An Ontology-based Integrated Visual Modeling System for Hydrological Forecasting

Reliable and accurate forecasting of key hydrological variables is of importance for regional water resource planning and management. To tackle the problems of low modeling efficiencies that are encountered in implementation of data-driven models using conventional modeling techniques, this paper presents the design and functionalities of a water resources ontology aided integrated visual modeling system that could be used to establish forecasting models easily and flexibly. The most significant features of integrated visual modeling system prototype are reflected in two aspects: firstly, water resources ontology was employed to address the semantic heterogeneity issues exiting in different data repositories and associate semantic information with data retrieval process. Meanwhile, a visual modeling workbench with standard modeling interfaces was established to facilitate model wrapper, data exchange, modular linking and interactive simulation, thus greatly saving efforts and time to implement fast modeling of the complex, non-linear, and dynamic hydrological process, while hiding the complexity of underline code and no programming is required. The system's feasibility and capability was demonstrated through a case study of one-day ahead streamflow forecasting in Shule River basin located in the arid inland region of northwest China. The encouraging simulation results show that the system can greatly simplify the process of implementing interactive hydrological forecast and appeal in offering a positive prospect of improving the efficiency and robustness of current practice in hydrological modeling.