Publications

Bian, J; Hu, XL; Shi, LS; Min, LL; Zhang, YC; Shen, YJ; Zhao, FH; Zha, YY; Lian, X; Huang, JS (2024). Improving the evapotranspiration estimation by considering the effect of flux footprint climatology. JOURNAL OF HYDROLOGY, 631, 130769.

Abstract
Eddy covariance (EC) is a common approach for observing evapotranspiration (ET). In the EC system, there is a varying flux footprint climatology which causes a spatial mismatch between the forcing data from the ET model and EC observations, generating extra errors in estimating ET. This study endeavored to reduce the error caused by the spatial scale inconsistency between forcing data and ET measurements employing joint observations from EC system, unmanned aerial vehicle (UAV), and satellites under the flux footprint theory and the Penman-Monteith (P-M) model. The joint observations were performed in 2020 at the LuanCheng (LES) and YuCheng (YES) sites. Surface conductance for the varying flux footprint climatology was retrieved from EC observations using the inverted P-M model. The vegetation indices from various remote sensing platforms, including UAV, MODIS, Landsat 8, and Sentinel-2, were retrieved and employed to evaluate surface conductance and ET within the flux footprint climatology. The findings indicated that varying flux footprint climatology had a significant impact on the accuracy of ET estimation. The UAV, Landsat 8, and Sentinel-2 data all generated good ET estimations (LES: R-2 > 0.72, RMSE < 0.81 mm day(-1); YES: R-2 > 0.74, RMSE < 0.91 mm day(-1)) in the 80 % and 90 % flux footprint climatology. In contrast, MODIS, Landsat 8, and Sentinel-2 obtained worse ET performance at LES (R-2 < 0.65, RMSE > 0.88 mm day(-1)) and YES (R-2 < 0.67, RMSE > 0.90 mm day(-1)) when flux footprint climatology was disregarded. Among the four remote sensing platforms, UAV remote sensing images integrated with the flux footprint model significantly enhanced the accuracy of ET estimation (LES: R-2 = 0.83, RMSE = 0.66 mm day(-1); YES: R-2 = 0.84, RMSE = 0.55 mm day(-1)). This study demonstrated that considering flux footprint climatology can acquire accurate ET estimates.

DOI:
10.1016/j.jhydrol.2024.130769

ISSN:
1879-2707