USDA's research and operational programs used remotely sensed data and related technologies to monitor, assess, and administer agricultural and forestry resources. The Agriculture Research Service (ARS) enhanced remote-sensing knowledge and developed productive applications at research facilities located throughout the United States. At Weslaco, Texas, the ARS Integrated Farming and Natural Resources Unit used airborne video data integrated with Global Positioning System (GPS) and Geographic Information System (GIS) technologies to detect and map the aquatic weeds, hydrilla, and water hyacinth in the Rio Grande River in extreme southern Texas. The unit also developed and delivered GIS maps to the Texas Parks and Wildlife Department and the Lower Rio Grande River Water Districts, which can use the maps to control the spread of these weeds. At Weslaco, aerial digital images in conjunction with yield monitor data were collected from 20 grain sorghum fields owned by Rio Farms, Inc., Monte Alto, Texas. Rio Farms managers have used the aerial images and yield maps, produced from the yield monitor data, to improve farm management.
The ARS Hydrology Laboratory initiated work to develop methods for retrieving soil moisture information from satellite-borne microwave sensors. Scientists conducted experiments at ARS facilities in Oklahoma using aircraft prototypes of these instruments. At the ARS Jornada Experimental Range, the Hydrology Laboratory collected laser scanning data and visible, thermal infrared, and video imagery to infer surface temperature, albedo, vegetation indices, roughness, and other land-surface characteristics. These parameters are to be used as inputs for land-surface models, coupled with atmospheric models. In preparation for the launch of NASA's Earth Observing System (EOS) AM1 satellite (now called Terra), the Hydrology Laboratory has flown multispectral thermal infrared sensors (TIMS and MASTER) over the Jornada Experimental Range to estimate surface emissivity and temperature. ARS will use the latter to estimate the surface-sensible heat flux. This approach has been successfully demonstrated with data acquired over the El Reno Grazing Lands facility in Oklahoma as part of the Southern Great Plains 97 experiment.
At ARS laboratories in Ames, Iowa, ARS researchers continued to develop methods for detecting weeds within corn and soybean canopies. Scientists measured and recorded the leaf spectra of weeds, corn, and soybeans grown in single species plots and in competition. ARS made concurrent reflectance measurements with a broadband radiometer of plots with different weed densities and species composition. To discriminate nutrient stress and its potential impact on yields, ARS measured the reflectance of corn produced from various populations and nitrogen management strategies. To determine the effect of soil background on crop detection, scientists developed a seasonal reflectance library for corn, soybeans, and wheat with different tillage and crop residue practices.
The ARS Great Plains System Research Unit in Ft. Collins, Colorado, the ARS Southwest Watershed Research Center in Tucson, Arizona, and Michigan State University brought private, nonprofit, and public-sector groups together in Arizona, Colorado, New Mexico, and Nebraska to develop techniques for estimating the amount and spatial distribution of total standing, green, and senescent biomass in grassland ecosystems. The project was designed to demonstrate how remote sensing can be cost-effectively used in managing these diverse and important natural resources.
In Phoenix, Arizona, the U.S. Water Conservation Laboratory (USWCL), in cooperation with the University of Arizona, designed and tested a system for acquiring images from sensors mounted on agricultural implements. Scientists used data from the system to verify a new chlorophyll index that shows promise as an indicator of crop fertilizer needs. In a separate study using the Free Air Carbon dioxide Enrichment (FACE) facility, USWCL used measurements of plant reflectance and temperature to assess the response of wheat and sorghum crops to increased atmospheric carbon dioxide. This research will improve our ability to monitor and understand how various global change scenarios will affect agricultural productivity and carbon sequestration. USWCL scientists cooperated with NASA's Stennis Space Center, through a Space Act Agreement, to develop products and applications to manage crops and soils using data from multispectral airborne sensors. Scientists have used the data to develop methods for detecting crop water stress and improving irrigation scheduling with minimal ground-based inputs and no image calibration. USWCL scientists, as members of the NASA Landsat 7 Science Team, combined imagery from the Landsat 7 Enhanced Thematic Mapper-Plus (ETM+) sensor with a grassland growth model to produce daily maps of grassland biomass, root biomass, and soil moisture over a semiarid watershed. These maps allow for better rangeland management and provide a greater understanding of rangeland ecology for addressing soil erosion and drought.
