Astron. Astrophys. 364, 732-740 (2000)

3. Location and morphology

The radio emission from RX J0852.0-4622 is superposed upon a highly structured region of the Vela remnant. Much of this confusing emission is of similar surface brightness to that seen from the new SNR. Furthermore, the very bright, thermal region RCW 38 lies almost adjacent to the southeastern boundary of RX J0852.0-4622. The peak flux of RCW 38 is approximately 150 Jy beam^-1 in the 2.42-GHz data.

The presence of this confusing radio structure, both thermal and non-thermal, meant that RX J0852.0-4622 was not recognised as an SNR from pre-existing radio observations of the region. Prior to the X-ray discovery of RX J0852.0-4622 the non-thermal emission in this region was thought to emanate from the Vela SNR.

3.1. The 2.42-GHz morphology

The filtered 2.42-GHz image presented by Combi et al. (1999) clearly shows the SNR to have a shell-like radio morphology. This is even apparent in unfiltered maps of the region, such as that presented in Fig. 1. Indeed, the emission now known to be associated with RX J0852.0-4622 can be recognised in the radio images presented by Duncan et al. (1995, 1996). Combi et al. (1999) also identify several additional features within their radio image, designated "A" through "D", which they suggest may represent extensions to the radio shell. These will be considered in more detail in Sect. 3.4. It should be noted that - possibly as a result of their filtering procedure - the 2.42-GHz image presented by Combi et al. (1999) does not show either the HII region RCW 38, or the bright, non-thermal emission from Vela-X to the west.

Fig. 1. Unfiltered, Parkes 2.42-GHz survey image of RX J0852.0-4622. Although confused with larger structures, radio emission from the SNR - approximately in diameter in the centre of the image - is clearly visible. To the west, the brighter emission associated with Vela-X is visible. To the southeast lies the bright HII region RCW 38. The angular resolution is , and the rms noise is approximately 17 mJy beam-1. The grey-scale wedge is labelled in units of Jy beam-1. Contour levels are: 5, 10, 20, 50 and 100 Jy beam-1.

Fig. 2 shows a spatially-filtered image of the region surrounding RX J0852.0-4622. This image has been filtered using the "bgf" algorithm (e.g. Sofue & Reich 1979), implemented within the NOD2 software package. A number of filtering resolutions were used, and it was (qualitatively) determined that the emission from RX J0852.0-4622 was optimally enhanced with a filtering resolution of approximately to (in agreement with Combi et al. 1999).

Fig. 2. The optimally-filtered, Parkes 2.42-GHz survey image of RX J0852.0-4622. A filtering beamwidth of has been used, such that larger-scale emission components have been eliminated. The SNR is now more obvious, although it is still heavily confused with emission from the much larger Vela remnant. The angular resolution is , and the rms noise is approximately 17 mJy beam-1. The grey-scale wedge is labelled in units of Jy beam-1. The upper panel shows radio contours, with levels of: 5, 10, 20, 50 and 100 Jy beam-1. The lower panel shows contours of X-ray intensity - from the resolution images of Snowden et al. (1997) - with levels of: 0.30, 0.40, 0.52, 0.65, 0.80, 1.0 and  c ounts s-1 arcmin- 2.

A filtering resolution of was used for the radio data presented in Fig. 2. This figure shows both the emission from RX J0852.0-4622 and the confusing structure more clearly. Comparing Fig. 2 with the unfiltered data presented in Fig. 1, it can be seen that removal of the large-scale structure does not have a major effect on the appearance of the field.

The radio image of the new remnant is dominated by two opposing arcs. Some fainter radio emission is visible on the remnant's western side, although there is no obvious counterpart to the east. The brightest section of the radio shell lies to the northwest, and appears approximately coincident with the brightest region of the X-ray image. Comparing the radio with the X-ray emission (Fig. 2, lower panel), we see that the distributions of both are generally similar, at least in as much as can be discerned from the cluttered radio field.

