
Racine, WI, USA). Samples were collected using an iron hoe with a wooden handle and a
soil auger (Figure 3.8). The bedding samples were cool on ice upon collection and
refrigerated when returned to lab at 1.0 ºC.

65 
 
 (a) (b)
Figure 3.8 - Equipments used to collect the bedding sample: iron hoe (a) and (b) storage in
plastic bags.

Bedding compost moisture contents throughout this study were measured by drying at
105°C for approximately 24 h (Figure 4.9). Moisture content (dry-basis) of bedding
compost was defined by Equation 4.1 (Sales, 2008; Melo, 2011).

MC ∙ 100% (4.1)

Where: mw is the mass of wet material (g) and md is the mass of dry material (g)
determined by placement of the sample in a convective oven at 105oC for 24 hours. MCdry-
basis is expressed as a percentage.

(a) (b) (c)

Figure 3.9 - Determination of moisture content in compost material: (a) wet weight of
sample, (b) samples inside the oven, and (c) dry weight of sample.

66 
 
3.3. Physical, chemical, bacterial and thermal properties of compost materials
3.3.1. Sample preparation and characteristics of materials
Fresh compost bulking materials were collected from 42 CBP barns in Kentucky
(Figure 3.1). The bedding compost samples were collected during farms visits between
mid-October 2010 and early-February 2011.
Samples of bedding compost were collected from the surface layer (0.10 m deep) in
nine different locations in the resting area of a CBP barn (Figure 3.5). Each location is the
center of a rectangular area of a 3 x 3 m grid that divides the CBP resting area into 9 equal
areas. Samples were collected using iron hoe and soil auger (Figure 3.8). A 20 liters
container was filled with incremental quantities of bedding collected from the nine
location. The bedding compost samples were immediately refrigerated upon return to the
lab at 1.0 ºC, see Figure 3.10.

(a) (b)
Figure 3.10 - The bedding compost samples refrigerated in container upon collection.

3.3.2. Physical properties of compost materials

Physical analyses including particle size distribution, bulk density, particle density,
porosity, water holding capacity were performed to characterize the compost materials.

3.3.2.1. Particle size distribution

As-received compost was allowed to air-dry for 48 hours before the determination of
the particle size distribution (Figure 3.11a). Dried compost was poured in graduated
volume cylinders sieved in a sieve shaker (Ro-Tap Model B, W. S. Tyler, Inc., Mentor,
67 
 
OH, USA) with sieves vertically aligned in series in a decreasing mesh screen opening
order: 25.00 mm, 8.00 mm, 5.60 mm, 4.75 mm, 2.00 mm and a pan of the bottom (Figure
3.11b). Compost bedding material was sieved in the shaker for 3 minutes. The amount of
compost retained by each screen was poured in a beaker and its weight determined.

(a) (b)
Figure 3.11 - As-received compost drying (a) and (b) shaker and sieves used to characterize
particle size ranges.

The total compost bedding sample weight (Wtotal) can be assumed as the sum of all
sieved particle ranges plus bottom pan (Maia, 2010). So, Wtotal is given by Equation 3.2:

W W W W W W W (3.2)

Where, Wcourse, WR2, WR3, WR4, WR5, and Wfines represent weights of course material, R2,
R3, R4, R5 and fines (pan), respectively.

Particle weight fractions for ranges 2, 3, 4, and 5 ( , , , and

respectively) were determined by Equation 3.3:
#
F #
% 100 (3.3)

Where, WR# represents the weight for particles sizes of the three ranges.

Compost weight fraction of fines on the pan, (%), were calculated as in Eq. 3.4.

F % 100 (3.4)

68 
 
 Finally, weight particle fractions were estimated as in Equation 3.5.

F % F % F % F % F % F %
F % (3.5)

3.3.2.2. Bulk density, particle density, and porosity

Compost bulk density was measured by adding 1.6 liters of air-dried compost into a 2.0
liters graduated volume cylinder (0.05 m height and 0.20 m diameter). The container was
filled with bedding compost material, and then the material was slightly compacted to
ensure absence of large void spaces and, then the material was vibrated using a vibrating
Jigsaw (Black & Deck, Model JS515, MD, USA) for 30 seconds (Figure 3.12). The bulk
density can be calculated by dividing the weight of the material by the compost filled
volume of the cylinder (Ahn et al., 2009; Sales, 2008).

Figure 3.12 - Cylinder with compost bulking material vibrating.

Compost particle density was determined by pouring 20 g of as-received compost into

five graduated cylinders and adding a portion of 0.10 liter of methanol (90%) to each
cylinder up to fill to the 0.10 liter mark (Figure 3.13). The remaining unused methanol
volume represented the volume of particles in the compost filled cylinder (Sales, 2008).

69 
 
Figure 3.13 - The methanol and other equipments used to determinate particle density.

Porosity was determined for each collected CBP bedding compost from the results of
bulk and particle density by using Equation 3.6:

 ρ 
ε = 1 - b  (3.6)
 ρ p 


where: ε is the overall porosity (%), ρb is the bulk density (g cm-3) and ρp is the particle
density (g cm-3).

3.3.2.3. Water holding capacity (WHC)

The WHC was determined using 30 grams of the as-received compost that was
saturated with deionized water in a beaker, stirred for 3 minutes and left to rest for 7
minutes to absorb water. Saturated compost was poured into three Buchner funnels on top
of Erlenmeyer flasks to drain excess water (Whatman #41 filter paper), covered with
parafilm, and taken to an environmental chamber at 25ºC and 58% RH, where they drained
for 12 hours (Figure 3.14). Drained compost was weighed (Wi) and dried (Ws) in a
convection oven for 24 hours at 105ºC. The water holding capacity (g water/g dry material)
is calculated as (Ahn et al., 2009):

WHC =
 W -W  +MC  M
i s i
(3.7)
1-MC   W 
i i

where: MC is the initial moisture content of sample and M is the mass of the wet sample.

70 
 
 (a) (b) (c)
Figure 3.14 - As-received compost saturated with deionized water (a), environmental
chamber (b) controlled for temperature and relative humidity and (c)
interior of the unit with the six compost samples.

3.3.3. Thermal property measurement

Thermal properties were determined with transient heat dissipation device (KD2,
Decagon, Pullman, WA, USA). The thermal properties, thermal conductivity (k, W/m K)
and thermal resistivity (rho, °C cm/W), were measured for varying particle size ranges
(0.00mm < Finer < 2.00mm < PS2 < 4.75 mm < PS3 < 5.60 mm < PS4 < 8.00 mm <
Coarser < 25.00 mm), water content (30, 45 and 60%), and different static (fluffy material,
0.0, 0.1, 0.2, 0.3, and 0.4 MPa) and dynamic compaction degree (0.1, 0.2, 0.3, and 0.4
MPa).
The thermal properties was measured using a handheld device. It consists of a
handheld controller and sensors made from thin-wall stainless steel tubing with 2.4 mm
diameter x 100 mm long (see KD2 Pro® Operator`s manual, 2008). This probe was
inserted into the bedding compost medium.
First, thermal conductivity and thermal resistivity were measured for all samples in
different particle size ranges (0.00mm < Finer < 2.00mm < PS2 < 4.75 mm < PS3 < 5.60
mm < PS4 < 8.00 mm < Coarser < 25.00 mm), as described earlier (Figure 3.15).
Wet sample of compost bulking material with different moisture content (30%, 45%
and 60%) was produced in a concrete mixer that mixed the bedding material with the added
water for 3 minutes (Figure 3.16a).  The desired moisture contents (30%, 45%, and 60%
w.b.) was achieved by adding a known water amount for each particle size range’s initial
moisture contents (Table 3.1). If the initial moisture contents was higher than the selected
MC, a 50 g of compost was weighed and left to air dry until reaching the target MC.

71 
 
Figure 3.15 - Thermal properties sensor used to determinate the thermal conductivity and
thermal resistivity in different particle size.

Table 3.1 - Amount of water added to each particle size range to reach 30%, 45%, and 60%
(w. b.) moisture contents. Sample amount = 30g. Number of samples = 3.
Barn Initial Moisture (%) Water added or removed (g) to again
number Average S. deviation 30% 45% 60%
1 37.5 ± 11.8 -5.3 6.9 28.2
2 49.0 ± 7.3 -13.6 -3.7 13.7
3 49.1 ± 5.8 -13.7 -3.8 13.6
4 36.2 ± 6.3 -4.4 8.0 29.8
5 52.2 ± 5.6 -15.9 -6.6 9.7
6 39.6 ± 9.6 -6.8 4.9 25.5
7 46.0 ± 3.2 -11.4 -0.9 17.5
8 61.6 ± 2.5 -22.6 -15.1 -2.0
9 41.9 ± 6.6 -8.5 2.9 22.7
10 51.7 ± 7.3 -15.5 -6.1 10.4
11 61.2 ± 2.4 -22.3 -14.8 -1.6
12 60.4 ± 2.8 -21.7 -14.0 -0.5
13 57.3 ± 4.3 -19.5 -11.1 3.4
14 57.5 ± 4.2 -19.7 -11.4 3.1
15 54.7 ± 2.7 -17.6 -8.8 6.6
16 66.5 ± 2.2 -26.1 -19.6 -8.2
17 56.0 ± 4.3 -18.6 -10.0 5.0
18 64.8 ± 0.8 -24.9 -18.0 -6.1
19 68.4 ± 1.9 -27.5 -21.3 -10.5
20 60.3 ± 3.3 -21.7 -14.0 -0.4
21 58.8 ± 3.3 -20.6 -12.6 1.5
22 64.0 ± 1.6 -24.3 -17.3 -5.0
23 67.3 ± 1.3 -26.7 -20.3 -9.2
24 52.0 ± 4.4 -15.7 -6.3 10.1
25 66.4 ± 1.3 -26.0 -19.5 -8.1
26 61.7 ± 3.7 -22.6 -15.2 -2.1
Continue...

72 
 
Table 3.1 - Amount of water added to each particle size range to reach 30%, 45%, and 60%
(w. b.) moisture contents. Sample amount = 30g. Number of samples = 3.
Barn Initial Moisture (%) Water added or removed (g) to again
number Average S. deviation 30% 45% 60%
27 64.6 ± 2.8 -24.7 -17.9 -5.8
28 47.4 ± 2.6 -12.4 -2.2 15.7
29 67.8 ± 1.3 -27.0 -20.7 -9.7
30 61.2 ± 2.3 -22.3 -14.7 -1.5
31 61.1 ± 2.1 -22.2 -14.6 -1.4
32 57.3 ± 2.8 -19.5 -11.2 3.4
33 62.6 ± 5.6 -23.3 -16.0 -3.2
34 68.9 ± 2.0 -27.8 -21.7 -11.1
35 69.5 ± 0.9 -28.2 -22.3 -11.8
36 71.8 ± 0.8 -29.9 -24.4 -14.7
37 64.3 ± 3.5 -24.5 -17.6 -5.4
38 61.7 ± 6.4 -22.7 -15.2 -2.2
39 66.5 ± 1.0 -26.1 -19.6 -8.1
40 64.6 ± 1.7 -24.7 -17.9 -5.8
41 70.1 ± 2.7 -28.7 -22.9 -12.7
42 69.1 ± 1.2 -27.9 -21.9 -11.4

Different methods used to determined particle density and the amount of water added to
the compost to reach 30, 45, and 60% (w.b%) moisture content are described in Liberty
(2002) and Sales (2008).
A fluffy bedding material was weighed and the PVC cylinder (detailed cylinder
apparatus description in the following paragraphs) was filled. The PVC cylinder was then
vibrated using vibrating Jigsaw (Black & Deck, Model JS515, MD, USA) for 60 seconds
and resulting volumes were measured (Figure 3.16b).

(a) (b)
Figure 3.16 - Concrete mixer (a) used to mix the compost bulking material and (b) PVC
cylinder vibrated by a vibrating Jigsaw.
73 
 
 For study of different degrees of compaction, a mechanical pressing device was
designed and constructed at the shop of the Biosystems and Agricultural Engineering at
Department of the University of Kentucky. The press device was assembled and located in
the Agricultural Air Quality Laboratory of the same Department (Figure 3.17).

(a) (b)
Figure 3.17 - Mechanical pressing device designed (a) and (b) constructed to provide
different degrees of static and dynamic compaction.

Material compaction was achieved by a mechanical pressing device that was

developed to pack the sample down in the PVC container. The PVC cylinder had 0.27 m
height and diameter of 0.15 m. This PVC cylinder had a series of three equally spaced
(0.075, 0.150, and 0.225 m) holes through the entire height. Pipe flange was used to hold
the PVC cylinder on the base frame assembly using knurled hold down knobs (see
Appendix C - Figure 1). The compaction level inside of a PVC cylinder was adjusted using
a valve and manometer that controlled the air pressure inside of steel cylinder chamber that
moved the steel piston (Figure 3.18). A Moisture drain valve prevented water condensation
in the line of air pressure. The pressures applied (0.0, 0.1, 0.2, 0.3, and 0.4 MPa) were
based on studies conducted by van der Tol et al. (2003) and van der Tol et al. (2002).
The Decagon KD2-Pro (Decagon Devices Inc.) was used to evaluate the thermal
properties (thermal conductivity and thermal resistivity) associated with type pressure
applied (static and dynamic pressure). This sensor uses a metal probe (0.0024 m diameter x
0.100 m long) that heats up the material and then reads the temperature decay as the
material dissipates the heat to calculate its thermal properties. The probe was used with the
entire length of the needle embedded inside the compost. Measurements were made by
74 
 
placing the sensor probe into the three holes over cylinder and recording the measurement
after one minute, which was the recommended procedure for this sensor. Dynamic pressure
was applied when varying the opening and closing the air in the valve control (Figure
3.18b), during 12 times per minute. This range of moisture content (30%, 45%, and 60%)
was selected to simulate the conditions of the bedded compost found in the field. The
Decagon KD2-Pro sensor was calibrated by factory and performance verification standards
were observed during this study.

Manometer
Manometer
Valve

Security
button

Moisture drain valve PVC tube

Valve of control
(a) (b)
Figure 3.18 - Valve and manometer (a) and (b) mechanical pressing device components
used to compress the bedding compost samples.

3.3.4. Chemical properties of the compost bedding material

Each compost barn pack was subdivided in three equal areas within each of the nine
spots (Figure 3.5) from which 0.4 liters bedding sample were collected during the visiting.
The bedding samples were collected from the surface and kept cooled upon collection at
0°C till it was placed in a -40°C freezer and held for chemical analyses. Sub-samples were
ground as-is before analysis for Carbon (C), nitrogen (N), phosphorus (P), potassium (K),
calcium (Ca), magnesium (Mg), zinc (Zn), Copper (Cu), manganese (Mn), and iron (Fe).
The analysis was performed at the Division of Regulatory Services at the University of
Kentucky. The solid compost was dried in an oven at 75°C for 24 hours, ground to pass a 2
mm screen, and stored at room temperature prior to analysis. Nitrogen (%) and Carbon (%)
were determined via combustion. Macro nutrients (P, K, Ca, Mg) and micro nutrients (Zn,

75 
 
Cu, Mn, and Fe) were extracted by digestion using a combination of HCl and H2SO4 acids.
The digestate was assayed by Inductively Coupled Plasma Spectrometer for concentration
determination. The P, K, Ca, Mg and Zn are reported in a weight percent basis and Zn, Cu,
and Fe in ppm.
The following section summarizes information collected by a collaborator (Black,
2012). It is included to give a more details about chemical analysis.

3.3.5. Bedding bacterial of the compost bedding material

Compost samples were collected from nine equally spaced locations throughout the
barn described previously (Figure 3.5). Half of a cup was collected from each location,
thoroughly mixed together, and stored in a -57°C freezer to later be thawed for analysis of
bacterial content. The samples were sent to three different laboratories for analysis.
University of Kentucky in Laboratory of Soil analyzed the CBP bedding and fresh bedding
material for the following pathogens on a wet weight basis: Coliform, E. Coli,
Staphylococcal species, Streptococcal species, and Bacillus.
The samples were thawed in a refrigerator. Fifty cubic centimeters (cc) of bedding
material was measured using a sterile placed into a plastic bag (Ziploc®, Johnson & Son,
Racine, WI, USA). Two‐hundred fifty cc of sterile distilled water was added to the bedding
material which was mixed and allowed to stand for 10 minutes. The sample was mixed
again, a liquid sample was removed by pipette and serial 10‐fold dilutions were made in
sterile Brain Heart Infusion broth. Sample dilutions were plated (200 μL) on colistin
naladixic acid (CNA) agar (BBL, Sparks, Md.), MacConkey agar (BBL, Sparks, Md.), and
thallium sulfate‐crystal violet‐B toxin blood (TKT) agar medium. Colony counts were
determined for each sample after 24 hours of incubation at 37°C. Bacterial groups were
identified as coliforms (lactose‐positive colonies on MacConkey's agar, which include
Klebsiella by visual identification), streptococcus species (growth on TKT agar), Bacillus
species (growth on CNA agar and gram‐positive) and coagulase negative staphylococci
(growth on the CNA agar and catalase activity). Bacteria counts are expressed as colony
forming units (cfu) / mL of bedding sample.
The following section summarizes information collected by a collaborator (Black,
2012). It is included to give a more details about bacterial analysis.

76 
 
3.4. Study of natural ventilation in reduced model of CBP barn
3.4.1. Wind tunnel
A wind tunnel was built to evaluate the effect of air flow through reduced structure
model (1:16) of CBP barn with alternate ridge designs. The purpose of this wind tunnel
was to provide a uniform and controllable air flow in the test section that passes over this
reduced model. This wind tunnel was composed of an air intake, flow straightening section,
test section, and fan section. All studies were carried out in the wind tunnel of Biosystems
and Agricultural Engineering at Department of the University of Kentucky.
The wind tunnel was a low speed open circuit type with dimensions of 3.25 m wide,
8.85 m length, and a height of 3.15 m (Figure 3.19). The internal dimensions were 3.15 m x
8.75 m x 3.05 m comprising a total volume of 84.0 m3. Concrete blocks were used on the
base of the wind tunnel. Two by six wood boards were used to build the framing and sheets
of plywood were installed on the internal side of the framing members on the top, bottom
and both of side walls. Two metal doors were used for access to internal sections of the
wind tunnel.

b d

Figure 3.19 - Wind tunnel 3D model design is showing placements of (a) windows, (b)
doors, (c) dampers, and (d) concrete blocks.

The axial fan diameter was approximately 1.22 m and was capable of producing 2.33
m3 s-1 of airflow at a maximum 380 rpm and against a 10.0 Pa pressure increase. The fan
section can be easily altered to accommodate smaller fan sizes. Given the 9.60 m2
cross‐section of the wind tunnel, this allows for a maximum mean wind speed of 2.0 m s‐1.
The fan had three cambered blades, also made from sheet metal, and full power
requirement for fan operation was 0.746 kW (1.0 hp). The fan was set in a custom welded
77 
 
steel frame that was mounted on a fiber glass box. This fiber glass box was attached to a
sheet of plywood. Thereafter sixteen metallic clamps pressure were used to hold whole
structure (fan and fiber glass box) in the end wall (Figure 3.20).

Figure 3.20 - View of the fan mounted on the wind tunnel.

This wind tunnel was equipped with four metal dampers (1.52 m x 1.52 m) to control
the wind speed. The dampers were assembled at air intake of the wind tunnel and were
controlled by a handle on the outside. The room controls were located adjacent to the test
section. The dampers were made with stainless steel and were fitted with actuators which
has a rotation speed of 15s for 0 - 90 degree pivot.
Turbulence in test section was dampened by the installation of a plastic honeycomb
plates. This honeycomb plate was placed at 1.77 m down flow from the dampers to
promote a laminar flow of air. For optimum benefit, the length of honeycomb was 6 times
of the diameter of the hole (Rae and Pope, 1984). The honeycombs were supported by
metal frame. This frame was made from aluminum channel to hold and seal the edges of
the honeycomb. The channels were secured to one another with rivets. The honeycombs
slid snuggly into the settling chamber and were attached to the walls on each side.
This wind tunnel had two clear viewing windows (test section) with different sizes .
These windows were covered by 2.40 x 1.20 m and 1.10 x 1.20 m, respectively, of
Plexiglas with 0.003 m thickness. Those windows were placed at 3.60 m and 6.65 m,
respectively, from the upwind edge of the window. The Plexiglas windows were secured in
the wind tunnel with screws and silicone to seal and to prevent air infiltration. Care was
taken to drill holes through the Plexiglas for the screws to prevent cracking. Smoke was
used to attempted to fix air infiltration problems which arose. The length of the test section
78 
 
was chosen to be around 2.0 m to be able to get a high enough Reynolds number (Re <
2100) on a flat plate for anticipated wind tunnel uses.
Smoke was used primarily for flow visualization and photographs and videos were the
primary method of data recording. Airflow patterns were visualized using a fog machine
(power 1190 W, Rosco®, model 1700, CT, USA) which vaporized Glycerin for a smoke
source. The smoke tracer was introduced from the generator with a funnel on an end of a
duct and allowed to entrain smoke with the air using the negative pressure of the wind
tunnel as draw. Three light boxes, with 300 W incandescent lamps, focused light at three
different points along the length of the wind tunnel. The bright light combined with the
black inner surfaces allowed for photographing of air movement by both camera and
camcorder.
Airflow patterns were assessed using both air flow visualization and by measuring the
airspeed along the length of the test section. Airflow visualization allowed for a qualitative
image of the airflow patterns.
The final step in constructing of wind tunnel was sealing air leaks using foam sealant.
Figure 3.21 illustrates the metal dampers, plastic honeycombs, and fog machine used in the
wind tunnel.
Further information regarding constructions and calibration of the wind tunnel and
others details used in this study can be seen in Lopes (2012).

(a) (b) (c)

Figure 3.21 - Details of (a) metal dampers, (b) plastic honeycombs, and (c) fog machine.

79 
 
3.5. Reduced model
3.5.1. Determination of pertinent variables
Airflow through the roof ridge opening of a building acts essentially as fluid flow
through an orifice. The flow is governed by viscous forces, inertia forces, gravitational
forces, buoyancy forces and geometric relationships. Heat transfer from the buildings is
controlled by fluid flow, fluid properties and building geometry and characteristics.
Assuming that the same phenomenon governs performance in the reduced model, a list of
pertinent variables affecting the ventilation characteristics were compiled (Table 3.2) and
reduced to the basic dimensions of length (L), mass (M), temperature (θ), and time (T). The
selection of pertinent variables was based on the assumptions that fluid flow is
incompressible, the building atmosphere will be without manure gases and added moisture,
and there will be significant internal heat source present (heat of compost pack).

Table 3.2 - Variables affecting ventilation characteristics.

Dimensional
Variable n◦ Description Units (S.I) Symbols
symbol
1 Building length m l L
2 Building width m w L
3 Building height m h L
4 Gravitational acceleration m s-2 g L T-2
5 Coefficient of heat transfer kg s−3 K−1 h M T−3 θ−1
6 Heat kg m2 s-2 Q M L2 T-2
7 Wind velocity m s-1 v L T-1
8 Temperature of air K T θ
9 Density of air kg m-3 M L-3
10 Specific heat of air m2 s-2 K-1 c L2 T-2 θ-1
11 Heat transfer coefficient kg s-3 K-1 U M T-3 θ-1
12 Dynamic viscosity of air kg m-1 s-1 M L-1 T-1
13 Thermal conductivity of air kg m s-3 K-1 k M L T-3 θ-1
14 Coefficient of thermal expansion βe k-1 θ-1

According Buckingham Pi Theorem, a group of ten (14 variables minus 4 dimensions)

independent pi terms (Table 3.3) was derived with the details of the determination shown
in Appendix D. The relationship among the pi terms for fluid flow and heat transfer can be
expressed as:

80 
 
Table 3.3 - List of pi terms.
Term nº Pi term
1 π w∙l
2 π h∙l
3 π g ∙l ∙ρ
4 π v ∙T ∙ρ ∙h
5 π Q∙l ∙v ∙ρ
6 π c ∙v ∙T
7 π U∙v ∙T ∙ρ
8 π μ ∙ l ∙v ∙ρ
9 π k∙l ∙v ∙T ∙ρ
10 π β ∙T

Thus, the function represented by equation 2.13 is:

f π ,π ,π ,π ,π ,π ,π ,π ,π ,π 0 (3.8)

The terms and have the dependent variables necessary to predict heat transfer
and fluid flow, respectively. According the equations 2.14 and 2.15 showed previously, can
be determination that (Murphy, 1950; Klin, 1966; Jentzsch, 2002):

π f π ,π ,π ,π ,π (3.9)
π f π ,π ,π ,π ,π ,π ,π ,π (3.10)

According Similitude theory, the equations 3.9 and 3.10 can apply to another system
(using index M) and the function needs to use the same variables and be submitted to the
same conditions of operation.

π f π ,π ,π ,π ,π (3.11)
π f π ,π ,π ,π ,π ,π ,π ,π (3.12)

So, the prediction of equation to and from and can be calculated using
the division of equation 3.9 by equation 3.11 and equation 3.10 by equation 3.12,
respectively:

π6 f (π1 , π 2 , π3 , π 4 , π5 )
= (3.13)
π 6M f (π1M , π 2M , π3M , π 4M , π5M )

81 
 
 π9 f (π1 , π 2 , π3 , π 4 , π5 , π 7 , π8 , π10 )
= (3.14)
π9M f (π1M , π 2M , π3M , π 4M , π5M , π 7M , π8M , π10M )

Therefore, the model is designed and operated to:

π π (3.15)
π π
π π
π π
π π
π π
π π
π π

Thus,

f π ,π ,π ,π ,π ,π ,π ,π f π ,π ,π ,π ,π ,π ,π ,π (3.16)

As a result,

π π (3.17)
π π (3.18)

Equations 3.17 and 3.18 predict of the behavior of heat transfer and fluid flow,
respectively. Those equations are valid if the criteria of the design and operation conditions
are acceptable, as showing in Appendix D, according Murphy (1950).
The design conditions (Table 3.3) must be met if the reduced model is to accurately
predict conditions quantitatively in prototypes. Design conditions 1 through 5 require the
model geometric dimensions to be 1 : n (n = 16, the geometric length scale) times the
prototype geometric dimensions. Design condition 6 through 10 establishes the ratio of
prototype wind velocity to model wind velocity. Using the same fluid in both the reduced
model and prototype systems requires the model wind velocity to be n times the prototype
wind velocity. Details about design conditions can be seen in Appendix D.
The tests were performed with the airflow in the reduced model geometrically and
dynamically similar to a full-scale CBP barn. To ensure that similarity of the characteristics

82 
 
of the air flow in the reduced model and real scale CBP barn, the flow regime, expressed
by the Reynolds number must be equal in both cases. Therefore:

Re Re (3.19)

 ρνL   ρνL 
  =  (3.20)
 μ Model  μ  Full

where, Re is the Reynolds number (dimensionless), L is the characteristic length describing

the geometry of the flow field (m), ρ is the specific mass of the air (kg ⋅ m−3); v is the
velocity of the air (m ⋅ s−1), and μ is the viscosity of the air (N ⋅ m−2 ⋅ s−1).

While the CFD simulated the real CBP barn and the reduced model, the equations
(3.20) were used for the similarity of the air speed results. More details in Appendix D.

3.5.2. Reduced model procedure

One of the conditions of scaling is geometric uniformity which requires that all
dimensions of the real object should be equally scaled. In this study all dimensions of the
CBP barn were proportionally reduced with exception of the thickness of the roof.
Reducing the thickness of the roof would result in a deformation in the shape of the roof
and hence would change the speed and behavior of the airflow.
The scale of the model depends on:
a) The height and width of the test section in the wind tunnel. A model that is too large
could block the airflow in the tunnel and would be subject to wall effects.
b) The thickness of the boundary layer. Thereby under natural conditions the CBP barn
is situated in the boundary layer, the reduced model should also be placed in the boundary
layer in the wind tunnel.

3.5.2.1. Experimental reduced model

A reduced model was constructed in accordance with the design conditions and has a
geometric length scale of 16 for testing. The five ridge vents studies (Appendix E - Figure
1) were constructed with a 0.057 m ridge outlet corresponding to a 0.913 m opening in the
prototype. The four wind directions investigated (North, South, East, and West) were to
represent wind directions found in the field by rotating the model in the wind tunnel.

83 
 
 The reduced model was built using balsa wood and wood glue was used to join pieces.
Two aluminum plates were used as the roof cover this model. This reduced model had an
overall width of 1.20 m, length of 1.33 m, high at the ridge of 0.50 m, and the slope of the
roof was 4:12, so that the scale factor is 1:16. The reduced model is a copy of a CBP barn
24 (Appendix B - Figures 43 and 44) with 36 dairy cows situated in Kentucky (USA). The
ventilation openings in the reduced model consisted of a 1.20 m wide ridge opening and
eave openings of 0.30 m in height. The top of front wall and end wall under the gable were
made of Plexiglas to accommodate airflow visualization. This reduced model was built on
a wood table equipped with four wheels. The table has an opening of 0.93 m x 1.33 m and
aluminum perforated plate was used to covered this opening. This opening was occupying
80.9% of the floor area that represent the compost pack area. Figure 3.22 is showing the 3D
model design used in this study.

Figure 3.22 - 3D model design used in this study. Dimensions in meters (m).

3.5.2.2. Heating system

Eighteen incandescent lights of 40 W each were installed as electrical heaters under the
perforated plate to simulate the sensible heat of microbial activity in the compost bedding.
The maximum power flux was ≈ 560 W m-2 and was controlled by dimmer varying the
intensity of the lights. Thus, it was varied to maintain the floor (compost area) at 45°C. The
reduced model floor was maintained at a constant temperature to simulate the heat gains
from the compost bedding in the occupied zone.
An insulated box was built below and around the sides of the opening to promote the
heat transmission only through the perforated plate. The perforated plate was cover with

84 
 
aluminum sheet and 2.5 kg of sawdust was spread out over the surface. The heating system
and the reduced model with perforated plate, that was used in this study, see Figure 3.23.

Lights 
Insulated box 

Dimmer

(a)

Perforated plate 

(b)
Figure 3.23 - The heating system (a) and (b) reduced model without roof used in this study.

3.5.2.3. Static pressure, airspeed and temperature measurements

Static pressure was read using two hoses connected with the static pressure sensor
(Setra Systems Model 265, series 0811) shown in Figure 3.24, which measure static
pressure from 0 to 62 Pa, with an advertized accuracy of 1%. One hose goes inside the
wind tunnel, while the other goes outside the wind tunnel. In this way, the sensor could
provide the difference in pressures from outside and inside the wind tunnel during the tests.
The tests were run with the static pressure range 0 to 10 Pa.

85 
 
 Static pressure sensor 

(a) (b)
Figure 3.24 - Static pressure sensor (a) and (b) hoses on the sensor for static pressure
measurement.

Airspeed and air temperature were measured using two hot‐wire anemometer (accuracy
of ±0.01 m/s; Testo®, model 425, Sparta, NJ, USA). One sensor was placed at the
beginning of the working section at 1.00 m height above the floor of the wind tunnel and
another sensor was placed inside the reduced model using a non-uniform measured grid.
This non-uniform grid ranged from 0.10 m to 1.20 m horizontally at 0.55 m intervals and
from the floor to ceiling vertically ranged from 0.02 m to 0.20 m at 0.09 m intervals. The
locations of the measurement positions used in the experimental analysis are illustrated in
Figure 3.25. The hot-wire anemometer inside the building was mounted on a frame, which
allowed the anemometer to be manually moved to each measurement position (see Figure
3.25) once measurements at each location were complete. The values measured were
replicated three times. The sampling period was fixed at 10 minutes per collection point.
The experiments were carried out with two reference airspeeds: 0.1 m s–1 (low wind
speed) and 1.0 m s–1 (high wind speed). Those airspeeds were selected based on the tunnel
capacity and the wall opening area of the reduced model in order to achieve the desired
Reynolds numbers (Re < 2100) for laminar airflow around and through the model.

86 
 
Figure 3.25 - The plan and elevation showing the distribution of measurement points
throughout the reduced model.

3.6. The CFD model

3.6.1. CFD representation of the reduced model geometry
In order to validate the CFD model, a 3D Rhinoceros® design of the experimental
reduced model was developed and imported into the CFD package. The dimensions of the
computational model were: length (Y direction) 1.35 m, width (X direction) 1.20 m, eave
height (Z direction) 0.27 m, and the ridge height 0.48 m. The computational model had a
total volume of 0.64 m3. With 78.1% of the total compost pack space provided for animal
occupation and bearing in mind the minimum specification for dairy cows housing at 9.3
m2/cow (Barberg et al., 2007; Janni et al., 2007). The effect of confined dairy cows on
airflow was not considered in this CFD model.
In this study, five different ridge vents were built, subdivided into: (a) Closed ridge -
CLR, (b) Open ridge - OPR, (c) Open ridge with chimney - ORC, (d) Elevated ridge -
87 
 
ELR, and (e) Overshot ridge - OVR. The dimensions of the ridge opening were 1.35 m in
length and 0.057 m wide. The eave openings were also 1.35 m in length and were 0.23 m in
height (0.31 m2). The total compost pack surface area of the building was 0.877 m2 and
total area of the side wall openings was 0.077 m2. The modifications in ridge vents were
adjusted in simulations that examined the influence of different types of the ridge design on
the ventilation performance.
The measured values obtained experimentally for an open reduced model of CBP barn
and subjected to natural ventilation were used to assign the boundary conditions of the
model (Figure 3.26 and Table 3.4). Two cases (shown in Table 3.4) were tested using four
building orientation (North, South, West, and East).

Figure 3.26 - Diagram of the modeled and simulated reduced model of CBP barn.

Table 3.4 – Boundary conditions utilized in the CFD model

Case Location Boundary condition Value
Average air speed 0.1 m s-1
Inlet
Air temperature 22.0 °C
1
Outlet Atmospheric pressure 0.0 Pa
Floor Temperature of surface compost 45.0 °C
Average air speed 1.0 m s-1
Inlet
Air temperature 22.0 °C
2
Outlet Atmospheric pressure 0.0 Pa
Floor Temperature of surface compost 45.0 °C

88 
 
3.6.2. Computational modeling
To improve the CFD accuracy, the study exerted much effort to find the appropriate
mesh quality and density, and optimum setting condition of wind velocity.  The mesh
quality is very important to get accurate data and computational convergence. Thus, due to
the geometric complexity of the reduced model of CBP barn, it was opted to utilize the
ANSYS ICEM CFD® software for construction of a tetrahedral computational mesh which
allows for obtaining results with fewer errors (Lee et al., 2007). This software has been
specifically designed to fulfill the task of automatically generating high quality tetrahedral
meshes, which are suitable for scientific computing using numerical methods such as finite
volume methods.
The tetrahedral volume element is very useful to maintain the high quality of the mesh
to a region using different forms of complex geometry to be modeled in the region of
interest. But, it requires more nodes than the mesh elements and results in a hexametrical
file size and computational time increased (Lee et al., 2007; Katz and Sankaran, 2011).
When the reduced model and CFD computed results were compared with each other,
the reduced model and CFD computed airflow distribution at a cross-section plane of the
reduced model were mainly comparable with each other.

3.6.3. Boundary conditions

The results obtained by simulation using CFD were compared with the corresponding
data obtained experimentally in the wind tunnel. In this study, the CFD boundary condition
was specified as the same as the wind tunnel boundary condition of test section.
The measured values obtained experimentally in the wind tunnel were used to assign
the boundary conditions of the model. Thus, it is necessary to provide the initial values of
variables such as air temperature and velocity, calculate the terms of air temperature and
velocity for each condition and solve the equations of conservation of energy, mass and
momentum
Air flow rates are associated with turbulent flows and, combined with heat transfer
rates, generate a complex system of coupled equations difficult to resolve. Therefore, the
CFD technique was utilized to solve the average Navier-Stokes and energy equations,
determining velocity, temperature and pressure by the finite volumes technique. The model
for non-isothermal fluid flow is described by equation of mass, continuity, energy and
species, simplified as follows (Kin et al., 2008).
89 
 
 ∙ ρU 0 (3.21)

∙ ρU 0 (3.22)

∙ ρUU p μ U U (3.23)

∙ k T ρC TU 0 (3.24)

m∙ C ∙ D C (3.23)

where, ρ is the specific mass (kg ∙ m ); t is the time (s); U is the velocity vector; p is the
pressure (N ∙ m ); is the dynamic viscosity of the fluid (kg ∙ m ∙ s ); T is the
temperature (K); Cp is the specific heat (W ∙ kg ∙K ); k is the thermal conductivity
(W ∙ m ∙K ); is the velocity component (m ∙ s ); CA is the concentration of specie
A (g ∙ m ); and D is the diffusion coefficient (m ∙ s ).

The Reynolds tensor was modeled using standard k-ε model, which evaluates viscosity
(μτ) from the ratio between turbulent kinetic energy (k) and dissipation of the turbulent
kinetic energy (ε). The ANSYS CFX® software belongs to the Department of Agricultural
and Environmental Engineering of the Federal University of Viçosa, and was employed to
program and simulate the proposed method. The following considerations were assumed:
(a) steady state, (b) incompressible flow, and (c) turbulent flow.
Two computational meshes were generated with different refinement levels using the
CFX Mesh computational program, with the objective of verifying effect of refinement on
local concentration gradients at the spatial levels.

3.6.4. Statistical analysis methods

a) Barns measurements, compost management practices, and herd characteristics
Descriptive statistics (mean, standard deviation, minimum, maximum, and quartiles)
were used to describe individual barn measurements, compost management practices, and
herd characteristics (Microsoft Excel, Microsoft Corp., Redmond, WA).

90 
 
b) WHC, moisture content and bacterial analysis of composts
The data were statistically analysed using Analysis of Variance and Correlation by
using computer software SigmaPlot®, version 11.0, Systat Software®, Inc (Wass, 2009) on
a Personal Computer.

c) Thermal properties as a function of particle size, moisture content, and compaction

degree
Statistical calculations were performed using Sisvar 4.6 (Ferreira, 2003) which is
developed by the Federal University of Lavras, Minas Gerais, Brazil. Average values for
thermal properties in each particle size, moisture content, and compaction degree ranges
were generally used for statistical analyses by ANOVA. For each ANOVA, a particle size,
moisture content, and compaction degree ranges consisted of a treatment. Differences
between groups were tested by Tukey`s test at significance levels of 5%.

d) Validation of the computational model

The statistical parameter used to evaluate measured (reduced model) versus simulated
(CFD model) air velocity and temperature were coefficient of determination (r2) and
Student's t-test. The r2 values indicate how consistently measured versus predicted values
follow a best fit line and can range from zero (no correlation) to 1.0 (perfect correlation).
The data were evaluated in the computer software SigmaPlot®, version 11.0, Systat
Software®, Inc (Wass, 2009) for statistical analysis and data representation. The p-value
was fixed at 5% (α < 0.05).

91 
 
 4 - RESULTS AND DISCUSSION

4.1. Introduction
Results are presented in four main sections:
a) Characterization and description of CBP barns (section 4.2): building layout and
dimensions, bedding pack characteristic, manure management practices, environmental
characteristics, cow characteristics, and producers responses;

b) Evaluate and determine the physical, bacterial, chemical, and thermal properties of
compost bulking materials (section 4.3): compost particle size distribution, particle density,
bulk density, porosity, water holding capacity, chemical properties of composts, bacterial
analysis of composts, and thermal properties of composts;

c) Development of predictive equations for thermal properties of compost as a function

of various combinations of moisture content, pressure, and particle size (section 4.3):

d) Develop computational fluid dynamics (CFD) model of compost barn with alternate
ridge designs and visually demonstrate impact on air flow through structure (section 4.4):

4.2. Characterization and description of CBP barns

4.2.1. Barn structure and layout
The compost dairy barn measurement and layout information collected included: barn
and pack dimensions, feed alley dimensions, waterer location and size, fan and light
description, stock density, sidewall height, as well as other barn specifications. General
barn characteristics for 42 CBP in Kentucky are shown in Appendix A (Table A.1).
Type and orientation of ridge, roof structure and type, roof pitch are showing in
Appendix A (Table A.2).
Ventilation system, cooling fan number, power, size, number of luminaries, and
illumination type are listed in Appendix A (Table A.3).
In general, the first CBP barn was constructed in Kentucky in 2002 and the CBP barn
number has increased to 58 by the end of 2010. The type of barn construction varied,
including barn dimensions, type of ridge, ridge orientation, ventilation system, etc.
CBP barns are a low capital-cost alternative to freestall barns. However, the lower
initial investment for bedded-pack barns may be offset by higher annual operating costs.

92 
 
Barn building costs ranged from $8.000,00 to $400.000,00. Barberg et al. (2007) reported
that the building costs ranged widely depending on the amount of on-farm labor utilized
and amenities added.
Barn ventilation rates are influenced building dimensions, orientation of ridge, sidewall
opening, eave opening, ridge opening, and wind speed. The average compost bedding pack
barn dimensions and were 49.1 m (length) and 21.9 m (width). The most frequent barn
length was between 25 to 50 m (45.2%), 33.3% was between 50 to 75 m, 11.9% was higher
than 75 m, and 9.5% was less than 15 m. The most of frequent barn width was between 15
to 30 m (61.9%), 19.0% was less than 15 m, 16.7% was 30 to 45 m, 2.4% was higher than
75 m. The majority of compost bedded pack barn had areas between 500 to 1000 m
(33.3%). The frequency distribution of compost bedding pack barn area is illustrated in
Figure 4.1.

Figure 4.1 - The frequency distribution of compost bedding pack barns area.

Majority of these barns had feed alleys and driveways. Only 26.2% of the barns, the
cows had free access to pasture, 9.5% was equipped with cow brush. A feed store and
anaerobic lagoon was present around these barns in 52.3%. Headlocks was found in 19% of
the CBP barns.
93 
 
 Cow water intake has a critical effect on her milk productivity and reproductive
performance since it regulates dry matter intake. A cow must have access to water when
needed to satisfy this need. Only 19.0% meet recommendations of 0.10 to 0.15 m per cow
(McFarland et al., 2007). In addition, cows did not have access to water at the barn in 24%
of the compost bedded pack barns. Cows in over 76.0% of the barns were critically short of
waterer space which would limit their milk production.
The recommended feedbunk (stock density) space per cow is 0.03 to 0.06 m / cow
(Smith and Harner, 2001). It was found that 29% of the compost bedded pack barns met
this recommendation with an additional 12% exceeding it. Only 31% of the compost
bedded pack barns had no feedbunks, although most of these had concrete alleyways
leading to nearby feedbunks. Inadequate bunk space was found at 25% of the compost
bedded pack barns. Overstocking at the feed bunk increases competition among cows and
may limit milk production.
Barn orientation will determine the intensity of solar radiation falling on the walls,
bedded and feed alley, and roof of the building. Barns oriented are generally an E-W
orientation in Kentucky, though the recommended orientation is site-specific based upon
natural wind flows. The typical prevailing winds in Kentucky come from the SW in the
summer and from the NW in the winter which would suggest the East-West ridge
orientation as the most effective. The frequency distribution of ridge orientation is listed in
Figure 4.2. Results show that the most frequent orientation is NE-SW (50.0%) with the
next highest being NW-SE (45.2%).
The vertical separation between the eave and ridge openings has a significant impact on
the pressure differences generated by the chimney effect (Chastain, 2000). Therefore, the
roof slope is an important consideration in the design of a naturally ventilated building. All
barns with a gable design are best ventilated with a roof pitch of 3:12 or 4:12 (Smith and
Harner, 2001). There are 42 measurements since 2 barns were found to have different
pitches on each side of the ridge and 2 barns were using hoop structure. Most of barns had
roof pitch of 4:12 (54.8%) and 14.3% had roof pitch of 3:12. Barns with roof pitches higher
than 4:12 or lower than 3:12 (MWPS-7, 2000), can limit the natural ventilation rate per
cow.

94 
 
Figure 4.2 - Frequency distribution of ridge orientation.

Sidewall opening height and eave height are provided to protect the cows from sun,
rain, and wind, thus helping to maintain proper ventilation and animal comfort. A 3.5m
sidewall opening is recommended for barns less than 12.2 m wide while a 4.2 m opening
height is recommended for barns wider than 12.2 m (Graves and Brugger, 1994).
Frequency distribution of sidewall opening height and eave height for the barns in this
study are presented in Figure 4.3. Results show that the most frequent sidewall height is
between 2.0 to 4.0 m (66.7%). In 66.7% of barns, the eave height had more than 4.0 m.
Some of these barns had low walls on outside perimeter to contain the compost and would
reduce the effectiveness of the measured eave height.
A proper eave overhang minimizes the entry of sun, snow, and rain through open
sidewalls. Gay (2009) recommend eave overhang with 1/3 of the sidewall height. Majority
of barns (54.8%) had eave overhand less than 1.0 m. In 35.7% of barns had eave overhang
between 1.0 m and 2.0 m and only 9.5% of barns had eave overhang with 2.0 m or more.
Figure 4.4 is showing the frequency distribution of eave overhang.

95 
 
Figure 4.3 - Frequency distribution of sidewall opening and eave height for the barns.

Figure 4.4 - Frequency distribution of eave overhang in barns.

Ridge design affects air movement by removing warm/hot moist air that collects under
the roof. The design of the ridge can help or inhibit barn ventilation depending on the wind
velocity and direction relative to the ridge line. Characteristics of the CBP barn ridge are
showed in Figure 4.5. The primary types of barn ridge found in the compost bedded pack
barn study were overshot ridges (59%), open ridge with cover (19%), capped ridge (12%),
open ridge without cover (5%), and hoop structure (5%).
96 
 
Figure 4.5 - Frequency distribution of the CBP barn ridge in Kentucky.

