Evaporation from water bodies (Ew) is a critical component of the global water cycle. However, existing evaporation products that include Ew often suffer from drawbacks such as coarse resolution, short time span, and high uncertainty. This study developed a 60-year (1960–2019) high-resolution (0.05º×0.05º) evaporation dataset for small shallow water bodies in China based on the Penman model. Two key factors affecting the accuracy of the Penman model were considered: the uncertainty of the empirical wind function and changes in heat storage in the water body. Specifically, we used large-size (20 or 100 m2) pan evaporation (Epan) observations from 21 sites as a benchmark to correct the wind function of the Penman model. A data-driven model was then developed to map the spatial distribution of the wind function coefficients across China. The corrected wind function significantly improved the accuracy of Epan estimates compared to the original wind function, with the Kling-Gupta efficiency (KGE) increased by 0.05∼0.10. To model the effect of heat storage changes on evaporation, an equilibrium temperature method was used. We also introduced an area-dependent scaling factor into the wind function to account for the effect of water body's size on Ew estimation. The reliability of the Ew algorithm was tested on two lakes using eddy-covariance flux observations, and simulations showed good agreement with observations. The Epan (20 m2 pan) dataset and its two components calculated from the radiative and aerodynamic terms of the Penman model can be accessed at https://osf.io/qd28m/. Users can utilize the two Epan components and the area-dependent scaling factor to estimate evaporation for water bodies of varying sizes. However, caution is needed when applying this dataset to deep water bodies, as it is designed for shallow water bodies.
Read full abstract