ESE Applications in the Agriculture and Forestry Markets
A sample of project highlights
By Ronald J. Birk and Roger L. King, Ph.D., P.E.

Agriculture and forestry are vitally important to every person on Earth. They are also two key industrial sectors of importance to both the U.S. and global economy for which modern remote sensing technology holds great promise. The intersection between these two industrial sectors and remote sensing technology occurred early in the history of the technology. In the early 70s, the advent of Landsat data led to studies of agricultural and forestry applications that had promising results. Such early projects as the Large Area Crop Inventory Experiment (LACIE) demonstrated the effectiveness of the technology in the mid-70s. Since then, a significant number of applications in both sectors have reached maturity, particularly for regional and management applications.
    Applications research continues and NASA scientists are using measurements from spaceborne, airborne, and ground-based instruments to develop information products for farmers and forest managers to use in their sustainable natural resource management practices. Through NASA science programs such as the Land Use and Land Cover, Pathfinder, BOREAS, Forest Ecosystem Dynamics, HAPEX-Sahel, and their advances in remote sensing applications, new levels of understanding of the Earth's surface and atmosphere have been achieved. In turn, these results are driving the gauges that will guide policy decision from local to global scales.

RDACS (Real-time Digital Airborne Camera System) Sensor developed by Spectral Visions captured vegetation in different stages of change due to Fall color change. The bands sensed capture the "red edge" which is that part of the spectrum that changes most rapidly when vegetation is changing or under stress.

Agricultural Applications
Key national and international examples of agricultural applications of satellite remote sensing include the National Agricultural Statistics Service (NASS) and the Foreign Agricultural Service (FAS). The NASS, a division of the U.S. Department of Agriculture, has used satellite remote sensing resources such as those provided by the Earth Science Enterprise (ESE) program to measure crop acreage, classify crop and vegetation type, conduct crop yield predictions, determine vegetation and crop health, and monitor droughts and floods. On a global basis, the FAS utilizes over 10,000 Landsat, SPOT, and Advanced Very High Resolution Radiometer (AVHRR) images per year to produce monthly global commodities production estimates.
    The portfolio of measurements have recently been augmented through NASA's Science Data Buy. SDB provides the science community with such products as wind vector profile measurements, 3 meter vertical accuracy digital elevation models (DEMs), high-resolution digital imagery, cloud monitoring information, and global coverage of orthorectified Landsat imagery. (See www.crsp.nasa.gov)
    Each of these data sets has valuable and practical applications. Many of the data sets can be integrated into sophisticated models to perform important predictive functions such as disaster risk assessment, crop yield predictions, and the basis of carbon permit allocations for forest management.
    A sample of four ESE-sponsored research/applications projects currently under way illustrates the richness of potential applications in this sector.
    Dr. Kevin P. Price of the Kansas Applied Remote Sensing (KARS) program at the University of Kansas uses remotely sensed data in improved crop type classification, separation of irrigated versus non-irrigated acreage, delineation of USDA Conservation Reserve Program (CRP) lands, and pre-harvest estimates of corn yields in Iowa and wheat yields in Kansas. Methods developed at KARS are now used for monitoring grassland and cropland health and vigor, and identifying impact areas due to natural hazards such as drought and severe weather. As a NASA Regional Earth Science Application Center (RESAC), KARS efforts are focused on agricultural lands of the Great Plains, with other activities in China, Africa, and Mexico. (See www.kars.ukans.edu)
    Dr. Susan Moran of the USDA-KARS Southwest Watershed Research Center (SWRC), located in Tucson, Arizona, works extensively with data provided through ESE. Methods have been developed to use Earth Observing System (EOS) sensors for monitoring crop transpiration rates, crop growth, soil moisture, and farm cultivation practices through a combined multispectral remote sensing and modeling approach. (See www.tucson.ars.ag.gov/rsg/landsat and www.uswcl.ars.ag.gov)
    The Upper Midwest Aerospace Consortium (UMAC), led by George Seielstad, University of North Dakota, brings Wyoming ranchers, farmers, state and federal agencies, and The Nature Conservancy together to study grazing lands for livestock. An overflight by Positive Systems collected multispectral data that enabled a rancher to point out the exact location of leafy spurge; an exotic, invasive weed blighting many western grasslands. Computer analysis by Photon Research Associates identified eleven other spots on a ranch too big for a single rancher to monitor where the weed was likely growing. Range management acquired a valuable new tool in maintaining healthy grasslands. (See www.umac.org)
    Research at Mississippi State University (MSU) is being conducted on detection of interrelated stresses to crops, including weed and insect pressure and plant nutrient requirements. Dr. Michael Cox is studying the site-specific effects of soil chemical and physical properties on soybean yield through measurements provided by remote sensing systems. (See www.erc.msstate.edu)