The width of the continuous ridge opening controls air exhausting from the barn. As the
width of a barn increases, the ridge opening should also increase. Figure 4.6 summarizes
the ratio of ridge opening to barn width on the compost bedded pack barns visited in
Kentucky. Figure 4.6 notes the present recommended ridge opening width of 0.07 m / 3.05
m of barn width with a minimum width requirement of 0.30 m for barns of less than 12.20
m width. There are two ridge opening values shown in Figure 4.6: a) apparent ridge
opening (0.02 meters / 3.05 meters of building width) and b) effective ridge opening (0.02
meters / 3.05 meters of building width). The apparent opening is the ridge opening width as
seen from inside the barn. The effective opening is the smallest width through which air
flows to the outside of the barn at the ridge.
Figure 4.6 indicates the compost barn ridge measurements as a ratio of the ridge
opening to barn width (meters / 3.05 meters). There were 17% of the barns with no ridge
opening. Based on the apparent ridge opening, 18% had inadequate ridge opening (< 2.5
0.02 meters / 3.05 meters), while 17% had marginal ridge openings (2.5 to < 0.07 meters /
3.05 meters), and 48% equaled or exceeded the recommended ridge opening. However,
based on the effective ridge opening, 53% had inadequate ridge opening, while 20% had
marginal ridge openings, and only 11% equaled or exceeded the recommended ridge
opening. The large majority of barns had ridge opening restrictions that could cause
inadequate natural ventilation that can significantly affect cows, particularly under heat
stress conditions.

97 
 
 (a)

(b)
Figure 4.6 - Ridge opening to barn width ratio for compost bedded pack barns: (a)
Apparent Ridge Opening Compost Barns - 0.02 meters / 3.05 meters width
and (b) Effective Ridge Opening Compost Barns - 0.02 meters / 3.05 meters
width.

The majority of compost bedded pack barns were equipped with box fans (42.8%) for
circulation, 30.9% using natural ventilation, 23.8% using High Volume Low Speed Fans
(HVLS) for circulation, and 2.5% Tunnel ventilation (hoop structures). Buildings that will
use tunnel ventilation require more elaborate ridge openings with some type of closure
system. In this case, it is recommended to consult a competent designer (Agricultural
Engineer) who has experience with cattle shelter ventilation for special situations.  Most of
HVLS fans used had diameter of 7.32 m. The average diameter of box fans was 0.98 m,
ranged from 0.46 m to 1.32 m.

98 
 
 Lighting in compost bedded pack barns was noted at each dairy visited. It was found
that 41% of compost bedded pack barns didn't have lights. High Intensity Discharge (HID)
was the most frequent form of light producers used. This was followed by Fluorescent
(14%), Compact fluorescent (12%), and Incandescent (10%). Total lighting wattage is also
noted. The light average power used in a barn was 124 W, ranged from 13 to 192 W.

4.2.2. Description of structure and layout in each barn

Barn 1 was built in October 2010 and cost $60,000 (Appendix A - Table A.1). The
dimensions of the barn were 51.2 m x 15.2 m, and a 1.48 m high wood sidewall
surrounding the pack. Waterers was outside of the pack area and the dimensions were 2.4
m x 0.8 m. Sidewall was built using seven wood board of 0.05 m x 0.20 m and sidewall
opening of 3.4 m. The feed alley was away from this barn, and the animals were feed in
separate of the building. There were no headlocks, see Figure 4.7. Geographic position of
the Barn 1 were 37° 13' 53'' N of latitude and 86° 20' 20'' W of longitude (Appendix A -
Table A.2) and ridge orientation of 249° East (E). Wood structure was used to build this
barn and metal roof was used to covered. The eave height was 4.87 m and eave overhang
was 1.22 m. The ridge type used was open ridge with cover. Roof pitch measured was 4:12
and ridge cap had dimensions of 0.51 m x 0.51 m, height of 0.51 m and ridge opening of
0.38 m. Drawings about Barn 1 can be seen in Appendix B (Figures 1 and 2). Forced
ventilation system was used with eight box fans (Appendix A - Table A.3). Lamps and cow
brush were not present this barn. One anaerobic lagoon was used close to the barn.

(a) (b)
Figure 4.7 - The main constructive characteristic of CBP barn 1.

Barn 2 was built in March 2008 and cost was $80,000 and including fans was
$110,000. The dimensions of the barn were 76.3 m x 30.3 m, see Appendix A (Table A.1).
99 
 
Concrete wall with 0.6 meter high was used around of the perimeter of the barn 2 and wire
cables were above the wall to prevent cows from falling or jumping out of this building
(Figure 4.8). The feed alley and waterers were in other building, away from barn 2, but the
animals had free access. This barn had sidewall opening of 3.48 m. Waterers was located
along the feed alley. There were no headlocks. A drive-by feed manager was located in
another building close to Barn 2. Geographic position of the Barn 2 were 36° 54' 32'' N
latitude and 86° 16' 42'' W longitude (Appendix A - Table A.2) and ridge orientation of
241° Northeast (NE). Type of construction in Barn 2 was steel-framed pole sheds and
stainless steel wire. The eave height was 3.67 m and eave overhang had dimension of 1.47
m. Metal roof was used to covered and the roof pitch was 4:12. The ridge type used was
open ridge with cover. Ridge cap had dimensions of 0.51 m x 0.51 m, height of 1.01 m and
ridge opening of 0.61 m. The Barn 2 was equipped with three High Volume Low Speed
(HVLS) Fans above the pack (Appendix A - Table A.3). Lighting utilized in barn 2 was
three high intensity discharge, with power of 175 watts each. Cow brush was present this
barn. An anaerobic lagoon was located close to the barn. See Appendix A (Figures 3 and 4)
for more details about CBP barn 2.

(a) (b)
Figure 4.8 - The main constructive characteristic of CBP barn 2.

There was no information about date of construction in Barn 3. The cost of this barn
was approximately $175,000. The barn dimensions were 87.8 m x 16.7 m (Appendix A -
Table A.1). Concrete feed alley and waterers were separated from the pack by low concrete
walls. Concrete wall was used around of the perimeter of this barn and wire cables were
above the wall to prevent cows from falling or jumping onto feed alley. This barn had open
sidewalls without curtains of 2.25 m. Waterers were located adjacent to the concrete wall
separating the pack area and the feed alley (Figure 4.9). In this barn , there were no

100 
 
headlocks. Geographic position in Barn 3 were 36° 25' 56'' N of latitude and 86° 44' 9'' W
of longitude, and ridge orientation of 36° Southwest (SW), as can be seen in Appendix A
(Table A.2). Steel-framed pole sheds and stainless steel wire were used to built this barn
and metal roof was used to covered. The eave height was 3.18 m and eave overhang was
2.35 m. The ridge type used was overshot, roof pitch measured was 4:12, overlap of the
ridge of 0.9 m, and ridge opening of 0.76 m. Drawings of Barn 3 can be seen in Appendix
B (Figures 5 and 6). The Barn 3 was equipped with seventeen box fans (Appendix A -
Table A.3). Lighting utilized was two high intensity discharge. Cow brush was and
anaerobic lagoon were not present around this barn.

(a) (b)
Figure 4.9 - The main constructive characteristic of CBP barn 3.

Barn 4 was built October 2010 and the total cost was $10,900, is detailed in Appendix
A (Table A.1). The dimensions of the barn were 51.5 m x 9.1 m. Waterers had dimensions
with of 2.4 m x 0.8 m. The feed alley was away from facility, and the animals were feed in
separate of the building (Figure 4.10). In this barn, the sidewall opening had 1.22 m. The
ridge orientation of the Barn 4 was 80° Southwest (SW) and the geographic position were
36° 42' 10'' N latitude and 86° 40' 34'' W longitude (Appendix A - Table A.2). Thirty
treated wood posts were used to build this barn. Posts had dimensions of 0.15 m x 0.15 m
and spaced of 1.83 m from each other. Two treated boards (0.05 m x 0.15 m) were used in
each sidewall. Tubular steel arches fastened to the tops and sides of the posts to form a
hooped roof were used and white poly vinyl tarp used to covered. The eave height was 4.57
m. Left-side eave overhang of this barn had width of 1.37 m and eave overhang over feed
area had width of 4.82 m. Only natural tunnel ventilation was used in Barn 4. Lightings and

101 
 
cow brush were not present this barn (Appendix A - Table A.3). Anaerobic lagoon was not
present around this barn. Drawings of Barn 4 can be seen in Appendix B (Figures 7 and 8).

(a) (b)
Figure 4.10 - The main constructive characteristic of CBP barn 4.

Barn 5 was built in 2003 and there was no information about costs of construction
(Appendix A - Table A.1). The dimensions of the barn were 48.7 m x 15.2 m. There were
no information about waterers. The dimensions of feed alley of 3.41 m x 48.7 m and one
concrete curb was separating the pack and the feed alley. This concrete wall had a height of
0.3 m and a wood fence was installed above this concrete wall to prevent cows from
jumping onto feed alley. A drive‐by feed manger was located on the left side of the barn.
Five wood boards (0.05 m x 0.10 m) were used in sidewall (Figure 4.11). Barn 5 had a
sidewall opening of 3.53 m. Geographic position founded in Barn 5 were 37° 44' 25'' N
latitude and 85° 44' 55'' W longitude and ridge orientation of 31° Southwest (SW) is
detailed in Appendix A (Table A.2). Wood pole and stainless steel plate were used to built
this barn and metal roof was used to covered. The eave height was 3.66 m. This barn had
two different widths in the eave overhang of 1.06 m and 0.76 m. The ridge type used was
overshot, roof pitch measured was 4:12, overlap of the ridge of 1.11 m, and ridge opening
of 0.76 m. Barn 5 was used natural ventilation system (Appendix A - Table A.3). There
were no information about type of lighting. Headlocks and cow brush were not present in
this barn. An anaerobic lagoon was not present around this barn. Drawings about Barn 5
can be seen in Appendix B (Figures 9 and 10).

102 
 
 (a) (b)
Figure 4.11 - The main constructive characteristic of CBP barn 5.

Barn 6 was built in June 2010 and there was no information about costs of construction
(Appendix A - Table A.1). The dimensions of the barn 6-A were 30.5 m x 16.4 m and barn
6-B were 33.0 m x 26.2 m. This barn had a wood fence wall separating the pack and feed
alley (Figure 4.12). Two concrete waterers were used on the pack area. Sidewall were
open during winter. The sidewall opening had 2.73 m. Geographic position of the Barn 6
were 37° 2' 27'' N of latitude and 85° 11' 2'' W of longitude (Appendix A - Table A.2). The
ridge orientation of this barn was 145° Northwest (NW). Wood structure was used to build
this barn and metal roof was used to covered. The eave height of barn 6-A and 6-B were
3.18 m and 4.27 m, respectively. The eave overhang was 0.61 m. The roof pitch measured
was 4:12. This barn had two different ridge type: capped ridge (Barn 6-A) and overshot
(Barn 6-B). The ridge opening in barn 6-B was 0.35 m, and overlap of the ridge of 0.83 m.
Twelve box fans were used in ventilation system (Appendix A - Table A.3). Five
incandescent lightings were used to illumination. Cow brush was not present in this barn.
Anaerobic lagoon was not present around this barn. Drawings of Barn 6 can be seen in
Appendix B (Figures 11 and 12).

Box fan

(a) (b)
Figure 4.12 - The main constructive characteristic of CBP barn 6.
103 
 
 Barn 7 was built in November 2008 and cost $50,000 (Appendix A - Table A.1). The
barn dimensions were 36.6 m x 16.3 m. Waterers was around of this barn and had
dimensions of 2.0 m x 0.6 m. This building had sixty headlocks along a covered drive‐by
feed manger. Sidewall opening had 3.65 m. The cows had a free access to pasture. A
tubular steel gates were used to access to another building (Figure 4.13). Geographic
position of the Barn 7 were 37° 46' 27'' N of latitude and 84° 38' 52'' W of longitude and
the ridge orientation of this barn was 40° South-west (SW), as can be seen in Appendix A
(Table A.2). Wood structure was used to build this barn and metal roof was used to
covered. The eave height was 3.65 m and eave overhang was 0.91 m. The ridge type used
was overshot and the roof pitch measured was 4:12. The ridge opening of 0.45 m and
overlap of the ridge of 0.64 m. Metal roof was used to covered and the roof pitch measured
was 4:12. Barn 7 was equipped with two High Volume Low Speed (HVLS) fans with 7.32
m of diameter and power of 1491.4 watts (Appendix A - Table A.3). Six High Intensity
Discharge (HID) lamps with 175 watts each were used to illumination. Cow brush was not
present in this barn. Anaerobic lagoon was not present around this barn. Details about
Barn 7 can be seen in Appendix B (Figures 13 and 14).

(a) (b)
Figure 4.13 - The main constructive characteristic of CPB barn 7.

Barn 8 was built in January 2011 and cost to built without concrete was $60,000. The
building dimensions of this barn were 65.8 m x 29.2 m, as can be seen in Appendix A
(Table A.1). A covered drive‐by feed manger was located on the center of the barn (Figure
4.14). This barn had an indoor concrete feed alley and 0.91 m high concrete wall around of
the barn. The sidewall opening had 2.7 m. This barn had installed guardrail fence above the
concrete wall to prevent cows from jumping onto feed alley or outside of the barn. The two
plastic waterers were outside of pack area. Geographic position in Barn 8 (Appendix A -
104 
 
Table A.2) were 36° 54' 55'' N of latitude and 86° 15' 1'' W of longitude and ridge
orientation of 210° Northwest (NW). Galvanized steel H-beam columns were used and
metal roof was used to covered of this barn. The eave height was 3.65 m. The ridge type
used was open ridge with cover and the roof pitch was 4:12. Ridge cap had dimensions of
0.48 m x 0.48 m, height of 0.61 m and ridge opening of 0.61 m. The Barn 8 was equipped
with 4 box fans over pack area (Appendix A - Table A.3). Lamps were not present in this
barn. One tractor was used to stirring the bedding compost using cultivators on skid loader.
Headlocks and cow brush were present this barn. One anaerobic lagoon was present
around this barn. See Appendix B (Figures 15 and 16) for more details about CBP barn 8.

(a) (b)
Figure 4.14 - The main constructive characteristic of CBP barn 8.

Barn 9 was built in September 2010 and cost of construction was $120,000 (Appendix
A - Table A.1). The building dimensions of this barn were 76.3 m x 27.6 m. This barn had
three waterers on the concrete scrape alley and dimensions of 3.1 m x 0.6 m, 2.1 m x 0.6 m,
and 1.2 x 0.6 m, respectively. A concrete wall separating the pack and the feed alley had
three walkways for cows to access the pack (Figure 4.15). Geographic position of the Barn
9 were 36° 54' 55'' N latitude and 86° 40' 34'' W longitude and the ridge orientation was 75°
Northwest (NW), as can be seen in Appendix A (Table A.2). This barn was built using
wood pole. The eave height was 4.57 m. The eave overhang had two different width of
0.30 m and 0.61 m. The ridge type used was overshot. Roof pitch measured was 4:12,
overlap of the ridge of 0.50 m and ridge opening of 0.46 m. Barn 9 equipped with twenty
eight box fans with 735.5 W each (Appendix A - Table A.3). Thirty high intensity
discharge (HID) lamps were used with 175 w each. Headlock was present in this barn. This
barn was not equipped with cow brush. Anaerobic lagoon was present around this barn.
Details about CBP barn 9 can be seen in Appendix B (Figures 17 and 18).

105 
 
 HID lamps 

Box fan 

(a) (b)
Figure 4.15 - The main constructive characteristic of CBP barn 9.

Barn 10 was built in Fall 2009 and cost $120,000 including fans and lamps (Appendix
A - Table A.1). The building and pack dimension of this barn were 61.0 m x 30.5 m.
Waterers was not present in this barn. A tubular steel gates were used to access to another
building (see Figure 4.16). The sidewall was built using three guardrail. In this barn, the
sidewall opening had 3.7 m. Geographic position were 36° 54' 55'' N latitude and 86° 15'
01'' W longitude (Appendix A - Table A.2). The ridge orientation was 349° Southeast (SE).
This barn was built using wood board. The eave height was 4.87 m and eave overhang was
0.76 m. The ridge type used was overshot and metal roof was used to covered this barn.
This barn had two different roof pitch of 3:12 and 2 1/2 :12. The ridge opening of 0.61 m,
and overlap of the ridge of 0.91 m. Five high intensity discharge (HID) lamps were used
with 70 watts each. Headlocks and cow brush were not present in this barn. Anaerobic
lagoon was around this barn. Barn 10 was equipped with fifteen box fans with 735.5 watts
each (Appendix A - Table A.3). Drawings of Barn 10 can be seen in Appendix B (Figures
19 and 20).

(a) (b)
Figure 4.16 - The main constructive characteristic of CBP barn 10.

106 
 
 Barns 11 and 12 were built in the same time, using the same constructive characteristics
and costs of construction. These barns were built in February 2010 and cost $150,000 each
with fans and labor. The building dimension of this barn were 91.5 m x 39.0 m, as can be
seen in Appendix A (Table A.1). Two different plastic waterers were present in pack area
with dimensions of 1.8 m x 0.6 m and 3.6 m x 0.5 m. The waterers were located adjacent to
the concrete wall separating the pack area and the feed alley. A covered drive-by feed
manger had dimension of 2.74 m. The wall separating the pack and the feed alley had five
walkways for cows to access the pack. This wall was built with concrete and had 1.10 m
high and one steel fence was installed on this wall to prevent cows from jumping onto the
feed alley (Figure 4.17). These barn had open sidewalls without curtains with 3.75 m. One
elevated walkway has been installed in the middle of each barn to help the handling of
cows. Tubular steel gates were used to access to another building.
Geographic positions of these barn were 37° 44'58'' N latitude and 84° 25' 53'' W
longitude and the ridge orientation was 23° Southwest (SW), as can be seen in Appendix A
(Table A.2). The type of these barns construction was wood and steel-framed pole. The
ridge type used was overshot with metal roof used to covered in each barn. These barns had
a different roof pitch configuration of 3:12 and 2 1/2 :12, overlap of the ridge of 0.61 m and
ridge opening of 0.46 m. Details about barn 11 and barn 12 can be seen in Appendix B
(Figures 21 and 22).
Five High Volume Low Speed (HVLS) fans with different diameter (three fans of 6.10
m and two fans of 7.32 m) were used under of pack area (Appendix A - Table A.3). Each
barn was equipped with six fluorescent (FLL) lamps of 192 watts. Headlocks and cow
brush were not present in this barn. Anaerobic lagoon was not present around this barn.

HVLS fan 

(a) (b)
Figure 4.17 - The main constructive characteristic of CBP barns 11 and 12.

107 
 
 Barns 13 and 14 were built in the same time, using the same constructive characteristics
and costs of construction. These barns were built in October 2010 and cost $85,000 each
barn, $140,000 for concrete, and total $400,000 without fans (Appendix A - Table A.1).
The barns dimensions were 61.0 m x 40.8 m. Barn 13 was 12.2 m away from Barn 14.
Eight plastic waterers were present in each barn and had dimensions of 1.8 m x 0.6 m. The
waterers were located adjacent to the concrete wall separating the pack area and the feed
alley. The drive-by feed manger was located in the middle of the barn (Figure 4.18). The
driveway had 4.87 m width. The concrete wall separating the pack and the feed alley had
five walkways for cows to access the pack. These barn had open sidewalls without curtains
with 2.96 m. Tubular steel gates were used to access to another building.
Geographic positions of these barns were 36° 59' 57'' N latitude and 85° 42' 38'' W
longitude (Appendix A - Table A.2). The ridge orientation was 329° Southeast (SE). These
barns were built using concrete floor and wood board. The eave height was 3.96 m and
eave overhang was 1.22 m. The ridge type used was open ridge without cover. The ridge
opening and height were 0.91 m. Metal roof was used to covered in each barn and the roof
pitch measured was 3:12. Barn 13 and 14 utilized natural ventilation (Appendix A - Table
A.3). Each barn was equipped with ten compact fluorescent lamps of 32 watts. Headlocks
was present in these barns. Anaerobic lagoon was present around these barns. Drawings of
CBP barns 13 and Barn 14 can be seen in Appendix B (Figures 23 and 24).

(a) (b)
Figure 4.18 - The main constructive characteristic of CBP barns 13 and 14.

Barn 15 was built in April 2010 and there were no information about costs of
construction (Appendix A - Table A.1). The barns and pack dimensions were 18.3 m x 18.3
m. The waterers was outside of the pack area and the dimensions were 3.6 m x 0.5 m
(Figure 4.19). Feed alley was located on the outside of the barn on the Northwest side.

108 
 
This barns had a 0.58 m high wall surrounding the pack. This barn had open sidewalls
without curtains with 3.96 m. The cows had free access to pasture. Geographic positions of
Barn 15 was 38° 10' 18'' N latitude and 85° 13' 43'' W longitude (Appendix A - Table A.2).
The ridge orientation was 334° Southwest (SW). These barns were built using concrete
floor, wood, and steel plate. The ridge type used was Capped ridge and metal roof was used
to covered. The end wall had a opening on the top. The eave height was 3.96 m and the
roof pitch of 4:12. More details see Appendix B (Figures 25 and 26). Barn 15 was equipped
with one High Volume Low Speed (HVLS) fans under of pack area (Appendix A - Table
A.3). These barns were not equipped with Lights. Headlocks and cow brush were not
present in this barn. An anaerobic lagoon was present around this barn.

HVLS fan

(a) (b)
Figure 4.19 - The main constructive characteristic of CBP barn 15.

There was no information about date and costs of construction of Barn 16. This barn
had dimensions of 14.6 m x 13.9 m, as can be seen in Appendix A (Table A.1). One
rectangular concrete waterers, 2.4 m x 0.6 m, was outside of pack area, adjacent to drive-by
feed manger (Figure 4.20). Opens sidewalls (2.22 m) without curtains were present in this
barn. Barn 16 had a concrete sidewall surrounds the pack and a wall separating the pack
from feed alley of 0.3 m high. Geographic positions of Barn 16 was 37° 8' 39'' N latitude
and 85° 18' 14'' W longitude (Appendix A - Table A.2). The ridge orientation was 243°
Northeast (NE). The type of these barns construction was concrete wall and wood in the
structure. The eave height was 3.50 m. This barn had a eave overhang of 1.65 m and eave
overhang over feed area of 1.83 m. The ridge type used was capped ridge and metal roof
was used to covered. This barn had a roof pitch of 5:28. Drawings of Barn 16 can be seen
in Appendix B (Figures 27 and 28). Barn 16 was equipped with three box fans under of
pack area and one box fan under feed alley (Appendix A - Table A.3). This barns was

109 
 
equipped with six incandescent lights of 100 watts each. Headlocks and cow brush were
not present in this barn. An anaerobic lagoon was present around this barn.

(a) (b)
Figure 4.20 - The main constructive characteristic of CBP barn 16.

There was no information about date of construction of Barn 17. The cost of
construction was $22,000 without fans (Appendix A - Table A.1). The barn dimensions
were 37.8 m x 13.7 m. Sidewalls were open (3.76 m) without curtain and two horizontal
stainless steel fence were used around this barn. Two plastic waterers were around this barn
and dimensions were 1.5 m each. Two tubular steel gates were used to access to another
building (Figure 4.21). Geographic position were 37° 1' 23'' N latitude and 85° 9' 29'' W
longitude (Appendix A - Table A.2). The ridge orientation was 240° Southeast (SE). These
barns were built using concrete floor and wood pole. The eave height was 4.92 m and eave
overhang was 1.32 m. The ridge type used was overshot and metal roof was used to
covered in each barn. The roof pitch measured was 4:12, ridge opening of 0.34 m, and
overlap of the ridge of 0.91 m. Details about barn 17 can be seen in Appendix B (Figures
29 and 30). Barn 17 was equipped with six box fans along of pack (Appendix A - Table
A.3). Eight incandescent lights of 100 watts each were used. Headlocks were present in this
barn. An anaerobic lagoon was present around this barn.

(a) (b)
Figure 4.21 - The main constructive characteristic of CBP barn 17.
110 
 
 Barn 18 was built in November 2008 and cost $90,000 (Appendix A - Table A.1). The
barn dimensions were 61.0 m x 24.4 m. The concrete driveway was outside of barn and had
width of 3.95 m. This barn had a small concrete curb and tubular steel gates was above this
curb, separating the pack and the feed alley (Figure 4.22). One waterers was outside of the
pack with dimension of 3.7 m x 0.6 m. This barn had two feed alley with dimensions of
30.6 m x 5.33 m and 61.0 m x 3.95 m. Sidewall was built using seven wood board of 0.05
m x 0.30 m. Geographic positions of this barn were 37° 29' 23'' N of latitude and 86° 51'
58'' W of longitude and ridge orientation of 90° South-west (SW), as can be seen in
Appendix A (Table A.2). This barn was built using wood posts, wood boards, and metal
roof was used to covered. The eave height was 4.9 m and eave overhang was 1.06 m. The
ridge type used was overshot, roof pitch measured was 3 1/2 : 12, overlap of the ridge of
0.23 m and ridge opening of 0.16 m. The barn 18 was equipped with twenty box fans and
eight fluorescents lamps of 192 W (Appendix A - Table A.3). Cow brush and headlock
were not present this barn. One anaerobic lagoon was used close to the barn. Drawings of
Barn 18 can be seen in Appendix B (Figures 31 and 32).

Box fan 

(a) (b)
Figure 4.22 - The main constructive characteristic of CBP barn 18.

Barn 19 was built in October 2006 and costs of construction was approximately $8,000
(Appendix A - Table A.1). This barn had dimension of 21.3 m x 14.0 m. The waterers and
feed alley were outside of this barn (see Figure 4.23). The cows had free access to pasture.
This building had not sidewall opening. Geographic positions of Barn 19 was 37° 34' 15'' N
latitude and 86° 08' 36'' W longitude (Appendix A - Table A.2). The ridge orientation was
28° Northwest (NW). This barns was built using wood board and metal plate. The ridge
type used was Capped ridge and metal roof was used to covered. The end wall had one

111 
 
opening on the top. The eave height was 4.26 m, eave overhang was 0.30 m, and roof pitch
of 3 1/2 : 12. Drawings of Barn 19 can be seen in Appendix B (Figures 33 and 34). Barn 19
was equipped with four box fans of 0.91 m (diameter). Two compact fluorescent lamps of
32 watts were used (Appendix A - Table A.3). Cow brush were not present this barn. One
anaerobic lagoon was used close to the barn.

(a) (b)
Figure 4.23 - The main constructive characteristic of CBP barn 19.

Barn 20 was built in 2003. There was no information about cost of construction
(Appendix A - Table A.1). The dimensions of this barn were 21.9 m x 18.0 m. Two
waterers, 0.70 m x 0.64 m, were on the pack area (Figure 4.24). One short concrete curb
and two steel pole were separating the pack area and the feed alley. This barn had a
covered feed alley width of 3.66 m and one closed sidewall with metal panels was used.
Geographic position were 37° 46' 25'' N latitude and 85° 21' 52'' W longitude and the
ridge orientation was 15° Southwest (SW), as can be seen in Appendix A (Table A.2). This
barn was built using wood pole, metal plate, and galvanized steel H-beam was used under
the roof metal. The eave height was 4.11 m and eave overhang of 0.23 m. The ridge type
used was overshot, roof pitch measured was 1 1/2 : 12 and 2 : 12, overlap of the ridge of
0.33 m and ridge opening of 0.23 m.
Barn 20 was equipped with six box fans of 0.91 m diameter and four box fans of 1.22
m diameter (Appendix A - Table A.3). Lamps, headlock and cow brush were not present
this barn. Anaerobic lagoon was not present around this barn. Details about barn 20 can be
seen in Appendix B (Figures 35 and 36).

112 
 
 (a) (b)
Figure 4.24 - The main constructive characteristic of CBP barn 20.

Barn 21 was built in June 2006 and costs of construction was $30,000 with fans. The
barn dimensions were 36.6 m x 15.2 m. Sidewalls were open without curtain (Figure 4.25).
Eleven wood post were uses on each side and spaced of 3.66 m from each other. Plastic
waterers was situated in another barn and the cows had a free access to the building. This
barn was equipped with two tubular steel gates. This building had sidewall opening of 4.57
m. Geographic position were 37° 20' 49'' N latitude and 85° 21' 41'' W longitude (Appendix
A - Table A.2). The ridge orientation was 233° Northeast (NE). This barn was built using
concrete sidewall curb, wood pole, and metal roof was used to covered. The eave height
was 4.57 m and eave overhang was 1.67 m. The ridge type used was overshot, roof pitch
measured was 4:12, ridge opening of 0.91 m, and overlap of the ridge of 0.91 m. Drawings
of Barn 21 can be seen in Appendix B (Figures 37 and 38). Barn 21 was equipped with six
box fans along of pack of 0.91 m diameter and two box fans of 1.22 m diameter above of
pack area, as can be seen in Appendix A (Table A.3). This barn was equipped with three
high intensity discharge (HID) lamps of 175 watts. Headlocks and cow brush were not
present in this barn. Anaerobic lagoon was not present around this barn.

(a) (b)
Figure 4.25 - The main constructive characteristic of CBP barn 21.
113 
 
 Barn 22 was built in 2002 and costs of construction was $40,000 (Appendix A - Table
A.1). The barns dimensions were 60.9 m x 18.9 m. The waterers was outside of pack area,
but located adjacent in the endwall. The waterers dimensions were with 1.8 m x 0.6 m. This
barn had the same feed alley that the barn 23 and the cows had a free access (Figure 4.26).
This feed alley had width of 5.48 m. This barns had a 0.35 m high concrete wall both side
of the pack. Sidewalls were open without curtain and hot wires were used above the
sidewall curb height. Geographic positions of Barn 22 was 37° 03' 58'' N latitude and 86°
18' 33'' W longitude (Appendix A - Table A.2). The ridge orientation was 35° Northwest
(NW). This barn was built using concrete floor, wood, and steel plate. Two zinc gutter
were used on both side and two PVC pipe were carry the rainwater to a concrete box. The
eave height was 4.33 m and eave overhang was 0.60 m. The ridge type used was Capped
ridge, roof pitch of 4:12, and metal roof was used to covered. Natural ventilation system
was used in the Barn 22 (Appendix A - Table A.3). Twelve high intensity discharge lamps
(150 watts each) were used above the pack area. Headlocks and cow brush were not present
in this barn. Anaerobic lagoon was not present around this barn. Drawings of Barn 22 can
be seen in Appendix B (Figures 39 and 40).

Barn 22
Barn 23

(a) (b)
Figure 4.26 - The main constructive characteristic of CBP barn 22.

Barn 23 was built in 2007 and costs of construction was $85,000 (Appendix A - Table
A.1). The barns dimensions were 60.9 m x 15.2 m. A plastic waterers of 3.6 m x 0.6 m was
outside of pack area (Figure 4.27). Sidewalls were open without curtain and three wood
boards were used above the sidewall curb. The sidewall curb height of 0.35 m. Geographic
positions of Barn 23 was 37° 03' 58'' N latitude and 86° 18' 34'' W longitude (Appendix A -
Table A.2). The ridge orientation was 305° Southeast (SE). This barn was built using
concrete floor, wood, and metal plate. Two zinc gutter were used on both side and PVC

114 
 
pipe were carry the rainwater to a concrete box. The eave height was 4.33 m and eave
overhang was 0.60 m. The ridge type used was Capped ridge, roof pitch of 4:12, and metal
roof was used to covered. Details about barn 23 can be seen in Appendix B (Figures 41 and
42). Barn 23 was natural ventilation system (Appendix A - Table A.3). Six high intensity
discharge lamps (150 watts each) were used on this barn. Headlocks and cow brush were
not present in this barn. Anaerobic lagoon was not present around this barn.

(a) (b)
Figure 4.27 - The main constructive characteristic of CPB barn 23.

Barn 24 was built in 2009 and costs of construction was $35,000 (Appendix A - Table
A.1). The barns dimensions were 30.5 m x 18.9 m. Plastic waterers had dimensions of 1.8 m
x 0.6 m and was outside of pack area. Curtain was used in one sidewalls during the winter,
see Figure 4.28. This barn had sidewall curb height of 0.35 m. Geographic positions of
Barn 24 was 37° 03' 58'' N latitude and 86° 18' 33'' W longitude (Appendix A - Table A.2).
The ridge orientation was 286° Southeast (SE). This barn was built using concrete floor,
wood, and metal plate. Two zinc gutter were used on both side. The eave height was 3.05
m and eave overhang was 0.60 m. The ridge type used was Capped ridge, roof pitch of
4:12, and metal roof was used to covered. Barn 24 was using natural ventilation system
(Appendix A - Table A.3). This barn was not equipped lamps. Headlocks and cow brush
were not present in this barn. Anaerobic lagoon was not present around this barn. See
Appendix B (Figures 43 and 44) for more details about CBP barn 24.

115 
 
 (a) (b)
Figure 4.28 - The main constructive characteristic of CBP barn 24.

Barn 25 was built in February 2009 and cost of building was $204,000. The dimensions
of this barn were 51.2 m x 38.4 m (Appendix A - Table A.1). Concrete wall (0.6 meter
high) and three wood board were used around of the perimeter of this barn (Figure 4.29).
Concrete feed alley and waterers were separated from the pack by low concrete wall. The
feed alley had width of 3.35 m. A drive-by feed manager was located outside of this
building. Geographic positions of the Barn 25 were 37° 36' 44'' N latitude and 85° 15' 25''
W longitude (Appendix A - Table A.2) and ridge orientation of 291° Southeast (SE). This
barn was built using concrete floor and wood posts and board. The eave height was 4.87 m.
The ridge type used was open ridge with cover, metal roof was used to covered and the
roof pitch was 3:12. Ridge cap had dimensions of 1.12 m x 1.12 m, height of 0.32 m and
ridge opening of 1.58 m. Details about barn 25 can be seen in Appendix B (Figures 45 and
46). The Barn 25 was equipped with three HVLS fans above the pack area (Appendix A -
Table A.3). Twenty four high intensity discharge, 175 watts each, were used in this barn.
Cow brush was not present in this barn. This barn was equipped with headlocks along of
the feed alley. Anaerobic lagoon was not around of the building.

HVLS fan 

(a) (b)
Figure 4.29 - The main constructive characteristic of CBP barn 25.

116 
 
 Barn 26 was built in June 2008 and cost of construction was $85,000, as can be seen in
Appendix A (Table A.1). This barn had dimensions of 60.9 m x 24.1 m. This barn had
three waterers on the pack area and dimensions of 0.61 m x 0.46 m, 1.12 m x 0.46 m, and
0.76 m x 0.46 m. The feed alley had width of 4.26 m and a steel cable and hot wires was
used above the sidewall curb height (Figure 4.30). Concrete wall was separating the pack
and the feed alley in three walkways. Curtains were used in one sidewall of the building.
Geographic position of the Barn 26 were 37° 00' 05'' N latitude and 87° 55' 58'' W
longitude and the ridge orientation was 265° Southeast (SE), as can be seen in Appendix A
(Table A.2). This barn was built using concrete floor, wood, and metal plate. The eave
height was 4.06 m. The eave overhang of this barn had width of 1.22 m. The ridge type
used was overshot, roof pitch measured was 3 1/2:12, overlap of the ridge of 1.60 m, and
ridge opening of 0.91 m. Metal roof was used to covered this building and insulation was
placed under this metal roof. Drawings of Barn 26 can be seen in Appendix B (Figures 47
and 48). Barn 26 was equipped with three High Volume Low Speed (HVLS) fans of 7.32
m and 1493.2 watts. (Appendix A - Table A.3). Two high intensity discharge lamps of 150
watts each were used. This barn has equipped with headlock, but cow brush was not
present. Anaerobic lagoon was present around this barn.

(a) (b)
Figure 4.30 - The main constructive characteristic of CBP barn 26.

Barn 27 was built in September 2010 and cost of construction was $65,000 (Appendix
A - Table A.1). The barn dimensions were 36.6 m x 18.0 m. This barn had a concrete feed
alley width of 1.73 m. Small concrete wall and wire mesh was separating the pack area and
the feed alley (Figure 4.31). Two plastic waterers had dimensions of 2.44 m x 0.38 m. The
feed alley width of 3.66 m. Sidewall was built using concrete curb and wire mesh. Curtains
were used during the winter and the endwall was closed using metal plate. Two zinc gutter

117 
 
were used on both side. Geographic positions of the Barn 27 were 36° 54' 23'' N of latitude
and 87° 20' 34'' W of longitude and ridge orientation of 330° Southeast (SE), as can be seen
in Appendix A (Table A.2). This barn was built using concrete floor, wood, and metal
plate. The eave height was 2.61 m and eave overhang was 0.91 m. The ridge type used was
overshot. Roof pitch measured was 4:12, overlap of the ridge of 0.96 m and ridge opening
of 0.40 m. Barn 27 was used natural ventilation system. Eight compact fluorescents lamps
were used of 13 watts each (Appendix A - Table A.3). Headlock and cow brush were not
present this barn. Anaerobic lagoon was not present around this barn. Drawings of Barn 27
can be seen in Appendix B (Figures 49 and 50).

(a) (b)
Figure 4.31 - The main constructive characteristic of CBP barn 27.

Barn 28 was built in August 2007 and costs of construction was $50,000 (Appendix A -
Table A.1). The barn dimensions were 36.6 m x 15.2 m. One sidewall was closed with
curtain (Figure 4.32). One plastic waterers was situated in another building and the cows
had a free access. This barn was closed to the labor. Two zinc gutter were used on both
side. Geographic position were 37° 00' 49'' N latitude and 87° 22' 14'' W longitude
(Appendix A - Table A.2). The ridge orientation was 37° Southwest (SW). This barn was
built using concrete floor, wood pole, metal panels, and metal roof was used to covered.
The eave height was 3.86 m and eave overhang was 0.81 m. The ridge type used was
overshot, roof pitch measured was 4:12, ridge opening of 0.35 m, and overlap of the ridge
of 0.60 m. Details about barn 28 can be seen in Appendix B (Figures 51 and 52). Barn 28
was equipped with six box fans along of pack of 1.22 m diameter and 1118.5 watts power,
as can be seen in Appendix A (Table A.3). Fourteen compact fluorescent lamps of 13 watts
were used. Headlocks and cow brush were not present in this barn. Anaerobic lagoon was
present around this barn.

118 
 
 (a) (b)
Figure 4.32 - The main constructive characteristic of CBP barn 28.

Barns 29 was built in February 2006 and cost of $105,000 with concrete. The building
dimension of this barn were 60.9 m x 23.7 m, as can be seen in Appendix A (Table A.1).
Three plastic waterers of 3.66 m x 0.56 m were outside of the pack area. The waterers were
located adjacent to the concrete wall separating the pack area and the feed alley. This barn
had width of feed alley of 4.26 m. The concrete curb was separating the pack and the feed
alley had three walkways for cows to access the pack area (Figure 4.33). Geographic
positions of Barn 29 were 36° 35' 37'' N latitude and 87° 09' 20'' W longitude (Appendix A
- Table A.2). The ridge orientation was 284° Southeast (SE). This barn was built using
concrete floor, wood pole, metal panels, and metal roof was used to covered. The eave
height was 4.11 m and eave overhang was 1.22 m. The ridge type used was overshot, roof
pitch measured was 3:12, ridge opening of 0.91 m, and overlap of the ridge of 1.52 m.
Drawings of Barn 29 can be seen in Appendix B (Figures 53 and 54). Barn 29 was
equipped with three HVLS fans (7.32 m) with 2237.1 watts each (Appendix A - Table
A.3). This barn was not using lamps. Headlocks was not present in this barn. This barn was
equipped with cow brush. Anaerobic lagoon was not present around this barn.

HVLS fan

Cow brush 

(a) (b)
Figure 4.33 - The main constructive characteristic of CBP barn 29.
119 
 
 Barn 30 was built in December 2009 and cost of construction was $35,000 for building
and $120,000 for filling (Appendix A - Table A.1). This barn had dimensions of 45.7 m x
23.4 m. This barn had two waterers on the pack area and dimensions of 3.05 m x 0.56 m.
The feed alley had width of 3.05 m (Figure 4.34). Concrete wall was separating the pack
and the feed alley in two walkways. Geographic position of the Barn 30 were 36° 38' 53'' N
latitude and 87° 09' 35'' W longitude and the ridge orientation was 253° Northeast (NE),
see Appendix A (Table A.2). This barn was built using concrete floor, wood, and metal
plate. The eave height was 4.70 m and eave overhang of 1.22 m. The ridge type used was
overshot, roof pitch measured was 3 1/2:12, overlap of the ridge of 1.22 m, and ridge
opening of 0.61 m. Metal roof was used to covered this building and insulation was placed
under this metal roof. Details about barn 30, see Appendix B (Figures 55 and 56). Barn 30
was equipped with two HVLS fans of 7.32 m with 1493.2 watts each. This barn was not
equipped with lamps (Appendix A - Table A.3). This barn has equipped with cow brush,
but headlock was not present. Anaerobic lagoon was present around this barn.

HVLS fan 

(a) (b)
Figure 4.34 - The main constructive characteristic of CBP barn 30.

Barn 31 was built in November 2009 and cost of building was $207,000. The building
dimensions of were 45.7 m x 43.0 m (Appendix A - Table A.1). Concrete curb of 0.2 m
high was used around of the perimeter of this building and one wire cable was used above
this wall (Figure 4.35). Concrete feed alley was separated from the pack area by low
concrete wall. The width of feed alley was 3.35 m. A drive-by feed manager was located
outside of this barn. This barn was away another building of 7.31 m. Two zinc gutter were
used on both side. Geographic positions of the Barn 31 were 36° 40' 52'' N latitude and 86°
52' 37'' W longitude (Appendix A - Table A.2) and ridge orientation of 345° Southeast

120 
 
(SE). This barn was built using concrete floor, galvanized steel H-beam columns, and metal
roof was used to covered of this barn. The eave height was 2.81 m and eave overhang had
dimension of 1.83 m. The ridge type used was open ridge with cover, metal roof was used
to covered and the roof pitch was 4:12. Ridge cap had dimensions of 0.23 m x 0.23 m,
height of 0.20 m and ridge opening of 0.40 m. Drawings of barn 31 can be seen in
Appendix B (Figures 57 and 58). The Barn 31 was using natural ventilation system
(Appendix A - Table A.3). This barn was not equipped with lamps. Cow brush and
headlocks were not present in this barn. Anaerobic lagoon was around of the building.

(a) (b)
Figure 4.35 - The main constructive characteristic of CBP barn 31.

Barn 32 was built in December 2007 and cost of construction was $30,000. This barn
dimensions were 30.5 m x 18.3 m. The waterers had dimensions of 3.66 m x 0.56 m and
was outside of the pack area (Appendix A - Table A.1). The feed alley was another
building. Concrete wall was used around of the perimeter of this building and curtains were
used above this concrete wall during the winter (Figure 4.36). This barn had one gate on
each endwall. Geographic position of the Barn 32 were 36° 54' 22'' N latitude and 87° 06'
26'' W longitude and the ridge orientation was 90° Southwest (SW), see Appendix A
(Table A.2). This barn was built using concrete floor, wood, metal panels were used in the
endwall, and metal roof was used to covered this building. The eave height was 4.14 m.
The eave overhang of this barn had width of 1.22 m. The ridge type used was overshot,
roof pitch measured was 4:12, overlap of the ridge of 1.22 m, and ridge opening of 0.76 m.
Barn 32 was equipped with two HVLS fans of 7.32 m with 1493.2 watts each. Three
fluorescent lamps of 96.0 watts were used in this building (Appendix A - Table A.3). Cow
brush and headlock were not present in this barn. Anaerobic lagoon was present around this
barn. Drawings about CBP barn 32 can be seen in Appendix B (Figures 59 and 60).

121 
 
 HVLS fan

(a) (b)
Figure 4.36 - The main constructive characteristic of CBP barn 32.

Barn 33 was built in November 2010 and cost of construction was $86,000 (Appendix
A - Table A.1). This barn had dimensions of 60.9 m x 24.1 m. This barn had two waterers
of 3.66 m x 0.61 m on the pack area. The concrete feed alley had width of 2.43 m. This
concrete feed alley was separated from the pack area by three guardrail fence. Concrete
wall was separating the pack and the feed alley in two walkways of 4.88 m each.
Geographic position of the Barn 33 were 36° 54' 24'' N latitude and 86° 13' 38'' W
longitude and the ridge orientation was 74° Southwest (SW), see Appendix A (Table A.2).
This barn was built using concrete floor, wood board, and metal panels were used in the
endwall (Figure 4.37). The eave height was 4.14 m and eave overhang of 2.43 m. The ridge
type used was overshot, roof pitch measured was 3 1/2:12, overlap of the ridge of 0.91 m,
ridge opening of 0.91 m, and metal roof was used to covered. Drawings of Barn 33 can be
seen in Appendix B (Figures 61 and 62). The Barn 33 was using natural ventilation system
(Appendix A - Table A.3). This barn was not equipped with lamps. Cow brush and
headlocks were not present in this barn. Anaerobic lagoon was around of the building.

(a) (b)
Figure 4.37 - The main constructive characteristic of CBP barn 33.

122 
 
 Barn 34 was built in October 2010. First of all, this barn was converted tobacco barn
and then added a new building. The cost for renovation this building was $6,000 and cost
for built a new part was $20,000 (Appendix A - Table A.1). The barns dimensions were 51.2
m x 19.5 m. The waterers was not present in the pack area. Four guardrail fence were used
in the sidewall. The labor was close to this building, as can be seen in Figure 4.38.
Geographic positions of Barn 34 were 36° 51' 04'' N latitude and 85° 54' 25'' W longitude
(Appendix A - Table A.2). The ridge orientation was 112° Northwest (NW). This barn was
built using wood boards, metal panels, metal roof was used to covered this building and
insulation was placed under this metal roof. The eave height was 5.90 m and eave overhang
was 2.75 m. The roof pitch of measured was 3.2:12. This barn had two ridge type
configuration. The tobacco barn had five chimneys, opening of chimneys was 0.43 m and
height of chimney was 0.23 m. The ridge type used was overshot in the new building,
overlap of this ridge of 1.17 m, and ridge opening of 0.61 m. Drawings of Barn 34 can be
seen in Appendix B (Figures 63 and 64). Barn 34 was using natural ventilation system
(Appendix A - Table A.3). This barn was not equipped lamps. Headlocks and cow brush
were not present in this barn. Anaerobic lagoon was not present around this barn.

