Submitted to: Society of Range Management
Publication Type: Abstract Only
Publication Acceptance Date: December 28, 2004
Publication Date: January 24, 2005
Citation: Skirvin, S.M., Moran, M.S. 2005. Application of geospatial modeling and remotely sensed data for rangeland studies. Proc. 57 Annual Meeting of the Soc. for Range Manage., Jan. 23-30, Salt Lake City, UT, Oral Presentation. Interpretive Summary: Computer simulation models are useful for testing scientists' understanding of ecosystem processes: model results can be compared with data acquired under existing conditions, which helps scientists to understand if the model needs to be changed, or if some of the input values have to be modified. Ecosystem simulation models can also be used to compare alternative future outcomes of management decisions and their impacts on erosion and vegetation. Many existing models were written when computers were less powerful and fewer sources of data were available. One model for rangelands, SPUR (Simulation of Production and Utilization of Rangelands), was initially released in 1987 and has been updated and modified several times. The latest modification allows SPUR to use spatial data, such as maps and satellite remote sensing data that can cover large portions of the landscape, as well as other kinds of data that can only be measured over small areas. This version of SPUR, known as SESPUR (Spatially Explicit SPUR) is being tested in semi-arid rangeland at the Walnut Gulch Experimental Watershed in southeastern Arizona.
Technical Abstract: The increasing availability of geospatial data and tools, including geographic information systems (GIS) for display and analysis, is providing opportunities to expand the use of computer models of rangeland production. One such model is SPUR (Simulation of Production and Utilization of Rangelands), which was initially released in 1987 and has subsequently been revised for use in a number of areas in the U.S., including rangelands of the Great Plains and the Great Basin. SPUR can be used as a decision support tool to evaluate hydrologic and erosion changes in response to management decisions, as well as vegetation growth and utilization by grazing animals. A spatially explicit implementation of SPUR (SESPUR) has been developed at the USDA-ARS Southwest Watershed Research Center in Tucson, Arizona, and is under evaluation for use in semi-arid rangelands of the southwest. The goal of SESPUR development is to utilize available geospatial information for accurate simulation of existing conditions and for development of scenarios of potential future conditions. Current work is focusing on the integration of remotely sensed data such as Landsat and MODIS-based LAI and soil moisture, as well as distributed hydrologic information from other spatial models such as SWAT (Soil and Water Assessment Tool) and KINEROS. Ongoing calibration and validation of SESPUR is based on historic and newly acquired data for the Walnut Gulch Experimental Watershed near Tombstone, Arizona.