In recent years, there has been a notable increase in the damage and interruptions caused by flooding. A commentary piece published today in the American Geophysical Union journal Earth’s Future features researchers from the University of California, Irvine, and the University of Bristol in the U.K. The U.K. researchers also work for Fathom, a flood risk intelligence company. They are urging scientists to improve their modeling of these risks and are warning against sensationalized reporting of these dangers in the media.
The authors of the paper advocate for a shift away from the traditional “bathtub modeling” method for mapping flood hazards. This method assumes that floodwaters spread out in a uniform way, similar to a still body of water. While this approach can be a simple way to visualize flooding impacts, the authors argue that it often results in an overly simplified and less accurate representation of flood risks compared to more sophisticated methodologies. They recommend using dynamical modeling, which relies on physics-based equations instead.
Brett Sanders, the co-author and a Chancellor’s Professor of civil & environmental engineering at UC Irvine, stated, “Bathtub models can either overestimate or underestimate flooding risks. One major error stems from their failure to incorporate protective systems like storm drains, levees, and pumps.”
He and his fellow researchers—Oliver Wing, Fathom’s chief scientific officer and an honorary research fellow at the University of Bristol, alongside Paul Bates, a hydrology professor at Bristol and Fathom’s chairman—highlight that bathtub models struggle to reflect at least six crucial factors (as shown in the accompanying graphic). These factors include: the reduction of flood severity from event dynamics and terrain friction; increased tidal effects due to ocean tide resonance in coastal bays; the influence of flood defenses like levees that may fail during extreme weather but still help limit inland flooding; the rise of the groundwater level; the surfacing of groundwater influenced by rising sea levels and changing water cycles; and the pumping of groundwater in low-lying areas to prevent flooding from high water tables.
The research team reviewed various studies related to flood risk and assessed the accuracy of bathtub models with the critical success index (CSI), scoring flood extent accuracy on a scale of 0 to 1, where 1 is a perfect match to real-world measurements. The CSI for bathtub models frequently falls below 0.5, which is much lower than the desirable 0.65 threshold for local relevance and useful results in impact assessments.
“A CSI under 0.5 shows these models perform worse than random guessing,” Wing said. “A chimpanzee could do a better job than a bathtub model at identifying flood risk areas.”
The researchers noted that bathtub modeling is commonly found in brief, high-profile journal articles that capture significant media attention. While the limitations and uncertainties inherent to bathtub modeling are often recognized in scholarly discussions, the sensationalized portrayals of flooding in public communications—complete with striking depictions of submerged cities—can misrepresent the actual risks.
“Accurate flood risk maps are crucial for everyone, including homeowners, businesses, insurers, banks, and governmental bodies,” Bates stated. “We all have a role in minimizing flood-related losses, but it all begins with reliable information.”
According to Sanders, trustworthy flood risk models are necessary for effectively involving communities in adaptation initiatives and developing equitable mitigation strategies. Poorly constructed models can result in misguided actions.
“Flood predictions must resonate with people, not only to enhance understanding of potential risks but also to influence the investments and policies that will be adopted to address these challenges,” Sanders explained. “Several studies have indicated that residents in high-risk areas are less likely to trust flooding projections that don’t align with their lived experiences. Research that oversimplifies flooding without considering real data hinders meaningful action.”
This study received support from the U.S. National Science Foundation, the National Oceanic and Atmospheric Administration, and the Natural Environment Research Council in the U.K.