(a) (b)
Figure 4.38 - The main constructive characteristic of CPB barn 34.

Barn 35 was built in the Fall of 2006 and cost of building was $60,000. The dimensions
of this barn were 36.5 m x 12.1 m (Appendix A - Table A.1). This barn had a concrete wall
of 0.3 m high (Figure 4.39). Concrete feed alley and waterers were separated from the pack
by low concrete wall and three steel pole. The concrete feed alley had width of 2.35 m. A
drive-by feed manager was located outside of this building. This barn had two metal
waterers of 3.0 m x 0.4 m on the pack area. Geographic positions of the Barn 35 were 36°
59' 08'' N latitude and 86° 57' 12'' W longitude (Appendix A - Table A.2) and ridge
123 
 
orientation of 117° Northwest (NW). This barn was built using concrete floor, metal
panels, and galvanized steel H-beam columns. The eave height was 4.72 m and eave
overhang had dimension of 1.52 m. The ridge type used was open ridge with cover, metal
roof was used to covered and the roof pitch was 4:12. Ridge cap had dimensions of 0.76 m
x 0.76 m, height of 0.58 m and ridge opening of 1.12 m. Drawings of Barn 35 can be seen
in Appendix B (Figures 65 and 66). Barn 35 was equipped with two HVLS fans of 7.32 m
and 1493.2 watts (Appendix A - Table A.3). Four incandescent (INC) lamps of 100.0 watts
were used above the pack area. Cow brush and headlocks were not present in this barn.
Anaerobic lagoon was around of the building.

HVLS fan 

Waterers

(a) (b)
Figure 4.39 - The main constructive characteristic of CBP barn 35.

Barn 36 was built in April 2004. There was no information about the cost of
construction (Appendix A - Table A.1). This barn dimensions were 43.9 m x 14.6 m,
respectively. The waterers was outside of this barn. The concrete feed alley has width of
3.6 m. Wood board fence was used around of the pack area, see Figure 4.40. Geographic
position of the Barn 36 were 37° 12' 19'' N latitude and 84° 59' 29'' W longitude and the
ridge orientation was 335° Southwest (SW), as can be seen in Appendix A (Table A.2).
This barn was built using concrete floor, wood boards, and metal panels. The eave height
was 4.6 m and metal roof was used to covered this barn. The eave overhang of this barn
had width of 1.22 m. This barn had different ridge configuration. The ridge type used was
overshot and closed ridge, details are showing in Appendix B (Figures 67 and 68). The roof
pitch measured was 4:12, overlap of the ridge of 0.74 m, and ridge opening of 0.43 m. Barn
36 was equipped with eight box fans along of pack of different diameters (Appendix A -
Table A.3). This barn was not equipped with lamps. Cow brush and headlock were not
present in this barn. Anaerobic lagoon was present around this barn.

124 
 
 Box fan 

(a) (b)
Figure 4.40 - The main constructive characteristic of CBP barn 36.

Barn 37 was built 2007 and the total cost, $17,000 (Appendix A - Table A.1). The
dimensions of the barn were 45.7 m x 10.6 m. This barn was built close to the barn 38.
Waterers was outside of this barn and had dimensions with of 0.7 m x 0.3 m (Figure 4.41).
The feed alley was away from facility, and the animals were feed in separate of the
building. The geographic position of the Barn 37 were 37° 12' 19'' N latitude and 84° 59'
29'' W longitude and the ridge orientation was 224° Northeast (NE), details are showing in
Appendix A (Table A.2). Thirty treated wood posts were used to build this barn. Drawings
are showing in Appendix B (Figures 69 and 70). Three treated wood boards, were used in
each sidewall. Tubular steel arches fastened to the tops and sides of the posts to form a
hooped roof were used and white poly vinyl tarp used to covered. The eave height of this
barn was 4.88 m. Only natural tunnel ventilation was used in Barn 37. Lamps and cow
brush were not present this barn (Appendix A - Table A.3). Anaerobic lagoon was not
present around this barn.

Waterers 

(a) (b)
Figure 4.41 - The main constructive characteristic of CBP barn 37.

Barn 38 was built in Summer of the 2011. The costs of construction was $25,000 and
$30,000 with feed alley. The building dimension were 36.6 m x 21.6 m (Appendix A -
125 
 
Table A.1). This barn was built close to the barn 37 (Figure 4.42). Sidewalls were open
without curtain. Plastic waterers was situated in another barn and the cows had a free
access to this building. This barn was built using concrete sidewall curb, wood pole, and
metal roof was used to covered. Sixteen wood post were uses on each side and spaced of
2.28 m from each other. Geographic position of the Barn 38 were 37° 20' 49'' N latitude
and 85° 21' 41'' W longitude (Appendix A - Table A.2). The ridge orientation was 233°
Northeast (NE). One hot wires were used between Barn 37 and Barn 38 to prevent cows
have access of this barn. One opening was on top of each endwall. The eave height was 4.9
m and eave overhang was 0.61 m. Drawings are showing in Appendix B (Figures 71 and
72). The ridge type used was overshot, roof pitch measured was 4:12, ridge opening of 0.44
m, and overlap of the ridge of 0.45 m. Only natural tunnel ventilation was used in Barn 38,
as can be seen in Appendix A (Table A.3). This barn was not equipped with lamps. Cow
brush were not present this barn (Appendix A - Table A.3). Anaerobic lagoon was not
present around this barn.

Barn 38
Barn 37

(a) (b)
Figure 4.42 - The main constructive characteristic of Barn 38.

Barns 39 and Barn 40 were built in the same time, using the same constructive
characteristics and costs of construction. These barns were built in July 2010 and the costs
of constructions were $80,000 for both barn including fans and concrete (Appendix A -
Table A.1). The barns dimensions were 48.7 m x 21.9 m. Two waterers were on the pack
area and had dimensions of 1.8 m x 0.6 m. The concrete feed alley width was 3.60 m. This
concrete feed alley was separated from the pack area by two guardrail fence, as can be seen
in Figure 4.43. Concrete curb wall was used above around this barn and two guardrail
fence was used above of this concrete curb wall. Curtains were used on the sidewall during
the winter. Geographic position of these barns were 36° 58' 10'' N latitude and 85° 38' 36''

126 
 
W longitude and the ridge orientation was 17° Southwest (SW), as can be seen in
Appendix A (Table A.2). These barn were built using concrete floor, wood posts, and metal
panels. The eave height was 4.84 m and metal roof was used to covered these barns. These
barns had eave overhang of 0.81 m. The ridge type used was overshot. The roof pitch
measured was 5:12, overlap of the ridge of 0.91 m, and ridge opening of 0.61 m. Drawings
are showing in Appendix B (Figures 73 and 74). Barn 39 and Barn 40 were equipped eight
box fans above of pack area, details can be seen in Appendix A (Table A.3). Five
fluorescent lamps of 192 watts were used in each barn. Cow brush and headlock were not
present in this barn. Anaerobic lagoon was present around this barn.

Box fans 

(a) (b)
Figure 4.43 - The main constructive characteristic of CBP barns 39 and Barn 40.

Barn 41 was built in 2001 and cost of building was $14,000 (Appendix A - Table A.1).
The dimensions of this barn were 30.5 m x 15.2 m. The feed alley and waterers were
outside of this barn. The waterers had diameter of 1.5 m. Two opening were on the top and
the bottom of the sidewall around the perimeter of this barn (Figure 4.44). Geographic
positions of the Barn 41 were 36° 39' 54'' N latitude and 85° 42' 58'' W longitude
(Appendix A - Table A.2) and ridge orientation of 102° Northwest (NW). This barn was
built using metal panels and wood columns. The eave height was 3.5 m and eave overhang
of 0.3 m. The ridge type used was open ridge with cover, metal roof was used to covered
and the roof pitch was 4:12. Ridge cap had dimensions of 0.54 m x 0.54 m, height of 0.27
m and ridge opening of 1.0 m. Barn 41 was equipped with eight box fans of 0.61 m and
246.0 watts (Appendix A - Table A.3). Lamps, headlocks and cow brush were not present
in this barn. Anaerobic lagoon was around of the building. More details are showing in
Appendix B (Figures 75 and 76).

127 
 
 (a) (b)
Figure 4.44 - The main constructive characteristic of CBP barn 41.

Barn 42 was built in June 2010 and costs of construction was $25,000. The building
dimension of this barn were 36.6 m x 15.2 m (Appendix A - Table A.1). One sidewalls was
closed with metal panels, but this sidewall had two openings on the top and bottom. Four
wood boards were used around of this barn (Figure 4.45). Twenty three wood posts were
used on each side and spaced of 1.66 m from each other. Waterers was situated outside of
the barn and the cows had a free access to another building. Two tubular steel gates were
used to access to another building. Geographic positions of the Barn 42 were 36° 41' 06'' N
latitude and 85° 42' 15'' W longitude (Appendix A - Table A.2). The ridge orientation was
204° Northeast (NE). This barn was built using wood posts, metal panels, and metal roof
was used to covered. The eave height was 3.8 m and eave overhang was 0.6 m. The ridge
type used was overshot. The roof pitch measured was 4:12, ridge opening of 0.35 m, and
overlap of the ridge of 0.57 m. Barn 42 was equipped with eight box fans along of pack of
1.32 m diameter, see Appendix A (Table A.3). Six incandescent lamps were used above the
pack area. Headlocks and cow brush were not present in this barn. Anaerobic lagoon was
present around this barn. Drawings are showing in Appendix B (Figures 77 and 78).

Box fans 

(a) (b)
Figure 4.45 - The main constructive characteristic of CBP barn 42.
128 
 
4.2.3. Compost bedding pack management
The highest percentage (45.2%) of barns are bedded with kiln dried wood shavings or
sawdust, followed by the green sawdust (42.9%) and the blend of both materials (11.9%).
Fine, dry wood shavings or sawdust is the ideal material for bedding (Endres, 2009). Most
of the barns in Kentucky had their compost tilled and aerated to a depth of 0.11 to 0.23 m
twice a day in the summer and winter time, while cows were at the milking parlor. The
most often used stirring tool was a chisel plow on a small tractor.
A large number of farmers cleaned out the pack entirely once a year. In this case, a load
of clean material (shavings or sawdust) was added after removal of the used bedding to a
depth 0.30 m in order to start the new pack. Typically, a load of fresh material was added
every one to five weeks, varying by season, weather conditions, barn size, and cow density.
Some farmers left about 0.15 m of old material in barn to help initiate microbial activity.
There were a small number of dairies adding a thin layer of fresh bedding when the bedded
pack became moist enough for it to stick to the cows body after they rise from lying down
on the bedded pack.
The average pack dimensions for the 42 compost barns we studied were 47.7 m (±19.1)
and 16.1 m (±5.9). The average stock density in this study was 8.9 m2 per cow, ranging
from 0.8 to 25.8 m2 per cow. The suggested minimum space per cow in these building is
7.4 m2 per cow (Janni et al., 2007). According to Endres (2009), some producers in
Minnesota provided 9.4 m2 per cow.
Temperatures of bedding compost tended to be lower on barns 16, 22, 23, 24, 29, 35,
36, and 38 which utilized green sawdust rather than kiln dried wood shavings or sawdust
for bedding (t test, P > 0.05). The temperatures were greater in the areas of the pack that
were fluffier, that were not heavily soiled or packed by the cows. This observation is
consistent with the need for oxygen for microbial activity that promotes composting.
Average bedding moisture content in the top (20 cm) was 59.0% (± 9.0) wet basis
ranged from 36.2 to 71.8%. Details can be seen in Appendix A (Table A.4). Moisture
content between 40.0% and 65.0% is generally recommended for composting most
materials (NRAES-54, 1992). Excessive moisture addition needs to be avoided. The
moisture content can vary owing to the addition of fresh bedding, weather, and cow density
in the sampling area (Barbeg et al.. 2007). More details about compost bedding pack
management in each barn can be seen below.

129 
 
4.2.3.1. Management practices in each barn
Barn 1 was bedded with green sawdust (Figure 4.46a). The pack was aerated twice a
day in the summer and the winter time using a chisel plow (Figure 4.46b), while the cows
were milked. The pack was aerated to a depth of 0.28 m, which is the recommended depth
of 0.25 to 0.30 m (Janni et al., 2007). This barn was not cleaned out during the year, the
pack had dimensions of 41.5 m x 15.3 m. The stock density was 10.0 m2 per cow
(Appendix A - Table A.1).
The average bedding temperatures at surface and 0.10 m and 0.20 m depths were
17.1°C (±1.3), 31.9°C (±10.2), and 35.2°C (±9.4), respectively. Paired t-test statistical
analysis indicated that the averages for the two depths were different (P < 0.05). Details
can be seen in Appendix A (Table A.4).
The average bedding moisture was 44.5% (±8.0). The minimum bedding moisture was
30.7% and the maximum was 56.7% (Appendix A - Table A.5). The first and third
quartiles were 40.4% and 49.8%, respectively.

(a) (b)
Figure 4.46 - Green sawdust (a) and (b) chisel plow used to aeration used in barn 1.

Barn 2 was using green sawdust to bed the pack (Figure 4.47a). This barn used a chisel
plow on a small tractor (Figure 4.47b) to till the pack twice a day in the summer and the
winter time, while the cows were milked. The pack was aerated to a depth of 0.40 m. The
pack dimensions were 53.4 m x 30.3 m and the stock density was 9.3 m2 per cow
(Appendix A - Table A.1). This barn was cleaned out once a year.
The average of the pack temperature in the depth 0.1 m [41.5°C (±10.6)] was less than
the average of the depth 0.2 m [53.4°C (±4.6)], as can been see in Appendix A (Table A.4).

130 
 
Paired t-test statistical analysis indicated that at the 0.05 significance level, the averages for
the two depths were different. The surface pack temperature measured was 16.5°C (±3.6).
The average bedding moisture was 49.0% (±7.3), ranging from 38.3% to 61.5%
(Appendix A - Table A.5). The first and third quartiles were 45.7% and 52.0%,
respectively. The moisture content in the bedding compost varied upon time since last
addition of fresh bedding, weather, and stock density in the sampling area.

(a) (b)
Figure 4.47 - Green sawdust (a) and (b) the chisel plow used in barn 2.

Barn 3 was bedded with kiln dried wood shavings or sawdust. According Changirath et
al. (2011), the cows prefer lying on drier and loose bedding (Figure 4.48). The bedded pack
was stirred twice a day in the summer and winter time, while the cows were being milked.
Moreover, a small tractor was pulling a chisel plow to promote the aeration. The bedding
compost was removed from this barn three a year and spread directly in the fields when
necessary. The pack area had dimensions of 87.8 m x 12.2 m and the stock density was 7.4
m2 per cow (Appendix A - Table A.1). This barn was cleaned out three times per year.
Averages bedding temperatures at surface, 0.1 m and 0.2 m depths were 16.3°C (±3.8),
36.2°C (±9.5), and 52.1°C (±4.6), respectively (Appendix A - Table A.4). Paired t-test
statistical analysis indicated that the averages for the two depths were different (P < 0.05).
The average bedding moisture was 49.1% (±5.8), which is the moisture content, needed
for composting (NRAES-54, 1992). The minimum and maximum bedding moisture was
between 43.4% and 58.7%. The first and third quartiles were 46.1% and 54.1%,
respectively. Details about bedding moisture content as seen in Appendix A (Table A.5).

131 
 
 (a) (b)
Figure 4.48 - Cows lying on the bedding compost of the Barn 3.

Barn 4 was bedded with mix material (sawdust mixed with other material as available).
Tilling was performed once a day in the summer and winter time, using a chisel plow with
0.45 m shovels (Figure 4.49a). The chisel plow tills to a depth of 0.25 m – 0.38 m. This
keeps the bedded pack fairly loose, well aerated, and the surface dry. In fact, this bedded
pack was the driest at surface to 0.25 m depth. During cultivation water vapor was visible,
indicating a higher temperature of the compost in comparison to the air (Figure 4.49b).
This barn was not cleaned out during the year. The pack dimensions were 51.5 m x 9.2 m
and the stock density (5.1 m2 per cow). Details can be seen in Appendix A (Table A.1).
The average surface temperature measured was 15.5°C (±3.5). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 41.1°C (±18.7) and 52.4°C (±7.6),
respectively (Appendix A - Table A.4). Paired t-test statistical analysis indicated that these
two averages were not different (P < 0.05).
The average bedding moisture was 36.2% (±6.3). as can be seen in Appendix A (Table
A.5). The minimum bedding moisture was 27.8% and the maximum was 44.9%. The first
and third quartiles were 30.5% and 40.0%, respectively.

(a) (b)
Figure 4.49 - Chisel plow used the bedding compost in Barn 4.

132 
 
 Barn 5 was using kiln dried wood shavings or sawdust in the pack (Figure 4.50). This
bedding material has been widely used in compost dairy barn throughout Minnesota with
different success (Barberg et al., 2007; Janni et al., 2007; Janni et al., 2005). Aeration was
performed once a day in the summer time and three times a day in the winter time, using a
tractor to pull a chisel plow. The pack was aerated to a depth of 0.23 m. The pack
dimensions were 42.7 m x 10.7 m. This barn was cleaned out twice a year. The stock
density (5.3 m2 per cow) was less than suggested minimum (7.4 m2 per cow) by Janni et al.
(2007). Details can be seen in Appendix A (Table A.1).
The average bedding temperatures at surface and two depths (0.1 m and 0.2 m) were
22.2°C (±1.3), 26.0°C (±6.1), and 27.2°C (±6.6), respectively, as can been see in Appendix
A (Table A.4). Paired t-test statistical analysis indicated that the averages for the two
depths were not different (P < 0.05).
The average bedding moisture was 52.2% (±5.6). The minimum bedding moisture was
43.7% and the maximum was 61.7%. The first and third quartiles were 47.5% and 55.2%,
respectively. More details about moisture content of bedding see Appendix A (Table A.5).

(a) (b)
Figure 4.50 - Barn 5 was using kiln dried wood shavings or sawdust in the pack.

Barn 6 was bedded with kiln dried wood shavings or sawdust (Figure 4.51a). The
loosest and driest compost was found in this barn with a fluffy consistency. The pack was
aerated once a day in the summer time and twice a day in the winter time, while the cows
were milked, using rotary tiller on a tractor (Figure 4.51b). This barn was not cleaned out.
The bedded pack had dimensions of 24.4 m x 12.2 m and the stock density was 4.0 m2 per
cow. Details can be seen in Appendix A (Table A.1).

133 
 
 The surface pack temperature measured was 16.2°C (±3.1). The average of the pack
temperature in the depth 0.1 m [34.5°C (±15.0)] was high than the average of the depth 0.2
m [34.0°C (±10.1)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were not different (P < 0.05).
The average bedding moisture was 39.6% (±9.6), which is less than recommended by
NRAES-54 (1992). The minimum and maximum bedding moisture was between 22.9%
and 51.8%. The first and third quartiles were 32.0% and 47.1%, respectively. Details about
bedding moisture content can be seen in Appendix A (Table A.5).

(a) (b)
Figure 4.51 - Rotary tiller used to till the bedding compost in barn 6.

Barn 7 was bedded with kiln dried wood shavings. The pack was mixed twice a day in
the summer and winter time, using a rotary tiller (Figure 4.52). This barn was cleaned out
once a year and the pack was aerated to a depth of 0.20 m. This barn had pack dimensions
of 36.6 m x 12.5 m and the stock density was 11.7 m2 per cow (Appendix A - Table A.1).
The average bedding temperatures at surface and two depths (0.1 m and 0.2 m) were
13.3°C (±4.7), 38.3°C (±6.0), and 42.2°C (±8.6), respectively (Appendix A - Table A.4).
Paired t-test statistical analysis indicated that the averages for the two depths were not
different (P < 0.05).
The average bedding moisture was 46.0% (±3.2). The minimum bedding moisture was
40.8% and the maximum was 49.7% (Appendix A - Table A.5). The first and third
quartiles were 44.4% and 48.9%, respectively.

134 
 
 (a) (b)
Figure 4.52 – Rotary tiller (was used to till the bedding compost of barn 7.

Barn 8 was bedded using kiln dried wood shavings or sawdust. This barn was not
cleaned out a year. The pack was aerated once a day in the summer and winter time using a
chisel plow (Figure 4.53). The pack was aerated to a depth of 0.25 m. The compost
appeared relatively wet and did stick to the boots when walking on the surface and small
lumps of bedding pack material did not smell when broken and rubbed against the palms of
our hands. According Klaas et al. (2010), this process can indicate absence of anaerobic
processes. The pack dimensions were 65.9 m x 7.9 m and the stock density was 11.0 m2 per
cow. Details can be seen in Appendix A (Table A.1).
Average bedding temperatures at surface and two depths (0.1 m and 0.2 m) were
17.8°C (±4.1). 24.1°C (±4.8), and 24.4°C (±3.5), respectively (Appendix A - Table A.4).
Paired t-test statistical analysis indicated that the averages for the two depths were not
different (P < 0.05).
The average bedding moisture was 61.6% (±2.5). The minimum bedding moisture was
58.2% and the maximum was 65.9% (Appendix A - Table A.5). The first and third
quartiles were 59.5% and 63.5%, respectively.

(a) (b)
Figure 4.53 – Barn 8 was aerated using a chisel plow.
135 
 
 Barn 9 was using sawdust mixed with other material as available. The pack was aerated
once a day in the summer time and twice a day in the winter time. A tractor was aerating
the pack using a rotary tiller and a chisel plow (Figure 4.54). This barn was cleaned out
once a year. The compost pack appeared to be compacted and chunks were seen after
stirring. Compaction is expected to reduce aeration and microbial activity compared to the
fluffy loose packs (Janni et al., 2007). The pack dimensions were 76.3 m x 18.5 m and the
stock density was 13.5 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was 22.3°C (±11.7). The average of the pack
temperature in the depth 0.1 m [38.6°C (±13.2)] was less than the average of the depth 0.2
m [42.5°C (±14.8)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were different (P < 0.05).
The average bedding moisture was 41.9% (±6.6). The minimum bedding moisture was
33.6% and the maximum was 53.9% (Appendix A - Table A.5). The first and third
quartiles were 37.5% and 45.7%, respectively.

(a) (b)
Figure 4.54 – Barn 9 was aerated using a chisel plow.

Barn 10 was bedded with kiln dried wood shavings or sawdust (Figure 4.55a). The
bedded pack was stirred twice a day in the summer and winter time, using a chisel plow
(Figure 4.55b), while the cows were being milked. The stirring mixes manure and urine on
the surface into the bedded pack to provide a fresh surface for cows to lie down on the pack
after milked (Janni et al., 2007). This barn was cleaned out twice a year. The pack
dimensions were 61.0 m x 30.5 m. Approximately 184 cows were kept in the bedded pack
that provided a stock density of 10.1 m2 per cow (Appendix A - Table A.1).

136 
 
 The averages bedding temperatures at surface and 0.1 m and 0.2 m depths were 17.3°C
(±5.6), 35.6°C (±8.6), and 41.4°C (±11.4), respectively. Paired t-test statistical analysis
indicated that the averages for the two depths were different (P < 0.05). Details can be seen
in Appendix A (Table A.4).
The average bedding moisture was 51.7% (±7.3). The minimum and maximum bedding
moisture was between 38.3% and 64.0%, respectively. The first and third quartiles were
48.7% and 54.5%, respectively. The average moisture content of bedding in Barn 10 is
given in Appendix A (Table A.5).

(a) (b)
Figure 4.55 - Barn 10 was aerated using a chisel plow.

Barns 11 and 12 were bedded using sawdust mixed with other material as available.
The wet compost was found in these barns, with mainly slurry consistency and did stick to
the boots when walking on the surface. The bedding materials were aerated to a depth of
0.23 m, once a day in the summer and winter time while cows are being milked. A tractor
was pulling a chisel plow to promote the aeration (Figure 4.56). The bedded packs had the
same dimensions in both barns of 91.5 m x 15.3 m and the stock density were 9.3 m2 per
cow (Appendix A - Table A.1). These barns were cleaned out twice a year.
The average surface temperature measured in CBP barns 11 and 12 were 15.2°C (±2.5)
and 18.5°C (±12.5), respectively (Appendix A - Table A.4). Paired t-test statistical analysis
indicated that there were not significant (P < 0.05) differences in average temperatures for
the two depths in Barn 11, but there were significant (P < 0.05) differences in average
temperatures for the two depths in Barn 12.
The average bedding moisture in Barns 11 and 12 were 61.2% (±2.4) and 60.4% (±2.8),
respectively. Details can be seen in Appendix A (Table A.5).

137 
 
 (a) (b)
Figure 4.56 - Barns 11 and 12 were bedded using sawdust.

Barns 13 and 14 were bedded using kiln dried wood shavings or sawdust (Figure
4.57a). The compost packs were aerated twice a day in the summer and winter time, using
rotary tiller (Figure 4.57b). Water vapor was visible during aeration, indicating a higher
temperature of the compost in comparison to the air. These barns were not cleaned out
during the year. These barns had the same pack dimensions of 61.0 m x 13.7 m and the
same stock density of 8.8 m2 per cow (Appendix A - Table A.1).
The average surface temperatures measured in Barns 13 and 14 were 11.7°C (±5.3) and
8.9°C (±1.6), respectively. Details about bedding temperature can be seen in Appendix A
(Table A.4). Paired t-test statistical analysis indicated that in both barns the averages
temperatures for the two depths of were not different (P < 0.05).
The average bedding moisture in Barn 13 was less [36.2% (±6.3)] than Barn 14 [57.3%
(±4.3)], ranged from 50.6% to 63.4% in Barn 13 and ranged from 49.8% to 64.0% in Barn
14 (Appendix A - Table A.5).

(a) (b)
Figure 4.57 - Rotary tiller was used to aerated the bedding compost in barns 13 and 14.

138 
 
 Barn 15 was bedded using kiln dried wood shavings or sawdust (Figure 4.58). The
compost pack was aerated twice a day (summer and winter time), using chisel plow. The
bedding materials were aerated to a depth of 0.18 m. The bedding compost was cleaned out
once a year and the bedding compost was spread directly in the fields when necessary. This
bedding compost was appearing with fluffy consistency. The pack had dimensions of 18.3
m x 18.3 m and the stock density was 9.3 m2 per cow (Appendix A - Table A.1).
The average surface temperature measured was 17.2°C (±10.2). The average bedding
temperatures across 0.1 m and 0.2 m depths were 52.3°C (±7.5) and 58.0°C (±10.3),
respectively. Paired t-test statistical analysis indicated that these two averages were
different (P < 0.05). Temperature is directly proportional to the biological activity within
the composting bedded-pack. Maintaining a temperature of 52°C or more for 3 to 4 days
favors the destruction of fly larvae and keeping the pack surface dry (Janni et al., 2005).
Average bedding moisture was 54.7% (±2.7), ranging from 51.0 to 58.6% (Appendix A
- Table A.5). The first and third quartiles were 53.0% and 56.2%, respectively.

(a) (b)
Figure 4.58 - The bedding compost in Barn 15 was appearing with fluffy consistency.

Barn 16 was bedded with green sawdust. The wet compost was found in this barn, with
mainly slurry consistency. The pack was aerated once a day in the summer and winter time.
A tractor was pushing a chisel plow (Figure 4.59). The bedding materials were aerated to a
depth of 0.20 m, which is less than minimum recommended depth of 0.25 m to 0.30 m
(Janni et al., 2007). This barn was cleaned out twice a year. The pack had dimensions of
14.6 m x 10.1 m and the stock density was 7.7 m2 per cow (Appendix A - Table A1).
The average bedding temperatures at surface and 0.1 m and 0.2 m depths were 8.9°C
(±2.9), 23.5°C (±12.6), and 32.1°C (±7.2), respectively. Paired t-test statistical analysis

139 
 
indicated that at the 0.05 significance level, the averages for the two depths were not
different (Appendix A - Table A.4).
The average bedding moisture was 66.5% (±2.2), which is high than the recommended
for composting of 45% to 60% moisture (NRAES-54, 1992). The minimum bedding
moisture was 64.3% and the maximum was 70.6% (Appendix A - Table A.5). The first and
third quartiles were 64.6% and 67.1%, respectively.

(a) (b)
Figure 4.59 - Barn 16 was aerated using chisel plow.

Barn 17 was bedded using kiln dried wood shavings. This barn was cleaned out once a
year. The pack was aerated twice a day (summer and winter time), using a chisel plow
(Figure 4.60), and the pack was aerated to a depth of 0.30 m. The compost appeared
relatively wet and dried fine material was spread out on top of the compost, however, it still
sticks to the boots, when one was walking on the surface. The pack had dimensions of 37.8
m x 13.7 m and the stock density was 4.3 m2 per cow (Appendix A - Table A.1).
The average bedding temperatures at surface and 0.1 m and 0.2 m depths were 11.7°C
(±5.0), 37.4°C (±16.6), and 36.6°C (±17.0), respectively (Appendix A - Table A.4). Paired
t-test statistical analysis indicated that the averages for the two depths were not different (P
< 0.05).
The average bedding moisture was 56.0% (±4.3), ranged from 50.2% to 62.4%
(Appendix A - Table A.5). The first and third quartiles were 52.7% and 59.1%,
respectively.

140 
 
 (a) (b)
Figure 4.60 - Barn 17 was aerated using chisel plow.

Barn 18 was using kiln dried wood shavings or sawdust. This package was very wet,
when walking atop it, the boots were sunk in up to ankles in some spots (Figure 4.61a). The
bedding material was aerated twice a day in the summer and winter time, using chisel plow
(Figure 4.61b) while cows are being milked. The bedded pack had dimensions of 61.0 m x
24.4 m and the stock density was 12.9 m2 per cow. This value is high than suggested
minimum (7.4 m2 per cow) by Janni et al. (2007). Compost barns will not perform well if
the cows are overcrowded. Details can be seen in Appendix A (Table A.1).
The average surface temperature measured was -5.1°C (±3.5). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 12.3°C (±9.7) and 20.9°C (±10.9),
respectively (Appendix A - Table A.4). Paired t-test statistical analysis indicated that the
averages for the two depths were different (P < 0.05). An anaerobic pack shows lower
temperatures which leads to reduced pathogen kill and odor problems (Graves, 1999).
The average bedding moisture was 64.8% (±0.8), which was above NRAES-54 (1992)
recommended level of 50% to 60%. When moisture is too high, the pack becomes
anaerobic (lack of oxygen), rate of microbial decomposition will slow, and the composting
will again be too slow (Bewley and Taraba, 2009). The minimum bedding moisture was
63.5% and the maximum was 66.0%. The first and third quartiles were 64.2% and 65.6%,
respectively (Appendix A - Table A.5).

141 
 
 (a) (b)
Figure 4.61 - Barn 18 was aerated using chisel plow.

Barn 19 was bedded with Kiln dried wood shavings. This barn was aerated twice a day,
using chisel plow and rotary tiller on a tractor. During cultivation water vapor was visible,
indicating a higher temperature of the compost in comparison to the air (Figure 4.62). The
pack was cleaned out twice a year and had dimensions of 11.6 m x 14.0 m. The stock
density was 6.7 m2 per cow (Appendix A - Table A.1).
The pack surface temperatures [-0.9°C (±3.8)]. The average bedding temperature
measured at 0.1 m depth [39.7°C (±5.6)] was less than average bedding temperatures
measured at 0.2 m depth [52.8°C (±3.2)]. Paired t-test statistical analysis indicated that the
averages for the two depths were different (P < 0.05). The average temperature of bedding
in this barn is given in Appendix A (Table A.4).
The average bedding moisture was 68.4% (±1.9), as can be seen in Appendix A (Table
A.5). The minimum bedding moisture was 64.5% and the maximum was 71.0%. The first
and third quartiles were 68.0% and 69.5%, respectively. A wet bedded pack is more
vulnerable to compaction.

(a) (b)
Figure 4.62 - Water vapor was visible during cultivation in Barn 19.

142 
 
 Barn 20 was bedded with kiln dried wood shavings or sawdust (Figure 4.63). New fresh
and dried fine material was spread out on top of the compost once a day. The pack was
aerated once a day. A tractor was used to pull a chisel plow. The bedding materials were
aerated to a depth of 0.18 m. This barn was cleaned out twice a year. The pack dimensions
were 22.0 m x 12.0 m and the stock density was 5.3 m2 / cow (Appendix A - Table A.1).
The averages bedding temperatures at surface and 0.1 m and 0.2 m depths were 23.7°C
(±1.2), 44.0°C (±12.4), and 45.0°C (±11.6), respectively. Paired t-test statistical analysis
indicated that at the 0.05 significance level, the averages for the two depths were not
different (Appendix A - Table A.2).
The average bedding moisture was 60.3% (±3.3). The minimum and maximum bedding
moisture was between 53.6% and 63.0%, respectively. The first and third quartiles were
58.3% and 62.7%, respectively. The average moisture content of bedding in Barn 20 is
given in Appendix A (Table A.5).

(a) (b)
Figure 4.63 – Dried fine material was spread out on top of the compost in Barn 20.

Barn 21 was bedded with kiln dried wood shavings. The compost appeared relatively
wet and fresh and fine material was spread out on top of the pack once a day. The bedding
material was aerated to a depth of 0.20 to 0.25 m once a day (summer time) and twice a
day (winter time) while cows were away at the parlor, using a chisel plow on a small
tractor (Figure 4.64). This barn was cleaned out once a year. The pack had dimensions of
36.6 m x 15.3 m. The stock density was 7.9 m2 / cow (Appendix A - Table A.1).
The surface pack temperature measured was 4.5°C (±2.4). The average of the pack
temperature in the depth 0.1 m [46.1°C (±32.5)] was less than the average of the depth 0.2
m [50.2°C (±14.7)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were not different (P < 0.05).
143 
 
 The average bedding moisture was 58.8% (±3.3). The minimum bedding moisture was
55.0% and the maximum was 64.5% (Appendix A - Table A.5). The first and third
quartiles were 56.8% and 59.2%, respectively.

(a) (b)
Figure 4.64 - Chisel plow was used to aerate the pack compost of barn 21.

Barn 22 was bedded with green sawdust. The loosest and driest compost was found in
the pack with a fluffy consistency (Figure 4.65). The pack was aerated once a day while the
cows were milked. A tractor was used to pull the chisel plow. This barn was cleaned out
twice a year. The bedded pack had dimensions of 61.0 m x 15.3 m and the stock density
was 10.2 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was 6.9°C (±2.7). The average of the pack
temperature in the depth 0.1 m [28.8°C (±5.4)] was high than the average of the depth 0.2
m [31.6°C (±3.2)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were different (P < 0.05).
The average bedding moisture was 52.0% (±4.4), ranging from 43.9% to 58.3%
(Appendix A - Table A.5). The first and third quartiles were 50.0% and 53.9%,
respectively.

Figure 4.65 - Bedding compost was fluffy consistency in Barn 22.

144 
 
 Barn 23 was bedded with green sawdust (Figure 4.66). The bedding compost was
appearing compacted and wet, with mainly slurry consistency. The pack was aerated once a
day while the cows were milked. A tractor was used to pull the chisel plow. This barn had
pack dimensions of 61.0 m x 15.3 m and was cleaned out twice a year. The stock density
(25.9 m2 per cow) was high than suggested minimum (7.4 m2 per cow) by Janni et al.
(2007). Compost barns will not perform well if the cows are overcrowded.
The average bedding temperatures at surface was 6.0°C (±2.6). Paired t-test statistical
analysis indicated that at the 0.05 significance level, the averages for the two depths were
different. When comparing the pack temperature of the two depths, was found that the
average of the depth 0.1 m was less than the average of the depth 0.2 m (Appendix A -
Table A.4). The bedding temperatures indicated that the bedding material used for this barn
was not composting, although the aerated pack was biologically active.
The average bedding moisture was 67.3% (±1.3), which is high than the recommended
for composting of 45% to 60% moisture (NRAES-54, 1992). The minimum bedding
moisture was 65.9% and the maximum was 69.7% (Appendix A - Table A.5). The first and
third quartiles were 66.3% and 68.1%, respectively.

Figure 4.66 - The bedding compost in Barn 23 was appearing compacted and wet.

Barn 24 was bedded using green sawdust (Figure 4.67). The pack had dimensions of
30.5 m x 15.3 m and was cleaned out twice a year. This farmer added some amount of
fresh media every week. It is common for farmers to have to add more media in the winter
due to the cold weather inhibiting airflow in the barn thus reducing the drying of the
surface of the pack (Changirath et al., 2011). The pack was aerated once a day while the
cows were milked, using a chisel plow. The stock density of this barn was 12.9 m2 per cow
(Appendix A - Table A.1).

145 
 
 The average bedding temperatures at surface and 0.1 m and 0.2 m depths were 7.2°C
(±2.5), 15.3°C (±5.5), and 19.8°C (±4.9), respectively (Appendix A - Table A.4). Paired t-
test statistical analysis indicated that the averages for the two depths were different (P <
0.05).
The average bedding moisture was 64.0% (±1.6), which is high than the recommended
by NRAES-54 (1992). The minimum bedding moisture was 62.5% and the maximum was
67.3% (Appendix A - Table A.5). The first and third quartiles were 62.9% and 64.2%,
respectively.

Figure 4.67 - Fresh media was added every week in Barn 24.

Barn 25 was bedded with kiln dried wood shavings or sawdust. Aerating was done
twice a day with a chisel plow that turns over the compost 0.20 to 0.25 m below the surface
(Figure 4.68). This farmer was trying to minimize the aeration to two or three times per
week or less. The reason given was to lower labor and fuel costs with seeming little benefit
to the bedded pack through drying or mixing. Fresh media was added to the pack every two
days as needed to keep the pack dry. The pack had dimensions of 51.2 m x 29.9 m and was
not cleaned out. The stock density was 9.3 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was 7.0°C (±4.2). The average of the pack
temperature in the depth 0.1 m [17.1°C (±3.5)] was high than the average of the depth 0.2
m [20.2°C (±2.3)]. as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were different (P < 0.05).
The average bedding moisture was 61.2% (±2.3). The minimum bedding moisture was
58.9% and the maximum was 64.5% (Appendix A - Table A.5). The first and third
quartiles were 59.4% and 63.8%, respectively.

146 
 
 (a) (b)
Figure 4.68 - Aerating was done with a chisel plow in CBP barn 25.

Barn 26 was bedded with green sawdust. The dimensions of the bedded pack were 61.0
m x 15.6 m. Approximately 135 cows were kept in the bedded pack that provided a stock
density of 7.0 m2 per cow (Appendix A - Table A.1). Aerating was done with a chisel plow
(Figure 4.69) twice a day in the summer and winter time. Fresh media was added to the
pack once a week in the winter.
The average bedding temperatures at surface and two depths (0.1 m and 0.2 m) were
6.8°C (±4.5), 39.5°C (±11.2), and 39.5°C (±9.8), respectively, as can been see in Appendix
A (Table A.4). Paired t-test statistical analysis indicated that the averages for the two
depths were not different (P < 0.05).
The average bedding moisture was 57.3% (±2.8). The minimum bedding moisture was
54.1% and the maximum was 63.3% (Appendix A - Table A.5). The first and third
quartiles were 55.0% and 58.5%, respectively.

(a) (b)
Figure 4.69 - Aerating was done with a chisel plow in CPB barn 26.

Barn 27 was bedded with kiln dried wood shavings or sawdust (Figure 5.70a). Aerating
was done with a chisel plow (Figure 4.70b) twice a day (summer and winter time) when the
cows are being milked. The depth of the bedding media layer was approximately 0.15 m to
147 
 
0.20 m. This barn was not cleaned out during the year. The dimension of the bedded pack
was 36.6 m x 12.2 m. The stock density was 8.6 m2 per cow (Appendix A - Table A.1).
The average surface temperature measured was 18.2°C (±9.1). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 39.4°C (±23.3) and 43.3°C (±14.7),
respectively. (Appendix A - Table A.4). Paired t-test statistical analysis indicated that these
two averages were not different (P < 0.05).
The average bedding moisture was 67.8% (±1.3). as can be seen in (Appendix A -
Table A.5). The minimum bedding moisture was 66.2% and the maximum was 70.7%. The
first and third quartiles were 67.0% and 68.2%, respectively.

(a) (b)
Figure 4.70 - Aerating was done with a chisel plow in CBP barn 27.

Barn 28 was bedded with green sawdust (Figure 4.71a). This barn had pack dimensions
of 36.6 m x 15.3 m and was tilling once a day (summer) and twice a day (winter), thereby
mixing and aerating the pack using chisel plough (Figure 4.71b). This farmer removed the
bedded pack media from this barn one time per year in summer, and then immediately
applied it to the fields. The stock density was 7.0 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was 9.0°C (±4.2). The average of the pack
temperature in the depth 0.1 m was less than the average of the depth 0.2 m (Appendix A -
Table A.4). Paired t-test statistical analysis indicated that these two averages were different
(P < 0.05).
The average bedding moisture was 57.3% (±2.8), ranging from 54.1% to 63.3%. The
first and third quartiles were 55.0% and 58.5%, respectively. Details can be seen in
Appendix A (Table A.5).

148 
 
 (a) (b)
Figure 4.71 - Bedded with green sawdust and chisel plough used in CBP barn 28.

Barn 29 was bedded with green sawdust and the bedded pack had dimensions of 61.0 m
x 14.6 m (Figure 4.72a). This package was very wet, when walking atop it, the boots were
sunk in up to ankles in some spots. Tilling was performed once daily (summer and winter)
using a chisel plow with 0.25 m shovels (Figure 4.72b). The pack appeared to be
compacted and chunks were seen after stirring. The stock density was 10.0 m2 per cow
(Appendix A - Table A.1).
The average surface temperature measured was 5.4°C (±3.4). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 15.5°C (±6.7) and 18.0°C (±5.6),
respectively. (Appendix A - Table A.4). Paired t-test statistical analysis indicated that the
averages for the two depths were different (P < 0.05). Details can be seen in Appendix A
(Table A.4).
The average bedding moisture was 61.1% (±2.1). The minimum bedding moisture was
58.2% and the maximum was 64.5%. The first and third quartiles were 59.2% and 62.4%,
respectively. The average moisture content of bedding is given in Appendix A (Table A.5).

(a) (b)
Figure 4.72 - Barn 29 was bedded with green sawdust.

149 
 
 Barn 30 was bedded with green sawdust. This barn was not cleaned out during the year.
Aerating was done with a chisel plow twice a day (summer and winter) when the cows are
being milked. The depth of the bedding media layer was approximately 0.40 to 0.45 m. The
wet compost was close to the scrape alley (Figure 4.73). This barn had pack dimensions of
45.8 m x 18.8 m and the stock density was 8.8 m2 per cow (Appendix A - Table A.1).
The average surface temperature measured was 14.2°C (±9.8). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 49.2°C (±7.8) and 52.9°C (±4.5),
respectively (Appendix A - Table A.4). Paired t-test statistical analysis indicated that the
averages for the two depths were different (P < 0.05). Details can be seen in Appendix A
(Table A.4).
The average bedding moisture was 47.4% (±2.6). The minimum bedding moisture was
43.4% and the maximum was 51.2%. The first and third quartiles were 46.2% and 49.0%,
respectively. The average moisture content of bedding is given in Appendix A (Table A.5).

(a) (b)
Figure 4.73 - Barn 30 was bedded with green sawdust.

Barn 31 was using green sawdust and a load of green sawdust was spread in early in the
winter. Tilling was done twice a day (summer and winter). A tractor was used to pull a
chisel plow (Figure 4.74). The depth of the compost media layer was approximately 0.60
m. The bedded pack had dimensions of 45.8 m x 31.7 m and the stock density was 11.0 m2
per cow (Appendix A - Table A.1).
The surface pack temperature measured was 11.8°C (±8.4). The average of the pack
temperature in the depth 0.1 m [32.8°C (±17.2)] was less than the average of the depth 0.2
m [35.1°C (±13.3)], as can been see in Appendix A (Table A.4). Paired t-test statistical

150 
 
analysis indicated that at the 0.05 significance level, the averages for the two depths were
not different.
The average bedding moisture was 61.7% (±3.7). The minimum bedding moisture was
56.8% and the maximum was 68.7% (Appendix A - Table A.5). The first and third
quartiles were 60.6% and 62.3%, respectively.

(a) (b)
Figure 4.74 - Barn 31 was using green sawdust and chisel plow was used to till.

Barn 32 was bedded with green sawdust (Figure 4.75a). Aerating was done twice daily
(summer and winter) with a chisel plow that turns over the compost 0.15 m – 0.20 m below
the surface (Figure 4.75b). The dimensions of the bedded pack were 30.5 m x 18.3 m.
Approximately 47 cows were kept in the bedded pack that provided a stock density of 11.9
m2 per cow (Appendix A - Table A.1).
The average surface temperature measured was 6.7°C (±4.4). When comparing the
pack temperature of the two depths, it was found that the average of the depth 0.1 m was
less than the average of the depth 0.2 m (Appendix A - Table A.4). Paired t-test statistical
analysis indicated that the averages for the two depths were different (P < 0.05).
The average bedding moisture was 66.4% (±1.3), which was above NRAES-54 (1992),
recommended level of 45% to 60%. The minimum bedding moisture was 64.8% and the
maximum was 68.3%. The first and third quartiles were 65.6% and 67.3%, respectively.
The average moisture content of bedding in this barn is given in Appendix A (Table A.5).

