Evaluation of Class-A Pan Coefficient Models for Estimation of Reference Crop Evapotranspiration for Dry Sub-humid Climates

Abstract

Evaporation and evapotranspiration are important parameters for various agricultural activities, which may be estimated using a pan coefficient value obtained from several models. It is a function of wind speed, temperature, relative humidity and fetch length. For dry sub-humid climate of Varanasi, Snyder model was found to be best for estimating the pan coefficient with an average value of 0.79, which closely agreed with pan coefficient value estimated by the standard FAO-56 model. The maximum and minimum average daily pan evaporation for the region was 8.8 mm.day-1 and 2.5 mm.day-1, respectively, in the month of May and December. The reference evapotranspiration estimated using pan coefficient obtained from Snyder method showed lowest MAE of 0.24, RMSE of 0.30, agreement Index of 0.99, percentage error of estimate of 6.51 % and efficiency of 96 per cent. The Mann-Kendall’s non-parametric test used for identifying trend showed an increasing trend for pan coefficient data series, and a decreasing trend for pan evaporation data series at 5 % significance level. Tempoal variation in pan coefficient may be computed for Varanasi meteorological station to estimate reference evapotranspiration.