Author: Curt A. Reynolds, reynoldsc@fas.usda.gov
United States Department of Agriculture (USDA), Foreign Agricultural Service (FAS),
Production Estimates and Crop Assessment Division (PECAD)
Submitted to the Third International
Conference on Geospatial Information in Agriculture and Forestry,
Denver, Colorado, November 5-7, 2001
Abstract
The Production Estimates and Crop Assessment Division (PECAD) of the U.S. Department of Agriculture’s (USDA) Foreign Agricultural Service (FAS) recently has made efforts to display nearly real-time agrometeorology data to the public. Several different input databases, climate normals, crop models, and data extraction routines are utilized to produce these images. These data sources and models are briefly described with references to their original sources.The main data output products provided by PECAD include spatial images for estimated precipitation and temperature; cumulative precipitation; percent normal of precipitation; temperature departures from normal; and top- and sub-layer soil moisture. Future outputs envisaged include: crop stage and crop stress for wheat and corn; relative yield reductions for corn, wheat, and soybeans; and NDVI anomalies for most major agricultural regions in the world. It is anticipated that most of these output products will be made available on the Internet in the near future.
PECAD relies on several different data sources to monitor weather anomalies that affect crop production and quality of agricultural commodities. The two main agrometeorological input data sources are:
- Ground meteorological station measurements from the World Meteorological Organization (WMO), and
- Gridded weather data that integrates the WMO ground station data with satellite imagery.
Both these data sets are entered and stored in PECAD’s CADRE (Crop Condition Data Retrieval and Evaluation) database management system (DBMS) on a daily basis. After the agrometeorological data is downloaded, several data extraction routines are executed to display the following spatial data for different regions in the world:
- Actual and cumulative precipitation (in mm)
- Average, minimum, and maximum temperatures (in ºC)
- Precipitation and temperature comparisons to long-term normals
- Temperature departures from normal (in ºC)
- Percent normal of precipitation (in %)
- Snow depth (in cm)
In addition, several crop models and data reduction algorithms are executed for both the station and satellite agrometeorological data sets. These models include crop calendars, crop hazards, and several different crop yield reduction models to assess crop conditions. Most of these crop models and computer tools were developed by researchers from various U.S. universities, government agencies, and private contractors.
Most of these models use daily potential evapotranspiration (PET) which is first calculated from minimum and maximum temperatures and station location (latitude, longitude, and elevation) according to the FAO 56 Penman-Monteith equation (Allen, et al, 1998). Precipitation, PET, and soil water-holding capacity are then utilized to run a two-layer soil moisture model within in CADRE to estimate:
- Top- and sub-layer soil moisture (in mm)
- Percent soil moisture in both soil layers (in %)
Many of the data sets produced by the crop models are not yet on-line, but PECAD in the near future plans to provide output data from these models. Future output products provided by PECAD will include the following:
- Crop stage and crop stress for wheat and corn,
- Relative yield reductions for corn, wheat, and soybeans,
- Normalized Difference Vegetation Index (NDVI) images and anomalies.
Over the past twenty years, most of these agrometeorological variables and crop models were not made available to the public. However, current efforts are being made to display these spatial databases on a nearly real-time basis (every ten-days). The maps and graphs provided are specifically designed to monitor adverse weather conditions over the main agriculture regions in the world. Regional PECAD analysts then utilize this agrometeorological data, other high-resolution satellite imagery, FAS attaché crop reports, in-county sources, wire services, and personal knowledge to estimate national crop production by the 12th of each month.
Some basic information about the agrometeorological input data, climate normals, crop models, and data extraction routines are described below.
Daily station data are originally from the
Global Telecommunication System (GTS) of the WMO, which is a global network of more than 6000 stations (refer to Figure 1).
|
Figure 1. WMO Station Locations |
USDA’s Joint Agricultural Weather Facility (JAWF) receives the WMO station data from NOAA and provides PECAD with the following daily station data:
- Minimum and maximum temperature
- Precipitation
After downloading the WMO station data, PECAD then runs
additional data reduction algorithms, soil moisture, crop stress, and relative
crop yield reduction models, as well as CADRE extraction routines for the
station data. The station data and
output products are then compared to the satellite data to screen the satellite
and station data sets for obvious errors.
The
Air
Force Weather Agency (AFWA) began to develop the Agricultural Meteorology
Model (AGRMET) in 1981 (Cochrane, 1981). The
AGRMET algorithms have evolved over several decades of work and they are
constantly changing. The
AGRMET model basically converts satellite data into spatial agrometeorological
information and utilizes WMO ground station data in the algorithms whenever
possible. These spatial
agrometeorological data sets are then stored in CADRE as 1/8-mesh grid cells.