151 
 
 (a) (b)
Figure 4.75 - Chisel plow was used to aeration the bedding compost in CBP barn 32.

Barn 33 was using kiln dried wood shavings or sawdust (Figure 4.76a). The pack was
aerated twice a day (summer and winter) using chisel plow (Figure 4.76b). The pack was
not cleaned out. The compost pack appeared to be compacted and chunks were seen after
stirring. The compost appeared relatively wet and did stick to the boots when walking on
the surface. The bedded pack had dimensions of 61.0 m x 15.6 m and the stock density was
10.8 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was -1.5°C (±16.1). The average of the pack
temperature in the depth 0.1 m [33.0°C (±27.0)] was less than the average of the depth 0.2
m [37.4°C (±25.4)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were different (P < 0.05).
The average bedding moisture was 69.5% (±0.9). The minimum bedding moisture was
68.2% and the maximum was 70.5% (Appendix A - Table A.5). The first and third
quartiles were 68.7% and 70.4%, respectively.

(a) (b)
Figure 4.76 - Chisel plow was used to aeration the pack compost in CBP barn 33.

Barn 34 owner was using kiln dried wood shavings or sawdust for bedding (Figure
4.77). This barn was stirred at least two time each day (summer and winter) when the cows

152 
 
are being milked. The pack was aerated 0.35 m depth, however experienced compost barn
operators suggest that the pack should be stirred to a depth 0.25 m to 0.30 m twice a day
(Janni et al., 2007). Bedded pack stirring was done using a cultivator attached to the back
of a tractor. The bedded pack had dimensions of 29.3 m x 11.0 m and the stock density was
3.0 m2 per cow (Appendix A - Table A.1).
The average bedding temperatures at surface and 0.1 m and 0.2 m depths were -0.8°C
(±3.7), 27.6°C (±29.3), and 23.3°C (±3.9), respectively. Paired t-test statistical analysis
indicated that the averages for the two depths were not different (P < 0.05). The average
temperature of bedding in this barn is given in Appendix A (Table A.4).
The average bedding moisture was 68.9% (±2.0), which is high than the recommended
for composting of 45% to 60% moisture (NRAES-54, 1992). The minimum bedding
moisture was 66.2% and the maximum was 71.9% (Appendix A - Table A.5). The first and
third quartiles were 67.3% and 70.3%, respectively.

(a) (b)
Figure 4.77 - Barn 34 was using kiln dried wood shavings or sawdust for bedding.

Barn 35 was bedded with green sawdust (Figure 4.78a), but it is not recommended to
start with the compost dairy barn using green sawdust (Crary et al.. 2005). The bedded
pack was aerated once a day using rotary tiller on a tractor. The rotary tiller was mounted
in 3 point diggers on the back of a tractor (Figure 4.78b). This farmer initially started with
0.35 m to 0.40 m depth of green sawdust. The bedded pack had dimensions of 36.6 m x 8.5
m and the stock density was 10.3 m2 per cow. This barn was cleaned out 2.5 times per year.
The surface pack temperature measured was -3.5°C (±2.6). The average of the pack
temperature in the depth 0.1 m was less than the average of the depth 0.2 m. as can be seen
in Appendix A (Table A.4). Paired t-test statistical analysis indicated that these two
averages were different (P < 0.05).

153 
 
 The average bedding moisture measured was 62.6% (±5.6), ranging from 48.7% to
69.4% (Appendix A - Table A.5). The first and third quartiles were 62.5% and 64.6%,
respectively.

(a) (b)
Figure 4.78 - Barn 35 was aerated using rotary tiller.

Barn 36 was bedded with green sawdust. Aerating was done twice daily with a chisel
plow that turns over the compost 0.15 m – 0.20 m below the surface. The dimensions of the
bedded pack were 43.9 m x 14.6 m. The stock density was 6.9 m2 per cow (Appendix A -
Table A.1). The wet compost was found in this pack, with mainly slurry consistency
(Figure 5.79). This barn was cleaned out three times per year. Addition of fresh material is
recommended when the bedded pack becomes moist enough for it to stick to the cows.
Average bedding temperatures at surface and 0.1 m and 0.2 m depths were 3.3°C
(±2.5), 15.7°C (±6.4), and 18.2°C (±5.8), respectively (Appendix A - Table 4). Paired t-test
statistical analysis indicated that the averages for the two depths were different (P < 0.05).
The average bedding moisture [71.8% (±0.8)] was the largest that has been found
among the farms studied, ranging from 70.8% to 73.3% (Appendix A - Table A.5). The
first and third quartiles were 71.3% and 71.9%, respectively. Bedding needs to stay dry for
the benefit of keeping cows clean and somatic cell count (SCC) low.

(a) (b)
Figure 4.79 - Bedded with green sawdust and chisel plow used in CBP barn 36.

154 
 
 Barn 37 was bedded with green sawdust. Tilling was done once a day with a chisel
plow that turns over the compost 0.15 m – 0.20 m below the surface. The same chisel plow
was used in this barn and Barn 38 (Figure 4.80). A load of green sawdust was added every
three weeks. In the winter time, fresh green sawdust was necessary to be added more
frequently. This barn was cleaned out twice a year. Care must be taken when removing the
bedded pack during clean out, or during pack stirring when the bedded pack is less than
0.30 m deep, to avoid disturbing the base of the barn (Janni et al., 2005). The pack had
dimensions of 45.8 m x 10.7 m and the stock density was 10.0 m2 per cow.
The average surface temperature measured was 4.7°C (±2.9). The average of the pack
temperature in the depth 0.1 m and 0.2 m were 20.2°C (±15.3) and 25.3°C (±16.6),
respectively. Paired t-test statistical analysis indicated that the averages for the two depths
were different (P < 0.05). Details can be seen in Appendix A (Table A.4)
The average bedding moisture was 64.3% (±3.5), which was above NRAES-54 (1992)
recommended level of 45% to 60%. The minimum bedding moisture was 57.0% and the
maximum was 68.7%. The first and third quartiles were 63.4% and 66.5%, respectively.
The average moisture content of bedding in this barn is given in Appendix A (Table A.5).

(a) (b)
Figure 4.80 - Bedded with green sawdust and chisel plow used in CBP barn 37.

Barn 38 was bedded with green sawdust (Figure 4.81). The compost appeared relatively
loose and dry and did not stick to the boots when walking on the surface. The bedded pack
was aerated once a day, at least 0.20 m deep. According to Endres (2009), the bedded pack
should be aerated twice daily, at least 0.25 m deep, to keep it aerobic and fluffy, hence
biological activity generates heat and helps dry the pack. This barn was cleaned out twice a
year and a load of fresh green sawdust was added every three weeks. The pack had
dimensions of 36.6 m x 21.7 m and the stock density was 15.9 m2 per cow.

155 
 
 The average bedding temperatures at surface was 3.8°C (±2.3). Paired t-test statistical
analysis indicated that at the 0.05 significance level, the averages for the two depths were
different. When comparing the pack temperature of the two depths, was found that the
average of the depth 0.1 m was less than the average of the depth 0.2 m (Appendix A -
Table A.4).
The average bedding moisture was 61.7% (±6.4). The minimum bedding moisture was
50.5% and the maximum was 67.7% (Appendix A - Table A.5). The first and third
quartiles were 56.9% and 66.7%, respectively.

Figure 4.81 - Barn 38 was bedded with green sawdust.

Barns 39 and 40 were bedded using green sawdust (Figure 4.82a). These barns were
stirred at least twice a day (summer and winter) when the cows are being milked. The
packs were aerated 0.18 m depth. Bedded packs stirring were done using a chisel plow
attached to the back of a tractor (Figure 4.82b). These barns had the same bedded pack
dimensions of 48.8 m x 22.0 m. The stock density of the barn 39 and 40 were 12.2 m2 per
cow and 12.6 m2 per cow, respectably. Details can be seen in Appendix A (Table A.1).
The average surface temperature measured in Barns 11 and 12 were -1.7°C (±4.5) and
0.0°C (±8.3), respectively. Paired t-test statistical analysis indicated that there were
significant (P < 0.05) differences in average temperatures for the two depths in both of
barns (Appendix A - Table A.4).
The average bedding moisture in Barns 39 and 40 were 66.5% (±1.0) and 64.6% (±1.7),
respectively. Details can be seen in Appendix A (Table A.5).

156 
 
 (a) (b)
Figure 4.82 - Bedded with green sawdust and chisel plow used in CBP barns 39 and 40.

Barn 41 was bedded with kiln dried wood shavings or sawdust. This farmer used chisel
plow (Figure 4.83) to aeration once a day (summer and winter). The pack was aerated 0.15
m depth. According to Endres (2009), some producers in Minnesota are aerating the pack
deeper (about 0.35 – 0.40 m) using a chisel plow type of equipment, and they have
observed a reduction in bedding needs and increased pack temperatures. Clean bedding
was added every time the bedded pack became moist enough for it to adhere to the cows
and a load of fresh material was added every three weeks. The pack had dimensions of 30.5
m x 15.3 m and the stock density was 5.2 m2 per cow (Appendix A - Table A.1).
The surface pack temperature measured was -4.0°C (±2.3). The average of the pack
temperature in the depth 0.1 m [25.2°C (±15.1)] was less than the average of the depth 0.2
m [30.0°C (±5.9)], as can been see in Appendix A (Table A.4). Paired t-test statistical
analysis indicated that these two averages were different (P < 0.05).
The average bedding moisture was 70.1% (±2.7), which is high than the recommended
for composting of 45% to 60% moisture (NRAES-54, 1992). The minimum bedding
moisture was 65.2% and the maximum was 71.7% (Appendix A - Table A.5). The first and
third quartiles were 69.1% and 71.7%, respectively.

(a) (b)
Figure 4.83 - Kiln dried wood shavings and chisel plow used in CBP barn 41.
157 
 
 Barn 42 was bedded with mix material. Tilling was performed twice a day in the
summer and winter time, using a chisel plow with 0.20 m shovels (Figure 4.84). The chisel
plow tills to a depth of 0.14 m – 0.18 m. This farmer cleaned out once a year and leave
about 0.05 m of old material in the barn to help initiate microbial activity. This pack was
cleaned out one time per year. The bedded pack had dimensions of 53.7 m x 14.9 m and the
stock density was 8.0 m2 per cow, as can been see in Appendix A (Table A.1).
The bedding temperatures measured at two depths were 18.2°C (±25.1) and 24.8°C
(±5.1) and the pack surface temperatures was 0.9°C (±3.5). Paired t-test statistical analysis
indicated that these two averages were different (P < 0.05).
The average bedding moisture was 69.1% (±1.2). The minimum bedding moisture was
67.6% and the maximum was 70.7%. The first and third quartiles were 68.3% and 70.2%,
respectively. More details about Barn 42 are given in Appendix A (Table A.1).

(a) (b)
Figure 4.84 - Chisel plow was used to aeration in CBP barn 42.

4.2.4. Environment characteristics

The average temperature gradient (taken at 0.05 m and 1.20 m above the pack) was
-0.52 °C · m-1 (± 1.0), indicating that there was a reduction of the air temperature along the
height measurement. Paired t-test statistical analysis indicated that at the 0.05 significance
level, the averages air temperatures for the two heights were different. The average surface
temperatures on the pack were similar to the average ambient temperature. During the
winter, the temperature would be the first indicator of changes in bed performance (Taraba
and Bewley, 2010). Air temperatures tended to be lower inside barns which utilized
Overshot or Capped ridge, mechanical ventilation system and ridge direction of SW or NW
(Table 4.1). Building temperatures must remain above freezing most of the time to prevent
frozen manure in the pack. This type of building requires insulation (curtains) and a
controlled ventilation system. Most compost packs were warmer than the outdoor air
158 
 
temperature. This indicates that the packs were heating and microbial activity was
occurring.
The average relative humidity gradient in this study was 2.35 % · m-1 (± 0.5). Thus, the
majority of barns tend to increase relative humidity along the height. Paired t-test statistical
analysis indicated that the averages relative humidity for the two heights were different (P
< 0.05). In some barns, sawdust bedding was used to keep cattle comfortable and dry in
higher amounts which increases barn operating costs.
The average of air velocity gradient was -0.22 m s-1 · m-1 (± 0.3), indicating that the air
velocity decreased along the height. Paired t-test statistical analysis indicated that at the
0.05 significance level, the averages air velocity for the two heights were different. Thus,
during the season of fall and winter in Kentucky, the barn orientation was not ideal for air
movement since air generally moving along the length meets higher resistance than air
moving across the building.  Sufficient air exchange is needed in cold weather to remove
moisture from the pack and extend the time between bedding addition. Overall, the low
sidewalls of the barns promoted high air velocity across the barns. Air movement through
the barn should be sufficient to maintain inside temperature only slightly above outside
temperature in winter and slightly below outside temperature in summer. A considerable
differences between air velocity indoor and outdoor were observed in barns 1, 5, 6, 11, 12,
13, 14, 16, 17, 20, 22, 23, 24, 25, 37, and 38. Buildings and other obstacles (trees, hills, and
tractors) around of these barns blocked for airflow and results in more high relative
humidity, as can be seen in Table 4.1. On the other hand, high airflow during the winter
can reduce the air temperature indoor and the pack temperature. In those conditions,
windbreaks become very important. In the winter, side curtains can be an important
investment to reduce barn ventilation and winter winds from extracting too much heat from
a compost bed (Taraba and Bewley, 2010).

159 
 
Table 4.1 - Average air temperatures, relative humidity and air velocity inside and outside in 42 CBP barns in Kentucky.
Air Temperature (°C) Relative Humidity (%) Air velocity (m/s) Outside weather
Air Air
0.05 m 1.2 m 0.05 m 1.2 m 0.05 m 1.2 m R. H.
Barn temp. veloc.
Std Std Std Std Std Std
Aver. Aver. Aver. Aver. Aver. Aver. (°C) (%) (m/s)
Dev Dev Dev Dev Dev Dev
1 16.7 ± 7.1 17.8 ± 0.6 79.9 ± 2.7 80.7 ± 1.2 1.3 ± 1.4 2.1 ± 2.1 16.7 73.0 0.9
2 20.2 ± 0.7 20.9 ± 0.5 43.1 ± 3.8 34.8 ± 3.5 0.7 ± 0.6 0.5 ± 0.3 24.6 16.5 0.5
3 22.4 ± 7.1 21.6 ± 1.2 34.0 ± 9.3 22.3 ± 1.1 0.6 ± 0.3 0.8 ± 0.8 24.6 16.5 1.7
4 15.4 ± 1.0 14.9 ± 1.0 59.4 ± 3.9 53.6 ± 2.3 1.0 ± 0.4 1.4 ± 1.0 15.0 38.0 2.7
5 24.7 ± 1.4 24.8 ± 1.4 70.1 ± 3.6 66.9 ± 5.1 0.7 ± 0.4 1.0 ± 0.8 26.8 55.7 0.3
6 16.2 ± 1.6 15.2 ± 0.9 88.4 ± 1.1 88.6 ± 1.0 0.4 ± 0.2 0.5 ± 0.3 19.4 84.0 0.3
7 15.9 ± 1.2 16.1 ± 0.8 48.1 ± 4.8 39.7 ± 2.6 0.8 ± 0.7 1.7 ± 1.2 16.2 36.0 2.3
8 19.4 ± 3.3 20.3 ± 0.5 52.3 ± 5.2 43.5 ± 1.8 0.5 ± 0.2 1.1 ± 0.5 21.4 35.5 1.3
9 23.3 ± 2.3 23.3 ± 1.7 40.8 ± 8.4 30.5 ± 2.2 0.4 ± 0.2 0.7 ± 0.7 21.4 35.5 1.3
10 20.0 ± 0.5 20.1 ± 0.3 49.8 ± 5.6 40.9 ± 1.9 0.5 ± 0.2 0.8 ± 0.2 18.8 38.9 0.8
11 14.9 ± 1.3 14.0 ± 1.1 85.1 ± 4.6 84.6 ± 5.5 0.4 ± 0.2 0.5 ± 0.2 9.4 98.0 0.3
12 15.6 ± 1.6 15.0 ± 3.1 86.1 ± 1.7 84.4 ± 3.1 0.3 ± 0.1 0.4 ± 0.3 9.4 98.0 0.3
13 12.8 ± 3.0 12.6 ± 2.8 81.3 ± 2.2 80.9 ± 2.4 1.8 ± 0.6 2.8 ± 1.1 9.3 84.6 1.0
14 10.5 ± 1.5 9.8 ± 2.1 83.7 ± 2.2 82.8 ± 1.6 2.8 ± 1.5 3.1 ± 1.9 9.3 84.6 1.0
15 15.3 ± 2.1 14.0 ± 2.1 58.7 ± 5.6 53.9 ± 3.4 0.7 ± 0.6 0.9 ± 0.7 12.9 52.0 2.2
16 9.7 ± 1.2 8.6 ± 0.7 92.0 ± 1.6 91.9 ± 1.7 1.5 ± 0.2 1.9 ± 0.8 8.9 67.0 1.5
17 10.9 ± 1.5 9.2 ± 0.6 88.7 ± 2.1 89.5 ± 1.5 0.6 ± 0.5 0.6 ± 0.5 6.7 95.0 0.6
18 0.6 ± 7.0 -1.9 ± 0.7 56.7 ± 1.2 55.2 ± 1.7 0.5 ± 0.5 0.7 ± 0.6 0.1 43.8 6.3
19 -0.8 ± 4.8 -4.2 ± 1.7 57.7 ± 6.7 59.9 ± 5.3 0.4 ± 0.3 0.5 ± 0.6 -3.4 63.2 0.5
20 25.7 ± 0.9 26.0 ± 0.5 42.4 ± 13.5 33.4 ± 1.0 1.6 ± 0.8 2.0 ± 1.1 25.9 33 1.4
21 4.3 ± 1.5 2.9 ± 0.5 62.9 ± 6.2 62.3 ± 1.6 1.0 ± 0.6 1.5 ± 0.6 -0.4 50.0 3.2
22 7.9 ± 0.4 7.1 ± 0.3 80.8 ± 2.4 81.3 ± 1.6 0.7 ± 0.4 0.8 ± 0.5 6.9 77.7 0.7
23 8.0 ± 0.9 7.6 ± 0.7 90.6 ± 1.5 91.3 ± 1.4 0.5 ± 0.5 0.6 ± 0.5 6.9 77.7 0.5
Continue...

160 
 
Table 4.1 - Average air temperatures, relative humidity and air velocity inside and outside in 42 CBP barns in Kentucky.
Air Temperature (°C) Relative Humidity (%) Air velocity (m/s) Outside weather
Air Air
0.05 m 1.2 m 0.05 m 1.2 m 0.05 m 1.2 m R. H.
Barn temp. veloc.
Std Std Std Std Std Std
Aver. Aver. Aver. Aver. Aver. Aver. (°C) (%) (m/s)
Dev Dev Dev Dev Dev Dev
24 7.7 ± 1.3 7.2 ± 1.2 89.2 ± 1.8 84.4 ± 15.6 0.9 ± 0.4 1.0 ± 0.4 6.9 77.7 0.7
25 7.3 ± 0.7 6.2 ± 0.7 86.3 ± 2.2 88.8 ± 1.3 0.4 ± 0.2 0.6 ± 0.3 5.7 83.5 0.5
26 5.8 ± 1.4 6.5 ± 6.4 61.8 ± 2.7 58.9 ± 5.8 0.6 ± 0.2 0.5 ± 0.3 3.3 58.6 1.5
27 12.1 ± 4.0 10.5 ± 3.5 59.6 ± 6.8 55.3 ± 5.7 0.3 ± 0.1 0.2 ± 0.1 5.4 42.4 3.4
28 6.9 ± 1.4 6.5 ± 1.0 56.2 ± 4.0 53.0 ± 3.6 0.3 ± 0.1 0.3 ± 0.2 -0.6 55.0 0.9
29 6.1 ± 1.3 6.0 ± 1.2 58.8 ± 3.8 54.0 ± 1.9 0.5 ± 0.2 0.7 ± 0.3 3.8 52.3 2.6
30 9.0 ± 4.1 7.3 ± 1.0 57.0 ± 13.3 54.0 ± 1.9 1.1 ± 0.7 1.3 ± 0.7 8.1 49.9 1.6
31 10.7 ± 2.8 9.8 ± 1.7 53.1 ± 4.1 47.4 ± 1.7 0.4 ± 0.2 0.6 ± 0.4 5.4 42.4 3.4
32 3.1 ± 1.5 2.6 ± 1.5 67.1 ± 2.8 67.5 ± 2.3 0.3 ± 0.1 0.2 ± 0.1 0.4 64.4 1.0
33 0.0 ± 1.7 -1.2 ± 1.0 64.0 ± 4.9 66.3 ± 3.9 0.9 ± 0.5 1.1 ± 0.6 -2.6 61.5 1.4
34 1.4 ± 0.9 0.7 ± 0.6 66.4 ± 3.4 67.6 ± 1.5 0.4 ± 0.2 0.3 ± 0.1 0.4 68.8 3.1
35 2.5 ± 4.6 2.4 ± 1.9 49.6 ± 4.2 50.6 ± 3.1 0.3 ± 0.2 0.5 ± 0.3 0.7 51.7 1.7
36 3.6 ± 2.3 3.7 ± 0.6 75.9 ± 2.5 76.1 ± 2.3 0.3 ± 0.2 0.2 ± 0.1 1.7 71.8 0.5
37 5.3 ± 0.9 4.7 ± 0.7 80.0 ± 1.7 80.0 ± 1.1 0.5 ± 0.3 0.5 ± 0.3 1.1 89.0 0.5
38 5.7 ± 0.4 5.3 ± 0.6 82.4 ± 1.8 82.7 ± 1.9 0.4 ± 0.2 0.4 ± 0.2 1.1 89.0 0.4
39 -2.0 ± 1.0 -2.6 ± 0.9 70.4 ± 1.8 69.7 ± 2.3 1.4 ± 0.7 3.7 ± 6.2 -3.4 65.3 2.5
40 -0.5 ± 3.1 -1.6 ± 0.9 66.6 ± 2.2 64.8 ± 2.5 0.8 ± 0.4 1.2 ± 0.5 -3.4 65.3 2.5
41 -1.4 ± 0.6 -2.0 ± 1.7 59.6 ± 2.8 59.8 ± 1.7 0.8 ± 0.5 0.6 ± 0.3 -2.1 59.2 2.5
42 -0.4 ± 1.1 -1.0 ± 1.3 64.2 ± 3.1 62.7 ± 2.6 0.3 ± 0.3 0.6 ± 0.5 -2.4 54.6 2.5

161 
 
4.2.5. Herd characteristics
The following section summarizes information collected by a collaborator (Black,
2012). It is included to give a more full picture of the benefits of this type barn.
The cow herd size, as reported by the dairymen, was adjusted for breed size and milk
production. The average herd size on the pack in the study was 89.7 cows, ranging from 19
to 184 cows and dry cow numbers ranged from 3 to 75 dry cows with an average of 25.1
dry cows (Figure 4.85). Dry cows were housed on pasture, in a separate or the same area of
the compost barn as the lactating cows, or in an alternative housing area. Breeds consisted
of Jersey (6.8%), Holstein (72.7%), and mix (20.5%). Average milk production of cows
was 27.2 kg per day with a mean fat and protein percentage of 3.85% and 2.35%,
respectively.

Figure 4.85 - Herd size and dry cows number in 42 CBP barns.

4.2.6. Producer Responses

All producers visited were very satisfied with their compost barns. They observed that
cows were more comfortable (24.2%) and that this housing system resulted in increased
cow cleanliness (12.1%), see Table 4.2 below for more details.
162 
 
Table 4.2 - Producer cited benefits of 42 CBP barns.
Benefits Percent (%)
Improved cow comfort 24.1
Improved cow cleanliness 12.1
Low maintenance 9.5
Good for heifers, lame, fresh, problem, and old cows 8.6
Natural resting position (no stalls) 7.8
Improved feet and legs 6.9
Proximity to parlor (compared to pasture) 6.9
Decreased SCC 5.2
Increased heat detection 5.2
Ease of manure handling 2.6
Increased dry matter intake (compared to pasture) 2.6
Increased production 2.6
Increased longevity 2.6
Fewer leg and teat injuries 1.7
Minimizes time standing on concrete 1.7

However, results of the survey revealed some aspects of the CBP barn producers would
change. Their main concern was the size or capacity of the barn (25%). Further, many
producers indicated the need for retaining wall (10%) and curtains (8.3%). Other
recommended facilities changes are cited in Table 4.3.

Table 4.3 - Recommended facilities changes of CBP barns producers.

Recommended Percent (%)
Increase size or capacity of the barn 25.0
Higher sidewalls and improved ventilation 20.0
Add a retaining wall 10.0
Add curtains 8.3
More fans 8.3
Larger ridge vent 8.3
No posts in pack 6.7
Change number or location of waterers 6.7
Change location or size of feed bunk 6.7

163 
 
 The majority of producers were concerned about the supply of bedding. Many
producers were seeking secure bedding supply (17.7%) and alternative bedding sources
that will work in the CBP (9.7%). Producer recommendations also included the
importance of maintaining the pack (8.1%), and stirring the pack two times per day or
more frequently (14.5%), using kiln-dried shavings (9.7%), avoid the use of straw, wheat
straw, corn fodder, or soybean fodder or pine as a bedding material (9.7%), and
maintaining the pack by keeping moisture low (8.1%), as can be seen in Table 4.4. Some
producers (4.8%) reported that touring other barns was an important influence on their
choice of structure. Producers also reported that they spoke with other producers managing
a CBP barn, they created their own plans, they reviewed freestall plans and designs,
economics, and ease of management. Producers learned that using kiln-dried shavings,
keeping the pack stirred, and avoiding the use of straw in the bedding, would help to
prevent other mistakes.

Table 4.4 - Producer recommendations and lessons learned of CBP barns producers.
Recommendations and lessons learned Percent (%)
Secure a bedding supply 17.7
Do not use straw, wheat straw, corn fodder, bean fodder, or pine 9.7
Add bedding frequently 6.5
Tour other barns 4.8
Stir two times per day or more frequently 14.5
Keep pack maintained and moisture low 8.1
Build the barn large 6.5
Add curtains 4.8
Use kiln-dried shavings 9.7
Minimum of 9.3 m2 per cow 8.1
Designated tractor for stirring 4.8
Do not start pack during winter 4.8

164 
 
4.3. Physical, chemical, and thermal properties of compost materials
4.3.1. Compost particle size distribution
Dried compost was sieved to measure the portion of each particle size range (PS2, PS3,
PS4, and PS5) comprising the total amount. Three samples of each particle size range were
used to determinate the compost particle size distribution. Particle size distributions on a
weight basis (for compost from CBP barns using Green sawdust, Kiln-dried shavings or
sawdust, and Mix compost) are presented in Figures 4.86.
A comparison of the particle weight fraction distributions content in Figure 4.86 reveals
that these compost materials from the CBP barns using different bedding compost materials
are very different. The following percentages and standard deviations were found: in Green
sawdust, in the particle size ranges together (PS2 + PS3 + PS4 + PS5) comprised 60.95 ±
10.0% of the total weight, while coarser material represented 3.46 ± 3.4% and finer
material, 35.60 ± 11.3%. In Kiln-dried shavings or sawdust, the compost particle size
distribution were 25.44 ± 13.1% (finer material), 25.45 ± 8.5% (PS2), 4.12 ± 1.0% (PS3),
10.50 ± 3.0% (PS4), 28.1 ± 14.2% (PS5), and 6.32 ± 8.1% (coarser material). In this plot
(Figure 4.85), the distribution values of Mix materials increased in the order of Finer < PS2
< PS5 < PS4 < PS3 < Coarser fractions. In CBP barns using Mix compost material, the
lower and higher compost particle size distribution were 14.38% (Coarser > 25.00 mm) and
48.66% (Finer < 2.00 mm), respectively.
The analysis of variance between of compost particle ranges from all CBP barns was
obtained and is shown in Figure 4.87 at the significance level of p < 0.05. The Tukey test
shows that there was a significant difference between average values of compost particle
ranges. This results indicate that the distribution of mass weight increased in the order of
Coarser = PS3 < PS4 < PS5 = PS2 < PS2 = Finer. The average finer index expressed as
weight percent in the samples studied was 30.1%. Details about statistic analysis can be
seen in Appendix F (Table 1).

165 
 
  
Figure 4.86 - Particle Weight Fractions for compost bedded from CBP barn using Green
sawdust, Kiln-dried shavings or sawdust, and Mix compost.

Figure 4.87 - Analysis of variance between compost particle ranges for all bed compost
materials. Values followed by different letters are significantly different (P <
0.05; Tukey).

4.3.2. Particle density, bulk density, and porosity

Particle density, bulk density, and porosity of 42 compost bulking material results are
shown in Table 4.5.
Particle density of all compost material studied ranged from 0.86 to 1.11 g cm-3 with an
average of 0.98 g cm-3 and a standard deviation of 0.02 (Table 4.5). Generally, the compost
materials of CBP barns 9, 11, 12, 13, 16, 20, 21, 22, 23, 24, 26, 27, 29, 30, 32, 33, 36, 40,
166 
 
41, and 42 show high particle density (≥ 1g cm-3) compared with other materials (Table
4.5). It should be noted that the obtained values are somewhat higher than expected, but
evaporation of water increase the solids content, particularly with well performing beds or
high cow density. Cows that spend time in pasture may return soil to bed.
Results and standard errors of bulk density for compost bed material collected from
each barn are shown below (Table 4.5). The average bulk density was 0.37 ± 0.02 g cm-3
with a range of 0.15 to 0.54 g cm-3. Higher bulk density suggests that the compost have less
pore space and are more compact. The highest bulk density value was 3.6 times that of the
lower bulk density value.
The average porosity of the bedding material in all compost barns was 62.65 ± 2.14%
with a range of 47.43 to 77.32%. In general, the average porosity increased when the bulk
density decreased, significantly higher porosity was found in Barns 41 (77.32%). The
variation of average porosity in all compost bulking material was similar to those bulk
density in the reverse order due to the inverse relationship between bulk density and
porosity (Table 4.5). So, high bulk density and low porosity might cause poor aeration and
limiting oxygen concentration uptake by microorganisms and inhibiting microorganisms
growth. Previous researchers have reported the optimum and minimum range of porosity
for aerobic process in compost systems was 85 to 90%, and 30%, respectively (Schulze,
1962; Rynk, 1992; Nappi and Barberis, 1993).

Table 4.5 - Particle density, bulk density, and porosity in 42 compost-bulking materials, as
well as standard deviations (n = 3).
Particle density Bulk density Porosity
Barn -3 -3
(g cm ) (g cm ) (%)
1 0.95 ± 0.06 0.50 ± 0.06 77.12 ± 8.23
2 0.98 ± 0.06 0.47 ± 0.01 72.30 ± 3.30
3 0.92 ± 0.04 0.38 ± 0.02 65.41 ± 2.25
4 0.86 ± 0.04 0.31 ± 0.01 65.40 ± 1.84
5 0.93 ± 0.07 0.47 ± 0.03 60.53 ± 3.26
6 0.91 ± 0.05 0.45 ± 0.01 60.50 ± 3.40
7 0.97 ± 0.06 0.37 ± 0.01 62.07 ± 1.65
8 0.99 ± 0.05 0.47 ± 0.13 53.63 ± 11.28
9 1.04 ± 0.05 0.54 ± 0.02 47.43 ± 3.61
10 0.94 ± 0.06 0.45 ± 0.01 51.92 ± 2.30
11 1.01 ± 0.03 0.44 ± 0.04 56.52 ± 4.34
12 1.05 ± 0.07 0.45 ± 0.02 56.70 ± 4.75
Continue...

167 
 
Table 4.5 - Particle density, bulk density, and porosity in 42 compost-bulking materials, as
well as standard deviations (n = 3).
Particle density Bulk density Porosity
Barn -3 -3
(g cm ) (g cm ) (%)
13 1.00 ± 0.01 0.35 ± 0.03 67.37 ± 2.18
14 0.99 ± 0.03 0.32 ± 0.01 65.21 ± 2.01
15 0.92 ± 0.03 0.28 ± 0.05 68.64 ± 6.60
16 1.01 ± 0.02 0.24 ± 0.01 76.72 ± 1.57
17 0.98 ± 0.04 0.44 ± 0.01 55.31 ± 2.83
18 0.90 ± 0.02 0.40 ± 0.02 56.00 ± 2.18
19 0.92 ± 0.11 0.27 ± 0.01 71.99 ± 4.26
20 1.06 ± 0.01 0.48 ± 0.03 54.81 ± 2.62
21 1.01 ± 0.02 0.24 ± 0.01 75.90 ± 1.02
22 1.00 ± 0.01 0.39 ± 0.01 60.94 ± 0.96
23 1.02 ± 0.06 0.36 ± 0.00 64.73 ± 2.43
24 0.97 ± 0.01 0.28 ± 0.01 70.80 ± 1.10
25 1.00 ± 0.03 0.39 ± 0.01 61.11 ± 0.41
26 1.11 ± 0.01 0.42 ± 0.01 62.29 ± 0.75
27 0.94 ± 0.05 0.34 ± 0.02 64.10 ± 0.94
28 1.04 ± 0.05 0.43 ± 0.01 58.88 ± 2.64
29 1.06 ± 0.04 0.35 ± 0.02 66.83 ± 1.20
30 0.97 ± 0.04 0.37 ± 0.00 61.68 ± 2.11
31 1.04 ± 0.02 0.39 ± 0.01 62.58 ± 1.81
32 1.00 ± 0.00 0.42 ± 0.01 57.80 ± 0.65
33 0.93 ± 0.02 0.33 ± 0.02 64.66 ± 2.52
34 0.97 ± 0.02 0.27 ± 0.02 71.80 ± 2.12
35 1.01 ± 0.01 0.40 ± 0.02 59.73 ± 1.78
36 0.95 ± 0.00 0.41 ± 0.01 57.26 ± 1.44
37 0.91 ± 0.01 0.38 ± 0.06 58.34 ± 6.60
38 0.87 ± 0.04 0.38 ± 0.02 56.11 ± 4.06
39 1.03 ± 0.02 0.40 ± 0.02 61.39 ± 0.78
40 1.06 ± 0.03 0.34 ± 0.03 57.50 ± 1.87
41 1.05 ± 0.00 0.20 ± 0.03 77.32 ± 2.67
42 1.05 ± 0.01 0.32 ± 0.04 54.20 ± 3.78
Average 0.98 ± 0.02 0.38 ± 0.02 62.65 ± 2.14

4.3.3. Water holding capacity

The water holding capacity (WHC) was evaluated in this study to determine whether
the beds operated under moisture conditions that would allow drainage of contaminated
water to compost bed base which could infiltrate into groundwater. Since WHC has very
important relationship with physical and biological properties of bed compost materials, the
moisture levels were set based on the WHC to find the real structure and texture conditions
168 
 
of compost providing aeration and microbial activity. So, WHC was achieved in the first 12
hours of natural drainage from an initial moisture content greater than the media WHC.
The values of WHC and moisture content in 42 bed compost materials are illustrated in
Figure 4.88. The average WHC of all bed compost was at 72.7% on a wet weight basis
with a range of 53.6 to 79.8%. WHC was lower for bed compost in CBP barn 1 to the other
materials tested and was able to hold water up to 53.6% w.b. moisture content. As
expected, the smaller particle size range was able to hold more water (Table 4.5). WHC
increases with increasingly fine particle size (decreasingly porosity), as can be seen in
Figure 4.88. Coarse particle size of compost have a lower WHC since they are high in large
pores subject to free drainage. Fine particle size have a greater occurrence of small pores
that hold water against free drainage, resulting in a higher WHC values. Therefore, beds
typically operate under MC conditions that would not allow drainage. Good operating
range was 40-60% which is below WHC and that groundwater drainage would not occur
when bed managed properly.
The present study indicates a positive relationship between WHC and moisture content
(r = 0.383, P < 0.05), see Figure 4.88. Thereby, WHC increased with increasing moisture
content. Ahn et al. (2008) reported that WHC increased with increasing moisture content
until certain level. They suggested that moisture promotes aggregation of particles, and
then decreased porosity. Finally, the lower porosity decreases the WHC.
The WHC method was adequate to estimate the maximum amount of moisture that
could be added to the bed compost without the occurrence of adequate drainage or runoff,
thus avoiding contamination of the soil or water outside the barn.

169 
 
  
Figure 4.88 - Water holding capacity (w. b., %), initial moisture content (w. b., %), and
porosity (%) of 42 bed compost materials.

170 
 
4.3.4. Chemical properties of composts
Results related to total nitrogen (N), total carbon (C), and carbon-nitrogen ratio (C:N)
of each bed compost material are shown in Figure 4.89. Each value is an average of nine
samples collected as previously described (Figure 3.5). Further results regarding other
chemical compounds are shown in Appendix A (Table 6).
The average total C was 42.7 ± 3.8% with a range of 29.8 to 47.1%. One of the major
constituent of the compost, C, was most abundant (47.1%) in CBP barn 19. Only in the
barn 9 contain average total C lower than 30.0%. Shane et al. (2010) reported a range total
C of 12.7 to 20.1% in six compost barns in Minnesota collected in two different depth.
Thus, the bed composts in Kentucky have not stabilized to the extend in this study. In
general, many of the beds had not been operating properly, but as a result, the dairy
producers learned how to improve beds to operate properly.
The average total N was 1.6% ranged from 0.5 to 2.5%. Shane et al. (2010) reported
an average total N of 0.76% with a range of 0.61 to 0.89% during the winter. Barberg et al.
(2007) reported an average total N of 2.54% with a range of 0.57 to 4.22%. Russelle et al.
(2009) reported average N of 1.12% in eight compost barns in Minnesota. Janni et al.
(2007) reported an average total N of 0.99% six compost barns in Minnesota. Overall, the
total N from the 42 bed compost materials tested in the study was similar to those reported
previously by Barberg et al. (2007) and Shane et al. (2010). Nitrogen level of compost
material is an important factor in the determination of C:N ratio of the composted material.
Microorganisms for building up their structure of the body require it (Iqbal et al., 2010).
The average C:N ratio was 28.5 ± 8.1. The C:N ratio of barn 10 (15.4) was lowest due
to high nitrogen contents as compared to barn 12 (26.2). In barn 20 had the highest C:N
ratio (43.9) to the other materials tested. Shane et al. (2010) reported the highest C:N ratio
found was 26.0 during winter. Barberg et al. (2007) reported an average of 19.5. Russelle et
al. (2009) reported a range of 11.2 to 20.9. Janni et al. (2007) reported an average C:N ratio
of 15.5. For effective composting C:N ratio of the compost should be ranged 20 to 25
(Diaz et al., 1993). In another words, a C:N ratio below 25:1 may emit ammonia odor,
which may influence the ammonia levels in the compost barns (Rosen et al., 2000). In
general, in the present study it was balance enough. However, the barns 1, 3, 6, 9, 10, and
18 will require an additional source of nitrogen to obtain the optimum bedding compost
C:N ratio between 20 and 25.

171 
 
Figure 4.89 - Total Carbon (C), Total Nitrogen (N), and C:N ration presents in 42 bedding
compost.

4.3.5. Bacterial analysis of composts

Figure 4.90 shows the pack bacterial count (Coliform, E. Coli, Staphylococcal,
Streptococcal, and Bacillus) and the associated moisture content in the bedding compost
material of 42 CBP barns.
Bedding compost always contained bacterial, but the variation was large between
different sampling sites and times. However, Coliforms were not present in Barn 15 that
had a higher compost temperature (Appendix A - Table 4). Coliforms, E. Coli, Bacillus and
Streptococcus count were higher in the barn 4 that showed lower moisture content (36.2%).
The average Coliforms, E. Coli, Streptococcus and Bacillus content in the bedding material
of barn 4 were 247.5 ± 12.0 cfu/g, 173.0 ± 18.38 cfu/g, 3595.0 ± 148.49 cfu/g, and 1810.0
± 608.11 cfu/g, respectably. Also, the environmental Staphylococcus were highest in the
barn 6 that had an average of 2.700 ± 0.00 cfu/g. The present study indicates a inverse
relationship between moisture content and bacterial counts (Coliforms, r = 0.377, P < 0.05;
E. Coli, r = 0.448, P < 0.05; Bacillus, r = 0.372, P < 0.05). There are no significant
relationships between bacterial counts (Streptococcal and Staphylococcal) with moisture
content (P > 0.05).

172 
 
Figure 4.90 - Bacterial analysis of bedding compost samples in 42 CBP barns.

4.3.6. Thermal properties of composts

4.3.6.1. Thermal properties of composts in different moisture content and compaction
degree
The thermal conductivity of unsegregated bed compost materials varied from 0.081 to
0.625 W m-1 K-1 for Green sawdust, 0.071 to 0.618 W m-1 K-1 for Kiln-dried shavings or
sawdust, 0.059 to 0.766W m-1 K-1 for Mix, and 0.105 to 0.406 W m-1 K-1 for average of all
compost material tested, depending upon the moisture content (30, 45, and 60%) and static
compaction degree (fluffy, 0.0, 0.1, 0.2, 0.3, and 0.4 MPa) within the experimental range of
the variables. An increasing trend in the thermal conductivity of all compost was also
observed with the increase in both moisture content and static compaction degree for all
compost material tested (Figure 4.91). The compost with low moisture content exhibited a
low thermal conductivity because of the presence of a relatively high fraction of air (a poor
conductor). The thermal conductivity increased with compaction due to a reduction in the
void space (Chandrakanthi et al., 2005).

173 
 
a) Thermal Conductivity:
Multiple regressions analysis for compost materials showed that there is also a linear
relationship between thermal conductivity (k) and moisture content (MC) and static
compaction degree (SC) as follows:
a) Green sawdust:
kgreen = -0.0619 + 0.311 · SC + 0.478 · MC R2 = 0.838 (4.1)

b) Kiln-dried shavings or sawdust:

kkiln = -0.0679 + 0.342 · SC + 0.487 · MC R2 = 0.877 (4.2)

c) Mix:
kMix = -0.0537 + 0.360 · SC + 0.434 · MC R2 = 0.829 (4.3)

d) Average of all compost material:

kAverage = -0.0612 + 0.338 · SC + 0.467 · MC R2 = 0.852 (4.4)

The response surface presented in Figure 4.91 was almost flat due to the fitted linear
model, except at high moisture content and high static compaction degree, which showed
an abrupt increase. The analysis of variance (Table 4.6) indicates a greater effect of the
static compact degree (high F-value) than that of the moisture content on the thermal
conductivity for Green sawdust and Kiln. However, the analysis of variance indicates a
greater effect of the moisture content than that of the static compaction degree on the
thermal conductivity for Mix. The magnitudes of respective regression coefficients in
equations 4.1, 4.2, 4.3, and 4.4 also confirmed the results. The models seem adequately
fitted based on the observation of high F-value as well as high coefficient of determination
(R2). The models accounted for 83.8% (Green sawdust), 87.7% (Kiln-dried shavings),
82.9% (Mix), and 85.2% (average of all compost material) variation in the thermal
conductivity within the experimental range of input variables.
The maximum differences between measured and estimated values of the thermal
conductivity were within ± 0.059 W m-1 K-1 (Green sawdust), ± 0.060 W m-1 K-1 (Kiln-
dried shavings), ± 0.066 W m-1 K-1 (Mix), and ± 0.065 W m-1 K-1 (average of all compost
material) in which define desirable levels of accuracy by the thermal sensor for estimating
the thermal conductivity (Figure 4.92).
Figure 4.91 revealed that the thermal conductivity of bed compost also increased with
increase in moisture content and static compaction degree. The thermal conductivity of leaf
174 
 
compost increased linearly with increase water content and compaction degree, represented
by volume fractions of air (Chandrakanthi et al., 2005). Details about statistic analysis can
be seen in Appendix F.

(a) (b)

(c) (d)
Figure 4.91 - Effect of moisture content and static compaction on thermal conductivity of
bed compost materials: (a) Green sawdust, (b) Kiln-dried shavings or
sawdust, (c) Mix, and (d) average of all compost material.

175 
 
Table 4.6 - Analysis of variance for effect of moisture content and static compaction degree on thermal conductivity of bed
compost materials.

Source of variation Sum of squares Degree of freedom Mean sum of squares Fcal-value* Probability
Green sawdust
Regression 0.1130 2 0.05630 38.863 <0.001
MC 0.0508 1 0.05080 35.030 <0.001
SC 0.0617 1 0.06170 42.551 <0.001
Residual 0.0217 15 0.00145
Total 0.1340 17 0.00790
Kiln-dried shavings or sawdust
Regression 0.1250 2 0.06270 53.561 <0.001
MC 0.0612 1 0.06120 52.307 <0.001
SC 0.0642 1 0.06420 54.871 <0.001
Residual 0.0176 15 0.00117
Total 0.1430 17 0.00841
Mix
Regression 0.1190 2 0.05950 36.375 <0.001
MC 0.0681 1 0.06810 41.524 <0.001
SC 0.0509 1 0.05090 31.036 <0.001
Residual 0.0246 15 0.00164
Total 0.1440 17 0.00845
Average of all compost material
Regression 0.1190 2 0.0593 42.971 <0.001
MC 0.3380 1 0.3380 43.333 <0.001
SC 0.0467 1 0.0467 42.608 <0.001
Residual 0.0207 15 0.0207
Total 0.1390 17 0.0082
*Highly significant.

