ABSTRACT Understanding the evapotranspiration of mountainous catchments is vital for the effective management of water resources. Evapotranspiration is affected by changes in both field conditions, such as changes in land cover and soil, and external conditions, such as climate. However, the magnitude and interaction of these effects is not well understood. The Shirasaka Experimental Watershed of Ecohydrology Research Institute, established in 1929, has undergone precise rainfall‐runoff and climate monitoring. During this period, the catchment experienced forest recovery following historical anthropogenic degradation and climate change. We first hypothesized that since the catchment has undergone substantial land cover changes, annual evapotranspiration has increased primarily due to changes in the land surface condition and climate change having a minor effect. To test this, we studied and measured changes in vegetation and soil, supplemented historical climate data such as precipitation, temperature, and solar radiation, calculated actual and potential evapotranspiration, and analysed long‐term trends using the Mann‐Kendall test. We then separated the effects of changes in land surface conditions and external conditions on annual evapotranspiration using catchment water and energy balances, as implemented in the Budyko model. The bare land is disappearing, and the stand volume and soil depth generally increased, yet evapotranspiration did not increase consistently, demonstrating that the hypothesis was rejected. Simultaneously assessing the impacts of land surface change and climate change demonstrated that relative impacts have changed with time. Between the mid‐1950s and the 1980s, annual evapotranspiration was about 50 mm lower than the long‐term mean and suggested that lower solar radiation and/or air pollution lowered evapotranspiration. While in the 2010s, changes in land surface conditions decreased evapotranspiration by more than 100 mm, although the climate conditions have changed in the direction of increasing evapotranspiration. Long‐term data show significant changes in watershed evapotranspiration caused by previously unconsidered factors, necessitating further investigation.
Bei et al. (Wed,) studied this question.