The coordinate system for the
1/8-mesh grid cells is a 512 by 512 grid laid on a polar stereographic
projection for the northern and southern hemispheres (refer to Figure 2). The projection is true at 60-degree latitude, with the grid cell
resolution ranging from 51-km at the poles to 25-km near the equator. More information on the 1/8-mesh grid reference system can be found in
Hoke, et al., (1981).
|
Figure 2. 1/8-Mesh Grid Reference System for the Northern Hemisphere (from Hoke, et al, 1981) |
|
Figure 3 illustrates the distribution of agrometeorological grid cells downloaded and archived by PECAD on a daily basis. |
The agrometeorology data sets downloaded daily for the grid cells in Figure 3 include the following:
- Minimum and maximum temperature (Kopp, 1995)
- Precipitation (Hollinger, 1989, and Kiess and Cox, 1988)
- Snow depth (Luces, et al, 1986)
- Solar and longwave radiation (Shapiro, 1987, Idso, 1981, and Watchmann, 1975)
- Potential and actual evapotranspiration (Mahrt and Ek, 1984, and Ek and Mahrt, 1991)
The references for derivation of these agrometeorological
variables are indicated above. After downloading the above agrometeorology data
sets, soil moisture, crop stress, and relative crop yield reduction models as
well as CADRE extraction routines are executed to display the various maps and
graphs on the Internet.
The current global precipitation
is estimated by blending four different data sources together. The SSM/I
(Special Sensor Microwave /Imager) is the first source. Rain rates are formulated from the brightness temperatures as documented
by Hollinger (1989). The second
source uses the Real Time Nephanalysis Cloud Model (RTNEPH)
described by Kiess and Cox (1988). The third and final data sources utilize WMO
ground station data and geostationary
satellites.
Development of the CADRE’s DBMS
began in 1979 (Tingley, 1988) and it was one of the first GIS (Geographic
Information Systems) DBMS designed specifically for global agricultural
monitoring. CADRE is the
operational outgrowth of the LACIE (Large Area Crop Inventory Experiment) and
AgRISTARS (Agriculture and Resources Inventory Surveys Through Aerospace Remote
Sensing) programs which began in 1974 and 1980, respectively (Boatwright and
Whitefield, 1986). The main
cooperating agencies for the LACIE and AgRISTARS programs were the National
Oceanic and Atmospheric Administration (NOAA), National Aeronautics and Space
Administration (NASA), and U.S. Department of Agriculture (USDA). These programs were the first joint effort by the U.S. government to use
satellite imagery to continuously monitor and assess crop production over
selected areas of the world. The
AgRISTARS program followed the LACIE program and it developed many operational
models and NOAA-AVHRR processing procedures currently used by PECAD and CADRE.
Twenty years ago, the original
developers of CADRE had a unique geospatial vision although FORTRAN card readers
were still used, storage capacity of computers were limited, and GIS (Geographic
Information Systems) software were not yet available. CADRE’s current DBMS is essentially the same DBMS of twenty years ago. A flow chart of CADRE’s data stream is shown in Figure 4. Changes in operating systems and upgrading the baseline datasets are the
main differences between the CADRE’s old and current systems. For example, CADRE was originally loaded in a DEC mainframe, later
transferred to a DEC VAX system, and currently resides on a DEC UNIX server. Switching to different operating systems also entailed converting
CADRE’s data archive, data extraction routines, and crop models into different
operating environments. The latest
switch to a UNIX environment made it possible to display output data from CADRE
on the Internet and on a nearly real-time basis (every ten days).
|
Figure 4. Input Data Sets, Models, and Extraction Tools for CADRE |
The most recent upgrades to CADRE involved transferring new digital baseline
databases developed by other agencies into CADRE’s 1/8-mesh grid reference
system. For example, monthly grid cell normals for rainfall and
temperature were recently extracted and imported into CADRE by utilizing the
Leemans and Cramer (1991) 16-bit global climate images.
These images also are available in 8-bit format from the FAO
Climate Maps. These grid cell
normal values are utilized in the precipitation and temperature time series
graphs provided on-line by PECAD.
Other baseline normal data sets were improved by incorporating monthly
station normals (from 1960-1990) provided by the WMO and NOAA. The soil water-holding capacities in CADRE were also upgraded by
extracting values derived by the FAO
(1996) Digital Soil Map of the World (DSMW). In addition, elevation and
monthly potential evapotranspiration normals for the grid cells were derived
from digital datasets developed by Row and Hastings (1997), and Ahn and
Tateishi (1994), respectively.
