Fixing Flooding for the Southeast’s Future

Flooding dominated the headlines of summer 2025. Atypical storms and rising rivers in the Texas Hill Country washed away an entire summer camp. Glacial snow melt, combined with flash river floods, caused hundreds of deaths in Pakistan. As the Atlantic hurricane season hits its peak, Americans wait to see if another storm may be as unexpectedly devastating as 2024’s Hurricane Helene. 

Flooding can be an existential threat, affecting everything from infrastructure to health. Georgia Tech researchers are developing solutions to monitor and forecast flooding, as well as restore ecosystems to prevent future flooding. These efforts support communities’ resilience in the face of climate change and keep the U.S. secure.

A hiker observes the rapid retreat of coastal glaciers in Greenland, where melting icebergs contribute to rising sea levels and intensified wave activity.

Sea levels rise when glaciers melt. With warming air and water, these glaciers melt faster. The situation worsens when the dark meltwater on glaciers absorbs more sunlight, causing even more melting. School of Earth and Atmospheric Sciences (EAS) Professor Alex Robel wants to better understand how fast glaciers and ice sheets melt as the climate changes. His lab compares simple equations describing how meltwater moves on glaciers with satellite data.

“These simple equations go into the large-scale models used to predict future sea level rise,” Robel said. “The models help policymakers and government leaders make decisions about how much flooding from sea level rise we should be planning for.”

Robel works with School of Civil and Environmental Engineering (CEE) Assistant Professor Chris Lai to determine how specific water flows melt glaciers. A major meltwater contributor is a plume, which forms when meltwater comes out from underneath the glacier into the ocean. This creates a feedback loop that agitates the ocean and brings more warm ocean water toward the glaciers. Lai makes ice blocks in his lab and tests this melting on a microscale.

“If you have an ice cube in a cup of room-temperature water, the cube melts. The same thing happens on a larger scale with glaciers,” Lai said. “It's not a problem that ice melts in water, but rather how quickly it melts. Our work is to reduce the uncertainty of knowing how quickly glaciers will melt and the sea level will rise.”

In his research, Lai gets down to the fundamentals of how heat gets from water into ice. The researchers create a 3-inch-thick, meter-tall ice block that replicates glaciers. The ice block is immersed in a salt solution that mimics seawater, which the researchers agitate to replicate waves and plumes.

Robel uses Lai’s data on temperature, salinity, and melt rate to build even more accurate computer models of glacier melt. Better predictions of sea level rise from glaciers affect more than the Earth’s polar regions. This information can also help coastal communities around the world better estimate flood impact from storms.

Robel and Lai’s work is funded by the National Science Foundation (NSF) and NASA.

Post-hurricane flooding inundates residential areas and transportation infrastructure, with low-lying terrain overwhelmed by storm surge and excessive rainfall.

When a hurricane hits, the most destructive and fatal part isn’t the wind — it’s the flooding.

“There are two types of floods during hurricanes: saltwater, surge-driven flooding, which is driven by strong winds and similar to a tsunami, and freshwater flooding caused by torrential rainfall,” said Ali Sarhadi, an EAS assistant professor. “When a hurricane makes landfall in coastal areas, these two types of flooding can occur together and produce compound flooding, which is much more destructive than either individual type of flooding on its own.”

Climate change and global warming can intensify the severity and frequency of these types of flooding in different ways. First, a warming atmosphere can hold more water vapor, which causes more intense rainfall from these storms. More intense hurricanes can also cause more surge-driven flooding. Sea level rise will also provide more ocean water in coastal areas, which also means more surge-driven flooding. To compound this, many communities are built in flood-vulnerable areas, which brings more exposure and potential for damage.

Sarhadi’s lab has been working with physics-based models to simulate how climate change may intensify the risk of hurricane-driven flooding. This work will also help them identify which U.S. areas and communities will be more vulnerable to these impacts in the coming years. The outcome of his research could help policymakers and communities prepare and build smarter as the risk of these storms worsens.

Hurricane Helene, for example, decimated western North Carolina, an area that doesn’t typically see hurricanes. Helene’s torrential rains on already saturated soils from previous rain, combined with river flooding, led to the storm’s catastrophic consequences.

