H . Michael Mogil, John C.

Monro,
NWS's Flash Flood warning and
1 2

and Herbert S. G r o p e r 1

National Weather Service, NOAA

Disaster preparedness Programs Silver Spring, Md. 20910

Abstract
Flash floods have become the nation's number one stormy
weather killer in the 1970s. Deaths from flash floods are now
approaching 200 each year compared to an average of less
than 70 per year during the preceding 30-year period. Losses
from flash floods are now nearly 10 times what they were in
the 1940s.
To combat these escalating losses the National Weather
Service (NWS) is expanding its efforts to improve forecasting
and warning of flash floods and to improve the public's re-
sponse to flash flood threats. This paper summarizes the
present and planned NWS operational flash flood warning
and disaster preparedness programs.
FIG. 1. Average annual flash flood deaths by decades. Figures
are 90% of all flood-related deaths reported by the National
1. Introduction Climatic Center, NOAA.
Flash floods are a major natural disaster warning prob- damage from floods and flash floods now averages more
lem facing the nation. During the 1970s the average than $1.0 billion annually compared to about $125 mil-
annual death toll from flash floods has tripled from the lion during the 1940s and $375 million during the 1960s. 3

rate during the 1940s (Fig. 1) and has now become the Figures 2 and 3 show damage from some recent flash
nation's number one stormy weather killer. Property floods (also, see cover of May 1978 B U L L E T I N ) .
1 Disaster Preparedness Staff. s Based on reported losses; we have not made adjustments
2 Office of Hydrology. for inflation.

FIG. 2. Damage resulting to the John Enos home on Murphy Creek, mi from the
Applegate River, near Med ford, Oreg., after a flash flood in January 1974.
690 Vol. 59, No. 6, June 1978
Unauthenticated | Downloaded 04/08/24 02:53 PM UTC
Bulletin American Meteorological Society 691

There are many reasons for this escalating trend in the NWS flood forecasting procedures used on larger
deaths and property losses. For example, an ever in- streams generally cannot respond fast enough. Also,
creasing number of vacationers are visiting flash flood there is the difficult problem in predicting the amount,
prone recreational areas in the summer—the time of timing, and location of flash flood producing rainfall.
highest vulnerability to flash floods. People have also Therefore, a special warning and preparedness system
been establishing themselves at an increasing rate in is needed to ensure the public safety.
flood plains across the nation. As this development oc- Although most flash floods are the result of intense
curs, normal drainage is disrupted. In addition, mobile localized thunderstorm activity or slowly moving/
home and transient trailer parks are often located near stationary thunderstorms or lines of thunderstorms
rivers. People continue to believe "it can't happen here." (NOAA, 1972, 1976a, 1977b, 1977c; Randerson, 1976),
With more than 15 000 communities and recreational some occur in conjunction with tropical cyclones (Groper
areas subject to flash floods, the potential for disaster and Dunlap, 1972; NOAA, 1973) and with extratropical
grows each year. The AMS has recently issued a state- cyclones (NOAA, 1977a).
ment of concern about flash floods (AMS, 1978). Since 1971, more than 1000 significant flash floods
The National Weather Service (NWS) has defined a have occurred nationwide, with every state having ex-
flash flood as a flood that follows the causative event perienced at least one such flood. Since 1945, more than
(excessive rain, dam or levee failure, etc.) within a few 3000 counties have received Red Cross assistance follow-
hours. Because of the short warning times involved and ing floods or flash floods (Fig. 4). Although many of these
the limited number of real-time precipitation reports, did not make national headlines, they have had a

FIG. 3. A boulder-strewn scene of destruction along Myers Creek in southern California follow-
ing rains from Tropical Storm Kathleen, in September 1976. Note road washout in background
and boulders on bridge in lower left corner of photo. (Photo courtesy of Union-Tribune Pub-
lishing Co., San Diego.)

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

692 Vol. 59, No. 6, June 1978

FIG. 4. Flood and flash flood events by counties for the period 1945-76.
Figures are based on Red Cross assistance.

