Dissertation
IMPROVED CHARACTERIZATIONS OF DESIGN STORM AND DROUGHT: STOCHASTIC, NONSTATIONARY AND MULTIVARIATE APPROACHES
Doctor of Philosophy (PhD), Washington State University
01/2019
Handle:
https://hdl.handle.net/2376/111797
Abstract
The overall objective of the dissertation is to develop a new framework which brings together robust and effective data-mining and statistical approaches to improve estimation of the design storm and drought, as well as their potential temporal and spatial changes. The first part of the dissertation is focused on improving the Intensity-Duration-Frequency (IDF) curves, which are often used for designing hydraulic infrastructures, by addressing the main limitations in the traditional regional frequency analysis (RFA). The contribution of this framework includes: 1) development of a fully automated and robust regionalization process based on Fuzzy C-means (FCM) clustering technique and introduction of a new data pooling method to weight the precipitation data from stations within a climatological homogenous regions; 2) quantification of uncertainty in selecting the probability distribution models and estimation of their parameters; 3) incorporation of potential changes nonstationary of the storms using a Bayesian augmented L-moment approach; 4) application of multivariate approach to account the potential correlations among the intensity, frequency and duration of extreme precipitations. Similar to the first part, the second part of the dissertation aims at providing a comprehensive framework for drought assessment that can be used in different climatic regions, watershed conditions, and for various drought types. To this end, the study contributes in: 1) providing a comprehensive and probabilistic assessment of multivariate drought risk using copula distributions taking into the impacts of future climate; 2) developing a new a multivariate and non–stationary standardized drought index based on precipitation, snowmelt, streamflow and reservoir storage which can simultaneously account for the different physical forms of drought within a changing climate. Overall, the research aims at developing probabilistic characterization of extreme hydrological events and providing an efficient decision-making tool for designing infrastructure and managing water resources taking into consideration the possible changes in the risk of extreme hydrological events.
Metrics
10 File views/ downloads
32 Record Views
Details
- Title
- IMPROVED CHARACTERIZATIONS OF DESIGN STORM AND DROUGHT: STOCHASTIC, NONSTATIONARY AND MULTIVARIATE APPROACHES
- Creators
- MD RUBAYET MORTUZA
- Contributors
- Yonas Demissie (Advisor)Akram Hossain (Committee Member)John Miller (Committee Member)Jennifer Adam (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Civil and Environmental Engineering, Department of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 139
- Identifiers
- 99900581813401842
- Language
- English
- Resource Type
- Dissertation