NASA and IBM Unveil AI Model 'Surya' to Predict Solar Flares with Unprecedented Accuracy

NASA and IBM built an AI to predict solar flares before they hit Earth

An AI model trained on years of data from NASA's Solar Dynamics Observatory can predict the sun's future appearance and potentially flag dangerous solar flares An artificial intelligence model trained on NASA satellite imagery can forecast what the sun will look like hours into the future - even predicting the appearance of solar flares that may warn of dangerous space weather for Earth. "I love to think of this model as an AI telescope where you can look at the sun and you can understand the moods," says Juan Bernabé-Moreno at IBM Research Europe. The sun's moods matter because outbursts of solar activity can bombard Earth with high-energy particles, X-rays and extreme ultraviolet radiation. These can disrupt GPS and communications satellites, and potentially harm astronauts and even people on commercial airlines. Solar flares can be followed by coronal mass ejections, which may disrupt Earth's own magnetic field and create geomagnetic storms capable of knocking out power grids. Bernabé-Moreno and his colleagues at IBM and NASA trained an AI model called Surya, after the Sanskrit word for sun, on nine years of data from NASA's Solar Dynamics Observatory. The satellite captures ultra-high-resolution images of the sun in 13 different wavelengths. The AI model learned to identify patterns in the visual data and generate images of what the sun would look like from the observatory's point of view in the future. When tested on historical solar flare data, the Surya model predicted the occurrence of a solar flare within the next day with 16 per cent better accuracy than a standard machine learning model. It could also generate the visual image of a flare the observatory would see up to two hours in the future. "The power of AI is that it has the ability to learn the physics in a more roundabout way - it kind of develops an intuition for how the physics works," says Lisa Upton at Southwest Research Institute in Colorado. Upton says she is especially interested in whether the Surya model can help predict solar activity on the far side of the sun and at the poles, where NASA's scientific instruments can't make direct observations. Surya does not explicitly attempt to model the far side of the sun, but it has still proven successful in predicting how the sun will look several hours in the future, when part of the far side has rotated into view, says Bernabé-Moreno. But it is unclear if the AI model can address existing challenges in predicting exactly how solar activity may impact Earth, says Bernard Jackson at the University of California, San Diego. That is because there is currently no way to directly observe the magnetic field configurations between the sun and Earth, which is what determines the paths of the high-energy particles travelling outward from our star. Bernabé-Moreno says the model is currently intended for use by scientists, but future integrations with other AI systems that can harness Surya's capabilities to answer basic questions about future solar activity might make it more accessible to power grid operators or satellite constellation owners as part of an early warning system.