176 
 
 (a) (b)

(c) (d)
Figure 4.92 - Estimated thermal conductivity of bed compost materials versus measured values for (a) Green sawdust, (b) Kiln-
dried shavings, (c) Mix, and (d) average of all compost material.
177 
 
 Thermal conductivity of bed compost materials as a function of moisture content (30,
45, and 60%) and dynamic compaction degree (0.1, 0.2, 0.3, and 0.4 MPa ) are shown in
Figure 4.93. Generally, the thermal conductivity increases with increasing moisture content
and dynamic compaction. Theoretically, these values can be predictable because
compaction over time has an direct effect, because it increases the bulk density of compost
by reducing porosity, which increases the thermal conductivity. Thermal conductivity
showed highest values at 60% of moisture content and 0.4 MPa of dynamic compaction
degree of Green sawdust (0.643 W m-1 K-1), Kiln-dried shavings or sawdust (0.626 W m-1
K-1), Mix (0.788 W m-1 K-1), and average of all compost material (0.244 W m-1 K-1).
Multiple regressions analysis for compost materials showed that there is also a linear
relationship between thermal conductivity (k) and moisture content (MC) and dynamic
compaction degree (DC) as follows:
a) Green sawdust:
kgreen = -0.131 + 0.269 · DC + 0.689 · MC R2 = 0.925 (4.5)

b) Kiln-dried shavings or sawdust:

kkiln = -0.127 + 0.271 · DC + 0.680 · MC R2 = 0.956 (4.6)

c) Mix:
kMix = -0.122 + 0.193 · DC + 0.623 · MC R2 = 0.934 (4.7)

d) Average of all compost material:

kAverage = -0.123 + 0.274 · DC + 0.664 · MC R2 = 0.944 (4.8)

The analysis of variance indicates a greater effect of the dynamic compact degree (high
F-value) than that of the moisture content on the thermal conductivity for all material
tested, see Table 4.7. The magnitudes of respective regression coefficients (R2) in
equations 4.5, 4.6, 4.7, and 4.8 also confirmed the results. Thermal conductivity presented
an excellent relationship with moisture content and dynamic compaction degree. To fit the
curves to the observed data, a linear equation was used (Equations 4.5, 4.6, 4.7, and 4.8).
This approach is a functional relationship selected purely by their mathematical flexibility
to fit the experimental data points, being the main advantage that it does not require the
determination of any additional compost material parameters. Authors in the literature have
also used these type of equations to relate the compost moisture content and static
compaction degree with different compost properties (Singh et al., 2007; Ahn et al., 2009).

178 
 
The models seem adequately fitted based on the observation of high F-value as well as high
coefficient of determination (R2). The models accounted for 92.5% (Green sawdust),
95.6% (Kiln-dried shavings or sawdust), 93.4% (Mix), and 94.4% (average of all compost
material) variation in the thermal conductivity within the experimental range of input
variables. The maximum differences between measured and estimated values of the
thermal conductivity were within ± 0.037 W m-1 K-1 (Green sawdust), ± 0.033 W m-1 K-1
(Kiln-dried shavings or sawdust), ± 0.045 W m-1 K-1 (Mix), and ± 0.036 W m-1 K-1
(average of all compost material), which is desirable accuracy for estimating the thermal
conductivity (Figure 4.94). The statistic analysis can be seen in Appendix F.

(a) (b)

(c) (d)
Figure 4.93 - Effect of moisture content and dynamic compaction on thermal conductivity
of bed compost materials: (a) Green sawdust, (b) Kiln-dried shavings or
sawdust, (c) Mix, and (d) average of all compost material.
179 
 
Table 4.7 - Analysis of variance for effect of moisture content and dynamic compaction degree on thermal conductivity of bed
compost materials.

Source of variation Sum of squares Degree of freedom Mean sum of squares Fcal-value* Probability
Green sawdust
Regression 0.09650 2 0.048200 55.323 <0.001
MC 0.01090 1 0.010900 12.500 0.006
DC 0.08560 1 0.085600 98.165 <0.001
Residual 0.00785 9 0.000872
Total 0.10400 11 0.009480
Kiln-dried shavings or sawdust
Regression 0.09430 2 0.047100 98.891 <0.001
MC 0.01100 1 0.011000 23.060 <0.001
DC 0.08330 1 0.083300 174.63 <0.001
Residual 0.00429 9 0.000477
Total 0.09860 11 0.008960
Mix
Regression 0.08160 2 0.04080 63.737 <0.001
MC 0.01180 1 0.01180 18.4375 0.002
DC 0.06980 1 0.06980 109.062 <0.001
Residual 0.00576 9 0.00064
Total 0.08730 11 0.00794
Average of all compost material
Regression 0.09060 2 0.0453 76.520 <0.001
MC 0.01120 1 0.01120 18.918 0.002
DC 0.07940 1 0.07940 134.121 <0.001
Residual 0.00533 9 0.000592
Total 0.09600 11 0.00872
*Highly significant.

180 
 
 (a) (b)

(c) (d)
Figure 4.94 - Estimated thermal conductivity of bed compost materials versus measured values for (a) Green sawdust, (b) Kiln-
dried shavings, and (c) Mix.

181 
 
 The analysis of variance between thermal conductivity and moisture content (30, 45,
and 60%) and static compaction (fluffy, 0.0, 0.1, 0.2, 0.3, and 0.4 MPa) degree for all bed
compost materials were obtained and are show in Figure 4.95 at the significance level of p
< 0.05. The Tukey test shows that there was a significant difference between average
values of moisture content and static compaction degree. These results indicate that the
thermal conductivity increased as moisture content decreased in the order of 60% > 45%
> 30% and thermal conductivity increased with the static compaction degree in order of 0.4
MPa > 0.3 MPa > 0.2 MPa > 0.1 MPa > 0.0 MPa = fluffy. Details about statistic analysis
see Appendix F (Table 16).

(a)

(b)
Figure 4.95 - Analysis of variance between thermal conductivity and moisture content (a)
and static compaction degree (b) for all bed compost materials. Values
followed by different letters are significantly different (P < 0.05; Tukey).
182 
 
b) Thermal resistivity:
The variations in thermal resistivity of bed compost materials with moisture content
(30, 45, and 60%) and static compaction degree (fluffy, 0.0, 0.1, 0.2, 0.3, and 0.4 MPa) are
presented in Figure 4.96. The values of thermal resistivity of Green sawdust, Kiln-dried
shavings or sawdust, Mix, and average of all compost material varied from 160.0 to 1240.5
°C cm / W, 161.7 to 1384.3 °C cm / W, 130.6 to 1253.3 °C cm / W, 269.5 to 962.6 °C cm /
W, respectably, for experimental range of the variables. As depicted in Figure 4.96, an
increasing trend in the thermal resistivity of bed compost was observed with the decrease in
both moisture content and static compaction degree.
Multiple regression analysis showed that there is a linear relationship between
dependent variable of the thermal resistivity (rho) and the independent variables of
moisture content (MC) and static compaction degree (SC) as follows:
a) Green sawdust:
rhogreen = 1229.382 - 825.693 · SC - 1099.412 · MC R2 = 0.938 (4.9)

b) Kiln-dried shavings or sawdust:

rhokiln = 1282.485 - 972.177 · SC - 1164.851 · MC R2 = 0.929 (4.10)

c) Mix:
rhoMix = 1229.610 - 987.615 · SC - 948.268 · MC R2 = 0.916 (4.11)

d) Average of all compost material:

rhoAverage = 1247.159 - 928.495 · SC - 1070.8 · MC R2 = 0.932 (4.12)

The equations 4.9, 4.10, 4.11, and 4.12 were the first order linear model, so the
response surface shown in Figure 4.96 were almost flat.
Analysis of variance (ANOVA) table was constructed to evaluate the individual effect
of independent variables on the thermal resistivity (Table 4.8). The high F-values of
regression confirmed the adequacy of the linear fitted model which accounted for 93.8%
(Green sawdust), 92.9% (Kiln-dried shavings or sawdust), and 91.6% (Mix) variations of
the thermal resistivity within the experimental range of the studied input variables.
Comparison showed that the maximum differences between measured and estimated values
of thermal resistivity were within ± 75.0 °C cm / W, ± 101.3 °C cm / W, ± 111.7 °C cm /
W, and ± 87.316 °C cm / W which are desirable accuracy for estimating the thermal
resistivity (Figure 4.97). Details about statistic analysis can be seen in Appendix F.
183 
 
 Comparing F-values of the moisture content and static compaction degree showed that
the effect of moisture content on the thermal resistivity was higher (high F-value) than the
effect of static compaction degree (low F-value), see Table 4.8.

(a) (b)

(c) (d)
Figure 4.96 - Effect of moisture content and static compaction on thermal resistivity of bed
compost materials: (a) Green sawdust, (b) Kiln-dried shavings or sawdust,
(c) Mix, and (d) average of all compost material.

184 
 
Table 4.8 - Analysis of variance for effect of moisture content and static compaction degree on thermal resistivity of bed
compost materials.

Source of variation Sum of squares Degree of freedom Mean sum of squares Fcal-value* Probability
Green sawdust
Regression 684279.297 2 342139.648 113.361 <0.001
MC 357928.582 1 357928.582 118.592 <0.001
SC 326350.715 1 326350.715 108.129 <0.001
Residual 45272.289 15 3018.153
Total 729551.586 17 42914.799
Kiln-dried shavings or sawdust
Regression 862548.929 2 431274.464 98.276 <0.001
MC 496191.886 1 496191.886 113.069 <0.001
SC 366357.043 1 366357.043 83.483 <0.001
Residual 65826.234 15 4388.416
Total 928375.163 17 54610.304
Mix
Regression 754863.347 2 377431.674 82.195 <0.001
MC 512075.963 1 512075.963 111.517 <0.001
SC 242787.384 1 242787.384 52.873 <0.001
Residual 68878.560 15 4591.904
Total 823741.907 17 48455.406
Average of all compost material
Regression 762214.467 2 381107.233 103.520 <0.001
MC 452603.798 1 452603.798 122.940 <0.001
SC 309610.669 1 309610.669 84.099 <0.001
Residual 55222.387 15 3681.4920
Total 817436.854 17 48084.521
*Highly significant.

185 
 
 (a) (b)

(c) (d)
Figure 4.97 - Estimated thermal resistivity of bed compost materials versus measured values for (a) Green sawdust, (b) Kiln-
dried shavings or sawdust, (c) Mix, and (d) average of all compost material.

186 
 
 Figure 4.98 shows the relationship between thermal resistivity as a function of moisture
content (30, 45, and 60%) and dynamic compaction (0.1, 0.2, 0.3, and 0.4 MPa) for bed
compost materials. The values of thermal resistivity of Green sawdust, Kiln-dried shavings
or sawdust, Mix, and average of all compost material varied from 155.5 to 1092.0 °C cm /
W, 159.7 to 1097.5 °C cm / W, 126.9 to 1128.6 °C cm / W, and 277.7 to 801.9 °C cm / W
respectably, depending upon the moisture content and dynamic compaction degree within
the experimental range of the variables.
An decreasing trend in the thermal resistivity of compost materials were also observed
with the increase in moisture content and dynamic compaction degree (Figure 4.98).
Multiple regression analysis showed that there is also a linear relationship between thermal
resistivity (rho) and moisture content (MC) and dynamic compaction degree (DC) as
follows:
a) Green sawdust:
rhogreen = 1187.716 - 497.389 · DC - 12.442 · MC R2 = 0.987 (4.13)

b) Kiln-dried shavings or sawdust:

rhokiln = 1211.623 - 532.158 · DC - 12.924 · MC R2 = 0.959 (4.14)

c) Mix:
rhoMix = 1205.911 - 566.171 · DC - 11.412 · MC R2 = 0.979 (4.15)

d) Average of all compost material:

rhoAverage = 1201.750 - 531.906 · DC - 12.259 · MC R2 = 0.985 (4.16)

The estimated values were compared with the measured thermal resistivity values. The
linear relationship between measured and estimated values (Fig. 4.99) shows high
correlation for all the materials. The estimated thermal resistivity from equations 4.13,
4.14, 4.15, and 4.16 depends on the moisture content and dynamic compaction degree of
the compost materials. These interdependent relationships between thermal resistivity and
the two measured parameters are presented in Figure 4.99. The maximum difference
between measured and estimated values of the thermal resistivity were within ±32.1 °C cm
/ W (Green sawdust), ±68.7 °C cm / W (Kiln-dried shavings or sawdust), ±30.7 °C cm / W
(Mix), and ±34.7 °C cm / W (average of all compost material), see Figure 4.99.

187 
 
 The analysis of variance (see Table 4.9) indicates a greater effect of the dynamic
compaction degree (high F-value) than that of the moisture content on the thermal
resistivity. The magnitudes of respective regression coefficients also confirmed the results
(Equations 4.13, 4.14, 4.15, and 4.16). The models showed adequately fitted based on the
observation of high F-value as well as high coefficient of determination (R2). Details about
statistic analysis can be seen in Appendix F.

(a) (b)

(c) (d)
Figure 4.98 - Effect of moisture content and dynamic compaction on thermal resistivity of
bed compost materials: (a) Green sawdust, (b) Kiln-dried shavings or
sawdust, (c) Mix, and (d) average of all compost material.

188 
 
Table 4.9 - Analysis of variance for effect of moisture content and dynamic compaction degree on thermal resistivity of bed
compost materials.

Source of variation Sum of squares Degree of freedom Mean sum of squares Fcal-value* Probability
Green sawdust
Regression 315739.270 2 157869.635 333.989 <0.001
MC 37109.325 1 37109.325 78.508 <0.001
DC 278629.945 1 278629.945 589.469 <0.001
Residual 4254.112 9 472.679
Total 319993.382 11 29090.307
Kiln-dried shavings or sawdust
Regression 343148.017 2 171574.009 128.428 <0.001
MC 42478.822 1 42478.822 31.796 <0.001
DC 300669.195 1 300669.195 225.060 <0.001
Residual 12023.558 9 1335.951
Total 355171.575 11 32288.325
Mix
Regression 282514.235 2 141257.117 262.347 <0.001
MC 48082.414 1 48082.414 89.299 <0.001
DC 234431.821 1 234431.821 435.393 <0.001
Residual 4845.929 9 538.437
Total 287360.164 11 26123.651
Average of all compost material
Regression 312966.440 2 156483.220 296.951 <0.001
MC 42438.574 1 42438.574 80.533 <0.001
DC 270527.865 1 270527.865 513.368 <0.001
Residual 4742.691 9 526.966
Total 317709.130 11 28882.648
*Highly significant.

189 
 
 (a) (b)

(c) (d)
Figure 4.99 - Estimated thermal resistivity of bed compost materials versus measured values for (a) Green sawdust, (b) Kiln-
dried shavings or sawdust, (c) Mix, and (d) average of all compost material.

190 
 
 The analysis of variance between thermal resistivity and moisture content (Figure
4.100a) and static compaction degree (Figure 4.100b) for all bed compost materials were
obtained at the significance level of p < 0.05. The Tukey test shows that there was a
significant difference between average values of thermal resistivity in different moisture
content and static compaction degree. This results indicate that the distribution of moisture
content decreased in the order of 30% > 45% > 60% and the distribution of static
compaction degree decreased in order of fluffy < 0 MPa < 0.1 MPa < 0.2 MPa < 0.3 MPa <
0.4 MPa. Details about statistic analysis can be seen in Appendix F (Table 17).

(a)

(b)
Figure 4.100 - Analysis of variance between thermal resistivity and moisture content (a)
and static compaction degree (b) for all bed compost materials. Values
followed by different letters are significantly different (P < 0.05; Tukey).
191 
 
 Thermal conductivity relationship to moisture content (30, 45, and 60%) and dynamic
compaction (0.1, 0.2, 0.3, and 0.4 MPa ) are shown in Figure 4.101. There was significant
difference between the thermal conductivity measurements taken on all bed compost
materials for different moisture content and dynamic compaction degree (p < 0.05). This
results indicate that the thermal conductivity increased in the order of 30% < 45% < 60%
moisture content and increased in order of 0.1 MPa < 0.2 MPa < 0.3 MPa < 0.4 MPa
dynamic compaction degree. Details about statistic analysis can be seen in Appendix F.

(a)

(b)
Figure 4.101 - Analysis of variance between thermal conductivity and moisture content (a)
and dynamic compaction degree (b) for all bed compost materials. Values
followed by different letters are significantly different (P < 0.05; Tukey).

192 
 
 The Tukey test was used to evaluate the moisture content (30, 45, and 60%) and
dynamic compaction (0.1, 0.2, 0.3, and 0.4 MPa ) on the thermal resistivity each had a
significant effect as shown by the ANOVA (Appendix F - Table 19). This results indicate
that the thermal resistivity decreased in the order of 30% > 45% > 60% moisture content
(Figure 4.102a). and the distribution of decreased in order of 0.1 MPa > 0.2 MPa > 0.3
MPa > 0.4 MPa dynamic compaction degree (Figure 4.102b).

(a)

(b)
Figure 4.102 - Analysis of variance between thermal resistivity and moisture content (a)
and dynamic compaction degree (b) for all bed compost materials. Values
followed by different letters are significantly different (P < 0.05; Tukey).

193 
 
4.3.6.2. Thermal properties of composts in different particle size
Thermal conductivity and thermal resistivity of granular materials are affected by
particle contact quality. The number of contacts per particle depend on the packing density,
particle shape and particle size distribution. Their effects on thermal conductivity and
thermal resistivity are explored in this section using selected compost materials and the
thermal needle probe technique (Decagon KD2-Pro sensor).
Figure 4.103 showed the influence of particle size in the thermal conductivity and
thermal resistivity for bed compost samples using Green sawdust, Kiln-dried shavings or
sawdust, and Mix. Thermal conductivity and thermal resistivity were plotted versus particle
size distribution, just that these two properties were directly observed data. Values of
thermal properties and thermal resistivity were not measured in samples of coarser material
because it was not possible to insert the probe of sensor into the sample.
The thermal conductivity increases with the decrease of particle size, while thermal
resistivity decreases with the increase of particle size. However, as seen in Figure 4.103,
the overall thermal conductivity and thermal resistivity varied with each particle size of bed
compost materials. In general, the thermal conductivity increased with increasing particle
size, while thermal resistivity decreased with increasing particle size what is the exact
opposite of what was expected (Yun and Santamarina, 2008; Ahn et al., 2009). This
behavior can probably be explained by considering the higher moisture retention of larger
particles once compost materials were air dried at the same time. It could possibly have
changed the values of the thermal properties. Thus, the effect of particle size on thermal
conductivity was more pronounced at higher particle size than at lower particle size.
However, the trend of thermal conductivity as a function of particle size, in some barns,
was very consistent at particle size below than 4.75 mm. It was also observed that thermal
conductivity increased drastically with large particle sizes higher than 8.00 mm.
The effect of particle size on thermal conductivity was more pronounced for the higher
particle size than at lower particle size. However, the tendency of lower thermal
conductivity as a function of particle size was very consistent at particle sizes below than
4.75 mm. It was also observed that thermal conductivity increased drastically (> 85%) with
large particle sizes higher than 8.00 mm. Thermal resistivity shows a decrease as the
compost particle size increases after the 4.75 and 5.60 mm range in Mix material (Fig.
4.103c).

194 
 
 The average values and standard deviation of thermal resistivity for all particle sizes
were 585.1 ± 117.0°C cm / W (Green sawdust), 600.8 ± 140.8°C cm / W (Kiln-dried
shavings or sawdust), and 572.4 ± 109.0°C cm / W (Mix), respectively. In a compost
material with smaller particle sizes, a fraction of the particles form a resistivity pathway,
which could result in enhanced thermal resistivity.
The analysis of variance between thermal conductivity and compost particle size
ranges for all bed compost materials (moisture content ≈ 30%) is shown in Figure 4.104.
The Tukey test with a significance level of p < 0.05 was applied. There was significant
differences between the thermal conductivity and compost particle size. In general, the
values of thermal conductivity were higher in compost materials with particle sizes larger
than 8.00 mm. Details about statistic analysis can be seen in Appendix F (Table 2).
The thermal conductivity of bed compost materials varied from 0.088 to 0.608 W m−1
K−1 depending upon the particle size. An increasing trend in the thermal conductivity of
bed compost materials was also observed with the increase particle size (Figure 4.104),
which is in general agreement with previous findings of some researchers. Investigations of
Ahn et al. (2009) showed that the thermal conductivity of 0.03 to 0.05 W m−1 K−1 to dry
sawdust and 0.03 to 0.06 W m−1 K−1 to dry wood shavings.

195 
 
 (a)

(b)

(c)
Figure 4.103 - Changes in thermal conductivity (W / m K) and thermal resistivity with
different compost particle size of compost materials: (a) Green sawdust, (b)
Kiln-dried shavings or sawdust, and (c) Mix.

196 
 
  
Figure 4.104 - Analysis of variance between thermal conductivity and compost particle
ranges (mm) for all bed compost materials. Values followed by different
letters are significantly different (P < 0.05; Tukey).

Thermal resistivity ranged from 170.2 °C cm / W to 1139.3 °C cm / W with a mean

value of 590.7 °C cm / W and associated standard deviation of 163.9 °C cm / W. However,
changes in thermal resistivity depending on the weights of each particle in a bed mixture.
The analysis of variance between thermal resistivity and particle size ranges for all bed
compost materials was obtained and show in Figure 4.105 at the significance level of p <
0.05. The Tukey test shows that there was a significant difference between the values of
thermal resistivity and compost particle size. The values of thermal resistivity for all bed
compost materials were statistically higher in lower particle size (Finer and PS2). The
average values of thermal resistivity in particle size between 4.75 and 8.00 mm (PS3 and
PS4) were statistically similar. This results are already expected due to increased particle
contact and particle shape effect with the probe of the sensor, thereby increasing the
resistance of the heat transfer. More Details about statistic analysis, see Appendix F.

197 
 
Figure 4.105 - Analysis of variance between thermal resistivity (°C cm / W) and compost
particle ranges (mm) for all bed compost materials. Values followed by
different letters are significantly different (P < 0.05; Tukey).

4.4. Develop computational fluid dynamics (CFD) model of compost barn

4.4.1. Mesh details
The most important aspect of the CFD model, as regards its meshing, was the eave and
ridge openings areas which, in order to resolve the small length scale and sharp gradient in
the airflow, required a high-resolution mesh. On the other hand, near the top of the
computational domain, cells could be relatively large. The mesh type is very important for
accurate CFD calculation. The meshes can be mainly composed of hexahedral and
tetrahedral volume mesh elements for a 3-D computational domain (Lee et al., 2007).
Because of the complicated shapes of the CBP barn, a tetrahedral mesh was built in
commercial ANSYS ICEM CFD® and a test of the different meshes was carried out using
this software. Various size of tetrahedral meshes were used, and after several levels of
previously evaluated refinement, no significant differences (p < 0.05) in the air velocity
and air temperature were encountered. Thus, the selected mesh was composed of 162,191
nodes and 683,222 elements (Figure 4.106). Such a meshing system was found to conform
easily to the exact boundary of the scale model.

198 
 
Figure 4.106 - Computational grid mesh detail of the scale model: (a) sidewall and compost
pack area, (b) external structure, (c) barn model with roof, and (d) the
volume mesh.

Although designing meshes was very time consuming, it was one of the most important
procedures to effectively increase the accuracy of CFD results. Unlike the CFD models
developed for other engineering fields, the natural ventilation study of agricultural
buildings requires a very big computational domain because a vertically and horizontally
sufficient area should be secured around the dairy barns. This larger volume needs to be
enough to for simulating natural air flow. If the distance between the dairy barn and the
boundary walls of the computational domain was not sufficiently secured, the faster airflow
stream could be compared with the natural airflow resulting in unreal airflow around the
leeward vent opening (Lee et al., 2007). In this study, the size of the CFD domain used
during the simulations was chosen in order to ensure that the position of the outer
boundaries did not compromise the CFD solution.
The computational domain was 2.40 x 2.40 x 1.20 m and the internal measured barn
model area was 1.20 x 1.33 m. The airflow data in the CBP barn is much more important
than that of the external area, but the adjacent area of the CBP barn is also very critical to

199 
 
make a reliable air velocity distribution around the structure. To effectively the total mesh
numbers, the dense meshes should be located in the CBP barn model and the mesh size had
to be designed gradually bigger as with distance from the CBP barn. The study also found
that it was very critical not to make a big and sudden change of mesh size between adjacent
meshes. It was assumed that the mesh quality including density and shape around the
external surface of the CBP barn model affected the external airflow and air pressure
distribution of the CBP barn, which greatly influenced the natural ventilation. Details about
refinement of the computational tetrahedral mesh can be seen in Figure 4.107.

Figure 4.107 - Refinement detail of the computational tetrahedral mesh.

4.4.2. Validation of reduced model and simulations

Figures 4.108 to 4.115 shows the values of the average standard errors for the variables
studied (air velocity and temperature) for all treatments tested (Closed ridge - CLR; Open
ridge - OPR; Open ridge with chimney - ORC; Elevated ridge - ELR; and Overshot ridge -
OVR) in each section and three different heights of data collection.
The simulations were performed to validate the CFD model and showed that the
average standard errors between the simulated and measured values for air velocity in all
treatments tested, using an air speed of 0.1 m s-1, wind direction south to north, and three
different heights, were 0.02 m s-1 (± 0.01), 0.02 m s-1 (± 0.01), 0.01 m s-1 (± 0.01), 0.01 m
s-1 (± 0.01), and 0.02 m s-1 (± 0.01), respectively. Thus, the average standard errors for air
200 
 
temperature were 0.22°C (± 0.17), 0.65°C (± 0.74), 0.74°C (± 0.54), 0.08°C (± 0.88), and
0.82°C (± 0.91), respectively (see Figure 4.108). However, testing the other condition (air
speed of 1.0 m s-1), the highest the average standard errors between the simulated and
measured values for air velocity and temperature values were 0.09 m s-1 (± 0.12) and
0.93°C (± 0.71), respectively, as can be seen in Figure 4.109.
The highest average standard errors between the simulated and measured values for air
velocity (0.02 m s-1 ± 0.01) and temperature (0.95°C ± 1.01), using air speed of 0.1 m s-1
and wind direction of north to south (perpendicular to ridge) were found in treatment
ORC, see Figure 4.110. Thus, using the other air speed (1.0 m s-1), the highest average
standard errors between the simulated and measured values for air velocity and temperature
were 0.13 m s-1 (± 0.10) and 1.17°C (± 1.12), respectively (see Figure 4.111).
The average standard error between simulated and measured air velocity and
temperature, using air speed of 0.1 m s-1and wind direction of east to west were 0.01 m s-1
with a range from 0.00 to 0.03 m s-1, and 1.13°C with a range from 0.05 to 3.35°C,
respectively (see Figure 4.112). To other air speed tested (1.0 m s-1), the average standard
error between simulated and measured air velocity and temperature were 0.23 m s-1 with a
range from 0.00 to 0.47 m s-1, and 0.85°C with a range from 0.05 to 3.05°C, respectively
(see Figure 4.113).
The highest the average standard errors between simulated and measured values were
0.01 m s-1 (± 0.01) for air velocity and 0.72°C (± 0.58) for temperature, using air speed of
0.1 m s-1and wind direction of west to east (Figure 4.114). However, using air speed (1.0 m
s-1), the highest average standard error between simulated and measured variables studied
were 0.18 m s-1 (± 0.06) and 0.78°C (± 0.37) for these treatments tested (see Figure 4.115).

201 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.108 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 0.1 m s-1, and wind
direction of south to north, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

202 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.109 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 1.0 m s-1, and wind
direction of south to north, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

203 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.110 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 0.1 m s-1, and wind
direction of north to south, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

204 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.111 - Average air velocity (a, b, and c) and temperature (c, d, and e) standard errors for air speed of 1.0 m s-1, and wind
direction of north to south, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

205 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.112 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 0.1 m s-1, and wind
direction of east to west, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

206 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.113 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 1.0 m s-1, and wind
direction of east to west, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

207 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.114 - Average air velocity (a, b, and c) and temperature (d, e, and f) standard errors for air speed of 0.1 m s-1, and wind
direction of west to east, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

208 
 
 (a) (b) (c)

(d) (e) (f)

Figure 4.115 - Average air velocity (a, b, and c) and temperature (c, d, and e) standard errors for air speed of 1.0 m s-1, and wind
direction of west to east, at nine locations and three different heights: (a and d) 0.02 m, (b and e) 0.11 m, and (c
and f) 0.20 m. Types of ridge: Closed ridge (CLR); Open ridge (OPR); Open ridge with chimney (ORC);
Elevated ridge (ELR); and Overshot ridge (OVR).

209 
 
 Figures 4.116 to 4.119 shows the regression plots for each variable studied and all
treatments tested. The regression equations presented by these figures can be seen in Tables
4.10 to 4.18.
In general, there is a good fit of the model with coefficients of determinations (R2) of
0.841, 0.671, 0.882, 0.718 and 0.754 (Table 4.10) for air velocity (Figure 4.116a) and
0.986, 0.881, 0.729, 0621, and 0.787 (Table 4.11) for temperature (Figure 4.116b) in CLR,
OPR, ORC, ELR and OVR types of ridges, respectively, using air speed of 0.1 m s-1 and
wind direction of south to north, while, the linear and angular coefficients are significant (t
test, P > 0.05). The coefficients of determination (R2) to air speed of 1.0 m s-1 were 0.916,
0.894, 0.909, 0.911, and 0.882 for air velocity (Figure 4.116c and Table 4.10) and 0.802,
0.723, 0.676, 0.716, and 0.819 for temperature (Figure 4.116d and Table 4.11) in CLR,
OPR, ORC, ELR and OVR types of ridges, respectively.

(a) (b)

(c) (d)

Figure 4.116 - Regression plots for air velocity and temperature in closed ridge (CLR),
open ridge (OPR), open ridge with chimney (ORC), elevated ridge (ELR),
and overshot ridge (OVR), using air speed of 0.1 m s-1 (a and b) and 1.0 m
s-1 (c and d), and wind direction south to north. Thin lines and curves
represent the linear or polynomial regression with the best values of r2
fitted for each data sets.
210 
 
Table 4.10 - Adjusted equations between the values of air velocity measured (Vmes) and
simulated (Vsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
south to north.
Ridge type Equations R2
Air speed of 0.1 m s-1
CLR Vsim = 0.8206 Vmes - 0.001 0.841
OPR Vsim = 9.2789 Vmes + 0.2068 0.671
ORC Vsim = 1.133 Vmes - 0.00006 0.882
ELR Vsim = 9.261 Vmes + 0.1776 0.718
OVR Vsim = 0.9899 Vmes - 0.0032 0.754
Air speed of 1.0 m s-1
CLR Vsim = 0.917 Vmes - 0.0094 0.916
OPR Vsim = 0.9072 Vmes + 0.0126 0.894
ORC Vsim = 0.9961 Vmes + 0.0069 0.909
ELR Vsim = 0.8163 Vmes + 0.0785 0.911
OVR Vsim = 0.824 Vmes + 0.0974 0.882

Table 4.11 - Adjusted equations between the values of air temperature measured (Tmes) and
simulated (Tsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
south to north.
Ridge type Equations R2
Air speed of 0.1 m s-1
CLR Tsim = -0.0039 Tmes2 + 1.2108 Tmes - 3.0157 0.986
OPR Tsim = 0.855 Tmes + 3.0072 0.881
ORC Tsim = 0.8344 Tmes + 3.6989 0.729
ELR Tsim = 0.7427 Tmes + 5.986 0.621
OVR Tsim = 0.1668 Tmes2 - 7.678 Tmes + 111.51 0.787
Air speed of 1.0 m s-1
CLR Tsim = 0.1002 Tmes2 - 4.2852 Tmes + 68.81 0.802
OPR Tsim = 0.1375 Tmes2 - 6.0948 Tmes + 90.602 0.723
ORC Tsim = 0.1374 Tmes2 - 6.2277 Tmes + 93.83 0.676
ELR Tsim = 0.0952 Tmes2 - 4.0725 Tmes + 66.549 0.716
OVR Tsim = 0.0897 Tmes2 - 3.8039 Tmes + 63.248 0.819

211 
 
 The variables studied (air velocity and temperature), using air speed of 0.1 m s-1 and
wind direction of north to south, gave a fair correlation of measured versus simulated
values with r2 of 0.774, 0.794, 0.915, 0.892, 0.855 (Figure 4.117a) and 0.983, 0.766, 0.729,
0.617, 0.787 (Figure 4.117b), respectively. The slopes values of simulated versus measured
air velocities regression for each treatment were 0.8598, 1.1416, 0.9035, 1.0833, and
0.9332, respectively, see Table 4.12. Good correlations were observed between measured
and simulated air velocity (R2 of 0.909, 0.949, 0.905, 0.856, and 0.875) and temperature
(R2 of 0.702, 0.800, 0.701, 0.487, and 0.666) values to air speed of 1.0 m s-1, see Figures
4.117 (c and d), and Table 4.13. In both condition, the linear and angular coefficients are
significant (t test, P > 0.05).

(a) (b)

(c) (d)

Figure 4.117 - Regression plots for air velocity and temperature in closed ridge (CLR),
open ridge (OPR), open ridge with chimney (ORC), elevated ridge (ELR),
and overshot ridge (OVR), using air speed of 0.1 m s-1 (a and b) and 1.0 m
s-1 (c and d), and wind direction north to south. Thin lines and curves
represent the linear or polynomial regression with the best values of r2
fitted for each data sets.

212 
 
Table 4.12 - Adjusted equations between the values of air velocity measured (Vmes) and
simulated (Vsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
north to south.
Ridge type Equations R2
Air speed of 0.1 m s-1
CLR Vsim = 0.8598 Vmes + 0.0013 0.774
OPR Vsim = 1.1416 Vmes + 0.00004 0.794
ORC Vsim = 0.9035 Vmes + 0.0044 0.915
ELR Vsim = 1.0833 Vmes - 0.0011 0.892
OVR Vsim = 0.9332 Vmes + 0.0089 0.855
Air speed of 1.0 m s-1
CLR Vsim = 0.9041 Vmes + 0.0286 0.909
OPR Vsim = 0.9905 Vmes - 0.0044 0.949
ORC Vsim = 1.0093 Vmes - 0.0093 0.905
ELR Vsim = 0.8159 Vmes + 0.0393 0.856
OVR Vsim = 0.9329 Vmes - 0.0065 0.875

Table 4.13 - Adjusted equations between the values of air temperature measured (Tmes) and
simulated (Tsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
north to south.
Ridge type Equations R2
Air speed of 0.1 m s-1
CLR Tsim = -0.0111 Tmes2 + 1.565 Tmes - 7.1559 0.983
OPR Tsim = 0.7181 Tmes + 6.4647 0.766
ORC Tsim = 0.8344 Tmes + 3.6989 0.729
ELR Tsim = 0.6528 Tmes + 8.1661 0.617
OVR Tsim = 0.1668 Tmes2 - 7.678 Tmes + 111.51 0.787
Air speed of 1.0 m s-1
CLR Tsim = 0.1067 Tmes2 - 4.7236 Tmes + 75.701 0.702
OPR Tsim = 0.0979 Tmes2 - 4.2643 Tmes + 69.769 0.800
ORC Tsim = 0.1097 Tmes2 - 4.7301 Tmes + 74.09 0.701
ELR Tsim = 0.1159 Tmes2 - 5.1249 Tmes + 80.221 0.487
OVR Tsim = 0.1035 Tmes2 - 4.5587 Tmes + 73.654 0.666

213 
 
 For equations of the estimated variables studied, working with air speed of 0.1 m s-1 and
wind direction of east to west, analysis linear regression and polynomial were performed
between the simulated and measured data, where R2 values were 0.433, 0.704, 0.639,
0.648, and 0.478, respectively, to air velocity (Figure 4.18a) and 0.715, 0.531, 0.815,
0.631, and 0.766, respectively, to air temperature (Figure 4.18b). More details can be seen
in Table 4.14 and 4.15. Additionally, using with air speed of 1.0 m s-1 and wind direction of
east to west, the measurement variable studied were modeled using the linear regression
and polynomial and good agreements were observed between the measured and simulated
air velocities and temperatures functions, see Figure 4.18c, Figure 4.18d, and Table 4.15.

(a) (b)

(c) (d)

Figure 4.18 - Regression plots for air velocity and temperature in closed ridge (CLR), open
ridge (OPR), open ridge with chimney (ORC), elevated ridge (ELR), and
overshot ridge (OVR), using air speed of 0.1 m s-1 (a and b) and 1.0 m s-1 (c
and d), and wind direction east to west. Thin lines and curves represent the
linear or polynomial regression with the best values of r2 fitted for each data
sets.

214 
 
Table 4.14 - Adjusted equations between the values of air velocity measured (Vmes) and
simulated (Vsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
east to west.
Ridge type Equations r2
Air speed of 0.1 m s-1
CLR Vsim = -20.274 Vmes2 + 4.5246 Vmes - 0.1479 0.433
OPR Vsim = 0.7207 Vmes + 0.0224 0.704
ORC Vsim = -19.65 Vmes2 + 4.4375x - 0.1438 0.639
ELR Vsim = 0.7064 Vmes + 0.0258 0.648
OVR Vsim = 0.566 Vmes + 0.0389 0.478
Air speed of 1.0 m s-1
CLR Vsim = 0.7827 Vmes + 0.2913 0.479
OPR Vsim = 1.0003 Vmes + 0.0487 0.595
ORC Vsim = -0.1256 Vmes2 + 1.3607 Vmes - 0.1947 0.846
ELR Vsim = -1.7216 Vmes2 + 3.7201 Vmes - 0.9359 0.612
OVR Vsim = -1.4101 Vmes2 + 3.1535 Vmes - 0.711 0.666

Table 4.15 - Adjusted equations between the values of air temperature measured (Tmes) and
simulated (Tsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
east to west.
Ridge type Equations r2
Air speed of 0.1 m s-1
CLR Tsim = 0.0861 Tmes3 - 5.8277 Tmes2 + 131.4 Tmes - 963.72 0.715
OPR Tsim = 0.6551 Tmes2 - 29.899 Tmes + 364.25 0.531
ORC Tsim = 0.8488 Tmes + 3.4548 0.815
ELR Tsim = 0.0567 Tmes3 - 3.8884 Tmes2 + 88.787 Tmes - 651.8 0.631
OVR Tsim = 0.0788 Tmes3 - 5.4173 Tmes2 + 123.82 Tmes - 918.22 0.766
Air speed of 1.0 m s-1
CLR Tsim = 0.2298 Tmes2 - 10.12 Tmes + 134 0.870
OPR Tsim = 0.2583 Tmes2 - 11.641 Tmes + 153.95 0.823
ORC Tsim = 0.3328 Tmes2 - 16.719 Tmes + 232.84 0.675
ELR Tsim = 0.0467 Tmes3 - 3.192 Tmes2 + 72.602 Tmes - 525.92 0.443
OVR Tsim = 0.5561 Tsim2 - 25.94 Tsim + 325.46 0.720

215 
 
 Good relationship was observed in the equations adjusted of variables studied measured
and simulated using air speed of 0.1 m s-1 and wind direction of west to east (Figure 4.19).
The adjusted equations of air velocity and temperature had lowest R2 of 0.404 and 0.451,
respectively. Overall, the models performed similarly with the exception of OVR, which
produced the largest values of air velocity measured. In this treatment, there was a positive
and linear relationship between the average of measured and simulated air velocity (slope =
0.5333, R2 = 0.404). Table 4.17 shows models efficiencies for air speed of 1.0 m s-1. The
results show a good correlation between measured and simulated air velocity and
temperature at five treatment tested. These are demonstrated by the correlation coefficients
(R2). In both conditions, these equations showed the linear and angular coefficients
significant (t test, P > 0.05).

(a) (b)

(c) (d)

Figure 4.19 - Regression plots for air velocity and temperature in closed ridge (CLR), open
ridge (OPR), open ridge with chimney (ORC), elevated ridge (ELR), and
overshot ridge (OVR), using air speed of 0.1 m s-1 (a and b) and 1.0 m s-1 (c
and d), and wind direction west to east. Thin lines and curves represent the
linear or polynomial regression with the best values of r2 fitted for each data
sets.

216 
 
Table 4.16 - Adjusted equations between the values of air velocity measured (Vmes) and
simulated (Vsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
west to east.
Ridge type Equations r2
Air speed of 0.1 m s-1
CLR Vsim = -56.112 Vmes2 + 10.518 Vmes - 0.3899 0.532
OPR Vsim = 0.9799 Vmes2 + 0.5333 Vmes + 0.0289 0.704
ORC Vsim = -10.669 Vmes2 + 2.8147 Vmes - 0.0728 0.709
ELR Vsim = -4.2914 Vmes2 + 1.3652 Vmes + 0.0002 0.570
OVR Vsim = 0.5333 Vmes + 0.0396 0.404
Air speed of 1.0 m s-1
CLR Vsim = 3.8045 Vmes3 - 11.707 Vmes2 + 11.943 Vmes - 2.9693 0.828
OPR Vsim = -1.6165 Vmes2 + 3.6835 Vmes - 1.0394 0.701
ORC Vsim = -1.2797 Vmes2 + 3.0085 Vmes - 0.648 0.524
ELR Vsim = -1.3077 Vmes2 + 3.1393 Vmes - 0.8013 0.724
OVR Vsim = -1.0585 Vmes2 + 2.6656 Vmes - 0.6049 0.725

Table 4.17 - Adjusted equations between the values of air temperature measured (Tmes) and
simulated (Tsim) using air speed of 0.1 m s-1 and 1.0 m s-1, and wind direction
west to east.
Ridge type Equations r2
Air speed of 0.1 m s-1
CLR Tsim = 0.1986 Tmes2 - 8.7515 Tmes + 119.33 0.787
OPR Tsim = 0.2498 Tmes2 - 11.159 Tmes + 147.52 0.544
ORC Tsim = 0.1744 Tmes2 - 7.7566 Tmes + 109.19 0.451
ELR Tsim = 0.1629 Tmes2 - 7.2151 Tmes + 102.96 0.550
OVR Tsim = -0.1701 Tmes3 + 12.439 Tmes2 - 301.37 Tmes + 2444.2 0.829
Air speed of 1.0 m s-1
CLR Tsim = 0.3824 Tmes2 - 17.386 Tmes + 220.51 0.808
OPR Tsim = 0.2695 Tmes2 - 11.971 Tmes + 155.71 0.826
ORC Tsim = 0.1932 Tmes2 - 8.6856 Tmes + 120.36 0.720
ELR Tsim = 0.2054 Tmes2 - 9.4132 Tmes + 130.75 0.676
OVR Tsim = 0.2539 Tmes2 - 11.701 Tmes + 157.69 0.729

217 
 
 In general, the experimental values of air velocity and temperature obtained in the nine
points and three high do not significantly differ from those encountered when simulating
the model with CFD. This is because the R2 values were mostly high than 0.700, indicating
that the CFD model is capable of predicting real operating conditions of the open scale
model of CBP barn with natural ventilation. Thus, the results prove to be promising, given
that the simulated system depends on several variables such as geometric characteristics of
the reduced model, roughness, temperature of the compost, orientation of the reduced
model, type of ridge, others.
The experimental and CFD simulation values of air velocity and temperature obtained
in nine locations and three different heights are presented in Figures 4.120 to 4.126.
The mean values and standard deviations of the air velocities measured for each
treatment tested using air speed of 0.1 m s-1 and wind direction of south to north were 0.07
m s-1 (±0.04), 0.04 m s-1 (±0.03), 0.06 m s-1 (±0.03), 0.04 m s-1 (±0.03), 0.05 m s-1 (±0.03),
respectively. However, the values and standard deviations of the air velocities simulated
were 0.06 m s-1 (±0.04), 0.05 m s-1 (±0.04), 0.07 m s-1 (±0.04), 0.05 m s-1 (±0.04), 0.05 m
s-1 (±0.04), respectively, see Figure 4.120. The mean values and standard deviations of the
air temperature measured for each treatment tested were 24.39°C (±1.61), 24.93°C (±2.01),
24.72°C (±1.79), 24.70°C (±1.88), and 24.94°C (±1.89), respectively. Moreover, the
highest mean values and standard deviations of the air velocities and temperature measured
and simulated using air speed of 1.0 m s-1 were 0.75 m s-1 (±0.37) and 0.73 m s-1 (±0.37),
respectively, and 24.81°C (±2.44) and 24.74°C (±2.00), respectively, see Figure 4.121.
The highest mean values measured and simulated air velocity, using air speed of 0.1 m
s-1and wind direction of north to south were 0.06 m s-1 with a range from 0.00 to 0.13 m s-1
and 0.06 m s-1 with a range from 0.00 to 0.14 m s-1, respectively. In the same study, the
highest mean values measured and simulated air velocity were 25.26°C with a range from
23.00 to 30.00°C and 25.27°C with a range from 23.05 to 29.15°C, respectively, see Figure
4.122. Thus, analyzing the other air speed tested (1.0 m s-1), the highest mean values
measured and simulated air velocity and temperature were 0.68 m s-1 (±0.43) and 0.68 m s-1
(±0.39), respectively, and 24.98°C (±2.02) and 24.70°C(±1.86) respectively (Figure 4.123).
The highest mean values measured and simulated of air velocity and temperature, using
air speed of 0.1 m s-1and wind direction of east to west, were 0.10 m s-1 (±0.01) and 0.10 m
s-1 (±0.02), respectively, for air velocity and 23.89°C (±1.57) and 23.55°C (±1.14),

218 
 
respectively, for temperature (Figure 4.124). However, using air speed of 1.0 m s-1, the
highest mean values measured and simulated values of variables studied were 0.96 m s-1 (±
0.23) and 0.99 m s-1 (±0.28), respectively, and 24.16°C (±0.65) and 23.47°C (±1.27),
respectively, for these treatments tested as shown in Figure 4.125.
The mean values and standard deviations of the air velocities simulated and observed,
using air speed of 0.1 m s-1 and wind direction of west to east, were 0.09 m s-1 (±0.02) and
0.09 m s-1 (±0.02), respectively, resulting in an average absolute deviation of 0.01 ± 0.01
and an average error of 0.02, respectively (Figure 4.126). The mean values of the air
temperature simulated and observed were 23.27°C and 23.47°C, respectively, which
resulted in an average absolute deviation of 0.74 ± 0.13 and 0.69 ± 0.10, respectively and
an average error of 1.08 and 1.02, respectively. However, the mean values and standard
deviations of the air velocities simulated and observed, using air speed of 1.0 m s-1 and the
same wind direction, were 0.94 m s-1 (±0.22) and 0.95 m s-1 (±0.22), respectively, proving
in an average absolute deviation of 0.01 ± 0.02 and an average error of 0.02, respectively,
see Figure 4.127. In the same study, the mean values and standard deviations of the air
temperature simulated and observed were 23.67°C (±1.20) and 23.44°C (±1.12),
respectively. In this case, the average absolute deviation 0.90 ± 0.27 and 0.73 ± 0.07,
respectively.
Given that the absolute average errors calculated between the values of air velocities
and temperature measured and simulated are very close to or less than 0.41 m s-1 and
2.29°C, respectively, the accuracy values of the anemometer (±0.01 m s-1) and thermometer
(±1.0°C) and that the magnitude of these errors would have little influence on the final
result for the calculation of the variables studied. Thus, the computer models can be
considered suitable for the proposed use.

