A dynamic estimate of a soil wetness index for the Mackenzie River Basin from SSM/I measurements

This work aims at providing dynamic estimates of a soil wetness index in the Mackenzie River Basin, in northwest Canada. The method used in this study is based on the Basin Wetness Index (BWI), which is computed using the brightness temperature remotely sensed by SSM/I in the 19, 37 and 85 GHz channels and the soil surface temperature. In its basic formulation, the BWI uses two empirical parameters that are constant in both time and space. The basin heterogeneity and the temporal evolution of the vegetation state suggest that these parameters could vary. An alternative approach is proposed that allows for a reassessment of the empirical constants at the reception of each new image. The index was computed on a daily basis for the summer season of 1999, pixel wise (625 km/sub 2/) over the entire surface of the Mackenzie Rive Basin (1.8/spl times/10/sup 6/ km/sup 2/), which roughly comprises 20% of Canada. The BWI estimates reliability was assessed with a combination including in-situ measurements and hydrological modeling.