Comparison and Recalibration of Equations for Estimating Reference Crop Evapotranspiration in Thailand

Abstract

In Thailand, the FAO-56 Penman-Monteith (FAO-56 PM) method is now widely recommended for estimating reference crop evapotranspiration (ETo) in spite of requiring more comprehensive weather data than other methods. However, in many cases of missing climatic data, the FAO-56 PM method may not be practically employed. Therefore, this study was conducted to evaluate the performance of alternative equations which require less input data, namely, the Priestley-Taylor, Jensen-Haise, Hargreaves, and pan evaporation methods. In addition, the empirical coefficients in some of these equations were recalibrated to determine the best alternative method for ETo estimation under Thailand's climatic conditions. The analysis based on three datasets of 30-year monthly averaged climatic data from 125 weather stations of the Thai Meteorological Department during 1966–2011 showed that with the default values of empirical coefficients, although the pan evaporation method with an assumed constant pan coefficient may be considered the most accurate, its consistency in ETo prediction needs further improvement. Nonetheless, after recalibration, no significant improvement was noticeable from the pan evaporation method. Conversely, the accuracy of the Priestley-Taylor equation was significantly improved with the average relative absolute error reduced to about 5% while maintaining a high precision with both calibration and validation datasets. Therefore, the Priestley-Taylor method with a newly calibrated value of α = 1.092 is highly recommended for using as the alternative method to the FAO-56 PM method. Nevertheless, under circumstances with very limited data, the pan evaporation and Hargreaves methods with the new empirical coefficients derived in this research resulted in about 10% of the average relative absolute error for estimating ETo under Thailand's climatic conditions.