Free Access
Issue
A&A
Volume 529, May 2011
Article Number A159
Number of page(s) 15
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201016311
Published online 22 April 2011

© ESO, 2011

1. Introduction

A supernova explosion releases energy of about 1051 erg into the interstellar space. Observations of supernova remnants (SNRs) are important not only to the understanding of the objects themselves but also to the properties of the ambient interstellar medium. Many intense and large SNRs were identified in the first radio surveys at low frequencies (e.g. Brown & Hazard 1953; Westerhout 1958). Sensitive and high angular resolution Galactic plane surveys subsequently revealed a large number of faint or compact SNRs (e.g. Reich et al. 1988; Brogan et al. 2006). To date, 274 Galactic SNRs have been catalogued by Green (2009). Observations of SNRs at many wavelengths are needed to derive reliable spectra. Polarization observations of SNRs are useful to reveal the magnetic field properties of SNRs.

The Sino-German λ6 cm polarization survey of the Galactic plane has been carried out with the Urumqi 25-m radio telescope since 2004. It covers 10° ≤  ≤ 230° in Galactic longitude and |b| ≤ 5° in Galactic latitude (Sun et al. 2007; Gao et al. 2010; Sun et al. 2011; Xiao et al. 2011). The Urumqi λ6 cm polarization system is suitable for observations of large and faint SNRs. The surface brightness limit for the survey is Σ1   GHz = 3.9 × 10-23   W   m-2   Hz-1   sr-1, below the faintest SNRs known today. Using this system, we have studied the magnetic fields and the spectral index distribution of SNR G74.0−8.5 (Cygnus Loop, Sun et al. 2006), G126.2+1.6 and G127.1+0.5 (Sun et al. 2007), G156.2+5.7 (Xu et al. 2007), G180.0−1.7 (S147, Xiao et al. 2008), G130.7+3.1 (HB3, Shi et al. 2008), and G65.3+5.7 (Xiao et al. 2009). In addition, Foster et al. (2006) disapproved G166.2+2.5 (OA 184) as being a SNR.

In this paper, we analyze and discuss 16 large SNRs in the survey region of 10° ≤  ≤ 230° with an apparent size exceeding 1° in our map. Smaller SNRs will be discussed elsewhere (Sun et al., in prep.). We will briefly describe the survey and the SNR sample in Sect. 2. In Sect. 3, we present their total intensity and polarization images, and discuss the properties of each SNR in detail. A summary is given in Sect. 4.

2. The λ6 cm survey and the large SNRs

The Sino-German λ6 cm polarization survey of the Galactic plane has been performed with a dual-channel λ6 cm receiving system constructed at MPIfR and mounted on the Urumqi 25-m radio telescope. The receiving system had a system temperature of about 22   K Ta, the central frequency was 4800 MHz, and the half power beam width (HPBW) was . The observations were made by scanning in both and b directions. Stokes parameters I, U, and Q were stored simultaneously for each map. The system set-up and the data reduction were described in detail by Sun et al. (2006, 2007) and Gao et al. (2010).

The total intensity and polarization calibration was made with observations of the primary polarization calibrator, 3C 286 (flux density at λ6 cm S6   cm = 7.5 Jy, polarization angle PA = 33°, and percentage polarization Π = 11.3%), and the secondary calibrators, 3C 48 (S6   cm = 5.5 Jy, PA = 108°, Π = 4.2%) and 3C 138 (S6   cm = 3.9 Jy, PA = 169°, Π = 10.8%). Instrumental effects in Stokes U and Q maps were corrected for the leakage of total intensity I into the polarization channels as described by Sun et al. (2007).

