Astron. Astrophys. 364, 683-688 (2000)

3. Results and discussion

3.1. Star formation in the cloud

Fig. 1 and Fig. 2 show the J-H/H-Ks color-color diagram and the Ks/H-Ks color-magnitude diagram of the sources from 2MASS data. In Fig. 1 the solid curve is the locus of points corresponding to unreddened main sequence stars (Koornneef 1983). The two dashed lines are parallel to the reddening vector with magnitude of Av=30. They form the reddening band (drawn from the base and tip of the unreddened main sequence) and bound the region in which stars with normal photosphere fall (also see Hunter et al. 1995). Similarly, in Fig. 2, the solid line represents the locus of unreddened main sequence and the dashed arrow shows the direction and magnitude of the reddening vector. We did a search in the SIMBAD database to find out any known sources around IRS1 with a search radius of 4.5´. Apart from the IRAS source and its H₂O maser source, a star of A0V type (HD37317, 7.79 mag in Ks) was found in the SIMBAD search. This star's co-ordinates are 05h 39m 19.9s +35^o 38´ 30", that is approximately 2.8´ south-west from the IRAS source. We found from the 2MASS point source catalogue the colors of the star to be H-Ks=0.05 mag and J-H=0.031 mag. The colors of IRS1 are H-Ks=0.93 mag and J-H=1.45 mag.

Fig. 1. Color-color diagram of the sources extracted from the 2MASS data. The filled circles represent Class I type sources. The solid line represents unreddened main-sequence stars and the dashed lines are parallel to the reddening vector with magnitude of Av=30. Also shown are the positions of IRS1, star # 12 (from Table 1), and HD37317

Fig. 2. Color-magnitude diagram. The symbols have same meaning as in Fig. 1.

Stars in Fig. 1 can be divided into three main groups a) those lying on the left side of the reddening band, b) those lying in the reddening band and c) those lying on the right side of the reddening band. The stars in the first group (a) plotted as open circles can be further divided into two sub-groups: one, those having J-H color less than 2 and second, those with J-H color more than 2. The former subgroup are mostly foreground stars as supported by their low values of Av (less than 5 mag), and the stars in the latter subgroup could be spurious detections since they have J and H magnitudes fainter than 17 and 16 mag respectively. The second group (b) mostly consist of normal stars with low values of Av (10) and background stars with high values of Av (10) and are also plotted as open circles. The third group (c) contains stars showing excess emission in H and Ks. Such sources are mostly YSOs (Lada & Adams 1992; Lada et al. 1993; Gomez et al. 1994). YSOs can be further divided into Class I, Class II and Class III type sources based upon their Spectral Energy Distributions (SEDs) (Strom et al. 1989; Kenyon et al. 1993; Hartmann 1998). We have plotted sources redder than IRS1 and falling on the right hand side of the reddening band as `filled circles' and are considered to be Class I type or protostars (since IRS1 is a known protostar; also see Lada & Adams 1992). Sources which show low H-Ks color (1.5) and J-H color (1.0) and also lying on the right hand side of the strip of reddening lines are also plotted as `open circles'. All these sources are faint in the Ks band (14 to 15.2 mag). Even though most of them are within the limit of completeness, we will need deep K band images to verify the existence of the sources. Their Av values range from 13 to 23 mag, suggesting that they could be Class II type sources (Kenyon et al. 1993).

It is clear from the J-H/H-Ks color-color diagram that the region around IRS1 is undergoing a phase of star formation. Three out of five Class I sources detected (excluding IRS1) are found within one arcmin radius of IRS1. Two of these sources are brighter than 14.2 mag in K' and are detected in the Mt. Abu images. A comparison of magnitudes of these two sources #9 and #12 (see Table 1) show that one of them #(12) has varied over the time of observation between 2MASS (February 1998, as given in the header of the FITS images) and Mt. Abu (January 2000). The star has become fainter by 1.1 mag in K' and H bands in about 23 months. In the Mt. Abu J band image the star is not detected. Since the difference in magnitude is much larger than the photometric errors and the decrease in the brightness is consistent in K' and H bands, we believe that this is real. The star also shows extreme reddening in the J-H/H-Ks color diagram, a typical characteristic of Class I sources and presence of circumstellar material. Variability in low-mass protostars is known (Hartmann 1998). The total luminosity depends on the mass accretion rate, and low mass protostars like FU Orionis type of objects have shown variability of several magnitudes in the optical wavelength (Bell et al. 1995, and references therein). It is possible that we have witnessed a FU Orionis kind of behavior from the star. However, we need further NIR observations to verify this.

3.2. Photometry and spectral energy distribution of IRAS 05361+3539

IRAS 05361+3539 (IRS1) is a massive luminous YSO and is associated with an ultra-compact HII region (Shepherd & Churchwell 1996). The low-resolution NVSS contour plot shows emission towards IRS1 at mJy level (4) (Condon et al. 1998). The FIR IRAS fluxes of IRS1 at 100µm, 60µm, 25µm, and 12µm are 1310Jy (upper limit), 29.15Jy, 6.72Jy, and 1.18Jy respectively. We estimated the spectral index of IRS1 to be 1.1 between 60µm and 1.12µm (Jband). The source therefore belongs to the group of Class I type sources (Strom et al. 1989). The SED is shown as a plot (crosses) of Log/Log in Fig. 3 where is the wavelength in µm and F is the flux in Watts/meter². We determined the temperature distribution of the circumstellar matter by fitting a model (Anandarao et al. 1993) to the observed SED. We have assumed photospheric temperature of 20,000K, stellar radius equal to seven solar radii, and plane parallel geometry for the dust shells. The dashed line in Fig. 3 shows the model. The derived dust temperatures are 800K at 4AU and 80K at 400AU from the central source. The uncertainties in the dust parameters could be as large as 10-20% due mainly to the inherent non-uniqueness of the model and to some extent to the uncertainties in the assumed stellar parameters. These results confirm that IRS1 is a Class I type source. The range of dust shell parameters derived from the model seem to support the accretion disk scenario (Hartmann 1998; Adams et al. 1987) in which case, the grain heating is due to two processes: one due to the reprocessing of UV photons and the other due to viscous heating in the disk (Hillenbrand et al. 1992). Following Hillenbrand et al. (1992) we estimate the minimum radius of an accretion disk attributing the entire 25µm flux to the accretion heating of the grain (i.e. T=140K). We derive radius of the disk to be 160 AU if the inclination is 0^o (face on).

