Discovering patterns in the universe

Dr Margaret Joan Geller is a pioneering astrophysicist, currently researching at the Harvard-Smithsonian Center for Astrophysics and the Smithsonian Astrophysical Observatory. Her research into the spatial distribution of galaxies fundamentally enriched our knowledge of the structure of the universe, and the way these enormous patterns have grown over the history of the universe. She continues to develop deeper maps of the galaxy through her HectoMAP project. Research Features was privileged to speak to Geller about her illustrious career – and about the remarkable advancements made in astronomy over her decades in the field. 

When did you first become interested in cosmology and astronomy and why?

In some ways, my entry into astrophysics was an accident. I was an undergraduate at Berkeley from 1966–1970. It was not an easy time. The physics department had an honors program and as part of that program Charles Kittel became my mentor. Kittel made important contributions to the foundation of solid state physics, now called condensed matter physics.

Kittel advised me against doing research in solid state physics and said, ‘You should choose a field that will be peaking ten years after your PhD. Then you have the opportunity to lay part of the foundation.’ He thought astrophysics was one of those fields.

I had never taken an astronomy course and I didn’t really know how to respond. To make matters even more disorienting, Kittel told me that I should go to graduate school at Princeton, then the top physics department in the country. I had never thought about Princeton because Princeton didn’t accept women when I applied as an undergraduate.

I decided to humour Kittel and apply to Princeton. If I was admitted, I would follow his advice and study astrophysics. I was admitted.

Because I had an NSF graduate fellowship, I could work with any faculty member. Jim Peebles, who won the Nobel Prize in Physics in 2019, was then an assistant professor. He was trying to understand the way galaxies were distributed in the universe. Geometric problems were always my favourite and I asked Peebles to be my thesis supervisor. He agreed.

You are a pioneer in mapping the nearby universe. Could you explain how you managed to accomplish your incredible achievement?

In the mid-1980s when my colleagues and I thought about mapping a slice of the universe, the idea was in its infancy. Mapping the universe is a big idea, but like many big ideas, it started in a small way.

In the mid-1980s, the general picture was that massive clusters of galaxies were randomly distributed in a sea of randomly distributed individual galaxies and small groups of galaxies.

I wondered whether there were larger patterns: the continents, and oceans of the universe. By thinking about the continents and oceans on earth, I realised that a thin strip around the earth crosses both continents and oceans … and reveals that both kinds of features are big. The 3D analogue of a strip is a slice. Mapping a slice of the nearby universe was a manageable project with our 1.5m telescope.

I suggested to my colleague, John Huchra, and a graduate student, Valerie de Lapparent, that we measure redshifts for a thousand galaxies in a strip across the sky.

“We were all speechless with awe when we saw the pattern traced by the thousand galaxies in our first slice of the universe.”

At that time, acquisition of redshifts was a slow process. We observed one galaxy at a time and each observation took half an hour or more. To measure a redshift, we use a diffraction grating to disperse the light from a distant galaxy into its colours. The result is a spectrum, a rainbow of galaxy light. This spectrum contains the tell-tale signature of the elements in the galaxy. The expanding space of the universe shifts these features to redder, longer wavelengths relative to their positions in a laboratory on Earth. The redshift gives us the distance to each galaxy. In the nearby universe, the distance is simply proportional to the redshift. Once we have the redshift, we have a three-dimensional position for each galaxy: we have the latitude and longitude on the sky and the distance from us. With these coordinates, we can construct a 3D map.

We were all speechless with awe when we saw the pattern traced by the thousand galaxies in our first slice of the universe. All the galaxies are in thin structures surrounding vast nearly empty regions, voids, that are more than two hundred million light-years across. The overall pattern in the now iconic map looks like a dancing stick figure. We called the overall structure ‘bubble-like’ to give a visual picture of the broader implications. Today this pattern is known as the cosmic web.

It was thrilling to be one of the first three people ever to see this amazing and beautiful pattern. I will never forget that feeling of wonder.

Soon scientists and members of the public all over the world shared the amazement. The presence of a clear pattern inspired both observational and theoretical investigations of the nature of the patterns and their history. John Huchra and I extended the survey across a larger region of the sky until it included 15,000 galaxies. Other deeper surveys of the universe and simulations have shown that the pattern revealed by our first steps into the universe is the basic arrangement of galaxies on large scales.

You also co-discovered hypervelocity stars – stars that travel at significantly higher velocities than a galaxy’s other stellar bodies – in 2005. What do you now know about these rarely observed phenomena?

Hypervelocity stars are a fascinating but rare phenomenon. We now know that at least some of these stars are ejected when a binary star passes close enough to the central massive black hole in the Milky Way. The gravitational tidal force of the black hole disrupts the binary: one member falls into the black hole and the other is ejected into the halo of the galaxy at a high enough velocity to escape from the Milky Way.