devastating effect on scores of communities. But it is buildings and/or judicious use of physical controls such
the catastrophic flash floods, such as Black Hills in 1972 as reservoirs, levees, and channel improvement may be
(237 deaths) and the Big Thompson Canyon in 1976 appropriate. But these must pass the tests of economic
(139 deaths), that highlight the vulnerability of the feasibility and environmental compatability. The highly
American people to these events (Fig. 5). localized nature of flash floods and the thousands of
To reduce loss of life and lessen damage to property vulnerable locations require the NWS, with other fed-
the NWS has developed flash flood warning and disaster eral agencies and state and local groups, to develop
preparedness programs. Their responsibility for issuing effective and fully coordinated programs.
weather-related warnings goes back to the establishment This paper will give an overview of the NWS's flash
of the Signal Corps in 1870 and the Organic Act of 1890. flood warning and disaster preparedness programs and
Although interest in forecasting severe thunderstorms how the NWS works with other groups in preparing for
and tornadoes has had a long history (Ludlum, 1970), flash floods. We will also include some proposed changes,
interest in flash flood and excessive rainfall forecasting as well as ongoing and planned NWS changes, to the
is more recent. In fact, a national flash flood warning warning and preparedness programs.
program was not established until 1971, and there have
been only limited studies into the hydrological and
meteorological aspects of flash floods. 2. Flash flood watch/warning program
Although the NWS was active in preparedness in the
late 1950s and the 1960s, a disaster preparedness program The basic forecast program begins at the National Mete-
was not formally authorized by Congress until 1971, and orological Center (NMC) in Camp Springs, Md. Here
the primary emphasis was on hurricanes. The program computers process meteorological observations and pro-
has since been expanded to include severe local storms, duce maps of observed and forecast weather systems
winter storms, floods, and flash floods. During its first (Brown and Fawcett, 1972; Fawcett, 1977). The Quanti-
7 years of existence, this program has been instrumental tative Precipitation Branch (QPB) of NMC routinely
in making the public aware of weather hazards. A major issues rainfall forecasts 6 times per day. These indicate
public safety and educational program has been de- areas in which rainfall amounts are expected to equal
veloped to improve response to warnings. or exceed selected forecast values (NOAA, 1976b). This
From the national viewpoint, resolving the flash flood information is transmitted on teletypewriter and fac-
problem cannot fall exclusively on the NWS. Preventive simile for use by River Forecast Centers (RFCs) in their
measures by local and state governments and individual river flood prediction program and by Weather Service
citizens are also required if the damage and loss of life Forecast Offices (WSFOs) for use in operational forecast
resulting from flash floods are to be effectively reduced. programs.
For example, effective flood plain management seeks to This guidance, based largely on synoptic-scale analyses
minimize the occupation of potentially hazardous flood and forecasts, cannot be used to specify localized exces-
plains. In some instances, flood proofing of essential sive convective rainfall amounts. It has, however, been

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

Bulletin American Meteorological Society 693

FIG. 5. Some recent significant flash flood events. Since the map was prepared, flash floods
have struct southern California twice (February and March 1978) killing nearly 40 people and
causing more than $100 million in damages.

very useful in specifying synoptic-scale rainfall patterns based on a diagnosis of synoptic-scale data (Mogil and
and amounts. Groper, 1976; Maddox et al., 1977; Maddox and Chap-
At the present time, the NMC QPB is restructuring its pell, 1978; Hughes and Longsdorf, 1978). There are also
program to give additional support to the flash flood guidelines for predicting extratropical storm rainfall
program. Fewer fixed-issuance time, fixed-forecast period (Smith and Younkin, 1972). In addition, some recent
products will be issued; instead, the QPB will issue studies (NOAA, 1976a; Johnson et al., 1978) suggest that
"outlooks" such as those issued by the National Severe a comparison between observed and climatological mois-
Storms Forecast Center (Kansas City, Mo.) and will ture values (Lott, 1976) can be a useful indicator of
monitor weather situations more closely and coordinate convective rainfall potential. Macroscale numerical
with field offices more frequently (NWS, 1978). model guidance can be helpful, too, especially when the
Currently, the flash flood warning program consists of heavy rainfall is tied to large-scale weather systems
the following elements: (Estelle, 1974; Hovermale et al., 1975). However, it re-
mains extremely difficult to accurately forecast rainfall
1) flash flood watches and warnings; amounts from decaying tropical storms (NOAA, 1972)
2) local flash flood warning programs; and from thunderstorms (NOAA, 1976a; Olson, 1977).
3) stream stage forecasts; Local NWS offices issue warnings only when flash
4) flash flood alarm systems (FFAS); flooding is observed and reported or when flash flood
5) combinations of the above. producing rainfall is indicated by radar, automated rain
The RFCs prepare stream stage forecasts (Colton and gages, or by rainfall observers. The NWS also solicits
Burnash, 1978) and hydrologic flash flood guidance for reports from CB groups and ham radio operators
use by WSFOs. The flash flood guidance is based in part (Weigel, 1977). Flash flood warnings are usually issued
on drainage basin configuration and past rainfall. The for periods of <4 h. They may be valid for a single
RFCs also develop local flash flood forecast procedures drainage basin, although they are more commonly
and recommend applicable flash flood warning systems issued for several counties.
to local communities. The WSFOs and WSOs also issue statements about
The WSFOs are responsible for issuing flash flood thunderstorm rainfall that may cause drainage and/or
watches. Flash flood watches are usually valid for pe- small stream flooding. Radar can be especially useful
riods of 12 h or less and may affect all or part of a because it gives an areal perspective and can be used
WSFO forecast area. Coordination among WSFOs is to quickly estimate accumulated rainfall. For example,
necessary to ensure watch continuity across state borders. the following can be done:
To be most effective, watches should be issued prior
to the onset of heavy rainfall. There are some general 1) Subjective radar scope evaluation. Nearly stationary
guidelines for predicting thunderstorm rainfall potential lines of thunderstorms, or slowly moving thunder-

