Skip to main content

Roadmap to the Stars: Accelerating Innovation in Multi-Messenger Astrophysics

Photograph of the Milky Way Galaxy with beige, blue, and purple surrounded by scattered stars

In a new Expert Recommendation appearing in Nature Reviews Physics, representatives from multiple communities, including big data analytics, artificial intelligence, high performance computing, physics, and astronomy, come together to discuss a community roadmap for integrating innovations from the big data Revolution into multi-messenger astrophysics (MMA).

The paper, “Enabling real-time multi-messenger astrophysics discoveries with deep learning,” co-authored by NCSA’s Eliu Huerta, is the product of collaboration between government, academic, and industry groups looking to bring the exciting benefits of data-driven analysis and AI-based processing to MMA.

MMA combines data from cosmic messengers such as gravitational waves, neutrinos and electromagnetic waves. Recent accomplishments of this science program include the observation of two colliding neutron stars in gravitational and electromagnetic waves, an observational breakthrough that revolutionized our understanding of neutron star physics, gravitational waves, cosmology, and fundamental physics.

The realization of the science goals of multi-messenger astrophysics in the big data era will only be accomplished by synergistic efforts among disparate communities. To contribute to this international endeavor, the Gravity Group at the National Center for Supercomputing Applications initiated a science program at the interface of artificial intelligence, large scale computing and gravitational wave astrophysics. This article is a milestone in our community building efforts to harness the big data revolution to accelerate discovery in multi-messenger astrophysics.

Eliu Huerta, NCSA Gravity Group Head

As next-generation instruments such as the Large Synoptic Survey Telescope start observing the sky with unprecedented fidelity, complemented by gravitational wave and neutrino observations, it will be critical to address upcoming computational grand challenges with fresh strategies. “The NSF Office of Advanced Cyberinfrastructure is pleased to be supporting the exploration of new approaches and technologies that address data-driven challenges in Multi-Messenger Astrophysics,” says Manish Parashar, director of the Office of Advanced Cyberinfrastructure at the US National Science Foundation (NSF).

“I am very happy to see the opportunity to build a community that brings domain experts together with data scientists on equal footing to maximize and accelerate discovery,” says Zhizhen (Jane) Zhao, assistant professor in the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign. “I hope the community building and collaboration will lead to innovative ideas and methods that are transferable to other science and engineering fields.”

While the possibilities of real-time discovery in MMA are “inspiring and ambitious,” computer scientist Elise Jennings of the Argonne National Laboratory notes that “they also highlight new challenges in merging AI with high-performance computing.”

Daniel S. Katz, assistant director for Scientific Software and Applications at NCSA, sees potential in those challenges: “This is an exciting opportunity to build new software and applications using modern best practices.”

Along with applications, computing infrastructure needs to keep pace with — and in some cases forge a path for — scientific innovation. Tom Gibbs, manager of developer relations at NVIDIA, notes that NVIDIA is “delighted” to be actively involved in “groundbreaking efforts applied to grand challenge problems.” Jack Wells, director of science for the Oak Ridge Leadership Computing Facility at Oak Ridge National Laboratory, is similarly enthusiastic about bringing Oak Ridge’s capabilities to bear on these efforts:

“The US Department of Energy’s leadership computing facilities are unparalleled in their ability to meet the present-day data challenges… of both the volume and complexity of this data. Only through a collaborative effort that includes leadership-class infrastructures will accurate neural network models be fully developed for the advanced analyses of these rich and interconnected scientific data sets,” says Wells.

“The advent of multi-messenger astronomy represents one of the most exciting periods in modern physics, astronomy, and cosmology,” says Edward Seidel, Founder Professor of Physics at Illinois.

The paper’s authors are optimistic about the implications of their work: “We hope this article will advance the young science of multi-messenger astrophysics as we approach the era of huge data sets for transient phenomena. By using AI to link gravitational wave data with observations by ground-based telescopes… [we] will be able to exploit to the maximum the scientific discovery potential of our leading-edge observatories over the next decade or more,” predicts Bernard Schutz, deputy director of the Data Innovation Research Institute at Cardiff University and emeritus director of the Max Planck Institute for Gravitational Physics (Albert Einstein Institute) in Potsdam, Germany.

Jinjun Xiong, co-director for the IBM-ILLINOIS Center for Cognitive Computing Systems Research (C3SR) at IBM Research, echoes Schutz’s sentiment: “Multi-messenger astrophysics provides an unprecedented opportunity for both the scientific community and the computing community to push the frontiers of unknowns.”


The National Center for Supercomputing Applications at the University of Illinois Urbana-Champaign provides supercomputing and advanced digital resources for the nation’s science enterprise. At NCSA, University of Illinois faculty, staff, students and collaborators from around the globe use these resources to address research challenges for the benefit of science and society. NCSA has been advancing many of the world’s industry giants for over 35 years by bringing industry, researchers and students together to solve grand challenges at rapid speed and scale.

Back to top