IBM and NASA Develop a Digital Twin of the Sun to Predict Future Solar Storms

The tool models the sun using AI, and its developers say it can anticipate solar flares 16 percent more accurately and in half the time of current prediction systems. The Sun's most complex mysteries could soon be solved thanks to artificial intelligence. On August 20, IBM and NASA announced the launch of Surya, a foundation model for the sun. Having been trained on large datasets of solar activity, this AI tool aims to deepen humanity's understanding of solar weather and accurately predict solar flares -- bursts of electromagnetic radiation emitted by our star that threaten both astronauts in orbit and communications infrastructure on Earth. Surya was trained with nine years of data collected by NASA's Solar Dynamics Observatory (SDO), an instrument that has orbited the sun since 2010, taking high-resolution images every 12 seconds. The SDO captures observations of the sun at various different electromagnetic wavelengths to estimate the temperature of the star's layers. It also takes precise measurements of the sun's magnetic field -- essential data for understanding how energy moves through the star, and for predicting solar storms. Historically, interpreting this vast amount of diverse and complex data has been a challenge for heliophysicists. To address this challenge, IBM says that Surya's developers used the SDO data to create a digital twin of the sun -- a dynamic virtual replica of the star that is updated when new data is captured, and which can be manipulated and more easily studied. The process began with unifying the various data formats fed into the model, allowing it to process them consistently. Next, a long-range vision transformer was employed -- AI architecture that enables detailed analysis of very high-resolution images and the identification of relationships between their components, regardless of their distance. The model's performance was optimized using a mechanism called spectral gating, which reduces memory usage by up to 5 percent by filtering out noise in the data, thereby increasing the quality of the processed information. Its developers say that this design gives Surya a significant advantage: Unlike other algorithms that require extensive labeling of the data that's fed to them, Surya can learn directly from raw data. This allows it to quickly adapt to different tasks and deliver reliable results in less time. During testing, Surya demonstrated its versatility in integrating data from other instruments, such as the Parker Solar Probe and the Solar and Heliospheric Observatory (SOHO), two other spacecraft that observe the sun. Surya also proved to be effective in various predictive functions, including predicting flare activity and solar wind speed. According to IBM, traditional prediction models can only predict a flare one hour in advance based on signals detected in specific regions of the sun. In contrast, "Surya provided a two-hour lead by using visual information. The model is thought to be the first to provide a warning of this kind. In early testing of the model, the team said they achieved a 16 percent improvement in solar flare classification accuracy, a marked improvement over existing methods," the company said in a statement. NASA stresses that, although the model was designed to study heliophysics, its architecture is adaptable to different fields, from planetary science to Earth observation. "By developing a foundation model trained on NASA's heliophysics data, we're making it easier to analyze the complexities of the sun's behavior with unprecedented speed and precision," said Kevin Murphy, NASA's director of data science, in a statement. "This model empowers broader understanding of how solar activity impacts critical systems and technologies that we all rely on here on Earth." The risk posed by abnormal solar activity is not minor. A major solar storm could directly affect global telecommunications, collapse electrical grids, and disturb GPS navigation, satellite operations, internet connections, and radio transmissions. Andrés Muñoz-Jaramillo, a solar physicist at the Southwest Research Institute in San Antonio, Texas, and lead scientist on the project, emphasized that Surya's goal is to maximize the lead time for these possible scenarios. "We want to give Earth the longest lead time possible. Our hope is that the model has learned all the critical processes behind our star's evolution through time so that we can extract actionable insights."

Meet Surya, a New AI Model From NASA and IBM That Can Predict Solar Flares

The new AI model may help NASA and NOAA better forecast auroras. Today's solar flare forecasting is pretty straightforward: Earth has several instruments pointed at the sun that monitor it, and when a solar flare erupts, NOAA predicts whether it'll hit Earth and reports it via the Space Weather Prediction Center. But NASA and IBM may be able to do this faster and with a little more accuracy thanks to a new artificial intelligence model named Surya. Surya, which is Sanskrit for "Sun," is a 366M-parameter AI model developed for the purpose of analyzing the various cool things the sun does. It's an open-source model available on GitHub for anyone to play with. Per the GitHub page, the model "learns general-purpose solar representations through spatiotemporal transformers, enabling state-of-the-art performance in solar flare forecasting, active region segmentation, solar wind prediction and EUV spectra modeling." The AI model has been trained on data about our nearest star, and uses that data to predict things like whether a solar flare is likely to hit Earth. IBM says that it's trained using data from NASA's Solar Dynamic Observatory, which has been monitoring the sun since 2010, along with eight other research centers. "We want to give Earth the longest lead time possible," said Andrés Muñoz-Jaramillo, solar physicist at SouthWest Research Institute and lead researcher on Surya. "Our hope is that the model has learned all the critical processes behind our star's evolution through time so that we can extract actionable insights." In short, researchers are hoping to use AI to forecast when a solar flare may hit Earth, giving the longest possible warning that a geomagnetic storm is approaching. Of course, since aurora borealis comes from the effect that geomagnetic storms have on the Earth's magnetic field, that would also mean we would know when auroras are happening much further in advance. The primary reason for NASA's Solar Dynamic Observatory was to piece together the underlying physics of the sun, but as IBM notes, the process has been slow going and science still doesn't know as much as it would like to about how the sun works. Prior to Surya, NASA was using things like flashes of light in the sun's corona to partially predict solar flares. NOAA has its own prediction methods as well that work well, but have limitations. IBM says the inclusion of Surya may help make those predictions more accurate and timely. "We've been on this journey of pushing the limits of technology with NASA since 2023, delivering pioneering foundational AI models to gain an unprecedented understanding of our planet Earth," said Juan Bernabé-Moreno, IBM's director in charge of scientific collaboration with NASA. "With Surya we have created the first foundation model to look the sun in the eye and forecast its moods."