Forestry Applications
The U.S. Forest Service and commercial forestry firms make good use of the data provided by Landsat, SPOT, and AVHRR to analyze forest types, stand ages, and associated land use and land cover information. A key application of remote sensing for forest management is monitoring forest fires and assessing fire hazards using the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) data provided by ORBIMAGE. Example applications include discrimination of forest health and type, forest harvest monitoring, forest fire monitoring, and timber disease monitoring. The instruments likely to be of greatest value include Landsat 7, Moderate Resolution Imaging Spectrometer (MODIS), Advanced Spaceborne Thermal Emission Reflector Radiometer (ASTER), and SeaWiFS.
    Dr. Sasan Saatchi, Jet Propulsion Laboratory, uses remote sensing for forestry science in the Africa Mapping Project. The tropical forest of Africa is 18% of the world's total and covers over 20 million square kilometers of land. The region faces deforestation with various degrees of intensity. The pace of deforestation varies from one country to another and accurate deforestation data are needed to understand forest changes and their causes. Saatchi uses high-resolution satellite imagery to map areas of forest clearing and types of land cover for the entire central African tropical region to provide large-scale quantitative estimates of the extent of deforestation and other forms of land cover changes. (See www-radar.jpl.nasa.gov/africamap/index.html)
    Dr. Emilio Moran, University of Indiana, conducts research on the study of populations in the Amazon Basin, specifically those experiencing changes leading to shifts in their use of resources. Traditional methods of research do not apply because of the transitory nature of settlements in forested frontiers-with up to 85% of the population moving within the first decade of settlement. (See www.indiana.edu/~speaweb/fcltydir/r_moran.html)
    Forestry research based on available and planned satellite and aerial remote sensors to monitor forest inventory, yield, and wildlife habitat is being conducted at MSU. Dr. David Evans investigates the use of imaging Light Detection and Ranging (LIDAR) and digital multispectral frame cameras to improve forest inventory and monitoring. These systems provide forest resource analysts with new tools to measure species, individual tree height and crown measurements, stem counts, canopy closure, and characterizing internal stand structures. (See www.cfr.msstate.edu/forestry/SITL/SITL.htm)

Verifying the Utility of Remote Sensing Data
Scientists support verification and validation of remote sensing data for research in agriculture and forestry; as presented in a vignette of the work being conducted at Purdue University.
    Dr. Chris J. Johannsen is director of the Laboratory for Applications of Remote Sensing (LARS) at Purdue University. LARS is NASA's verification and validation site for agriculture. The site is part of the NASA Stennis Space Center's effort to develop a nationwide network of ground truth sites to test commercial sensors against performance specifications and customer requirements. The ground truth data base for agriculture consists of information for each agricultural field over 5 acres in size showing crop species, crop variety, planting date, plant row directions, and any cultural practices performed on the field. (See http://dynamo.ecn.purdue.edu/~biehl/LARS)

A Sample Commercial Application
An example commercial application is the decision support system that has been developed by Pacific Meridian for fire management practices. Wildfires burned close to 6 million acres in 1996, a 170% increase over the previous 5 years. Pacific Meridian Resources developed FIRE!, an ARC/INFO-based fire behavior model to help mitigate the effects of wildfires. Based on the FARSITE model, FIRE! integrates fuels and topographic data with weather, wind settings, and initial fuel moistures to predict forest fire behavior across both time and space. Pacific Meridian will have access to ESE measurements to support this fire management tool. (See www.pacificmeridian.com)

A Rising Capability to Meet Growing Needs
An expanding variety of remote sensing tools are becoming available to help the agriculture and forest products industries increase productivity to meet the demands of expanding economies and populations. Creative partnerships among NASA, academia, and industry are important catalysts in the innovations of these new tools.
    ESE is supporting applications research centers such as the Remote Sensing Technologies Center (RSTC) at MSU to conduct research in agriculture and forestry and wildlife. Through basic research in understanding the phenomenology within a broad range of disciplined areas, MSU has developed a highly characterized field laboratory capability. New remote sensing technologies and applications will be tested in agriculture and forestry uses through the assets and capabilities of the RSTC.

About the Authors:
Ronald J. Birk works at Intermap Technologies, Inc., in Bay St. Louis, Mississippi. Dr. Roger L. King, P.E., works in the Remote Sensing Technologies Center at Mississippi State University in Mississippi State, Mississippi.

Back