The known hypervelocity stars provide some insight into the stellar population in the centre of the Milky Way. Unfortunately, hypervelocity stars are too rare to be used as tracers of the distribution of dark matter in the halo of the Milky Way.

The HectoMAP Cluster Survey is a project that you lead, which maps the middle-aged universe. What have you discovered so far?

The HectoMAP survey is a slice of the universe that reaches fifteen times deeper into the universe than our original survey. The slice includes 110,000 redshifts.

When we look out in space we look back in time. HectoMAP shows us a picture of the way galaxies are distributed over the last six billion years of the history of the universe. HectoMAP is a dense survey that reveals the clusters of galaxies and voids in the middle-aged universe. The pattern is strikingly similar to the pattern in the nearby universe.

We used the hundreds of clusters of galaxies in the HectoMAP survey to trace the way gravity in an expanding universe makes clusters grow as the universe ages. Over the last six billion years, these huge systems of galaxies increase their mass by about 50%. We measured this increase directly from the data and showed that the observed growth agrees with predictions of large computer simulations of the development of patterns in the universe.

Could you please briefly tell our readers about dark matter, and about your project SHELS?

Dark matter in the universe is an enduring mystery. This puzzle has been with us since 1937 when Fritz Zwicky discovered that most of the matter in the Coma cluster of galaxies must be dark. Vera Rubin later showed that galaxies like our own Milky Way live in dark matter haloes. We now know that 85% of the matter in the universe is dark. We know a lot about where the dark matter is but we still don’t know what it is.

Like HectoMAP, SHELS is a redshift survey that peers into the distant universe. In addition to the redshift surveys, the regions of these two surveys are covered by very deep imaging surveys carried out on large telescopes like the Japanese Subaru telescope on Mauna Kea, Hawaii.

A redshift survey shows us the distribution of galaxies or, equivalently, light-emitting material. We trace the total distribution of matter in the universe by using the deep images to make maps called weak lensing maps. By comparing the galaxy and weak lensing maps we can explore the way galaxies trace the dark matter on very large scales.

“HectoMAP shows us a picture of the way galaxies are distributed over the last six billion years of the history of the universe.”

Galaxies in our HectoMAP and SHELS surveys act as lenses that produce small but coherent distortions of the much more distant galaxies in the deep Subaru images. By measuring and analysing those distorted images of very faint distant galaxies we can construct a map of the dark matter distribution. We then check the relationship between the redshift survey galaxy maps and the weak lensing maps.

What are you currently researching?

Weak lensing is a powerful technique for making maps and for measuring the total mass of systems of galaxies. We are now using the spectacular Subaru imaging data in the HectoMAP region to measure the way dark matter is distributed in the clusters of galaxies in the survey.

We are also investigating the way galaxies evolve over the last six billion years of the history of the universe. The galaxies that we select for observation in the HectoMAP survey are called quiescent because they are no longer forming stars. These galaxies grow in size as the universe ages. There is general agreement that the galaxies grow because they merge with smaller galaxies, but the details of the picture are murky. Because HectoMAP is a large survey we can use it to unravel some of these details.

What obstacles and barriers have you faced in your career and how did you overcome these?

The main obstacle I have faced in my career is that I am a woman. As I mentioned earlier in this interview, women were not even admitted to many major universities when I was an undergraduate.

Sadly, the more I achieved the worse the obstacles became. I have been able to persevere because I have worked with some amazing younger colleagues. Some of my protégés have collaborated with me for years. Others have provided steady reminders of rich friendships. I have also had generous support from some senior colleagues.

What do you think will be the key research areas for astronomy in the future?

There is plenty of the universe left to map. A new project to map the most recent eleven billion years of the history of the universe began on the 4m telescope on Kitt Peak in January 2022. A dedicated instrument called DESI can acquire 5,000 redshifts at a time for distant galaxies and quasars. This project will measure the parameters that define our universe with greater precision. The maps will reveal the evolution of the cosmic web over 11 billion years out of the 14-billion-year history of the universe.

Spectacular James Webb Space Telescope (JWST) images that explore the young universe will tell us how the first stars and galaxies formed. One of the first JWST images shows young galaxies lensed by a massive foreground cluster of galaxies. A JWST spectrum shows us a young galaxy as it was only six hundred million years after the big bang. During the ten-year lifetime of JWST continuing observations will test our picture that dark matter haloes form first and gas condenses within them to form the stars in galaxies.

My career has spanned an extraordinary time in astronomy. Technological advances in digital detectors, fibre optics, robotics, and computing have made our time the era when we can first map and understand our largest surroundings, the universe. The adventure will continue to surprise and enchant us.

Contact
Dr Margaret J Geller

E: [email protected]

Discovering patterns in the universe