2. The radio maps
In Fig. 1 the NVSS-map of G16.2-2.7 at 1.4 GHz is shown. Here the maps are plotted with the "Skyview" package (Ebert et al. 1998). The circular shell structure of the SNR has the angular diameter ´, while the integral flux density in the map, Jy, is much lower than the value extrapolated from the spectrum (see below). The large uncertainty of the NVSS flux depends strongly on the background level definition and could only be a lower limit because the NVSS suffers from lack of zero-spacing data and its images are insensitive to smooth radio structures much larger than several arcmin.
The flux weighted centroid of the source has the Galactic coordinates: and or equatorial ones: RADEC1950 = , -16o09´50".
Recently Gaensler (1998) has investigated the nature of the the bilateral SNRs, one of them, G03.8-0.3, has a bilateral structure very similar with G16.2-2.7. Their surface brightness, angular sizes are close. We could estimate the value of defined to be the acute angle between the symmetry axis of the SNR and the Galactic plane. We fit the symmetry axis using only the bright circular arcs. This gives a value . Thus the symmetry axis is aligned close to the Galactic plane (see Fig. 1). It is not clear whether the central weak filament with a brightness of nearly 1.5 mJy/beam () located close to the symmetry axis is real and associated with this SNR or not. The new radio mapping are needed.
The sources in the field of this map from the NVSS source catalog are marked by white crosses and the numbers around them annotate the fractional polarization in per cent. In Table 1 a list of the NVSS sources with detectable linear polarization is given. This table is obtained by the select program in the astrophysical catalogs data base CATS (Verkhodanov el al. 1997). We have reduced it to six columns: number, coordinates (B1950), flux density, polarized intensity or power of polarization in per cent; the last column indicates whether the source is a part of the SNR or not. It is remarkable that details (included in the NVSS source catalog) of bright western and eastern arcs of the shell are highly polarized (6-20% from the last column in Table 1). Probably two relatively bright sources (14,15) are background extragalactic ones. The weak filament-like weak source (11) in the center is highly polarized (p=43%) and it might be a detail of the radio shell with a highly ordered magnetic field. The NVSS data show that the bright eastern arc of G03.8-0.3 is also polarized to 10% at 1.4 GHz.
Table 1. NVSS sources around of the SNR G16.2-2.7
The source G16.2-2.7 is visible in the Effelsberg surveys maps at 1.4 and 2.7 GHz (Reich et al. 1990a, Reich et al. 1990b, here RRF and RFRR) as a extended source, which has not been included in the catalogs of these surveys because its apparent size is over 16´. We cut the maps from the original maps with a "postage stamps" procedure at Max-Planck-Institut fuer Radioastronomie (MPIfR) web-site, which allowed us to cut small images from a single survey map. In Fig. 2 the ISSA (IRAS) grey scale and white contour map at 60 µm (Beichman et al. 1988) is superposed on the black contour plot of the 2.7 GHz intensity map. There is no clear relation between radio and infrared radiation. Nevertheless, we have estimated the infrared flux to be approximately 250 and 700 Jy at 60 and 100 µm respectively within the region of the radio shell. These fluxes are higher than times the radio ones.
We used the radial profiles from the NVSS map and the 2.7GHz map to fit the spatial parameters of the shell SNR. Model profiles of a optically thin synchrotron hollow circular shell with an outer diameter D, width R and a random magnetic field was discussed by Rosenberg (1970) for Cas A. The best fit with the real radial profiles at 1.4 and 2.7 GHz gives D= and R=. This initial radial profile was convolved with the -beam of the Effelsberg survey at 2.7 GHz. In Fig. 2 we compare this smoothed model profile with the mean one from map at 2.7 GHz. We see that these profiles are very similar within uncertainties of the background level and smoothing effects.
The low value of the flux at 1.4 GHz obtained from the NVSS image could be explained with the above model. If a background with a filtering window of is subtracted from the image with such a radial profile, then the total flux reduced to only 35% of the initial value becomes visible mainly in the bright limb. Therefore, probably the real flux of the source is about 2 Jy, which roughly coincides with the obtained spectrum. Thus SNR G16.2-2.7 is likely a hollow, almost spherical shell of optically thin smoothly distributed radio emission.
Unfortunately there are not any extended or filamental details in the Digital Sky Survey II image of the SNR area that could be recognized as a optical counterpart.
© European Southern Observatory (ESO) 1999
Online publication: December 2, 1999