Regional estimates of evapotranspiration (ET) are critical for a wide range of applications. Satellite remote sensing is a promising tool for obtaining reasonable ET spatial distribution data. However, there are at least two major problems that exist in the regional estimation of ET from remote sensing data. One is the conflicting requirements of simple data over a wide region, and accuracy of those data. The second is the lack of regional ET products that cover the entire region of northern China. In this study, we first retrieved the evaporative fraction (EF) by interpolating from the difference of day/night land surface temperature ( T ) and the normalized difference vegetation index (NDVI) triangular-shaped scatter space. Then, ET was generated from EF and land surface meteorological data. The estimated eight-day EF and ET results were validated with 14 eddy covariance (EC) flux measurements in the growing season (July-September) for the year2008 over the study area. The estimated values agreed well with flux tower measurements, and this agreement was highly statistically significant for both EF and ET (p <0.01), with the correlation coefficient for EF (R2 =0.64) being relatively higher than for ET (R2 =0.57). Validation with EC-measured ET showed the mean RMSE and bias were 0.78 mm d-1 (22.03 W m-2 ) and 0.31 mm d-1 (8.86 W m-2 ), respectively. The ET over the study area increased along a clear longitudinal gradient, which was probably controlled by the gradient of precipitation, green vegetation fractions, and the intensity of human activities. The satellite-based estimates adequately captured the spatial and seasonal structure of ET. Overall, our results demonstrate the potential of this simple but practical method for monitoring ET over regions with heterogeneous surface areas.
展开▼
