For more than a decade, large, moveable telescopes tucked away on a remote, high-altitude site in the Inyo Mountains, about 250 miles northeast of Los Angeles, have worked together to paint a picture of the universe through radio-wave observations.
Known as the Combined Array for Research in Millimeter-wave Astronomy, or CARMA, the telescopes formed one of the most powerful millimeter interferometers in the world. CARMA was created in 2004 through the merger of the Owens Valley Radio Observatory (OVRO) Millimeter Array and the Berkeley Illinois Maryland Association (BIMA) Array and initially consisted of 15 telescopes. In 2008, the University of Chicago joined CARMA, increasing the telescope count to 23.
CARMA's higher elevation, improved electronics, and greater number of connected antennae enabled more precise observations of radio emission from molecules and cold dust across the universe, leading to ground-breaking studies that encompass a range of cosmic objects and phenomena—including stellar birth, early planet formation, supermassive black holes, galaxies, galaxy mergers, and sudden, unexpected events such as gamma-ray bursts and supernova explosions.
"Over its lifetime, it has moved well beyond its initial goals both scientifically and technically," says Anneila Sargent (MS '67, PhD '78, both degrees in astronomy), the Ira S. Bowen Professor of Astronomy at Caltech and the first director of CARMA.
On April 3, CARMA probed the skies for the last time. The project ceased operations and its telescopes will be repurposed and integrated into other survey projects.
Here is a look back at some of CARMA's most significant discoveries and contributions to the field of astronomy.
These CARMA images highlight the range of morphologies observed in circumstellar disks, which may indicate that the disks are in different stages in the planet formation process, or that they are evolving along distinct pathways. The bottom row highlights the disk around the star LkCa 15, where CARMA detected a 40 AU diameter inner hole. The two-color Keck image (bottom right) reveals an infrared source along the inner edge of this hole. The infrared luminosity is consistent with a 6M Jupiter planet, which may have cleared the hole.
Newly formed stars are surrounded by a rotating disk of gas and dust, known as a circumstellar disk. These disks provide the building materials for planetary systems like our own solar system, and can contain important clues about the planet formation process.
During its operation, CARMA imaged disks around dozens of young stars such as RY Tau and DG Tau. The observations revealed that circumstellar disks often are larger in size than our solar system and contain enough material to form Jupiter-size planets. Interestingly, these disks exhibit a variety of morphologies, and scientists think the different shapes reflect different stages or pathways of the planet formation process.
CARMA also helped gather evidence that supported planet formation theories by capturing some of the first images of gaps in circumstellar disks. According to conventional wisdom, planets can form in disks when stars are as young as half a million years old. Computer models show that if these so-called protoplanets are the size of Jupiter or larger, they should carve out gaps or holes in the rings through gravitational interactions with the disk material. In 2012, the team of OVRO executive director John Carpenter reported using CARMA to observe one such gap in the disk surrounding the young star LkCa 15. Observations by the Keck Observatory in Hawaii revealed an infrared source along the inner edge of the gap that was consistent with a planet that has six times the mass of Jupiter.
"Until ALMA"—the Atacama Large Millimeter/submillimeter Array in Chile, a billion-dollar international collaboration involving the United States, Europe, and Japan—"came along, CARMA produced the highest-resolution images of circumstellar disks at millimeter wavelengths," says Carpenter.
A color image of the Whirlpool galaxy M51 from the Hubble Space Telescope (HST). A three composite of images taken at wavelengths of 4350 Angstroms (blue), 5550 Angstroms (green), and 6580 Angstroms (red). Bright regions in the red color are the regions of recent massive star formation, where ultraviolet photons from the massive stars ionize the surrounding gas which radiates the hydrogen recombination line emission. Dark lanes run along spiral arms, indicating the location where the dense interstellar medium is abundant.
