Hundreds of similar watch parties are planned around the world, with participants eager to glimpse the universe from Rubin’s extraordinary perspective.

The images captured over this 10-year survey will be stitched into time-lapse videos, revealing motion, patterns and phenomena — expected and unexpected — in regions we have never seen because until now we haven’t had the technology to capture these images or make sense of the huge amount of data the LSST will harvest.

We have it now — on a perch about 8,800 feet high in the Andes Mountains of central Chile, where the Rubin Observatory is nearing completion.

But where should we point this one-of-a-kind camera? What will we look for? Which filter will we use and when? How do we organize all of this?

Bianco is leading the team that is developing the strategy to make the most of this 10-year quest. The team provides input on critical technical decisions and has identified four primary science goals:

• Understanding the nature of dark matter and dark energy

• Creating an inventory of the solar system

• Mapping the Milky Way

• Exploring objects that change position or brightness over time, called “transients”

“Normally, surveys either go deep — where you can see very faint things — or they go wide and shallow so you can see a large area of the sky,” Bianco said. “The idea here is to do all three. It will be the deepest survey there has ever been and it will cover the whole Southern Hemisphere sky. What we will provide is a four-dimensional view — including the time domain aspect.

“It’s a kick-ass camera, so it can do it all with very high resolution. We’ll be able to see individual stars instead of seeing them clumped together.”

It’s a nimble design, too, Bianco said. Though the telescope is massive, it can move very quickly. And it can scan the entire Southern Hemisphere’s sky in about three nights’ time.

There’s a lot going on up there. Supernova explosions, for example, have special “wow” powers. They are seen when the core of a massive star collapses, producing a fireball.

“One of the goals of LSST is to discover thousands of supernovas every night,” Gizis said. “With LSST’s gigantic image, there will be stars exploding and you’ll always be getting some supernova.

“Hopefully we’ll be able to discover all kinds of rare things we’ve never been able to characterize before because we didn’t have a telescope pointing at the right place at the right moment. Hopefully we’ll see all of the really rare stuff — unusual types of stars blowing up in unusual places. Somewhere in the volume of space that is within 10 billion light years of the sun, lots of stars are exploding somewhere.” 

Much in our universe is invisible to us. What will we learn if we can see more using this new telescope? What could it mean if we understood dark matter and dark energy? How will theories stand up to new observations and information?

“It might be that it all holds up,” Gizis said. “Or we may need to make some adjustments. And it could be that there’s something fundamentally wrong. Maybe the laws of physics are different than we think. Maybe dark energy is something else that we haven’t thought of yet. It could be that Newton’s laws of motion that work perfectly well on Earth or in our solar system — maybe it’s something quite different when you’re talking about distances of millions of light years.

“One way we make progress is by making more and more precise measurements of how galaxies are behaving at different distances.”

Eight science teams with more than 2,000 members are part of the LSST ecosystem, all working as a connected network with distinct objectives.

Bianco, for example, is the deputy project scientist and interim head of science for the Rubin Observatory. From 2014 through 2021, she chaired the collaboration that deals with “time astronomy,” looking for transient and variable stars, things that move, stars that explode and disappear.

Gizis was the chair of the team that will map the Milky Way, our home galaxy, bringing his extensive expertise in brown dwarfs (the smallest stars), his expertise in astrometry — capturing precise measurements — along with calibration and testing, ensuring that the software and camera are producing accurate positions for the stars.

Hong, who is new to UD’s faculty and a core faculty member of the Data Science Institute, is developing algorithms that may help make sense of the enormous amount of data that will be collected — more than 9 billion gigabytes each year.

At least 10 UD students also will be part of the work, Bianco said.

“We’ve been very fortunate to have Federica Bianco decide to come to the University, in my opinion,” Gizis said. “I’m thrilled that she’s here. She’s a rising star. And Prof. Hong is developing new methodologies in AI and machine learning — fancy statistics — and he has an interest in applying those to astronomy. This is a nice place to demonstrate how powerful the algorithms are.”

With more than 10 million changes expected in the telescope’s field of vision each night — more than 36.5 billion over the course of LSST’s survey — the data lift is heavy indeed.

