Optimum parameter selection for the morphometric description of watersheds: A case study of central Nigeria

Abstract

Hydrological models are very useful for predictions in many ungauged basins across the world. There are many hydrological models available for discharge data generation with different complexities and varied input parameter requirements. Studies have shown that models with many input parameters do not necessarily perform better than those with few input parameters. Basin morphometric parameters play significant roles in the conversion of rainfall to runoff and obtaining good estimates of these parameters for use in runoff models is sometime challenging as Inaccurate input into models can propagate errors and make the models to perform poorly. This study employs the method of principal component analysis to reduce the number of morphometric parameters required to run a runoff model without losing any major information. Parameters for five selected study basins in central Nigeria were measured and analysed. The result shows that three morphometric parameters (Fitness Ratio, Ruggedness Number and Watershed Eccentricity) can adequately represent other parameters as an input into a runoff model for the basins. This reduces significantly the time and effort needed to compute all the parameters which in actual fact may not improve the quality or efficiency of the runoff model.