More than 100 people died in the area, with a total of 248 fatalities across all affected states.

“Hurricane Helene proved to us that hurricanes are no longer only a coastal problem. Many people died because of the intensifying flooding,” Sarhadi said. “Our models try to address the extent to which hurricane-driven flooding in coastal and inland areas may intensify. From that, we can learn how much we need to enhance the resilience of cities and communities to minimize fatalities and damage.”

His team is also developing physics-informed, generative AI-based forecasting models to predict the hazards related to hurricanes. The goal of these models is to be better prepared for anticipated hazards and to warn communities to evacuate ahead of time.

Sarhadi's work is mostly funded by NSF.

A flood-level reflective sign in Charleston to let vehicles know how deep the water is. [Photo by John Taylor]

First responders are crucial during storms. But in some cities, storms can threaten their ability to respond. In Charleston, South Carolina, for example, the main hospital is in a flood-prone area, making it challenging for ambulances to traverse the city when it rains. During flooding events in some cities, the emergency medical services (EMS) can take up to 10 times as long to reach patients in severe cases. CEE researchers John Taylor and Neda Mohammadi conduct research that helps first responders understand which areas might be flooding and how badly.

“My group thought that if we could better understand flooding on the streets, we could provide better information on flooding to ambulances, helping them route better — not necessarily in a shorter distance but a faster time,” Taylor said.

Taylor’s lab creates digital twins, which are AI models that replicate a real-world environment. To create these digital twins for downtown Charleston, his group combines multiple data sources: data on sensed water depth of certain Charleston streets, data from sensors placed around the city for livestream footage of the streets, and streaming data on road closures from the police department. Together, these data sources can help them develop better predictions of which areas flood.

They have collaborated with Charleston to test how well the sensors and digital twins work during hurricane season. Based on their results, the project could expand to Savannah and eventually to the entire Southeast.

The Partnership for Innovation, an initiative of Georgia Tech’s Enterprise Innovation Institute, funds Taylor’s work in Charleston.

Georgia Tech graduate students and faculty Karl Lang and Frances Rivera-Hernández doing fieldwork in Puerto Rico. [Photo courtesy of Karl Lang]

When we think of flooding, we often picture rivers rising after being inundated with rain or hurricanes pushing ocean water into cities. But that’s not the only type of flooding. Outburst flooding — when a dam bursts or a reservoir overflows — can also be detrimental to communities. Karl Lang began researching outburst floods as a Ph.D. student and has continued this work as an EAS assistant professor.

“I am a geomorphologist, which is someone who studies how landscapes change over time,” he said. “I don’t directly study flooding; I study the impact of the flooding on the landscape.”

Lang first started studying outburst floods when he found unusual mineral deposits in the Himalayas. These minerals suggested the area was flooded thousands of years ago. Knowing the flood history of an area is important for accurately modeling how often it might flood now.

With climate change and melting glaciers, these outburst floods are even more common, and models must be updated. In 2023, the Himalayan region of Sikkim, India, experienced a deadly outburst flood when a landslide caused a lake to overflow. These floods have the compounding hazard of containing a lot of sediment from the lake, meaning communities are likely to be buried by sediment. Better modeling is crucial for preserving infrastructure and preventing mass casualties.

“These outburst floods are essentially a wave, like a tsunami, that starts from an initial point and then propagates over the landscape,” Lang said. “So, we use an adapted tsunami model to simulate these outburst floods.”

Lang’s work can be applied to multiple regions, from the Himalayas to the flooded areas of western North Carolina after Hurricane Helene. Recently, he collaborated with EAS Assistant Professor Frances Rivera-Hernández and CEE Regents’ Professor Rafael Bras, tracking how Puerto Rico’s landscape transforms after hurricanes, which often cause landslides and consequent outburst flooding.

Lang and Rivera-Hernández travel to Puerto Rico annually to measure how hurricanes change rivers, from the channel width to vegetation amounts. Using drones, they estimate these changes and build more accurate models. Their work can help communities decide what to rebuild.

Lang’s research is funded by NSF and NASA.

Joel Kostka and Michael Hodges, a shellfish biologist in the South Carolina Department of Resources, determining the elevation of degraded marsh habitat. [Photo courtesy of Joel Kostka]

Coastal wetlands provide the first line of defense from storms and natural hazards. These mangroves and salt marshes can slow down tidal surges and soak up excess water before it flows into cities. But overdevelopment and storms are destroying these important habitats.

