Estimation of actual evapotranspiration and water stress in the Lijiang River Basin, China using a modified Operational Simplified Surface Energy Balance (SSEBop) model

Abstract

Due to the simultaneous impacts of economic development and climate change, the Lijiang River Basin in China—the largest karst tourist attraction in the world—has experienced dramatic water shortages during the dry season. As actual evapotranspiration (${\mathrm{ET}}_a$) plays a critical role in the water cycle, accurate estimation of ${\mathrm{ET}}_a$ and water stress are important for sustainable water resources management. In this paper, we mapped the distribution of daily ${\mathrm{ET}}_a$ using a modified Operational Simplified Surface Energy Balance (SSEBop) model in combination with Landsat 8 images and assessed water stress using the Crop Water Stress Index (CWSI) during the dry season in the Lijiang River Basin. In general, the daily ${\mathrm{ET}}_a$ simulated by the SSEBop model with aerodynamic resistance value of 110 s m⁻¹ was higher than that of satellite-based actual evapotranspiration products (i.e., MOD16A2 and Penman-Monteith-Leuning (PML)_V2 actual evapotranspiration products in this study). Aerodynamic resistance plays a critical role in the estimation of energy fluxes in the SSEBop model and should be readjusted and calibrated with available datasets to improve the model’s performance in estimating actual evapotranspiration for particular regions. Readjusted values between 20 and 35 s m⁻¹ of aerodynamic resistance produced reasonable agreement with satellite-based actual evapotranspiration products in the Lijiang River Basin. In addition, insufficient ground-level measurements of actual evapotranspiration might have increased the uncertainty of the SSEBop model’s performance. The achievement of higher accuracy in the estimation of actual evapotranspiration and water availability will require establishing local flux towers, particularly in forested areas, to collect evapotranspiration, temperature and other in situ data. For different land-cover classes, forest areas exhibited the highest actual evapotranspiration, whereas farmland and built-up areas had the lowest actual evapotranspiration values compared to the other land-cover classes. All land-cover classes, especially farmland areas, experienced severe water stress. Inadequate precipitation as a result of climate change, combined with high actual evapotranspiration will result in less water being available for the Lijiang River Basin. Additional water is required to compensate for evapotranspiration and support plant growth in the Lijiang River Basin during the growing season.