Link: NASA and IBM release Surya, an open-source AI model trained on over a decade's worth of NASA solar data to predict solar flares and winds (Peter Hall/MIT Technology Review)

NASA and IBM's collaboration has produced Surya, a new open-source machine learning model designed to enhance predictions of solar activities. Surya, which utilizes over a decade of solar data from NASA, aims to warn scientists early about potentially hazardous solar flares.

Solar storms can disrupt technology on Earth and in space, making accurate predictions crucial for mitigation. Surya can forecast solar flares, potentially doubling the warning time compared to current technologies.

The model was trained using data from NASA’s Solar Dynamics Observatory, amassing over 250 terabytes of solar imagery. This massive dataset helps Surya predict various solar events with greater accuracy.

Despite advancements, predicting the exact timing and intensity of solar flares remains a significant challenge. Surya's effectiveness lies in spotting hidden patterns that could precede solar destabilizations.

Moreover, IBM researcher Juan Bernabe-Moreno sees potential in using Surya to explore connections between solar and Earth's weather patterns. Understanding these links could lead to breakthroughs in predicting Earth weather events influenced by solar activity.

Beyond solar flares, Surya may provide insights into the physics of other celestial bodies. NASA and IBM envision Surya as a versatile tool akin to large language models, broad in application and rich in potential discoveries.

 #

--

Yoooo, this is a quick note on a link that made me go, WTF? Find all past links here.