Abstract

Estimation of land surface evapotranspiration (ET) is one of the most difficult tasks in the field of hydrology and water resources. In this study, ET in the Golmud Region is estimated using remote sensing techniques, which own the characteristics of high efficiency, precision, large-scale and visualization. The Golmud Region has a specific geographic feature and it is an important mineral production base in China. However, the health of ecosystem in this region is just enduring severe water resources stress. Surface energy balance system model being an applicable model with good theoretic basis to estimate land surface ET with less meteorological data requirement, and the improved model is used to estimate land surface ET and analyzes the relationship with the water level in the Golmud Region on the basis of MODIS data. Results show that annual ET is single-peak distributed in month scale with a well seasonal variation. Spatially, Land surface spatial distribution of ET from south to north, from high altitude to low altitude mountain plain, ET following a decrease in the ground water table, so come to that the groundwater depth is the most important controlling factor of ET. The improved model is also suitable for other similar areas.

Key words: Land surface evapotranspiration, remote sensing, MODIS, Golmud Region