The 3rd USGS Modeling Conference (7-11 June 2010)

Paper No. 6
Presentation Time: 9:45 AM

VEGETATION INDEX METHODS FOR ESTIMATING EVAPOTRANSPIRATION BY REMOTE SENSING


GLENN, Edward P., Department of Soil, Water and Environmental Science, University of Arizona, Tucson, AZ 85721 and NAGLER, Pamela L., USGS, Sonoran Desert Research Station, 1110 E. South Campus Drive, Tucson, AZ 85711, eglenn@ag.arizona.edu

Most remote sensing methods for evapotranspiration (ET) have used thermal infrared bands on satellite sensors to measure surface temperatures to provide an estimate sensible heat flux from the land surface. ET is then calculated as a residual in the Simplified Surface Energy Balance (SSEB) equation. Recently, simpler methods that use visible and near infrared (NIR) bands have been developed as alternatives to SSEB methods. Visible and NIR bands are used to calculate vegetation indices (VIs) that are used to estimate green foliage density over the landscape. In many agricultural and natural ecosystems plant transpiration is the major component of ET and there is often a strong correlation between VIs measured by satellite sensors and ET measured at ground stations. VI methods cannot detect plant stress or bare soil evaporation and these must be estimated separately. This preliminary review discusses the need for VI methods for ET; their physical and biological justification; examples of different approaches and applications from agricultural and natural biomes; their limitations; and a synthesis of SSEB and VI methods, showing how they can be combined to provide maximum information about ET from vegetated landscapes.