|Kustas, William - Bill|
Submitted to: Water Resources Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/12/1998
Publication Date: N/A
Interpretive Summary: An energy balance model has been developed to use land cover and soil moisture estimates derived from remote sensing data for applications over land surfaces. With ancillary meteorological data the model is capable of simulating energy fluxes, including water use by vegetation and soil/substrate, over the daytime period with just a single daily remote sensing observation of soil moisture. The model was tested with data collected from a subhumid watershed. Good agreement between observations and model predictions of the fluxes was found. With satellite-based remote sensing systems the model has potential for mapping evapotranspiration over complex landscapes and the capability to monitor water use at watershed and regional scales. This information is important in order to assess the impact of agricultural management practices on local and regional climate and to address environmental impact issues. In addition, the ability to monitor water use over large regions will assist resource managers to assess the vegetation health of agricultural and natural ecosystems.
Technical Abstract: A two-source energy balance model was applied to remotely sensed near-surface soil moisture maps generated from passive microwave data collected during the Washita '92 Experiment conducted in the Little Washita watershed, a subhumid basin. Soil parameters for the model were derived from a soil texture data base. Vegetation parameters were derived from a combination of a land use/land cover data base and a vegetation index map created from a SPOT satellite image. Maps of midday Bowen ratio were created using the microwave images. The daily images of the Bowen ratio indicated areas with low vegetation cover or senescent vegetation which were drying out significantly while other areas with higher vegetation cover showed smaller increases in response to a drying soil surface. This result agrees with the surface flux observations which showed minor changes in the Bowen ratio for the experimental period. The reliability of the model was evaluated by comparing predicted daytime fluxes to observed daytime fluxes from four sites. Differences between modeled and observed fluxes were within 25%, on average. Model simulated surface temperatures were compared to sites containing ground-based observations. Differences between modeled and measured surface temperatures averaged approximately 2 K. These results indicate that the two-source model can reliably map surface fluxes at large scales using remotely sensed surface soil moisture under the conditions that existed during Washita '92.