Kansas Water Science Center
Delineation and Evaluation of Hydrologic-Landscape Regions in the United States Using Geographic Information System Tools and Multivariate Statistical Analyses
By David M. Wolock, Thomas C. Winter, and Gerard McMahon
Hydrologic-landscape regions in the United States were delineated by using geographic information system (GIS) tools combined with principal components and cluster analyses. The GIS and statistical analyses were applied to land-surface form, geologic texture (permeability of the soil and bedrock), and climate variables that describe the physical and climatic setting of 43,931 small (approximately 200 km2) watersheds in the United States. (The term "watersheds" is defined in this paper as the drainage areas of tributary streams, headwater streams, and stream segments lying between two confluences.) The analyses based on similarities in land-surface form, geologic texture, and climatic characteristics. The percentage of explained variance (R-squared value) in an analysis of variance was used to compare the hydrologic-landscape regions to 19 square geometric regions and the 21 U.S. Environmental Protection Agency level-II ecoregions. Hydrologic-landscape regions generally were better than ecoregions at delineating regions of distinct land-surface form and geologic texture. Hydrologic-landscape regions and ecoregions were equally effective at defining regions in terms of climate, land cover, and water-quality characteristics. For about half of the landscape, climate, and water-quality characteristics, the R-squared values of square geometric regions were as high as hydrologic-landscape regions or ecoregions.
Wolock, D.M., Winter, T.C., and McMahon, Gerard, 2004, Delineation and evaluation of hydrologic-landscape regions in the United States using geographic information system tools and multivariate statistical analyses: Environmental Management, v. 34, p. S71-S88.
Additional information about watershed processes and modeling can be found at: http://ks.water.usgs.gov/pages/watershed-processes
To request a paper copy of this abstract, email: email@example.com