Jump to Main Content
The effect of a hydrological model structure and rainfall data on the accuracy of flood description in an upland catchment Land Reclamation
- Wałęga, Andrzej, Książek, Leszek
- Annals of Warsaw University of Life Sciences, SGGW 2015 v.47 no.4 pp. 305-320
- computer software, highlands, hydrograph, hydrologic models, land use, meteorological data, model validation, multivariate analysis, rain, rivers, uncertainty, watersheds
- The effect of a hydrological model structure and rainfall data on the accuracy of flood description in an upland catchment. The aim of this paper was to determine the influence of a hydrological model structure and rainfall- -related data on flood parameters obtained from a simulation. The study included an upland river Stobnica, right tributary of the Wisłok. The following assumptions were investigated: (i) the greater number of rainfall stations, the more accurate a flood description, i.e. the resulting hydrograph much better describes the actual flood, (ii) a distributed parameter model provides a more precise description of a catchment response to rainfall than a lumped parameter model. All calculations were performed using HEC-HMS 3.4 software. The analyses showed that increasing the number of rainfall stations slightly improved the model performance (by on average 4.1%). Furthermore, it was showed that in the catchment characterized by low topographical variability and stable land use, more reliable flood simulation results were obtained in the lumped parameter model than in the distributed parameter model. Considering the calibration process slightly improved the model performance, irrespective of its structure and the number of rainfall stations. Multivariate analysis of variance (MANOVA) revealed that the resulting differences in the model efficiency for individual variants were not significant. Considering limited empirical evidence on rainfall-runoff episodes, uncertainty of these results is probably high and thus they should be treated as a starting point for further studies.