Separation Of Soil Evaporation And Vegetation Transpiration By MODIS Data For Central And Northern China

Evapotranspiration(ET) plays a crucial role in the hydrologic system. To estimate evapotranspiration quantitatively in a large scale, remote sensing data has been used in a number of models and shows its applicability in the estimation of evapotranspiration. In this paper, evapotranspiration for central and northern China was derived from MODIS data. In arid and semi-arid regions, soil evaporation can be considered as the minimum water requirement for bare area, while evapotranspiration can be considered as the minimum water demand for the area covered by vegetation. Hence the separation of soil evaporation and vegetation transpiration is valuable for efficient water resources management. In this study, the land surface temperature-fractional vegetation coverage(Ts-f) trapezoid method was applied in conjunction with an operational two-layer model. A modified algorithm for the determination of actual dry/wet edges(MADE) of the Ts-f trapezoid was proposed, which is an improvement of the original method based on Ts-VI(vegetation index) triangle developed by Ronglin Tang(2010). The MADE algorithm was then integrated with the two-layer model to estimate the latent heat flux (evaporation and transpiration). It’s showed that the retrieved latent heat flux is in good agreement with FLUXNET data obtained from Department of Biogeochemical Integration. The root mean square error of monthly ET is below 25 W/m2. The result demonstrated that the accuracy of the modified algorithm to determine dry/wet edges in the Ts-f trapezoid was satisfactory. Finally, the spatial and temporal distribution of soil evaporation and vegetation transpiration of central and northern China was further investigated in this study