Assessment of Pan Evaporation Modeling Using Bootstrap Resampling and Soft Computing Methods

This study evaluated combined bootstrap resampling and neural network models for estimating daily pan evaporation (PE) in the Republic of Korea. Two different support vector machines (SVMs), epsilon-support vector regression (epsilon-SVR) and nu-support vector regression (nu-SVR), were developed for the local implementation of SVMs. Five-input combination models (epsilon-SVR5 and nu-SVR5) were found to be generally the best for the local implementation of SVMs. Optimal SVMs, including epsilon-SVR5 and nu-SVR5, were employed to develop bootstrap-based support vector machines (BSVMs) for two weather stations. The ensemble PE was estimated by averaging the output of 50 individual BSVMs. A Mann-Whitney U test was performed to compare the observed and ensemble of bootstrap resamplings for the training data of the PE. The uncertainty associated with PE estimation using BSVMs was evaluated. Results indicated that BSVMs could improve confidence in PE estimation and that ensemble PE using BSVMs is more stable and reliable than that using SVMs. The conventional multiple regression model (CMRM) was applied for the test period and compared with the optimal SVMs and BSVMs. It was found that BSVMs performed better than optimal SVMs and CMRM. (C) 2014 American Society of Civil Engineers.