Observations

Observations of the 12CO(1-0) transition at 115.271 GHz were made during two periods, 1996 April-October and 1997 April-June, using the Mopra antenna, operated by the Australia Telescope National Facility, CSIRO. At this frequency the effective diameter of the antenna is 15 m, producing a half-intensity beam width of 43 arcsec. A cryogenic 3-mm SIS receiving system was used, for which a receiver temperature (SSB) of 130 K was assumed for all observations.

Initially observations were made on a 2 arcmin grid covering an area of 1.5 deg x 2 deg. This produced a coarsely sampled map which allowed us to trace the overall molecular-cloud distribution. A total of 7000 pointings were obtained. Follow-up observations were then made on a finer but still under-sampled grid of 1 arcmin towards areas of interest. We used an observing technique with a reference position offset by 2 degrees to avoid a contribution from off-source CO emission. One-minute integrations were made at each `source' position, with a five-minute reference observation every thirty minutes.

The spectra were obtained using a digital correlator which was configured into 2 bands, each consisting of 1024 channels extending over 64 MHz. One band was tuned to the CO transition, giving a velocity resolution of 0.20 km/s. For each observing run the band was centred on a radial velocity of -20 km/s (LSR). The second band was tuned to the 86-GHz SiO transition. A pointing accuracy of better than 10 arcsec was obtained periodically observing SiO masers near Carina.

At a wavelength of 2.6-mm, atmospheric absorption is significant. We adopted the `chopper-wheel' method (Ulich & Haas 1976) to correct for this attenuation. By measuring the signal from a black-body absorbing paddle every 30 mins by means of a method described by Hunt (1997), observed intensities were converted to a temperature scale corrected for atmospheric effects. To provide a temperature scale compatible with that used at the 15-m Swedish ESO Sub-millimetre Telescope (SEST), we observed the SEST calibration sources in Orion and M17 on a daily basis throughout each observing period and compared the CO temperatures to those obtained with the SEST. To make the temperature scales consistent an average scaling factor of 1.5 was applied to the Mopra data. The final intensity calibration is accurate to 10-15 percent.

The data were processed using a combination of data reduction packages on ATNF facilities. For the preliminary reduction simple baselines were subtracted from the individual spectra and Hanning smoothing was applied. The frequency scale was corrected to velocity (LSR) without a diurnal variation correction, introducing a maximum error of 0.5 km/s. A spectral cube (RA, Dec, velocity) was made using a Gaussian algorithm to interpolate between each spatial point. A smoothing radius of 0.8 arcmin was applied, giving an effective FWHM beam of 1.75 arcmin.

© Copyright Astronomical Society of Australia 1997