Performance evaluation of different empirical models for reference evapotranspiration estimation over Udhagamandalm, The Nilgiris, India.

Evapotranspiration (ET) is an important component of the hydrological cycle and reliable estimates of ET are essential for assessing crop water requirements and irrigation management. Direct measurement of evapotranspiration is both costly and involves complex and intricate procedures. Hence, empirical models are commonly utilized to estimate ET using accessible meteorological data. Given that empirical methods operate on various assumptions, it is essential to assess their performance to pinpoint the most suitable methods for ET calculation based on the availability of input data and the specific climatic conditions of a region. This study aims to evaluate different empirical methods of ET in the tropical highland Udhagamandalam region of Tamil Nadu, India, utilizing sixty years of meteorological data from 1960-2020. In this study, 8 temperature-based and 10 radiation-based empirical models are evaluated against ET estimates derived from pan evaporation observation and the FAO Penman-Monteith method (FAO-PM), respectively. Statistical error metrics indicate that both temperature and radiation-based models perform better for the Udhagamandalam region. However, radiation-based models performed better than the temperature based models. This is possibly due to the high humidity of the study region throughout the year. The results suggest that simple temperature and radiation-based models using minimum meteorological information are adequate to estimate ET and thus find potential application in agricultural water practices, hydrological processes, and irrigation management.