Fig. 3. SED of IRS1. Crosses represent observed fluxes and dashed line model values. The flux F is in Watts/meter2 and wavelength in µm. See text for details.

3.3. Morphology of IRAS 05361+3539 and Detection of a NIR Jet

Fig. 4 shows K', H₂ (emission line + continuum) and Br (emission line + continuum) images of IRS1. Fig. 5 gives the contour map of H₂ superposed on the K' image. The plate scale in Fig. 4 is 1"/pixel for the K' image and 0.5"/pixel for the narrow band images. The NIR images reveal (Fig. 4, Fig. 5) that the IRS1 source is associated with a nebulosity extended up to 5" in the northern direction and a filamentary structure of length 6" in the east (seen more prominently in the H₂+continuum image).

Fig. 4. K' (top), molecular hydrogen line (2.12µm line + continuum) (middle) and Br (emission line + continuum) (bottom) images of IRS1. The stars are identified with numbers as given in Table 1 and the arrows indicate the extended nebulosity. The x and y axes are RA and DEC (J2000) respectively.

Fig. 5. Contours of the molecular hydrogen (line + continuum) image on the K' image, the contours are from 4 to 9 levels. The x and y axes are same as in Fig. 4.

The eastern filamentary structure bends beyond 7" from IRS1 and continues another 20" in the southeast direction. Since we do not have a continuum free H₂ emission line image we cannot quantify the amount of H₂ emission from the region. However, we qualitatively argue from the brightness in the K' and the H₂+continuum image that at least 40% of the total brightness in the narrow band image is due to pure H₂ emission. Further, the filamentary structure matches with the eastward jet found in the low resolution CO map of Shepherd & Churchwell (1996). Therefore, it is likely that the H₂ emission is tracing the jet from the YSO. The bending of the jet could be because of physical obstruction of its flow due to the presence of putative dense matter. We do not see significant structure in the Br + continuum image. If one assumes a particle density of 10⁵ /cm³, then there cannot be sufficient flux of energetic UV photons available from a B2 star to ionize the matter except in optically thin regions.

The structure in the northern direction is six times brighter than the jet. There are three distinct possibilities regarding the nature of this structure. This could be an unresolved star closely associated with the IRS1 source. However, in the 2MASS point source catalogue there is no entry corresponding to that position. We will need high-resolution images in K band to prove that the northern structure is an unresolved star.

A second alternative is that the extended nebulosity is due to the presence of the ultra-compact HII region. From the near-infrared morphology, it appears that this could be a cometary UCHII region (Wood & Churchwell 1989). Supposing that the central star has a relative motion with respect to the parent molecular cloud, say a few km/s, such a supersonic motion can create a region of low density behind the star (Hughes & Viner 1976; Weaver et al. 1977). If we consider a dynamical time of 5.310⁴ years (Shepherd & Churchwell 1996) and a speed of 1km/s to 2km/s (Jones & Walker 1988) then, in as many years the star would have traveled a distance of 0.1 pc (upper limit). The extent of the extended nebulosity in the H₂ + continuum image corresponds to about 0.04 pc (5", see Fig. 4). The density behind the central star should be much less (10³ particles/cm³) for the UV photons to reach up to 0.04 pc.

NIR spectroscopy of UCHII regions have shown that they are bright in Br emission line (Armand et al. 1996; Doherty et al. 1994). for the UCHII region can be calculated from Av using the relation /Av=0.125 (for R_v=5.0 extinction law of Cardelli et al. 1989), and it turns out to be 1.87 mag. The density of matter can also be calculated from the relation between Av and column density, Av/N_H=1 10^-21 mag/cm² (Cardelli et al. 1989). We determined the density of matter within 1 pc of the source assuming that most of the extinction is from within 1pc of the source. This gives a lower limit of density of 5x10³particles/cm³. The upper limit can be 10⁵ particles/cm³, if we consider that most of the extinction is within 0.05 pc. The value of 0.05 pc has been assumed based on the fact that very close to the star up to 5000AU the density can be as high as 10⁸ particles/cm³ and up to 0.5 pc it can be 10⁵ particles/cm³ (Churchwell 1997). A sizable Stromgren sphere with the lower limit of particle density can be obtained for a central B2 type star. But, we do not see any significant brightness in the Br + continuum image compared to the K' band image except at the core of the nebulosity. A deeper Br image will be required for dertermining the extent of the nebulosity.

As a third possibility, the extended nebulosity can be considered as a dense clump of molecular hydrogen in the parent molecular cloud. We found that the region is bright in the H₂+continuum image and only qualitatively we can say that pure H₂ emission is present. H₂ emission can arise due to either UV fluorescence, or collisionally excited by the impact of stellar winds from IRS1 (Genzel 1992).

From the above arguments, it appears that it is difficult to determine from the present data the nature of the extended nebulosity. High resolution radio continuum images and medium resolution NIR spectroscopy are necessary to resolve this issue.

© European Southern Observatory (ESO) 2000

Online publication: January 29, 2001
helpdesk.link@springer.de