Modeling of Monthly Pan Evaporation Using M5P Machin Learning Technique

Abstract

The purpose of this study is to investigate the ability of M5P model trees machine learning technique for estimating monthly pan evaporation from meteorological data. The M5 method as it is implemented in the WEKA system is used to generate trees models. Three different M5P models comprising various combinations of monthly climatic variables (temperature, wind speed, and relative humidity) are developed to evaluate effect of each of these variables on evaporation estimations. Two error statistics namely root mean squared error and coefficient of determination are used to measure the performance of the developed models. Monthly meteorological data of Emara station in Missan, south of Iraq is used in this study as a case study. The results demonstrated that the M5P models whose inputs are wind speed, relative humidity and temperature performed the best among the input combination tried in the study. It was found that M5P could be employed successfully in modeling evaporation process from the available climatic data.