Credit: Jin Koda
Stars form in "clouds" of gas, consisting primarily of molecular hydrogen, that contain as much as a million times the mass of the sun. "We do not understand yet how the diffuse molecular gas distributed over large scales flows to the small dense regions that ultimately form stars," Carpenter says.
Magnetic fields may play a key role in the star formation process, but obtaining observations of these fields, especially on small scales, is challenging. Using CARMA, astronomers were able to chart the direction of the magnetic field in the dense material that surrounds newly formed protostars by mapping the polarized thermal radiation from dust grains in molecular clouds. A CARMA survey of the polarized dust emission from 29 sources showed that magnetic fields in the dense gas are randomly aligned with outflowing gas entrained by jets from the protostars.
If the outflows emerge along the rotation axes of circumstellar disks, as has been observed in a few cases, the results suggest that, contrary to theoretical expectations, the circumstellar disks are not aligned with the fields in the dense gas from which they formed. "We don't know the punch line—are magnetic fields critical in the star formation process or not?—because, as always, the observations just raise more questions," Carpenter admits. "But the CARMA observations are pointing the direction for further observations with ALMA."
Molecular gas in galaxies
CARMA was used to image molecular gas in the nearby Andromeda galaxy. All stars form in dense clouds of molecular gas and thus to understand star formation it is important to analyze the properties of molecular clouds.
Credit: Andreas Schruba
The molecular gas in galaxies is the raw material for star formation. "Being able to study how much gas there is in a galaxy, how it's converted to stars, and at what rate is very important for understanding how galaxies evolve over time," Carpenter says.
By resolving the molecular gas reservoirs in local galaxies and measuring the mass of gas in distant galaxies that existed when the cosmos was a fraction of its current age, CARMA made fundamental contributions to understanding the processes that shape the observable universe.
For example, CARMA revealed the evolution, in the spiral galaxy M51, of giant molecular clouds (GMCs) driven by large-scale galactic structure and dynamics. CARMA was used to show that giant molecular clouds grow through coalescence and then break up into smaller clouds that may again come together in the future. Furthermore, the process can occur multiple times over a cloud's lifetime. This new picture of molecular cloud evolution is more complex than previous scenarios, which treated the clouds as discrete objects that dissolved back into the atomic interstellar medium after a certain period of time. "CARMA's imaging capability showed the full cycle of GMCs' dynamical evolution for the first time," Carpenter says.
The Milky Way's black hole
CARMA worked as a standalone array, but it was also able to function as part of very-long-baseline interferometry (VLBI), in which astronomical radio signals are gathered from multiple radio telescopes on Earth to create higher-resolution images than is possible with single telescopes working alone.
In this fashion, CARMA has been linked together with the Submillimeter Telescope in Arizona and the James Clerk Maxwell Telescope and Submillimeter Array in Hawaii to paint one of the most detailed pictures to date of the monstrous black hole at the heart of our Milky Way galaxy. The combined observations achieved an angular resolution of 40 microarcseconds—the equivalent of seeing a tennis ball on the moon.
"If you just used CARMA alone, then the best resolution you would get is 0.15 arcseconds. So VLBI improved the resolution by a factor of 3,750," Carpenter says.
Astronomers have used the VLBI technique to successfully detect radio signals emitted from gas orbiting just outside of this supermassive black hole's event horizon, the radius around the black hole where gravity is so strong that even light cannot escape. "These observations measured the size of the emitting region around the black hole and placed constraints on the accretion disk that is feeding the black hole," he explains.
In other work, VLBI observations showed that the black hole at the center of M87, a giant elliptical galaxy, is spinning.
CARMA also played an important role in following up "transients," objects that unexpectedly burst into existence and then dim and fade equally rapidly (on an astronomical timescale), over periods from seconds to years. Some transients can be attributed to powerful cosmic explosions such as gamma-ray bursts (GRBs) or supernovas, but the mechanisms by which they originate remain unexplained.