“A lot of what we do depends on automation, artificial intelligence and machine learning,” Bianco said. “That is really the revolution here — the scale up.”

The project will enable decades of science, Bianco said, with many projects emerging only after all 10 years of data are captured.

"It’s really exciting to be a part of a project of this magnitude,” said Siddharth Chaini, a doctoral student in Bianco’s group. “I am working on preparing for the ‘data tsunami’ the Rubin Observatory will bring. A large part of my work involves developing machine learning and applying astrostatistical methods to enable scientific discovery within the billions of objects Rubin will see.

“But what I find most incredible, beyond the groundbreaking ability of this Observatory, is the global collaborative role in this project,” Chaini said. “I had the opportunity to take on a leadership role in the scientific community interested in the discovery of the unknown, or ‘anomaly detection.’ Through this role, I have been fortunate enough to witness just how supportive the broader Rubin community is, with researchers willing to go above-and-beyond — as scientists, but also as human beings. This environment results in a culture where every researcher involved contributes their piece to build something where the project is much greater than the sum of the parts. I love this."

Easton Honaker, a doctoral student in Gizis’ group, agreed.

“Being part of an organization and collaboration as big as the Rubin community is an incredible opportunity to meet and work with scientists from across the world,” he said. “Even more amazing is that, despite everyone working on widely different topics, we are all working together to make the Rubin project a success.

“Personally, I’m very excited to use the survey to build the deepest, largest and most complete sample of brown dwarfs — the coldest, faintest and smallest stars in the universe to date. This will help us understand how stars in the Milky Way have formed and evolved. The sheer volume of data and longevity of the project makes it a truly career-defining opportunity. On top of that, being able to bring the much-anticipated first look moment home to UD and share it with the public is an honor. I hope we will connect to a large audience and help draw interest in astronomy and inspire future scientists.”

Bianco said an important aspect of the project is the vast number of scientists that will be able to access the data, which is public in the U.S. and Chile and accessible to many international partners.

UD’s experts will be looking at LSST data from much different angles.

Federica Bianco

Bianco, associate professor of physics and astronomy and a core faculty member of the Data Science Institute, specializes in transient objects and things that move in the sky — things that appear, disappear, explode.

“I’m interested in things that change over time,” she said. “And one of the frontiers is to do these analyses very rapidly. As soon as something changes, you can follow up with other telescopes.”

Networks of astronomers around the world monitor such developments. When an interesting change is reported, they inspect the location with the telescope available to them and provide data from various angles.

“Nature is tricky,” Bianco said. “You learn a lot more in the early phases of transience, so you really want to know as soon as possible. A new nascent supernova is a coveted observation in the field. You can learn so much stuff in the first few hours. And that means the images have to be reduced in real time.

“Our goal is to send information about everything that changes in the night sky within 60 seconds. Anything detected to have changed gets identified. Information is collected and distributed to anybody in the world within 60 seconds of when you take the image. And that amounts to 10 million things that will change every night.”

The 3,200-pixel camera will capture so much data it would take hundreds of high-resolution screens to display a single image.

And the enormous volume of data LSST will collect cannot just be downloaded — there’s far more than can be transmitted. Instead, Bianco said, scientists write code and upload it to LSST’s computing platform to do their analysis remotely.

“You select those that are most urgent and most interesting,” she said. “And how do you build such an algorithm? What does it mean to be the ‘most interesting’?”

Always more questions.

John Gizis

Gizis, UD’s Annie Jump Cannon Professor of Astronomy and associate director of UD’s Data Science Institute, is a former co-chair of the LSST team that will oversee the mapping of the Milky Way and he continues to work with that team. He is leading a working group on the solar neighborhood, looking at the stars that are nearest to us, while others use the same telescope to observe the most distant stars.

“I’m looking for faint things, with low luminosity that have been overlooked in previous surveys,” he said.

He is keenly interested in low-mass stars and brown dwarfs and has used many different telescopes to investigate these objects, including land-based telescopes and the space-based Hubble Telescope and James Webb Space Telescope.

“Brown dwarfs are smaller and dimmer than the sun,” he said, “and there is not a single one that we can just see with our eyes, even though they are much closer than the stars that we can see.”

