ADJUSTING FOR BACKGROUND SOIL EVAPORATION WHEN INTERPOLATING EVAPOTRANSPIRATION BETWEEN SATELLITE OVERPASS DATES

Satellite based surface energy balance modes are now routinely operated to produce evapotranspiration (ET) products on an operational basis for use in water resources management. To produce estimates of ET at field scale Landsat satellite imagery is commonly used. The Landsat imagery is well suited for surface energy balance estimations because of the onboard thermal imager and the high resolution. Two Landsat satellites are currently in operation, Landsat 5 and Landsat 7, each flying in the same orbit but eight days apart. The return time for both satellites is every 16 days.

Because of cloudiness and other atmospheric disturbances including jet contrails and smoke, not all images or portions of images are suited for processing. Although cloud cover varies between different areas, it is common to find only one good image per month suited to be processed. Based on this one snapshot of ET the monthly and ultimately seasonal ET are determined by interpolating a relative ET fraction (ETrF) between image dates using for example a cubic spline function and multiplying by a reference ET calculated from weather data for each day of the month. A potential shortfall when performing the interpolation is whether local precipitation events, such as afternoon summer rain storms are fully accounted for, or if these events unduly dominate the ETrF image. If such an event happens within a few days prior to the satellite overpass time, the residual evaporation soil may be overestimated when that image is used as a basis for estimating the monthly ET. The result is an error in the water balance.

We therefore adjust the ET_rF derived for the satellite overpass date from the METRIC surface energy balance model for background evaporation from soil caused by rainfall so that the final ET map represents average conditions for the month. The method currently used for this this adjustment is to establish a daily soil water balance for bare soil for each day of the month and ratio the average evaporation over the month from bare soil to the evaporation on the image date. This ratio is modified to account for shading effects of vegetation using the Normalized Difference Vegetation Index (NDVI) = NDVI_{bare soil} , with no adjustment made for areas with full vegetation cover, represented by NDVI = NDVI_{full cover}.