Estimating surface energy fluxes, particularly evapotranspiration (ET), from satellite observations has proven to be a challenging task, because the single "snapshot" images routinely obtained from high-spatial-resolution satellites do not provide enough temporal information.A new two-step approach (called Disaggregated Atmosphere Land Exchange Inverse or DisALEXI) has been developed to combine the high temporal resolution of GOES (Geostationary Operational Environmental Satellite) with the high spatial resolution of Landsat to estimate crop ET on the 10 - 100 meter scale without requiring any local observations.

The first step uses surface brightness-temperature-change measurements about four hours apart in the morning from the GOES satellite to estimate average surface fluxes on the scale of about 5 km with an algorithm known as ALEXI. The second step disaggregates the GOES 5-km surface-flux estimates by using high-spatial-resolution images of vegetation-index and surface temperature, such as from ASTER, Landsat or aircraft, to produce high-spatial-resolution maps of surface fluxes.

We have applied the DisALEXI approach to disaggregate fluxes from the 5-10 km ALEXI results to 30 to 60 meters using LandSat vegetation and temperature data. The goal of this work is to validate the ALEXI and DisALEXI methods against actual Oklahoma Mesonet flux observations. Specifically, we attempt to isolate the LandSat pixel containing a Mesonet station for direct flux comparison. This application of DisALEXI has allowed us to properly validate our ALEXI methods, while evaluating the shortfalls of point land-surface flux measurements as collected by the Mesonet. Results are encouraging (root-mean-square difference again within ~50 Wm-2). Many of the discrepancies between 30-meter DisALEXI flux estimates and Mesonet fluxes associated with instruments calibrations (e.g, GOES and atmospheric corrections), flux "footprint" issues associated with weather condition variability, and non-closure of Mesonet fluxes caused again by variable weather and surface heating conditions. Such errors are typical of many flux measuring systems and procedures.

One example of DisALEXI is shown below. Other comparisons can be seen at http://kang.ssec.wisc.edu/~alexi/disagg/.