As a side project at Carnegie, I took electronic snapshots of the remaining unidentified southern X-ray sources from the EMSS. Some celestial objects, such as the Large and Small Magellanic Clouds and Alpha Centauri, are only visible from the southern hemisphere. So, while I was in Chile, I snapped images through the duPont 100-inch telescope of southern EMSS source positions with no obvious optical counterparts.
One night, as I took these images, I saw the most amazing sight on my computer screen. As the image I had just taken displayed itself, rather than the nondescript collections of stars and galaxies that I was used to seeing in these blank fields, I saw a dense bees' hive of little faint smudges: a distant cluster of galaxies. I could barely see these smudges through our reddest filter, and a subsequent image through a bluer filter showed no galaxies whatsoever. The fact that the cluster galaxies showed up through the reddest filter but not through a bluer filter meant that the galaxies were very far away—nearly halfway across the Universe. That was my first view of the cluster MS1054-0321.
Like meeting someone who would subsequently become your best friend years later, this cluster consistently appeared in my life after that night. A year later, I was observing on Mount Palomar, desperately trying to discover the exact distance to this cluster by measuring the spectrum of at least one of its galaxies. An image of a galaxy is relatively easy to get, because all of the photons more or less pile on top of each other, making a nice significant spot on the detector. But a spectrum is much more difficult to make, because the photons, the light, are dispersed in a line across the detector, rather than in a single spot. The dispersion of the light depends on the energies of the photons, so we can tell what energy the light has. A spectrum tells us much more about an astronomical object than an image does; in an image we lose information by stacking photons without regard to their energies. But a spectrum, in the bands of light sorted by energy, can reveal how fast the object is moving towards us or away from us, the elemental components of the gas and their ionization states, the temperatures of the gas and much more. Here I wanted to measure how much the Universe had ''stretched'' (expanded) since the time the light had been emitted. This expansion is also known as ''redshift;'' the redshift of the light emitted by the cluster galaxies would tell me how far away that cluster was.
I obtained an extremely noisy spectrum and was only able to estimate a redshift of about 0.8, which means that the spectrum of the galaxy was stretched by 80% (0.8) to longer wavelengths. Later, Isabella Gioia, from the University of Hawaii, took advantage of the remarkable skies above the Canada-France-Hawaii telescope and found a redshift of 0.83 for a few of the galaxies. With such a high redshift, the cluster was confirmed to be nearly halfway across the known Universe.
This distance was very significant, because in the standard model of cosmology and the formation of large-scale structure in the Universe, such clusters weren't supposed to be there. If this cluster was not only distant, but also hot and massive, its mere existence was going to be tough to explain in the context of these models.
X-ray astronomy had moved into the spotlight again in 1990 with the launch of a joint German/UK/US satellite observatory called ROSAT. ROSAT's window on the electromagnetic spectrum was too low an energy to enable us to measure cluster gas temperatures, but its excellent imaging detectors meant we could measure the shapes of the cluster, by taking a picture of the X-ray emission coming from the cluster gas. I began to apply for time to study some of the EMSS clusters in my thesis; but I was excited by a grand-scale project that Simon Morris, another postdoctoral researcher at Carnegie, was participating in to measure the velocities of galaxies in EMSS clusters with redshifts of 0.3 to 0.5. By measuring the velocities of these cluster galaxies, Morris and his collaborators could tell how much matter was really there in each cluster. Such measurements had never been done well before for clusters so distant.
I did not feel I could intrude on their project, but I could start a project of my own to observe the highest-redshift clusters of galaxies in the EMSS, the clusters with redshifts greater than 0.5, including my friend MS1054-0321. To test the waters, I applied to ROSAT to observe the most luminous cluster in the EMSS, MS0451-03 at redshift 0.55. When that proposal was awarded observing time, I conceived of a project that would study all of the high-redshift EMSS clusters. These were some of the only clusters known in the distant Universe. Because the distances to the clusters of galaxies is so immense, light takes a long time to get from these clusters to us on Earth. This time delay means we're seeing the clusters, not as they are now, but as they were billions of years ago, when the Universe was only about half as old as it is now. Clusters of galaxies are also very rare objects. Studying how samples of rare objects change with time was sure to be interesting and possibly relevant to cosmology.
With the support of my colleagues John Stocke and Isabella Gioia, we began applying for more observing time. We speculated that perhaps the lumpiness of these clusters would tell us something about how structure forms, and the density of the Universe.
The density of the Universe is a quantity that astronomers are rather keen to measure. If the average density of the Universe is the critical density or less, the Universe will continue to expand forever. If the average density is greater than the critical density, it will eventually stop expanding and collapse. The critical density is actually very low, the equivalent of a few hydrogen atoms in a space the size of a typical closet. The value Om is defined to be the ratio of the average density of matter divided by the critical density. A critical density universe (or, equivalently, an Qm = 1 universe) is a very appealing theoretical model because it is so simple, and has a plausible explanation in the form of the inflationary Universe, a key variant on the Big Bang theory of the Universe's origin.
Theoretical work by Doug Richstone and others suggested that, in a high-density universe, clusters would be as lumpy in the past as they are today. Optical observers, using galaxy positions and velocities as evidence, were jumping up and down at conferences saying that the Universe must be critical density, because even local clusters of galaxies were so lumpy. I thought that a better test must be to compare local clusters with distant clusters, since ''lumpiness'' is difficult to define, but comparing the lumpiness of clusters nearby with that of distant clusters might be easier to quantify. If distant clusters were more lumpy than they are today, that would support the idea of a less dense universe.
But I happened upon an even more compelling test of the density of the Universe in a comment by Monique Arnaud in one of her 1992 papers in the European journal Astronomy and Astrophysics. In this paper she and her colleagues describe one of the hottest (and therefore the most massive) clusters of galaxies ever discovered. It was so hot and massive, she explained, that its very existence in the volume of space where it was discovered was a challenge to the dense Universe (Qm = 1) theory. She was not the first to say that the existence of a very massive cluster of galaxies was very difficult to explain in the context of an Om = 1 universe—Dr. Jim Peebles of Princeton University in his 1989 paper with Drs. Ruth Daly and Roman Juszkiewicz was one of the first—but she was the first to apply this concept to an X-ray cluster in an X-ray survey, and it was Monique's paper that planted the idea in my head of making this test.
In 1993, we were lucky again—Mark won a Hubble fellowship to go to Johns Hopkins University (JHU) and I was offered an Institute fellowship with equivalent prestige at the Space Telescope Science Institute, just across the street from the physics and astronomy department at JHU. We packed up our Siberian husky and moved across the country to Baltimore.
Yet another X-ray satellite was launched in 1993, jointly by the Japanese and the US. This satellite was named ASCA (Advanced Satellite for Cosmology and Astrophysics)—really a pun for an ancient Japanese word for ''flying bird'' or Asuka. ASCA was perfectly suited to take spectra of clusters of galaxies, and therefore was perfectly suited to take a cluster's temperature. I immediately applied to observe MS0451-03 in the first round and was rewarded with a first-priority, early mission observation!
Was this article helpful?