2 Observations

2.1 KOSMA observations

Between 1989 and 1993 we have used the old (before the 1996 refurbishing) KOSMA 3-m telescope at Gornergrat (Switzerland) to observe ¹²CO(J=1-0, J=2-1 and J=3-2) and ¹³CO(J=1-0, and J=2-1) towards MBM 32. Based on a quick ¹²CO(1-0) map consisting of 555 spectra on a 4$^\prime$ raster made with a relatively low velocity resolution (0.93 km s^-1), we obtained a higher velocity resolution map (0.15 km s^-1) on a 2$^\prime$ raster of a slightly smaller region. Subsequently we observed parts of the latter region on 2$^\prime$or 1$^\prime$ rasters in the ¹²CO(2-1, and 3-2), and ¹³CO(1-0, and 2-1) transitions. One position (at offset (8$^\prime$, 19$^\prime$)) was also observed in C¹⁸O(2-1) and ¹³CO(3-2) (see Schreiber et al. [1993]). The observational parameters are summarized in Table 1. In addition to the maps we made in ¹²CO(2-1), ¹³CO(2-1), and in ¹²CO(3-2), deeper observations in there transitions (typical rms noise levels ( $T_{\rm {R}}^*$) of 0.08, 0.03, and 0.10 K) were made along a line through MBM 32 in order to study the line ratios. The raster size was in this case 1$^\prime$ for both (2-1) transitions and 40 $^{\prime\prime}$ for ¹²CO(3-2). All offsets in this paper are given with respect to the MBM position $\alpha (1950)=9^{\rm h}28^{\rm m}42^{\rm s}$, $\delta (1950)=+66$$^\circ$5$^\prime$. During all observations we frequently observed a calibration position towards offset (4$^\prime$, 4$^\prime$) and found that intensities typically have uncertainties of about 10%. The telescope and receivers used have been described by Winnewisser et al. ([1986], [1990]). The pointing accuracy was usually 10 - 15 $^{\prime\prime}$, but during daytime observations pointing errors might have been 30 - 40 $^{\prime\prime}$ due to thermal movements of the building. At 115 and 110 GHz we used frequency switching and at other frequencies we observed in position switching mode using an off-source position 30$^\prime$ south of MBM 32. At 115 and 110 GHz the atmospheric transmission was estimated by means of skydips. At higher frequencies the sky transmission was estimated by measuring the radiation temperature of blank sky at the elevation of MBM 32. Analogous to the standard chopper wheel method, we then corrected the intensities to the $T_{\rm A}^*$ scale. The backend consisted of a Medium Resolution (channel width 167 kHz; MRS), or a High Resolution (channel width 38.6 kHz; HRS) acousto-optical Spectrometer.

 

 

Table 1: Parameters of KOSMA observations

	12CO(1-0)		13CO(1-0)	12CO(2-1)	13CO(2-1)	12CO(3-2)
$T_{\rm {rec}}$ (K)	355-400		310-360	370-400	400-470	320-350
$T_{\rm {sys}}$(zenith) (K)	1000-1600		740-820	1000-1800	1200-1800	950-1550
beamsize ($^\prime$)	3.9		3.9	2.2	2.2	1.2
$\eta'$	0.76		0.76	0.67	0.67	0.52 (89/90)
						0.47 (90/91)
						0.60 (91/92)
backend	MRS	HRS	HRS	MRS	MRS	MRS
chan. width (km s-1)	0.43	0.082	0.086	0.21	0.22	0.14
vel. res. (km s-1)	0.93	0.15	0.16	0.45	0.47	0.30
rms noise ( $T_{\rm {R}}^*$) (K)	0.19	0.44	0.13	0.18	0.07	0.19
# positions	555	475	126	540	61	310
raster ($^\prime$)	4	2	2/4	2	2	1

All intensities presented in this paper are on the scale $T_{\rm R}^*=T_{\rm A}^*$/$\eta'$. The sizes of clumps in MBM 32 are several arcminutes, but the whole cloud has an extent of 0 $.\!\!^\circ$5 to 1$^\circ$. Therefore, to compare the intensities in the different transitions we corrected them by using the average value $\eta'$ of the telescope efficiencies derived from Jupiter ( $\eta _{\rm {mb}}$) and from Moon measurements ( $\eta _{\rm fss}$). At 345 GHz the efficiencies used differ for the three observing winters. The Jupiter measurements were in all cases made in good weather, but it is not known whether the efficiencies were constant during each winter or not. The values which we used are listed in Table 1. For the 115 and 110 GHz measurements we applied in addition a correction $\eta$ $_{\rm {R}}$$\eta$ $_{\rm {rss}}=0.91$, which is already corrected for in the calibration at the higher frequencies. After applying these corrections we found a ratio $\int T_{\rm R}^*$(¹²CO(1-0)) HRS/MRS of 1.25 (which we cannot explain), and to be able to use both data sets (and since the ¹³CO data used for LTE column densities were also made with the HRS), we corrected the MRS data with this additional factor.

2.2 Effelsberg HI observations

We have used the Effelsberg 100-m telescope to obtain a fully-sampled 21 cm H I map of MBM 32. The beam size at 1420 MHz is 9 $.\mkern-4mu^\prime$2 and we observed the cloud at the same 4$^\prime$ grid as in the ¹²CO(1-0) MRS map. The observations were made in frequency switching mode. The velocity resolution of 0.32 km s^-1 was obtained with a 1024 channel autocorrelator. The rms noise in the spectra is 0.25 - 0.40 K, with the lower value for positions which could be observed twice. The spectra were corrected for stray radiation (Kalberla et al. [1982]) and calibrated using the standard position S7 (Williams [1973]).