Assessing Different Formulae for Estimating Potential Evapotranspiration in Xinjiang

Abstract

【Objective】 Potential evapotranspiration (ET) is an important parameter for irrigation management and can be calculated by different methods. In this paper we evaluate different methods for estimating ET in Xinjiang province of China. 【Method】 The analysis was based on meteorological data measured from 1970 to 2015 from 16 whether stations across the province. The value of ET was calculated by the following models: comprehensive method, radiation method, mass transfer method and temperature method. The technique for order of preference by similarity to ideal solution (TOPSIS) was used to rank applicability and suitability of these models for different regions based on their accuracy and sensitivity. 【Result】 On average across Xinjiang, ET calculated by the radiation method was closer to the standard value than by other models. Among all models we compared, the Makkink model was more applicable, giving most accurate result across the province, with its correlation coefficient with the standard value for all 16 weather station being more than 0.95, the associated NSE close to 1 and average root mean square errors over the 16 stations being 25.28 mm/month. 【Conclusion】 In southern Xinjiang, the Makkink model and the Penman-Monteith model are most accurate, followed by Priestley-Taylor (P-T), Hargreaves (Harg) and temperature method. In northern Xinjiang, the radiation method worked better, followed by the Penman and the Penman-Monteith models.

Keywords

• potential evapotranspiration
• sensitivity analysis
• topsis
• xinjiang