IBM, NASA cook up AI model to predict solar tantrums

Open source Surya system promises early alerts for space weather that can fry satellites and grids Boffins at IBM and NASA have concocted an AI model to help predict the weather, but this time it is taking on space weather that might disrupt satellites and spacecraft, possibly even terrestrial power grids and the internet. IBM's hope is that the model has learned enough of the Sun's behavior so that it is able to generate actionable insights - in other words, warn of impending danger Big Blue and the US space agency are open sourcing their latest AI model, named Surya, the Sanskrit word for the Sun. This has been developed to predict the kind of violent solar flare-ups that can send a torrent of charged particles toward the Earth, potentially damaging electronic infrastructure. A famous example of solar activity causing havoc is the Carrington Event of 1859, when a geomagnetic storm caused telegraph systems in Europe and North America to fail, sometimes catching fire because of induced currents. Modern-day electronic infrastructure is much more complex and sensitive. The Surya model has been trained using nine years' worth of data from NASA's Solar Dynamics Observatory (SDO) satellite, which has been closely monitoring the Sun for some time. IBM said that its scientists, along with those from NASA and eight other research centers, used the data to build their foundation model of solar physics. "Think of this as a weather forecast for space," said Juan Bernabe-Moreno, Director of IBM Research Europe for Ireland and UK. "Just as we work to prepare for hazardous weather events, we need to do the same for solar storms." He added that Surya is intended to anticipate what will happen next. IBM's hope is that the model has learned enough of the Sun's behavior so that it is able to generate actionable insights - in other words, warn of impending danger. The researchers claim a 16 percent improvement in solar flare classification accuracy in early tests, and hail this as a substantial improvement compared to previous methods. Surya has also been developed to visually predict solar flares for the first time, providing a high-resolution image of where the flare is predicted to occur up to two hours ahead. NASA and IBM previously worked together on an open source AI climate model called Prithvi that was developed to predict weather patterns here on Earth, while using fewer compute resources. Building on that model but trying something different, the researchers say they gave the Surya model sequential images from SDO, then had it envision what the observatory would see an hour in the future. They were then able to check the accuracy of its prediction against actual observed data. However, IBM says the solar images used to train the model are ten times larger than typical AI training data, requiring a specialized multi-architecture solution to handle it all. Surya is available on Hugging Face, GitHub, and IBM's TerraTorch library for fine-tuning geospatial AI models. The research team is also open sourcing SuryaBench, a set of curated data sets and benchmarks intended to simplify building and evaluating applications for space weather forecasting. ®

IBM and NASA create first-of-its-kind AI that can accurately predict violent solar flares

