Astron. Astrophys. 334, 1016-1027 (1998)

2. Observations

Most of the observations of the 183 GHz line were carried out on Jan 16th, 1994 with the IRAM-30m telescope located on Pico Veleta, the day before the observations of the same line in Orion were conducted (Cernicharo et al., 1994). The receivers, conveniently tuned at the frequencies of the three lines that were observed (CO p-H₂ O and SiO v =1 the arrangement of the spectrometers, the meteorological conditions and uncertainties in the calibration were similar to those stipulated for the Orion observations and are described in Cernicharo et al. (1994). The reference positions were obtained by oscillating the subreflector every two seconds ("Wobbling" procedure) between two symmetrical positions separated by in azimuth of the source. Weather conditions during that period were excellent and the 183 GHz line was detected in twenty one stars. Two more objects, which could not be observed in 1994, were detected at 183 GHz in December 89 and in January 91, thus a total of 23 sources were detected at 183 GHz. In addition, some of the objects observed in 1994 were also observed in 1989 and in 1991.

On December 24th, 1989, the 2mm receiver was tuned in double band, and the frequency of the image band was found to be 191.2 GHz. Since this frequency is near the limit of the receiver band, the image band was in effect attenuated by 4 dB. The receiver was tuned in single band on Jan 13th, 1991, and an attenuation of 7 dB for the image band was obtained. In all the observational runs, the calibration was made by using two absorbers at different temperatures, which allowed measuring the emissivity of the atmosphere. During these two periods, the temperature of the atmosphere was C and the relative humidity was less than 10%. In order to verify the accuracy of the calibration, the opacity of the atmosphere was measured by means of the "skydip" procedure. The achieved values as far as the opacity is concerned were close to 1, matching the values obtained from the calibration when two absorbers were used. Typical values of the system temperature were 8 10³ and 5 10³ K in 1989 and 1991, respectively.

In 1991 and 1994 we made some measurements of the atmospheric emission at 183 GHz, which are carefully analyzed in Pardo et al. (1996), together with absorption measurements toward continuous sources (Venus, Saturn, Mars). Fig. 1a shows a spectrum of atmospheric emission, and Fig. 1b shows the absorption of atmospheric water vapour in the direction of Venus. The atmospheric opacity can be derived from these spectra. With a spillover contribution of 27 K and a 7 dB attenuation of the image band, the ATM atmospheric model (Cernicharo, 1985) fits into an opacity of 0.96 in the line center and 0.7 in the wings for the spectrum shown in Fig. 1a. With the same parameters and assuming a temperature of 340 K for Venus, the model fits an opacity of 1.08 for the spectrum shown in Fig. 1b. These values are slightly lower than those found in 1989. We estimate that there is an uncertainty in the intensity scale (depending on elevation) amounting to a factor 1.5-2 for an atmospheric transmission of 20-40% at the zenith.

Fig. 1. a Spectrum of atmospheric emission around 183 GHz; b spectrum of atmospheric absorption around 183 GHz in the direction of Venus

Fig. 2. a  upper, b  lower, c  next page. Line profiles observed towards the stars of the sample. The water lines are labeled by their frequency. The dotted line in each panel indicates the LSR velocity of the star. The panel corresponding to a given source is indicated in Table 1 (F2)

Fig. 2. (continued)

The p-H₂ O line at 325 GHz was observed on February 12th, 1994 with the IRAM 30m telescope in Pico Veleta. On that date, we detected the line in eight of the stars observed at 183 GHz. The 325 GHz line was first observed in star forming regions by Menten et al. (1990) using the CSO telescope in Mauna Kea (Hawaii). Similarly, it has been detected in U Her by Melnick et al. (1993) and in several other stars by Yates et al. (1995). We used an SIS receiver tuned in double side band, with a receiver temperature of K. The spectrometers consisted of a 256 100 kHz filter bank and an autocorrelator of 1024 channels and 80 MHz bandwidth, giving velocity resolutions of 0.09 and 0.07 km s^-1, respectively. The observational procedure was the same as for the observations at 183 GHz. The atmospheric absorption at 325 GHz is higher than at 183 GHz, thus only under exceptional meteorological conditions the line could be observed. The opacity at the zenith was during the observational period, and the transmission at the elevation at which the sources were observed was typically 15-20%. Uncertainty in the intensity scale amounted for a factor of .

Finally, the o-H₂ O line at 22 GHz was observed with the 100m Effelsberg telescope on Jan 27th, 1994. We have observed the line in fifteen of the sources observed at 183 GHz. The HPBW of the telescope is at 22 GHz. The spectrometer was an autocorrelator with a frequency resolution of MHz (0.82 km s^-1). Typical system temperatures were of order of 300 K. The main beam efficiency of the telescope at 22 GHz was .

The observations on the lines p-H₂ O at 325 GHz, o-H₂ O at 22 GHz, and those of p-H₂ O at 183 GHz conducted in 1994 were carried on at time interval of about one month and half. Since the pulse period of most of the observed Mira stars is around one year or even longer, the phases of the stars in the periods during which the three water lines were observed did not vary significantly.

All the data are calibrated in units of main beam brightness temperature () for the intensity scale. In order to obtain the intensities in Janskys, the values of the 183 GHz lines have to be multiplied by six. Figs. 2a-c display the observed lines towards our star sample, and the column labeled F₂ in Table 1 indicates the panel in Fig. 2 which displays the observations of each source.

Table 1. Coordinates, IRAS flux densities in the 12, 25, 60, and 100 µm bands, spectral type and variability type of the observed sources. Columns F₂ and F₅ indicate the panels in Figs.  2 and 5, respectively, which display the observations made for each source. Type of object: M indicates Miras, SRb semiregular variables, SRc semiregular supergiants, SG supergiants, OH and IRC mean OH/IR and IRC objects

© European Southern Observatory (ESO) 1998

Online publication: June 2, 1998

helpdesk.link@springer.de