NCSA NEWS

Good CARMA

Story posted November 9, 2006

With support from the National Science Foundation, CARMA (the Combined Array for Research in Millimeter-wave Astronomy) joins telescopes from two earlier millimeter arrays -- the Berkeley-Illinois-Maryland Association array and Caltech's Owens Valley Radio Observatory array -- to form a more powerful astronomical tool. A grand opening for the high-altitude array was held in the spring, and NCSA's Trish Barker recently spoke with CARMA Director Anneila I. Sargent, the Benjamin M. Rosen Professor of Astronomy at Caltech, about the new opportunities CARMA will provide to observe and understand galaxies, molecular clouds forming clusters of stars, newly born stars emerging from their clouds, comets, and the cosmic radiation left-over from the Big Bang.

Q: Can you describe what CARMA is and why it is significant?
A: It's a combination of existing telescopes/antennas from Caltech's Owens Valley Radio Observatory array and the Berkeley-Illinois-Maryland Association (BIMA) array. It comprises at the moment six 10-meter telescopes from the Owens Valley array and nine 6-meter telescopes from the BIMA array.

I'd really like to emphasize that this is a novel and innovative array; it's not just the sum of some old parts put together. The changes we've introduced in terms of new technology and software make this a state-of-the-art instrument.

Having this new array with the mix of telescope sizes on a higher altitude site means that we get a combination of high sensitivity, high resolution, and the capability to image astronomical sources over wide fields of view at these millimeter wavelengths with better fidelity than has been achieved anywhere else.

Q: Many people probably think of a telescope as something that captures visible light, but these telescopes are looking at a different wavelength.
A: These are radio telescopes. They detect radiation at millimeter wavelengths. We're very sensitive to thermal radiation from forming stars, forming planets from the gas of the interstellar medium, and from other galaxies both near and far.

Often the most interesting regions of the universe -- those where stars and planets are forming or the earliest epochs when galaxies are in their earliest stages of evolution -- are obscured from optical view by intervening dust. These dust particles are minute -- often comparable in size to the wavelength of visible light. As a result, they can prevent optical radiation from astronomical sources from reaching you directly. At millimeter wavelengths you can detect the emission from interstellar gas molecules and dust directly, and also infer what is happening behind the dust.

Q: How do the multiple telescopes in the array work together?
A: The size of a telescope is related to the degree of detail you can see. If you have a larger diameter telescope, you can collect more radiation, and you can also see tinier detail. However, the wavelength of the radiation you are detecting also affects the degree of detail achievable. The longer the wavelength, the bigger the telescope has to be to reach the same level of detail. To get the same detail at millimeter wavelengths as you do with an optical telescope, you would have to have a telescope the size of a football field -- obviously out of the question. The solution is to use an array of interconnected smaller telescopes. Then the degree of detail depends not on the size of the individual telescopes in the array but on how far apart you can separate them and still get a coherent signal.

Q: And in addition to being used by the collaborating institutions (California Institute of Technology, University of California at Berkeley, University of Illinois at Urbana-Champaign, and University of Maryland), CARMA is accessible to the broad community of astronomers, correct?
A: We believe this is one of the most important things that we will do. Illinois is going to help us by ensuring we have an easily accessible archive, so scientists who may not be black-belt interferometrists will still be able to pose a problem and get an answer. CARMA needs to be accessible to observers on a variety of levels. (See sidebar for more on the CARMA cyberinfrastructure.)

Q: What's particularly advantageous about the new site in the Inyo Mountains?
A: It's about twice as high as the elevation of the previous arrays [approximately 7,500 feet above sea level] so that the deleterious effect of water vapor in the Earth's atmosphere decreases by a factor of two. We also can separate the telescopes by rather large distances. The combination of the increased sensitivity, lower atmospheric opacity, the increased collecting power of the telescopes, coupled with new or upgraded electronics, means that we can make detailed, wide area maps faster than has been possible with any other instrument to date.

Q: How will CARMA continue to develop?
A: We see ourselves as place where students and post-docs will come and get first-hand experience with a state-of-the-art instrument. This is really where we expect to teach and inspire instrumentalists and scientists for the future. The CARMA we're building today needs to continue to evolve to stay alive and attract the best researchers of the next generation.

CARMA's cyberinfrastructure

Athol Kemball, leader of the radio astronomy imaging team at NCSA, explains the tools to transfer, process, and store data that are an integral part of the CARMA project.

Q: Can you speak for a minute about the need in astronomy in general, and with the CARMA project in particular, for cyberinfrastructure support?
A: Modern telescopes are faced by two new challenges: increasing data volumes and the need to make telescopes easily accessible to the broadest user community. The data rates are being driven exponentially by advances in electronics and detectors. These data rates and the need for open access make custom interactive reduction untenable as a routine means of analyzing the data; smarter scientific workflows are needed. Automated pipeline reduction is essential for new telescopes such as CARMA so that the telescope can be used by the broadest possible astronomical community. Cyberenvironments in turn offer the best means of providing access by the community to the underlying compute and data resources needed for CARMA.

Q: Can you describe the capabilities that were developed for CARMA?
A: The CARMA data archive resides at NCSA, and we will also host the CARMA data reduction pipeline. NCSA has participated closely in the community consortium that built the software system for CARMA, and we lead the team responsible for cyberenvironments, archiving, calibration, and imaging in the CARMA consortium. During construction we contributed to the overall flow of science data and meta-data from the telescope to the archive at NCSA and all cyberinfrastructure components involved in this data flow. NCSA also invests in community codes for high-performance calibration and imaging. This application cyberinfrastructure is a vital part of building the automated science workflows for CARMA. Our current focus is on deploying a cyberenvironment to host these capabilities.

Q: Can you describe the process of developing the CI component of CARMA?
A: The CARMA computing team is a distributed community collaboration, including members from the University of Maryland, University of California, Berkeley, Caltech, the University of Chicago, and NCSA, and it is governed by the requirements of the user community.

In our cyberinfrastructure development we made extensive use of past experience with the archive for the Berkeley-Illinois-Maryland Array (BIMA), which is also hosted at NCSA. We reused infrastructure from this prior project, augmenting this with more recent developments in shared cyberinfrastructure development at NCSA and elsewhere in the cyberinfrastructure community. The success of a project such as CARMA requires extensive reuse of common cyberinfrastructure components wherever possible.

For more information: http://www.mmarray.org/intro.html