Collection System Planning: Considering Climate Change and Deep Uncertainty

Storm sewers are designed to safely convey runoff from rainfall events of a certain maximum intensity and/or depth. Depending on when the sewers were designed and built, in many areas of the world it is highly likely that the design storm frequency and intensity has changed, with storms getting more intense and the return interval of storms decreasing as climate change progresses towards higher temperatures. This is resulting in extreme flooding events occurring with increasing frequency and damage. In 2023, floods and flash floods caused around 2.1 billion U.S. dollars' worth of property and crop damage across the United States. Assessing the current and projected intensity and changing frequency of extreme storms is critical for utilities and municipalities to make changes to current sewer collection practices, to prepare for surface storage of flood waters, and to decide on future design standards for stormwater management. There are three aspects to stormwater planning that are not always understood or fully addressed. 1.Current patterns of extreme rainfall have already changed from the assumptions used in past studies. This requires a re-calculation of changes to extreme rainfall frequency and intensity already being experienced. 2.Rainfall intensity and frequency will continue to change in the future due to climate change. Understanding the magnitude of the change, as well as the differences in projected change depending on the storm duration and return interval is critical to preparing for future floods. 3.The challenge of understanding the uncertainty associated with extreme rainfall frequency estimates, both for current and future conditions cannot be underestimated. There is a need to understand the range of projections, as well as to extend the projections beyond the 5- to 10-year return interval often associated with collection system design. The purpose of this presentation is to address all three aspects of stormwater planning and present some tools and approaches to address each challenge. This paper will present: - Three practical methods for updating current extreme rainfall probabilities and making projections of climate induced changes. The approaches present innovative and unique methods of assessing changes for current extreme rainfall as well as projected changes that account for duration and return interval of rainfall events. These techniques have been developed and applied both in the US and abroad, and have been benchmarked against several existing Federal and State projections. - An innovative approach to developing synthetic storms and stochastically generating long precipitation time series that will help explore the full range of potential extreme rainfall events. This method can help better evaluate both current and future projected probabilities of extreme rainfall events. The issue of uncertainty in current and projected extreme rainfall increases the complexity of making stormwater management decisions but is critical for a more complete understanding of flood risk. The presentation will show that a semi-parametric stochastic approach, which generates synthetic storms of varying durations and volumes, allows a better exploration of current and future uncertainty in the estimates of extreme rainfall. It also provides the necessary data to develop annual exceedance probability estimates for return intervals that far exceed the length of rainfall data from rain gauges. This innovative method, developed in the past year, has been applied to projects in Arizona and New Mexico. Understanding flood risk when designing storm sewers can no longer rely on the tried-and-true process of using published estimates of extreme rainfall based on past rain gauge data, or even of NOAA frequency estimates. Climate change and a better understanding of the variability inherent in current data make it critical to broaden our assessments to meet the challenges of climate change and the uncertainties in the rainfall data we use. By exploring the full range of current and projected extreme rainfall and their associated return intervals, better decisions can be made in designing collection system upgrades, as well as to plan for surface conveyance of stormwater to avoid excessive flood damage during extreme rainfall events.