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

694 Vol. 59, No. 6, June 1978
storms that show redevelopment on the upstream techniques for estimating rainfall amounts from satellite
side of the storm, provide important clues to heavy imagery. Provisional procedures have been published by
rainfall events. Examination of echo tops and in- NESS for this (Scofield and Oliver, 1977) based on a
tensities is important, too. However, heavy rain study of convective rainfall events over the central
can fall from echoes that have relatively low tops. United States for one summer season, and several NWS
2) Manually digitized radar (MDR). The MDR offices are now evaluating these procedures. Results of
values are based on a relationship between theoreti- one of these initial studies indicated that use of the
cal rainfall rates and echo intensity (Moore et al., NESS procedures in conjunction with radar data
1974) and can be used to keep track of accumu- yielded good estimates of heavy convective rainfall
lated rainfall. Although there appears to be a (Johnson et al., 1978).
good relationship between total storm rainfall and Following the disastrous Johnstown flash flood, the
MDR totals for long-duration storms (NWS, 1977), NESS and NHEML techniques were used to estimate the
MDR is not an infallible flash flood alerting tool convective rainfall. Both yielded excellent estimates
under certain meteorological conditions. In addi- (Scofield, 1978; Woodley et al., 1978). Real-time evalua-
tion, other factors such as terrain and previous tion of the NESS techniques is planned for this year.
rainfall have to be considered. In several recent
flash floods (e.g., Kansas City), flash flooding was 3. Local flash flood warning system
already in progress before critical MDR values were
reached (NOAA, 1977c). The NWS watch/warning program is as accurate as the
3) Radar Digitizer and Processor (RADAP) data. This available data and the meteorological/hydrological state
system automatically computes numerous radar fields of the art permits. However, a locally operated com-
and displays them digitally. One of these fields is munity warning system is, at present, the most effective
accumulated rainfall for 6 km by 10 km areas (Saffle, means of preventing loss of life and reducing property
1976). The RADAP uses a convective rainfall rela- damage from flash flooding. This is because it is respon-
tionship of Z = 55R (where R is the rainfall esti-
16 sive to NWS warnings and can also activate itself when
mate and Z is the reflectivity), which more than required by localized weather conditions that NWS may
doubles the rainfall estimates obtained from the not be aware of. The design of such systems will depend
Marshall-Palmer relationship (Greene and Clark, upon site specific problems (e.g., geography, population
1974). This experimental equipment is only in use in the flood plain, flood control structures, and warning
at five network radar locations at the present time. time) and resources available for program implementa-
Very limited evaluation of total storm rainfall esti- tion and maintenance. The systems can range from the
mates obtained by evaluating the maximum (peak) simple to the sophisticated. Some equipment and com-
rainfall data collected by RADAP shows excellent munication options of the local flash flood warning sys-
agreement with observed rainfall totals (Greene and tem are given below:
Saffle, 1978). Equipment:
plastic rain gage
All of the above require the use of a Z-R relationship. staff river gage
However, because of the many assumptions used in de- recording river gage
riving such a relationship (Greene, 1971), there has been tipping bucket rain gage
electronic river level sensors
considerable variation among researchers as to what the automated rainfall reporting stations
relationship should be (Battan, 1973). In some parts of Communication System—from observer to flash flood
the United States (for example, in the High Plains from coordinator:
eastern Colorado to the Dakotas) the presence of hail telephone
and/or the evaporation of precipitation as it falls ham radio
CB radio
through low-level drier air can make it difficult to rely police, fire, and other emergency radio
on any Z-R relationship. Also, researchers at NOAA's Communication System—from automated equipment to
Environmental Research Laboratories in Boulder (Mad- flash flood coordinator:
dox et al., 1977) and others have suggested that for a telephone
variety of factors rainfall estimates may be significantly hardwire
line of sight radio
underestimated at large ranges. Some attention has been satellite relay radio
placed recently (e.g., Brandes, 1974; Woodley et al., The essential elements of such a local program are
1974; Cain and Smith, 1976) on techniques for calibrat-
ing radars using rainfall observations. However, their the following:
procedures are not being used operationally. 1) volunteer rainfall and stream gage observers;
Satellite data may also be used for estimating rainfall 2) a reliable and rapid local communication system
amounts. Both the National Environmental Satellite with emergency backup;
Service (NESS) (Oliver and Scofield, 1976) and the Na- 3) a flash flood warning coordinator and alternate;
tional Hurricane and Experimental Meteorology Labora- 4) forecast procedures developed by NWS hydrolo-
tory (NHEML) (Griffith et al1976) have developed gists;

