SOIL MOISTURE ESTIMATION USING REMOTE SENSING & GIS TECHNIQUES

Abstract

With the increasing population, there is a requirement to increase crop production and productivity, which leads to increasing demand for irrigation water. So, there is a need to use optimum water in irrigation, and for this, an accurate estimation of irrigation requirements is necessary and helps increase the irrigation efficiency. With the advent of satellite technology, remote sensing methods are continuously being used for soil moisture estimation. This study is focused on field-level soil moisture estimation using microwave remote sensing techniques for wheat crops. The study was carried out at the Nagla-aimad village. Initially, several wheat fields were identified, and then soil moisture data was observed with a Time Domain Reflectometry (TDR) instrument for 12 days in synchronization with Sentinel-1 at two depths for the entire crop period. Secondly, a semi-empirical Water cloud model(WCM) and its modified version were used, proving its accuracy in soil moisture estimation as the literature suggests. Sentinel-1A SAR-C data were used for extracting the backscattering coefficient, and Sentinel-2 data were used to derive the vegetation index of the study area. Further, localized models were prepared for each date to check the accuracy of the selected methodology with different vegetation indices. The localized models showed a good correlation of soil moisture with the co-polarized backscattering coefficient derived from the Sentinel-1 satellite imageries. The entire data is assembled and grouped in two groups for model calibration and validation for the two different depths. The linear regression approach was used to determine the parameters, and the model performance was evaluated using four various statistical indicators. For surface soil moisture (SSM), results showed that the model performed satisfactorily with R2 and RMSE of 0.6 and 5.84%, respectively, for the model calibration period and 0.55 and 6.36% for validation. The model's overall performance shows a good agreement between the observed and estimated soil moisture. Similarly, the model was developed for 20cm depth, and the results do not show a very good agreement between the observed and estimated soil moisture. Finally, soil moisture maps were prepared in the GIS platform for the collected data to know the Spatio-temporal variation of soil moisture in the area. It can be concluded that the developed model for the study area can be used for soil moisture monitoring and irrigation scheduling over the wheat crop. Furthermore, this study will help the policymakers for allocation of water within the command area.