Evaluation of long-term spatio-temporal soil moisture variation over the lower Mahanadi River Basin of Odisha, India: a case study

The aim of the current study is to simulate the long-term spatiotemporal soil moisture variation in the lower Mahanadi basin using the Soil and Water Assessment Tool (SWAT) model. The model was calibrated and validated using stream discharge data for the periods from 2005 to 2012 and 2013 to 2017, respectively. The coefficient of determination (R2) and Nash-Sutcliffe-Efficiency (NSE) values were 0.81 and 0.79 during calibration, and 0.78 and 0.74 during validation. The spatiotemporal annual and seasonal soil moisture content maps were prepared sub-basin-wise for the study period (2005-2017). The depth of the simulated soil moisture content varies with the rainfall, land use, and soil types of the lower Mahanadi basin. The upper region of the lower Mahanadi basin shows a higher simulated soil moisture content compared to the downstream region. The amounts of average annual rainfall in agricultural land, barren land, deciduous forest, and wasteland were 416.25, 439.06, 403.04, and 409.90 mm, respectively, which correspond to 118.18, 111.48, 126.78, and 121.50 mm/m of soil moisture content. The seasonal soil moisture maps showed that after harvesting the kharif rice crop, a sufficient amount of soil moisture was available during the post-monsoon season. Therefore, short-duration pulses and oilseed crops are recommended in this region to utilize the residual simulated soil moisture content, which could bring unutilized areas into cultivation and enhance farmer's income. Further, the simulated soil moisture content was compared with the satellite-derived SMAP-based soil moisture content, and a reasonably good agreement was found between the observed and simulated soil moisture content. The overall result showed that the SWAT model can reasonably simulate the spatiotemporal variation of soil moisture content.