Resources Science ›› 2020, Vol. 42 ›› Issue (10): 2035-2046.doi: 10.18402/resci.2020.10.19

Previous Articles    

Verification and comparison of three high-resolution surface evapotranspiration products in North China

HE Shaoyang1,2(), TIAN Jing1(), ZHANG Yongqiang1   

  1. 1. Key Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2020-05-06 Revised:2020-08-11 Online:2020-10-25 Published:2020-12-25
  • Contact: TIAN Jing;


The North China Plain is a critical agricultural region in China and one of the most severe water deficient areas in the world. Surface evapotranspiration (ET) is the largest component of water resource consumption, therefore obtaining accurate ET data is an important basis for water resource management on the North China Plain. In this study, accuracy verification and spatiotemporal comparison of three global high-resolution ET products were conducted in North China in order to provide a reference for the selection of a high-resolution ET data product that is more suitable for the North China Plain and can better serve the purpose of water resource research and management. Through the comparison with the eddy correlation measurement, the research showed that the PML_V2 product had the highest accuracy in North China, followed by SSEBop_V4 and MOD16A2, with correlation coefficients of 0.81, 0.74, and 0.52, respectively. The root mean square errors were 0.87, 1.52, and 1.44 (mm/d), respectively. PML_V2 showed the highest consistency with the fluctuation trend observed at the site. The correlation between the estimated and observed ET of the three products in the growing season of wheat was higher than that of maize. SSEBop_V4 ET and PML_V2 ET estimates had the highest correlation with the observed ET in wheat season and maize season, respectively. Comparatively, PML_V2 and SSEBop_V4 are more similar in spatial distribution, with the highest correlation coefficient of 0.76. The spatial distribution of MOD16A2 is very different from that of the other two products. The biggest difference of the three products appears in the cultivated land area. In 2003-2018, MOD16A2 showed a clear trend of increase for three land use types, while SSEBop_V4 and PML_V2 showed no obvious change.

Key words: evapotranspiration, high resolution, accuracy verification, spatiotemporal change, eddy covariance, Google Earth Engine (GEE), growing season, North China Plain