Spatiotemporal trends and evapotranspiration estimation using an improvised SEBAL convergence method for the semi-arid region of Western Rajasthan, India

Abstract

The study demonstrates how to estimate evapotranspiration (ET) for the Western Rajasthan region of India utilizing remotely sensed images with the Surface Energy Balance Algorithm for Land (SEBAL). Landsat 8 and Moderate Resolution Imaging Spectroradiometer (MODIS) satellite inputs were used to compute seasonal and annual ET on the Google Earth Engine platform. The assessment utilizing the SEBAL algorithm, in combination with the Food and Agriculture Organization (FAO) Penman–Monteith (PM) and Hargreaves methods, demonstrates that SEBAL has adequate reliability for estimating ET for a spatially large extent in semi-arid regions when evaluated with the Hargreaves method. The values of R2, root-mean-square error (RMSE), and mean bias error (MBE) for FAO-PM were 0.63, 1.65 mm/day, and 1.28 mm/day, respectively. For the Hargreaves method, the values of R2, RMSE, and MBE were 0.96, 0.41 mm/day, and −0.31 mm/day, respectively. The study showed that the northern region witnessed the highest ET due to the availability of abundant surface water for irrigation. Overall, the results demonstrate the SEBAL model's effectiveness in estimating evapotranspiration. A downward trend in ET is observed in the region, mainly due to changing climatic conditions. HIGHLIGHTS Improvised SEBAL model integrated with GEE demonstrates satisfactory performance for regional scale ET estimation.; Hargreaves method performed relatively better for selected semi-arid to arid study area.; Study demonstrates an inverse correlation of LST with NDVI.; Improvised SEBAL model leads to faster and more efficient convergence of the iterative process.; Downward trend in pan evaporation–based ET observed.;

Keywords

• evapotranspiration
• google earth engine
• remote sensing
• sebal
• semi-arid