A physically based and machine learning hybrid approach for accurate rainfall-runoff modeling during extreme typhoon events

Accurate rainfall-runoff modeling during typhoon events is an essential task for natural disaster reduction. In this study, a novel hybrid model which integrates the outputs of physically based hydrologic modeling system into support vector machine is developed to predict hourly runoff discharges in Chishan Creek basin in southern Taiwan. Seven storms (with a total of 1200 data sets) are used for model calibration (training) and validation. Six statistical indices (mean absolute error, root mean square error, correlation coefficient, error of time to peak discharge, error of peak discharge, and coefficient of efficiency) are employed to assess prediction performance. Overall, superiority of the present approach especially for a longer (6-h) lead time prediction is revealed through a systematic comparison among three individual methods (i.e., the physically based hydrologic model, artificial neural network, and support vector machine) as well as their two hybrid combinations. Besides, our analysis and in-depth discussions further clarify the roles of physically based and data-driven components in the proposed framework.