| This thesis makes an endeavor to improve the accuracy of rainfall and runoff predictions in three aspects, model inputs, selection of models, and data-preprocessing techniques. Seven input techniques, namely, linear correlation analysis (LCA), false nearest neighbors, correlation integral, stepwise linear regression, average mutual information, partial mutual information, artificial neural network (ANN) based on multi-objective genetic algorithm, are first examined to select optimal model inputs in each prediction scenario. Representative models, such as K-nearest-neighbors (K-NN) model, dynamic system based model (DSBM), ANN, modular ANN (MANN), and hybrid artificial neural network-support vector regression (ANN-SVR), are then proposed to conduct rainfall and streamflow forecasts. Four data-preprocessing methods including moving average (MA), principal component analysis (PCA), singular spectrum analysis (SSA), and wavelet analysis (WA), are further investigated by integration with the abovementioned forecasting models.;Proposed data-driven models are firstly employed for predictions of monthly and daily series of rainfall and runoff. The comparison of seven input techniques indicates that LCA is able to identify model inputs reasonably and SSA performs the best in four data preprocessing techniques. In terms of prediction models, modular models tend to perform better than global models for daily series data whereas the advantage of the former over latter is not salient for monthly series data. Proposed models are also used to perform daily rainfall-runoff (R-R) prediction where model inputs consist of previous rainfall and runoff observations. Compared to models depending solely on previous flow data as inputs, these R-R models make more accurate predictions. Finally, we extend point prediction to interval prediction using the ANN-SSA R-R model for the purpose of evaluating uncertainty of prediction where the uncertainty estimation based on local errors and clustering (UNEEC) method is compared with the bootstrap method. Results indicate that UNEEC performs better in locations of low flows whereas the bootstrap method proves to be well suited in locations of high flows.;One of major contributions of this research is the exploration of a viable modeling technique of coupling data-driven models with SSA for hydrological predictions. |
