AI predicts destructive solar winds in advance

A team of scientists from a North American university has developed an artificial intelligence (AI) model capable of predicting the speed of destructive solar winds four days in advance, EFE reported on Wednesday.

“By combining advanced AI with solar observations, we can issue early warnings that help protect critical technologies on Earth and in space,” said Dattaraj Dhuri, a researcher at the Center for Astrophysics and Space Science at New York University Abu Dhabi (CASS), located in the United Arab Emirates, in a statement.

Current AI language models analyze text to predict solar activity, but the new system, developed in Abu Dhabi, analyzes images of the Sun to identify patterns related to changes in the solar wind.

The result is a 45% improvement in forecast accuracy compared to current operational models. The new model also represents a 20% improvement compared to previous AI-based approaches.

The team of scientists trained the AI ​​model using high-resolution ultraviolet images combined with historical solar wind records. The ultraviolet images are from NASA's Solar Dynamics Observatory.

The solar wind is a continuous stream of charged particles released by the Sun. When these particles accelerate, they can trigger space weather events that disrupt Earth's atmosphere, knocking satellites out of orbit. Particle acceleration can also trigger intense solar storms, which disrupt communications systems, power grids, and air traffic.

In 2022, a phenomenon related to solar winds led to the loss of 40 Starlink satellites from the aerospace company SpaceX, proving the urgent need to improve solar forecasts, according to EFE.

"This is a major step forward in protecting satellites, navigation systems, and energy infrastructure on which modern life depends," said Dattaraj Dhuri.

The breakthrough demonstrates how AI can solve one of the most complex challenges in space science: forecasting solar winds, according to EFE.

More reliable forecasts allow scientists and engineers to better prepare for space weather events, strengthening resilience to critical infrastructure disruptions, according to the same source.