Financial Roadmap to Resilience: Scaling Site Specific Modeling for More Realistic Watershed-Wide Planning

Financial Roadmap to Resilience: Scaling Site Specific Modeling for More Realistic Watershed-Wide PlanningAbstractOver the past five years, Federal and State level grant funding for resilience projects has significantly increased (BRIC, Infrastructure Investment and Jobs Act, etc.). Prioritizing where those investment dollars are best spent can be a challenge due to the high sensitivity of the benefit /cost models to the expected flooding depth. Typically, regional scale flood hazard analysis data is used to determine the feasibility of a project, with more detailed modeling to follow as a way to confirm benefit / cost information. However, the complexities of hydrologic and hydraulic processes often mean that accuracy is sacrificed when assessing regional scale flood risk assessments. If the ultimate benefit cost ratio determined from the detailed project specific model is drastically different than the initial regional model, it calls in to question whether the initial prioritization of projects was successful. In this presentation we will demonstrate how GIS and machine learning can be coupled with local site-specific modeling to produce detailed hazard information at a watershed-scale. This ultimately allows site specific information like stormwater capacity, detailed terrain information, infiltration data, and future conditions to increase confidence in regional feasibility level cost-benefit analyses. We will provide comparisons of the following data for several sites to examine how this new process can benefit resiliency planning: 1.FEMA flood hazard information 2.Regional scale hazard modeling 3.Detailed, project specific modeling 4.GIS & machine learning predictions (Flood Predictor) 5.Observed data Planning directors, as well as development and community infrastructure managers will learn how machine learning can be coupled with existing site specific hydraulic modeling efforts to rapidly scale detailed results to other areas throughout the watershed that are lacking good hazard information. Financial administrators can use this information to better invest in projects with greater confidence acknowledging this proactive and adaptive process has strengthened the success curve and ROI to the community.This paper was presented at the WEF Collection Systems and Stormwater Conference, April 9-12, 2024.SpeakerSmith, CurtisPresentation time16:15:0016:45:00Session time15:45:0016:45:00SessionIntegrated PlanningSession number09Session locationConnecticut Convention Center, Hartford, ConnecticutTopicFlooding, Hydrology & Hydraulics, Machine Learning, Septic to Sewer, TMDLs, Water Quality, Watershed ManagementTopicFlooding, Hydrology & Hydraulics, Machine Learning, Septic to Sewer, TMDLs, Water Quality, Watershed ManagementAuthor(s)Smith, CurtisAuthor(s)C. Smith1Author affiliation(s)Stantec 1SourceProceedings of the Water Environment FederationDocument typeConference PaperPublisherWater Environment FederationPrint publication date Apr 2024DOI10.2175/193864718825159384Volume / Issue Content sourceCollection Systems and Stormwater ConferenceCopyright2024Word count14