A new artificial intelligence model can detect early, hidden signs of solar eruptions nearly nine hours before they become visible. This breakthrough offers a crucial early warning system for potentially damaging space weather events that can impact Earth. An NJIT-led research team developed the AI system, named EarlyDetect, in collaboration with Princeton University and NASA's Ames Research Center.The findings were published on August 14, 2026, in the Journal of Geophysical Research: Machine Learning and Computation.[news+7]
New Tool Gives Early Warning for Solar Storms
Before dark sunspots appear on the Sun's surface, a new active region begins to show subtle signs of its formation. These active regions are where powerful solar eruptions can start.Until now, scientists struggled to capture these faint signals.The EarlyDetect model can forecast the emergence of these solar active regions an average of 9.24 hours in advance.[news+5]
Jonas Tirona, an undergraduate researcher at the New Jersey Institute of Technology (NJIT) and the study's lead author, developed the approach.He worked with computer scientists and solar physicists from NJIT, Princeton University, and NASA's Ames Research Center."The most valuable thing this work shows is that we can use machine learning to predict when solar active regions will emerge in advance," Tirona said.[news+5]
How Scientists Found Hidden Solar Activity
The EarlyDetect model analyzes hourly acoustic power maps and magnetic field measurements from NASA's Solar Dynamics Observatory (SDO).Active regions, which are magnetically intense areas, begin forming over several hours or days.As magnetic fields rise toward the Sun's surface, they leave faint signatures in acoustic waves.Scientists detect these wave patterns using helioseismology, which is the study of solar vibrations.[news+7]
The AI model uses a Transformer architecture, a type of technology also found in large language models.It learns patterns in solar data.Researchers made a surprising discovery: standard data filtering techniques, meant to highlight short-term patterns, actually removed the faint fluctuations most useful for prediction.By discarding this noise-canceling approach, the model could capture the critical signals indicating an active region's imminent rise.[techexplorist+5]
Protecting Earth from Sun's Powerful Blasts
Early warnings about solar eruptions are crucial for protecting modern technology and infrastructure on Earth.Solar storms pose significant risks to our technology-dependent society.Powerful solar events can energize Earth's ionosphere, leading to major GPS errors or complete loss of satellite communication signals.[news+6]
These storms also threaten power grids.Geomagnetically induced currents from coronal mass ejections can overload transformers and trigger widespread power outages.In commercial aviation, solar flares can disrupt radio communications and navigation systems.They also expose high-altitude flights to increased radiation.The stakes are even higher for human spaceflight, as astronauts on missions to the Moon or Mars need protection from solar radiation.[science+7]
"That early warning could allow satellite communications companies or power grid companies to prepare and potentially mitigate damage from solar storms," Tirona explained.Such preparation might include temporarily taking sensitive systems offline or moving satellites to different orbits.[news+2]
The Next Steps for Space Weather Forecasts
While EarlyDetect shows great promise, it is not yet ready for real-time forecasting.The model sometimes produces false alarms or late predictions.Also, an emergence warning does not guarantee that a solar flare or coronal mass ejection will follow, as many active regions never produce major eruptions.The research team needs to validate the approach across many more solar events.[techexplorist+6]
To encourage further research and collaboration, the team released the Solar Active Region Emergence Dataset (SolARED) and the SAR Portal.These are the first public resources specifically for studying active region emergence.NJIT data scientist Mengjia Xu noted that this dataset provides a shared resource for both the machine learning and heliophysics communities to develop and test new prediction methods.[techexplorist+5]
Other AI efforts are also advancing solar forecasting. In August 2025, NASA, in partnership with IBM, launched the Surya Heliophysics Foundational Model.Surya can generate visual predictions of solar flares two hours into the future, surpassing existing benchmarks by 16 percent.This model was trained on nine years of SDO data, analyzing 250 terabytes of high-resolution solar images.Surya is open-source, allowing researchers worldwide to use and explore it.[science+6]
This new AI model represents a significant step forward in predicting solar activity. By providing earlier warnings, scientists hope to give Earth-based industries and space missions more time to prepare for and lessen the impact of powerful solar storms.[news+2]
