Remote Sensing Approach for Estimating Evapotranspiration Using Satellite-Based Energy Balance Models in Al Hamra, Oman

Abstract

AbstractThe conventional physical- and climatological-based models estimating actual evapotranspiration (ETa) do not provide accurate ETa, especially for regional-scale which is a challenge that needs to be tackled for the accurate assessment of irrigation practices. In this study, two energy balance models (1) Surface Energy Balance Algorithm for Land (SEBAL) and (2) Mapping Evapotranspiration at High Resolution with Internalized Calibration (METRIC) were applied in the Al Hamra region of Oman to estimate ETa using Landsat 8 satellite imagery. The aim of the study was to estimate ETa using the SEBAL and METRIC model for the period of two years and validate the outcome against modified Penman–Monteith (PM) model. In addition, this study highlighted procedural differences between METRIC and SEBAL models. The results showed that METRIC model estimated ETa successfully with a R2 value of 0.81 as compared to SEBAL model (R2 = 0.17). Study validated against PM model showed that the METRIC model outperformed the SEBAL with MAE and RMSE values of 0.30 and 0.36 mm as compared to 2.67 and 1.42 mm, respectively. Study concluded that the SEBAL model used cumulative daily net radiation, while METRIC used cumulative reference evapotranspiration to extrapolate ETa on daily basis. Although METRIC performed well than SEBAL, the manual selection of hot and cold pixel in METRIC was a difficult process and it creates vulnerability of inaccurate estimates of ETa. However, from this study, we can conclude that the METRIC model can be useful in providing high spatial and temporal ETa estimates. Furthermore, the approach used in this paper can be generalized and pave the way for further developments of the studied models.KeywordsRemote sensingSEBALMETRICLandsat 8Energy balance