Developing Floodplain Designations that Account for Future Climate Change: A Case Study on Implementation of the Community Risk and Resiliency Act (CRRA) in New York, USA

Abstract:

Energy stakeholders working with floodplain managers use rules and guidelines to protect utilities and other critical infrastructure within the regulatory floodplain, which is typically delineated by the 100-year flood event. There is, however, growing concern that the 100-year floodplain designation does not adequately account for future flood risk due to climate change. In 2014 the New York State (NYS) government enacted the Community Risk and Resiliency Act (CRRA) to require state agencies to consider future flood risk in specified regulations and funding programs. This presentation describes recent NYS efforts to (1) evaluate actionable, science-based approaches to map future “climate-informed” flood risk and (2) develop a cost-effective statistical approach to scale-up map production statewide. First, we produced flood hazard maps for a 3,500 square mile pilot region covering two NYS counties and compared three approaches to modifying the 100-year floodplain designation to account for climate change: design-flow multipliers, upper confidence bounding, and freeboard addition. A comparative analysis of the three approaches shows significant differences in terms of area of the flood hazard and the number of affected structures. The advantages and limitations of each approach are discussed. Second, we developed a random forest statistical model that uses landscape characteristics to estimate or “predict” a climate-informed floodplain extent. The random forest model correctly predicted the location of flood hazard areas in the pilot region, compared to the mapped floodplain, with an average accuracy, precision, and recall of 86%, 86%, and 77% respectively. The results suggest the statistical model could provide useful information about flooding in unmapped river reaches, especially reaches that that are close to mapped areas, in high relief terrain, and not heavily regulated by built infrastructure. This case study illustrates the use of statistical mapping approaches to predict effects of climate change on riverine floodplains. These approaches can be used to develop flood risk management strategies to increase resilience and reduce the exposure of energy resources and infrastructure to local flood hazards.