Online material

	Fig. 2 Unresolved sources. The images are scaled logarithmically. North is to the top of the page and east is to the left. The contours typically represent 1, 5, 25 and 75 percent of the peak flux.
Open with DEXTER

Appendix A: Notes on individual objects

Here we discuss the images and modelling results for each resolved MYSO.

A: G263.7759−00.4281

The 20 μm morphology of G263.775986−0100.4281 (IRAS 08448-4343) is reminiscent of the cavities of a bipolar outflow seen close to edge on. The direction of the elongation seen in the MIR image is consistent with that seen in the NIR via 2MASS images. The radio luminosity of the object is consistent with a jet rather than a Hii region (Hoare et al. 2007). This supports the notion that the bulk of the MIR emission arises in the walls of cavities evacuated by an outflow. Furthermore, the extension in the NW/SE direction is aligned with the H₂ jet detected by Giannini et al. (2005), confirming that this object drives an outflow and the MIR morphology is associated with outflow activity.

The SED of this object can be reproduced with the selected model with an inclination of i = 87, which is consistent with the notion that this object is seen close to edge on. The intensity profile of the model is not entirely consistent with the data. However, the axis-symmetric model used cannot account for the slight asymmetry of the image. Indeed, the model can approximately recreate the extension of the southern part of the observed bipolar structure but not the northern half (see Fig. 7, panel A). The model cannot reproduce asymmetry at these wavelengths in an edge on configuration. Therefore, we surmise that the model reproduces the data as well as can be expected.

B: G265.1438+01.4548

This object exhibits a fairly symmetric morphology in the MIR. If the generic model featuring outflows is correct, this would suggest that this object is seen at a relatively low inclination. Modelling of this object’s CO bandhead emission is consistent with this scenario (Bik & Thi 2004; Wheelwright et al. 2010). The SED and intensity profile are relatively well reproduced by the generic model with an inclination of i = 32, which is consistent with the hypothesis that this object is viewed at a low inclination. The model slightly under-predicts the flux at distances greater than  ~2′′   from the centre of the intensity distribution. However, examination of the image reveals that this may not be related to the source as the core of the emission is concentrated within 2′′. Therefore, this slight discrepancy is neglected.

C: G268.3957−00.4842

G268.3957−00.4842 exhibits a relatively symmetric morphology in the MIR. As before, it is suggested that this is the result of a relatively low inclination. Indeed, the object’s SED and intensity profile are reproduced by a model with an inclination of i = 30, consistent with the hypothesis that this object is observed at a low inclination.

D: G269.1586−01.1383

The 20 μm image of G269.1586−01.1383 incorporates an additional, localised source of flux approximately 7′′   to the NE. Both sources exhibit an extended, slightly cometary morphology reminiscent of a Hii region. However, radio observations indicate that while the northern source is a Hii region (Urquhart et al. 2007), the southern source is likely to be a YSO. This is confirmed via low resolution NIR spectra. The spectrum of the southern source exhibits H₂ line emission, suggesting that it drives an outflow.

The complex appearance of this object is difficult to recreate with the 2D, axis-symmetric codes of Whitney et al. In principle, a bipolar cometary morphology could be recreated by warm dust emission from the walls of cavities carved by a bipolar outflow with a large opening angle. However, the image contains no hint of a bipolar structure. Therefore, it might be expected that the model cannot reproduce the intensity profile of this source, as is found to be the case. We note that this object exhibits a notably different morphology to the rest of the sample, which may indicate it represents a different phase of massive star formation than the other MYSOs. This is partially substantiated by the fact that this object is associated with a Hii region.

E: G310.0135+00.3892

G310.0135+00.3892 (IRAS 13481-6124) was observed with the VLTI and AMBER by Kraus et al. (2010). These authors reconstructed images of this object in the K-band and detected a disc seen under moderate inclination. In addition, Kraus et al. (2010) discovered signs of an outflow orientated in the NE/SW direction, perpendicularly to the disc. The 20 μm morphology of this object is slightly extended in the NE/SW direction, i.e. along the direction of the outflow. This is consistent with the notion that the majority of the MIR emission traces warm dust in the walls of outflow cavities.

The observed SED and intensity profile are reproduced relatively well by a model with an inclination of i = 32, close to the intermediate inclination derived by Kraus et al. (2010). The model under-predicts the flux at distances greater than  ~1.5′′   from the peak of the intensity distribution. However, examination of the image reveals that at this distance the morphology is not symmetric. The axis-symmetric model cannot reproduce this morphology. Therefore, it is surmised that the model reproduces the data as well as possible and this slight discrepancy is neglected.

F: G318.0489+00.0854

The 20 μm image of G318.0489+00.0854 exhibits a slightly cometary morphology. This object is associated with two Hii regions detected via radio observations, both within several arcseconds (Urquhart et al. 2007). The near/far ambiguity over the distance to this source has yet to be resolved. As a result, it is impossible to unambiguously model its intensity profile and SED. Therefore, we do not attempt to reproduce the observations of this object.

