A world-class observatory in the desert
The ALMA Observatory is operated at two distinct sites, far away from comfortable living conditions of modern civilization.
The ALMA Operations Support Facility (OSF) is the base camp for the every-day, routine operation of the observatory. It is located at an altitude of about 2900 meters, quite high compared to standard living conditions, but still quite acceptable for scientific projects in astronomy of similar scope. However, the OSF will not only serve as the location for operating the Joint ALMA Observatory, it is also the Assembly, Integration and Verification (AIV) station for all the high technology equipment before being moved to the Array Operations Site (AOS), located at 5000 meters altitude.
The ALMA antennas are located on the Chajnantor plateau of the Chilean Andes at an altitude of 5000 meters. This site offers exceptional atmospheric transparency at millimeter and sub-millimeter wavelengths.
The geographical location of ALMA is
latitude: -23.029° ; longitude: -67.755°
The Operations Support Facility (OSF)
The OSF is, in many aspects, the center of activities of the ALMA project. The OSF is the focal point of all antenna Assembly-Integration-Verification (AIV) activities. Antenna assembly is done at the OSF site at three separate areas, one each for the antennas provided by North America (VERTEX), Japan (MELCO), and Europe (AEM Consortium).
The OSF is also the center for activities associated with commissioning and science verification as well as Early Science operation. During the operations phase of the observatory it is the workplace of the astronomers and of the teams responsible for maintaining proper functioning of all the telescopes.
The Array Operations Site (AOS): The second highest building in the world
The construction of the AOS Technical Building started in October 2005 and the outer shell was completed by mid 2006. Inside construction work was completed in summer 2007. Human operations at the AOS is limited to an absolute minimum, due to the high altitude. The AOS Technical Building houses the ALMA Correlator . Digitized signals received from the radio telescopes are processed here and further transmitted to the data storage facilities located at the OSF.
A High Altitude Road with Super-Highway Dimensions
The construction of the OSF and AOS sites and their access required substantial efforts of the ALMA project. Obviously, there was no access to these two remote locations (see Figure below). The OSF site, located at 2900 meters altitude, is about 15 kilometers away from the closest public road, the Chilean highway No. 23. The AOS is another 28 kilometers away from the OSF site. Thus, one of the first projects to be accomplished by ALMA was to construct an access road not only to the OSF but also to the AOS road, 43 km in length, not only at high altitudes, but also with sufficient width to regularly transport a large number of large radio telescopes with a diameter of 12 meters.
Access to the AOS and OSF facilities
When the sky is clear, the principal sources of atmospheric attenuation are the molecular resonances of water vapor, oxygen and ozone. The resonances of water vapor and oxygen are pressure broadened and cause attenuation far from the resonance frequencies. The figure below plots the absorption versus frequency. Below 30 GHz the absorption is dominated by the weak transition of H20 at 22.2 GHz, and rarely exceeds 20% in the zenith directions. The oxygen bands in the 53-67 GHz band are considerably stronger, and no astronomical observations can be made from the ground in this band. A similar effect happens with the isolated 118 GHz O2 line, which makes observations impossible in the 116-120 GHz band. There is a series of strong water vapor lines at 183, 325, 380, 448, 475, 557, 621, 752, 988 and 1097 GHz and higher. Observations can be made in the windows between these lines at dry locations. This is where the different ALMA Observing Bands have been defined, as shown in the plot below which shows the atmospheric transmission at the ALMA site with the different observing bands overlaid. The bands plotted in red and blue are available for early science (bands 3, 4, 6, 7, 8, 9 and 10).
No online corrections for the attenuation of the signal due to atmospheric absorption are made for ALMA. Instead, absorption is dealt with at the calibration stage. This is done in two different ways. Firstly, by comparing the observed flux of a well-known flux calibrator source to its known flux. One disadvantage of this technique is that some calibrators may be time variable sources, or resolved at different frequencies and/or array configuration. Constant monitoring of the known calibrators helps overcoming the variability problem, while comparison with model images of the calibrator can be used to compare the observed flux of an 'over-resolved' calibrator (i.e when the interferometer is missing some of the calibrator's flux).
A second method for correcting the atmospheric attenuation of the signal from an astronomical source is to measure the total system temperature. This is done several times during a typical observation and is used in the calibration phase. The measurement is done using the chopper-wheel method. This does not measure the atmospheric opacity directly but, by combining measurements of the sky and of a chopper-wheel at known temperatures, makes it possible to eliminate the effects of atmospheric absorption and to measure the component of the antenna temperature due to the source in the absence of the atmosphere (i.e the actual flux density of the source). The bandpass calibration ensures that the atmospheric absorption is removed correctly across the observed bandwidth.
At radio and sub-mm wavelengths, the most important non-uniformly distributed quantity in the troposphere is the water vapor density. In the atmosphere, N2, O2, Argon and CO2 are well mixed. Water is not. Variations in the water vapor distribution in the atmosphere that move across the interferometer cause phase fluctuations that degrade the measurements. Fluctuations in water in the atmosphere may cause refraction of a few arcseconds, causing the source to 'move'. First radio astronomers called this anomalous refraction, although the term is rather inappropriate as this effect is closely related to Kolmogorov turbulence. Most of the fluctuations are located in inversion layers (warm dry air on top of cold wet air).
The excess propagation path in a particular direction due to water vapor can be estimated from measurements of the brightness temperature in the same direction at frequencies near water vapor resonances. The technique, called Water Vapor Radiometry, is currently in place to correct for atmospheric phase fluctuations in ALMA data.
Accurate phase calibration is a critical requirement for ALMA, and the baseline design of ALMA uses a 183 GHz receiver (mounted slightly off-axis from the astronomical beam) to measure a strong atmospheric water line. Under various assumptions about the atmospheric pressure and temperature, and the location of the turbulence, the electrical path above each antenna can be derived.