Using remote-sensing technology, the ARS Remote Sensing and Modeling Laboratory (RSML) and the Hydrology Laboratory initiated a long-term experiment to evaluate the economic and environmental impact of four alternative farming practices on surface and subsurface water quality. They mapped subsurface flow patterns with Ground Penetrating Radar and linked them to crop yields and remotely sensed hyperspectral data. Their remote-sensing assessment of the spatial and temporal variability of crops will benefit farmers and various agricultural industries by providing a watershed-scale demonstration site at which crop yields, profitability, and environmental impact can be compared under identical hydrogeological setting and climatic conditions.
In a field study cosponsored by Stennis Space Center and ARS, RSML investigated the use of Airborne Visible and Infrared Imaging Spectrometer (AVIRIS) hyperspectral data to detect nitrogen deficiency and water stress in crops. Scientists conducted the study at Shelton, Nebraska, over an irrigated corn field with 20 variable-rate nitrogen plots. RSML acquired coverage from three NASA AVIRIS flights during the growing season and made extensive ground measurements of the agronomic and spectral characteristics to facilitate the development of models to identify crop stress. Over spring wheat fields in Montana, North Dakota, South Dakota, and Minnesota, RSML identified and monitored crop stress throughout the growing season. Through cooperative research with the USDA's Foreign Agricultural Service (FAS), scientists evaluated models for assessing spring wheat production in Siberia and Kazakhstan. Also, RSML used fluorescent sensing to detect ozone damage, ultraviolet radiation damage to vegetation, and the effect of increased carbon dioxide. RSML demonstrated that remote sensing can accurately assess tropospheric environmental problems.
Researchers at the ARS Western Integrated Cropping Systems Research Unit in Shafter, California, continued to develop procedures for detecting and managing water stress and pest infestations. The Shafter Airborne Multispectral Remote Sensing System (SAMRSS), which acquires high-resolution imagery in the visible, near-infrared, and thermal infrared, was flown on 30 missions. Using this imagery and data from ground-based collections, researchers developed procedures for detecting early infestations of spider mites in cotton and the onset of water stress. Researchers at Shafter cooperated with colleagues from Opto-Knowledge Systems, Inc., and NASA in studying the use of hyperspectral imagery from AVIRIS in agricultural applications. In cooperation with Cotton, Inc., researchers purchased a cotton yield monitor and installed it on a cotton picker to determine spatial relationships between yields and remotely sensed variables. They determined that the spatial variability in cotton yields across a field can best be estimated using midseason multispectral imagery.
The ARS Rangeland Resources Research Unit at the High Plains Grasslands Research Station in Cheyenne, Wyoming, collected Very Large Scale Aerial (VLSA) images of soil erosion plots at the Central Plains Experimental Range. To improve rangeland management, researchers there compared measurements made from this aerial imagery with hand-collected data of bare ground, plant cover, microtopography and microflow patterns. In cooperation with the Department of the Interior's Bureau of Land Management, the State of Wyoming, and private ranchers, the unit has used VLSA and space imagery for rangeland monitoring that is low-cost and defendable in court. An additional critical component of this research will relate important aspects of rangelands like leaf area of dominant functional plant types to carbon dioxide fluxes and carbon cycling.
At Reynolds Creek Experimental Watershed, the ARS Northwest Watershed Research Center (NWRC) used satellite imageryLandsat 5 Thematic Mapper and Satellite Pour l'Observation de la Terre (SPOT) 3HRVto classify accurately shrub steppe and subalpine plant communities. A practical procedure for classifying and mapping intermountain plant communities using satellite imagery was provided to the Bureau of Land Management's Snake River District in Idaho. NWRC also has evaluated multispectral digital aerial photography as a tool to determine stream shading by riparian vegetation and its effects on surface water quality in rangelands. Using Landsat imagery, NWRC quantified leaf area indices for rangeland plant communities for input to water and energy balance models that can be used to estimate rangeland plant production. NWRC also has worked with NASA to develop the use of synthetic aperture radar to map the distribution of frozen soil and soil-water content in rugged terrain.