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

Bulletin American Meteorological Society 695

FIG. 6. Flash flood alarm system locations. Some communities not shown in the figure
have also installed alarm systems.

5) a warning dissemination plan; a day staffing. It receives a continuous signal from the
6) an adequate preparedness plan (including public river station that provides information on the system's
education). operational status. When the critical level is reached at
the river station, an audible and visible alarm is acti-
Normally, inexpensive rainfall gages and staff gages for vated. During the major flooding and flash flooding that
river data will be sufficient. However, in some areas occurred in eastern Kentucky, western Virginia, and
where time is critical and observers are not available, West Virginia on 2-5 April 1977, four FFASs pro-
automated precipitation gages and electronic stream vided invaluable notification of critical river levels to
level sensors are essential. Event-reporting instrumenta- officials in the communities (NOAA, 1977a).
tion is useful, too (Burnash and Twedt, 1978). The FFAS is only one of several devices now available
The NWS assists communities in establishing their for automated river level sensing. Automated precipita-
flash flood warning program. Thus, NWS will survey the tion sensors have also been developed. This type of
area, recommend appropriate equipment and network equipment can be a valuable complement to existing
design, and provide some of the necessary equipment volunteer observer networks, especially in remote water-
(e.g., staff gages and plastic rain gages). The NWS has shed areas. Costs of such automated equipment range
also purchased and installed 67 electronic river level from about $200 to over $3000 per site.
sensors (FFASs) as part of a demonstration program A critical element in local programs is the stream level
(Fig. 6).
The FFAS has three components: a river station, an forecast procedure. NWS hydrologists prepare simplified
intermediate station, and an alarm station. The river procedures for predicting stream levels, given an index
station senses the critical water level for flash flooding. of antecedent soil moisture conditions and reported
The intermediate station provides power to the river rainfall amounts (Monro and Anderson, 1974). Hydro-
station and couples the river station's signal output to logic staffs at RFCs routinely furnish the index to com-
the alarm station. The alarm station is located in an munities that have local flash flood warning programs.
appropriate place in the community and has 7-day, 24 h The forecast procedure usually includes a set of tables

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

696 Vol. 59, No. 6, June 1978
and/or graphs. These are used by the community's flash cations Commission, the Defense Civil Preparedness
flood warning coordinator to predict potential flooding. Agency (DCPA), and the National Industrial Advisory
The success of the local warning program depends on Committee (which represents the broadcast industry)
a rapid and reliable local communications system. A began a joint effort to develop state and local disaster
redundant configuration is recommended. warning dissemination procedures for the Emergency
Broadcast System (EBS). Once operational area plans are
4. Dissemination implemented, NWS offices will be able to relay short-
The NWS relies upon many systems to disseminate fuse warnings rapidly to many broadcast stations by
watches, warnings, and statements. These include com- simply calling one station.
mercial radio and television, cable television, NOAA The NWS also uses several civil defense and law en-
Weather Wire Service (teletypewriter), NOAA Weather forcement communications systems in its warning pro-
Radio, hotline telephones, and the national press wire gram. The most important of these is the National
services. Warning System (NAWAS), a multipoint inter- and
The NOAA Weather Radio is an extremely important intrastate telephone hotline system funded by the DCPA.
vehicle for disseminating critical weather information. This system connects state and area warning points,
This is an NWS-operated VHF-FM system with three local municipalities, and many NWS offices. There are
special frequencies (162.40, 162.475, and 162.55 MHz), also a few statewide microwave emergency communica-
having an effective range of ~65 km. In March 1978 tions systems in operation (e.g., KEWS—Kentucky
there were approximately 170 stations in operation; by Emergency Warnings System). In addition, amateur
the end of 1979 the NWS expects to have more than 300 radio and CB groups can be used to disseminate warn-
stations in operation that will be within listening range ing information as well as to gather storm reports.
of 90% of the U.S. population. An improved NWS communications system called
The NOAA Weather Wire Service is a teletypewriter AFOS (Automation of Field Operations and Services)
system devoted exclusively to weather information. Com- is scheduled for nationwide implementation beginning
mercial radio and television, police, some local civil this year (NOAA, 1976c). This system will provide NWS
defense offices, and others subscribe to this service. offices with high-speed data handling and display capa-
Commercial radio and television and cable television bilities by means of on-site minicomputers linked to-
represent other important vehicles for warning dis- gether in a nationwide network (Johnson and Giraytys,
semination. They receive NWS watches and warnings 1975). Weather information will be displayed on TV-like
from the NOAA Weather Wire Service, NOAA Weather screens instead of on paper, and warning messages will
Radio, national press wire services, or by telephone and likely be received by adjacent weather offices in seconds
are urged to rebroadcast the information immediately. rather than minutes. The AFOS system will have fore-
About two years ago, the NWS, the Federal Communi- cast applications as well (Klein, 1976).

