Estimating Evapotranspiration of Paddy Field and Teak Plantation Using Remote Sensing

Abstract

               The objective of this study was to investigate suitable models that can estimate actual evapotranspiration (ET_a) by using remote sensing data. To achieve this, the study compared evapotranspiration in a paddy field in Sukhothai with a teak plantation in Lampang using remotely
sensed data. The investigation period covered two years and eight months from January 2002 to September 2004. 17 AVHRR images from NOAA/ satellite were used for the study. 

               It was found that the relationship between ET_a and remote sensing parameters such as normalized difference vegetation index (NDVI), land surface temperature (LST) and surface albedo (Sur_alb) of each site, were highly significant. The suitable model for each site and each period were as follows: in the paddy field for the whole period ET_a = -1.21+ 0.73 (NDVI) + 0.19 (LST) + 4.02 (Sur_alb), r² = 0.71, during rice planting season ET_a = 3.99 + 0.83 (NDVI) + 0.01 (LST) +1.61(Sur_alb), r² = 0.65 and after rice planting season ET_a = -1.93 + 1.42 (NDVI) + 0.2 (LST) + 7.44 (Sur_alb), r² = 0.85.
The derived equation for the teak plantation was ET_a = -2.87 + 0.18 (NDVI) + 0.27 (LST) + 0.57(Sur_alb): r² = 0.4, rainy season ET_a = 3.7 - 0.60 (NDVI) + 0.07 (LST) + 4.57 (Sur_alb) , r² = 0.64. In addition, the best suited equation for the dry season was ET_a = -4.23-0.0004(NDVI)+0.37(LST)-5.8 (Sur_alb), r² = 0.60.

               It can be concluded that the remote sensing approach is applicable for ET_a estimation. Advantages of using remote sensing are that it provides spatial distribution for large-areas and it is relatively easy to find out NDVI, surface albedo and surface temperature when compared to other measurement methods.