Accurate assessment of flood risk is critical to protecting lives and property worldwide. The design and safe operation of dams, levees, culverts, bridges, storm drainage infrastructure, and many nuclear facilities are informed by estimates of an “upper bound” of possible precipitation. In particular, dams and nuclear facilities in populated areas are often referred to as “critical” or “high hazard” due to the risk to life and property a failure presents. These structures might want to be built to withstand the most extreme storm or flood considered possible at that location.
Now, this might seem counterintuitive…
In engineering practice, this concept is called Probable Maximum Precipitation (PMP), and it is defined as the “theoretical maximum precipitation for a given duration under modern meteorological conditions” (WMO 2009). In the United States, PMP is generally estimated using a deterministic “moisture maximization method” (also referred to as the storm-based approach), which combines observations of historical extreme precipitation events in regions relevant to the location of interest with storm maximization assumptions.
Limitations of Traditional PMP Estimation
PMP estimation methods and the federally published extreme precipitation data they produce are frequently cited as being in need of update and improvement (NRC 1994; Tomlinson and Kappel 2009; England et al. 2020; Wright et al. 2021). Due to the rarity of extreme events, old storms with limited observational data are often used to define the upper bound of precipitation. Observations of many important old storms are limited in spatial and temporal coverage, and sometimes of dubious quality. This reduces confidence in flood hazard assessments used in dam safety evaluations and leads to unknown or uncertain societal risk.
Modernising PMP Estimation with Numerical Weather Models
Recent advancements in high-resolution weather modeling, particularly the ability to simulate convection explicitly, present an opportunity to improve upon traditional PMP estimation methods. Dynamical weather models produce spatially and temporally continuous precipitation estimates, often at considerably higher resolution than observations or historical reanalysis datasets. Because these data are produced by solving physical equations of the atmosphere (in contrast to interpolation methods historically employed to make up for limited observations), dynamical model representation of storm physics and evolution also reduces reliance on spatial, temporal, and physical assumptions that currently underpin PMP estimation (Mukhopadhyay and Kappel 2016).
Dynamical model output also provides coverage in remote, data-sparse regions (e.g., complex and/or high-elevation topography) and can resolve precipitation type (snow, rain, hail) as opposed to approximations based on algorithms using surface temperature or similar. Furthermore, these models on both weather and climate time scales are likely to be critical to informing updates to PMP, which incorporate the role of climate change in the anticipated increase of extreme precipitation (Mahoney et al. 2018b; McCormick et al. 2020).
Simulating Historical Extreme Events
The Twentieth Century Reanalysis (20CR) project has enabled the reconstruction of three-dimensional atmospheric states as far back as the mid-nineteenth century (Compo et al. 2011). This provides the necessary observational data to initialize and drive high-resolution weather model simulations of historical extreme precipitation events.
Researchers have leveraged the 20CR to dynamically downscale past events, including the 1888 New England Blizzard (Michaelis and Lackmann 2013) and a severe 1925 foehn storm in Europe (Stucki et al. 2015). These studies demonstrate the potential to explore historical weather extremes in unprecedented detail, with implications for understanding and assessing flood risk.
A Case Study in Modernising PMP Estimation
The Colorado–New Mexico Regional Extreme Precipitation Study (CO-NM REPS) recently applied numerical weather modeling to supplement existing data and advance both deterministic and probabilistic PMP estimation methods. As part of this project, researchers used high-resolution, convection-permitting dynamical downscaling of 20CR ensemble members to simulate seven historic storms that currently control PMP estimates in the western United States.
For the November 1909 Rattlesnake, Idaho, event, a modest ensemble of four WRF (Weather Research and Forecasting) Model simulations was found to provide valuable insights (CODNR and NMOSE 2018). The WRF-simulated precipitation fields were incorporated into the PMP calculation process, improving the spatial distribution and accounting for precipitation type (rain versus snow) compared to the original storm analysis based on sparse observations.
In contrast, for the September 1923 Savageton, Wyoming, event, the WRF simulations were unable to replicate the extreme precipitation magnitudes reported in historical records (CODNR and NMOSE 2018). This highlighted the need to carefully evaluate the reliability of historical observations, especially in data-sparse regions, and the potential for numerical modeling to identify questionable data.
Opportunities and Challenges
The CO-NM REPS project demonstrated multiple stakeholder and user benefits from incorporating numerical weather model data into PMP estimation. The results were ultimately used to update the State of Colorado Dam Safety Rules, officially passed in January 2020, which now allow for the use of high-resolution climate data and dynamical modeling techniques.
However, the exploration also uncovered several challenges and areas for improvement. Computational cost and the selection of appropriate 20CR ensemble members remain concerns, as does the integration of ensemble information versus individual simulations. Establishing a priori experimental design criteria and objective skill metrics are recommended to double-check that the most robust application of numerical modeling to PMP estimation.
While the current PMP concept and associated methods may require fundamental reimagining, the criticality of maintaining safe and usable estimates for the foreseeable future renders it an essential component of hydro-engineering practice. Opportunities to supplement and improve upon existing approaches, such as the historical event downscaling method demonstrated in the CO-NM REPS, can offer near-term benefits to society. As the field of extreme precipitation estimation evolves, numerical weather modeling is poised to play an increasingly important role in enhancing flood risk assessment and resilience.
Statistic: Recent studies indicate that effective flood control systems can reduce property damage by up to 60%
