Scientists mapped where a once-in-150-year solar storm would drop the US grid first

Scientists have built the most detailed map yet of how a severe solar storm would move through the American power grid, and it shows which regions would likely go dark first. A study published this month in the journal AGU Advances combines physics, engineering and economic modeling across more than 10,000 substations and 16,000 transmission lines nationwide. It finds the East Coast and the Northern Plains carry the highest blackout risk, largely because of what sits beneath those regions rather than anything utilities built above ground.

Mapping 10,464 substations and 16,256 transmission lines

The research team, led by Edward J. Oughton and Dennies K. Bor and published on September 4 in AGU Advances, assembled a model of the contiguous United States transmission network containing 10,464 substations and 16,256 transmission lines. Substation locations came from OpenStreetMap, while the transmission-line data came from the federal Homeland Infrastructure Foundation-Level Data program. The result is a national research model built from publicly available infrastructure data, not a complete operating map supplied by every utility, but detailed enough to estimate storm impacts substation by substation. The researchers built the model specifically because earlier assessments of space-weather risk tended to be siloed by discipline, with physicists studying the storms themselves and engineers studying grid vulnerability largely separately, leaving few attempts to connect either one to the resulting economic damage.

Why bedrock, not just wiring, decides who loses power first

Solar storms begin when the sun ejects plasma and magnetic field during a coronal mass ejection. When that material reaches Earth and interacts with the planet’s magnetic field, it can induce geoelectric fields in the crust and ocean below. Those induced currents then flow along whatever conducting path is available, including through the extra-high-voltage transformers that step electricity up and down across the grid. According to the AGU research summary, the East Coast and Northern Plains are especially exposed because the bedrock underneath is highly resistive. Instead of letting induced currents drain harmlessly into the ground, that resistive rock forces the current to build up in the grid itself, raising the odds that a transformer overheats or trips offline.

A 100-year storm and a 250-year storm, compared

Rather than modeling a single worst case, the team ran the grid model against storms of varying severity. A storm with a 100-year return period was projected to cut power to roughly 3.5 million people and 91,000 businesses, with daily economic losses near $1.22 billion. A more extreme 250-year storm, the most intense scenario the researchers evaluated, would affect up to 5 million Americans and more than 135,000 businesses, producing direct losses of about $980 million a day and total daily losses of roughly $1.81 billion once cascading economic effects are included. Framing the storms by return period, the same approach flood planners use for a “100-year flood,” lets utilities weigh the cost of hardening a substation against the statistical odds that a storm severe enough to knock it out will actually occur in a given decade.

Testing the model against the 2024 Gannon storm

To check whether the simulation reflected real-world behavior, the researchers compared its predicted effects at electric substations against measurements the Tennessee Valley Authority collected during the May 2024 geomagnetic storm, commonly called the Gannon storm. That event was one of the strongest to hit Earth in two decades and gave researchers rare, real utility-grade data to validate a model that otherwise has to rely on simulated scenarios rather than an actual continent-wide blackout. The Gannon storm, named for the late space weather scientist Jennifer Gannon, produced auroras as far south as Mexico and knocked GPS-guided farm equipment off course during peak planting season, contributing to an estimated $500 million in crop losses even though it never caused the kind of grid failure the new study models. That gap between a storm severe enough to disrupt GPS nationwide and one severe enough to actually collapse sections of the grid is part of what the researchers were trying to quantify.

Where the Carrington benchmark fits in

The study’s own worst-case scenario is a 250-year storm, but a CNN interactive built around the same research frames the ultimate benchmark differently: a Carrington-class event, named for the intense 1859 storm that set telegraph lines on fire, is described there as a roughly once-in-150-year occurrence or rarer. That distinction matters because a Carrington-class storm would sit beyond even the 250-year scenario the AGU Advances team modeled directly. Space weather has already caused real trouble on a smaller scale: in 1967, a solar storm jammed early-warning radar systems used by the United States military, a disruption serious enough that it briefly worried Cold War-era officials before solar forecasters identified the true cause.

What the vulnerability map means for grid hardening

For utilities and grid planners, the value of the new map is less about predicting exactly when a severe storm will hit and more about knowing where to spend hardening dollars first. Substations in high-resistivity zones are candidates for neutral current-blocking devices and reinforced transformers precisely because they cannot rely on the ground to absorb a storm’s induced currents. The study’s authors note that future work still needs to examine cascading grid dynamics and multihazard interactions, such as what happens if a geomagnetic storm strikes during a heat wave or a hurricane, when demand on the grid is already elevated and repair crews are stretched thin.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview