Journal article
Non-stationary flood frequency analysis in southern Germany
Proceedings of the Seventh International Conference on Hydroscience and Engineering
13 Apr 2007
Abstract
The paper proposes an implementation of an extended Gumbel distribution and Log-Pearson III (LP3) distribution respectively in a non-stationary extreme discharge study. Three types of time dependant functions are proposed for the statistical distribution parameters. Among the three types, a modified logistic regression function is applied to the Gumbel scale parameter as well as the scale and shape parameters of the LP3. Simulated Annealing is employed as optimization algorithm for parameter estimation towards an exploration of the maximum likelihood. Based on the extended Gumbel and LP3 distributions, significance tests and trend analysis are carried out through bootstrap re-sampling. Discharge data, made up of annual maxima obtained from ten gauging stations located in southern Germany, is used as a case study. The results demonstrate satisfactory non-stationary parameter fitting and flood estimation using both extended distribution functions. The study is an attempt to provide an alternative approach for a more reliable estimation of the design return flood for engineering purposes. Through the extended non-stationary setting, the study gives an impression of the impact of climate/landuse change on flood occurrences and magnitudes. It can also serve as a useful tool for studying climate change scenarios along with climate model simulations.
Metrics
15 File views/ downloads
4 Record Views
Details
- Title
- Non-stationary flood frequency analysis in southern Germany
- Creators
- Yi He (Author) - Drexel University (1970-)András Bárdossy (Author) - Drexel University (1970-)Jürgen Brommundt (Author) - Drexel University (1970-)
- Publication Details
- Proceedings of the Seventh International Conference on Hydroscience and Engineering
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- DU; College of Engineering; Civil, Architectural, and Environmental Engineering
- Identifiers
- 991014632177204721