Weeks-ahead solar storm warnings

- Researchers used physics-informed AI to produce solar-storm warnings that extend forecasting windows to weeks. (x.com) - The technique is designed to help utilities and satellite operators prepare earlier for geomagnetic storms. (x.com) - Early coverage emphasizes potential operational benefits for grids and spacecraft resilience during active solar periods. (x.com)

Solar storms can knock out satellites, scramble GPS and strain power grids, and researchers now say they can see some of the risk building weeks earlier than before. (ucar.edu) The new system, called PINNBARDS, was developed by the National Science Foundation National Center for Atmospheric Research and Southwest Research Institute and described in research published February 18, 2026, in *The Astrophysical Journal*. (swri.org, iopscience.iop.org) Space-weather forecasters usually get useful warning only hours before an eruption, because many tools rely on small magnetic signatures that become predictive shortly before a flare or coronal mass ejection. PINNBARDS is built to work earlier by tracing the problem back to where those dangerous regions form inside the Sun. (swri.org, nasa.gov) The basic idea is that solar active regions — the magnetically tangled patches that spawn big eruptions — do not appear at random. They cluster along warped magnetic bands, and those bands carry clues about deeper magnetic structures in the Sun’s tachocline, a transition layer below the surface. (swri.org, iopscience.iop.org) PINNBARDS uses a physics-informed neural network, which means the artificial intelligence is trained with both observations and equations from magnetohydrodynamics, the branch of physics that describes electrically charged gas and magnetic fields. The team said that lets the model reconstruct hidden magnetic conditions beneath the solar surface instead of only spotting patterns in past data. (ncarprojects.ucar.edu, meteorologicaltechnologyinternational.com) For its demonstration, the team used a February 14, 2024 synoptic magnetic map from the Solar Dynamics Observatory’s Helioseismic and Magnetic Imager to infer subsurface conditions and generate starting points for forward simulations of solar magnetic evolution. Those simulations were run on NSF NCAR’s Derecho supercomputer. (swri.org, ucar.edu) The practical target is earlier notice for operators who cannot do much with a 30-minute or same-day alert. The research teams said longer lead times could help agencies and companies prepare satellites, communications networks, GPS-dependent systems, astronauts and electric grids for periods of higher risk. (ucar.edu, nasa.gov) That need is not theoretical. NASA says geomagnetic storms during the 2003 Halloween events disrupted communications, aircraft and spacecraft, and a 1989 storm caused a 12-hour blackout across Quebec. (nasa.gov) The researchers are not claiming full operational forecasts are solved. Their papers and statements describe PINNBARDS as a “first step” that opens the door to predicting where and when large, flare-producing active regions are likely to emerge weeks in advance. (ucar.edu, swri.org) What changed in 2026 is not the Sun’s behavior but the forecasting window: researchers say they can now begin estimating solar-storm risk before the most dangerous regions even break through the surface. (meteorologicaltechnologyinternational.com, ncarprojects.ucar.edu)

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.