Thesis
Identifying flood causing mechanisms in major cities in the United States using information theory and artificial neural network
Washington State University
Master of Science (MS), Washington State University
05/2020
DOI:
https://doi.org/10.7273/000004199
Handle:
https://hdl.handle.net/2376/125144
Abstract
Floods have significant impacts on societies than any other natural calamities. Urbanization, in addition to introducing significant impervious areas, concentrates people within small areas, increasing flood vulnerability and flood losses. Thus, a better understanding of the causes of these floods is crucial to mitigate and manage future impacts. In this study, different potential flood mechanisms were analyzed for some of the major cities in the United States using Information Theory (IT) and Artificial Neural Network (ANN). The flood mechanisms considered include rainfall, snowmelt, soil moisture, and temperature. The non-linear causal relationships between the flood mechanism and flood events of different magnitudes were assessed using partial mutual information (PMI) and ANN methodologies. The procedure can be considered as a two-stage process, where first the potential flood mechanisms are filtered using their information content to predict floods and then filtered based on how well the different combinations of mechanisms predict the observed floods. The time lagged data of the different mechanisms were filtered through the PMI, which identifies the relevant mechanisms that are used as input variables for the ANN simulation of floods. Then the combination of the relevant mechanisms was selected based on the minimum AIC value from ANN prediction of flood, which represents the best model. The methodologies were tested for Seattle, New York, Los Angeles, Chicago, Denver, and Houston, which represent the different climatologic regions and flood causes in the US.
Metrics
2 File views/ downloads
14 Record Views
Details
- Title
- Identifying flood causing mechanisms in major cities in the United States using information theory and artificial neural network
- Creators
- Rakib Ahmed Siddique
- Contributors
- Yonas Demissie (Advisor) - Washington State University, Civil and Environmental Engineering, Department of
- Awarding Institution
- Washington State University
- Academic Unit
- Engineering and Applied Sciences (TRIC), School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Identifiers
- 99900896438101842
- Language
- English
- Resource Type
- Thesis