August 01, 2024

Improving Flood Hazard Predictions Through Modeling

Spatially explicit models predict the likelihood of flooding and human impacts using simulations of thousands of storms in southeast Texas.

An example simulation of flood water levels after a synthetic storm events.

[Reprinted under a Creative Commons Attribution 4.0 International License (CC BY 4.0) from Perez, G., et al. "Advancing Process-Based Flood Frequency Analysis for Assessing Flood Hazard and Population Flood Exposure." Journal of Hydrology 639 (131620), (2024). DOI:10.1016/j.jhydrol.2024.131620.‌]

The Science

This study presents a new method to better predict flood hazards and their impact on people. By using advanced computer models, researchers simulated 5,000 different storm scenarios on a region in southeast Texas to understand how floods behave over large areas. They found their model could accurately estimate flood patterns at different levels of flooding, such as floods that happen once every 100 or 500 years. This approach provides a detailed understanding of flood behavior, including peak water flow and flood extent, offering valuable insights into the complex relationships between rainfall, flooding, and the number of people affected.

The Impact

The approach’s large number of storms and spatially detailed nature enables users to pinpoint flood-prone areas and understand the potential impact of flooding on communities. By refining flood maps, this approach could lead to more informed planning and smarter, more effective protection measures. Ultimately, the new method offers a pathway toward safer, more resilient communities that are better equipped to face the challenges of severe flooding events on the U.S. Gulf Coast.

Summary

This study aims to enhance flood mitigation planning by using advanced models of a 2,000-square-kilometer watershed in southeast Texas, an area vulnerable to severe flooding. By simulating 5,000 hypothetical storm scenarios, researchers achieved statistically robust predictions of flooding’s frequency and intensity. This approach goes beyond traditional methods by not only predicting peak riverine flows but also estimating the extent of flooding and the number of people at risk.

At the heart of this study is the innovative use of Integrated Surface-Subsurface Hydrological Models (ISSHMs) alongside Stochastic Storm Transposition (SST). SST generates a broad range of storm events that align with historical storm patterns of the watershed, while ISSHMs offer a spatially explicit, comprehensive simulation of flood dynamics able to capture both riverine and rain-on-pavement flooding, including the effects of preexisting soil moisture conditions. Combining the two approaches allows researchers to build a direct connection from rainfall to riverine flows, inundation patterns, and therefore community and infrastructure impacts.

Principal Investigator

Ethan Coon
Oak Ridge National Laboratory
[email protected]

Program Manager

Sally McFarlane
U.S. Department of Energy, Biological and Environmental Research (SC-33)
Urban Integrated Field Laboratories
[email protected]

Funding

This study was supported by the Southeast Texas Urban Integrated Field Laboratory.

References

Perez, G., et al. "Advancing Process-Based Flood Frequency Analysis for Assessing Flood Hazard and Population Flood Exposure." Journal of Hydrology 639 (131620), (2024). https://doi.org/10.1016/j.jhydrol.2024.131620.