Remotely Sensed ET Computed by SEBAL — Implications for Water Resource Planning in California

Abstract

Quantitative estimates of consumptive water use by crops, landscape, and natural vegetation are necessary for water resource planning. Availability of satellite remotely-sensed data offers an alternative to current estimating techniques. Evapotranspiration (ET) is predicted as a residual of the energy balance by the Surface Energy Balance Algorithm for Land (SEBAL) using routine weather data and satellite sensed thermal radiances. The California Department of Water Resources (CDWR) is required by state law to develop a state-wide water plan every five years. Computations of consumptive water use are an integral part of the water planning process. Remotely-sensed ET computed by SEBAL offers potential advantages compared to the current labor- and data-intensive estimates techniques. For 2002, annual actual ET for California was computed by SEBAL using MODerate-resolution Imaging Spectroradiometer (MODIS) satellite images (1 km by 1 km square pixels) and California Irrigation Management Information System (CIMIS) weather data. The annual ET of irrigated, agricultural lands computed by SEBAL is compared to the CDWR current water plan estimates. The hydrologic implications of these different estimates to the recommendations and options discussed in the plan are explored. The advantages, disadvantages and estimated accuracy of the current ET computation method and the remotely-sensed based SEBAL method are compared and contrasted.