Predicting Uncertainty: IRIS researchers develop models to more accurately predict and map flood zones

Predicting Uncertainty: IRIS researchers develop models to more accurately predict and map flood zones

Flooding in urban areas cost Americans over $106 billion dollars from 1960 through 2016 in damage, and took lives and livelihoods. Predicting which areas are most likely to flood amid changing land use and rainfall can be expensive and complicated—and past methods of drawing flood maps fail to capture the inherent uncertainty in flood predictions.

Institute for Resilient Infrastructure Systems alum Dr. Tim Stephens, now at Dynamic Solutions LLC, and IRIS Director Dr. Brian Bledsoe, developed a simplified, cost-effective method that accurately captures the uncertainty in flood predictions.

In their recent paper published in Water, the authors demonstrate what Stephens calls, “the applicability of a practical simplified approach for quantifying uncertainty in flood hazard estimates compared to other alternatives,” by modeling flooding in two urban watersheds: Proctor Creek in Atlanta, GA and Bronx Wash in Tucson, AZ.

Stephens and Bledsoe hope that this approach will produce maps that more realistically show flood zones–and with less effort and cost compared to previous methods. Overall, this should result in improved maps that are more frequently updated.  

“The simplified approach is more accessible to a broader audience, such as those with limited time, budget, or technical resources. It can be implemented with a small increase in the level of effort for traditional regulatory flood hazard studies, making the incorporation of uncertainty much more approachable, viable, and cost effective,” Stephens said of the research.

This research is especially important because it comes at a time when future flooding is cast into high uncertainty due to climate and land use change, as flooding impacts more people and communities.

To learn more about this research, visit the full publication here.

Writer: Sarah Buckleitner

Contact: Tim Stephens and Brian Bledsoe (bbledsoe@uga.edu)