When Hurricane Helene hit the southeastern United States in September 2024, it brought widespread devastation and left communities grappling with power outages, damaged infrastructure and the overwhelming task of recovery. Just weeks later, Hurricane Milton followed, compounding the damage in already affected areas and pushing response teams to the limit. Amid these back-to-back disasters, a specialized team from the Department of Energy's Oak Ridge National Laboratory was deployed to the field, using cutting-edge technology to gather critical data for both immediate recovery and future preparedness.
Led by ORNL's Andrew Duncan, the group dispatched two teams to the field, spending six days across Florida, North Carolina and East Tennessee. The team, including Zach Ryan, Brandon Stockwell, Jairus Hines, Matt Larson and Jason Richards, conducted over 100 drone flights, collecting more than 400 gigabytes of geospatial data. These flights, conducted in the wake of both Hurricanes Helene and Milton, focused primarily on energy infrastructure such as power poles and transmission lines. The data gathered will fuel ORNL's research into developing more effective disaster response tools, using machine learning models to detect damaged infrastructure.
"We set out to collect training data, but the impact went beyond that," said Duncan. "While we were out there, local authorities began reaching out to us once they realized what we could do. They saw the value in the data we were collecting and began asking us for assistance with their own recovery efforts."
The team used their drones to gather aerial imagery, which was processed and shared almost immediately with responders through ORNL's Mapster technology, a software platform that provides real-time geospatial data to aid in disaster response.
"With Mapster, within 10 minutes of landing the drone, we were able to push the data to ORNL, the Federal Emergency Management Agency and other emergency response partners," Duncan added.
The team's quick deployment and immediate impact showcased ORNL's ability to collect and share valuable data in real time, even under the challenging conditions of post-disaster environments. A major limiting factor in disaster response is simply the amount of time it takes to translate collected data into useful information. The complete data workflow demonstrated during both Helene and Milton showcases ORNL's broader research efforts aimed at reducing the data-to-decision timeline and developing tools that will allow responders to act faster and more efficiently in future disasters.
As the team flew over the impacted areas, the scale of the devastation was clear. Duncan recalled a particularly emotional moment in Ezell Beach, where there was little left but piles of debris.
"In some areas, there was nothing to recover," he said. "It was devastating to see how completely the storm had wiped out entire neighborhoods. In these small communities, people's homes and lives were scattered across the streets. It really hit home why we're doing this."
The ability to provide immediate assistance proved invaluable to local utility workers. In one instance, Duncan described how the team helped a worker responsible for identifying downed power poles, a task that normally would take days. "We had a map ready for him in an hour," Duncan said. "It was amazing to see how much faster we could help these workers pinpoint the damage and get to work restoring power."
ORNL's efforts didn't stop with data collection on the ground. In the background, DOE's Environment for Analysis of Geo-Located Energy Information, or EAGLE-ITM, platform played an essential role in monitoring utility outages throughout both hurricanes. EAGLE-I was used extensively by state, local and federal agencies throughout their responses to Helene and Milton. ORNL has managed and maintained the platform since 2016.
"EAGLE-I helps us track where the biggest utility outages are, especially in rural areas where smaller counties might not get as much attention," said Aaron Myers, who leads the platform's development at ORNL.
During the response to Hurricane Helene, EAGLE-I saw a notable increase in users - including FEMA responders, decision-makers in DOE's Energy Response Center, and local emergency management officials - reflecting the platform's growing relevance in disaster management.
"We've built a lot of trust within the response community," Myers said. "This time, the focus wasn't on teaching responders how to use the tool. They were asking us how to interpret the data and deconflict information from field reporting to ensure a consistent assessment of impacts on the ground."
Ultimately, EAGLE-I's accurate, real-time data gave decision-makers a clearer picture of where power outages were most severe, allowing them to prioritize recovery efforts in both large cities and more rural, underserved areas.
Looking ahead, ORNL's work during storms is setting the stage for significant advancements in disaster response. The aerial data collected by the unmanned aerial system teams will be used to train machine learning models that can automatically detect damage to critical infrastructure, reducing the need for manual assessments and potentially saving valuable time during future recovery efforts. Meanwhile, ongoing improvements to EAGLE-I aim to integrate even more advanced data analytics and remote sensing capabilities, ensuring that responders can make informed decisions quicker than ever before.
For the teams on the ground, the work was both challenging and deeply rewarding. Known as the "Duck Squad"- a nickname born from their Drone Unit Camera Kit (DUCK) - Duncan's group took pride in their ability to help communities when they needed it most.
"It's one thing to develop these technologies in the lab but being able to apply them in a real disaster environment and see how much they can help - it was incredibly fulfilling," Duncan said.
ORNL's role in supporting the responses to Hurricanes Helene and Milton highlights the power of science and in innovation in disaster response. As both the UAS teams and EAGLE-I continue to evolve, they are poised to become even more integral in helping communities recover from future storms, ensuring that the science and technology deployed today will pave the way for faster, more efficient recovery in the years to come.
EAGLE-I and ORNL's support to disaster response and recovery efforts are funded by the DOE Office of Cybersecurity, Energy Security, and Emergency Response, or CESER.
UT-Battelle manages ORNL for DOE's Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE's Office of Science is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science .
