Bridging the GAP between modis and fluxnet: validation of new high spatial resolution satellite-based estimates of evapotranspiration using fluxnet observations water fluxes from the land surface to…

Bridging the GAP between modis and fluxnet: validation of new high spatial resolution satellite-based estimates of evapotranspiration using fluxnet observations water fluxes from the land surface to the atmosphere play a critical role in climate system and knowledge of their spatial and temporal variation is essential to a wide range of hydrological, ecological applications, and water management applications. Satellite remote sensing provides unparalleled opportunity for providing consistent estimatesof evapotranspiration (ET) across local, regional and global scales. Although ET cannot be directly determined by remote sensing, significant recent advances have been made in models that translate satellite observations to estimates of ET. Validation of these models remains problematic, primarily due to the mismatch in spatial scales between the satellite observations and the ground-based measurements of ET that precludes direct comparison of the satellite estimates to field measurements. Model intercomparisons are informative yet there is still the need to validate the models with actual flux measurements by tower-based eddy covariance measurements and catchment water balances. The latter are at a suitable spatial scale for comparison with remote sensing estimates, but are limited in temporal resolution to annual time scales. Tower fluxes have high temporal resolution but are not at a comparable spatial scale as the remote sensing estimates. We propose a remote sensing ET methodology using our previously validated model (fisher ET AL. 2008, 2009) driven entirely with modis data at a spatial scale that would allow direct validation using the fluxnet tower network of ET observations. All remote sensing ET methods are currently limited by the 20-25KM spatial resolution of the required net radiation (RN) which is too coarse for validation with ~1KM tower flux measurements. We propose to estimate RN using newly developed methods (RYU ET AL. 2008, bisht and bras in press) that can be driven entirely with modis remote sensing observations. These 1KM all-sky estimates of RN, combined with modis-based estimates of air temperature (tair), water vapor pressure (EA), ndvi and EVI, will allow the first direct evaluation of a global ET methodology using in situ flux measurements throughout the fluxnet network.