219 
 
Figure 4.120 -  Differences between air velocities and temperature measured experimentally and CFD model, using air speed
simulated of 0.10 m s-1 and wind direction to north, at 9 locations and 3 different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

220 
 
Figure 4.121 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 1.0 m s-1 and direction to north, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

221 
 
Figure 4.122 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 0.1 m s-1 and direction to south, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

222 
 
Figure 4.123 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 1.0 m s-1 and direction to south, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

223 
 
Figure 4.124 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 0.1 m s-1 and direction to west, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

224 
 
Figure 4.125 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 1.0 m s-1 and direction to west, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m,
and (c) 0.20 m.

225 
 
Figure 4.126 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 0.1 m s-1 and direction to east, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m, and
(c) 0.20 m.

226 
 
Figure 4.127 -  Differences between air velocities and temperature measured experimentally and simulated with CFD, using air
velocity of 1.0 m s-1 and direction to east, at nine locations and three different heights: (a) 0.02 m, (b) 0.11 m, and
(c) 0.20 m.

227 
 
4.4.3. Airflow patterns
Airflow patterns in the reduced models were analyzed using smoke produced by
vaporized Glycerin. These studies were conducted at a wind velocity of 0.1 and 1.0 m s-1.
Similar inside airflow distributions were observed in all in simulated and measured five
ridge type evaluated when compared with CFD model evaluated with the same wind
directions. Small differences in flow patterns inside reduced model and CFD were
observed with different ventilation ridge openings. The smoke was visually observed when
it was passed through the ridge opening.

4.4.4. Computational simulation

Application of tests with adapted values and calculated variables found in the field
were performed to exemplify the application of computational models developed for a CBP
barn equipped with five different ridge types and four wind directions.

4.4.4.1. Applying computational simulation of the reduced model

The main conditions used to validate the computational model were used in this study.
However, the values (1.35 x 1.20 x 0.27 m) used in these CFD simulations correspond to
CBP barn 24, in which the value of air speed and heat flux of the floor (compost area) were
0.04 m s-1 and 0.1 W m-2 (details about these values calculated can be seen in Appendix E),
respectively. The values of air temperature used was 6.9°C (see Table 4.18).

Table 4.18 – Boundary conditions utilized in the CFD model

Case Location Boundary condition Value
Average air speed 0.04 m s-1
Inlet
Air temperature 6.9°C
1
Outlet Atmospheric pressure 0 Pa
Floor Heat flux 0.1 W m-2

The results of the simulations performed by the computational models provide visual
demonstration of the impact on air flow through structure and information of the effect of
the air velocity in cooling an area of the floor (compost area), as can be seen in Figures
4.128 to 4.172.

228 
 
 Such simulations can help engineers and dairy producers in project development and
management of the CBP barns. Applying various scenarios and allowing the simulation to
aid in the understanding the whole system.
Figures 4.128 to 4.135 shows the simulated air velocity vectors and temperature
distribution vertically and on the surface of the floor in CBP barn equipped with closed
ridge (CLR).
Figures 4.128a and 4.128b showed that the air entered the CBP barn through the ridge
opening at one side, and then exited through the sidewall opening located at the other side.
So, the air flow velocity was reduced due to presence of the wall and fences on both side of
the barn. It resulted in a internal vortex at the center of the feed alley and compost area.
In the Figure 4.129, one can observe the effect of cooling the compost surface area near
the alleyway. From the Figure 4.131, the higher surface temperature in the center of the
compost area (after down flow from wall) is due to a reduced air velocity which results in
lower cooling in this area. Figures 4.132 to 4.135 shows the simulated results of air
velocity and surface temperature on the compost area, when predominant winds enter the
installation at east to west and west to east. At the Figures 4.132 and 4.134, it can be seen
that there is a considerable variation in the direction of the air velocity vectors. Inside of
the trusses structure, it can be seen that the main flow tends to turn opposite the main wind
direction.
In Figures 4.133 and 4.135, it can be seen observed that the vortex of the airflow
promoted the worst conditions of air circulation and the surfaces temperature of the
compost area were higher on the both sides of the barn.
The air velocity distribution in a vertical plan situated in the center of the barn allows
us to determine the inside air speed more precisely. Figure 4.136 shows the air velocity
distribution in a vertical plan situated in the center of the CBP barn equipped with closed
ridge (CLR) in four wind direction. As evidenced in Figure 4.136, the higher air velocity
values was observed near to the compost area in east and west winds. However, in the near
ridge region, both these winds direction failed to perform well with air velocity values
around 0.1 m s-1. Outside air velocity values were higher in south and north wind
directions. This aspect coincides with that of other studies performed by Norton et al.
(2010b), Majdoubi et al. (2009), Lee et al. (2007), and Koenig et al. (1978).

229 
 
 (a)
Wind direction 

(b)
Figure 4.128 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with closed ridge (CLR) and wind direction of south to north.

(a) (b)
Figure 4.129 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with closed ridge (CLR) and wind direction of south to north.

230 
 
 (a)
Wind direction 

(b)
Figure 4.130 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with closed ridge (CLR) and wind direction of north to south.

(a) (b)
Figure 4.131 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with closed ridge (CLR) and wind direction of north to south.

231 
 
 (a)
Wind direction 

(b)
Figure 4.132 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with closed ridge (CLR) and wind direction of east to west.

(a) (b)
Figure 4.133 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with closed ridge (CLR) and wind direction of east to west.

232 
 
 (a)
Wind direction 

(b)
Figure 4.134 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with closed ridge (CLR) and wind direction of west to east.

(a) (b)
Figure 4.135 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with closed ridge (CLR) and wind direction of west to east.

233 
 
Figure 4.136 - Profile of mean horizontal wind speed in the centre of the CBP barn
equipped with closed ridge (CLR) in four wind direction.

Figures 4.137 to 4.145 shows the simulated results of CBP barn equipped with open
ridge (OPR) in four wind direction.
According to the Figures 4.137 and 4.139, these barns produced higher ridge vent flows
for the south and north wind directions at wind velocities above 0.07 m s-1. Wind direction
comparisons indicated that south winds consistently produced the highest ridge vents flows
(Figure 137a), however the east and west generated lowest ridge vent flows (Figures
4.141a and 4.143a). Since wind direction had a large influence on ridge vent flow, building
orientation with respect to wind direction should be considered in the design of naturally
ventilated, open ridge, assuming the reduced model exhibit similar airflow characteristics.
Figures 4.141 and 4.143 shown the reduction of air velocity inside the trusses structure.
It can be seen that the main flow tends to turn opposite the main wind direction. Highest
values of temperature on the surface of the compost area were observed with wind
direction of north (Figure 4.140). By comparison, wind direction of east and west shown
the lowest values of surface temperature on the compost area (Figures 4.142 and 4.144,
respectively).
Figure 4.145 shows the air velocity distribution in a vertical plan situated in the centre
of the CBP barn equipped with open ridge (OPR) in four wind direction. Examination of
the air velocity inside the barn above the compost area surface, demonstrates that the
values are very close at 0.2 m height from ground. However, one can remark that the air
velocity values in south and north winds directions rapid increase at ridge area.

234 
 
 (a)
Wind direction 

(b)
Figure 4.137 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge (OPR) and wind direction of south to north.

(a) (b)
Figure 4.138 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge (OPR) and wind direction of south to north.

235 
 
 (a)
Wind direction 

(b)
Figure 4.139 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge (OPR) and wind direction of north to south.

(a) (b)
Figure 4.140 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge (OPR) and wind direction of north to south.

236 
 
 (a)
Wind direction 

(b)
Figure 4.141 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge (OPR) and wind direction of east to west.

(a) (b)
Figure 4.142 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge (OPR) and wind direction of east to west.

237 
 
 (a)
Wind direction 

(b)
Figure 4.143 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge (OPR) and wind direction of west to east.

(a) (b)
Figure 4.144 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge (OPR) and wind direction of west to east.

238 
 
Figure 4.145 - Profile of mean horizontal wind speed in the centre of the CBP barn
equipped with open ridge (OPR) in four wind direction.

Figures 4.146 to 4.153 shows the simulated results of CBP barn equipped with open
ridge with chimney (ORC) in four wind direction. Figures 4.146 and 4.148 show
considerable variation in the direction of the air velocity vectors, and consequently the flow
is far from being two-dimensional. Near the ridge opening, it can be seen that center of
turbulence is located above the roof in the right (Figure 4.146a) and left (Figure 4146b)
side. From the Figure 4.148a, the north wind direction, which produced the highest ridge
vent flows, also generated high air velocities inside the structure. So, the contour plot of the
value of the vertical component of wind speed which is also shown in Figures 4.146 and
4.148 reveals periodic air circulation loops due to buoyancy forces with horizontal axes
which develop perpendicularly to the wind.
Distribution of surface temperature on the compost area are represented in Figures
4.147, 4.149, 4.151, and 4.153. It can be observed that the distribution temperature on the
compost area in wind direction north (Figure 4.149) is more uniform than east (Figure
4.151) and west (Figure 4.153). This result can be explained by the fact that the air velocity
distribution is almost uniform in near entire compost area. In contrast, when the wind
direction were east and west, the air movement tends to stabilize, generating lower air
velocity above the compost area. Figure 4.154 shows the air velocity distribution in a
vertical plan situated in the center of the CBP barn equipped with open ridge with chimney
(ORC) in four wind direction. The air velocity value above the compost area is lower in
south and north wind direction due to drag effect of the sidewall and sidewall planks. The
air velocity values were lower near ridge opening in east and west wind direction.

239 
 
 (a)
Wind direction 

(b)
Figure 4.146 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of south.

(a) (b)
Figure 4.147 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of south
to north.

240 
 
 (a)
Wind direction 

(b)
Figure 4.148 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of north
to south.

(a) (b)
Figure 4.149 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of north
to south.

241 
 
 (a)
Wind direction 

(b)
Figure 4.150 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of east to
west.

(a) (b)
Figure 4.151 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of east to
west.

242 
 
 (a)
Wind direction 

(b)
Figure 4.152 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of west
to east.

(a) (b)
Figure 4.153 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with open ridge with chimney (ORC) and wind direction of west
to east.

243 
 
Figure 4.154 - Profile of mean horizontal wind speed in the centre of the CBP barn
equipped with open ridge with chimney (ORC) in four wind direction.

Figures 4.155 to 4.162 shows the simulated results of CBP barn equipped with elevated
ridge (ELR) in four wind direction. The higher ridge vent outlet flow was visualized in
CBP barn equipped with ELR ridge type using east and west winds direction (Figure
4.155a and 4.157a). The wind flows up the windward wall surface and above the roof of
the barns with east and west winds (Figures 4.159a and 4.161a) but hits the roof inside the
barn with south and north winds (see Figures 4.155a and 4.157a). In addition, a lower
pressure area is created above the roof in the east and west winds than above the south and
north winds, thereby generating higher ridge opening flows.
Figures 4.155b, 4.157b, 4.159b, and 4.161b represents the air temperature distribution
vertically by convection and Figures 4.156, 4.158, 4.160, and 4.162 shown the temperature
distribution on the surface of the compost area. The worst uniformity in surface
temperature was observed in compost area tested with north wind direction (Figures 4.158)
in relation to other winds direction, which may be due to the fact that air flux over the
compost area is less uniform.
Figure 4.163 shows the air velocity distribution in a vertical plan situated in the center
of the CBP barn equipped with elevated ridge (ELR) in four wind direction. In general,
this shows that the average air speed in the space between the top of the compost area and
the barn roof is approximately half that of the outside wind, for all wind directions. Air
velocity is much lower inside the barn than outside and its flows through ridge opening
systematically higher in south and north wind directions.

244 
 
 (a)
Wind direction 

(b)
Figure 4.155 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with elevated ridge (ELR) and wind direction of south to north.

(a) (b)
Figure 4.156 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with elevated ridge (ELR) and wind direction of south to north.

245 
 
 (a)
Wind direction 

(b)
Figure 4.157 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with elevated ridge (ELR) and wind direction of north to south.

(a) (b)
Figure 4.158 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with elevated ridge (ELR) and wind direction of north to south.

246 
 
 (a)
Wind direction 

(b)
Figure 4.159 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with elevated ridge (ELR) and wind direction of east to west.

(a) (b)
Figure 4.160 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with elevated ridge (ELR) and wind direction of east to west.

247 
 
 (a)
Wind direction 

(b)
Figure 4.161 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with elevated ridge (ELR) and wind direction of west to east.

(a) (b)
Figure 4.162 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with elevated ridge (ELR) and wind direction of west to east.

248 
 
Figure 4.163 - Profile of mean horizontal wind speed in the centre of the CBP barn
equipped with open ridge with chimney (ORC) in four wind direction.

Figures 4.164 to 4.171 shows a representation of air velocity vectors, and temperature
distribution vertically and on the surface of the floor when overshot ridge (OVR) was
equipped in four wind direction, respectably.
The turbulence of the airflow formed above the roof in simulated barn with south winds
(Figure 4.164a) is smaller than simulated barn with north winds (Figure 4.166a). Therefore,
as can be seen in Figures 4.168a and 4.170a, the turbulence zone was situated after barn at
windward gable wall height in simulated barn with east and west winds. The south and
north winds direction created the lowest air velocity values differences (Figures 4.164 and
4.166) and east and west winds produced the highest air velocity values differences
(Figures 4.168 and 4.170). Airflow enters through both sides the opening wall and owing to
density differences between the inside and outside air, immediately drops to the floor. This
airflow then travels toward the center of the building, heating up as it passes over the
compost area before it eventually rises (see Figures 4.164 and 4.166). A similar observation
has already been reported by Norton et al. (2010a).
The highest surface temperature distribution differences on compost area was observed
in the east and west winds direction (Figures 4.169 and 4.171). The lowest surface
temperature differences on compost area was generally noted south winds (Figure 4.165).
Figure 4.172 shows the air velocity distribution in a vertical plan situated in the center
of the CBP barn equipped with overshot ridge (OVR) in four wind direction. The air
velocity show approximately the same values at 0.02 m above the compost area for all
winds direction tested, but values strongly decreases near the ridge opening in the east and
west winds direction, then increases progressively from the ridge opening.
249 
 
 (a)
Wind direction 

(b)
Figure 4.164 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with overshot ridge (OVR) and wind direction of south to north.

(a) (b)
Figure 4.165 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with overshot ridge (OVR) and wind direction of south to north.

250 
 
 (a)
Wind direction 

(b)
Figure 4.166 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with overshot ridge (OVR) and wind direction of north to south.

(a) (b)
Figure 4.167 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with overshot ridge (OVR) and wind direction of north to south.

251 
 
 (a)
Wind direction 

(b)
Figure 4.168 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with overshot ridge (OVR) and wind direction of east to west.

(a) (b)
Figure 4.169 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with overshot ridge (OVR) and wind direction of east to west.

252 
 
 (a)
Wind direction 

(b)
Figure 4.170 - Air velocity (a) and temperature (b) distribution vertically in the CBP barn
equipped with overshot ridge (OVR) and wind direction of west to east.

(a) (b)
Figure 4.171 - Air temperature distribution on the surface of the floor inside the CBP barn
equipped with overshot ridge (OVR) and wind direction of west to east.

253 
 
Figure 4.172 - Profile of mean horizontal wind speed in the centre of the CBP barn
equipped with overshot ridge (OVR) in four wind direction.

Overall, one can summarise the complex airflow as follows:

a) In general, wind direction indicates that air would approximately move along the
length of the barns as opposed to the width or across the barn. The ridge openings
noticeably increased the inside air velocities for all the wind directions studied except for
the east and west direction. However, in this case, low air velocities were observed inside
the trusses structure and near the ridge opening;
b) The airflow patterns were visually higher in all ridge openings tested with wind
direction of south and north. With east and west winds higher airflows were observed in the
windward ends of the ridge vents than was noted downwind. The airflow patterns for east
and west winds also showed that inside the trusses structure the air generally flowed in the
opposite direction of the outside winds. With a south and north wind the inside airflow near
the trusses structures were generally in the same direction as the outside air. South and
north winds produced a swirling action inside the all CFD models with most of the air
exiting from the opening wall. This condition was probably due to a slight difference in the
outside airflow;
c) CBP barn model geometry effects the surface temperature difference on compost
area were visibly significant through the air temperature distribution on the surface of the
floor inside the CBP barns in the previous figures for all ridge opening tested and direction
of the wind; and
d) Air loops fed by buoyancy forces due to the inside air temperature differences
develop perpendicularly between the top of the compost area and bottom of the roof vent
openings.
254 
 
4.4.4.2. Evaluation of the types of ridge vents and wind direction
Maintaining active composting in a compost bedded pack barn is a real challenge
during the winter months. It becomes difficult to start or restart a compost bedded pack
barn in winter with low bacterial activity since heat losses can easily exceed heat
generated. Because of this, to select the best condition of ventilation in the winter weather
with reduced cooling of the surface of the bed, the CFD models were evaluated in function
of five types of ridge vents (CLR, OPR, ORC, ELR, and OVR) and four wind direction
(North, South, East, and West) to promote greater percentage heated surfaces of the floor.
The analyzes of variance for variables five types of ridge vents and four wind direction
in function of percentage of heated floor area are showing in Table 4.19. There is not a
statistically significant difference (p> 0.05) between types of ridge vents (P = 0.716) and
wind direction (P = 0.874).

Table 4.19 - Analysis of variance to heated floor area in different types of ridge vents and
wind direction for all CFD models tested.
SV DF SS MS Fcal-Value* Probability
Ridge 4 1620.038 405.010 0.531 0.716
Direction 3 526.051 175.350 0.230 0.874
Residual 12 9155.593 762.966
Total 19 11301.682
Source of Variation (SV), degree of freedom (DF), sum of squares (SS), mean sum of squares (MS), and probability (P).

The mean values of percentage of heated floor area in five types of ridge vents and
four wind direction are showing in Table 4.20. Table 4.20 shows that the highest
percentage of heated floor of the surface area inside the CFD models were observed in
treatments ORC and W to E (Tukey test; p < 0.05). Type of ridge vent and wind direction
that promoted greater cooling surface of the floor were the treatments ELR and S to N
(Tukey test; p < 0.05). Details about statistical analysis can be seen in Appendix F (Table
20).

255 
 
Table 4.20 - Mean values of percentage of heated floor area for different types of ridge
vents and wind direction. Values followed by different letters are
significantly different (P < 0.05; Tukey).
Treatments Means values (%)
Type ridge design
ORC 45.292 a
OPR 27.512 b
CLR 24.462 c
OVR 23.497 d
ELR 19.457 e
Wind direction
W to E 32.366 a
E to W 32.176 b
N to S 27.922 c
S to N 19.712 d

Once CFD modeling returns results based on temperature variations of the total area in
function of the composting area size into barns, highest heating and wind direction
percentage can probably be a indicative where the composting process will be occurring
with better efficiency or, in other words, what kind of ridge opening will be better to
maintain ideal microbiological conditions in terms of aeration and temperature with range
of 45 a 60°C, as described early. These results showed that the best ridge vent and wind
direction in the winter weather was observed in the open ridge with chimney (ORC) and
West to East, respectively. Therefore, open ridge with chimney (ORC) positioned from
West to East can possibly be the best barn design with better efficiency among those barns
who were studied.

256 
 
 5 - CONCLUSION

The following conclusions were drawn based on the experimental study of 42 compost
bedded pack (CBP) barns in Kentucky:

a) Overall, dairies producers visited in Kentucky were very satisfied with their compost
barns. They observed that cows were more comfortable and that this housing system
resulted in increased cow cleanliness. The compost bedding pack (CBP) barns in
Kentucky of this study had a variety of building designs. Majority of these barns had
feed alleys and driveways. Most frequent barn orientation was NE to SW. The primary
types of barn ridge found in the barns were overshot ridges. Most of barns had roof pitch
of 4:12. The majority of barns was equipped with box fans. Air temperatures tended to
be lower indoor the barns which utilized overshot or capped ridge, forced ventilation
system and ridge direction of SW or NW. The average relative humidity in this study
was approximately 65.0%. In general, wind direction was approximately transverse to
the barns length direction and the air velocity was 1.0 m s-1. In addition, air quality and
possible dust issues in compost dairy barns need to be evaluated.

b) The bedding materials more used in the CBP barns was kild-dried sawdust. However,
most dairies producers preferred to use green sawdust rather than the kild-dried sawdust,
because this material was harmful or beneficial to the composting bed. Moreover, green
sawdust was not absorb as much water as kiln-dried sawdust. Pack temperatures
indicated that some of the bedding materials were not able to support microbial activity
and produce heat. Average Pack moisture content was 59.0% (± 9.0) wet basis. Our
observations indicate moisture content should be between 40-60%. Most of the barns in
Kentucky had their compost aerated to a depth below 0.25 m, twice a day in the summer
and winter time, while cows were away at the parlor, most often using a chisel plow on a
small tractor. Bedding management is critical to encourage microbial activity, minimize
pathogen exposure, and maintain cow cleanliness. Further research is necessary
regarding the effect of different climatic conditions on the composting processes.

c) A comparison of the particle weight fraction distributions content reveals that these
bedding compost materials found in CBP barns are very different. The results indicated
that the distribution of mass weight increased in the order of Coarser to Finer. The
257 
 
 average finer index expressed as weight percent in the samples studied was 30.1%,
implying excessive water retention and low aeration.

d) Generally, some compost bedding materials showed high particle density (≥ 1g cm-3)
compared with other materials. Indeed, these values were higher than expected, but can
likely be tied to the presence of carbonates and other inorganic media such as soil
particles contained within the compost samples. The higher bulk density value was 3.6
times that of the lower bulk density value. This investigation also showed that porosity
is potentially dependent (reverse order) on bulk density although the statistical
significance of this relation was high.

e) Water Holding Capacity (WHC) increases with increasingly fine particle size. Coarse
particle size of compost have a lower WHC since they are high in large pores subject to
free drainage. Fine particle size have a greater occurrence of small pores that hold water
against free drainage, resulting in a higher WHC values. In this study, WHC increased
with increasing moisture content until certain level. Therefore, compost bedding
material with good physical and chemical properties helps to increase the WHC of
bedding compost, reduces natural drainage and runoff of the moisture (urine and faeces)
through surface of compost, and consequently improves their moisture content. Further
research is needed to investigate the influence of moisture content measurement on the
determination of WHC.

f) Pack chemical characteristics were almost similar to all bedding materials. Thus, many
dairy producers can use the bedding compost to fertilizer their crop fields and avoid
over-applying nutrients and reduce water pollution. Bedding compost always contained
bacterial, but the variation was large between different sampling sites and times.
However, Coliforms was not present in barns that had a higher compost temperature.
Coliforms, E. Coli, Bacillus and Streptococcus count were higher in the barns that
showed lower moisture content.

g) Generally, the thermal conductivity increased with increasing particle size, while
thermal resistivity decreased with increasing particle size. This behavior can probably
be explained by considering the higher moisture retention of larger particles once
compost materials were air dried at the same time. It could possibly have changed the
258 
 
 values of the thermal properties. Thus, the tendency thermal conductivity as a function
of particle size was very consistent at particle size below than 4.75 mm and thermal
resistivity showed an pronounced decrease in compost particle size between 4.75 and
5.60 mm. Thus, results above indicate that the thermal properties are strongly dependent
on particle size. However, further studies should be carried out to confirm their positive
effects.

h) In the present study, an increasing trend in the thermal conductivity of all compost
materials was also observed with the increase in both moisture content and static and
dynamic compaction degree for all compost material tested. In contrast, an increasing
trend in the thermal resistivity of bed compost was observed with the decrease in both
moisture content and static and dynamic compaction degree. The first order regression
models developed in this study represent the relationship of degree of moisture content
and static and dynamic compaction versus thermal conductivity and thermal resistivity
as well. The resulting thermal properties of compost bedding materials can be used to
develop heat transfer models for the design of more optimal temperature control in
compost bedding pack systems.

i) A 3-D CFD model was established to graphically show the air velocity and temperature
distribution in a CBP barns using different ridge vents opening with natural ventilation.
The developed model provided good agreement with experimental measurements and it
was able to identify the impact on air flow through structure in a CBP barns. To date
most published works on CFD simulation for dairy building have been 2-D, and only
ventilation has been modeled with 3-D CFD. We hope this developed model can
contribute to the optimum design for CBP barn construction. It should also provide
guidelines for modifying existing CBP barns towards more uniform and optimum
temperature and air movement distribution, thus increasing air quality and cow comfort.
Research employing this model or other models with similar characteristics is needed to
simulate more accurately the effect on air flow through structure in CBP barns with
dairy cows. These results showed that the best ridge vent and wind direction in the
winter weather was ORC positioned from West to East can possibly be the best barn
design with better efficiency among those barns who were studied.

259 
 
 6 - REFERENCES

ABU-HAMDEH, N. H. Thermal Properties of Soils as affected by Density and Water

Content. Biosystems Engineering, v. 86, n. 1, p. 97–102, 2003.

ABU-HAMDEH, N. H. Soil and water: measurement of the thermal conductivity of sandy

loam and clay loam soils using single and dual probes. Journal of Agricultural
Engineering Research, v. 80, n. 2, p. 209-216, 2001.

AGNEW, J. M.; LEONARD, J. J.; FEDDES, J.; FENG, Y. A modified air pycnometer for
compost air volume and density determination. Canadian Biosystems Engineering, v. 25,
p. 27-35, 2003.

AGNEW, J. M.; LEONARD, J. J. The physical properties of compost. Compost Science

and Utilization, v. 11, n. 3, p. 238-264, 2003.

AHN, H.; SAUER, T. J.; RICHARD, T. L.; GLANVILLE, T. D. Determination of thermal

properties of composting bulking materials. Bioresource Technology, v. 100, p. 3974–
3981, 2009.

AHN, H. K.; RICHARD, T. L.; GLANVILLE, T. D. Laboratory determination of compost

physical parameters for modeling of airflow characteristics. Waste Management, v. 28, p.
660–670, 2008.

AHN, H. K.; SAUER, T. L.; RICHARD, T. L.; GLANVILLE, T. D. Determination of

thermal properties of compost bulking materials by using various methods. In ASABE
Annual International Meeting, Portland, Oregon. 2006.

ALBRIGHT, L. D. Environmental Control for Animals and Plants. 4th edition, The
American Society of Agricultural Engineers, St. Joseph, MI. 1990.

ANSYS CFX 13.0, “User manual”, ANSYS Europe Ltd., 2011.

ARMSTRONG, D.V.; HILLMAN, P.E.; MEYER, M.J.; SMITH, J.F.; STOKES, S.R.;
HARNER, J.P. Heat stress management in freestall barns in the western U.S. In:
Proceedings…Western Dairy Management Conference. p. 87-95, 1999.

AYNSLEY, R. M.; MELBOURNE, W. H.; VICKERY, B. J. Architectural

aerodynamics. Applied Science Publishers Ltd., London, Ont. 250p., 1977.

BAHRAMI, M. 2012. Dimensional Analysis and Similarity Advantages of dimensional

analysis. Simon Fraser University. Available at: http://www.sfu.ca. Accessed 26 May 2012.

BARBERG, A. E.; ENDRES, M. I.; SALFER; J. A.; RENEAU, J. K. Performance and

welfare of dairy cows in an alternative housing system in Minnesota. Journal of Dairy
Science, v. 90, p.1575-1583, 2007.

260 
 
BARLOW, J. B.; RAE W. H.; POPE, A. Low-Speed Wind Tunnel Testing, 3rd edition;
John Wiley & Sons, 1999.

BASSANEZZI, R. C. Ensino–aprendizagem com modelagem matemática. São Paulo:

Contexto, 2002.

BEFFA, T. 2002. The composting biotechnology: a microbial aerobic solid substrate

fermentation complex process. The composting process and management, Compag
Technologies International. Available at: http://www.compag.ch. Accessed 20 May 2012.

BERNAL, M. P.; ALBURQUERQUE, J. A.; MORAL, R. Composting of animal manures

and chemical criteria for compost maturity assessment. A review. Bioresource
Technology, v. 100, p. 5444–5453, 2009.

BERNAL, M. P.; LOPEZ-REAL, J. M.; SCOTT, K. M. Application of natural zeolites for

the reduction of ammonia emissions during the composting of organic wastes in a
composting simulator. Bioresource Technology, v. 43, p. 35–39, 1993.

BEWLEY, J. M.; TARABA, J. L. Compost Bedded Pack Barns in Kentucky. Extension

Fact Sheet ID-178, 2009.

BJERG, B.; SVIDT, K.; ZHANG, G.; MORSING, S.; JOHNSEN, J. O. Modeling of air
inlets in CFD prediction of airflow in ventilated animal houses. Computers and
Electronics in Agriculture, v. 34, n. 1-3, p. 223-235, 2002.
BJERG, B.; SVIDT, K.; ZHANG, G.; MORSING, S. The effect of pen partitions and
thermal pig simulators on airflow in a livestock test room. Journal of Agricultural
Engineering Research, v. 77 n. 2, p. 317-326, 2000.

BLACK, R. A. Unpublished data. 2012. 100p. M.S. Thesis; Animal & Food Sciences,
University of Kentucky.

BLACK, R. A.; TARABA, J. L.; DAY, G. B.; DAMASCENO, F. A.; NEWMAN, M. C.;
AKERS, K. A.; BEWLEY, J. M. Relationships among temperature, moisture, bacterial
counts, and animal hygiene in compost bedded pack barns. In: American Dairy Science
Association (ADSA), New Orleans, Louisiana, v. 89. p. 239-239, 2011a.

BLACK. R. A. ; TARABA, J. L. ; DAY, G. B. ; DAMASCENO, F. A. ; BEWLEY, J. M. .

An overview of compost bedded pack barn management in Kentucky. In: American Dairy
Science Association (ADSA), New Orleans, Louisiana, v. 89. p. 239-239, 2011b.

BLACKBURN, T. H. Benthic mineralization and bacterial production. In: Blackburn T. H.,

Sorensen J (ed.). Nitrogen cycling in coastal marine environments. Wiley, New York.
1988.

BOULARD, T.; WANG, S. Experimental and numerical studies on the heterogeneity of

crop transpiration in a plastic tunnel. Computers and Electronics in Agriculture, v. 34, n.
1–3, p. 173–190, 2002a.

261 
 
BOULARD, T.; KITTAS, C.; ROY, J.C.; WANG, S. Convective and ventilation transfers
in greenhouses, part 2: determination of the distributed greenhouse climate. Biosystems
Engineering, v. 83, p. 129–147, 2002b.
BRADSHAW, P.; PANKHURST, R. C. The design of low-speed wind tunnels. Progress
in aeronautical sciences, v. 6, p. 1-69, 1964.

BRADY, N. C.; WEIL, R. R. The Nature and Properties of Soils. 13th ed. Prentice Hall.
Upper Saddle River, NJ. 2002.

BRANDT, A. Multilevel adaptive solutions to boundary value problems. Mathematics of

Computation, v. 31, p. 333-390, 1977.

BRIAUD, J. L.; CHAOUCH, A. Hydrate melting in soil around hot conductor. Journal of
Geotechnical and Geoenvironmental Engineering, v. 123, n. 7, p.645–653, 1997.

BRISTOW, K. L. Measurement of thermal properties and water content of unsaturated

sandy soil using dual-probe heat-pulse probes. Agricultural and Forest Meteorology, v.
89, p. 75–84, 1998.

BRISTOW, K. L.; WHITE, R. D.; KLUITENBERG, G. J. Comparison of single and dual

probes for measuring soil thermal properties with transient heating. Australian Journal of
Soil Research, v. 32, p. 447-464, 1994.

BROUK, M. J.; SMITH, J. F.; HARNER, J. P. Fan placement and heat stress abatement in
four-row freestall barns. Kansas Dairy Day Report of Progress, v. 110, p.1-8, 2001.

BRUIJN, P. J; VAN HANEGHEM, I. A.; SCHENK, J. An improved nonsteady-state probe

method for measurements in granular materials. Part 1: Theory. High Temperatures -
High Pressures, v. 15, p. 359-366, 1983.

CHANDRAKANTHI, M.; MEHROTRA, A. K.; HETTIARATCHI, J. P. A. Thermal

conductivity of leaf compost used in biofilters: An experimental and theoretical
investigation. Environmental Pollution, v. 136, p. 167-174, 2005.

CHANGIRATH, S.; HALBACH, T. R.; DORFF, R. 2011. Media and Media Mix
Evaluation for Dairy Barn Compost Bedding Systems. Department of Soil, Water and
Climate. Available at: http://www.soils.um.edu/. Accessed 16 May 2012.

CHASTAIN, J.P. Designing and managing natural ventilation systems. In: Proceedings…
Dairy Housing and Equipment Systems: Managing and planning for profitability. NRAES
publication, v. 129, p. 147-163, 2000.

CHASTAIN, J. P.; HIATT, R. S. Supplemental lighting for dairy milk production.

National Food and Energy, Council, Columbia, MO 65203, p. 20, 1998.

262 
 
CHASTAIN, J. P. Pressure gradients and the location of the neutral pressure axis for
low rise structures under pure stack conditions. 1987. M.S. Thesis, Department of
Agricultural Engineering, University of Kentucky, Lexington KY.

CHOI, K.; ALBRIGHT, L. D.; TIMMONS, M. B. An application of the k-ε turbulence

model to predict air distribution in a slot ventilated enclosure. Transactions of the ASAE,
v.31, p.1804–1814, 1988.

CLARKE, S.; HOUSE, H. Using Less Energy on Dairy Farms. Dairy and Beef Housing
and Equipment Engineer, OMAFRA, Clinton. Paper No. 10-067, 2010.

CORNELL WASTE MANAGEMENT INSTITUTE (CWMI, 2006). Using Manure Solids

as Bedding. Department of Crop and Soil Sciences. Available at:
http://cwmi.css.cornell.edu/bedding.htm. Accessed 15 May 2012.

CORNELL UNIVERSITY COMPOSTING (CUC, 2003). The Science and Engineering of

Composting. Available at: http://www.cfe.cornell.edu/compost/science.html. Accessed 16
May 2012.

COYNE, M. S. Biological denitrification. In: Schepers, J.S.; Raun, W. (Eds.). Nitrogen in

agricultural systems. Agronomy Monograph 49, Madison, WI. p. 197-249, 2008.

DAHL, G. E.; ELSASSER, T. H.; CAPUCO, A. V.; ERDMAN, R. A.; PETERS, R. R.

Effects of long day photoperiod on milk yield and circulating insulin-like growth factor-l.
Journal of Dairy Science, v. 80, p. 2784-2789, 1997.

DAMASCENO, F. A.; YANAGI JUNIOR, T.; LIMA, R. R. ; GOMES, R. C. C. ;

MORAES, S. R. P. Avaliação do bem-estar de frangos de corte em dois galpões comerciais
climatizados. Ciência e Agrotecnologia, v. 34, p. 1031-1038, 2010a.

DAMASCENO, F. A.; AMARAL, A. G.; MARTINS, M. A.; SARAZ, J. A. O.; BAÊTA,

F. C. Dinâmica do fluido computacional para simulação da temperatura e velocidade do ar
em sistema de aquecimento avícola. In: Congresso Brasileiro de Engenharia Agrícola, 38.
2010b, Vitória. Anais... Vitória: Sociedade Brasileira de Engenharia Agrícola.

DAVIDSON, E. A. Fluxes of nitrous oxide and nitric oxide from terrestrial ecosystems. In
ROGERS, J.E.; WHITMAN, W.B. (Eds.) Microbial production and consumption of
Greenhouse gases: Methane, nitrogen oxides and halomethanes. American Society for
Microbiology, Washington, DC, p. 219-236, 1991.

DIAZ, L. F.; SAVAGE, G. M.; EGGERTH, L. L.; GOLUEKE, C. G. Composting and

Recycling. Municipal Solid Waste. Lewis Publishers, USA. v. 33, p. 127-180, 1993.

DIPPEL, S.; DOLEZAL, M.; BRENNINKMEYER, C.; BRINKMANN, J.; MARCH, S.;
KNIERIM, U.; WINCKLER, C. Risk factors for lameness in freestall-housed dairy cows
across two breeds, farming systems, and countries. Journal of Dairy Science, v. 92, p.
5476-5486, 2009.

263 
 
EGAN, R. K.; HELLICKSON, M. A. Ridge vent and wind direction effects on ventilation
characteristics of a model open-front livestock building. Transactions of the ASAE, No.
75-4523. ASAE, St. Joseph, Mich. 1977.

ENDRES, M. I.; JANNI, K. A. 2009. Compost Bedded Pack Barns for Dairy Cows. The
Cooperative Extension Service. Available at: http://www.bioforsk.no. Accessed 08 May
2012.

ENDRES, M. I. Compost bedded pack barns – Can they work for you?. In: Proceedings…
Western Canadian Dairy Seminar, Alberta, Canada, p. 271, 2009.

EPSTEIN, E. The science of composting. Lancaster, Pennsylvania: Technomic Publishing

Company, p. 487, 1997.

FERREIRA, D. F. SISVAR - Sistema de análise estatística para dados balanceados.

Lavras: UFLA/DEX, 2003.

FERRY, M. New features of the MIGAL solver. In: Proceedings... 9th PHOENICS User
Conference, Moscow Power Engineering Institute, Moscow, Russia, p. 1–23, 2002.

FITZGERALD, M. Build a Wind Tunnel. Tech Directions, p. 6, 2005.

FOX, R. W.; McDONALD, A. T. Introdução à mecânica dos fluidos, 4º ed., Rio de

Janeiro: LTC, 662p, 1998.

GAJALAKSHMI, S.; ABBASI, S. A. Solid Waste Management by composting: state of

the art. Critical Reviews in Environmental Science and Technology, v. 38, p. 311-400,
2008.

GAY, S. W. Bedded-pack Dairy Barns. Virginia Cooperative Extension, p. 442-124,

2009.

GEBREMEDHIN, K. G.; WU, B. Simulation of flow field of a ventilated and occupied

animal space with different inlet and outlet conditions. Journal of Thermal Biology, v. 30
n. 5, p. 343-353, 2005.

GEBREMEDHIN, K. G.; WU, B. X. Characterization of flow field in a ventilated space

and simulation of heat exchange between cows and their environment. Journal of
Thermal Biology, v. 28, n. 1, p. 301–319, 2003.

GINKEL, J. T. V.; RAATS, P. A. C.; VAN HANEGHEM, I. A. Bulk density and porosity
distributions in a compost pile. Netherlands Journal of Agricultural Science, v. 47, p.
105-121, 1999.

GORECKI, J. P. Túneis aerodinâmicos: passado, presente e futuro. In: II Encontro

Nacional de Ciências Térmicas, Águas de Lindóia, SP, 53p., 1988.

264 
 
GOYAL, S.; DHULL, S. K.; KAPOOR, K. K. Chemical and biological changes during
composting of different organic wastes and assessment of compost maurity. Bioresource
Technology, v. 96, n. 14, p. 1584-1581, 2005.

GRAVES, R. E. 1999. Dairy Manure Handling. Agricultural and Biological Engineering

Extension. Penn State University. Available at http://www.abe.psu.edu. Accessed 14 May
2012.

GRAVES, R. E.; BRUGGER, M. F. Naturally ventilated freestall barns. In: Proceedings…

Expansion Strategies for Dairy Farms: Facilities and Financial Planning , NRAES-77:
Ithaca, NY, p. 409-417, 1994.

HANDRECK, K. A. Particle size and the physical properties of growing media for
containers. Communications Soil Science Plant Analysis, v. 14, n. 3, p. 209–222, 1983.

HARRAL, B. B.; BOON, C. R. Comparison of predicted and measured air flow patterns in
a mechanically ventilated livestock building without animals. Journal of Agricultural
Engineering Research, v.66, p.221–228, 1997.

HASHEMI, M.; HERBERT, S.; CHICKERING-SEARS, C.; WEIS, S.; GRADIL, C.;
PURDY, S.; HUYLER, M.; PROSTAK, R. Bedding Options for Dairy Cows. 2001.
UMass Extension Crops, Dairy, Livestock, Equine . Available at: www.umass.edu/cdl.
Accessed 08 May 2012.

HAUG, R. T. Compost Engineering: Principles and Practice. Technomic Publishing

Company, Inc., Lancaster, PA, 655p., 1980.

HENRY, C.; SULLIVAN, D. The Nitrogen Cycle in Managing Nitrogen from

Biosolids. 1999.

HIRSCH, C. Numerical Computation of Internal and External Flows. Butterworth-

Heinemann. 2007.

HOFF, S. J.; JANNNI, K. A.; JACOBSON, L. D. Three-dimensional buoyant turbulent

flows in a scaled model, slot-ventilated, livestock confinement facility. Transactions of
the ASAE, v.35, p.671–686, 1992.

HUSAINIE, S. N.; QAMAR, A. 2012. Wind tunnel & airfoil drag analysis. Available at:
http://whoknowshowto.com. Accessed 27 May 2012.

HUTCHINSON, B. R.; RAITHBY, G. D. A Multigrid method Based on the Additive

Correction Strategy. Numerical Heat Transfer, v. 9, p. 511-537, 1986.

INCROPERA, F. P.; DEWITT, D. P. Fundaments of heat transfer and mass. 6th

Edition, John Wiley and Sons, New York, 2001.

265 
 
IQBAL, M. K.; SHAFIQ, T.; AHMED, K. Characterization of bulking agents and its
effects on physical properties of compost. Bioresource Technology, v. 101, p. 1913–1919,
2010.

IWABUCHI, K.; KIMURA, T.; OTTEN, L. Effect of volumetric water content on thermal
properties of dairy cattle feces mixed with sawdust. Bioresource Technology, v. 70, p.
293–297, 1999.

JANNI, K. A.; ENDRES, M. I.; RENEAU, J. K.; SCHOPER, W. Compost dairy barn
layout and management recommendations. Applied Engineering in Agriculture, v. 23, p.
97-102, 2007.

JANNI, K.; RENEAU, J.; SCHOPER, W. Composting Bedded Pack Barns for Dairy
Housing. University of Minnesota, Dairy Extension, Regional Extension Educator-Dairy.
2005.

JENTZSCH, R. Estudo de modelos reduzidos destinados à predição de parâmetros

térmicos ambientais em instalações agrícolas. 2002. 103p. PhD. Dissertation.
Department of Agricultural Engineer, Federal University of Viçosa, Viçosa.

JOHNSON, A. E.; HANCOCK, P. E. Some aspects of centrifugal fan characteristics in

blower wind tunnels. Aero Journal, Royal Aeronautical Society, v. 101, p. 481, 1997.
JOYE, S. B; HOLLIBAUGH, J. T. Influence of sulfide inhibition of nitrification on
nitrogen regeneration in sediments. Science, v. 270, p. 623–625, 1995.

KARTHIKEYAN, R.; BHANDARI, A. Anaerobic biotransformation of aromatic and

polycyclic aromatic hydrocarbons in soil microcosms: A REVIEW. Journal of Hazardous
Substance Research, v. 3, n. 1, p. 1-19, 2001.

KATZ, A.; SANKARAN, V. Mesh quality effects on the accuracy of CFD solutions on
unstructured meshes. Journal of Computational Physics, v. 230, n. 20, p. 7670-7686
2011.

KIM, K.; YOON, J. Y.; KWON, H. J.; HAN, J. H.; EEK SON, J.; NAM, S. W.;
GIACOMELLI, G. A.; LEE, E. I. B. 3-D CFD analysis of relative humidity distribution in
greenhouse with a fog cooling system and refrigerative dehumidifiers. Biosystems
Engineering, v. 100, n. 2, p. 245-255, 2008.

KLAAS, I. C.; BJERG, B.; FRIEDMANN, S.; BAR, D. Cultivated barns for dairy cows.
Dansk Veterinærtidsskrift, v. 9, p. 20-29, 2010.
KLIN, S. J. Similitude and approximation theory. New York: McGraw-Hill, 1966.

KOENIG, T. J.; HELLICKSON, M. A.; ROEPKE, W. L. Building geometry and wind

effects on model open-front beef building ventilation. Transactions of the ASAE, Paper
No. 77-4041, 1978.

266 
 
KÖLTZSCH, P.; WALDEN, F. Ähnlichkeitstheorie und modelltechnik. Heft 1. Freiberg:
TU Bergakademie Freiberg, 1990.

KRISTENSEN, E. Organic matter diagenesis at the oxic/anoxic interface in coastal marine

sediments, with emphasis on the role of burrowing animals. Hydrobiologia, v. 426, p. 1-
24, 2000.

