The ATCA Cosmic Microwave Background Radiation Anisotropy Experiment

The Observing Strategy

Our observing strategy has been to make full Earth-rotation synthesis observations of `empty' fields in a special ultra-compact 122-m array configuration -- with five 22-m diameter antennas located 30.6 m apart in an E-W line -- in order to maximize the brightness sensitivity of the imaging.

Novel observing strategies were adopted to eliminate any spurious signals, for example, the array phase centre was offset about 10 primary beam widths from the antenna pointing centre in order to keep any correlator based offset errors well outside the primary beam region of interest; bandwidth decorrelation was avoided by observing in multi-channel continuum mode. The fields observed were selected to be relatively devoid of strong discrete sources and located at declinations about -50^o to avoid shadowing and maximize surface brightness sensitivity. These observations were made at the highest available frequency of 8.7 GHz to minimize discrete source confusion.

The three baselines between antennae spaced 61 m apart are used to construct a model of the foreground confusion; this is then subtracted from all of the visibility data. Images with high surface brightness sensitivity are synthesized at 3 cm using just the 30-m baselines and they are examined for CMB anisotropy; these images have an angular resolution of 2 arcmin. 6-cm observations in a scaled 244-m configuration help identify extended (on the 2-arcmin scale) foreground synchrotron/optically-thin thermal sources that may be resolved in 60-m baselines in the 3-cm observations.

Because the confusion is determined simultaneously and using baselines between the very same antennae, errors in the estimation of confusion owing to variability in the foreground sources and calibration errors are reduced. And because the confusion is estimated as a model fit to visibility data, deconvolution errors are avoided.