The model's training data comes from nine years' worth of images from the Solar Dynamic Observatory (SDO) satellite, which orbits the Earth, snapping pictures every 12 seconds. (Image credit: IBM-NASA-Surya) IBM and NASA scientists have unveiled a groundbreaking artificial intelligence (AI) model that can predict the sun's ferocious outbursts more accurately than ever, giving us a chance to react to dangerous and disruptive solar activity. The new AI model, known as "Surya" (Sanskrit for the sun), absorbs the raw images that are captured by the Solar Dynamic Observatory (SDO) satellite -- which has been staring directly into the sun for the last 15 years -- and processes them quicker than any humans can. Using this raw data, which IBM representatives said in a statement researchers have barely scratched the surface of, the foundational model can predict violent outbursts before they happen. That way, we can protect astronauts and equipment in space, and even plan for disruption to power grids and communications systems on Earth. "We've been on this journey of pushing the limits of technology with NASA since 2023, delivering pioneering foundational AI models to gain an unprecedented understanding of our planet Earth," Juan Bernabé-Moreno, the director of IBM Research Europe for the U.K. and Ireland who is in charge of scientific collaboration with NASA, said in the statement. "With Surya we have created the first foundation model to look the sun in the eye and forecast its moods." Solar activity has a growing impact on our lives the further we venture into space, and the more we rely on technology on Earth. Solar flares and coronal mass ejections can knock out satellites, disrupt airline navigation, trigger power blackouts and pose a radiation risk to astronauts, making accurate solar weather prediction increasingly important. Related: AI is entering an 'unprecedented regime.' Should we stop it -- and can we -- before it destroys us? Forecasting storms on Earth is notoriously difficult, the scientists said, and predicting solar storms is even tougher. When solar flares erupt through the sun's magnetic field, it takes eight minutes for that light to reach our eyes -- this lag (the eight minutes in which we have no visibility over what has happened) means that scientists need to be even further ahead. The Soraya AI model is comparable to the separate "Prithvi" family of AI models. These models process gigantic volumes of satellite data to create a more accurate representation of Earth in order to better predict its climate and weather, alongside completing other tasks such as mapping deforestation, measuring the impact of flooding and projecting the effect of extreme heat. The Soraya model is an open-source, 360-million-parameter system designed to learn solar representation through eight Atmospheric Imaging Assembly (AIA) channels and five Helioseismic and Magnetic Imager (HMI) products. AIA is designed to provide different views at the top of the sun's atmosphere, known as the solar corona -- taking images that span 1.3 solar diameters in multiple wavelengths to improve the understanding of the physics behind what we can observe in the sun's atmosphere. HMI, meanwhile, is an instrument that studies oscillations and the magnetic field at the sun's surface. The system can accurately forecast solar dynamics, solar wind and solar flares, and detect extreme ultraviolet (EUV) spectra.. The scientists say that the novel architecture of Soraya means it can learn the underlying physics behind solar evolution. They outlined their findings in a study uploaded Aug. 18 to the arXiv preprint database, meaning that it has not yet been peer-reviewed. "This is an excellent way to realize the potential of this data," Kathy Reeves, a solar physicist at the Harvard-Smithsonian Center for Astrophysics, who was not involved in the study said in the statement. "Pulling features and events out of petabytes of data is a laborious process and now we can automate it." The model's data comes from the SDO, which orbits the Earth, snapping pictures every 12 seconds. These images capture the sun at different wavelength bands to take the temperature of its layers, which vary from 5,500 degrees Celsius on the surface to up to 2 million degrees Celsius at the corona. SDO also captures magnetic activity, with emerging sunspots revealed in white light while other imaging tools check the speed of bubbles on the surface and track the twisting of the sun's magnetic lines. Researchers trained Soraya by taking a nine-year excerpt of this data, first harmonizing the different layers -- meaning the different types of data are amalgamated to create a more holistic picture -- and then experimenting with different AI architectures to process it. With Soraya, they challenged the model to take sequential images and then predict what SDO would see an hour into the future -- checking these predictions against the actual observed images. The sun also has various quirks that the scientists attempted to hardcode into the model -- including the fact that the sun rotates faster at its equator than at its poles. Yet, remarkably, they found that Soraya was more effective at learning these quirks on its own, from the data, than through any human input. In testing, the AI model could forecast whether an active region was likely to set off a solar flare an hour before it happened, and in some experiments, they achieved predictions within two hours (when led by visual information). This represents a 16% improvement on existing prediction methods, IBM representatives said in the statement. The team has made the AI model open-source, and it is now available on GitHub and Hugging Face -- an open-source platform that hosts AI models and datasets. SorayaBench, a curated set of datasets and benchmarks aimed at helping researchers better understand the behavior of the sun, is also available to access freely.