LEE, I. B.; SADANOR, S.; SUNG, S. H. Evaluation of CFD accuracy for the ventilation
study of a naturally ventilated broiler house. Japan Agricultural Research Quarterly
(JARQ), v. 41, n. 1, p. 53-64, 2007.

LEE, I. B.; SHORT, T. Two-dimensional numerical simulation of natural ventilation in a

multi-span greenhouse. Transactions of the ASAE, v.43, n.3, p.745-753, 2000.

LIBERTY, K. R. Yard-Waste Compost Biofilters for Ammonia Adsorption and

Biotransformation. 2002. 374p. Ph.D. Dissertation, Biosystems and Agricultural
Engineering Department, University of Kentucky.

LINDGREN, B.; JOHANSSON, A. V. Design and Evaluation of a Low-Speed Wind-

Tunnel with Expanding Corners. Technical Reports, Royal Institute of Technology.
Stockholm, Sweden, 2002.

LOPES, I. M. Evaluation of transitions for testing agricultural ventilation fans with

the Fan Assessment Numeration System (FANS). 2012. 100p. M.S. Thesis; Biosystems
and Agricultural Engineering, University of Kentucky.

MAGNUSSON, M.; CHRISTIANSSON, A.; SVENSSON, B. Bacillus cereus spores

during housing of dairy cows: Factors affecting contamination of raw milk. Journal of
Dairy Science, v. 90, n. 6, p. 2745-2754, 2007.

MAIA, G. D. N. Ammonia Biofiltration and Nitrous Oxide Generation as Affected by

Media Moisture Content. 2010. 269p. PhD. Dissertation. Biosystems and Agricultural
Engineering, University of Kentucky. Kentucky.

MAIER, R. M.; PEPPER, I. L.; GERBA, C. P. Environmental Microbiology. Academic

Press, San Diego CA, 2000.

MAINTENANCE OF BACTERIAL STRAINS (MBS). Science In The Real World.

Microbes In Action, p. 1-3, 1999.

MAJDOUBI, H.; BOULARD, T.; FATNASSI, H.; BOUIRDEN, L. Airflow and

microclimate patterns in a one-hectare Canary type greenhouse: An experimental and CFD
assisted study. Agricultural and Forest Meteorology, v. 149, p. 1050-1062, 2009.

MALALASEKERA, W.; VERSTEEG, H. K. An introduction to computational fluid

dynamics – The finite volume method. 1st Ed., Pearson Education Limited, Essex,
England, 1995.

267 
 
MALISKA, C. R. Transferência de Calor e Mecânica do Fluidos Computacional. LTC,
Rio de Janeiro, 1995.

MCFARLAND, D. F.; GRAVES, R. E.; TYSON, J. T.; WILSON, T. H. Ridge openings

for naturally ventilated dairy shelters. Cooperative Extension, Penn State, 2007.

MEGONIGAL, J. P.; HINES, M. E.; VISSCHER, P. T. Anaerobic Metabolism:

Linkages to Trace Gases and Aerobic Processes in Schlesinger. Elsevier-Pergamon,
Oxford, UK. 2004.

MELO, L. D. Moisture control methodology for gas phase compost biofilters. 2011.
182p. M.S. Thesis; Biosystems and Agricultural Engineering, University of Kentucky.

MENEGALI, I. ; TINOCO, I. F. F. ; BAETA, F. C. ; CECON, P. R. ; GUIMARAES, M.

C. C. ; CORDEIRO, M. B. Ambiente térmico e concentrações de gases em instalações para
frangos de corte no período de aquecimento. Revista Brasileira de Engenharia Agrícola
e Ambiental, v. 13, p. 984-990, 2009.

MIDWEST PLAN SERVICE (MWPS-32). Mechanical ventilating systems for livestock

housing. 7th edition, Midwest Plan Service, Agricultural and Biosystems Engineering
Department, Iowa State University, 2000.

MIDWEST PLAN SERVICE (MWPS-7). Dairy Freestall Housing and Equipment. 7th
edition, Midwest Plan Service, Agricultural and Biosystems Engineering Department, Iowa
State University, 2000.

MIRANDA, R. C. Simulação de tanque de condicionamento de reagentes por

Mecânica dos Fluidos Computacional. 2009. 67p. M.S. Thesis; Department of
Chemistry, Federal University of Minas Gerais, Belo Horizonte.

MITCHELL, C. D.; ROSS, P. A. Model study of air flow in two calf houses. Farm
Building Progress, p. 19-22, 1977.

MOHSENIN, N. N. Thermal Properties of Foods and Agricultural Materials. Gordon

and Breasch, New York, 1980.

MOZAMMEL, H; SHAHAB, S; TONY, B; SUDHAGAR, M; LADAN, J; LIM, J.

Interaction of Particle size, Moisture content and Compression Pressure on the Bulk
density of Wood chip and Straw. In: Proceedings… CSBE/SCGAB Annual Conference.
The Canadian Society for Bioengineering, 2006.

MUELLER, H. J.; KRAUSE, K. H. Flow simulation for design of ventilation systems in

animal houses. Building Simulation, New York, p.932-936, 2007.

MUELLER, T. J. Handbook of Flow Visualization. Yang, W. J.(Ed.): Hemisphere, New

York, 1989.

268 
 
MUKHTAR, S.; KALBASI, A.; AHMED, A. Carcass Disposal: A Comprehensive
Review. National Agricultural Biosecurity Center, Kansas State University, p.85, 2004.

MURPHY, G. C. E. Similitude in engineering. New York: Ronald Press, 1950.

NÄÄS, I. A.; MOURA, D. J. BUCKLIN, R. A. FIALHO, F. B. An algorithm for

determining opening effectiveness in natural ventilation by wind. Transactions of the
ASAE, v. 41, n. 3, p. 767-771, 1998.

NAKASAKI, K.; SHODA, M.; KUBOTA, H. Effect of temperature on composting of

sewage sludge. Applied and Environmental Microbiology, v. 500, p. 1526–1530, 1985.

NAPPI, P.; BARBERIS, R. Compost as growing medium: chemical, physical and

biological aspects. Acta Horticulturae, v. 342, p. 249–256, 1993.

NATURAL RESOURCES CONSERVATION SERVICES (NRCS), 2010. Part 637

Environmental Engineering, National Engineering Handbook, U.S. Department of
Agriculture, Washington, D.C.

NORTHEAST REGIONAL AGRICULTURAL ENGINEERING SERVICE (NRAES-76).

Guideline for Planning Dairy Freestall Barns. Ithaca, N.Y.: Northeast Regional
Agricultural Engineering Service. 1995.

NORTHEAST REGIONAL AGRICULTURAL ENGINEERING SERVICE (NRAES-54).

On-Farm Composting Handbook. Waste Management Publications. 1992.

NORTON, T.; GRANT, J.; FALLON, R.; SUN, D. W. A computational fluid dynamics
study of air mixing in a naturally ventilated livestock building with different porous eave
opening conditions. Biosystems Engineering, v. 106, p. 125-137, 2010a.

NORTON, T.; GRANT, J.; FALLON, R.; SUN, D. W. Optimizing the ventilation
configuration of naturally ventilated livestock buildings for improved indoor environmental
homogeneity. Building and Environment, v. 45, p. 983–995, 2010b.

NORTON, T.; SUN, D.; GRANT, J., FALLON, R.; DODD, V. Applications of
computational fluid dynamics (CFD) in the modelling and design of ventilation systems in
the agricultural industry: A review. Bioresource Technology, v. 98, p. 2386–2414, 2007.

OGILVIE, J. R.; BOYD, K. G. Air-movement in modified open front monoslope swine

barns. In: Proceedings… CSAE, Can. Soc. Agric. Eng., Ottawa, Ont., 1984.

PATANKAR, S.V.; SPALDING, D.V. A calculation procedure for heat, mass and
momentum transfer in three-dimensional parabolic flows. International Journal for Heat
and Mass Transfer, v. 15, p. 1787–1806, 1972.

PAWAR, S. R.; CIMBALA, J. M.; WHEELER, E. F.; LINDBERG, D. V. Analysis of

poultry house ventilation using computational fluid dynamics. Transactions of the
ASABE, v. 50, n. 4, p. 1373-1382, 2007.
269 
 
PEIGNE, J.; GIRARDIN, P. Environmental impacts of farm-scale composting practices.
Water, Air, and Soil Pollution, v. 153, p. 45-68, 2004.

PEREIRA, A. R.; MACHADO, E. C. Análise quantitativa do crescimento de

comunidade vegetal. Campinas: Instituto Agronômico, (Boletim Técnico), 114. 33p.
1987.

PETZEN, J.; WOLFANGER, C.; BONHOTAL, J.; SCHWARZ, M.; TIMOTHY, T.;
YOUNGERS, N. Eagleview compost dairy barn - case study. Farm viability institute.
2009.

RAE, W. H.; POPE, A. Low-speed wind tunnel testing. 2nd ed. John Wiley & sons,
1984.
RAGHOEBARSING, A. A.; POL. A.; VAN DE PAS-SCHOONEN, K. T.; SMOLDERS,
A. J. P.; ETTWIG, K. F.; RIJPSTRA, W. I. C; SCHOUTEN, S.; DAMSTE, J. S. S.; OP
DEN CAMP, H. J. M.; JETTEN, M. S. M; STROUS, M. A microbial consortium couples
anaerobic methane oxidation to denitrification. Nature, v. 440, p. 918-921, 2006.

REECE, F. N.; LOTT, B. D.; DEATON, J. W. Ammonia in the atmosphere during

brooding affect performance of broiler chikens. Poultry Science, v. 59, n. 1, p. 486-488,
1980.

ROBERTS, L. Technology concepts: Power hydrodynamics. Tech Directions, v. 60, n. 7,

p. 10, 2001.

ROBERTSON, L. A.; CORNELISSE, R.; DE VOS, P.; HADIOETOMO, R.; KUENEN, J.

G. Aerobic denitrification in various heterotrophic nitrifiers. Antonie van Leeuwenhoek,
v. 56, p. 289-299, 1989.

ROCHA, K. S. O. Modelagem e simulação do processo de resfriamento de grãos

armazenados em silos graneleiros por meio das leis de conservação de energia e
massa. 2012; 150p. PhD. Dissertation. Department of Agricultural Engineer, Federal
University of Viçosa, Viçosa.

ROSEN, C.; HALBACK, T. R.; MUGAAS, R. Composting and mulching: A guide to

managing organic yard waste. University of Minnesota Extension Publication. St. Paul,
Minnesota. 2000.

ROY, J. C.; BOULARD, T.; KITTAS, C.; WANG, S. Convective and ventilation transfers
in greenhouses, part 1: the greenhouse considered as a perfectly stirred tank. Biosystems
Engineering, v. 83, p. 1–20, 2002.

RUSSELLE, M. P.; BLANCHET, K. M.; RANDALL, G. W.; EVERETT, L. A. 2009.

Characteristics and nitrogen value of stratified bedded pack dairy manure. Online. Crop
Management. Available at: http://naldc.nal.usda.gov. Accessed 26 Feb 2012.

270 
 
RYNK, R. On Farm Composting Handbook. Northeast regional agricultural
engineering service. Cooperative Extension, Ithaca, NY. 1992.

SAINSBURY, D.W.B. Health problems in intensive animal production. In: CLARK, J.A.
Environmental aspects of housing for animal production. Butterworths: England, p.
439-54, 1981.

SALES, G. T. Assessment of Biofilter Media particle Sizes for Removing Ammonia.

2008. 143p. M.S. Thesis; Biosystems and Agricultural Engineering, University of
Kentucky.

SANTOS, A. A. C. Modelagem numérica com validação experimental de escoamento

através do bocal inferior de um elemento combustível nuclear. 2008. 164p. PhD.
Dissertation. Department of Mechanical Engineering, Federal University of Minas Gerais,
Belo Horizonte.

SCHAUBERGER, G., PIRINGER, M., PETZ, E. Steady state balance model to calculate
the indoor climate of livestock buildings, demonstrated for finishing pigs. International
Journal of Biometeorology, v. 43, p. 154–162, 2000.

SCHLOSS, P. D.; HAY, A. G.; WILSON, D. B.; WALKER, L. P. Molecular assessment

of inoculums efficacy and process reproducibility in composting using ARISA.
Transactions of the ASAE, v. 46, p. 919–927, 2003.

SCHULZE, K.L. Continuous thermophilic composting. Compost Science, v. 3, p.22–34,

1962.

SCHURING, D. J. Scale models in engineering – fundamentals and applications. New

York: Calspan Corporation, 1977.

SEEDORF, J.; HARTUNG, J.; SCHRODER, M.; LINKERT, K. H.; PEDERSEN, S.;
TAKAI, H.; JOHNSEN, J. O.; METZ, J. H. M.; GROOT KOERKAMP, P. W. G.; UENK,
G. H.; PHILLIPS, V. R.; HOLDEN, M. R.; SNEATH, R. W.; SHORT, J. L. L.; WHITE,
R. P.; WATHES, C. M. Survey of Ventilation Rates in Livestock Buildings in Northern
Europe. Journal of Agricultural Engineering Research, v. 70, p. 39-47, 1998.

SENGA, Y.; MOCHIDA, K.; OKAMOTO, N. Nitrous oxide in brackish Lake Nakaumi,
Japan II: The role of nitrification and denitrification in N2O accumulation. Liminology, v.
3, p. 21-27, 2002.

SHANE, E. M.; ENDRES, M. I.; JANNI, K. A. Alternative bedding materials for compost
bedded pack barns in Minnesota: a descriptive study. Applied Engineering in
Agriculture, v. 26, n. 3, p. 465-473, 2010.

SHINDO, S.; BRASK, O. A smoke generator for low speed wind tunnels. Technical
Note No. 69-1, University of Washington, Department of Aeronautics & Astronautics.
1969.

271 
 
SIMANGO, D. G.; SCHULTE, D. D. Effect of roof slope on ventilation of non-
mechanically ventilated, single-slope MOF swine buildings. Transactions of the ASAE,
Amer. Soc. Agric. Eng., St. Joseph, Mich, Paper no. 83-4025, 1983.

SINGH, T. N.; SINHA, S.; SINGH V. K. Prediction of thermal conductivity of rock

through physico-mechanical properties. Building and Environment, v. 42, p. 146–155,
2007.

SMITH, J. F.; HARNER, J. P. Fan Placement and Heat Stress Abatement in Four-row
Freestall Barns. Kansas State University Agricultural Experiment Station and Cooperative
Extension Service. 2001.

STROM, P. F. Effect of temperature on bacterial species diversity in thermophilic solid

waste composting. Applied and Environmental Microbiology, v. 50, p. 899–905, 1985.

STROUS, M.; JETTEN, M. S. M. Anaerobic oxidation of methane and ammonium.

Annual Review of Microbiology, v. 58, p. 99–117, 2004.

SUN, H. Theoretical and experimental study of a High Rise-Hog Building for

improved utilization and environmental quality protection. 2004. 200p. PhD.
Dissertation. Department of Agriculture Enginner and Food, University of Ohio.

SUN, S. H.; KEENER, R.; STOWELL, R.; MICHEL, F. C. Two-dimensional

computational fluid dynamics (CFD) modeling of air and ammonia distribution in a High-
RiseTM Hog Building. Transactions of the ASAE, v. 45, n. 5, p. 1559-1568, 2002.

SURFACE WATER INTEGRATED MANAGEMENT (SWIM), 2011. Characterizing

dairy manures and slurries. Available at: www.envirolink.govt.nz. Accessed 12 May 2012.

TAKAI, H.; MÖLLER, F.; IVERSEN, M.; JORSAL, S.E.; BILLE-HANSEN, V. Dust
control in swine buildings by spraying of rapeseed oil. In: Proceedings… Livestock
Environment Symposium, 4., 1993, St. Joseph: American Society of Agricultural
Engineering, p.726-33, 1993.

TARABA, J.; BEWLEY, J. Winter Management of Dairy Compost Bedded Pack

Barns. Extension Fact Sheet, 2010.

TATMAN, N. Wind Tunnel Design and Operation. 2006. 26p. PhD. Dissertation.
Department of Physic. Aerospace Engineering Virginia Tech, Virginia.

TEYE, F. K.; HAUTALA, M. Adaptation of an ammonia volatilization model for a

naturally ventilated dairy building. Atmospheric Environment, v.42, p.4345–4354, 2008.

TEYE, F. K. Microclimate and gas emissions in dairy buildings: Instrumentation,

theory and measurements. 2008. 65p. PhD. Dissertation. Department of Agrotechnology,
University of Helsinki.

272 
 
TINÔCO, I. F. F.; SARAZ, J. A. O.; GATES, R. S.; DAMASCENO, F. A.; MARIN, O. L.
Z. 3D-CFD modeling of a typical uninsulated and internal misting tunnel ventilated
Brazilian poultry house. Transactions of the ASAE, St. Joseph, Pittsburgh: ASAE. paper
No 1009150, p. 35-42, 2010.

TUCKER, C. B.; WEARY, D. M.; FRASER, D. Free-Stall Dimensions: Effects on

preference and stall usage. Journal of Dairy Science, v. 87, p. 1208-1216, 2004.

VAN DER TOL, P. P. J.; METZ, J. H. M.; NOORDHUIZEN-STASSEN, E. N.; BACK,

W.; BRAAM, C. R.; WEIJS, W. A. The vertical ground reaction force and the pressure
distribution on the claws of dairy cows while walking on a flat substrate. Journal of Dairy
Science, v. 86, p. 2875–2883, 2003.

VAN DER TOL, P. P. J.; METZ, J. H. M.; NOORDHUIZEN-STASSEN, E. N.; BACK,

W.; BRAAM, C. R.; WEIJS, W. A. The pressure distribution under the bovine claw during
square standing on a flat substrate. Journal of Dairy Science, v. 85, p. 1476–1481, 2002.

VAN GINKEL, C. T.; RAATS, P. A. C.; VAN HANEGHEM, I. A. Bulk density and
porosity distributions in a compost pile. Netherlands Journal of Agricultural Science, v.
47, p. 105–121, 1999.

VAN OUWERKERK, E. N. J.; PEDERSEN, S. Application of the carbon dioxide mass

balance method to evaluate ventilation rates in livestock buildings. In: Proceedings... XII
World Congress on Agricultural Engineering, Milano, Italy, p. 516–529, 1994.

VAN WAGENBERG, A. V.; BJERG, B.; BOT, G. P. A. Measurements and simulation of

climatic conditions in the animal occupied zone in a door ventilated room for piglets. In:
Proceedings… Agricultural Engineering International: The CIGR Journal of Scientific
Research and Development, v.4, p.20, 2004.

VEKARIYA, P.; SUBBAIAH, R.; MASHRU, H. Hydraulics of microtube emitters: a

dimensional analysis approach. Irrigation Science, v. 29, n. 4, p. 341-350, 2011.

VOKEY, F. Bedding key to compost dairy barn success. PRO-DAIRY, Northeast

DairyBusiness, p. 37-42, 2008.

WAGNER, P. E. 2002. Bedded pack shelters. http://capitaldairy.cas.psu.edu. Accessed 02

May 2012.
WALLIS, R. A. Axial Flow Fans and Ducts. Wiley, New York. 1983.

WARNER, R. C. The kinetics of the hydrolysis of urea and of arginine. The Journal of
Biological Chemistry, v. 142, p. 705-723, 1942.

WASS, J. A. SigmaPlot 11: Now with total sigmaStat integration. Scientific

Computing, v. 26, n. 1, p. 21-23, 2009.

273 
 
WEARY, D. M.; TASZKUN, I. Hock lesions and free-stall design. Journal of Dairy
Science, v. 83, p. 697-702, 2000.

WEINDORF, D. C.; WITTIE, R. Determining Particle Density in Dairy Manure Compost.

The Texas Journal of Agriculture and Natural Resource, v. 16, p. 60-63, 2003.
WELLER, S. A.; HELDMAN, D. R.; ESMAY, M. L. Air-flow characteristics of a scale
Model chamber. Transactions of the ASAE, St. Joseph, Mich, Paper No. 67-438, 1970.

WELLS, G. D. 2004. Dairy Barn Ventilation - Exhaust Fan Systems. University of

Vermont Extension. Available at: www.uvm.edu/extension. Accessed 12 May 2012.

WILSON, J. D.; BISHOP, R. G. A model study of alternative ventilation systems for a

broiler house. Transactions of the ASAE, St. Joseph, Mich, Paper No. 71-914, 1974.

WINCHELL, W.; DARBY, D.; BORG, R. Alberta feedlot management guide. 1st edition,
Engstrom, D. F. (edit.). Feeding pen design. Alberta Agriculture Food and Rural
Development. Edmonton, AB, 1996.

WORLEY, M. S.; MANBECK, H. B. Modelling particle transport and air flow in ceiling
inlet ventilation systems. Transactions of the ASAE, v. 38, n.1, p.231–239, 1995.

WRIGHT, P.; INGLIS, S. Biodrying Dairy Manure. In: Proceedings... International

Symposium on Composting and Compost Utilization. Emmaus, PA, p. 996-1007, 2002.

XIA, B.; SUN, D. W. Applications of computational fluid dynamics (CFD) in the food
industry: a review. Computers and Electronics in Agriculture, v. 34, n. 1, p. 5-24, 2002.

YUN, T. S.; SANTAMARINA, J. C. Fundamental study of thermal conduction in dry soils.

Granular Matter, v. 10, p. 197–207, 2008.

ZDANOWICZ, M.; SHELFORD, A.; TUCKER, C. B.; WEARY, D. M.; VON

KEYSERLINGK, M. A. G. Bacterial populations on teat ends of dairy cows housed in free
stalls and bedded with either sand or sawdust. Journal of Dairy Science, v. 87, n. 6, p.
1694-1701, 2004.

ZHANG, G; STROM, J. S.; LI, B.; ROM, H. B.; MORSING, S.; DAHL, P.; WANG, C.
Emission of Ammonia and Other Contaminant Gases from Naturally Ventilated Dairy
Cattle Buildings. Biosystems Engineering, v. 92, p. 355-364, 2005.

ZHANG, G.; MORSING, S.; BJERG, B.; SVIDT, K.; STROM, J.S. Test room for
validation of airflow patterns estimated by computational fluid dynamics. Journal of
Agricultural Engineering Research, v. 76, p. 141–148, 2000.

ZOES, V.; DINEL, H.; PARÉ, T.; JAOUICH, A. Growth substrates made from duck
excreta enriched wood shavings and source-separated municipal solid waste compost and
separates: physical and chemical characteristics. Bioresource Technology, v.78, p.21-30,
2001.

274 
 
 7 - APPENDICES

Appendix A:
Table A.1 - Barn characteristics of 42 CBP barns in Kentucky.
Table A.2 - Characteristics of roof and ridge in 42 CBP barns in Kentucky.
Table A.3 - Description of barn ventilation system and lights in CBP barns in
Kentucky.
Table A.4 - Average bedding temperature (°C) at surface and two depths in nine
different locations of CBP barns.
Table A.5 - Moisture content (%) of bedding in CBP barns in Kentucky.
Table A.6 - Bedding chemical characteristics in 42 bedding compost materials.

Appendix B:
Figure B.1 to B.78 - Drawings of CBP barns evaluated in this study.

Appendix C:
Figure C.1 - Compost compact device designed used in this study.

Appendix D:
D.1 - Determination of groups π Buckingham.
Table D.1 - Variables affecting ventilation characteristics in CBP barn.
Table D.2 - Pi terms and independent dimensionless found.
D.2 - The dimensionless groups found.
D.3 - Calculate of air speed and floor surface temperature to the full model.
Table D.3 - Thermal properties of the air at atmospheric pressure.

Appendix E:
Figure E.1 - Types of ridge vents used in this study.

Appendix F:
Table F.1 to F.20 - Statistical analyses of compost materials.

275 
 
 APPENDIX A
Table A.1 - Barn characteristics of 42 CBP barns in Kentucky.

Barn Pack Feed

Waterers
dimension dimensions Alley Stock density Cost
Barn Occupied
L W L W L W L
(m) (m) (m) (m) (m) (m) (m) (m2 / cow) (Dollars)
1 41.5 15.3 41.5 15.3 1.8 0.6 - 10.0 $60,000 Oct. 2010
2 76.3 30.3 53.4 30.3 - - - 9.3 $80,000 and $110,000 including fans Mar. 2008
3 87.8 16.8 87.8 12.2 3.7 0.6 87.8 7.4 $175,000 -
4 51.5 9.2 51.5 9.2 1.8 0.6 - 5.1 $10,900 Oct. 2010
5 48.8 15.3 42.7 10.7 - - - 5.3 - 2003
6 30.5 16.5 24.4 12.2 - - 30.5 12.4 - Jun. 2010
7 36.6 16.3 36.6 12.5 2.0 0.7 36.6 11.7 $50,000 Nov. 2009
8 65.9 29.3 65.9 7.9 1.2 0.9 65.9 3.7 $60,000 without concrete Jan. 2011
9 76.3 27.6 76.3 18.5 3.1 0.6 76.3 13.5 $120,000 Sep. 2010
10 61.0 30.5 61.0 30.5 - - - 10.7 $120,000 including fans and lamps Fall 2009
11 91.5 39.0 91.5 15.3 1.8 0.6 91.5 9.3 Feb. 2011
$150,000 each with fans and labor
12 91.5 39.0 91.5 15.3 1.8 0.6 91.5 9.3 Feb. 2011
13 61.0 37.8 61.0 13.7 1.8 0.6 61.0 8.8 $85,000 each, over $140,000 for Oct. 2010
14 61.0 37.8 61.0 13.7 1.8 0.6 61.0 8.8 concrete, total $400,000 without fans. Oct. 2010
15 18.3 18.3 18.3 18.3 3.7 0.6 - 2.9 - Apr. 2010
16 14.6 13.9 14.6 10.1 2.4 0.6 14.6 0.8 - -
17 37.8 13.7 37.8 13.7 1.5 1.5 21.4 4.3 $55,000 without fans 2008
18 61.0 28.4 61.0 24.4 3.7 0.6 91.5 12.9 $90,000 Nov. 2008
19 21.4 14.0 11.6 14.0 - - - 1.7 $8,000 Oct. 2006
20 22.0 18.0 22.0 12.0 0.7 0.7 22.0 0.9 - 7 years
21 36.6 15.3 36.6 15.3 - - - 7.9 $30,000 with fans June 2007
22 61.0 18.9 61.0 15.3 1.8 0.6 61.0 10.2 $40,000 2002
Continue...

276 
 
Table A.1 - Barn characteristics of 42 CBP barns in Kentucky.
Barn Pack Feed
Waterers
dimension dimensions Alley Stock density Cost
Barn Occupied
L W L W L W L
(m) (m) (m) (m) (m) (m) (m) (m2 / cow) (Dollars)
23 61.0 15.3 61.0 15.3 - - - 25.8 $85,000 2007
24 30.5 18.9 30.5 15.3 - - 30.5 12.9 $35,000 2010
25 51.2 38.4 51.2 29.9 2.4 0.8 102.5 9.3 $204,000 Fev. 2010
26 61.0 24.1 61.0 15.6 26.8 0.4 61.0 7.0 $85,000 Jun. 2009
27 36.6 18.0 36.6 12.2 2.4 0.4 36.6 8.6 $65,000 without labor Sep. 2011
28 36.6 15.3 36.6 15.3 2.4 1.2 - 12.1 $50,000 Aug. 2008
29 61.0 23.8 61.0 14.6 3.7 0.6 45.8 10.0 $105,000 with concrete Feb. 2007
30 45.8 23.4 45.8 18.8 3.1 0.6 33.6 8.8 $35,000 for building alone Dec. 2010
31 42.1 45.8 45.8 31.7 3.7 0.6 91.5 11.0 $207,000 Nov. 2010
32 30.5 18.3 30.5 18.3 3.7 0.6 - 11.9 $30,000 Dec. 2008
33 61.0 24.1 61.0 15.6 3.7 0.6 61.0 10.8 $86,000 Nov. 2010
34 51.2 19.1 0.0 0.0 - - 51.2 3.0 $20,000 Oct. 2010
35 36.6 12.1 36.6 8.5 3.1 0.4 36.6 10.4 $60,000 Fall 2007
36 43.9 14.6 43.9 11.0 - - 58.6 6,9 - May 2004
$17,000 (Hoop); $25,000 (Metal barn);
37 45.8 10.7 45.8 10.7 0.3 0.7 - 10.0 2007
and $30,000 with feed alley.
$17,000 (Hoop); $25,000 (Metal barn);
38 36.6 21.7 36.6 16.8 - - 26.8 12.3 2007
and $30,000 with feed alley.
$80,000 for both, including fans and
39 48.8 22.0 48.8 18.2 1.8 0.6 48.8 12.2 Jul. 2010
concrete.
$80,000 for both, including fans and
40 48.8 22.0 48.8 18.2 1.8 0.6 48.8 12.6 2002
concrete.
41 30.5 15.3 30.5 15.3 1.5 1.5 61.0 5.2 $14,000 Jun. 2010
42 53.7 14.9 53.7 14.9 - - - 8.0 $25,000 Jun. 2010

277 
 
Table A.2 - Characteristics of roof and ridge in 42 CBP barns in Kentucky.

Orientation Roof Roof Pitch

Barn Ridge
Direction Angle Structure Type V:H
1 Open ridge with cover E 249° Wood Metal 4:12
2 Open ridge with cover SW 241° Metal Metal 4:12
3 Overshot SW 36° Metal Metal 4:12
4 Hoop structure SW 80° Hoop Vinyl Hoop
5 Overshot SW 31° Wood Metal 4:12
6 Capped ridge NW 145° Wood Metal 4:12
7 Overshot SW 40° Wood Metal 4:12
8 Open ridge with cover NW 210° Metal metal 4:12
9 Overshot NW 75° Metal Metal 4:12
3:12
10 Overshot SE 349° Wood Metal
2 1/2 :12
11 Overshot SW 23° Wood Metal 3:12
12 Overshot SW 23° Wood Metal 2 1/2 :12
13 Open ridge without cover SE 329° Wood Metal 3:12
14 Open ridge without cover SE 329° Wood Metal 3:12
15 Capped ridge SW 334° Wood Metal 4:12
16 Capped ridge NE 243° Wood Metal 5:28
17 Overshot S 240° Wood Metal 4:12
18 Overshot SW 90° Wood Metal 3 1/2 :12
19 Capped ridge NW 28° Wood Metal 3 1/2 :12
1 1/2 : 12
20 Overshot SW 15° Metal Metal
2:12
21 Overshot NE 233° Wood Metal 4:12
22 Capped ridge SE 35° Wood Metal 4:12
23 Capped ridge SE 305° Wood Metal 4:12
24 Capped ridge SE 286° Wood Metal 4:12
25 Open ridge with cover SE 291° Wood Metal 3:12
26 Overshot SE 265° Wood Metal 3 1/2 :12
27 Overshot SE 330° Wood Metal 4:12
28 Overshot SW 37° Wood Metal 4:12
29 Overshot SE 284° Wood Metal 3:12
30 Overshot NE 253° Wood Metal 3 1/2 :12
Continue...

278 
 
Table A.2 - Characteristics of roof and ridge in 42 CBP barns in Kentucky.

Roof
Orientation Roof
Barn Ridge Pitch
Direction Angle Structure Type V:H
31 Open ridge with cover SE 345° Metal Metal 4:12
32 Overshot SW 90° Wood Metal 4:12
33 Overshot SW 74° Wood Metal 3 1/2 :12
34 Overshot NW 112° Wood Metal 3.2:12
35 Open ridge with cover NW 117° Metal Metal 4:12
36 Open ridge with cover SW 335° Wood Metal 4:12
37 Hoop structure NE 224° Vinyl Metal Hoop
38 Overshot NW 162° Wood Metal 4:12
39 Overshot SW 17° Wood Metal 5:12
40 Overshot SW 17° Wood Metal 5:12
41 Open ridge with cover NW 102° Wood Metal 4:12
42 Overshot NE 204° Wood Metal 4:12

279 
 
Table A.3 - Description of barn ventilation system and lights in CBP barns in Kentucky.

Ventilation system Light

Barn Type Power Size Power
Amount Type Amount
(W) (m) (W)
1 Box fans 8 745.7 1.3 - - -
2 HVLS Fans 3 1491.4 7.3 HID 3 175
3 Box fans 17 745.7 1.3 HID 2 175
4 Tunnel Ventilation - - - - - -
5 Natural ventilation - - - - - -
Box fans 10 494.6 0.5 Incandescent 10 100
6 Box fans 8 745.7 0.9 Incandescent 5
Box fans 2 745.7 1.2 Incandescent 5 100
7 HVLS Fans 2 1491.4 7.3 HID 6 175
8 Box fans 4 745.7 1.3 - - -
9 Box fans 4 745.7 1.3 HID 30 175
10 Box fans 15 745.7 1.3 HID 5 70
HVLS Fans 3 1491.4 6.1 Fluorescent 6 192
11
HVLS Fans 2 1491.4 7.3 - - -
HVLS Fans 3 1491.4 6.1 Fluorescent 6 192
12
HVLS Fans 2 1491.4 7.3 - - -
Compact
13 Natural ventilation - - - 5 32
Fluorescent
Compact
14 Natural ventilation - - - 5 32
Fluorescent
15 HVLS Fans 1 1491.4 7.3 - - -
16 Box fans 4 372.8 0.9 Incandescent 6 100
17 Box fans 6 372.8 0.9 Incandescent 9 100
18 Box fans 20 745.7 1.3 Fluorescent 8 192
Compact
19 Box fans 4 372.8 0.9 2 32
Fluorescent
Box fans 6 372.8 0.9 - - -
20
Box fans 4 559.3 1.2 - - -
Box fans 6 372.8 0.9 HID 3 175
21
Box fans 2 745.7 1.2 - - -
22 Natural ventilation - - - HID 12 150
23 Natural ventilation - - - HID 6 150
24 Natural ventilation - - - - - -
Continue...

280 
 
Table A.3 - Description of barn ventilation system and lights in CBP barns in Kentucky.

Ventilation system Light

Barn Type Power Size Power
Amount Type Amount
(W) (m) (W)
25 HVLS Fans 6 1491.4 7.3 HID 24 150
26 HVLS Fans 3 1491.4 7.3 HID 2 150
Compact
27 Natural ventilation - - - 11 13
Fluorescent
Compact
28 Box fans 6 1118.5 1.2 14 13
Fluorescent
29 HVLS Fans 3 1491.4 7.3 - - -
30 HVLS Fans 2 1491.4 7.3 - - -
31 Natural ventilation - - - - - -
32 HVLS Fans 2 1491.4 7.3 Fluorescent 3 96
33 Natural ventilation - - - - - -
34 Natural ventilation - - - Incandescent 4 100
35 HVLS Fans 2 1491.4 7.3 - - -
Box fans 2 186.4 0.6 - - -
36 Box fans 1 372.8 0.9 - - -
Box fans 6 745.7 1.1 - - -
37 Natural ventilation - - - - - -
38 Natural ventilation - - - - - -
39 Box fans 8 745.7 1.3 Fluorescent 5 192
40 Box fans 8 745.7 1.3 Fluorescent 5 192
41 Box fans 8 246.1 0.6 - - -
42 Box fans 8 745.7 1.3 Incandescent 6 100

281 
 
Table A.4 - Average bedding temperature (°C) at surface and two depths in nine different
locations of CBP barns. Bedding material: A (green sawdust). B (kiln-dried
shavings or sawdust), and C (mix).

Pack Temperature
Surface temperature
Bedding 0.1 m depth 0.2 m depth
Barn
material
Average Std Dev Average Std Dev Average Std Dev

1 A 17.1 ± 1.3 31.9 ± 10.2 35.2 ± 9.4

2 A 16.5 ± 3.6 41.5 ± 10.6 53.4 ± 4.6
3 B 16.3 ± 3.8 36.2 ± 9.5 52.1 ± 4.6
4 C 15.5 ± 3.5 41.1 ± 18.7 52.4 ± 7.6
5 B 22.2 ± 1.3 26.0 ± 6.1 27.2 ± 6.6
6 B 16.2 ± 3.1 34.5 ± 15.0 34.0 ± 10.1
7 B 13.3 ± 4.7 38.3 ± 6.0 42.2 ± 8.6
8 B 17.8 ± 4.1 24.1 ± 4.8 24.4 ± 3.5
9 C 22.3 ± 11.7 38.6 ± 13.2 42.5 ± 14.8
10 B 17.3 ± 5.6 35.6 ± 8.6 41.4 ± 11.4
11 C 15.2 ± 2.5 30.5 ± 12.0 34.4 ± 6.1
12 C 18.5 ± 12.5 31.5 ± 7.7 36.4 ± 6.9
13 B 11.7 ± 5.3 42.0 ± 10.7 43.6 ± 9.4
14 B 8.9 ± 1.6 40.3 ± 14.2 38.4 ± 14.9
15 B 17.2 ± 10.2 52.3 ± 7.5 58.0 ± 10.3
16 A 8.9 ± 2.9 23.5 ± 12.6 32.1 ± 7.2
17 B 11.7 ± 5.0 37.4 ± 16.6 36.6 ± 17.0
18 B -5.1 ± 3.5 12.3 ± 9.7 20.9 ± 10.9
19 B -0.9 ± 3.8 39.7 ± 5.6 52.8 ± 3.2
20 B 23.7 ± 1.2 44.0 ± 12.4 45.0 ± 11.6
21 B 4.5 ± 2.4 46.1 ± 32.5 50.2 ± 14.7
22 A 6.9 ± 2.7 28.8 ± 5.4 31.6 ± 3.2
23 A 6.0 ± 2.6 15.3 ± 6.0 18.9 ± 6.7
24 A 7.2 ± 2.5 15.3 ± 5.5 19.8 ± 4.9
25 B 7.0 ± 4.2 17.1 ± 3.5 20.2 ± 2.3
26 A 6.8 ± 4.5 39.5 ± 11.2 39.5 ± 9.8
27 B 18.2 ± 9.1 39.4 ± 23.3 43.3 ± 14.7
28 A 9.0 ± 4.2 45.8 ± 11.2 53.0 ± 7.0
29 A 5.4 ± 3.4 15.5 ± 6.7 18.0 ± 5.6
30 A 14.2 ± 9.8 49.2 ± 7.8 52.9 ± 4.5
Continue...

282 
 
Table A.4 - Average bedding temperature (°C) at surface and two depths in nine different
locations of CBP barns. Bedding material: A (green sawdust), B (kiln-dried
shavings or sawdust), and C (mix).
Pack Temperature
Surface temperature
Bedding 0.1 m depth 0.2 m depth
Barn
material
Average Std Dev Average Std Dev Average Std Dev

31 A 11.8 ± 8.4 32.8 ± 17.2 35.1 ± 13.3

32 A 6.7 ± 4.4 42.0 ± 6.7 44.2 ± 6.3
33 B -1.5 ± 16.1 33.0 ± 27.0 37.4 ± 25.2
34 B -0.8 ± 3.7 27.6 ± 29.3 23.3 ± 3.9
35 A -3.5 ± 2.6 8.8 ± 5.2 13.2 ± 4.9
36 A 3.3 ± 2.5 15.7 ± 6.4 18.2 ± 5.8
37 A 4.7 ± 2.9 20.2 ± 15.3 25.3 ± 16.6
38 A 3.8 ± 2.3 13.5 ± 8.8 17.2 ± 9.8
39 A -1.7 ± 4.5 22.8 ± 9.0 28.6 ± 8.7
40 A 0.0 ± 8.3 23.3 ± 15.0 27.7 ± 11.1
41 B -4.0 ± 2.3 25.2 ± 15.1 30.0 ± 5.9
42 C 0.9 ± 3.5 18.2 ± 25.1 24.8 ± 5.1

283 
 
Table A.5 - Moisture content (%) of bedding in CBP barns in Kentucky.

Std 1st 3rd

Barn 1 2 3 4 5 6 7 8 9 Average Min. Max.
Dev Quartile Quartile
1 49.8 53.8 43.2 44.1 38.2 40.4 30.7 56.4 43.7 44.5 8.0 30.7 56.4 40.4 49.8
2 52.0 56.0 61.5 45.7 49.7 49.0 38.3 39.7 49.2 49.0 7.3 38.3 61.5 45.7 52.0
3 46.3 46.5 46.1 56.8 43.4 54.1 58.7 46.1 44.3 49.1 5.8 43.4 58.7 46.1 54.1
4 37.9 30.5 27.8 34.8 44.3 29.0 36.4 44.9 40.0 36.2 6.3 27.8 44.9 30.5 40.0
5 61.7 56.6 46.9 55.0 52.2 47.5 55.2 43.7 51.3 52.2 5.6 43.7 61.7 47.5 55.2
6 42.9 36.9 41.0 32.0 22.9 31.3 50.1 51.8 47.1 39.6 9.6 22.9 51.8 32.0 47.1
7 49.7 49.4 44.4 45.1 45.1 40.8 48.9 48.1 42.7 46.0 3.2 40.8 49.7 44.4 48.9
8 63.5 61.4 59.5 58.2 65.9 61.0 63.8 61.7 59.3 61.6 2.5 58.2 65.9 59.5 63.5
9 38.8 37.5 34.4 33.6 43.7 53.9 45.7 41.3 47.7 41.9 6.6 33.6 53.9 37.5 45.7
10 51.8 45.4 51.4 38.3 64.0 48.7 57.5 53.6 54.5 51.7 7.3 38.3 64.0 48.7 54.5
11 64.9 62.4 60.9 61.8 56.2 62.4 62.3 59.5 60.7 61.2 2.4 56.2 64.9 60.7 62.4
12 65.6 60.2 59.5 62.4 60.5 55.9 60.1 61.7 57.5 60.4 2.8 55.9 65.6 59.5 61.7
13 62.4 54.7 55.7 59.8 63.4 50.6 60.0 57.5 53.7 57.5 4.2 50.6 63.4 54.7 60.0
14 58.7 57.5 60.1 64.0 55.8 58.9 49.8 51.9 58.7 57.3 4.3 49.8 64.0 55.8 58.9
15 56.2 56.0 53.6 55.4 51.0 51.0 57.2 52.9 58.5 54.7 2.7 51.0 58.6 53.0 56.2
16 69.2 70.6 67.1 66.2 66.4 64.6 64.3 65.9 64.4 66.5 2.2 64.3 70.6 64.6 67.1
17 61.6 62.4 52.7 55.7 50.2 59.1 52.6 56.5 53.7 56.0 4.3 50.2 62.4 52.7 59.1
18 65.1 64.6 64.2 64.0 65.6 65.6 66.0 63.5 65.1 64.8 0.8 63.5 66.0 64.2 65.6
19 69.2 69.5 70.2 68.0 71.0 67.3 68.3 68.0 64.5 68.4 1.9 64.5 71.0 68.0 69.5
20 57.0 62.7 53.6 60.9 62.7 62.0 58.3 63.0 63.0 60.3 3.3 53.6 63.0 58.3 62.7
21 63.4 64.5 55.0 56.7 59.1 56.9 58.8 59.2 55.4 58.8 3.3 55.0 64.5 56.8 59.2
22 58.3 56.5 43.9 53.9 48.3 52.6 50.5 49.9 53.4 52.0 4.4 43.9 58.3 50.0 53.9
23 67.0 68.8 69.7 66.7 68.1 65.9 67.2 66.2 66.1 67.3 1.3 65.9 69.7 66.3 68.1
24 62.9 62.9 67.3 63.6 64.2 62.5 63.8 62.9 65.6 64.0 1.6 62.5 67.3 62.9 64.2
25 64.5 63.7 61.1 64.0 59.2 58.9 60.1 59.4 59.5 61.2 2.3 58.9 64.5 59.4 63.8
Continue...

284 
 
Table A.5 - Moisture content (%) of bedding in CBP barns in Kentucky.