Crop information within CADRE was
also updated by compiling provincial crop yield and production data for most
countries monitored by PECAD. Administration
units at level-2 were utilized to store historical crop information because most
national crop statistics are reported at this scale. These national crop statistics were then converted into ARCVIEW shape
files and imported into CADRE’s 1/8-mesh grid cell format. Crop information updates are continuously required (average yield over
the last five years) in efforts to make the relative crop yield reduction models
more effective.
PECAD continuously assess and test
operational crop models over large areas for accuracy and improvements. The crop models desired, chosen, and developed by PECAD over the years
are those with minimum regression coefficients so that they have global
applications. For example, some models deemed to have good results by the
U.S. scientific community will not produce satisfactory results when implemented
globally because these models use too many localized regression coefficients or
were tested within limited number of areas.
The original soil moisture and
crop models used by CADRE began development in 1978 and were implemented by the
AgRISTARS program in the early 1980s, (Boatwright and Whitefield, 1986). These models are FORTRAN programs which extract daily station and gridded
agrometeorological data described above. Personnel from the USDA, NOAA, NASA,
and private contractors wrote the original crop models. The initial focus of crop model development was for the U.S. and Former
Soviet Union (FSU). However,
many of these models are currently being expanded for other counties throughout
the world.
The two-layer soil moisture
algorithm developed by the AgRISTARS program is the backbone algorithm that runs
the crop calendar (growth stage) and crop stress (alarm) models. The two-layer soil moisture model is a bookkeeping method that
accounts for the water gained or lost in the soil profile by recording the
amount of water withdrawn by evapotranspiration and replenished by
precipitation. The final aim of the soil moisture model is to estimate if soil
moisture storage between dry spells was adequate for maximum plant growth.
Crop calendar models developed by
the AgRISTARS program are accretion models that model the crop growth
incrementally, based on growing degree-days (or thermal units) for several
different types of crops and crop varieties. The crop calendar is a growing
degree-day algorithm that uses daily minimum and maximum temperature
measurements, as well as threshold temperatures defined by the particular crop
type. The current crop calendar
models are initialized by average start of season data derived from national
crop reports. Future efforts
will be made to initialize the crop models from start of season data derived
from agrometeorological data or vegetation index numbers (VIN) stored in CADRE. The VIN database in CADRE is derived from the red and near-infrared
channels from the NOAA-AVHRR satellite series as described by Bethel and Doorn,
(1998).
Crop-stress models developed by
the AgRISTARS programs use both the soil moisture and crop calendar algorithms,
as well as a hazard algorithm to alert analysts of abnormal temperature or
moisture stresses that may affect yields. These
hazard algorithms are based on temperature and soil moisture thresholds known to
be outside the optimal range of growing conditions and which may cause crop
damage at various crop stages. For example, optimal growing conditions for corn
is critical during the reproductive phase and the soil moisture and temperature
thresholds are most sensitive during this stage. Therefore, if the plant experiences extreme water deficits or temperature
conditions during the reproductive phase, the alarm model alerts the analysts of
the crop stress in the region.
In addition, new crop models are
constantly reviewed for possible integration into the operational systems of
CADRE. For example, other crop
models written by USDA and university researchers have been modified to run
specifically from CADRE input data. In these cases, PECAD has work directly with the author of the crop model so that
the model is running in-house with CADRE input data. These in-house models need a full growing season to get
results and analysts will not use the models unless they feel the model agrees
with ground conditions.
Most yield reduction models used
by PECAD were written by other individuals outside of the AgRISTARS project and
their models were customized to run from agrometeorological data extracted from
CADRE. Yield reduction models begin with the assumption of perfect conditions
and decrease yield predictions based on crop stresses. The goal of these crop models is to provide a yield estimate quantified
as tons per hectare. Most of these models have a crop water production function
algorithm which compares the crop yield with optimal water requirements to the
actual water (rainfall) received by the crop.
Since no model is correct at all
times for all geographic areas, analysts often run several models at once to
reduce reliance on one particular model. The most recent trusted yield reduction
models (used for lock-up analysis) are the Sinclair soybean (Sinclair, et al,
1991) and CERES wheat models (Richtie, et al, 1998). Of course, other information from other sources such as satellite
imagery, in-county sources, FAS attaché reports, wire services, and personal
knowledge are also used to decide how adverse weather conditions might have a
significant impact on crop production.
PECAD has over two decades of
experience in monitoring global weather data to maintain an early alert status
of agricultural conditions that may alter regional yield potentials or effect
international markets. During these
two decades, PECAD has developed numerous algorithms to monitor
agrometeorological variables and crop conditions throughout the world.