IBM and NASA made an open-source AI model for predicting solar weather

Last year, the most powerful geomagnetic storm in 20 years . It produced stunning aurora displays in parts of the US that are normally too far south to see them. Normally, such storms are a headache for energy providers. In 1989, for example, the Canadian province of blackout following a series of plasma ejections from the Sun. This time around, power companies were better prepared, and in the US and Canada, there weren't significant service disruptions. The episode highlighted the value of proper preparation against geomagnetic storms, and for the past couple of years, NASA and IBM have been working to give the scientific community and others a better way to predict solar weather. Today, they're releasing the result of their work, an open-source foundation model called Surya. Named after the Sanskrit word for the Sun, Juan Bernabe-Moreno, director of IBM Research Europe, UK and Ireland, describes the system as an "AI telescope for the Sun." IBM trained the model on nine years of high-resolution images from the (SDO), a satellite NASA has been using to study the Sun since 2010. Effectively, Surya applies machine learning to solar image interpretation and forecasting, and the results are promising. In early testing, Bernabe-Moreno says IBM found the model was 16 percent more accurate at answering the question "will there be a solar flare in the next 24 hours?" than past systems. Additionally, the model can generate visual predictions of what the SDO might see ahead of time. So far, using data from the most , IBM has found Surya can accurately predict what the sun will look like two hours ahead of time. "We're exploring the accuracy of even longer lead time predictions," Bernabe-Moreno told me. Two hours might not seem like a lot, but according to Bernabe-Moreno, who previously worked at one of Europe's largest energy companies, it could be a game changer for infrastructure providers, which have spent the last few decades building more responsive power grids. Moreover, Surya is a 366 million parameter model, meaning it's light enough to run on less powerful hardware. The irony of today's announcement is that it demonstrates the value of NASA's science team exactly at a time when its very existence is threatened. If you haven't been following what's been going on at the agency, President Trump plans to . The Solar Dynamics Observatory would be among the missions affected by the proposed cuts. It won't be cancelled like New Horizons and OSIRIS-APEX, but according to an , the mission would have its operating budget slashed from $14 million annually to $8 million per year. Policymakers from both parties have pushed back on the proposal, but with the Senate and House not yet in agreement on the agency's 2026 operating budget, and the appropriations deadline quickly approaching, NASA's fate is uncertain. Even if the cuts don't go through, the agency is on track to , or about 20 percent of its workforce, as part of the Trump administration's broader efforts to trim the number of workers across the federal government. Bernabe-Moreno says Surya wouldn't have come together without NASA's help. "From the very beginning, the science team at NASA have been informing us what the model should do, how to validate the model, and how to ensure it's robust," he said. If there's a silver lining to the funding situation, it's that the science team's work will live on. "The beauty of this model is that we created a capability -- an AI platform, if you want," said Bernabe-Moreno. "And this capability has applications beyond NASA." If you want to check out Surya for yourself, you can download the model from .

NASA and IBM Unveil AI That Helps Scientists Forecast Solar Storms

Experts recently identified major gaps in their ability to forecast space weather. Surya aims to fill them. Earlier this year, local and national officials gathered for a first-of-its-kind tabletop exercise to test their readiness for a severe solar storm. The simulation exposed major gaps in scientists’ ability to forecast space weather, which threatens critical infrastructure on Earth and in orbit. On Wednesday, August 20, IBM and NASA unveiled Surya: an open-source AI model that could begin to fill those gaps. Heliophysicists currently rely on complex computer models to monitor and predict the Sun’s activity. Surya improves upon the lead time and accuracy of existing solar forecasting technologies, allowing scientists to not only predict a solar flare two hours out but also visually pinpoint where it should occur on the Sun’s surface, according to IBM. “It’s very important to have a mechanism to look into the Sun and understand how these events [are going to happen], when they’re going to happen, why they're going to happen, and start predicting the occurrence so that we can be prepared,†Juan Bernabé-Moreno, director of IBM Research Europe for Ireland and the U.K., told Gizmodo. The surface of the Sun is a violent place. Our host star is constantly emitting bursts of energy such as solar flares, high-speed solar winds, and coronal mass ejections. When Earth is in the line of fire during one of these events, the onslaught of high-energy particles can trigger a geomagnetic storm in the upper atmosphere. Such storms result from disturbances in Earth’s magnetic field and can damage or disrupt satellites, power grids, and radio communication systems, according to NASA. Being able to anticipate these outbursts gives decision-makers critical lead time to protect vulnerable infrastructure, potentially avoiding billions of dollars in damage. According to a systemic risk analysis by Lloyd’s, a severe solar storm could result in losses to the global economy of $2.4 trillion over a five-year period. Bernabé-Moreno thinks of Surya as a powerful AI telescope that also lets you look into the future. Whereas traditional solar weather prediction relies on partial satellite views of the Sun’s surface, Surya trained on nine years of high-resolution solar observation data gathered by NASA’s Solar Dynamics Observatory. This telescope launched in 2010 and has been continuously observing the Sun for the past 15 years, capturing images every 12 seconds at various wavelengths to take the temperature of its layers and map magnetic activity. Heliophysicists will be able to use Surya in a variety of ways, but its most novel application is solar flare prediction, IBM senior research scientist and technical project lead Johannes Schmude told Gizmodo in an email. The model does this by generating an image of an event that the SDO satellite is likely to see, essentially predicting what the surface of the Sun will look like hours ahead. Testing showed that Surya can predict a solar flare two hours in advance with a 16% improvement in flare classification accuracy, but IBM is exploring the accuracy of even longer lead time predictions, according to Schmude. It’s important to note, however, that Surya trained on data from the previous solar cycle. “Testing the model’s applicability to Solar cycle 25 is one of the post-release tasks on our list, but we plan to explore continuous training and other fine-tuning with data from Solar Cycle 25,†Schmude said. Following the release of this open-source AI model, Bernabé-Moreno is excited for the scientific community to begin using it on a wider scale, identifying new applications and challenging its capabilities. “That is going to create utility,†he said. “That, for us, is the most important thing.â€

