Evolutionary Approach of Clustering to Optimize Hydrological Simulations

Modeling of hydrological systems and their dynamics in high spatio-temporal resolution leads to a better understanding of the hydrological cycle, thus it reduces the uncertainties in hydrologic forecasts. Simulation of such high-resolution, distributed and physically based models demands high performance computing resources. However, the availability of such computing resources is restricted in some domains. In this paper, we propose an approach to reduce computational costs by reducing hydrological model redundancies using similarities in functionality of hydrological model units. The approach applies K-Means clustering to detect similar model units and simulates only one representative unit of each cluster. The clustering is applied when rainfall is forced to the hydrological system and is based on the structure, current state and flux of the model units. Application of this evolutionary approach on a test case results in a 1.8x speedup over the original simulation run time and the RMSE of 0.0049 compared to the original simulation output.