SPATIAL VARIABILITY CONSTRAINTS TO MODELING SOIL-WATER AT DIFFERENT SCALES

There is increasing interest in modeling soil water content over relat ively large areas or scales. In general, the spatial variability of so il water content increases with scale, but it is not known how much or at which scales. High spatial variability constrains soil water model s by reducing the accuracy of input parameters, calibration and verifi cation data. It may also require representation of soil water in a spa tially distributed manner. Soil water content data were collected at t he Reynolds Creek Experimental Watershed at scales ranging from 12 m(2 ) to 2.3 x 10(8) m(2) to determine how scale affects spatial variabili ty. We found significant spatial variability at the 12-m(2) scale, whi ch could be described as random in large-scale models. The increase of spatial variability with scale was controlled by deterministic 'sourc es' such as soil series and elevation-induced climatic effects. The sa tellite-derived, soil-adjusted vegetation index showed that spatial va riability at the scale of Reynolds Creek (2.3 x 108 m(2)) is not rando m, and may have abrupt transitions corresponding to soil series. These results suggest a modeling strategy that incorporates soil series cha racterized by random spatial variability nested within the larger, ele vation-induced climatic gradient. The distinctions between soils and e levations is greatest early in the growing season and gradually dimini sh as the effects of differential precipitation and snowmelt timing ar e erased by evapotranspiration until late in the summer, when they vir tually disappear. These conclusions are landscape-dependent, so that r epresentation of spatial variability should be an explicit part of mod el development and application. (C) 1998 Elsevier Science B.V. All rig hts reserved.