You might see prettier skies, thanks to new tech from NASA and IBM

What if scientists could predict northern and southern lights like they could an eclipse? What if they could tell you where and when to be outside, within a narrow window, to see these vibrant displays? A new AI might make that possible. Today, IBM introduced Surya, an open-source foundational AI model that was developed in partnership with heliophysics scientists at NASA. "Surya is like an AI telescope for the sun that can also look into the future," explained Juan Bernabe Moreno, director of IBM research in Europe, the U.K., and Ireland. Not only can Surya model what the sun looks like now, but it can also predict our star's future behavior. This is key for understanding solar flares, and whether they will produce coronal mass ejections (CMEs) and subsequent geomagnetic storms, which cause northern lights. That's also important, as these can significantly disrupt life on Earth; a severe space weather risk scenario published by the London-based Lloyd's insurance marketplace presented possible global economic losses of up to $9.1 trillion over a five-year period.

IBM's and NASA's Surya AI model is designed to predict the next 'Carrington-class' solar storm - SiliconANGLE

IBM's and NASA's Surya AI model is designed to predict the next 'Carrington-class' solar storm IBM Corp. said today it's teaming up with experts from the National Aeronautics and Space Administration to create an artificial intelligence model that can provide an early warning about solar flares and coronal mass ejections that could cause serious disruption to life on Earth. The new open-source model, called Surya, named after the Sanskrit word for "sun," is available to download on Hugging Face. IBM said it's designed to interpret high-resolution solar images so it can forecast space weather conditions and help protect satellites in low-Earth orbit, power grids, telecommunications systems and other infrastructure that could be severely affected by strong solar flares. Scientists know only too well how damaging solar flares can be. Although the sun is about 93 million miles away, such events can knock out satellites, disrupt airline navigation and take energy grids offline, and they also pose major radiation risks to any astronauts who happen to be in orbit. Although these incidents are rare, they do happen. A case in point is the so-called "Carrington Event" in September 1859, which was a massive solar storm caused by a powerful coronal mass ejection from the Sun, directed at Earth. The flare was observed by the British Astronomer Richard Carrington 17 hours before the event, and resulted in spectacular auroras visible as far south as the Caribbean. It led to severe disruption of telegraph systems worldwide, causing operators to receive electric shocks and equipment to catch fire. There was even a report of two telegraph operators who shut down their systems, but were still able to communicate with one another given the electrical current in the atmosphere caused by the aurora. One study by Lloyd's suggests that if a similar event occurred today, it would likely cause extensive damage to electrical grids, satellites and communication systems, with economic losses estimated at about $2.4 trillion. There have been plenty of close calls in recent years. Less severe solar storms hit Earth in 1921 and 1938, causing massive radio disruption, while the March 1989 geomagnetic storms took down power grids in large parts of Quebec, Canada. Another "Carrington-class" solar storm occurred in July 2012, but its trajectory missed the Earth by a margin of just nine days. With humanity becoming increasingly dependent on space-based technology and focused on further exploration of the solar system, it's essential to be able to predict the behavior of the sun, and that's exactly what Surya has been designed to do. If experts can accurately forecast a solar flare, they might be able to shut down at-risk infrastructure systems and prevent damage from being done to them. Current solar weather prediction relies on partial satellite views of the Sun's surface, but Surya significantly improves on this. It's trained on the world's largest and