Next-gen tools for next-gen telescopes

UNIVERSITY PARK, Pa. — Technological advances in the latest generation of telescopes are allowing astronomers to peer deeper into the universe, analyze the light from distant objects more precisely, and gain new insights into the origin and evolution of the universe. But new technology is almost always accompanied by new challenges. These next-generation telescopes and the greatly enhanced quality of their observations are poised to detect cosmic phenomena at an unprecedented rate, creating equally unprecedented amounts of data, requiring new methods to make sense of it all.

One such telescope is located at the Vera C. Rubin Observatory in Chile, which started observing in June 2025. Using a 3,200-megapixel camera the size of a car, Rubin scans the entire visible southern sky every three to four nights, producing about 20 terabytes of data each night as part of the Legacy Survey of Space and Time (LSST). Another next-generation telescope, NASA’s Nancy Grace Roman Space Telescope, launched in August and will produce about 1.36 terabytes of data each day, adding up to more than 20,000 terabytes in its first five years. By comparison, NASA’s Hubble Space Telescope produced 172 terabytes of data in its entire 10-year run, and NASA’s James Webb Space Telescope is expected to produce about 1,000 terabytes at the end of its first five years.

“In order to make sense of this extraordinary amount of data and identify rare phenomena, we need to develop new algorithms in a scalable and efficient way,” said Charlotte Ward, Mercedes Richards Career Development Assistant Professor of Astronomy and Astrophysics. “My group helps develop flexible software and methods that enable scientific discovery across astronomy and astrophysics. We are driven by our own scientific goals, which include understanding how black holes form and grow over time, measuring the expansion of the universe with supernovae, and studying host galaxy environments of transient, or short-lived, phenomena. But we’re always thinking about how to make sure our data releases and software are relevant to the entire scientific community.”

Rubin’s wide field of view and extreme sensitivity to faint light will allow for the detection of real-time changes in the sky as well as rare events, including everything from supernovae, gamma-ray burst afterglows, and emissions from supermassive black holes. Roman will map the visible universe in optical and infrared wavelengths with high resolution, helping identify and study distant supernovae and supermassive black holes and their host galaxies, which may improve understanding of how black holes and stars evolve and lend insight into the origins of the universe. Both telescopes should also help answer questions about dark energy — a mysterious cosmic force that astrophysicists say makes up two-thirds of the known universe and is believed to be causing the universe to expand.

“We are actually very interested in using the two new telescopes and their data together,” Ward said. “Rubin has an ultra-wide field of view but is limited to optical wavelengths and has to deal with distortions and interference by the Earth’s atmosphere. Roman doesn’t have as wide of a field of view, but it has very high resolution because it’s in space without Earth’s atmosphere getting in the way. If we take the images from Rubin on the ground and from Roman in space and we model them together, we can overcome the individual challenges of each instrument. If we can combine the strengths of these telescopes and their surveys with new computational tools and machine learning techniques, we’ll have a much more enhanced scientific analysis.”

Ward and her lab are particularly interested in studying supermassive black holes, which have masses hundreds of thousands, or even millions to billions, times that of the sun. Supermassive black holes are frequently located at the centers of large galaxies and are often detected when gas is absorbed by the black hole, which gives off light that is visible from Earth.

“Now, there’s another way to detect black holes that was more recently discovered,” Ward said. “Sometimes a star falls too close to a supermassive black hole, such that it’s pulled apart by tidal forces and causes a temporary flare as the star is eaten up. We can sometimes detect that sudden brightening from the center of a distant galaxy, and these have turned out to be really useful probes for identifying when there is a hidden supermassive black hole at the center of a galaxy.”

These probes, she explained, can also reveal the mass of the supermassive black hole, as well as the stellar dynamics that led to the star being swallowed by the black hole. With the improved capabilities of Rubin, however, Ward hopes to learn more about supermassive black holes that are not located at the center of a galaxy.

“A fraction of supermassive black holes are wandering outside of their galaxy’s center,” she said. “They might have been ejected as two galaxies merged and it takes a while to get back to the center, or maybe two black holes merged and a remnant was kicked away. But these wandering black holes are really hard to find, as there isn’t much gas for them to eat up. With Rubin, we can look for these tidal disruption events and hopefully identify many more, so that we can study them collectively at the population level for the first time.”

Ward, who is originally from Australia, started studying these wandering black holes when she was a graduate student at the University of Maryland. She used data from Rubin’s predecessor, the Zwicky Transient Facility, but said that the data and algorithmic capabilities available at the time limited the questions she could ask in this area.

“That really guided my Ph.D. to be much more focused on building algorithms and software that enable the challenging scientific questions and overcome limitations in the data,” she said. “We were driven by the challenges that came up, and working through these challenges gave me an appreciation for how important it is to collaborate with different research groups and bring in lots of different technical skills and modern computational approaches to address these challenges.”

After finishing her doctoral degree, she served as a postdoctoral research associate at Princeton University, where she worked closely with the Rubin Observatory software team. Joining the faculty at Penn State, she said, was particularly appealing in part because of the support for interdisciplinary collaborations and the infrastructure that fosters it, like the Center for Astrostatistics and the Institute for Computational and Data Sciences.

“At Penn State, there is expertise in and value for the hard methodological and statistical work that underpins a lot of high-impact science,” she said. “It’s a friendly and collaborative environment, and it has been a great place for my research group to take on the difficult algorithmic challenges.”

Ward’s current research group includes several graduate and undergraduate students who, in addition to supporting methodological advances for new instruments like Rubin and Roman, are exploring a variety of questions using data from equally varied telescopes and surveys.

“There are some really big observatories scheduled to come online when our graduate students will be finishing up at Penn State and applying for postdocs and other jobs,” Ward said. “The skills that they'll get in this joint analysis of Rubin and Roman will really help them with the next-generation observatories.”

Editor's Note: This story originally appeared in the fall 2026 edition of the Science Journal, the Eberly College of Science magazine.