Performance evaluation of WRF-Noah Land surface model estimated soil moisture for hydrological application: Synergistic evaluation using SMOS retrieved soil moisture

Abstract

Summary This study explores the performance of soil moisture data from the global European Centre for Medium Range Weather Forecasts (ECMWF) ERA interim reanalysis datasets using the Weather Research and Forecasting (WRF) mesoscale numerical weather model coupled with the Noah Land surface model for hydrological applications. For evaluating the performance of WRF for soil moisture estimation, three domains are taken into account. The domain with best performance is used for estimating the soil moisture deficit (SMD). Further, several approaches are presented in this study to evaluate the efficiency of WRF simulated soil moisture for SMD estimation and compared against Soil Moisture and Ocean Salinity (SMOS) downscaled and non-downscaled soil moisture. In this study, the first approach is based on the empirical relationship between WRF soil moisture and the SMD on a continuous time series basis, while the second approach is focused on the vegetation cover impact on SMD retrieval, depicted in terms of growing and non-growing seasons. The linear growing and non-growing seasonal model in combination performs well with the NSE = 0.79, RMSE = 0.011 m; Bias = 0.24 m, in comparison to linear model (NSE = 0.70, RMSE = 0.013 m; Bias = 0.01 m). The performance obtained using WRF soil moisture is comparable to SMOS level 2 product but lower than the downscaled SMOS datasets. The results indicate that methodologies could be useful for modelers working in the field of soil moisture information system and SMD estimation at a catchment scale. The study could be useful for ungauged basins that pose a challenge to hydrological modeling due to unavailability of datasets for proper model calibration and validation.