Evaluating the Impact of Topography on Satellite-Derived Evapotranspiration Estimates in the High Atlas Mountains of Morocco

Abstract

Accurate evapotranspiration (ET) estimates in mountainous regions are needed for better understanding the hydrological cycle and managing water resources within watersheds. However, the complex topography of these areas can have significant effects on ET, making it challenging to monitor at all scales. In this study, we sought to improve the accuracy of thermal remote sensing-based ET estimates in the High Atlas region of Morocco by taking into account the effect of topography. To do this, we used two ET models, both driven by LANDSAT optical/thermal data: the Two-Source Energy Balance (TSEB) model and the contextual Water Deficit Index (WDI) model. The meteorological data (such as air temperature, wind speed, and humidity) used to force the models were taken from ERA5-Land reanalysis products and specifically disaggregated at 30 meters to account for elevation effects, while the solar radiation data were obtained using the Samani et al. method to consider sun exposure effects. We compared the ET estimates produced by both models to measurements taken at two Eddy covariance towers in the mountains at different elevations (900 and 3850 m.a.s.l). Our results showed that the TSEB model was able to accurately estimate ET in the region, with a high level of consistency (r² = 0.72, rmse = 43 Wm-2). The relative performance of both TSEB and WDI models was assessed. We also found that topography significantly influences ET in the High Atlas Mountains, emphasizing the importance of considering it when estimating ET at the watershed scale. This outcome can be used to better understand the hydrological cycle and manage water resources in mountainous areas.