2 Observations and data reduction

2.1 Source selection

To study spatial distribution and physical properties of YSO's in relatively close-by molecular clouds, we have selected from our catalogue of observational data (see Papers I and II), based on the IRAS Point Source Catalogue (PSC), the sources obeying the following criteria:

i)

having passed through all steps described in Papers I and II and being eventually identified as Class I protostellar candidates;

ii)

belonging to the VMR-D cloud (as designated by Murphy & May 1991), which is located at a distance of 700 $\pm$ 200 pc from the Sun (Paper I).

Both criteria essentially provide a flux limited sample of 12 IRAS objects with red colours [1 Jy < F₁₂ < F₂₅ < F₆₀, where $F_{i}=F_{\nu}(i)$ in the IRAS band centred at i $\mu$m] and associated to molecular gas with a low velocity (0 < $v_{\rm LSR}$ < 14 km s^-1) compatible with that of the D cloud itself (Murphy & May 1991).

In Table 1 the selected objects are listed along with their equatorial coordinates and the gas velocities detected by means of pointed spectral line observations (CO and/or CS) performed at high spatial resolution (Papers I/II). The identification given in the last column (# IRS) refers to our internal classification and may be useful in order to have a quick reference to the notes on individual sources given elsewhere (Papers I/II). In the Appendix we provide some complementary informations about the sources investigated here. The spatial distribution of the selected sources is depicted in Fig. 1, superimposed on the ¹²CO(1-0) integrated intensity contour plot (in galactic coordinates) provided by Murphy & May (1991) in their Fig. 5. The selected IRAS sources tend to cluster near the peaks of the molecular emission, namely their original birthplaces, thus giving support to their YSO's nature. Marginal exceptions to this occurrence are represented by IRS 67 and IRS 71 which lie in locations where the integrated intensity is lower; nevertheless we decided to include them for the sake of completeness.

  

Table 1: Selected IRAS sources

  

Figure 1: Spatial distribution of the IRAS Class I sources belonging to VMR-D cloud, superimposed on the 12CO(1-0) integrated intensity map given by Murphy & May (1991). Contour intervals are 1.3 K km s-1 from 1.3 K km s-1; sources are named according to the internal classification used in Table 1

2.2 Near-IR imaging

The imaging data were obtained in February 1993 with the IRAC2 near-infrared camera (Moorwood et al. 1992) on the ESO/MPI 2.2 m telescope at La Silla (Chile). The observations were carried out through standard J (1.25 $\mu$m), H (1.65 $\mu$m) and K (2.20 $\mu$m) broad band filters. The $256~\times~256$ pixels NICMOS3 array was used at a plate scale of 0.49 arcsec/pixel, resulting in a field of view of about $2~\times~2$ ${\rm arcmin}^2$on the sky. For each source and for each filter we obtained a set of 3 dithered frames offset from each other by $30 \hbox{$^{\prime\prime}$}$ in declination, resulting in total on-source integration times of 120 s, 180 s and 540 s at J, H and K, respectively. The data were reduced through standard IRAF routines and photometry was performed using the DAOPHOT package, in IRAF. Unfortunately, PSF fitting procedures did not work well with our data, so we carried out standard aperture photometry. More details are given in the Appendix.

The K brightness of extended sources was obtained by integrating the detected emission down to a $\sim\!5-10\sigma$ level with respect to the underlying background; J and H brightnesses were derived integrating over the same area. Instrumental magnitudes were converted to absolute values by comparison with ESO photometric standard stars (Bouchet et al. 1989) which also allowed us to check that the zero-points were exceptionally stable during the whole nights (rms variations were $\sim\!0.01~\div~0.03$ mag in J, H and K bands): the limiting magnitudes of this survey are $J \approx$ 18.5, $H \approx$ 18.0 and $K \approx$ 17.0. Air mass corrections were applied using typical values for the atmospheric extinction coefficients at La Silla (E_J= 0.125. E_H=0.080 and E_K=0.110 mag/airmass).

A total of $\approx 1300$ sources have been detected in K, $\approx 1200$ in H and $\approx 800$ in J; coordinates and magnitudes of all sources found in the K band are given in Table 4, while the obtained JHK images are displayed in Fig. 2. For sources undetected in some of the bands, upper limits were estimated in each frame examining the magnitude-error diagrams.

The completeness limit in the K band is not in principle the same for all frames because of the presence of local nebulosities; however, we obtained the K luminosity function for each field, which generally shows a turn-off at 15.5-16.5 mag (see our forthcoming paper). Thus, we conservatively define as completeness limit the value $K_{\rm compl}=15.5$ mag (i.e. more than one magnitude below the limiting magnitude), which corresponds to complete observations at a 10$\sigma$ detection level.

Some complementary images in the L' band ($3.8~\mu$m) were obtained with both the first IRAC $64~\times~64$ pixels Hg:Cd:Te array at the ESO 2.2 m telescope, and the $58~\times~62$ pixels InSb array at CTIO during the period 1991-1992. Despite these results are not presented here because of their poor quality, nevertheless we will refer to them in the following, since they help us in the identification of the IRAS counterparts: in fact the sources with an intrinsic IR excess definitely emerge in the L' band, often remaining the only bright object within the frame area.

  

Figure 2: J, H and K images of all observed fields. Each column refers to the same band (indicated above) and each row refers to the same field (indicated on the left-hand side). North is up and east to the left

 

Figure 2: continued

 

Figure 2: continued

2.3 NIR positions  

Absolute $\alpha$ and $\delta$ positions were derived in the observed fields using reference stars whose equatorial coordinates are given in the HST Guide Star Catalogue (GSC). Each K frame was searched for IR counterparts of GSC stars and the astrometric solution was derived. If a frame contained too few (or none) GSC stars, we first determined the astrometric solution for a Digitized Sky Survey (DSS) plate larger than but including the K frame field. Then IR counterparts of DSS stars were used to derive the astrometric solution for the K frame itself. We estimate that the positions given in this paper are accurate within $1 \hbox{$^{\prime\prime}$}$ both in RA and DEC.

2.4 Millimeter continuum observations

The observations were carried out during September 1992 with the SEST 15 m telescope at La Silla. The antenna fed a ³He-cooled bolometer of the MPIfR (Kreysa 1990). The filter set coupled to the atmospheric transmission window provides an effective wavelength around 1.25 mm. The beam size is 24$\hbox{$^{\prime\prime}$}$ (HPBW) and the chop throw is 70$\hbox{$^{\prime\prime}$}$.

The sources were located by pointing on a nearby radio quasar with strong millimetric fluxes; the pointing accuracy was most of the time better than 2$\hbox{$^{\prime\prime}$}$ and was checked each half an hour.

We observed 11 out of the 12 IRAS sources using integration times $n\times200$ s, with n depending on their expected 1.3 mm intensity, which had been approximately found by extrapolating the IRAS 100 $\mu$m flux. The atmospheric transmission was monitored by frequent sky dips. Uranus, Mars and Saturn were used as primary calibrators and several quasars as secondary calibrators, mainly to detect sky variations during the observations.

The 1.3 mm flux values were corrected for the overall system and atmospheric response through the gain-elevation curve, assuming that the spectra in the millimetric region can be approximated by $F = F_{0}(\frac{\nu}{\nu_0})^\alpha$, where $\alpha$ is the spectral index of the source. The results of the 1.3 mm photometry are listed in Col. 10 of Table 1 along with their uncertainties.