Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 364, 613-624 (2000)

Previous Section Next Section Title Page Table of Contents

2. Observations

2.1. Molecular observations

The observations were performed in 1996-1997 with the 15-m SEST telescope on La Silla, Chile. The telescope and its instrumentation are described by Booth et al. (1989). The most important parameters of our measurements are summarized in Table 1. Further details are given below.


Table 1. Parameters of molecular line observations.
*1) The system temperatures are given in the [FORMULA] scale.

The observations were performed with SIS receivers in a single-sideband mode. At 220 GHz, we used dual beam switching with a beam throw of [FORMULA] 12´ and 2 acousto-optical spectrometers in parallel: (1) a 2000 channel high-resolution spectrometer with a 86 MHz bandwidth, 43 kHz channel separation and 80 kHz resolution and (2) a 1440 channel low-resolution spectrometer (LR1) with a 1000 MHz total bandwidth, 0.7 MHz channel separation, and 1.4 MHz spectral resolution. The LR1 band was centered on the HNCO([FORMULA]-[FORMULA]) transition. However, it also covered C18O(2-1), SO(65-54) and other lines (see Table 1). The C34S(3-2) measurements at 145 GHz were made in a dual beam switching mode, too, using a high-resolution spectrometer as backend. The beam size is 23" (HPBW) at 230 GHz and 35" at 145 GHz.

Throughout the paper, we adopt the IRAS co-ordinates as the central source position ([FORMULA], [FORMULA]. The area of 40"[FORMULA]40" around this position was mapped with 10" spacing in SO(65-54), C18O(2-1), and HNCO([FORMULA]-[FORMULA]) and with 15" spacing in C34S(3-2). The peak position was also observed in 34SO(65-54).

The CO(2-1) map for IRAS 12326-6245 was sampled with 2/3 beamwidth intervals of 15[FORMULA] The observations were made in the position-switch mode with an OFF position of 30´ to the east and an integration time of 30 seconds per ON and per OFF position. Based on the repeated measurements of smaller parts of the map, the total integration time per ON position ranged between 1 minute and 3.5 minutes, resulting in a [FORMULA](rms) between 0.18 and 0.46 K. The pointing was checked every two to three hours with the source W Hya and was found to be accurate to better than 6[FORMULA]

We express the results in units of the main beam brightness temperature ([FORMULA]) assuming the main beam efficiencies [FORMULA] at 230 GHz and 0.52 at 220 GHz. The temperature scale was checked by observations of Orion A.

2.2. Bolometer observations

The continuum observations were performed in March 1996 with the one-channel He3-cooled SEST facility bolometer (Kreysa 1990). The bolometer has an equivalent bandwidth of [FORMULA] 50 GHz and a central frequency of [FORMULA] = 236 GHz ([FORMULA] = 1.27 mm). The beam size at this wavelength is 23" (HPBW).

The maps were obtained with the standard "double beam" technique (Emerson et al. 1979), i.e. by scanning the telescope continuously in azimuth over the source position while chopping with a frequency of 6 Hz and a beam throw of 67" in scan direction. The beam switching was performed with a focal plane chopper. A scanning velocity of 8"/second was used and the elevation spacing between adjacent scans was 8[FORMULA]

Three individual maps with a size of 5´[FORMULA] 4´ were combined to produce the final map of the source. The atmospheric transmission was measured by sky dips and amounted to [FORMULA] during the whole observing run. Telescope pointing and focus were checked frequently towards nearby quasars. The pointing was repeatable within 5[FORMULA] Uranus served as calibration standard, adopting a brightness temperature of 96 K (Griffin & Orton 1993). The average 1 [FORMULA] rms noise in the final map (outside the source) is 48 mJy/beam. Data reduction was performed with the MOPSI software package (R. Zylka).

2.3. Imaging in H, J, [FORMULA] and H2

The near-infrared (NIR) broad- and narrow-band imaging was performed using IRAC2b (Moorwood et al. 1992) at the ESO 2.2-m telescope in June, 1998, during the time slot alloted to the Max-Planck-Institut (MPI), Heidelberg. Dithered images were taken at five positions yielding an overall field-of-view of about 160"[FORMULA]160[FORMULA] The resulting total integration times for the central region are as follows: 800 s (J), 400 s (H), 200 s ([FORMULA]), and 20 minutes for the narrow-band filters BP4 (continuum, 2.105 µm) and H2 (2.122 µm). The limiting magnitudes (3 [FORMULA] point source detection) of the J, H, and [FORMULA] images are 17, 18, and 18.5 mag, respectively. In order to obtain a continuum-subtracted H2 image, the BP4 image was convolved to match the point spread function (PSF) of the H2 image. After this step, the continuum emission was subtracted from the H2 image. Nevertheless, residuals of bright stars are still present in the subtracted image (Fig. 2b).

The intrinsic image scale of lens C ([FORMULA]) was resampled to [FORMULA] during the mosaicking of the final image. The angular resolution as derived from stellar profiles is 1[FORMULA] The astrometry is based on stellar positions extracted from the DSS2 image of the region and is accurate to [FORMULA] 1[FORMULA]

2.4. N- and Q-band imaging

Diffraction limited thermal-infrared images were obtained in March, 1998 during Max Planck time using MANIAC (Böker et al. 1997) at the ESO 2.2-m telescope. The observations were performed with the common chopping/nodding technique at a pixel scale of [FORMULA] and a total on-source integration time of 3.5 min. A chopper throw of 30" was applied. The individual images were subject to a wavelet-filtering algorithm (Pantin & Starck 1996) which preserves both flux and spatial resolution while suppressing the noise considerably. These images were then resampled to half the original pixel size and combined using a shift-and-add algorithm to compensate for image motion. The coarse astrometry derived using the telescope offsets from the reference star indicated that two infrared sources are close to the unresolved ultracompact H II regions found by Walsh et al. (1998). Thus, the final astrometry was tied to the radio positions. The N and Q band photometry of the detected sources (above 3 [FORMULA] noise level) is based on the 12.13 µm and 21.34 µm MSX flux densities. The 3 [FORMULA] detections limits are 0.16 Jy in N and 11 Jy in Q, respectively. The individual flux contributions were derived using a multi-component PSF fitting algorithm. The MIR source 4 is either a binary or is associated with extended emission. For the derivation of the flux, it was treated as a binary source. We should stress that MIR3 is present in the Q band image at a 2 [FORMULA] level.

We also added mid-infrared data from the MSX-SPIRIT III point source catalog (Egan et al. 1999). The 1 [FORMULA] error ellipse of the point sources detected by MSX in the field is [FORMULA].

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: January 29, 2001