3.2. The 4.85-GHz PMN data

Fig. 3 shows data from the 4.85-GHz PMN survey from the same region as shown in the previous figures. Although this survey is not optimised for extended sources, the northern and southern sections of the limb-brightened shell stand out clearly.

Fig. 3. An image taken from the 4.85-GHz PMN survey, centred on the remnant RX J0852.0-4622. The angular resolution is , and the rms noise is approximately 8 mJy beam-1. The grey-scale wedge is labelled in units of Jy beam-1. The white contours denote intensities of: 0.2, 0.8, 3.0, 10, 30 and 80 Jy beam-1. The black circle (see Sect. 3.2) is centred on the X-ray coordinates of the source and is in angular diameter.

The black circle near the centre of Fig. 3 fits the outer boundary of the radio emission from both Fig. 2 and Fig. 3 well, and represents what we take to be the outer boundary of the radio emission from RX J0852.0-46221. This boundary is in angular diameter, and is centred on the X-ray centre of the SNR (as given by Aschenbach 1998). Within the uncertainties, this diameter is in agreement with that estimated by Combi et al. (1999), who quote a value of , based upon the (lower resolution) Parkes 2.42-GHz data alone.

Both the radio and X-ray data are consistent with a remnant centred on Galactic longitude , latitude . Thus, we suggest a Galactic designation of G266.2-01.2 for this SNR.

3.3. Confusing structure within the field

As can be seen from Fig. 2, a good deal of additional radio structure is visible in the vicinity of RX J0852.0-4622. Over the remnant itself, most of this structure takes the form of two diffuse "filaments", each of which is wide. These filaments are oriented approximately north-south over the new SNR. Beyond the northern boundary of the remnant, the filaments begin to curve towards the west.

From larger images of the Vela region, such as have been presented by Duncan et al. (1995, 1996), these filaments are known to curve around Vela-X, forming almost a full quadrant of a circle. The eastern arc (as seen in Fig. 2) is highly polarised (e.g. Duncan et al. 1996), and appears to represent the current boundary of the shock from the Vela supernova event.

Interestingly, the confusing filaments from Fig. 2 are almost completely absent from the PMN data. This is because of the observing and data processing procedures used as part of the PMN survey, coupled with the fact that the confusing filaments lie approximately parallel to the scanning direction of the telescope over the region of sky containing RX J0852.0-4622.

3.4. Extensions to the radio shell?

As mentioned in Sect. 3.1, Combi et al. (1999) identify a number of additional features within their radio image. These features were designated "A" through "D" in Fig. 1 of their paper, and apparently extend for relatively large angular distances beyond the edge of the RX J0852.0-4622 shell (up to almost twice the radius of the remnant). Combi et al. (1999) argue that these features may represent extensions to the radio shell of RX J0852.0-4622 (c.f. Aschenbach et al. 1995). It is of interest to consider these in more detail.

Examining the 2.42-GHz radio image of Combi et al. (1999), we find that radio features "A" and "C" appear to be sections of the much more extensive "arc" structures discussed in Sect. 3.3. These arcs can be traced in Fig. 2 for several degrees, up to the northern edge of the figure (i.e., beyond the boundary of RX J0852.0-4622). Larger radio images of the region show that these features continue for many degrees further, in both total-power and polarised intensity. Feature "B" appears to be an isolated, slightly extended source with no obvious connection to the new remnant (even in the unfiltered image presented in Fig. 1). Finally, feature "D" corresponds to the X-ray feature "D/D´" as identified by Aschenbach et al. (1995). Aschenbach (1998) notes that this feature is also a source of hard X-rays, but that this emission is associated with a much lower temperature spectrum than that from RX J0852.0-4622.

We suggest, therefore, that none of the possible "extensions" identified by Combi et al. (1999) are associated with RX J0852.0-4622.