"By looking at transients at different wavelengths—and, in particular, looking at them soon after they are discovered—we can understand the progenitors that are causing these bursts," says Carpenter, who notes that CARMA led the field in observations of these events at millimeter wavelengths. Indeed, on April 27, 2013, CARMA detected the millimeter-wavelength emission from the afterglow of GRB 130427A only 18 hours after it first exploded. The CARMA observations revealed a surprise: in addition to the forward-moving shock, there was one moving backward. This "reverse" shock had long been predicted, but never conclusively observed.
Getting data on such unpredictable transient events is difficult at many observatories, because of logistics and the complexity of scheduling. "Targets of opportunity require flexibility on the part of the organization to respond to an event when it happens," says Sterl Phinney (BS '80, astronomy), professor of theoretical astrophysics and executive officer for astronomy and astrophysics at Caltech. "CARMA was excellent for this purpose, because it was so nimble."
Multi-wavelength view of the redshift z=0.2 cluster MS0735+7421. Left to right: CARMA observations of the SZ effect, X-ray data from Chandra, radio data from the VLA, and a three-color composite of the three. The SZ image reveals a large-scale distortion of the intra-cluster medium coincident with X-ray cavities produced by a massive AGN outflow, an example of the wide dynamic-range, multi-wavelength cluster imaging enabled by CARMA.
Credit: Erik Leitch (University of Chicago, Owens Valley Radio Observatory)
Galaxy clusters are the largest gravitationally bound objects in the universe. CARMA studied galaxy clusters by taking advantage of a phenomenon known as the Sunyaev-Zel'dovich (SZ) effect. The SZ effect results when primordial radiation left over from the Big Bang, known as the cosmic microwave background (CMB), is scattered to higher energies after interacting with the hot ionized gas that permeates galaxy clusters. Using CARMA, astronomers recently confirmed a galaxy cluster candidate at redshifts of 1.75 and 1.9, making them the two most distant clusters for which an SZ effect has been measured.
"CARMA can detect the distortion in the CMB spectrum," Carpenter says. "We've observed over 100 clusters at very good resolution. These data have been very important to calibrating the relation between the SZ signal and the cluster mass, probing the structure of clusters, and helping discover the most distant clusters known in the universe."
Training the next generation
In addition to its many scientific contributions, CARMA also served as an important teaching facility for the next generation of astronomers. About 300 graduate students and postdoctoral researchers have cut their teeth on interferometry astronomy at CARMA over the years. "They were able to get hands-on experience in millimeter-wave astronomy at the observatory, something that is becoming more and more rare these days," Sargent says.
Tom Soifer (BS '68, physics), professor of physics and Kent and Joyce Kresa Leadership Chair of the Division of Physics, Mathematics and Astronomy, notes that many of those trainees now hold prestigious positions at the National Radio Astronomy Observatory (NRAO) or are professors at universities across the country, where they educate future scientists and engineers and help with the North American ALMA effort. "The United States is currently part of a tripartite international collaboration that operates ALMA. Most of the North American ALMA team trained either at CARMA or the Caltech OVRO Millimeter Array, CARMA's precursor," he says.
Following CARMA's shutdown, the Cedar Flats sites will be restored to prior conditions, and the telescopes will be moved to OVRO. Although the astronomers closest to the observatory find the closure disappointing, Phinney takes a broader view, seeing the shutdown as part of the steady march of progress in astronomy. "CARMA was the cutting edge of high-frequency astronomy for the past decade. Now that mantle has passed to the global facility called ALMA, and Caltech will take on new frontiers."
Indeed, Caltech continues to push the technological frontier of astronomy through other projects. For example, Caltech Assistant Professor of Astronomy Greg Hallinan is leading the effort to build a Long Wavelength Array (LWA) station at OVRO that will instantaneously image the entire viewable sky every few seconds at low-frequency wavelengths to search for radio transients.
The success of CARMA and OVRO, Soifer says, gives him confidence that the LWA will also be successful. "We have a tremendously capable group of scientists and engineers. If anybody can make this challenging enterprise work, they can."