highest-resolution heliophysics dataset, which was created to help researchers study and evaluate space weather prediction tasks. Some example tasks, which Surya was trained on, include solar flare prediction, solar EUV spectra forecasting, the speed of solar winds and the emergence of active regions on the Sun's surface. IBM's and NASA's researchers say early tests show that Surya has achieved a 16% improvement in the accuracy of solar flare classifications. They intend to continue training the model to visually predict solar flares for the first time, with the aim being to provide a two-hour warning before such events take place. IBM Research Europe Director Juan Bernabe-Moreno said Surya will provide weather forecasts for the sun. "Just as we work to prepare for hazardous weather events, we need to do the same for solar storms," he explained. "Surya gives us the unprecedented capability to anticipate what's coming and is not just a technological achievement, but a critical step toward protecting our technological civilization from the star that sustains us." Surya was reportedly trained on more than nine years' worth of high-resolution solar observation data from the NASA Solar Dynamics Observatory. The images are about 10 times larger than typical images used for AI training, so IBM had to develop a customized multi-architecture system to handle the increased spatial resolution. That's needed to resolve solar features with enough detail and context to try and understand what's happening. NASA Chief Science Officer Kevin Murphy said the combination of its deep scientific expertise and powerful AI is extremely promising for data-driven science. "By developing a foundation model trained on NASA's heliophysics data, we're making it easier to analyze the complexities of the Sun's behavior with unprecedented speed and precision," he said. "This model empowers broader understanding of how solar activity impacts critical systems and technologies that we all rely on here on Earth." Holger Mueller of Constellation Research Inc. said the model is a good example of where AI doesn't just replace humans, but does jobs that most humans cannot do. "Very few people in the world have the skills to even attempt to forecast solar weather, so AI changes the dynamic," the analyst said. Additionally, Mueller said the partnership between IBM and NASA shows how organizations can work together to continuously advance AI. "We have one partner with all of the data, and the other has the know-how and systems in place to build the models, and we will see many more of these kinds of partnerships in all areas." Surya is just the latest initiative in a long-running collaboration between IBM and NASA. Two years ago, they teamed up to create an AI model that's designed to analyze geospatial satellite imagery at unprecedented scale. It's one of the most powerful models within the Prithvi family, and is all about forecasting Earth-based risks, such as flooding and wildfires. Like Surya, it was open-sourced and made available on Hugging Face for the benefit of the broader research community. Last year, IBM and NASA released a second Prithvi model designed to provide short- and long-term weather projections and forecast climate change.

IBM, NASA Launch Open Source Model Surya for Solar Observation and Prediction | AIM

For the first time, a model can visually predict where a solar flare might occur, up to two hours in advance. IBM and NASA have released an open-source AI model to predict solar weather that could disrupt technology on Earth and in space. The model, named Surya, is available on the Hugging Face and uses high-resolution solar observation data to forecast solar activity, aiming to help protect satellites, navigation systems, power grids and telecommunications. Surya was developed using nine years of NASA's Solar Dynamics Observatory data and is designed to support scientists and industries in planning for solar storms. According to the IBM blog, solar flares and coronal mass ejections can disrupt airline navigation, damage satellites, and pose risks to astronauts. Juan Bernabe-Moreno, director of IBM Research Europe, UK and Ireland, said, "Think of this as a weather forecast for spac

