GIS and Remote Sensing in Hydrology, Water Resources and Environment (Proceedings of ICGRHWE held at the Three Gorges Dam, China, September 2003). IAHS Publ. 289, 2004, 392-400

1 Department of Civil Engineering, Graduate School of Engineering, Tohoku University, Aoba 06, Aramaki, Aoba, Sendai, Miyagi Prefecture 980-8579, Japan

2 Graduate School of Environmental Studies, Tohoku University, Aoba 06, Aramaki, Aoba, Sendai,
Miyagi Prefecture 980-8579, Japan

Abstract To establish a method for estimating evapotranspiration from NDVI data, cloud free images were compiled by a new algorithm using a temporally varying function and land use data. The NDVI data set was used for the estimation of evapotranspiration, and the estimated flux was compared with the estimation of a single layer model. The comparison shows that the single layer model produces less evapotranspiration than the NDVI method based on water balance. The distribution patterns of evapotranspiration estimated by the two methods were quite similar, indicating that the NDVI method is effective in estimating evapotranspiration.

Key words cloud-free image; evapotranspiration; GIS; NDVI; remote sensing