The estimation of evapotranspiration(ET) over heterogeneous land surface of arid and semi-arid regions is quite complicated and significant. By using NOAA/AVHRR remote sensing, NCEP reanalysis grid and meteorological station data, Li xingming from CAEERI estimated daily ET over the Heihe River basin based on the integrating surface energy algorithm by land model and the Penman method.

As for clear days, remote sensing model is used to retrieval the instantaneous ET to daily ET. FAO-17 Penman method is also used to estimate the same day′s reference crop ET by NCEP reanalysis grid data and the meteorological station data. The relationship between actual ET and the reference crop ET can be used to calculate the actual ET based on the results from FAO-17 Penman methodon the cloudy days.

These results are validated by the observation data and other study in Heihe River basin. Integrate the remote sensing data and grid data can enhances the resolution of the NCEP grid data from 0.25 degree to 1.1 km by remote sensing data, and also make up deficiency of the remote sensing data on the cloudy days. This work will provide reference for similar research in this region.