The ARS laboratory in Sidney, Montana, conducted remote-sensing research for noxious weed identification, mapping, and monitoring crop management. The laboratory used color aerial photography to map leafy spurge infestations in Theodore Roosevelt National Park and the Sheyenne National Grasslands of North Dakota. It subsequently interpreted, digitized, and incorporated the national park's imagery with similar data collected in 1993. The 5-year comparison has provided a valuable evaluation of leafy spurge growth, distribution, and dynamics within the park. The laboratory also acquired normal color aerial photography of salt cedar stands in Wyoming, Utah, Nevada, and California to develop baseline population levels for studies involving releases of biological control agents. Research continued on the use of aerial videography for monitoring crop development and yield prediction in western North Dakota and eastern Montana. The Montana research project used the remote imagery to develop and assess the success of precision agriculture cropping strategies. Researchers have initiated the same technologies in North Dakota to monitor potato crop development and production. A new hyperspectral radiometer and imaging system will be used to improve weed identification and mapping and provide calibrated data for the multitemporal evaluation of crop development.
The satellite remote-sensing program of FAS, operated by the Production Estimates and Crop Assessment Division (PECAD), remained a critical element in USDA's analysis of global agricultural production and crop conditions by providing timely, accurate, and unbiased estimates of global area, yield, and production. Satellite-derived early warning of unusual crop conditions and production anomalies enabled more rapid and precise determinations of global supply conditions. The FAS/PECAD analysts employed a proven "convergence of evidence" approach to crop assessmentincorporating NOAA Advanced Very High Resolution Radiometer (AVHRR), Landsat, and SPOT imagery; crop models; weather data; U.S. agricultural attaché reports; field travel; and ancillary data to forecast foreign grain, oilseed, and cotton production.
FAS/PECAD accurately forecast 1999-2000 global grain production to within roughly 3 percent of final output. Visual interpretation of high-resolution SPOT satellite imagery provided early warning of significant crop stress in key Russian winter-wheat-producing regions. An analysis of yield-simulation models and vegetative indices derived from AVHRR satellite data indicated where drought impact was most severe and identified crop areas that escaped damage. Subsequent harvest reports verified the FAS/PECAD early-season analysis.
FAS/PECAD personnel have participated in international agricultural research studies in Russia and Kazakhstan, in conjunction with USDA/ARS researchers, to expand and improve crop assessment resources for estimating wheat production in the former Soviet Union. The team has worked closely with agricultural and remote-sensing researchers from the Kazakhstan Space Research Institute and the Institute for Environmental Monitoring in Western Siberia to evaluate and refine the use of yield- simulation models and new satellite sensors. This cooperative research resulted in a valuable exchange of crop-forecasting technology.
FAS remote sensing supported Department of State assessments of food needs in the former Soviet Union, Indonesia, and North Korea. Also, FAS prepared detailed analyses of the record Argentine soybean crop, the bumper wheat crop in Australia, the bumper soybean crop in Brazil, drought in the Ukraine, dryness in China and North Korea, and flooding in Mexico.
The Farm Service Agency (FSA) continued to share with FAS the cost of analyzing imagery of the United States. A timely analysis of U.S. crop conditions, combined with weather data, crop model results, and GIS products, made possible the development of accurate and timely projections and comprehensive evaluations of crop disaster situations. During the 1999 growing season in the United States, the domestic analysts of FAS/PECAD provided early warning on anomalous crop conditions, including severe droughts in the Mid-Atlantic States and Eastern Corn Belt, as well as flooding from hurricanes and subsequent rainfall in North Carolina and southern Virginia. The impact of the sixth consecutive wet spring and early summer on agricultural interests in North and South Dakota was determined and reported in interagency briefings and published on the internal FSA/FAS Web site. FSA continued to be a partner in the National Aerial Photography Program (NAPP) and the National Digital Orthoquad Program (NDOP). FSA started field-reengineered business processes that combine the use of digital orthophotography, GIS, GPS, and satellite imagery to replace the use of hard-copy NAPP aerial photography and 35-millimeter slides.