The most recent additions to CADRE
have been introducing JAVA and PERL extraction routines to automatically
generate maps and graphs. These images are then displayed on the Internet and
are automatically updated on a decadal basis. These improvements should allow more users and professionals (both inside
and outside of PECAD) to view CADRE’s invaluable agricultural monitoring data
sets.
On-going modifications include
upgrading all crop models for global coverage because many of the original crop
models were only designed for a few specific countries. The final result will be
an upgraded geospatial information system with global coverage that will display
agrometeorological and crop monitoring information on a nearly real-time basis. In summary, CADRE is a geospatial
DBMS which stores numerous data sets and performs various modeling functions. A brief list of input data, climate normals, crop information and models,
and data extraction routines performed within CADRE include:
Time-series data sets
- Daily WMO station data (precipitation, min. and max. temperatures)
- Daily agrometeorological data derived from station and satellite data, and imported into 1/8-mesh grid cells (precipitation; min and max temperatures; snow depth; solar and longwave radiation; potential and actual evapotranspiration).
- Decadal VINs (vegetation index numbers) derived from LAC (Local Area Coverage, approx 1.1-km pixels) data of the NOAA-AVHRR satellite series and imported into 1/8-mesh grid cells.
- Biweekly VINs (vegetation index numbers) derived from GAC (Global Area Coverage, approx 8-km pixels) data of the NOAA-AVHRR satellite series and imported into 1/8-mesh grid cells.
Normal baseline data sets
- Normal precipitation and temperature values for WMO stations (from WMO and NOAA)
- Normal precipitation, temperature, potential evaporation, and elevation values imported into 1/8-mesh grid cells
(from GRID Geneva)
- Soil-water holding capacity imported into 1/8-mesh grid cells (from FAO’s DSMW, 1996)
- Biweekly VIN normals for the GAC data set.
Crop Information and Models (for wheat, corn, and soybeans)
- Crop type and average start of season
- Average yield and area planted
- Percent crop production within a country
- Two-layer soil moisture algorithm
- Crop calendars (based on growing-degree days)
- Crop stress or alarm models for corn and wheat (based on soil moisture and temperature thresholds)
- Crop water production functions to estimate relative yield reductions
- Crop models by Ricthie
CADRE extraction routines
- Automated maps and graphs generated every 10-days for display on Internet
- Interactive ARCVIEW 3.2 scripts for displaying station and grid cell data
- Interactive CADRE X-tract program for displaying graphs
AFWA |
Air Force Weather Agency |
AGREMET |
Agricultural Meteorology Model |
AgRISTARS |
Agriculture and Resources Inventory Surveys Through Aerospace Remote Sensing |
AVHRR |
Advanced Very High Resolution Radiometer |
CADRE |
Crop Condition Data Retrieval and Evaluation |
COTS |
Commercial Off-the Shelf Software |
DBMS |
Database management system |
DSMW |
Digital Soil Map of the World |
EUMETSAT |
Europe's Meteorological Satellite Organization |
FAO |
Food and Agriculture Organization |
FAS |
Foreign Agricultural Service |
FSU |
Former Soviet Union |
GIS |
Geographic Information Systems |
GMS |
Geostationary Meteorological Satellite |
GOES |
Geostationary Operational Environmental Satellites |
GTS |
Global Telecommunication System |
IIASA |
International Institute for Applied Systems Analysis |
JAWF |
Joint Agricultural Weather Facility |
LACIE |
Large Area Crop Inventory Experiment |
NASDA |
National Space Development Agency of Japan |
NDVI |
Normalized Difference Vegetation Index |
NOAA |
National Oceanic and Atmospheric Administration |
PECAD |
Production Estimates and Crop Assessment Division |
PET |
Potential evapotranspiration |
RTNEPH |
Real Time Nephanalysis Cloud Model |
SSM/I |
Special Sensor Microwave/Imager) |
USDA |
United States Department of Agriculture |
VIN |
Vegetation index numbers |
WMO |
World Meteorological Organization |
Ahn C.-H. and R. Tateishi.
1994. Development of a Global 30-minute grid Potential Evapotranspiration Data
Set. Journal of the Japan Soc. Photogrammetry and Remote Sensing, 33(2): 12-21.
Allen, R. G., L.S. Pereira, D. Raes,
and M. Smith. 1998. Crop evapotranspiration; Guidelines for computing crop water
requirements, FAO Irrigation and Drainage Paper 56, pp. 27-65.