Before he joined UD’s faculty, Gizis was part of a survey called 2MASS, which mapped the sky once and discovered hundreds of objects.

LSST, by comparison, will map the sky hundreds of times and provide tens of thousands of samples of objects.

“That will give us a much better handle on how stars change in time, how many of them there are and whether some of these things are escaped planets,” he said.

He’ll be looking for peculiar stars and typical stars.

“We’re motivated by physics questions, so we would like to make underlying models of the stars and galaxies and all of that,” he said. “We’re not just satisfied with measuring them and not knowing what the models mean.”

As with almost any scientific pursuit, there are debates about how these investigations should proceed.

“One of the main motivations for LSST is trying to better characterize the problems of cosmology, the problem of dark matter, the problem of dark energy. We can describe the effects of things we can’t see. And we know that galaxies are flying apart when gravity should be attracting them.”

Those are areas LSST aims to illuminate.

“Ideally, we’d have a collection of Einsteins that gave us a complete theory we could test,” he said. “They’d say: ‘Here’s the dark matter. We made it in a lab, and we can do things with it.’

“Instead, all we have is astronomy. Here's what we can figure out indirectly — the way stars and galaxies are moving. We hope to make progress on the astronomy side of the problem.”

David Hong

Hong, assistant professor of electrical and computer engineering, brings uncommon skills in data analysis to the effort. A special focus of his research is developing algorithms that can make sense of “messy” big data, discovering underlying patterns in high-dimensional data sets.

All of that is essential for making the most of the tsunami of data LSST will deliver.

“I tend to think about data geometrically,” he said. “If we had data points in just two or three dimensions, we could just visualize them and we’d see all sorts of patterns.”

LSST, though, will deliver data with thousands of dimensions.

“Often these data are in hundreds or thousands of dimensions,” Hong said. “Now you need algorithms. You can’t easily visualize it anymore.”

LSST presents an exciting new challenge for Hong, whose doctorate is in electrical engineering. He previously worked on questions related to algorithms for finding low-dimensional structure in heterogeneous data, motivated by data analysis problems that arise in medical imaging. For example, what if some data are high quality and some are low quality? How should data be combined in a single analysis when some are noisier than others?

“You need to account for heterogeneous quality in data,” he said. “You might do that by giving less weight to lower-quality, less-trustworthy data. The question is how much less weight? We made a surprising discovery in our work — the standard weights turned out to be sub-optimal and we derived new optimal weights.”

Much of the data in astronomy is of this heterogeneous quality, Hong said, and he has enjoyed getting acquainted with the massive amounts of astronomical data available online.

“I’m excited to work with astronomers to apply some of the work in my area of algorithms research to the challenging data analysis problems they face and to form new algorithms together,” he said. “I’m not trained as an astronomer, but I have always found astronomy fascinating.

“Astronomers have been doing big data since before it was cool. Lots of work has already been done, but when you’re thinking about the heterogeneity in the data, I might have some expertise to contribute. I’m excited to get involved and play some small part in this massive scientific effort.”

About the Vera C. Rubin Observatory

The Vera C. Rubin Observatory is a joint initiative of the U.S. National Science Foundation (NSF) and the U.S. Department of Energy’s Office of Science (DOE/SC). Its primary mission is to carry out the Legacy Survey of Space and Time, providing an unprecedented data set for scientific research supported by both agencies. It is named for astronomer Vera Rubin, who provided the first convincing evidence for the existence of dark matter. Rubin is operated jointly by NSF NOIRLab and SLAC National Accelerator Laboratory. NSF NOIRLab is managed by the Association of Universities for Research in Astronomy (AURA), and SLAC is operated by Stanford University.

About the LSST Discovery Alliance

The LSST Discovery Alliance is a nonprofit organization working to maximize the impact of Rubin’s LSST by supporting the global network of scientists who are part of the project and addressing obstacles that might otherwise hinder their advances and discoveries. With support from member institutions, government grants and donations from corporations and private citizens, the alliance provides funding and programs that complement the investments of the National Science Foundation and the Department of Energy in the construction and operations of LSST. The alliance is distinct from the Rubin Observatory teams that are constructing and will be operating the Rubin Legacy Survey of Space and Time (LSST).