Rainfall variability and effect of different spatial interpolation methods on streamflow modelling in Kalu Ganga Basin, Sri Lanka

Abstract

Streamflow modelling plays a vital role in hydraulic design activities. As rainfall is the major input for the modelling process, the accuracy of the output after modelling directly depends on the accuracy of used rainfall data values. Obtaining spatially distributed rainfall data to incorporate the spatial variation of rainfall is achieved by using different interpolation techniques. Depending on the interpolation method used, accuracy varies and it is necessary to identify the most suitable interpolation method for a specific basin. This study is focused on the Ellagawa sub basin in Kalu Ganga basin, Sri Lanka and it compares four Deterministic Spatial Interpolation Methods (DSIMs); Simple Average Method (SA), Thiessen Polygon Method (TP), Inverse Distance Weighting Method (IDW) and Inverse Distance & Elevation Weighting Method (IDEW) in both daily and monthly basis using Twoparameter water balance model and Four-parameter (ABCD) water balance model. After comparing Pearson Coefficient (PC), Mean Relative Error (MRE) and Nash-Sutcliffe Efficiency (NSE), TP Method was identified as the most suitable method for both daily and monthly data for Kalu Ganga basin and basins with similar characteristics.