Std 1st 3rd

Barn 1 2 3 4 5 6 7 8 9 Average Min. Max.
Dev Quartile Quartile
26 56.6 58.8 54.9 57.9 56.1 54.1 54.9 63.3 58.5 57.3 2.8 54.1 63.3 55.0 58.5
27 68.3 67.9 68.1 66.8 67.0 67.1 67.6 66.1 70.6 67.8 1.3 66.2 70.7 67.0 68.2
28 66.3 65.1 63.3 66.2 62.1 59.6 64.3 65.2 69.3 64.6 2.8 59.7 69.4 63.3 66.2
29 64.5 60.0 62.4 60.9 58.2 59.2 63.4 59.2 61.7 61.1 2.1 58.2 64.5 59.2 62.4
30 46.5 50.0 51.2 43.4 46.2 47.7 44.2 48.2 49.0 47.4 2.6 43.4 51.2 46.2 49.0
31 56.8 61.7 61.9 56.9 60.6 61.0 62.2 65.2 68.6 61.7 3.7 56.8 68.7 60.6 62.3
32 67.7 65.8 65.5 68.3 64.8 66.8 67.2 64.8 66.6 66.4 1.3 64.8 68.3 65.6 67.3
33 69.9 68.9 69.9 68.3 68.7 70.4 70.3 70.4 68.1 69.5 0.9 68.2 70.5 68.7 70.4
34 68.2 71.7 69.1 67.3 66.1 71.9 70.3 67.1 68.0 68.9 2.0 66.2 71.9 67.3 70.3
35 48.6 63.2 69.4 63.3 64.1 62.5 64.6 65.0 62.4 62.6 5.6 48.7 69.4 62.5 64.6
36 71.8 71.9 70.8 71.5 73.2 71.6 71.3 72.8 70.9 71.8 0.8 70.8 73.3 71.3 71.9
39 64.7 66.1 68.2 66.9 65.8 67.0 66.5 66.2 66.9 66.5 1.0 64.8 68.2 66.1 67.0
40 66.6 65.1 62.8 66.7 65.1 61.8 66.1 63.3 64.0 64.6 1.7 61.8 66.7 63.3 66.1
41 72.3 70.3 69.1 73.6 71.6 66.7 71.4 70.9 65.2 70.1 2.7 65.2 73.6 69.1 71.7
42 68.2 68.6 70.2 67.7 68.4 70.2 70.7 67.6 69.9 69.1 1.2 67.6 70.7 68.3 70.2

285 
 
Table A.6 - Bedding chemical characteristics in 42 bedding compost materials.
P K Ca Mg Zn Cu Mn Fe
Barn
(g/kg) (g kg) (g/kg) (g/kg) (ppm) (ppm) (ppm) (ppm)
1 0.5 1.6 2.3 0.7 168.0 30.0 436.9 7271.2
2 0.5 1.6 1.4 0.5 175.0 39.9 189.2 768.3
3 0.2 0.8 0.8 0.2 66.2 9.2 113.1 3624.0
4 0.2 0.4 0.6 0.2 40.7 8.1 121.0 764.9
5 0.4 1.1 1.5 0.5 137.3 32.4 150.3 1304.6
6 0.8 2.0 2.0 0.8 209.5 46.5 297.8 2143.4
7 0.5 2.2 3.6 0.7 202.7 55.8 238.7 1968.7
8 0.5 1.8 1.6 0.4 147.3 38.6 241.8 7281.7
9 0.6 1.7 2.0 0.7 161.7 35.0 818.9 9077.7
10 0.6 1.6 1.5 0.6 142.2 60.9 207.9 1647.1
11 0.3 0.9 4.1 0.5 87.0 16.6 183.2 1744.9
12 0.2 0.7 1.8 0.3 77.0 14.7 167.5 1870.8
13 0.3 1.0 1.5 0.4 94.2 15.0 129.7 3792.4
14 0.4 0.9 1.4 0.4 112.9 28.1 153.9 2935.6
15 0.4 1.4 1.4 0.4 96.1 44.4 165.3 938.8
16 0.5 1.6 1.6 0.5 152.2 55.4 208.0 966.6
17 0.4 1.3 2.7 0.7 123.8 30.2 163.6 1760.7
18 0.6 1.8 2.6 0.7 137.4 39.8 215.1 2756.7
19 0.4 0.9 1.2 0.4 99.8 44.4 166.0 480.8
20 0.2 0.6 0.7 0.2 32.7 13.1 241.2 4960.6
21 0.4 1.0 1.1 0.3 102.8 28.1 129.7 493.3
22 0.3 0.9 0.9 0.3 90.2 17.5 159.0 1428.9
23 0.3 1.0 0.8 0.3 85.3 17.6 140.1 733.2
24 0.4 1.9 1.0 0.4 75.8 12.1 277.1 980.7
25 0.5 1.8 1.4 0.5 139.3 37.3 192.3 922.4
26 0.4 1.0 0.9 0.4 71.9 17.7 210.8 5032.2
27 0.3 0.8 1.0 0.3 52.1 9.6 154.4 3434.2
28 0.4 1.4 1.1 0.4 92.4 19.5 571.6 7582.3
29 0.5 1.7 1.4 0.4 136.0 27.7 224.5 1661.6
30 0.3 1.2 1.1 0.3 140.0 39.3 188.3 471.4
31 0.2 0.7 1.2 0.3 36.5 11.2 196.0 1379.8
32 0.4 1.6 4.3 0.7 138.6 43.3 241.7 5812.8
33 0.3 0.9 0.9 0.3 65.8 9.6 155.5 7487.0
34 0.2 0.6 0.9 0.3 57.2 14.2 115.5 696.6
35 0.4 1.5 1.2 0.5 118.1 38.2 170.8 597.0
Continue...

286 
 
Table A.6 - Bedding chemical characteristics in 42 compost dairy barns in Kentucky.
P K Ca Mg Zn Cu Mn Fe
Barn
(g/kg) (g kg) (g/kg) (g/kg) (ppm) (ppm) (ppm) (ppm)
36 0.4 1.0 1.2 0.3 155.8 29.8 212.1 902.7
37 0.4 1.4 1.3 0.4 79.3 18.4 170.3 2646.6
38 0.3 0.9 1.0 0.3 66.4 12.1 159.0 4738.5
39 0.4 0.9 0.9 0.3 50.0 7.7 158.8 1963.1
40 0.3 0.9 0.9 0.3 67.3 10.0 168.1 1726.6
41 0.4 1.1 0.9 0.3 110.6 19.2 147.8 981.4
42 0.5 1.4 0.8 0.3 70.9 8.0 163.9 1831.2
Average 0.4 1.2 1.5 0.4 106.0 26.3 209.9 2606.2
Std dev. 0.1 0.4 0.9 0.2 44.2 14.8 125.8 2323.9

 
 

287 
 
 APPENDIX B

Figure B.1 - Details about end wall, barn section, and ridge of CBP Barn 1.
288 
 
  

Figure B.2 - Details about plan view and profile of CBP Barn 1.
289 
 
  

Figure B.3 - Details about end wall, barn section, and ridge of CBP Barn 2.

290 
 
  

Figure B.4 - Details about plan view and profile of CBP Barn 2.
291 
 
  

Figure B.5 - Details about end wall, barn section, and ridge of CBP Barn 3.
292 
 
  

Figure B.6 - Details about plan view and profile of CBP Barn 3.
293 
 
  

Figure B.7 - Details about end wall, barn section, and ridge of CBP Barn 4.
294 
 
  

Figure B.8 - Details about plan view and profile of CBP Barn 4.
295 
 
  

Figure B.9 - Details about end wall, barn section, and ridge of CBP Barn 5.
296 
 
  

Figure B.10 - Details about plan view and profile of CBP Barn 5.
297 
 
  

Figure B.11 - Details about end wall, barn section, and ridge of CBP Barn 6.
298 
 
  

Figure B.12 - Details about plan view and profile of CBP Barn 6.
299 
 
  

Figure B.13 - Details about end wall, barn section, and ridge of CBP Barn 7.
300 
 
  

Figure B.14 - Details about plan view and profile of CBP Barn 7.
301 
 
  

Figure B.15 - Details about end wall, barn section, and ridge of CBP Barn 8.
302 
 
  

Figure B.16 - Details about plan view and profile of CBP Barn 8.
303 
 
  

Figure B.17 - Details about end wall, barn section, and ridge of CBP Barn 9.
304 
 
  

Figure B.18 - Details about plan view and profile of CBP Barn 9.
305 
 
  

Figure B.19 - Details about end wall, barn section, and ridge of CBP Barn 10.
306 
 
  

Figure B.20 - Details about plan view and profile of CBP Barn 10.
307 
 
  

Figure B.21 - Details about end wall, barn section, and ridge of CBP Barns 11 and 12.

308 
 
  

Figure B.22 - Details about plan view and profile of CBP Barns 11 and 12.
309 
 
  

Figure B.23 - Details about end wall, barn section, and ridge of CBP Barns 13 and 14.
310 
 
  

Figure B.24 - Details about plan view and profile of CBP Barns 13 and 14.

311 
 
  

Figure B.25 - Details about end wall, barn section, and ridge of CBP Barn 15.
312 
 
  

Figure B.26 - Details about plan view and profile of CBP Barn 15.
313 
 
  

Figure B.27 - Details about end wall, barn section, and ridge of CBP Barn 16.
314 
 
  

Figure B.28 - Details about plan view and profile of CBP Barn 16.
315 
 
  

Figure B.29 - Details about end wall, barn section, and ridge of CBP Barn 17.
316 
 
  

Figure B.30 - Details about plan view and profile of CBP Barn 17.
317 
 
  

Figure B.31 - Details about end wall, barn section, and ridge of CBP Barn 18.
318 
 
  

Figure B.32 - Details about plan view and profile of CBP Barn 18.

319 
 
  

Figure B.33 - Details about end wall, barn section, and ridge of CBP Barn 19.
320 
 
  

Figure B.34 - Details about plan view and profile of CBP Barn 19.
321 
 
  

Figure B.35 - Details about end wall, barn section, and ridge of CBP Barn 20.
322 
 
  

 
 

Figure B.36 - Details about plan view and profile of CBP Barn 20.
323 
 
  

Figure B.37 - Details about end wall, barn section, and ridge of CBP Barn 21.
324 
 
  

Figure B.38 - Details about plan view and profile of CBP Barn 21.
325 
 
  

Figure B.39 - Details about end wall, barn section, and ridge of CBP Barn 22.
326 
 
  

Figure B.40 - Details about plan view and profile of CBP Barn 22.
327 
 
  

Figure B.41 - Details about end wall, barn section, and ridge of CBP Barn 23.
328 
 
  

Figure B.42 - Details about plan view and profile of CBP Barn 23.
329 
 
  

Figure B.43 - Details about end wall, barn section, and ridge of CBP Barn 24.

330 
 
  

Figure B.44 - Details about plan view and profile of CBP Barn 24.

331 
 
  

Figure B.45 - Details about end wall, barn section, and ridge of CBP Barn 25.

332 
 
  

Figure B.46 - Details about plan view and profile of CBP Barn 25.

333 
 
  

Figure B.47 - Details about end wall, barn section, and ridge of CBP Barn 26.
334 
 
  

Figure B.48 - Details about plan view and profile of CBP Barn 26.
335 
 
  

Figure B.49 - Details about end wall, barn section, and ridge of CBP Barn 27.
336 
 
  

Figure B.50 - Details about plan view and profile of CBP Barn 27.
337 
 
  

Figure B.51 - Details about end wall, barn section, and ridge of CBP Barn 28.
338 
 
  

Figure B.52 - Details about plan view and profile of CBP Barn 28.
339 
 
  

Figure B.53 - Details about end wall, barn section, and ridge of CBP Barn 29.
340 
 
  

Figure B.54 - Details about plan view and profile of CBP Barn 29.
341 
 
  

Figure B.55 - Details about end wall, barn section, and ridge of CBP Barn 30.
342 
 
  

Figure B.56 - Details about plan view and profile of CBP Barn 30.
343 
 
  

Figure B.57 - Details about end wall, barn section, and ridge of CBP Barn 31.

344 
 
  

Figure B.58 - Details about plan view and profile of CBP Barn 31.
345 
 
  

Figure B.59 - Details about end wall, barn section, and ridge of CBP Barn 32.
346 
 
  

Figure B.60 - Details about plan view and profile of CBP Barn 32.
347 
 
  

Figure B.61 - Details about end wall, barn section, and ridge of CBP Barn 33.
348 
 
  

Figure B.62 - Details about plan view and profile of CBP Barn 33.
349 
 
  

Figure B.63 - Details about end wall, barn section, and ridge of CBP Barn 34.
350 
 
  

Figure B.64 - Details about plan view and profile of CBP Barn 34.
351 
 
  

Figure B.65 - Details about end wall, barn section, and ridge of CBP Barn 35.
352 
 
  

Figure B.66 - Details about plan view and profile of CBP Barn 35.
353 
 
  

Figure B.67 - Details about end wall, barn section, and ridge of CBP Barn 36.
354 
 
  

Figure B.68 - Details about plan view and profile of CBP Barn 36.
355 
 
  

Figure B.69 - Details about end wall, barn section, and ridge of CBP Barn 37.
356 
 
  

Figure B.70 - Details about plan view and profile of CBP Barn 37.
357 
 
  

Figure B.71 - Details about end wall, barn section, and ridge of CBP Barn 38.
358 
 
  

Figure B.72 - Details about plan view and profile of CBP Barn 38.
359 
 
  

Figure B.73 - Details about end wall, barn section, and ridge of CBP Barn 39 and 40.
360 
 
  

Figure B.74 - Details about plan view and profile of CBP Barn 39 and 40.
361 
 
  

Figure B.75 - Details about end wall, barn section, and ridge of CBP Barn 41.
362 
 
  

Figure B.76 - Details about plan view and profile of CBP Barn 41.
363 
 
  

Figure B.77 - Details about end wall, barn section, and ridge of CBP Barn 42.
364 
 
  

Figure B.78 - Details about plan view and profile of CBP Barn 42.

365 
 
 APPENDIX C

Figure C.1 - Compost compact device designed used in this study.

366 
 
 APPENDIX D

D.1 - Determination of groups π Buckingham.

Table D.1 is showing the dimensional parameters that affect the ventilation
characteristics inside of CBP barn.

Table D.1 - Variables affecting ventilation characteristics in CBP barn.

Dimensional
Variable n◦ Description Units (S.I) Symbols
symbol
1 Building length m l L
2 Building width m w L
3 Building height m h L
4 Gravitational acceleration m s-2 g L T-2
5 Coefficient of heat transfer kg s−3 K−1 h M T−3 θ−1
6 Heat kg m2 s-2 Q M L2 T-2
7 Wind velocity m s-1 v L T-1
8 Temperature of air K T θ
9 Density of air kg m-3 M L-3
10 Specific heat of air m2 s-2 K-1 c L2 T-2 θ-1
11 Heat transfer coefficient kg s-3 K-1 U M T-3 θ-1
12 Dynamic viscosity of air kg m-1 s-1 M L-1 T-1
13 Thermal conductivity of air kg m s-3 K-1 k M L T-3 θ-1
14 Coefficient of thermal expansion βe k-1 θ-1

Determination of π groups
The number of dimensional parameter is n and r is the number of primary dimensions.
So, dimensional parameter is n = 14 and primary dimensions is k = 4. Using four selected
repeating variable (l , v , T , and ρ ) which together must contain L, M, T, and θ.
Therefore, n - k = 10 π groups, forming the dimensionless groups are obtained:
 
Number of physical quantities: n = 14
Number of basic dimensions: k = 4
Pi term number: n – k = 10

367 
 
 Physics quantities
Building length: l L
Wind velocity: v L2 T-1
Temperature of air: T θ
Density of air: M L-3

Determination of π terms
→ Variable 2: Building width, w
π l ∙v ∙ T ∙ ρ ∙ w   
π L ∙ L T ∙ θ ∙ M L ∙ L   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x 2y 3k 1 0 → x 1
y 0 → y 0
k 0 → k 0
z 0 → z 0
π l ∙ v ∙ T ∙ ρ ∙ w   
π w∙l
 
→ Variable 3: Building height, h
π l ∙v ∙ T ∙ ρ ∙ h   
π L ∙ L T ∙ θ ∙ M L ∙ L   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x 2y 3k 1 0 → x 1
y 0 → y 0
k 0 → k 0
z 0 → z 0
π l ∙ v ∙ T ∙ ρ ∙ h   
π h ∙ l

π → Variable 4: gravitational acceleration, g 

π l ∙v ∙ T ∙ ρ ∙ g   
π L ∙ L T ∙ θ ∙ M L ∙ M L   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   

368 
 
 x y 3k 2 0 → x 1
y 0 → y 0
z 0 → z 0
k 1 0
→ k 1
π l ∙v ∙T ∙ρ ∙ g   
 π g∙l ∙ρ  
 
→ Variable 5: Coefficient of heat transfer, hc
π l ∙v ∙ T ∙ ρ ∙ h   
π L ∙ L T ∙ θ ∙ M L ∙ M T θ   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x y 3k 0 → x 0
y 3 0 → y 3
k 1 0 → k 1
z 1 0
→ z 1
π l ∙v ∙T ∙ρ ∙ h   
π v ∙T ∙ρ ∙h

→ Variable 6: Heat, Q
π l ∙v ∙ T ∙ ρ ∙ Q   
π L ∙ L T ∙ θ ∙ M L ∙ M L T   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x y 3k 2 0 → x 3
y 2 0 → y 2
z 0 → z 0
k 1 0
→ k 1
π l ∙v ∙T ∙ρ ∙ Q    
π Q ∙ l ∙v ∙ρ
 
→ Variable 10: Specific heat of air, c
π l ∙v ∙ T ∙ ρ ∙ c   
π L ∙ L T ∙ θ ∙ M L ∙ L T θ   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   

369 
 
 x y3k 2 0 → x 0
y 2 0 → y 2
z 1 0 → z 1
k 0
→ k 0
π l ∙v ∙ T ∙ ρ ∙ c   
 π c ∙v ∙T  

→ Variable 11: Heat transfer coefficient, U 

π l ∙v ∙ T ∙ ρ ∙ U   
π L ∙ L T ∙ θ ∙ M L ∙ M T θ   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x y 3k 0 → x 0
y 3 0 → y 3
z 1 0 → z 1
k 1 0
→ k 1
π l ∙v ∙T ∙ρ ∙ U   
 π U∙v ∙ T ∙ ρ  
 
→ Variable 12: Dynamic viscosity of air,  
π l ∙v ∙T ∙ρ ∙   
π L ∙ L T ∙ θ ∙ M L ∙ M L T   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x2y 3k 1 0 → x 1
y
1 0 → y 1
z 0 → z 0
k 1 0 → k 1
π l ∙v ∙T ∙ρ ∙   
 π ∙ l ∙v ∙ρ
 
π → Variable 13: Thermal conductivity of air, k 
π l ∙v ∙ T ∙ ρ ∙ k   
π L ∙ L T ∙ θ ∙ M L ∙ M L T θ   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   

370 
 
 x y
3k 1 0 → x 1
y 3 0 → y 3
z 1 0 → z 1
k 1 0
→ k 1
π l ∙v ∙T ∙ρ ∙ k   
 π k∙l ∙v ∙T ∙ρ  
 
π → Variable 14: Coefficient of thermal expansion, βe 
π l ∙v ∙ T ∙ ρ ∙ β   
π L ∙ L T ∙ θ ∙ M L ∙ θ   
π L ∙ T ∙ M ∙ θ L ∙ T ∙ M ∙ θ   
x y 3k 0 → x 0
y 0 → y 0
z 1 0 → z 1
k 0 → k 0
π l ∙v ∙ T ∙ ρ ∙ β   
 π β ∙T   
 
Table D.2 - Pi terms and independent dimensionless found.
Term nº Pi term
1 π w∙l
2 π h∙l
3 π g ∙l ∙ρ
4 π v ∙T ∙ρ ∙h
5 π Q∙l ∙v ∙ρ
6 π c ∙v ∙T
7 π U∙v ∙T ∙ρ
8 π μ ∙ l ∙v ∙ρ
9 π k∙l ∙v ∙T ∙ρ
10 π β ∙T

However, the group calculated was certificated using the dimensions (LTM).
According Newton's Second Law, F m ∙ a, the mass can be written as m ρ ∙ ∀ and
if the volume had dimension of L3, m ∝ ρ ∙ L . The acceleration can be rewritten as
a dv/ds v ∙ dv/ds. So, F ∝ ρ ∙ V ∙ L (Fox and McDonald, 1998).

371 
 
 F m ∙ a ∴ F kg ∙ m ∙ s ∴F M∙L∙T ∴ M F∙L ∙T

Thus we get,
π l ∙ v ∙ T ∙ ∙ w
π L ∙ L T ∙ θ ∙ M L ∙ L
π L ∙ L T ∙ θ ∙ M L ∙ L
π 1

π l ∙ v ∙ T ∙ ∙ h
π L ∙ L T ∙ θ ∙ M L ∙ L
π L ∙ L T ∙ θ ∙ M L ∙ L
π 1

π l ∙ v ∙ T ∙ ∙ g
π L ∙ L T ∙ θ ∙ M L ∙ M L
π L ∙ L T ∙ θ ∙ M L ∙ M L
π 1

π l ∙ v ∙ T ∙ ∙ h
π L ∙ L T ∙ θ ∙ M L ∙ M T θ
π L ∙ L T ∙ θ ∙ M L ∙ M T θ
π 1

π l ∙ v ∙ T ∙ ∙ Q
π L ∙ L T ∙ θ ∙ M L ∙ M L T
π L ∙ L T ∙ θ ∙ M L ∙ M L T
π 1

π l ∙ v ∙ T ∙ ∙ c
π L ∙ L T ∙ θ ∙ M L ∙ L T θ
π L ∙ L T ∙ θ ∙ M L ∙ L T θ
π 1

372 
 
 π l ∙ v ∙ T ∙ ∙ U
π L ∙ L T ∙ θ ∙ M L ∙ M T θ
π L ∙ L T ∙ θ ∙ M L ∙ M T θ
π 1

π l ∙ v ∙ T ∙ ∙
π L ∙ L T ∙ θ ∙ M L ∙ M L T
π L ∙ L T ∙ θ ∙ M L ∙ M L T
π 1

π l ∙ v ∙ T ∙ ∙
π L ∙ L T ∙ θ ∙ M L ∙ M L T θ
π L ∙ L T ∙ θ ∙ M L ∙ M L T θ
π 1

π l ∙ v ∙ T ∙ ∙ β
π L ∙ L T ∙ θ ∙ M L ∙ θ
π L ∙ L T ∙ θ ∙ M L ∙ θ
π 1

D.2 - The dimensionless groups found

D.2.1 - Reynolds number (Re):
π μ ∙ l ∙v ∙ρ 1/R

The Reynolds number (Re) expresses the ratio of inertial forces to viscous forces,
defined by Re = V L ν-1 = V L μ ρ-1.

Where, V: velocity of the fluid, (m s-1); L: dimension, (m); ν: kinematic viscosity, (m2 s-1);
μ: dynamic viscosity, (kg m-1 s-1); and ρ: density of the fluid, (kg m-3).

D.2.2 - Eckert number (Ec):

π c ∙v ∙T 1/E

373 
 
 The Eckert number (Ec) is used to calculate flow. It is expresses the relationship
between a flow`s kinetic energy and enthalpy, thus is used to characterize dissipation. It is

defined as E kinect energy/enthalpy.

∙ ∆

where, Cp: specific heat of the fluid, (J kg-1 K-1); and ∆t: temperature difference of the flow,
(K).

D.2.3 - Prandtl number (Pr):

∙
μ∙c ∙k = Pr

Prandtl number (Pr) is the ratio of momentum diffusivity (kinematic viscosity) to

thermal diffusivity. The Pr is defined as:
ν Viscous diffusion rate Cp ∙ μ
Pr  =   =   = 
α Thermal diffusion rate k

where, α: thermal diffusivity, (m2 s-1); and k: thermal conductivity, (W m-1 K-1).

D.2.4 - Peclet number (Pe):

ρ ∙ c ∙ v ∙ l ∙ k R ∙P

Peclet number (Pe) is the ratio between the rates of heat transfer by conduction and
advection, defined by V ∙ L ∙ α R ∙P .

D.2.5 - Grashof number (Gr):

∙ ∙ ∙ ∙
G ∴ν

The Grashof number is a dimensionless quantity used in analyzing the velocity

distribution in free convection systems. In the systems encountered most commonly in
CBP barns, free convection is the natural tendency of the wind (air movement) to migrate
due to some driving force. In free convection, the driving force is a buoyancy force caused
by a temperature gradient, as the fluid would be at rest in the absence of temperature
variations. The Grashof number is analogous to the Reynolds number in forced convection.

374 
 
D.2.6 - Nusselt number (Nu):
h ∙ l ∙ k Nu

A dimensionless number used in the study of forced convection which gives a measure
of the ratio of the total heat transfer to conductive heat transfer, and is equal to the heat
transfer coefficient times a characteristic length divided by the thermal conductivity.

D.3 - Calculate of air speed and floor surface temperature to the full model
a) Air speed in the full model:
Re Re

∙ ∙ ∙ ∙

Then,

v∙L v∙L

So, scale used in this reduce model was 1:16,

VFull
Vmodel =
16

According Table 4.1, the air speed found in the Barn 24, which was used as a base to
build the reduced model, was 0.7 m s-1. Then,

V 0.7 m s / 16
V 0.04 m s

b) Temperature of the floor in the full model:

Using the Pi terms found in Appendix - Table C.2:

π Q∙l ∙v ∙ρ
375 
 
 Q∙l ∙v ∙ρ Q∙l ∙v ∙ρ

ρ ρ

Then,

Q∙l ∙v Q∙l ∙v

lmodel = 1.35 m (Figure 3.22);

vw model = 0.04 m s-1
lfull = 21.6 m
vw model = 0.7 m s-1 (Table 4.1)

So, using these values in the equation above:

Q ∙ 1.35 ∙ 0.04 Q ∙ 21.6 ∙ 0.7

Q 7.97 ∙ 10 ∙ Q

Calculating the heat flux in the compost area:

i) Average temperatures: floor surface (Ts), compost at 0.2 m deep (Tc), and the
environment (Tair):
Ts = 7.2°C = 280.35 K → Appendix - Table A.4 → Barn 24;
Tair = 6.9°C = 280.05 K → Table 4.1→ Barn 24;

ii) Properties of the mean air temperature (Values in Appendix - Table D.3):
Variable Value Unit (S.I)
ρ 1.2613 kg m-3
ν 1.39 ·10-5 m2 s-1
k 0.024626 W m−1 K−1
β 0.0035708 K-1
L 21.6 m
Pr 0.718

G g∙β∙ T T ∙L ∙ν
G 5.46 ∙ 10

376 
 
 For calculating the Nusselt number (Nu) of equation (D.2.6), is recommended by
McAdams (1954) using the equation (D.2.7) for Gr > 109 where the heat transfer
coefficient is independent of the length (L).
/
N 0.13 ∙ Gr ∙ Pr 951.4 (E.1)
Then, the coefficients of heat transfer is:

k ∙
h 1.08 W m-2 K−1
L

The heat transfer in the full scale CBP barn is:

Qfull = h · A · ∆T
Qfull = 106.8 W

So,
Q 7.97 ∙ 10 ∙ Q

The value of heat flux (q"model) is:

q"model ≈ 1 · 10-1 W m-2

Table D.3 - Thermal properties of the air at atmospheric pressure.

T ρ cp μ · 107 ν · 107 k · 107 α · 107 Pr
(K) (kg m-3) (kJ kg-1 K-1) (kJ kg-1 K-1) (m2 s-1) (W m-1 K-1) (m2 s-1)
200 1.7458 1.007 132.5 7.590 18.1 10.3 0.737
250 1.3947 1.006 159.6 11.44 22.3 15.9 0.720
300 1.1614 1.007 184.6 15.89 26.3 22.5 0.707
350 0.9950 1.009 208.2 20.92 30.0 29.9 0.700
Source: Incropera and DeWitt (2001).

377 
 
 APPENDIX E

(a) (b)

(c) (d)

(e)
Figure E.1 - Types of ridge vents: (a) closed ridge - CLR, (b) open ridge - OPR, (c)
elevated ridge - ELR, (d) overshot ridge - OVR, and (e) open ridge with
chimney - ORC, used in presented study. Dimensions in centimeters (cm).

378 
 
 APPENDIX F

Table F.1 - Analysis of variance to particle weight fractions for all compost material.
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
trat 5 28289.661189 5657.932238 81.537 0.0000
erro 246 17070.190471 69.391018
--------------------------------------------------------------------------------
Total corrigido 251 45359.851660
--------------------------------------------------------------------------------
CV (%) = 49.98
Média geral: 16.6668254 Número de observações: 252
--------------------------------------------------------------------------------
Teste Tukey para a FV trat
--------------------------------------------------------------------------------
DMS: 5.22235316385467 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 42
Erro padrão: 1.285366523386
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
Finer 30.104286 a
PS2 26.051429 a b
PS5 24.710714 b
PS4 10.243095 c
Coarser 4.797619 d
PS3 4.093810 d
--------------------------------------------------------------------------------

Table F.2 - Analysis of variance to thermal conductivity in different particle weight

fractions for all compost materials.
Variável analisada: Thermal conductivity

Opção de transformação: Variável sem transformação ( Y )

--------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
TRAT 4 0.747026 0.186757 47.816 0.0000
erro 205 0.800678 0.003906
--------------------------------------------------------------------------------
Total corrigido 209 1.547704
--------------------------------------------------------------------------------
CV (%) = 31.58
Média geral: 0.1978810 Número de observações: 210
--------------------------------------------------------------------------------
Teste Tukey para a FV TRAT
--------------------------------------------------------------------------------
DMS: 0.0375439678889858 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 42
Erro padrão: 0.00964333747961254
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
PS5 0.310738 a
PS4 0.200286 b
PS3 0.179524 b c
PS2 0.151833 c
Finer 0.147024 c
--------------------------------------------------------------------------------

379 
 
Table F.3 - Analysis of variance to thermal resistivity in different particle weight fractions
for all compost materials.
--------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
TRAT 4 2682694.497429 670673.624357 46.878 0.0000
erro 205 2932874.110238 14306.702977
--------------------------------------------------------------------------------
Total corrigido 209 5615568.607667
--------------------------------------------------------------------------------
CV (%) = 20.25
Média geral: 590.6966667 Número de observações: 210
--------------------------------------------------------------------------------
Teste Tukey para a FV TRAT
--------------------------------------------------------------------------------
DMS: 71.8551564486023 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 42
Erro padrão: 18.4563210082947
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
Finer 722.588095 a
PS2 682.300000 a
PS3 599.785714 b
PS4 548.421429 b
PS5 400.388095 c
--------------------------------------------------------------------------------

Table F.4 - Multiple linear regression to thermal conductivity in function moisture content
and static compaction degree for Green sawdust.

Multiple Linear Regression

Data source: Data 1 in Notebook2

Col 3 = -0.0619 + (0.311 * Col 1) + (0.478 * Col 2)

N = 18

R = 0.916 Rsqr = 0.838 Adj Rsqr = 0.817

Standard Error of Estimate = 0.038

Coefficient Std. Error t P VIF

Constant -0.0619 0.0350 -1.767 0.098
Col 1 0.311 0.0525 5.924 <0.001 1.000
Col 2 0.478 0.0732 6.529 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.113 0.0563 38.863 <0.001
Residual 15 0.0217 0.00145
Total 17 0.134 0.00790

Column SSIncr SSMarg

Col 1 0.0508 0.0508
Col 2 0.0617 0.0617

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001 Col 2 <0.001

380 
 
Table F.5 - Multiple linear regression to thermal conductivity in function moisture content
and static compaction degree for Kiln-dried shavings or sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook2

Col 3 = -0.0679 + (0.342 * Col 1) + (0.487 * Col 2)

N = 18

R = 0.937 Rsqr = 0.877 Adj Rsqr = 0.861

Standard Error of Estimate = 0.034

Coefficient Std. Error t P VIF

Constant -0.0679 0.0315 -2.154 0.048
Col 1 0.342 0.0472 7.233 <0.001 1.000
Col 2 0.487 0.0658 7.404 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.125 0.0627 53.561 <0.001
Residual 15 0.0176 0.00117
Total 17 0.143 0.00841

Column SSIncr SSMarg

Col 1 0.0612 0.0612
Col 2 0.0642 0.0642

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

Table F.6 - Multiple linear regression to thermal conductivity in function moisture content
and static compaction degree for Mix.
Multiple Linear Regression
Data source: Data 1 in Notebook2

Col 3 = -0.0537 + (0.360 * Col 1) + (0.434 * Col 2)

N = 18

R = 0.911 Rsqr = 0.829 Adj Rsqr = 0.806

Standard Error of Estimate = 0.040

Coefficient Std. Error t P VIF

Constant -0.0537 0.0373 -1.441 0.170
Col 1 0.360 0.0558 6.453 <0.001 1.000
Col 2 0.434 0.0779 5.578 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.119 0.0595 36.375 <0.001
Residual 15 0.0246 0.00164
Total 17 0.144 0.00845

Column SSIncr SSMarg

Col 1 0.0681 0.0681
Col 2 0.0509 0.0509

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

381 
 
Table F.7 - Multiple linear regression to thermal conductivity in function moisture content
and dynamic compaction degree for Green sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook5

Col 3 = -0.131 + (0.269 * Col 1) + (0.689 * Col 2)

N = 12

R = 0.962 Rsqr = 0.925 Adj Rsqr = 0.908

Standard Error of Estimate = 0.030

Coefficient Std. Error t P VIF

Constant -0.131 0.0376 -3.481 0.007
Col 1 0.269 0.0762 3.535 0.006 1.000
Col 2 0.689 0.0696 9.907 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.0965 0.0482 55.323 <0.001
Residual 9 0.00785 0.000872
Total 11 0.104 0.00948

Column SSIncr SSMarg

Col 1 0.0109 0.0109
Col 2 0.0856 0.0856

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 0.006
Col 2 <0.001

Table F.8 - Multiple linear regression to thermal conductivity in function moisture content
and dynamic compaction degree for Kiln-dried shavings or sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook5

Col 3 = -0.127 + (0.271 * Col 1) + (0.680 * Col 2)

N = 12

R = 0.978 Rsqr = 0.956 Adj Rsqr = 0.947

Standard Error of Estimate = 0.022

Coefficient Std. Error t P VIF

Constant -0.127 0.0278 -4.556 0.001
Col 1 0.271 0.0564 4.803 <0.001 1.000
Col 2 0.680 0.0515 13.218 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.0943 0.0471 98.891 <0.001
Residual 9 0.00429 0.000477
Total 11 0.0986 0.00896

Column SSIncr SSMarg

Col 1 0.0110 0.0110
Col 2 0.0833 0.0833

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

382 
 
Table F.9 - Multiple linear regression to thermal conductivity in function moisture content
and dynamic compaction degree for Mix.
Multiple Linear Regression
Data source: Data 1 in Notebook5

Col 3 = -0.112 + (0.00193 * Col 1) + (0.623 * Col 2)

N = 12

R = 0.966 Rsqr = 0.934 Adj Rsqr = 0.919

Standard Error of Estimate = 0.025

Coefficient Std. Error t P VIF

Constant -0.112 0.0322 -3.484 0.007
Col 1 0.00193 0.000450 4.294 0.002 1.000
Col 2 0.623 0.0596 10.442 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 0.0816 0.0408 63.737 <0.001
Residual 9 0.00576 0.000640
Total 11 0.0873 0.00794

Column SSIncr SSMarg

Col 1 0.0118 0.0118
Col 2 0.0698 0.0698

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 0.002
Col 2 <0.001

Table F.10 - Multiple linear regression to thermal resistivity in function moisture content
and static compaction degree for Green sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook6

Col 3 = 1229.382 - (825.693 * Col 1) - (1099.412 * Col 2)

N = 18

R = 0.968 Rsqr = 0.938 Adj Rsqr = 0.930

Standard Error of Estimate = 54.938

Coefficient Std. Error t P VIF

Constant 1229.382 50.603 24.295 <0.001
Col 1 -825.693 75.821 -10.890 <0.001 1.000
Col 2 -1099.412 105.728 -10.399 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 684279.297 342139.648 113.361 <0.001
Residual 15 45272.289 3018.153
Total 17 729551.586 42914.799

Column SSIncr SSMarg

Col 1 357928.582 357928.582
Col 2 326350.715 326350.715

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

383 
 
Table F.11 - Multiple linear regression to thermal resistivity in function moisture content
and static compaction degree for Kiln-dried shavings or sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook6

Col 3 = 1282.485 - (972.177 * Col 1) - (1164.851 * Col 2)

N = 18

R = 0.964 Rsqr = 0.929 Adj Rsqr = 0.920

Standard Error of Estimate = 66.245

Coefficient Std. Error t P VIF

Constant 1282.485 61.018 21.018 <0.001
Col 1 -972.177 91.427 -10.633 <0.001 1.000
Col 2 -1164.851 127.489 -9.137 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 862548.929 431274.464 98.276 <0.001
Residual 15 65826.234 4388.416
Total 17 928375.163 54610.304

Column SSIncr SSMarg

Col 1 496191.886 496191.886
Col 2 366357.043 366357.043

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

Table F.12 - Multiple linear regression to thermal resistivity in function moisture content
and static compaction degree for Mix.
Multiple Linear Regression
Data source: Data 1 in Notebook6

Col 3 = 1229.610 - (987.615 * Col 1) - (948.268 * Col 2)

N = 18

R = 0.957 Rsqr = 0.916 Adj Rsqr = 0.905

Standard Error of Estimate = 67.764

Coefficient Std. Error t P VIF

Constant 1229.610 62.417 19.700 <0.001
Col 1 -987.615 93.523 -10.560 <0.001 1.000
Col 2 -948.268 130.411 -7.271 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 754863.347 377431.674 82.195 <0.001
Residual 15 68878.560 4591.904
Total 17 823741.907 48455.406

Column SSIncr SSMarg

Col 1 512075.963 512075.963
Col 2 242787.384 242787.384

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

384 
 
Table F.13 - Multiple linear regression to thermal resistivity in function moisture content
and dynamic compaction degree for Green sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook2

Col 3 = 1187.716 - (497.389 * Col 1) - (12.442 * Col 2)

N = 12

R = 0.993 Rsqr = 0.987 Adj Rsqr = 0.984

Standard Error of Estimate = 21.741

Coefficient Std. Error t P VIF

Constant 1187.716 27.715 42.855 <0.001
Col 1 -497.389 56.135 -8.861 <0.001 1.000
Col 2 -12.442 0.512 -24.279 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 315739.270 157869.635 333.989 <0.001
Residual 9 4254.112 472.679
Total 11 319993.382 29090.307

Column SSIncr SSMarg

Col 1 37109.325 37109.325
Col 2 278629.945 278629.945

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

Table F.14 - Multiple linear regression to thermal resistivity in function moisture content
and dynamic compaction degree for Kiln-dried shavings or sawdust.
Multiple Linear Regression
Data source: Data 1 in Notebook5

Col 3 = 1211.623 - (532.158 * Col 1) - (12.924 * Col 2)

N = 12
R = 0.983 Rsqr = 0.966 Adj Rsqr = 0.959

Standard Error of Estimate = 36.551

Coefficient Std. Error t P VIF

Constant 1211.623 46.593 26.004 <0.001
Col 1 -532.158 94.373 -5.639 <0.001 1.000
Col 2 -1292.434 86.151 -15.002 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 343148.017 171574.009 128.428 <0.001
Residual 9 12023.558 1335.951
Total 11 355171.575 32288.325

Column SSIncr SSMarg

Col 1 42478.822 42478.822
Col 2 300669.195 300669.195

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

385 
 
Table F.15 - Multiple linear regression to thermal resistivity in function moisture content
and dynamic compaction degree for Mix.
Multiple Linear Regression
Data source: Data 1 in Notebook2

Col 3 = 1205.911 - (566.171 * Col 1) - (11.412 * Col 2)

N = 12 Missing Observations = 12

R = 0.992 Rsqr = 0.983 Adj Rsqr = 0.979

Standard Error of Estimate = 23.204

Coefficient Std. Error t P VIF

Constant 1205.911 29.580 40.768 <0.001
Col 1 -566.171 59.913 -9.450 <0.001 1.000
Col 2 -11.412 0.547 -20.866 <0.001 1.000

Analysis of Variance:
DF SS MS F P
Regression 2 282514.235 141257.117 262.347 <0.001
Residual 9 4845.929 538.437
Total 11 287360.164 26123.651

Column SSIncr SSMarg

Col 1 48082.414 48082.414
Col 2 234431.821 234431.821

The dependent variable Col 3 can be predicted from a linear combination of the independent variables:
P
Col 1 <0.001
Col 2 <0.001

386 
 
Table F.16 - Analysis of variance to thermal conductivity (k) in different moisture content
(MC) and static compaction degree (SC) for all compost materials.
Variável analisada: k

Opção de transformação: Variável sem transformação ( Y )

--------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
MC 2 15.878701 7.939351 1597.882 0.0000
SC 5 15.037659 3.007532 605.299 0.0000
erro 4528 22.498140 0.004969
--------------------------------------------------------------------------------
Total corrigido 4535 53.414501
--------------------------------------------------------------------------------
CV (%) = 35.10
Média geral: 0.2008428 Número de observações: 4536
--------------------------------------------------------------------------------
Teste Tukey para a FV MC
--------------------------------------------------------------------------------
DMS: 0.00601234749076705 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 1512
Erro padrão: 0.00181277626663453
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
30 0.135912 a1
45 0.187604 a2
60 0.279013 a3
--------------------------------------------------------------------------------
Teste Tukey para a FV SC
--------------------------------------------------------------------------------
DMS: 0.0103385638674625 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 756
Erro padrão: 0.00256365278182262
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
Fluffy 0.122601 a1
0 0.132481 a1
14.5 0.192725 a2
29 0.229508 a3
43.5 0.252398 a4
58 0.275344 a5
--------------------------------------------------------------------------------

387 
 
Table F.17 - Analysis of variance to thermal resistivity (rho) in different moisture content
(MC) and static compaction degree (SC) for all compost materials.
Variável analisada: RHO
Opção de transformação: Variável sem transformação ( Y )
-------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
MC 2 84768364.414369 42384182.207185 2399.893 0.0000
SC 5 119427953.545884 23885590.709177 1352.459 0.0000
erro 4528 79968392.717152 17660.864116
--------------------------------------------------------------------------------
Total corrigido 4535 284164710.677405
--------------------------------------------------------------------------------
CV (%) = 21.54
Média geral: 616.8897928 Número de observações: 4536
--------------------------------------------------------------------------------
Teste Tukey para a FV MC
--------------------------------------------------------------------------------
DMS: 11.335219789401 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 1512
Erro padrão: 3.41766962785626
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
60 451.055423 a1
45 613.748347 a2
30 785.865608 a3
--------------------------------------------------------------------------------
Teste Tukey para a FV SC
--------------------------------------------------------------------------------
DMS: 19.4915370284921 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 756
Erro padrão: 4.83331473942492
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
58 443.349735 a1
43.5 476.126455 a2
29 514.993519 a3
14.5 596.467460 a4
0 804.187037 a5
F 866.214550 a6
--------------------------------------------------------------------------------

388 
 
Table F.18 - Analysis of variance to thermal conductivity (k) in different moisture content
(MC) and dynamic compaction degree (DC) for all compost materials.
Variável analisada: k
Opção de transformação: Variável sem transformação ( Y )
--------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
MC 2 20.748868 10.374434 1671.811 0.0000
DC 3 2.790090 0.930030 149.872 0.0000
erro 3018 18.728215 0.006206
--------------------------------------------------------------------------------
Total corrigido 3023 42.267173
--------------------------------------------------------------------------------
CV (%) = 32.57
Média geral: 0.2418687 Número de observações: 3024
--------------------------------------------------------------------------------
Teste Tukey para a FV MC
--------------------------------------------------------------------------------
DMS: 0.00823049820589862 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 1008
Erro padrão: 0.00248118018777819
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
30 0.150047 a1
45 0.224783 a2
60 0.350777 a3
--------------------------------------------------------------------------------
Teste Tukey para a FV DC
--------------------------------------------------------------------------------
DMS: 0.0104170309506276 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 756
Erro padrão: 0.00286502009864341
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
15 0.198185 a1
29 0.231403 a2
44 0.258775 a3
58 0.279111 a4
--------------------------------------------------------------------------------

389 
 
Table F.19 - Analysis of variance to thermal resistivity (rho) in different moisture content
(MC) and dynamic compaction degree (DC) for all compost materials.
Variável analisada: RHO
Opção de transformação: Variável sem transformação ( Y )
-------------------------------------------------------------------------------
TABELA DE ANÁLISE DE VARIÂNCIA
--------------------------------------------------------------------------------
FV GL SQ QM Fc Pr>Fc
--------------------------------------------------------------------------------
MC 2 74382330.284365 37191165.142183 2456.087 0.0000
DC 3 8338082.144299 2779360.714766 183.548 0.0000
erro 3018 45699896.688360 15142.444231
--------------------------------------------------------------------------------
Total corrigido 3023 128420309.117024
--------------------------------------------------------------------------------
CV (%) = 24.50
Média geral: 502.2878968 Número de observações: 3024
--------------------------------------------------------------------------------
Teste Tukey para a FV MC
--------------------------------------------------------------------------------
DMS: 12.8568786636429 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 1008
Erro padrão: 3.87585682164862
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
60 320.108135 a1
45 483.815476 a2
30 702.940079 a3
--------------------------------------------------------------------------------
Teste Tukey para a FV DC
--------------------------------------------------------------------------------
DMS: 16.2724660910133 NMS: 0.05
--------------------------------------------------------------------------------
Média harmonica do número de repetições (r): 756
Erro padrão: 4.47545395863856
--------------------------------------------------------------------------------
Tratamentos Médias Resultados do teste
--------------------------------------------------------------------------------
58 443.390212 a1
44 470.972354 a2
29 511.965741 a3
15 582.823280 a4
--------------------------------------------------------------------------------
 

390 
 
Table F.20 - Analysis of variance to heated floor area (Area) in different ridge design
(Ridge) and wind direction (Direction) for all CFD models tested.

Two Way Analysis of Variance

Data source: Data 1 in Notebook1

General Linear Model (No Interactions)

Dependent Variable: Col 3

Normality Test: Passed (P = 0.373)

Equal Variance Test: Passed (P = 1.000)

Source of Variation DF SS MS F P
Col 1 4 1620.038 405.010 0.531 0.716
Col 2 3 526.051 175.350 0.230 0.874
Residual 12 9155.593 762.966
Total 19 11301.682 594.825

Power of performed test with alpha = 0.0500: for Col 1 : 0.0500

Power of performed test with alpha = 0.0500: for Col 2 : 0.0500

--------------------------------------------------------------------------------
Teste Tukey para a FV RIDGE
--------------------------------------------------------------------------------

DMS: 0 NMS: 0.05

--------------------------------------------------------------------------------

Média harmonica do número de repetições (r): 4

Erro padrão: 0
--------------------------------------------------------------------------------

Tratamentos Médias Resultados do teste

--------------------------------------------------------------------------------
ELR 19.457500 a1
OVR 23.497500 a2
CLR 24.462500 a3
OPR 27.512500 a4
ORC 45.292500 a5
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
Teste Tukey para a FV DIRECTION
--------------------------------------------------------------------------------

DMS: 0 NMS: 0.05

--------------------------------------------------------------------------------

Média harmonica do número de repetições (r): 5

Erro padrão: 0
--------------------------------------------------------------------------------

Tratamentos Médias Resultados do teste

--------------------------------------------------------------------------------
S to N 19.712000 a1
N to S 27.922000 a2
E to W 32.178000 a3
W to E 32.366000 a4
--------------------------------------------------------------------------------

391