A spatial neural fuzzy network for estimating pan evaporation at ungauged sites

摘要

Evaporation is an essential reference to the management of water resources.In this study, a hybrid model that integrates a spatial neural fuzzy networkwith the kringing method is developed to estimate pan evaporation atungauged sites. The adaptive network-based fuzzy inference system (ANFIS)can extract the nonlinear relationship of observations, while kriging is anexcellent geostatistical interpolator. Three-year daily data collected fromnineteen meteorological stations covering the whole of Taiwan are used totrain and test the constructed model. The pan evaporation (E_pan) atungauged sites can be obtained through summing up the outputs of thespatially weighted ANFIS and the residuals adjusted by kriging. Resultsindicate that the proposed AK model (hybriding ANFIS and kriging) caneffectively improve the accuracy of E_pan estimation as compared withthat of empirical formula. This hybrid model demonstrates its reliability inestimating the spatial distribution of E_pan and consequently providesprecise E_pan estimation by taking geographical features intoconsideration.