2008 Joint Meeting of The Geological Society of America, Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Gulf Coast Association of Geological Societies with the Gulf Coast Section of SEPM

Paper No. 7
Presentation Time: 8:00 AM-6:00 PM

Soil Property Mapping using Legacy Data and Pedometric Techniques: A Case Study Approach

THOMPSON, James A., Division of Plant and Soil Sciences, West Virginia University, Morgantown, WV 26506-6108, PERRY, Charles, USDA-Forest Service, 1992 Folwell Avenue, St. Paul, MN 55108, DEGLORIA, Stephen, Crop and Soil Sciences Department, Cornell University, 232 Emerson Hall, Ithaca, NY 14853-5601, PRESCOTT, Timothy, USDA-Natural Resources Conservation Service, 75 High St., Room 301, Morgantown, WV 26505 and MOORE, Amanda, USDA-Natural Resources Conservation Service, 1117 University Avenue #406, Morgantown, WV 26505, james.thompson@mail.wvu.edu

There are various methods of spatial interpolation and extrapolation that have been used to develop estimates of soil properties. Soil property maps are produced at many scales, derived primarily from generalized soil maps, with most property estimates generated for regional, national, or global applications. The objective of this study is create detailed soil property maps for multiple locations across the Northeast and North Central United States by developing generalized models using both point and polygon data sources. Point measures are derived from NRCS Soil Survey Division pedon data and Forest Service Forest Inventory and Analysis (FIA) program sample data. Polygon data come from the NRCS US General Soil Map (US GSM, or STATSGO2) and the Soil Survey Geographic Database (SSURGO). Various pedometric techniques, such as multivariate linear regression and regression trees, are used to develop statistical models from point data and complementary environmental covariate data. Soil property predictions from polygon data are generated using (i) measure-and-multiply approaches and (ii) spatial disaggregation techniques using environmental covariates. These two spatial predictions are combined to produce raster soil property maps. These efforts are part of a digital soil mapping initiative to develop global maps of selected soil properties specifically for the North American continent.