CARMA was created through the merger of the Owens Valley Radio Observatory (OVRO) Millimeter Array and the Berkeley-Illinois-Maryland Association (BIMA) Array. These arrays pioneered the development of millimeter wavelength interferometry. The Owens Valley Array was built and operated by Caltech in the Owens Valley near Big Pine California. It consisted of six 10.4-meter telescopes. The BIMA Array was started by the University of California Berkeley with three 6.1-meter telescopes and expanded over the years to nine telescopes with the addition of the University of Illinois at Urbana-Champaign and the University of Maryland as partners. The BIMA Array was located at the Hat Creek Radio Observatory in northern California. In 2004, the two groups, in cooperation with the National Science Foundation, agreed to merge the two arrays into a single array on a higher site, the current Cedar Flat location. The merger involved moving the antennas from their respective locations and the creation of completely new hardware and software array infrastructure. The construction and commissioning took roughly three years. The array began routine science operation in January 2007. The University of Chicago joined CARMA in the Summer of 2008, adding eight 3.5-meter antennas to CARMA. These antennas were initially operated as an independent array, and have recently been integrated with the other fifteen antennas to form a single array. CARMA has better imaging performance and better sensitivity than any of the original arrays. It is the largest array of its type in the northern hemisphere and serves as a training ground for students and young scientists.
The Cedar Flat site was chosen because of its altitude and convenience to
existing all-weather roads. The altitude of the site is an important factor
because water vapor in the atmosphere degrades the quality of observations.
At millimeter wavelengths, it is the total amount of water vapor in the
atmosphere that is important, not the presence or absence of clouds, although
dark rain clouds of course contain a lot of water vapor (and droplets), which
hurt observations. The high altitude Cedar Flat site, at 7200 feet,
typically has a factor of two less water vapor than the OVRO site, at 4000 feet.
At Cedar Flat, the array is located just off of California State Highway
168 which leads to Nevada. The road is maintained, and plowed in the
winter, by CalTrans. This facilitates our year-round 24-hour per day
operation of the array.
CARMA has receivers to observe in three bands which correspond to wavelengths of roughly 7 millimeters, 3 millimeters, and 1.3 millimeters. These wavelengths are traditionally referred to as part of the radio wavelength regime of light. Because the sky does not scatter light at these wavelengths, CARMA can observe the Universe both day and night. Of course, we must avoid the Sun itself, unless we are intending to observed the Sun -- which is done on occasion.
The surfaces of the antennas are parabola-shaped as are the surfaces of traditional optical telescopes, but the surface is much rougher than a traditional mirror. The key point is that the surface of the antenna needs to be very smooth and reflective compared to the wavelength that is being observed. In the case of CARMA, our shortest operating wavelength is roughly 1 millimeter, so the surface of the antenna needs to be about a factor of 10 smoother than that wavelength to properly reflect the radio waves. In fact, the surfaces of the antennas are accurate to 30 microns (1/30 of a wavelength) which means that they reflect very well for our purposes. However, since visible light has a wavelength of 0.5 microns, the surface is extremely rough, and a poor reflector, to visible light. We actually like the antenna surface a little rough because that means that it will not focus all of the light and heat from the Sun onto a single point which might melt the secondary mirror or one of the feed-legs supporting the secondary mirror.
The key to getting high quality images with fine detail is to have a telescope with good resolution. Resolution is the ability to make an image which separates two sources that are close together. The resolution of a telescope depends on the diameter of the telescope: larger diameter telescopes are capable of higher resolution. In fact, the resolution of a telescope depends on the diameter of the telescope measured in terms of the number of wavelengths across the diameter -- at the wavelength being observed. Thus, optical telescopes, working at 0.5 microns wavelength (blue-green color light), can obtain 1 arcsecond resolution with a 12 centimeter diameter telescope; a telescope working at 1.3 millimeter wavelength, needs to be 2000 times bigger to achieve 1 arcsecond resolution -- 300 meters in diameter! It rapidly becomes both expensive and technically difficult to build telescopes of this size. The less expensive solution is to build smaller telescopes which are placed near each other and whose signals are combined in such as way that they give the resolution corresponding to the separation of the telescopes rather that the size of the individual telescope. This is the advantage of an array.
The resolution of the array is determined by distance between the telescopes: the further the telescope are spread apart, the higher the resolution. The overall quality of the image -- the ability of the image to accurately reproduce the actual astronomical source -- is dependent in getting information from antennas at a broad range of separations. The CARMA antennas can be placed at separations ranging from 4 meters to 2 kilometers. At the maximum separation of 2 kilometers, CARMA has a resolution at 1.3 mm wavelength of 0.15 arcsecond, comparable to the Hubble Space Telescope at optical wavelengths.
The number of the antennas in the array is not a magic number. More antennas are better because they result in better images but, of course, more antennas cost more money to build and maintain. The exact number in the CARMA array is a convergence of our funding history and the requirements of the science that we do.
The three antenna sizes has its origins in the history of independent development of the arrays which joined into CARMA. The sizes of 10.4, 6.1 and 3.5 arose from different design considerations and different primary science goals in the origins of the three arrays. Now, in the combined CARMA array, the three different sizes have advantages in making high quality science images. However, if we could turn back the clock and start all over with the same pot of money, it is not clear that we would chose to build exactly these antennas.
CARMA is a scientific facility supported by the National Science Foundation and the CARMA member Universities. Time on the array is awarded on a competitive basis to astronomers around the world. The member Universities operate CARMA, supplying observers who actually run the array, technical people to maintain, upgrade, and develop the array, and support scientists who assist in producing science from the array. The primary purposes of CARMA are to: do world-leading science, develop next-generation instrumentation, and develop the next generation of scientists and technical leaders.
If you are using CARMA as a reference in a publication, please read this acknowledgement page with the specifics on how to list a standard CARMA reference.