The USDA Forest Service provided support to the Selection Committee for the NASA Research Announcement on Agriculture, Forestry, and Range Management by contributing three members to the selection panel. A proposal from the Forest Service's Fire Sciences Laboratory for Mapping Fire and Fuels Characteristics Using Remote Sensing and Biophysical Modeling for Operational Fire Management was selected by the committee. This project will provide researchers at the Forest Service and universities with vegetation maps of areas prone to wildfires, allowing firefighters to determine which plants are more likely to fuel wildfires and better predict the paths of such fires.
Project Redsky, a DoD/Forest Service experiment to detect fires using DoD satellites, continued in 1999. The Forest Service also has supported the implementation and testing of the Hazard Support System at the U.S. Geological Survey's Reston, Virginia, campus. This system, which warns of the outbreak of wildfires and volcanic eruptions, is a joint program among the DoD, the U.S. Geological Survey, NASA, and other Government agencies. The Langley Research Center FireSat Team provided airborne measurements during a series of controlled fires in Gila National Forest; the measurements were conducted to validate the performance of the Hazard Support System.
The Forest Service continued to study the use of Light Intersection Direction and Ranging (LIDAR) data to create three-dimensional structure maps for forested lands. A NASA C-130, carrying the Laser Vegetation Imaging Sensor (LVIS), mapped a 200-square-mile forest area in northern California as a precursor to the launch of the Vegetation Canopy Lidar (VCL) satellite scheduled for launch in September 2000. The California flights will allow scientists to acquire VCL-like data that will be used to fine-tune data analysis methods.
The National Agricultural Statistics Service (NASS) used remote-sensing data to construct area frames for statistical sampling, to estimate crop area, to create crop-specific land-cover data layers for GIS, and to assess crop conditions. For area frame construction, NASS combined digital Landsat and SPOT data with U.S. Geological Survey digital line-graph data, enabling the user to assign each piece of land in a State to a category, based on the percentage of cultivation or other variables. NASS implemented a new remote-sensing based area frame and sample for Mississippi. The remote-sensing acreage estimation project analyzed Landsat data of the 1998 crop season in Arkansas, North Dakota, and South Dakota to produce crop acreage estimates for major crops at State and county levels, plus a crop-specific categorization in the form of a digital mosaic of Thematic Mapper scenes distributed to users on a CD-ROM. For the 1999 crop season, NASS headquarters and several NASS field offices entered into partnership agreements with State organizations to decentralize the Landsat processing and analysis tasks. Technicians collected data for 1999 acreage estimation analysis in Arkansas, Illinois, Mississippi, New Mexico, and North Dakota. Vegetation condition images based on AVHRR data were used with conventional survey data to assess crop conditions. NASS employed this imagery to monitor the 1999 drought in Mid-Atlantic States.
Natural Resources Conservation Service (NRCS) continued its cooperative
partnership with Federal, State, and local agencies in developing 1-meter
digital ortho-imagery coverage of the Nation through both NDOP and NAPP.
By year's end, approximately 1,800 counties will have complete digital
orthoimagery coverage. NRCS and FSA jointly awarded an innovative contract
for the development of digital color infrared ortho-imagery for Hawaii.
Imagery acquired for Hawaii will be fully digital and integrated with
data collected by an onboard inertial measuring unit and dual-band GPS.
The inertial measuring unit and GPS data significantly reduce the need
for obtaining costly ground control to generate digital orthoimagery to
meet national map accuracy standards. NRCS continued to work with the
Massachusetts Institute of Technology to make seamless digital orthoimagery
data accessible over the Internet.