Aggregation Strategies for SSURGO Data: Effects on SWAT Soil Inputs and Hydrologic Outputs
- Sarah E. Gatzkea,
- Dylan E. Beaudettea,
- Darren L. Ficklina,
- Yuzhou Luob,
- Anthony T. O'Geenc and
- Minghua Zhang* *d
- a Dep. of Land, Air and Water Resources Univ. of California Davis, CA 95616
b Dep. of Land, Air and Water Resources Univ. of California Davis, CA 95616 and California Dep. of Pesticide Regulation 1001 St. Sacramento, CA 95814
c Dep. of Land, Air and Water Resources Univ. of California Davis, CA 95616
d Dep. of Land, Air and Water Resources Univ. of California Davis, CA 95616 and California Dep. of Pesticide Regulation 1001 I St. Sacramento, CA 95814
The USDA-NRCS Soil Survey Geographic (SSURGO) dataset provides the most comprehensive, detailed soil data coverage across the United States. Correct usage of these data within a hydrologic model depends on assumptions that define how soil property data are aggregated. To reduce data intensity and improve model efficiency, most hydrologic modeling studies using SSURGO assume that soil property data are adequately grouped into some notion of a “soil type” which is represented by the map unit, denoted as map unit key (MUkey), within SSURGO. However, the map unit is not intended for this purpose as continuity in map unit design or composition is not guaranteed between adjacent surveys of different vintages. This causes problems when several survey areas are used together, because similar soils are assigned a different map unit across the boundaries of soil survey maps. We present a methodology for aggregating soil data among multiple soil survey areas according to soil taxonomic information available in SSURGO. Results indicate that the aggregation method provides an acceptable representation of soil parameter values and distributions while eliminating the reliance on an arbitrary map unit for soil type identification. The results of the hydrologic modeling using the Soil and Water Assessment Tool (SWAT) in the San Joaquin River Watershed indicate that the commonly used aggregation method and the newly developed method satisfactorily estimated soil and surface hydrologic processes as compared to using a non-aggregated soil dataset. For the soil hydrologic processes, the SWAT model output from our aggregation method accurately estimated soil water content (mean difference compared to the non-aggregated soil dataset output of −4 mm for western San Joaquin River Watershed subbasins and 15 mm for the eastern San Joaquin River watershed subbasins) and lateral flow (3 mm for the western subbasins and 0.2 mm for the eastern subbasins) as compared to using a non-aggregated soil dataset. For the surface hydrologic processes, the SWAT model under predicted surface runoff (−0.5 mm for the western subbasins and −0.1 mm for the eastern subbasins) and sediment yield (−0.02 t/ha for the western subbasins and −9×10−4 t/ha for the eastern subbasins) as compared to using a non-aggregated soil dataset. While some variations were statistically significant, the differences were numerically small. The results show that soil taxonomy provides a robust framework for grouping soils.Please view the pdf by using the Full Text (PDF) link under 'View' to the left.
Copyright © 2011. Copyright © by the Soil Science Society of America, Inc.