Institute: District of Columbia
Year Established: 2014 Start Date: 2014-03-01 End Date: 2015-02-28
Total Federal Funds: $14,700 Total Non-Federal Funds: Not available
Principal Investigators: Valbona Bejleri
Abstract: Residents and businesses of Washington, DC’s Bloomingdale and LeDroit Park neighborhoods have historically experienced severe flooding during large storms. In some cases flooding has been a contributing factor in traffic incidents. In August 2012 the District appointed a Flood Prevention Task Force of experts and residents and has undertaken construction projects to alleviate some flooding problems in principal roadways, but longer term solutions are currently not scheduled to begin until 2022. A deeper understanding of factors that contribute to flooding conditions in the area is necessary to ensure that the District invests its resources most effectively to address this problem. The main objective of this research is to develop a model using Bayesian methodologies that will allow the District to carry out diagnostic, predictive, and intercausal reasoning around flooding events in the Bloomingdale and LeDroit Park neighborhoods. The model will be scalable to include effects of mitigation factors, allowing for what-if analysis. The proposed project will offer the opportunity for training students and District experts on how to use (and extend the use of) the model developed by incorporating more factors/predictors. Students will be trained to use modern Bayesian techniques and statistical software to perform data analysis. They will be engaged in research by solving real life problems. The study will assist researchers and regulators in their management decisions. The results of this project will be disseminated through water quality conferences, and/or mathematics and statistics meetings.