Evaluation of ERA5, ERA5-Land, GLDAS-2.1, and GLEAM potential evapotranspiration data over mainland China

Abstract

Study region: China. Study focus: Accurate estimation of potential evapotranspiration (PET) is essential for understanding climate change. Using ground-based pan evaporation measurements over continental China, the monthly scale PET data during 2000–2017 of ERA5, ERA5-Land, GLDAS-2.1/Noah, and GLEAM V3.8a are evaluated, from the perspectives of their consistency in spatiotemporal variation, and performance measures. Factors controlling the data quality of the four datasets are investigated from the perspective of their PET calculation models and meteorologically input data. New hydrological insights for the region: PETERA5 performs the best in mainland China among four gridded PET datasets with higher correlation coefficients (r) and smaller biases, which can well capture the temporal variation of Epan. The outstanding performance of PETERA5 in China mainly results from the utilization of the Penman-Monteith (P-M) equation which performs the best among several competing formulas for PET computation, as well as its better meteorological inputs for computing PET than other datasets. Although the PETERA5-Land is a replay of the land component of the ERA5 climate reanalysis, it exhibits substantial overestimation of PET values and temporal trends, particularly in coastal areas of Southern China and the eastern side of Northeastern China, mainly caused by the overestimation of its net radiation. The PETGLDAS shows significant overestimation, partly due to its overestimation of wind speed, but mostly due to its modified P-M equation with its parameterization of land surface conditions for computing PETGLDAS. The PETGLEAM underestimated PET generally mainly due to the joint effect of the use of the Priestley-Taylor equation with small P-T parameter α, and the underestimation of the net radiation input from ERA-Interim, especially in Northwest and Qinghai Tibet.

Keywords

• Potential evapotranspiration
• ERA5
• GLEAM
• GLDAS
• Model algorithm
• Meteorological input