FIG. 7. The "warning system" showing the complex interactions among agencies,
dissemination subsystems, and the public.

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

Bulletin American Meteorological Society 697

5. Disaster preparedness
For any community to respond positively to a flash flood,
the community officials and the individuals in the com-
munity must be prepared. Whereas most federal agen-
cies are involved in postdisaster relief and recovery pro-
grams, the NWS's program is committed to predisaster
planning and preparedness. For example, the NWS has
issued a "Guide for Flood and Flash Flood Preparedness
Planning" (Owen, 1977) and has developed and dis-
tributed literature about flash floods and flash flood
safety. Recently, the NWS introduced, in its prepared-
ness literature, the weather conscious logo—Owlie Sky-
warn. Owlie tells children in elementary schools safety
rules for all natural disasters.
At the present time, 18 WSFOs have been assigned a
dedicated disaster preparedness meteorologist. In addi-
tion, meteorologists at WSFOs and WSOs and hydrolo-
gists at WSFOs and RFCs support the flash flood warn-
ing and disaster preparedness programs. These people
and others are involved in the following activities: FIG. 8. The major variables affecting warning—effectiveness,
variation in, and/or existence. All of the variables on the
1) hold preparedness meetings with state, county, and left act to increase the potential for disastrous flash floods.
local officials, law enforcement agencies, school offi- All of those on the right act to mitigate that potential, but
cials, amateur radio and CB groups, and others to they have not been adequately addressed.
establish and maintain local warning communica- others have yet to be initiated because of personnel
tion systems and storm spotter and volunteer ob- and/or fiscal limitations. The following are some of the
server networks and to implement local flash flood initiatives the NWS and other agencies can take to
warning programs; improve warning dissemination and increase public
2) work with the mass media to ensure that flash flood awareness/response:
releases and safety information are rapidly and
reliably disseminated. 1) expand the local flash flood warning program for
3) encourage appropriate response by local officials all communities subject to flash flooding;
and the public by distribution of weather safety 2) install additional NAWAS drops at NWS offices,
literature (pamphlets, slides, films, and news re- RFCs, and at selected emergency action agencies;
leases). 3) complete the manning of the NWS flash flood and
disaster preparedness programs nationwide;
The NWS has also begun to examine the sociological 4) place more emphasis on predisaster efforts (e.g.,
aspects of its warning program (McLuckie, 1973, 1974). there should be funding to enable smaller com-
Particularly important is the complex linkage among munities with limited resources to plan for natural
warning, dissemination, and response (Fig. 7). Recent disasters by purchasing needed FFASs, rain gages,
disasters continue to show that people do not know how and communications equipment, and to support
to respond adequately to a flash flood threat (Gruntfest, plan maintenance);
1977; NOAA, 1977c). 5) encourage the private sector to develop inexpensive
The National Science Foundation (NSF) has recently automated precipitation and river alarms;
completed a research assessment of natural hazards 6) increase public understanding of and response to
(White and Haas, 1975) and the natural hazard warning hazardous weather (this could include a major flash
systems (Mileti, 1975) in the United States. Mileti's flood public awareness program in conjunction with
graphic illustration of inputs to warning clearly shows other federal, state, and private agencies involved
the difficulties in providing adequate warnings to in flash floods or recreational activities; the news
people in flash flood prone areas (Fig. 8). The University media, including outdoor recreation publications,
of Minnesota (Carter et al., 1977) is engaged in a 3-year should be involved heavily in this program);
NSF-funded study to assess community and organiza- 7) provide additional NOAA Weather Radio trans-
tional response to natural hazard warnings. mission capability, especially for highly used recre-
ational areas.
6. Improvements to the warning and
preparedness programs In the area of hydrometeorological forecasting and re-
There are numerous actions that could be taken to im- search, the following could be done:
prove the flash flood warning and disaster preparedness 1) ensure that all management and field personnel
programs. Some of these are already underway, but really understand flash flooding and that it can

