Spatial Patterns in Surface Energy Balance Components Derived from Remotely Sensed Data

Abstract

A relatively simple modeling approach for estimating spatially distributed surface energy fluxes was applied to two small watersheds, one in a semi-arid climate region and one in a sub-humid region. This approach utilized a combination of ground-based meteorological data and remotely sensed data to estimate ‘instantaneous’ surface energy fluxes at the time of the satellite or aircraft overpasses. The spatial resolution in the watershed grid cells, which was on the order of 100-400 km, was selected to be compatible with ground measurements used for validation. The model estimates of surface energy fluxes compared well with ground-based measurements of surface flux (typically within approximately 40 Wm−2). The model accuracy may be slightly less for bare soil surfaces due to an overestimation of the soil heat flux. In addition to demonstrating the feasibility of computing spatially distributed values of surface energy fluxes, these maps were used to qualitatively infer the dominant factors controlling the energy fluxes for the time period shortly following precipitation events in the basins. For the semi-arid watershed, values of sensible heat flux varied considerably over the watershed and displayed a pattern very similar to that of the spatially variable cumulative precipitation for at least one to eight days prior to the image acquisition. Due to the large fraction of exposed bare soil in a semi-arid ecosystem, even very small precipitation events had a strong influence on the pattern of sensible heat fluxes observed shortly after the event (less than 24 hours). For the sub-humid watershed, the fluxes tended to be more uniform across the watershed, and were influenced by a combination of precipitation total and land cover type.