Estimation of evapotranspiration using fuzzy state models.

Abstract

The suitability of fuzzy logic was examined for estimating evapotranspiration (ET). Two fuzzy ET models, one using two input weather parameters (solar radiation and relative humidity), and the other using three input weather parameters (solar radiation, relative humidity and day wind speed), were developed and applied to estimate grass ET. Independent weather parameters from sites representing arid and humid climates were used to test the models. Comparison of the fuzzy estimated ET values with direct ET measurements from grass-covered weighing lysimeters gave the standard error of the estimates (Syx) in the range of 0.22 to 0.97, and the squared correlation coefficients (r2) in the range of 0.72 to 0.90. The mean absolute per cent errors were in the range of 6.1% to 23.9%. The ET estimated using the fuzzy model with three input weather parameters were comparable to the ET estimated with the FAO Penman-Monteith equation at all the sites evaluated. The results show that fuzzy ET models can yield accurate, robust, and low-cost estimation of ET. However, at extremely low temperatures (below 10°C), both models overestimated ET.