Estimating of regional evapotranspiration for arid areas using LANDSAT thematic mapper images data : a case study for grape plantation

Abstract

In west southern mountains of Yemen grape crop has been considered as an important cash crop. Thus, water management for grape plantation in arid areas has become an important aspect to ensure a food produce. Except alfalfa, the water used by grape trees is greater than that of most crops. Conventional Point measurement of water needed by one Grape plantation cannot provide accurate estimate for all the orchards in a county. In fact, over a vast area, the point measurements technique is costly and unpractical. In this paper, a new approach is suggested to estimate detailed water requirement by grape plantation at a county scale. The proposed technique used LANDSAT-TM data and a modified SEBAL (Surface Energy Balance Algorithm for Land) to estimate evapotranspiration over grape plantation in wadi asser- Sana’a basin central Yemen mountains. The modified SEBAL model estimates evapotranspiration (ET) using the energy balance equations, for which the surface temperature and reflectance data from TM image data and metrological data from local weather station. The model calculates net radiation, soil and sensible heat flux, and evapotranspiration. Comparing the calculated results with those observed in point measurements in the field of Grape and alfalfa from the period 1995 to 1998 proves that the modified SEBAL also provides an accurate information. The average relative error between estimated and observed ET is 11.6%, and the average absolute error is 0.43 mm/day. This proposed technique has the potential to provide guidelines for various users, including government agencies on how to evaluate current water-usage schemes.