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

698 Vol. 59, No. 6, June 1978
occur in proximity to tornadoes and other severe instrumentation for real-time flash flood warnings. Pre-
weather; prints, Conference on Flash Floods: Hydrometeorological
2) encourage more local hydrometeorological studies Aspects (Los Angeles), AMS, Boston, pp. 96-101.
into the flash flood problem; Cain, D. E., and P. L. Smith, Jr., 1976: Operational adjust-
3) develop improved statistical guidance forecasts ment of radar estimated rainfall with rain gage data: A
for heavy rainfall; statistical evaluation. Preprints, 17th Conference on Radar
Meteorology (Seattle), AMS, Boston, pp. 533-538.
4) increase understanding of convective cloud-scale Carter, T. M., J. P. Clark, R. K. Leik, and G. A. Fine, 1977:
processes; Social factors affecting the dissemination of and response
5) continue development of techniques for estimating to warnings. Postprints, 11th Technical Conference on
rainfall based on satellite data; Hurricanes and Tropical Meteorology (Miami Beach), AMS,
6) continue efforts to calibrate radar by use of rain- Boston, pp. 302-309.
fall observations and initiate development of tech- Col ton, D. E., and R. J. C. Burnash, 1978: A flash flood warn-
niques for estimating rainfall from radar param- ing system. Preprints, Conference on Flash Floods: Hydro-
eters other than intensity (e.g., echo tops, verti- meteorological Aspects (Los Angeles), AMS, Boston, pp.
cally integrated liquid water content, Doppler 102-104.
signatures); Estelle, E. W., 1974: Characteristics of the models in forecast-
ing precipitation. NMC Tech. Attach. 74-4, Washington,
7) standardize the approach to small basin hydrologic D.C., 4 pp.
forecasting; Fawcett, E. B., 1977: Current capabilities in prediction at
8) develop a flash flood watch/warning verification the National Weather Service's National Meteorological
program that includes an evaluation of NMC Center. Bull. Amer. Meteor. Soc., 58, 143-149.
QPB guidance forecasts for significant rainfall Greene, D. R., 1971: Numerical techniques for the analysis
events; of digital radar data with applications to meteorology and
9) encourage hydrometeorology cross training at field hydrology. Ph.D. thesis, Texas A&M Univ., College Station,
offices; 125 pp.
10) establish a flash flood forecasting course. , and R. A. Clark, 1974: The operational use of digital
radar data for flash flood monitoring. Proceedings of Flash
In addition to these, we support the recommendations Flood Symposium, International Association of Hydrologi-
contained in the AMS Statement of Concern about flash cal Sciences, Paris, pp. 100-105.
floods (AMS, 1978). , and R. E. Saffle, 1978: Radar analysis of the 1977 Johns-
town flash flood. Preprints, Conference on Flash Floods:
Hydrometeorological Aspects (Los Angeles), AMS, Boston,
7. Summary pp. 176-180.
In this paper we have attempted to present a general Griffith, C. G., W. F. Woodley, D. W. Martin et al, 1976:
overview of the NWS's flash flood and disaster prepared- Rainfall estimation from geosynchronous satellite imagery
ness programs and to give some insight into needed pro- during daylight hours. NOAA Tech. Rept., ERL WMPO-7,
Miami, 106 pp.
gram improvements. Some of the recommendations have Groper, H. S., and D. V. Dunlap, 1972: Meteorological and
already been addressed; others will require longer-range hydrological analysis of the August 27-28, 1971, New
developmental efforts and/or budgetary support. Jersey flood. Water Resour. Bull., 8, 1044-1054.
Acknowledgments. We would like to thank Karl Johannes- Gruntfest, E. C., 1977: What people did during the Big
sen (Office of Meteorology and Oceanography), Robert Clark Thompson flood. Natural Hazard Research Working Paper
(Office of Hydrology), and Herbert Lieb (Disaster Prepared- 32, Institute of Behavioral Science, Univ. of Colorado,
ness Staff) for supporting us during the writing of this paper. Boulder, 64 pp.
We would also like to thank Lisa Grandier (Office of Hovermale, J., D. Marks, R. Chu, S. Scolnik, and R. Jones,
Meteorology and Oceanography) and Barbara Pakenham and 1975: Experimental forecasts of hurricane movement with
Desha DeVor (Disaster Preparedness Staff) for their assistance a limited-area fine-mesh grid. Paper presented at Ninth
in preparing this paper. We also want to thank those who Technical Conference on Hurricanes and Tropical Mete-
reviewed the draft manuscript for us. orology, AMS, Key Biscayne, Fla., 27-30 May.
Hughes, L. A., and L. L. Longsdorf, 1978: Guidelines for
References flash flood and small tributary flood prediction. NOAA
AMS, 1978: Statement of Concern: Flash floods—A national Tech. Memo., NWS CR-58 (revised), Kansas City, Mo.,
problem. Bull. Amer. Meteor. Soc., 59, 585-586. 10 pp.
Battan, L. J., 1973: Use of radar for precipitation measure- Johnson, G. A., J. D. Belville, and J. D. Ward, 1978: A flash
ments. Radar Observation of the Atmosphere, University flood aid—The limited area QPF. Preprints, Conference on
of Chicago Press, Chicago, pp. 84-113. Flash Floods: Hydrometeorological Aspects (Los Angeles),
Brandes, E. A., 1974: Radar rainfall pattern optimizing tech- AMS, Boston, pp. 21-28.
nique. NOAA Tech. Memo., ERL-NSSL-67, Norman, Okla., Johnson, R. E., and J. Giraytys, 1975: The AFOS experi-
16 pp. mental system. WMO Bull., 24, 86-92.
Brown, H. E., and E. B. Fawcett, 1972: Use of numerical Klein, W. H., 1976: AFOS forecast applications. Preprints,
guidance at the National Weather Service's National Mete- Sixth Conference on Weather Forecasting and Analysis
orological Center. /. Appl. Meteor., 11, 1175-1182. (Albany), AMS, Boston, pp. 36^3.
Burnash, R. J. C., and T. M. Twedt, 1978: Event-reporting Lott, G. A., 1976: Precipitable water over the United States,

