Soil Moisture and Ocean Salinity (SMOS) is a dedicated mission, which provides a ow of coarse resolution soil moisture data for hydrological applications. On the other hand Soil Moisture Decit (SMD) is a key indicator of soil water content changes and is valuable to a wider range of applications such as weather research, climatology, ood, drought, agricultural forecasting etc. In this paper, the SMOS Level 2 soil moisture was used for the estimation of SMD over the Brue catchment, Southwest of England, United Kingdom. Several approaches for estimation of SMD are performed using the Generalized Linear Model with di„erent families/link functions such as Gaussian/logit, Binomial/identity, Gamma/inverse and Poisson/log. ‡e overall performance obtained after all the techniques indicate that the Binomial with identity and Poisson with log link functions look promising for simulation of SMD from SMOS soil moisture with marginally high performance as compared to the other techniques.