“Coastal communities throughout the Southeast are now facing constant challenges from flooding and sea level rise,” said Joel Kostka, Tom and Marie Patton Distinguished Professor and associate chair for Research in the School of Biological Sciences. “These communities are looking to nature-based solutions to prevent flooding and erosion related to storms.”

Kostka’s lab works along the Southeast shoreline to bolster these wetlands. In Charleston, for example, they’re regrowing grass to restore a damaged marsh. They also work with “living” shorelines — restoring the marsh by adding live oysters, which naturally filter water and form barriers.

The research group’s most innovative effort is near St. Mary’s and Cumberland Island, Georgia. There, Kostka collaborates with the U.S. Navy to protect a submarine base from storms. The Navy dredges nearby waterways to enable ships and submarines to come in, and these waterways are surrounded by salt marsh. Kostka is working with the Navy and the U.S. Army Corps of Engineers to develop nature-based solutions that use the dredged sediment to fortify the marshes against rising sea levels.

“A marsh is considered a marsh for two reasons: One is the growth of grasses, and the other is because of how sediment accumulates there from rivers flowing out to the ocean,” Kostka explained. “Marsh grasses can take hold and grow where this sediment accumulates, but as we dam rivers, there’s less sediment.”

Kostka’s lab applies these solutions — planting grass, adding oysters, and spraying dredge sediment on grasses — to rebuild the coastline. As they use each method, the researchers also quantify the method’s effectiveness to figure out what the best solution might be for each wetland.

They collect GPS data on marsh elevation because elevation determines whether the marsh will be submerged as the sea level rises. They also track vegetation health. These two data pieces are put into an AI model that estimates the path of a storm surge to determine if the marsh will survive a major flood. Elevation needs to be just right: too high and the marsh dries out, but too low and the plants drown. All these efforts could prove monumental for coastal restoration and protection.

“These ecosystems provide habitat for wildlife, and people want to go down and catch fish and enjoy the coast,” Kostka said. “We need to have restoration solutions that support habitat for wildlife, because that's really why we have tourism on the Georgia coast.”

Kostka’s research is funded by the National Fish and Wildlife Foundation. He was recently named director of Georgia Tech for Georgia’s Tomorrow, a new initiative supported by the College of Sciences.

Jill Gambill is pictured with Hanif Haynes from the Pin Point Betterment Association. [Photo courtesy of Jill Gambill]

Sea level rise on Georgia’s coast affects its communities and its infrastructure. Since 2018, Georgia Tech has tracked how much water is rising with a smart sea level sensors project, led by School of Computer Science senior research scientist Russ Clark. Now the project’s goal involves local outreach as much as technology.

In 2023, Jill Gambill joined the Institute for People and Technology, which now houses the sea level sensors research under the CEAR Hub. Gambill’s role bridges the gap between community engagement and the sea level sensors Clark maintains. After joining Georgia Tech, Gambill moved to Savannah, where she is part of the community she’s helping.

To determine which sensor data is most useful, CEAR partners with governmental organizations like the City of Savannah and the Chatham Emergency Management Agency. Gambill also regularly attends community meetings, working with Harambee House, a community organization, as well as the Pin Point Betterment Association, which serves the Gullah Geechee community. Her work takes her from Savannah to Tybee Island to Brunswick.

“What’s special about Georgia Tech’s approach is that it really is community-based,” she said. “It's not researchers coming in to solve the problem and acting like we know better. These communities deal with flooding day in and day out, so we're trying to provide expertise, support, and connections to help them so they can implement solutions.”

That includes educating and training the community. CEAR partners on STEM nights at local schools to teach students how the sea level sensors work and what can be learned from their data. A monthlong summer camp at Savannah State University lets kids dive deeper.

Impacts are already being felt across these communities — Pin Point is working on stormwater improvements to improve flood resilience, for example. These partnerships will make coastal Georgia communities part of the solution, rather than an afterthought.

CEAR research is funded by NASA, the National Oceanic and Atmospheric Administration, and the National Fish and Wildlife Foundation.