Evaluation and Drivers of Four Evapotranspiration Products in the Yellow River Basin
Lei Jin,Shaodan Chen,Haibo Yang,Chengcai Zhang
DOI: https://doi.org/10.3390/rs16111829
IF: 5
2024-05-22
Remote Sensing
Abstract:Evapotranspiration is a key driver of water and energy exchanges between terrestrial surfaces and the atmosphere, significantly influencing ecosystem balances. This study focuses on the Yellow River Basin (YRB), where evapotranspiration impacts both ecological dynamics and human activities. By analyzing actual evapotranspiration data from 1982 to 2017, this research provides insights into its spatial and temporal patterns within the YRB. Furthermore, a comprehensive assessment and comparative analysis were performed on four distinct evapotranspiration product datasets: GLDAS-Noah, ERA5-Land, GLEAM v3.8a, and MOD16A2. Employing the Geodetector model, the research identified seven key influencing factors—the digital elevation model (DEM), slope, aspect, precipitation, temperature, soil moisture, and normalized difference vegetation index (NDVI)—and analyzed their impact on evapotranspiration variations, yielding the following insights: (1) Based on the monthly-scale actual evapotranspiration dataset from 1982 to 2017, the annual average evapotranspiration in the YRB fluctuated between 375 and 473 mm, with an average value of 425 mm. A declining trend in the region's overall evapotranspiration was discerned using the Theil–Sen median slope estimator and Mann–Kendall trend test. (2) The datasets from GLDAS-Noah, ERA5-Land, and GLEAM exhibited the highest correlation with the observed datasets, all exceeding a correlation coefficient of 0.96. In contrast, the MOD16A2 dataset showed the least favorable performance. The ERA5-Land dataset was particularly noteworthy for its close alignment with observational benchmarks, as evidenced by the lowest recorded root mean square error (RMSE) of 5.09 mm, indicative of its outstanding precision. (3) Employing the Geodetector model, a thorough analysis was conducted of the interactions between evapotranspiration and seven critical determinants. The findings revealed that precipitation and the NDVI were the most significant factors influencing evapotranspiration, with q-values of 0.59 and 0.42 in 2010, and 0.71 and 0.59 in 2015, respectively. These results underscore their pivotal role as the main drivers of evapotranspiration variability within the YRB. Conversely, the q-values for slope in 2010 and 2015 were only 0.01 and nearly zero, respectively, indicating their minimal impact on the dynamics of evapotranspiration in the YRB.
environmental sciences,imaging science & photographic technology,remote sensing,geosciences, multidisciplinary