Estimating the water budget of major US river basins via remote sensing

Abstract Nine satellite-based products, each of which provides information about land surface water budget terms, are used to estimate seasonal and annual variations in the water budget of the major river basins of the conterminous USA from 2003 to 2006. The remotely sensed terms are compared with gridded gauge precipitation, and estimates of evapotranspiration (E) and total water storage (TWS) derived from the Variable Infiltration Capacity (VIC) macroscale hydrology model. Among the remote sensing estimates, precipitation has the largest uncertainties. In general, apparent errors for E and TWS show substantial spatial variations, but the consistencies among these remote sensing products are greater than among precipitation products, possibly due in part to similarities in methodology, especially for TWS. Inferred run-off (as a residual of remote sensing estimates of precipitation, E, and TWS) is generally overestimated, due both to excessive precipitation and underestimation of combined E and terrestrial water storage change (TWSC) from remote sensing. Acknowledgements The authors would like to thank Dr Donald P. Chambers of the University of Texas at Austin for sharing the Gaussian smoothing technique used in connection with postprocessing VIC TWS estimates, and Drs Qiaozhen Mu and Steven Running at the University of Montana for providing their E product. We are also grateful to Dr James S. Famiglietti at University of California at Irvine for advice regarding use of the GRACE data, and to Dr Yang Hong at Oklahoma University for advice regarding the remotely sensed precipitation products. We also thank Dr Balaz Fekete and an anonymous reviewer for review comments that we believe improved the manuscript and Elizabeth Clark for her general comments and editorial suggestions. The GRACE data were processed by Dr Donald P. Chambers, supported by the NASA Earth Science REASoN GRACE Project, and are available at http://grace.jpl.nasa.gov. This work was supported by National Aeronautics and Space Administration under Grant NNG06GD79G to the University of Washington, and to Princeton University by subcontract.