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC

Bulletin American Meteorological Society 699

Vol. 1, Monthly means. NOAA Tech. Rept., NWS-20, Silver , 1977c: Kansas City flash flood of September 12-13, 1977
Spring, Md., 173 pp. —A report to the Administrator. Natural Disaster Survey
Ludlum, D. M., 1970: Early American Tornadoes, 1586-1870. Rept. 77-2, NOAA, Rockville, Md., 49 pp.
AMS, Boston, 219 pp. NWS, 1977: Relationship between manually digitized radar
Maddox, R. A., and C. F. Chappell, 1978: Meteorological as- (MDR) data sums and observed precipitation in long dura-
spects of 20 significant flash flood events. Preprints, Con- tion storms. Eastern Region Tech. Attach. 77-14, Garden
ference on Hydrometeorological Aspects of Flash Floods City, N.Y., 4 pp.
(Los Angeles), AMS, Boston, pp. 1-9. , 1978: Excessive rainfall potential outlook. Tech. Proc.
, F. Caracena, L. R. Hoxit, and C. F. Chappell, 1977: Bull. No. 239, Silver Spring, Md., 7 pp.
Meteorological aspects of the Big Thompson Flash Flood of Oliver, V. J., and R. A. Scofleld, 1976: Estimation of rainfall
31 July 1976. NOAA Tech. Rept., ERL 388-APCL41, from satellite imagery. Preprints, Sixth Conference on
Boulder, Colo., 83 pp. Weather Forecasting and Analysis (Albany), AMS, Boston,
McLuckie, B. F., 1973: The warning system—A social science pp. 242-245.
perspective. NWS, NOAA, Ft. Worth, Tex., 66 pp. Olson, D. A., 1977: Evaluation of 1976 warm season objective
, 1974: Warning—A Call to Action. NOAA, NWS, Ft. quantitative precipitation forecasts. NMC Tech. Attach.
Worth, Tex., 85 pp. 77-5, Camp Springs, Md., 5 pp.
Mileti, D. S., 1975: Natural Hazard Warning Systems in the Owen, H. J., 1977: Guide for flood and flash flood prepared-
United States: A Research Assessment. Monogr. NSF-RA ness planning. NWS, Silver Spring, Md., 47 pp.
E-75-013, Institute of Behavioral Science, University of Randerson, D., 1976: Meteorological analysis for the Las
Colorado, Boulder, 97 pp. Vegas, Nevada, flood of 3 July 1975. Mon. Wea. Rev., 104,
Mogil, H. M., and H. S. Groper, 1976: On the short-range pre- 719-727.
diction of localized excessive convective rainfall. Preprints, Saffle, R. E., 1976: D/RADEX products and field operation.
Conference on Hydro-Meteorology (Ft. Worth), AMS, Preprints, 17th Conference on Radar Meteorology (Seattle),
Boston, pp. 9-12. AMS, Boston, pp. 555-559.
Monro, J. C., and E. A. Anderson, 1974: National Weather Scofleld, R. A., 1978: Using satellite imagery to estimate rain-
Service River Forecasting System. J. Hydraul. Div., ASCE, fall during the Johnstown rainstorm. Preprints, Conference
100, 621-630. on Flash Floods: Hydrometeorological Aspects (Los
Moore, P. L., A. D. Cummings, and D. L. Smith, 1974: The Angeles), AMS, Boston, pp. 181-189.
National Weather Service manually digitized radar program , and V. J. Oliver, 1977: A scheme for estimating convec-
and some applications. NOAA Tech. Memo., NWS SR-75, tive rainfall from satellite imagery. NOAA Tech. Memo.,
Ft. Worth, Tex., 21 pp. NESS 86, Washington, D.C., 47 pp.
