Researchers at Oak Ridge National Laboratory (ORNL), part of the Department of Energy, are leveraging non-weather data from the national weather radar network to track airborne, non-meteorological events, with potential applications for improving emergency response.
Weather radar data, commonly showcased in news reports as colorful displays of precipitation movement, is gathered through a network of Doppler radars across the country. These radars emit energy pulses that are processed by algorithms to provide vital weather information, such as rain and sleet patterns. However, the data typically filtered out could also be repurposed to track items like smoke, ash, or debris in the air.
Precipitation shows up as spheres on a Doppler radar. In the dataset, spheres have a value of one. Anything else detected by the radar in the air shows up as an elongated item and has a value of less than one. In this way, we can remove the precipitation entries, those coded as one, and use the remaining data to see what was in the air, what direction it went, and what happened to it as it traveled over a distance.
Erik Kabela, Meteorologist, National Security Sciences, Oak Ridge National Laboratory
To explore the potential of this data, Kabela's team analyzed historical news events where debris was launched into the air. These included incidents such as the BP oil rig explosion in the Gulf of Mexico, the SpaceX Starship explosion in Texas, and the aftermath of the Oklahoma City bombing. By reviewing radar data from these events, the researchers visualized how debris clouds formed and moved, with their findings published in the Journal of Emergency Management.
Kabela is interested in adapting the outcome of this research for emergency responders.
“Non-weather data could tip off authorities before other types of indications are used. For responders already in the field, this data could tell them where a debris cloud is moving and provide indications of safe places to set up their operating station.”
Mike Benson, a Mechanical Engineer at ORNL supporting nuclear nonproliferation, noted that weather—a dataset people check daily—offers readily available information for various other applications.
A tremendous resource exists with data that was filtered out for the intended purpose but actually is really useful for other applications. It's a network capability in our country that’s well distributed.
Mike Benson, Mechanical Engineer, Oak Ridge National Laboratory
According to Benson, the next step is to develop a machine learning algorithm to automatically detect a cloud of debris and notify a person in real-time.
“The algorithm could detect emergency events in rural areas that haven’t been noticed yet by a person. The earlier an event can be detected, the faster services are deployed to mitigate impacts,” said Benson.
This data can also be used to support nuclear forensics investigations, a project that David Hooper of ORNL has been supporting for the past 15 years.
If an investigation were to happen, technicians would go into the field to collect samples. Weather radars could detect a plume to tell us how much debris is in the air and where it’s moving. This would help with tracking the data and providing safer information for the field technicians.
David Hooper, Incident Modeling and Computation Engineer, Oak Ridge National Laboratory
Hooper’s team focuses on creating software tailored to the U.S. government’s specific requirements for fallout modeling. By integrating radar data, these models can be enhanced to provide more accurate predictions of fallout impacts. In a forensic scenario, radar data could monitor the trajectory of a rising cloud, offering precise estimates of how winds are moving it, the influence of shear layers on its behavior, and the potential paths of debris as it descends to the ground.
“This would mean dealing less with a predictive model and more with a physical measurement, which for mission planning is always preferable,” said Hooper.
The Defense Threat Reduction Agency provided funding for this project.
Journal Reference:
Benson, J, M., et al. (2024) Weather radar utility in hazard detection and response. Journal of Emergency Management. doi.org/10.5055/jem.0868