Bethel, G. and
B. Doorn, 1998. USDA Remote Sensing
Technical and Systems Support for Operational Worldwide Agricultural Analysis, 1st
International Conference: Geospatial Information in Agricultural and Forestry,
Disney’s Coronado Springs Resort, Lake Buena Vista, Florida, USA, 1-3 June
1998
Boatwright, G.O.
and V.S. Whitefield. 1986. Early
Warning and Crop Condition Assessment Research. IEEE Transactions on Geosciences and Remote Sensing. Vol
GE-24, No 1., January.
Cochrane,
M.A. 1981. Soil Moisture and
Agromet Models, Technical Report, USAF Air Weather Service (MAC), USAFETAC,
Scott AFB, Illinois, TN-81/001, March 1981.
Ek, M., and L. Mahrt, 1991. OSU
1-D PBL Model User's Guide. A one-dimensional planetary boundary layer model
with interactive soil layers and plant canopy. Version 1.0.4. March 1991. Dept
of Atmospheric Sciences, Oregon State University.
FAO. 1996. The Digitized Soil Map of the World Including Derived Soil Properties, CD-ROM,
Food and Agriculture Organization, Rome.
Hoke, J.E., J.L. Hayes, and L.G. Renninger, 1981. Map Projections and Grid Systems for
Meteorological Applications, AFGWTC/TN-79/003 (AD-A100324), Air Force Weather
Agency, Offutt AFB, Nebraska.
Hollinger, J., 1989: DMSP Special Sensor Microwave /Imager Calibration/ Validation
Report, Volume I, Naval Research Laboratory, Washington D.C.
Idso, S. 1981. A set of equations for
full spectrum and 8 to 14ųm and 10.5 to 12.5ųm thermal radiation from
cloudless skies. Water Resources Research, 17:295-304.
Kiess, R.B. and W.M. Cox , 1988. The AFWGWC
Automated Real-Time Cloud Analysis Model, Technical Report AFGWTC/TN-88/001
(AD-B121615), Air Force Weather Agency, Offutt AFB, Nebraska.
Kopp, T.J. 1995. The Air Force Global Weather
Central Temperature Model, Technical Report AFGWTC/TN-95/004, Air Force
Weather Agency, Offutt AFB, Nebraska.
Leemans, R. and W. Cramer. 1991. "The IIASA database for mean monthly values of
temperature, precipitation and cloudiness on a global terrestrial grid".
Research Report RR-91-18. November 1991. International Institute of Applied
Systems Analyses, Laxenburg, pp. 61.
Luces, S.A.,
J.D. Matens, and S. J. Hall. 1986. The AFGWC Snow Analysis Model, Technical
Report AFGWTC/TN-86/001 (AD-A176202), Air Force Weather Agency, Offutt AFB,
Nebraska.
Mahrt, L. and M.
Ek. 1984. The influence of atmospheric stability on potential evaporation, J.
Climate and Appl. Meteor., 23:222-234.
Richtie,
J.T., U.Singh, D.C. Godwin, and W.T. Bowen. 1998. Cereal growth, development,
and yield. G.Y. Tsuiji, et al, (eds),
Understanding Options in for Agricultural Production, Luwer Academic Publishers,
Great Britian, 79-98.
Robine, K. 1998. Global
Agrometeorological Database Exploitation Using Avenue Scripts and Sybase SQL
Server. 18th Annual International ESRI User Conference, San Diego,
CA, 26-30 July, 1998.
Sinclair, T. R.,
S. Kitani, K. Hinson, J. Bruniard, and T. Horie. 1991. Soybean Flowering Date: Linear and Logistic Models
Based on Temperature and Photoperiod. Crop
Science, Vol 31, May-June, pp.786-790.
Shapiro, R.
1987. A Simple Model for the Calculation of the Flux of Direct and Diffuse Solar
Radiation Through the Atmosphere, AFGL-TR-87-0200, Air Force Geophysics Lab,
Hanscom AFB, MA.
Row III, W.R., and D.
Hastings. 1997. TerrainBase Global
Terrain Model, National Geophysical Data Center and World Data Center-A for
Solid Earth Geophysics, Boulder, Colorado U.S.A.
Tingley, W. 1988. Crop Condition Data Retrieval and Evaluation (CADRE) DBMS Dictionary.
Lockheed Engineering and Sciences Company, Inc. Unpublished.
Watchmann, R. 1975. Expansion of the atmospheric temperature moisture profiles in
empirical orthogonal functions for remote sensing applications. Digest of preprints,
Topical Mtg on Rem. Sens. of the Atm.,
Optical Society of America, Anaheim, CA.