Probable Maximums Should Probably Have Rare Probabilities: Implications of Modernizing Probable Maximum Precipitation

Historic and current Probable Maximum Precipitation (PMP) estimates assume precipitation is bounded and cannot be exceeded. These PMP estimates are based on subjective procedures using limited data, do not account for non-stationarity or statistical uncertainty, and have been exceeded in the past. Due to current PMP estimate limitations, and to ensure the safety of critical infrastructure, The National Academies of Sciences, Engineering and Medicine (NASEM) has proposed a revised PMP definition and vision for future PMP methodology. The proposed revised definition states PMPs should have an extremely low annual exceedance probability (AEP), for a specific time period. The shift from deterministic upper limits to probabilistic estimates alone will have significant implications for dam safety programs and project decision making. Additionally, the NASEM long term vision of PMPs using physics-based models for now and the future will introduce technical challenges in credibly extrapolating to rare AEPs while quantifying uncertainty for varying time periods. This paper provides thoughts on challenges applying these changes, including project specific examples, as well as dam owner preliminary options to address these challenges.