Evaluation of modeled high resolution virtual brightness temperatures compared to space-borne observations for the neckar catchment

摘要

Synthetic space-borne brightness temperatures have been generated by exploiting CMEM radiative transfer simulations at L-band and land-surface-atmosphere coupled models. The community land model CLM coupled with the atmospheric model COSMO have been used in order to create a full representation of subsurface, soil and vegetation of a catchment located in the south-west Germany, as a virtual reality framework (VR). These VR simulations are used as forcing for CMEM to simulate high resolution 1.4GHz brightness temperature (TB) at H&V polarization. Satellite-like synthetic observations are then generated including foot-print, orbital path and incidence angle. The satellite synthetic TBs are compared with real observations to show the degree of agreement on magnitude and dynamic. It has been found that the synthetic TBs have values of 20 to 60 K less compared to the mean distribution of the real satellite observations within the catchment. This biases vary according to the season with summer exhibiting the maximum and winter the minimum discrepancy. The general bias is largely explained by the fact that the catchment's soil moisture is overestimated by 45% when compared with satellite retrievals. A reduction applied to SM of 55% yields a partial reduction on TB bias, yet not enough to match the real satellite data. The remaining discrepancy is attributed to radiative effects introduced by inappropriate vegetation parametrization and inadequate orography assumptions.