Publications

Yu, B; Shang, SH; Zhu, WB; Gentine, P; Cheng, Y (2019). Mapping daily evapotranspiration over a large irrigation district from MODIS data using a novel hybrid dual-source coupling model. AGRICULTURAL AND FOREST METEOROLOGY, 276, UNSP 107612.

Abstract
Accurate knowledge of spatially distributed evapotranspiration (ET) is essential for better understanding the availability and utilization of water resources. Over the past several decades, remote sensing (RS) technique has been widely applied to acquire indispensable input data for the estimation of ET at different spatial and temporal scales. Several dual-source ET models that can partition evaporation and transpiration have been developed by utilizing remote sensing data and ancillary meteorological observations. However, due to the significant spatial variability of air temperature over heterogeneous land surface, these models typically cannot be applied in heterogeneous regions with scarce meteorological observations. In the present study, a new model, referred to as the HTEM-ABL model, is developed by coupling the Hybrid dual-source scheme and Trapezoid framework-based ET Model (HTEM) with a simple atmospheric boundary layer (ABL) model for ET estimation without ancillary air temperature observations. This model uses Moderate Resolution Imaging Spectroradiometer (MODIS) data as the main input and is evaluated with field observations over a large irrigation district in North China. Results show that the HTEM-ABL model can be well applied to the study area, and the accuracy of HTEM-ABL model is slightly superior to the original HTEM model. The root mean square errors between the calculated and measured ET are 0.63 mm/day and 29.6 mm/year at the field scale and the regional scale, respectively. Furthermore, the estimated air temperature has been proved to be accurate qualitatively and quantitatively, and provides another aspect to evaluate the HTEM-ABL model. The spatial and temporal distributions of ET produced by the HTEM-ABL model are in good agreement with both the distribution of land use types and the variations of crop growth stages. In addition, the HTEM-ABL model shows less sensitivity to land surface temperature compared with the original HTEM model.

DOI:
10.1016/j.agrformet.2019.06.011

ISSN:
0168-1923