On the morning of October 16, physicist Marcelle Soares-Santos was the only Brazilian among the 16 research group leaders giving a press conference at the headquarters of the National Science Foundation, USA, to announce the observation of a phenomenon that could transform our knowledge of the universe.
Soares-Santos, 36, is a professor at Brandeis University and a researcher at the Fermi National Accelerator Laboratory (Fermilab), one of the most important particle physics laboratories in the world. Born in Vitória, Soares-Santos has a degree in physics from the Federal University of Espírito Santo (UFES) and a master’s and doctorate in astronomy from the University of São Paulo (USP). She started her postdoctoral fellowship at Fermilab in 2010, and helped build one of the largest light detectors ever made: a 570-megapixel camera installed in a telescope in Chile, used to map 300 million galaxies in the Dark Energy Survey (DES). Today, she leads a DES team searching for light emitted by events that generate gravitational waves.
She spoke to FAPESP about the recently detected phenomenon and the potential use of gravitational waves to calculate the rate of expansion of the universe.
What was your role in the observations of the neutron star collision?
It started with my participation in the DES, which is designed to observe 300 million galaxies and discover how dark energy contributes to the architecture of the universe. When I started, the camera used in the observations was under construction, and I soon gained the respect of my colleagues for my work testing each of its components. In September 2012, when the camera was completed, the DES began to accumulate a sample of millions of galaxies. I used my experience in data analysis and my knowledge of the camera to study galaxy clusters, and I quickly reached a position of leadership. In July 2013, when the collaborative group that operates the Laser Interferometer Gravitational-Wave Observatory (LIGO) made a call for astronomers, I saw an opportunity for the DES to play an important part.
Where were you when the LIGO and Virgo observatories reported the collision of the neutron stars?
I was in my apartment in Chicago. The event was detected on August 17, the same day a moving truck was scheduled to take my things to Waltham, Massachusetts, where I was moving as part of my transfer to Brandeis University. I had just gone to bed at around 7:40 in the morning when the automatic LIGO alert went off on my phone. I had been working all night because there had been a collision of black holes on August 14, and I thought maybe something had gone wrong with the analysis of that event. I did not think it was going to be something new. I jumped out of bed and turned on my computer, and began planning with my colleagues for the observations in Chile after sunset. There was one group at the telescope and one at Fermilab. My apartment was almost completely empty; the only things left were my laptop, a chair, and the Internet router.
Have you been waiting to see a collision between neutron stars for a long time?
When LIGO entered its second phase of operation in 2015, the first events we expected to see were collisions involving neutron stars, not black holes. In nature, we expect systems with smaller masses, such as those of neutron stars, to exist in greater quantity, and we would therefore be more likely to observe them. It was a surprise to see that black holes with 10 to 30 times more mass than the Sun are so common [Rainer Weiss, from the Massachusetts Institute of Technology (MIT), and Kip Thorne and Barry Barish, both from the California Institute of Technology (Caltech), received the Nobel Prize in Physics in 2017 for the detection of gravitational waves emitted by merging black holes].
What were the next steps after the alert on August 17?
Because of the searches we conducted after detecting the black hole collisions, we had already been through the procedure three or four times. First, we mapped the sky and identified the region where the event probably occurred. Based on the position of the telescope in Chile and the number of hours we were able to observe per night, we calculated the search field and how to cover the most of it. For this event, the distance was small, and we made a list of galaxies. Whenever one of them appeared in the observed area, a group looked for signs of light emission in the images before they were processed by computer. We were one of the first groups to detect the light from the source of the gravitational waves. Shortly after our discovery, we sent an email to inform the other LIGO research partners and saw that another group had identified the same source 10 minutes earlier.
Was it the first observation sequence?
The first night. The observations lasted about an hour. After finding the most plausible candidate, we changed strategy for the next night. We focused our attention on this object to get as much information about it as possible and to monitor its development. We also did a second scan of the region in search of other possible sources. The area indicated by LIGO and Virgo was 30 square degrees [equal in size to about 150 full moons as seen from Earth]. We looked at 70 square degrees because we know from experience with previous events that LIGO conducts a more refined analysis after the initial mapping, which can change the object’s position in the sky.
Were there many objects in the indicated area?
The data suggested that the collision occurred 40 megaparsecs away (130 million light-years). There are about 50 galaxies in the space encompassed by a 30-square-degree angle and a distance of 40 megaparsecs. In an analysis that ignored the distance, we found 1,500 potential sources, but only one passed the three criteria established to exclude false candidates.
Why is it important to associate the emission of light with the source of the gravitational waves?
In the August event, we knew from the gravitational waves that two neutron stars had collided rather than two black holes, and that the collision took place 40 megaparsecs away. But they cannot tell us whether the result of the collision was a black hole or a neutron star, whether the collision generated heavy chemical elements, whether the environment around the event was disturbed, or whether the environment was different to most galaxies. It is only possible to know these things by combining gravitational waves with the optical light.
You studied gravitational waves in your master’s degree and neutron stars in your doctorate—was it all in preparation for observing an event like this?
My plan was to work in cosmology. In my undergraduate research training and in my master’s degree, I calculated the spectrum of primordial gravitational waves, which have existed since the beginning of the universe rather than having been generated by collisions. A future detector like the LISA Interferometer operated by the ESA [European Space Agency] could be sensitive enough to observe such gravitational waves in space. It was all very theoretical and I missed the observational side of things. For my doctorate, I studied galaxy clusters, the largest structures in the universe. The rate at which these clusters form is related to cosmology. If there is too much dark energy, they form more slowly. I developed algorithms to find such clusters in the Sloan Digital Sky Survey data, which scanned a third of the sky and mapped 500 million objects. I continued studying galaxy clusters at a postdoctoral level, but I became interested in gravitational waves again when LIGO announced the beginning of a new round of observations.
What can gravitational waves reveal about dark energy?
Cosmologists can use events such as neutron star collisions in a similar way to supernovae, which occur when massive stars explode. Supernovae are known as “standard candles” because they emit a known luminosity, allowing us to estimate how far they are from us. Neutron star collisions, however, are known as “standard sirens” because the way they are detected is similar to sound waves [although gravitational waves are different from sound waves]. The intensity of the gravitational waves detected on Earth depends on the mass of the stars, which can be calculated based on the type of wave detected. Once we have identified several more events like the one in August, it will be possible to measure distances on a cosmological scale. The August event allowed us to calculate the rate of expansion of the universe, called the Hubble constant, with an uncertainty of 15%. That is high, but it should decrease as we observe more events, which is important because there are still questions as to whether supernovae really can function as standard candles, or whether there may be variations between them that affect the measurements. Measurements based on cosmic microwave background radiation [microwave radiation emitted when the universe was only 380,000 years old] differ from those obtained using supernovae. Gravitational waves could be an alternative to these metrics.
What are your plans for the future?
To study gravitational waves as a way of finding out how much dark energy there is in the cosmos. We want to improve the detection of light in future events, the number of which should increase in the next round of LIGO-Virgo observations. And I intend to develop a strategy for mapping galaxies using the next generation of cameras and telescopes.