Erik LeitchThe Sunyaev-Zel'dovich Array, taken during the
Fall of 2005, Owens Valley Radio Observatory, eastern Sierra
Nevada.
| One of the most important developments of the last few years has been the detection, through observations of the cosmic microwave background (CMB) and supernova studies, of a form of energy that is accelerating the expansion of the universe. Dubbed dark energy by analogy with dark matter (though the designation signifies little more at this point than our state of ignorance about it), this appears to be the dominant component of the universal energy budget, accounting for roughly 70% of the universe's bulk. |
Erik Leitch |
The Sunyaev-Zel'dovich Array, taken during the
Fall of 2005, Owens Valley Radio Observatory, eastern Sierra
Nevada. |
While dark energy cannot be observed directly, we can infer its basic properties from its effect on structure formation in the universe. Just as an ecologist can learn about the food supply by studying how animal populations evolve with time, physicists can learn about dark energy by studying the population statistics of the universe's inhabitants — in this case, galaxy clusters.
As with any population study, the trick is to find your subjects, and to properly quantify any selection bias that your method of finding them might introduce.
The SZA gets its name from the means by which it detects galaxy clusters: the scattering of CMB photons as they pass through the hot ionized gas in the centers of clusters, known as the Sunyaev-Zel'dovich effect (SZE). In short, we use the CMB as a backlight against which galaxy clusters can be seen by the shadows they cast.
Because clusters are detected via the SZE, an interaction with the uniform background illumination provided by the CMB, and not by their own light, the detection is redshift-independent; any cluster massive enough to be seen with the SZA will be detected no matter how distant the cluster. This means that the SZA (and indeed any experiment which uses the SZE to detect clusters) will find all clusters above the detection threshold, all the way back to the epoch at which the clusters first formed.
The experiment consists simply of deep integration on selected patches of sky, a program which began in earnest in the fall of 2005. In principle, the minimum detectable mass should be a simple function of the available integration time and the amount of sky surveyed. In practice, however, the mass threshold for the survey will also depend on the details of the gas distribution and dynamics within the individual clusters. Although the initial survey is being conducted at 30 GHz, these secondary effects will be investigated with multi-wavelength, high-resolution followup observations, first with the SZA itself and later with the SZA integrated into the Combined Array for Millimeter-wave Astronomy (CARMA). The full correlator was installed in February 2005; after commissioning observations, and targeted science observations during the spring and summer, the SZA has been conducting a blank-field survey of ~12 square degrees since the Fall of 2005.
The SZA is not a single telescope, but an array of 8 telescopes operating together as an interferometer. An interferometer does not detect light in quite the same way as an ordinary telescope, by measuring the total power collected by a single dish; instead, it looks at differences between the light falling on pairs of telescopes. Like water waves, light waves can interfere with each other, producing a complex pattern of intensity enhancements where the waves constructively interfere, and nulls where they destructively interfere.
As light from a source washes over the array, an interferometer detects this interference pattern — hence the name. The source's structure on the sky can then be inferred from the interference pattern in much the same way that you might infer the size and shape of a stone thrown into a pond from the pattern of ripples left in its wake.
The native resolution of an interferometer depends not on the size of the individual telescopes (as with a traditional single telescope), but on their separation, pairs of telescopes with large separations providing sensitivity to small-scale structure, while short spacings are sensitive to large-scale structure on the sky. The 8 SZA telescopes are divided into a compact array of six telescopes, which provides maximum sensitivity to the (large-scale) emission from clusters, and two outlier telescopes, yielding maximum sensitivity to the (small-scale) emission from point sources. The SZA is thus two instruments in one: a cluster-finding machine, and a point-source subtractor which can be used to cleanly remove contaminating flux from bright radio sources.
Each SZA telescope is equipped with two HEMT-amplified receivers,
allowing operation of the array at either 30 GHz (Ka-Band) or 90 GHz
(Q-Band). The Ka-Band receivers use traditional monolithic HEMT
amplifiers, with typical noise figures in the range ~ 15 K. At
Q-Band, the receivers use MMIC amplifiers, with typical noise
temperatures of ~ 30 K (??). The SZA correlator uses a prototype of
the digital correlator designed for CARMA, configured to correlate in
16 contiguous 500-MHz bands at either frequency.
The SZA is located at an elevation of 4000 ft., at Caltech's Owens
Valley Radio Observatory, in the high desert of the eastern Sierra
Nevada. In the rain-shadow of the Sierras, the valley floor provides
a superb dry site for centimeter-wave astronomy, with typical
wintertime opacities at 30 GHz of 3%.
| One of the most important developments of the last few years has been the detection, through observations of the Cosmic Microwave Background (CMB) and supernova studies, of a form of energy that is accelerating the expansion of the universe. Dubbed dark energy by analogy with dark matter (though the designation signifies little more at this point than our state of ignorance about it), this appears to be the dominant component of the universal energy budget, accounting for roughly 70% of the universe's bulk. |  |
Model of the South Pole Telescope, showing the
10-meter dish inside its ground shield. As of February 2006, the
building and support tower for the telescope were complete. |