Abstract
Accurate estimation of surface energy fluxes from space at high spatial resolution has the potential to improve prediction of the impact of land-use changes on the local environment and to provide a means to assess local crop conditions. To achieve this goal, a combination of physically based surface flux models and high-quality remote-sensing data are needed. Data from the ASTER sensor are particularly well-suited to the task, as it collects high spatial resolution (15–90 m) images in visible, near-infrared, and thermal infrared bands. Data in these bands yield surface temperature, vegetation cover density, and land-use types, all critical inputs to surface energy balance models for assessing local environmental conditions. ASTER is currently the only satellite sensor collecting multispectral thermal infrared images, a capability allowing unprecedented surface temperature estimation accuracy for a variety of surface cover types. Availability of ASTER data to study surface energy fluxes allows direct comparisons against ground measurements and facilitates detection of modeling limitations, both possible because of ASTER's higher spatial resolution.
Surface energy flux retrieval from ASTER is demonstrated using data collected over an experimental site in central Iowa, USA, in the framework of the Soil Moisture Atmosphere Coupling Experiment (SMACEX). This experiment took place during the summer of 2002 in a study of heterogeneous agricultural croplands. Two different flux estimation approaches, designed to account for the spatial variability, are considered: the Two-Source Energy Balance model (TSEB) and the Surface Energy Balance Algorithm or Land model (SEBAL). ASTER data are shown to have spatial and spectral resolution sufficient to derive surface variables required as inputs for physically based energy balance modeling. Comparison of flux model results against each other and against ground based measurements was promising, with flux values commonly agreeing within ∼50 W m
−
2
. Both TSEB and SEBAL showed systematic agreement and responded to spatially varying surface temperatures and vegetation densities. Direct comparison against ground Eddy Covariance data suggests that the TSEB approach is helpful over sparsely vegetated terrain.