G: G332.2941+02.2799

G332.2941+02.2799 (IRAS 16019-4903) exhibits a significantly extended morphology in the NE/SW direction. This is also seen in the NIR in 2MASS images and at 10 μm (see Mottram et al. 2007). This object is known to be associated with several bipolar outflows and the direction of the most prominent outflow is aligned with the extension seen in the infrared images (Henning et al. 2000). As a result, the observed morphology is consistent with the notion that the MIR flux largely arises from warm dust in outflow cavity walls.

Since these observations were taken, it has been determined that the luminosity of G332.2941+02.2799 is 844 L_⊙ (see Mottram et al. 2011). This places the spectral type of this object at approximately B2/B3. Therefore, this object is an intermediate mass YSO rather than a massive YSO and should be excluded from the sample. Nonetheless, we attempt to recreate the observations with the generic model developed for the sample of MYSOs. The model cannot fully recreate the extension of the intensity profile observed, even with a high inclination and stellar luminosity. This may imply that the model developed for MYSOs is not directly applicable to intermediate mass YSOs. Alternatively, the strong outflow may cause shock heating in the cavities, which is not included in the model.

H: G332.986886−0100.4871

G332.986886−0100.4871 exhibits a fairly symmetric morphology. As before, it is suggested that this is the result of a relatively low inclination. The object’s SED is well reproduced by a model with an inclination of i = 15°, as is the object’s intensity profile. The model intensity profile deviates from that observed at a distance of  ~1.75′′ from the centre. However, this is where the object’s flux distribution falls to the level of the background and thus this discrepancy is not considered significant.

I: G339.622186−0100.1209

This object exhibits a slightly cometary morphology. As discussed in the case of G269.1586−01.1383, this is difficult to recreate at 20 μm with an axis-symmetric code. We note that this object is the most luminous in the sample with L = 5.2 × 10⁵   L_⊙. Mottram et al. (2011) find that there is not a significant population of radio quiet MYSOs with such a luminosity. Sources with luminosities greater than L ~ 10⁵   L_⊙ are almost inevitably associated with a Hii region. Therefore, it might be expected that this object is associated with a Hii region. While this object is a radio non-detection (Urquhart et al. 2007), it is possible that a weak Hii region has escaped detection due to the large distance to this object (13 kpc). This could explain the cometary morphology of this source.

A relatively inclined model (i = 60°) reproduces the object’s SED and intensity profile relatively well. The mm flux is under predicted, but at the large distance to this object, several sources could have been included in the mm beam, resulting in an over-estimation of the flux. We note that, although the model recreates the observed intensity profile, the model image exhibits bi-polar structure while the observed morphology is appears more mono-polar. As discussed above, this source may harbour a Hii region, which could explain this discrepancy.

J: G343.502486−0100.0145

The image of G343.502486−0100.0145 exhibits a complex, notably extended morphology. Extended nebulosity towards the SE is also visible in 2MASS and Glimpse IRAC images. The extension of the 20 μm image is broadly consistent with that seen at

shorter wavelengths, although these data exhibit a more complex morphology. This source was detected at 4.8 GHz (Urquhart et al. 2007), and is thus classified as a Hii in the RMS catalogue. Therefore, it might be expected that the model developed to recreate the observations of radio quiet MYSOs cannot reproduce the complex morphology exhibited by this object. Indeed, the model that recreates the majority of the object’s SED fails to match the spatial extension of the 20 μm flux. We suggest that this is likely due to some of the observed flux originating in dust surrounding a Hii region, rather than the central source.

K: G343.521386−0100.5171

The image of G343.521386−0100.5171 reveals three localised sources. The central source clearly dominates the flux. However, one of the secondary sources is within 2′′ of the primary source. It is apparent that secondary flux will contaminate the intensity profile of the primary. Indeed, the image reveals that this source is less centrally concentrated than its isolated counterparts. A limited range of angles was used in constructing the azimuthally averaged intensity profile of this object in an attempt to limit contamination, but it is not clear how effective this was. The model fails to simultaneously reproduce the broad central maximum of the image and the observed SED. We suggest that this is the result of contamination of the intensity profile by flux from the neighbouring secondary source.

L: G345.0061+01.7944

G345.0061+01.7944 exhibits an extended morphology similar to that of G332.2941+02.2799. This suggests that the MIR image traces outflow structure seen close to edge-on. Indeed, the NIR spectrum of this object exhibits H₂ emission indicative of outflow activity. Unlike the case of G332.2941+02.2799, a satisfactory fit to both the SED and the intensity profile was found. This is consistent with the hypothesis that the difficulties in reproducing the observations of G332.2941+02.2799 were the result of applying a model developed for high luminosity MYSOs to the case of an intermediate mass YSO.

M: G349.7215+00.1203

G349.7215+00.1203 is the second most luminous object in the sample with L = 3.1 × 10⁵   L_⊙. The image of this object reveals a cometary shaped source of flux to the NW of the primary object. Radio observations confirm that the additional source of flux is a Hii region (Urquhart et al. 2007). The model reproduces most of the features of the SED and intensity profile of this object. The model under-predicts the flux at  ~100 μm. However, this could be due to the measured flux including a contribution from the Hii region. The model intensity profile exhibits a slightly different slope than that observed, although it is essentially within the uncertainty. We suggest that this could also be due to the presence of the Hii region.

© ESO, 2012