A further argument against an association between these features and the new remnant is that the boundary of the RX J0852.0-4622 shock is well fitted (in both the PMN and Parkes 2.42-GHz survey data) by a circle. This is consistent with the radio morphologies of other young shell SNRs, such as Kepler (Dickel et al. 1988), Tycho (Dickel et al. 1991), and the remnant of SN1006 (Reynolds & Gilmore 1993), although we caution that RX J0852.0-4622 is considerably fainter than these and other young remnants. We also note that the higher resolution PMN image (Fig. 3), although not optimised for extended emission, shows no evidence for any connections between the features noted by Combi et al. (1999) and the shell of the new remnant.

3.5. The quasi-central source

The PMN image shows that a point-like (i.e. unresolved at a resolution of ) source lies approximately east of the apparent centre of the remnant. This source is not coincident with either of the two compact X-ray sources near the centre of the remnant that are discussed by Aschenbach et al. (1999). The PMN survey source catalogue (Wright et al. 1994) lists this source as PMN J0853-4620, with a 4.85-GHz flux of  mJy. Being relatively faint, our ability to detect PMN J0853-4620 in the 2.42-GHz data is compromised somewhat by confusion, coupled with beam dilution. Nevertheless, we can establish the 2.42-GHz flux to be  mJy. This leads to a spectral index for this source of (with ).

The radio spectral indices of pulsar emissions are generally much steeper than the estimated above for the source (e.g. Taylor et al. 1993). Furthermore, a flux of  mJy at a frequency of 4.85 GHz would be exceptionally high for a pulsar. It is much more likely, then, that the PMN J0853-4620 source is extragalactic in origin, rather than associated with RX J0852.0-4622.

3.6. Radio spectrum

The presence of the confusing structure noted in Sect. 3.3 makes accurate estimates of the integrated remnant flux density difficult. The values for the flux density of the SNR given below were estimated by integrating the emission within the boundary of RX J0852.0-4622, as defined by the circle seen in Fig. 3 (lower panel). The integrated area extended approximately one beamwidth beyond this circle, in order to include all the flux from the shell. The base level was determined from flux minima near the centre of the remnant, as well as beyond the eastern and southwestern edges of the shell. Fluxes contributed by the confusing "filaments" seen to the western and eastern sides of the remnant (as discussed in Sect. 3.3) were estimated and subtracted from the total, integrated flux. Note that the uncertainties in the integrated flux values are dominated by baselevel uncertainty, rather than by uncertainties in the flux estimates of the confusing structure.

We estimate the integrated fluxes of RX J0852.0-4622 at 2.42 and 1.40 GHz to be  Jy and  Jy, respectively. These values lead to a very uncertain estimate of the spectral index, with (). To better establish the spectral index of the remnant, the method of "T-T" plots was used (e.g. Turtle et al. 1962).

Estimates of the remnant spectral index were made from both filtered and unfiltered images, using the T-T plot technique. The northern section of the shell was found to have a consistent, non-thermal index of . The southern section of emission exhibited a much flatter spectrum of . We believe this latter value to be unreliable, due to the proximity of the southern section of the shell to the bright HII region RCW 38 and its associated emission. At the lower angular resolution of the 1.40-GHz data (to which the 2.42-GHz data are also smoothed, for the purposes of spectral index calculation), some of this thermal emission becomes confused with the southern arc of RX J0852.0-4622. We suggest that the value determined for the northern shell section better represents the new remnant's radio emission.

Extrapolating the measured integrated flux to a frequency of 1 GHz (using a spectral index of ), we determine a value of  Jy, leading to an average surface brightness at this latter frequency of  W m^{- 2} Hz^-1 sr^-1. The above values are summarised in Table 2.

Table 2. Characteristics of the radio emission from RX J0852.0-4622, determined from the Parkes data. The spectral index measured for the southern limb of the SNR appears to be confused with nearby thermal emission. The surface brightness is given in units of W m^{- 2} Hz^-1 sr^-1.

© European Southern Observatory (ESO) 2000

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