Notice of RetractionModeling daily runoff in a large-scale basin based on Support Vector Machines

Notice of Retraction

After careful and considered review of the content of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE´s Publication Principles.

We hereby retract the content of this paper. Reasonable effort should be made to remove all past references to this paper.

The presenting author of this paper has the option to appeal this decision by contacting TPII@ieee.org.

Support Vector Machine (SVM) based rainfall runoff models have been successfully applied to daily runoff modeling in many basins. Most of them are however designed for small or meso-scale basins rather than large-scale basins. One of aims in the present work is therefore to develop an SVM model with an optimized combination of input variables for daily stream flow simulating. Another aim is to compare the performance of SVM models with two different process-based hydrological models, namely TOPMODE and Xinanjiang model, in one day ahead stream flow forecasting. Yingluoxia basin, with a drainage area of 10009 km², is selected for testing them. The results show that the precipitation, evaporation and antecedent observed stream flow, are all necessary as inputs to SVM modeling for this basin. The optimized SVM model performs much better than TOPMODE and Xinanjiang model both for calibration period and the validation period in terms of Nash-Sutcliffe efficiency. The daily stream flows simulated by the SVM are in very good agreement with the observed ones, while those simulated by Xinanjiang and TOPMODEL significantly underestimate or overestimate the main peak-flows and are greatly different from the observed ones for low flow stages in both calibration stage and validation period. SVM models are promising tools for short term daily runoff forecasting even if in a large-scale basin.