NOAA, 1972: Black Hills flood of June 9, 1972—A report to Smith, W., and R. J. Younkin, 1972: An operationally useful
the Administrator. Natural Disaster Survey Rept. 72-1, relationship between the polar jet stream and heavy pre-
NOAA, Rockville, Md., 20 pp. cipitation. Mon. Wea. Rev., 100, 434-440.
, 1973: Final report of the disaster survey team on the Weigel, E. P., 1977: Allies to disaster fighters. NOAA, 7(4),
events of Agnes—A report to the Administrator. Natural 18-20.
Disaster Survey Rept. 73-1, NOAA, Rockville, Md., 45 pp. White, G. F., and J. E. Haas, 1975: Assessment of Research
, 1976a: Big Thompson Canyon flash flood of July 31- on Natural Hazards. Monogr. NSF-RA-E-75-001, Institute
August 1, 1976—A report to the Administrator. Natural of Behavioral Science, University of Colorado, Boulder, 487
Disaster Survey Rept. 76-1, NOAA, Rockville, Md., 41 pp. pp.
, 1976b: Facsimile Products, NOAA, NWS Forecasting Woodley, W. L., A. Olsen, A. Herndon, and V. Wiggert, 1974:
Handbook No. 1. NWS, Silver Spring, Md., -270 pp. Optimizing the measurement of convective rainfall in
, 1976c: Project development plan for AFOS. NWS, Silver Florida. NOAA Tech. Memo., ERL WMPO-18, Miami, 99
Spring, Md., 120 pp. pp.
, 1977a: Appalachian flood April 2-5, 1977—Preliminary , C. G. Griffith, J. Griffin, and J. Augustine, 1978: Satel-
disaster report. NOAA, NWS, Garden City, N.Y., 314 pp. lite rain estimation in the Big Thompson and Johnstown
, 1977b: Johnstown, Pennsylvania flash flood of July flash floods. Preprints, Conference on Flash Floods: Hydro-
19-20, 1977—A report to the Administrator. Natural Disas- meteorological Aspects (Los Angeles), AMS, Boston, pp.
ter Survey Rept. 77-1, NOAA, Rockville, Md., 60 pp. 44-51. •

announcements
Hydrodynamics of lakes—Call for papers there is no participation fee. Participants should register no
A symposium on the Hydrodynamics of Lakes, sponsored by later than 1 August 1978, stating whether a contribution will
the Hydraulics Laboratory of the Federal Institute of Tech- be submitted and, if so, for which session. A detailed program
nology, will be held 12-13 October 1978 in Lausanne, Switzer- will be sent to all who register before the closing date. Par-
land. It will feature four sessions: Mathematical modeling, ticipants must arrange hotel accommodations. The symposium
Measuring techniques, Data analysis, and Air-water interface. will be held at the ficole Polytechnique Federate de Lausanne
Papers are solicited for oral (20 min) and written presenta- (EPF-L), Avenue de Cour 33. Registration should be directed
tion. The written version is due prior to 15 August 1978. to: Prof. Walter H. Graf, Laboratoire d'Hydraulique
Symposium languages are English, French, and German, but (LHYDREP), Rue de Geneve 67, CH-1004 Lausanne, Switzer-
no simultaneous translation will be provided. Proceedings land (tel: (021) 24.13.41; telex: 24 478).
will be published and distributed during the symposium; Continued on page 719

Unauthenticated | Downloaded 04/08/24 02:53 PM UTC