Hydrological model assessment for flood early warning in a tropical high mountain basin
María Carolina Rogelis1, Micha Werner1,2, Nelson Obregón3, and Nigel Wright41UNESCO-IHE, P.O. Box 3015, 2601DA Delft, the Netherlands 2Deltares, PO Box 177, 2600MH Delft, the Netherlands 3Universidad Javeriana, KR 7 No 40-62, Bogotá, Colombia 4School of Civil Engineering, University of Leeds, Leeds LS2 9JT, UK
Received: 19 Jan 2016 – Accepted for review: 18 Feb 2016 – Discussion started: 02 Mar 2016
Abstract. A distributed model (TETIS), a semi-distributed model (TOPMODEL) and a lumped model (HEC HMS soil moisture accounting) were used to simulate the discharge response of a tropical high mountain basin characterized by soils with high water storage capacity and high conductivity. The models were calibrated with the Shuffle Complex Evolution algorithm, using the Kling and Gupta efficiency as objective function. Performance analysis and diagnostics were carried out using the signatures of the flow duration curve and through analysis of the model fluxes in order to identify the most appropriate model for the study area for flood early warning. The impact of varying grid sizes was assessed in the TETIS model and the TOPMODEL in order to chose a model with balanced model performance and computational efficiency. The sensitivity of the models to variation in the precipitation input was analysed by forcing the models with a rainfall ensemble obtained from Gaussian simulation. The resulting discharge ensembles of each model were compared in order to identify differences among models structures. The results show that TOPMODEL is the most realistic model of the three tested, albeit showing the largest discharge ensemble spread. The main differences among models occur between HEC HMS soil moisture accounting and TETIS, and HEC HMS soil moisture accounting and TOPMODEL, with HEC HMS soil moisture accounting producing ensembles in a range lower than the other two models. The ensembles of TETIS and TOPMODEL are more similar.
Rogelis, M. C., Werner, M., Obregón, N., and Wright, N.: Hydrological model assessment for flood early warning in a tropical high mountain basin, Hydrol. Earth Syst. Sci. Discuss., doi:10.5194/hess-2016-30, 2016.