In May 2024, people in Florida, Puerto Rico, and northern Mexico looked up and saw the northern lights with their naked eyes. The geomagnetic storm responsible was the strongest to hit Earth in over two decades, and it arrived well after Solar Cycle 25 had already begun showing signs of peaking. Now, as of mid-2026, monthly sunspot counts tracked by NOAA’s Space Weather Prediction Center confirm the cycle is sliding into its declining phase. Yet space weather researchers say the probability of another extreme storm, one capable of pushing vivid auroras deep into the subtropics and stressing power grids and satellites, is not declining with it. If anything, historical patterns suggest the danger is growing.
Solar Cycle 25 outperformed expectations, then started to fade
When an international panel co-chaired by NOAA and NASA issued its original Solar Cycle 25 forecast, the outlook was modest. The Sun had other plans. Activity climbed faster and higher than projected, prompting the Space Weather Prediction Center to issue a revised outlook describing a quicker, stronger peak than the panel had anticipated. A subsequent forecast update was effectively overtaken by the Sun’s own behavior, with observed sunspot indices running well above the baseline scenario.
By mid-2026, the SWPC’s running solar cycle progression chart shows those indices trending downward from their higher-than-expected maximum. Solar Cycle 25 has crested. But “crested” is not the same as “calm,” and the distinction matters enormously for anyone who cares about auroras, satellite safety, or grid reliability.
The May 2024 storm proved what a fading cycle can still do
The single most dramatic illustration came on May 10, 2024, when a barrage of coronal mass ejections from sunspot region AR3664 slammed into Earth’s magnetic field and drove a G5 “extreme” geomagnetic storm, the highest tier on NOAA’s five-level scale. The USGS Geomagnetism Program recorded a peak Dst of -351 nT, a measure of how severely Earth’s magnetic field was compressed. The Space Weather Prediction Center had issued its first G4 watch since 2005 in advance of the storm window, flagging multiple Earth-directed CMEs.
The last G5 event before that had occurred during the Halloween storms of October–November 2003. Before 2003, you have to reach back to the great Quebec blackout storm of March 1989, which knocked out Hydro-Québec’s entire grid for nine hours and left six million people without power. These are not frequent occurrences. They are the kind of events that define a solar cycle’s legacy.
The May 2024 storm was not just a visual spectacle. A peer-reviewed study published in Geophysical Research Letters documented the thermospheric response, showing that AR3664’s repeated CME launches caused measurable increases in upper-atmosphere density and drag on low-orbit satellites. Enhanced currents in near-Earth space affected GPS accuracy and satellite operations. The storm was a system-wide event, not merely a pretty light show.
Why the declining phase can be more dangerous, not less
The pattern is counterintuitive but well-documented across previous solar cycles. As sunspot numbers drop, the remaining active regions tend to be fewer but can be large, magnetically complex, and capable of launching sequences of fast CMEs in rapid succession. When multiple CMEs merge or compress the solar wind ahead of them on the way to Earth, the resulting geomagnetic impact can far exceed what any single eruption would produce. AR3664 followed this script almost perfectly, firing off repeated eruptions over several days that interacted with each other in interplanetary space before arriving as a combined punch.
The strongest statistical framework supporting this idea comes from research applying extreme-value statistics to the Ap geomagnetic index across many solar cycles. That work, circulated as a preprint on arXiv, estimates return levels for severe storms in the roughly 10- to 100-year range and finds that the probability of extreme geomagnetic activity is modulated by solar cycle phase. The conclusions are framed in confidence intervals rather than single-number predictions for any given year, but the overall message is clear: the years after a solar maximum are historically fertile ground for the biggest storms.
No official NOAA or NASA product currently assigns a specific probability to another G5-class storm occurring during the remainder of Solar Cycle 25’s decline. Forecast discussions acknowledge the potential, but they stop short of long-term odds. Readers should treat the “rising odds” framing as a reasonable interpretation of historical patterns, not a precise forecast for a specific season.
What forecasters can and cannot tell you right now
For short-range prediction, the Space Weather Prediction Center relies on the OVATION aurora model, which generates 30- to 90-minute forecasts of where auroras will be visible based on real-time solar wind measurements. The model’s “view line” represents the equatorward boundary of auroral visibility and is calibrated in part using citizen science reports, according to a study archived in the NASA Technical Reports Server. If a storm is already underway, OVATION can tell a skywatcher in, say, Kansas City whether the auroral oval might reach their latitude tonight.
What OVATION cannot do is tell anyone whether another G5 storm will arrive this year, next year, or at all before Solar Cycle 25 bottoms out. That gap between real-time nowcasting and long-term probability remains one of the biggest open problems in operational space weather forecasting. Until it narrows, the practical approach is to pair awareness of the lingering risk with close attention to SWPC watches, warnings, and the Kp-based storm scale when activity flares.
What this means for 2026 and beyond
For operators of power grids, satellite constellations, and aviation routes, the takeaway is that preparedness should not relax just because the sunspot count is falling. Contingency plans for rare but high-impact storms, including procedures for powering down sensitive equipment, adjusting satellite orbits to compensate for atmospheric drag, and rerouting polar flights when radiation levels spike, remain as relevant now as they were at the cycle’s peak.
For the rest of us, the message is simpler and more hopeful. The most spectacular aurora display of Solar Cycle 25 may not have happened yet. The May 2024 storm pushed naked-eye auroras to latitudes where many people had never seen them and never expected to. Another large, magnetically complex sunspot region, arriving during a period when the solar wind is already primed by earlier activity, could do it again. A quieter Sun on average does not rule out a single night that people remember for the rest of their lives.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
