Regresar

Hydrological analysis for a distributed and a semi-distributed model

Abstract:

This paper presents 2 different methods, increasing gradually in complexity, for pbkp_redicting the rainfall-runoff of a midsize catchment. The methods applied in this study are (1) a lumped conceptual semi-empirical method, being the NAM-module of the MIKE 11 model, and (2) a fully distributed, physically-based deterministic catchment model, the MIKE SHE model. The two methods are applied to the 465 km2 large Jeker basin, situated in the loamy belt region of Belgium. The landscape is rolling, and the soils are varying from sandy-loam to clay-loam. A 6-year continuous series of average daily hydrologic data were used for the calibration and validation. Mainly for reasons of workability of the distributed modelling approach the DTM is 600×600 m. The distributed model was calibrated and validated using a split-sample (SS) and a multi-site (MS) test, while only the split-sample test was used for the lumped model. In this paper the performance of two hydrologic models, a lumped conceptual and a distributed physical-based model are compared using hydrologic data of a medium sized catchment in the loamy region of Belgium. The performance of both models is compared after calibration and validation.