NASA and IBM unveil Surya AI to forecast solar flares

NASA and IBM have collaborated to create Surya, a newartificial intelligence model designed to forecast solar flares. Unveiled recently, Surya aims to enhance prediction accuracy and provide earlier warnings of potential geomagnetic storms. The model is intended to improve upon existing forecasting methods by leveraging advanced AI techniques to analyze solar activity. On January 5, 2022, an X 1.2 class solar flare was recorded, highlighting the type of event Surya is designed to predict. Current solar flare forecasting depends on instruments monitoring the sun and the National Oceanic and Atmospheric Administration (NOAA) predicting potential Earth impacts through its Space Weather Prediction Center. Surya represents an effort to refine and expedite this process. Surya, named after the Sanskrit word for "Sun," is a 366-million-parameter AI model developed to analyze solar phenomena. IBM emphasized that the model is open-source and accessible on GitHub, encouraging further development and exploration by the scientific community. The GitHub page states that the model "learns general-purpose solar representations through spatiotemporal transformers, enabling state-of-the-art performance in solar flare forecasting, active region segmentation, solar wind prediction, and EUV spectra modeling." The AI model uses extensive data about the sun to predict events such as solar flares that may impact Earth. IBM reports that Surya is trained using data from NASA's Solar Dynamic Observatory (SDO), which has monitored the sun continuously since 2010. Data from eight additional research centers have also contributed to the model's training dataset, creating a comprehensive foundation for analysis. Andrés Muñoz-Jaramillo, a solar physicist at Southwest Research Institute and lead researcher on Surya, stated, "We want to give Earth the longest lead time possible." He added, "Our hope is that the model has learned all the critical processes behind our star's evolution through time so that we can extract actionable insights." The primary objective is to provide more advanced warnings of approaching geomagnetic storms caused by solar flares. Improved forecasting of geomagnetic storms could lead to earlier alerts for auroras. Knowledge of when auroras are likely to occur could be available further in advance, since auroras result from the interaction of geomagnetic storms with Earth's magnetic field. The Solar Dynamic Observatory was established to understand the underlying physics of the sun. IBM indicates that progress in this area has been gradual, and a complete understanding of solar activity remains elusive. Prior to Surya, NASA used indicators, such as flashes of light in the sun's corona, to predict solar flares. NOAA also employs its own prediction methods, which, according to IBM, generally perform well but have limitations. IBM suggests that Surya's inclusion could improve both the accuracy and timeliness of solar flare predictions. Juan Bernabé-Moreno, IBM's director in charge of scientific collaboration with NASA, said, "We've been on this journey of pushing the limits of technology with NASA since 2023, delivering pioneering foundational AI models to gain an unprecedented understanding of our planet Earth." Bernabé-Moreno added, "With Surya, we have created the first foundation model to look the sun in the eye and forecast its moods."

NASA employs AI to better predict solar activity

NASA announced Wednesday that it is deciphering the sun's behavior with help from artificial intelligence. In a press release, the said that, in conjunction with among other partners, it has developed an AI model dubbed the Surya Heliophysics Foundational Model, which has been trained on nine years' worth of data from . Surya is Sanskrit for "sun." "We are advancing data-driven science by embedding deep scientific expertise into cutting-edge AI models," said NASA Chief Science Data Officer in the release. "By developing a foundation model trained on heliophysics data, we're making it easier to analyze the complexities of the sun's behavior with unprecedented speed and precision." "This model empowers broader understanding of how solar activity impacts critical systems and technologies that we all rely on here on Earth," he added. Using solar data, Surya can analyze solar flares and make predictions regarding how space weather might impact technology such as communication systems and satellites, as well as power grids. Surya can also forecast how UV from the sun affects the Earth's upper atmosphere and determine solar wind speed. "Our society is built on technologies that are highly susceptible to space weather," said Heliophysics Division Director in the release. "Just as we use meteorology to forecast Earth's weather, space weather forecasts predict the conditions and events in the space environment that can affect Earth and our technologies." "We want to give Earth the longest lead time possible," said solar physicist in a press release from . "Our hope is that the model has learned all the critical processes behind our star's evolution through time so that we can extract actionable insights." NASA also reports that while Surya is designed for Sun study, it can be adapted to engage several types of scientific explorations, including observing the Earth and conducting planetary science. Additionally, both the model and training datasets from Surya are available to try out online at Hugging Face, GitHub and in TerraTorch library for fine-tuning geospatial AI models. A benchmark dataset called "SuryaBench" has also been open sourced to the public. "We've been on this journey of pushing the limits of technology with NASA since 2023, delivering pioneering foundational AI models to gain an unprecedented understanding of our planet Earth," said Juan Bernabé-Moreno, the director in charge of the scientific collaboration with NASA. "With Surya we have created the first foundation model to look the sun in the eye and forecast its moods." Copyright 2025 United Press International, Inc. (UPI). Any reproduction, republication, redistribution and/or modification of any UPI content is expressly prohibited without UPI's prior written consent., source