Using remote sensing data in macroscale hydrological modelling

There has been considerable focus on the estimation of the spatial and temporal variability of the hydrological and energy budgets of the land surface using macroscale hydroclimatic models. However, performing large-scale applications is greatly complicated by the scarcity of land surface observations needed to force them. Remote sensing offers a potentially powerful alternative to the use of ground observations that historically have provided the sole forcings. In addition, the use of remotely sensed information for data assimilation in such models has begun. In this paper we summarize our experience using a predominantly remote sensing approach to hydrological modelling. Results are presented from the recent NASA EOS Interdisciplinary Science project whose main objectives were to: (a) develop and test a land surface hydroclimatic model, the VIC model, capable of using remotely sensed data for forcing and assimilation; and (b) develop and test remote sensing algorithms appropriate for hydroclimatic modelling. We also present results from a modelling experiment run over the Ohio River basin where the VIC model is driven first by ground-based data, and then by remotely sensed data. Lastly variability comparisons of satellite-derived vs modelled land surface temperature are discussed.