| Volumes and Issues Contents of Issue 8 |
www.hydrol-earth-syst-sci.net/14/1681/2010/
doi:10.5194/hess-14-1681-2010
© Author(s) 2010. This work is distributed
under the Creative Commons Attribution 3.0 License.
Possibilistic uncertainty analysis of a conceptual model of snowmelt runoff
Facultad de Ingeniería, Pontificia Universidad Católica de Valparaíso, Av. Brasil 2147, Valparaíso, Chile
Abstract. This study presents the analysis of predictive uncertainty of a conceptual type snowmelt runoff model. The method applied uses possibilistic rather than probabilistic calculus for the evaluation of predictive uncertainty. Possibility theory is an information theory meant to model uncertainties caused by imprecise or incomplete knowledge about a real system rather than by randomness. A snow dominated catchment in the Chilean Andes is used as case study. Predictive uncertainty arising from parameter uncertainties of the watershed model is assessed. Model performance is evaluated according to several criteria, in order to define the possibility distribution of the parameter vector. The plausibility of the simulated glacier mass balance and snow cover are used for further constraining the model representations. Possibility distributions of the discharge estimates and prediction uncertainty bounds are subsequently derived. The results of the study indicate that the use of additional information allows a reduction of predictive uncertainty. In particular, the assessment of the simulated glacier mass balance and snow cover helps to reduce the width of the uncertainty bounds without a significant increment in the number of unbounded observations.
Final Revised Paper (PDF, 1111 KB) Discussion Paper (HESSD)
Citation: Jacquin, A. P.: Possibilistic uncertainty analysis of a conceptual model of snowmelt runoff, Hydrol. Earth Syst. Sci., 14, 1681-1695, doi:10.5194/hess-14-1681-2010, 2010. Bibtex EndNote Reference Manager XML
