Effect of different methods for spatial interpolation of rainfall data for hydrological modeling in dry zone Mi Oya Basin, Sri Lanka

Abstract

For water resource management in ungauged basins, rainfall runoff hydrological modelling is applied to predict or estimate the response of watershed or drainage basin to rainfall by simulating the rainfall runoff generation. As climate data, mainly rainfall or precipitation is the given input to the model, and thus even a minor variability in the rainfall dataset would cause negative effects in the model causing the predictions and estimations in the hydrological system to infer less accurate results. Therefore, in the present study, the spatial variability of rainfall is taken into account via different spatial interpolation and computational methods and their effect is reviewed. The basin is selected as to study the variation of accuracy of simulation with different spatial interpolation methods through a model which takes precipitation as input and produces streamflow as output. The model is calibrated for a four-year period. Further, the validation is assessed for consecutive four-year period using the calibrated parameter values. Subsequently, the calibrated and validated model with the selected most suitable spatial interpolation method can be used in watersheds with similar catchment characteristics, and in our study, IDW method considerably outperforms other methods for rainfall runoff simulation at daily time-steps.