The λ6 cm survey region covers 10° ≤  ≤ 230° and |b| ≤ 5° of the Galactic plane. According to the SNR catalog of Green (2009), twenty-two known SNRs in our survey region have a size larger than 1°. In the λ6 cm survey, the size of G43.9+1.6 is less than 1°, while G127.1+0.5, G166.0+4.3 (VRO 42.05.01), and G189.1+3.0 (IC443) are larger than 1°. Thus 24 SNRs are left. Four SNRs, SNR G126.2+1.6 and G127.1+0.5 (Sun et al. 2007), SNR G180.0−1.7 (Xiao et al. 2008), and G130.7+3.1 (Shi et al. 2008), have previously been studied. SNRs G13.3−1.3 is omitted in this paper because it was not reliably detected in the λ6 cm survey. G28.8+1.5 is omitted because only a small section of the radio shell is detected and is confused by strong diffuse Galactic emission. Faint diffuse emission is detected in the area of SNR G32.0−4.9 and only some bright knots above 3σ noise level were seen in the survey map. We cannot unambiguously associate the diffuse emission with the SNR and therefore do not study this object. SNR G108.2−0.6 (Tian et al. 2007) was also skipped because it is confused by strong thermal emission from nearby H ii regions and cannot be separated. However, the two SNRs, G82.2+5.3 (W63) and G89.0+4.7 (HB21), are located at the b =  + 5° boundary of the λ6 cm survey and have been completely observed (Xiao et al. 2011). They were included. We therefore study 17 objects in this paper, where G192.8−1.1 is disproved as being a SNR in Sect. 3.1.

thumbnail Fig. 1

Polarization intensity (PI) in greyscale images overlaid by total intensity contours for 16 SNRs (more in next pages). Four of them, G69.0+2.7 (CTB 80), G89.0+4.7 (HB21), G160.9+2.6 (HB9), and G205.5+0.5 (Monoceros Nebula), are presented in page 5 by slightly larger plots. Observed polarization B-field vectors (i.e. PA of the observed E-vector plus 90°) of SNRs were overlaid every 6′ in case PI exceeds a certain threshold (see below). Their lengths are proportional to PI. The stars indicate the pulsar in G114.3+0.3 and pulsar wind nebula inside G106.3+2.7, G189.1+3.0 (IC 443) and CTB 80. The thick dashed line marks the boundary for flux density integration for G39.7−2.0 (W50), G65.1+0.6, G78.2+2.1, G82.2+5.3 (W63), G106.3+2.7, IC 443, CTB 80, and the western shell of the Monoceros Nebula. All contour lines run from the local 3σ level (see values for each SNR below) in different steps. The contour start level, contour step, and PI threshold for plotted B-vectors (all in mK TB) are for W50: 13.5, 13.5, 3.0; for G65.1+0.6: 3.9, 5.0, 4.0; for G78.2+2.1: 90.0, 120.0, 15.0; for W63: 9.0, 24.0, 8.0; for G93.7−0.2: 4.5, 12.0, 7.0; for G106.3+2.7: 6.0, 4.2, 3.2; for G114.3+0.3: 4.0, 4.0, 3.8; for G116.5+1.1: 3.6, 4.5, 7.0; for G166.0+4.3 (VRO 42.05.01): 2.1, 5.0, 3.5; for G179.0+2.6: 1.5, 3.5, 2.5; for G206.9+2.3 (PKS 0646+06): 3.0, 4.5, 3.0 for HB21: 3.3, 34.0, 7.5; for HB9: 3.0, 7.5, 7.0. For the Monoceros Nebula: 4.0, 6.0, 4.5. For IC 443, contours start at 7.2   mK  TB and run in steps of 2n  −  1 × 4.8   mK   TB (n = 1,2,3,...), the PI threshold is 7.0   mK  TB. For CTB 80, contours start at 3.3   mK  TB and run in steps of , and the PI threshold is 10.0   mK  TB.

Open with DEXTER

thumbnail Fig. 2

Flux densities at various frequencies were used to derive the integrated radio spectra for 15 SNRs (not including G205.5+0.5). The black dots indicate the newly derived flux density at λ6 cm from the Urumqi observations, and the triangles for the newly derived flux densities from the Effelsberg λ11 cm and λ21 cm surveys. The open circles are integrated flux densities taken from literature (see references in Table 1).

Open with DEXTER

3. Results and discussions

The SNR images were obtained after we filtered out the large-scale Galactic diffuse emission by using the technique of “background filtering” (Sofue & Reich 1979). A twisted hyper-plane was defined by the average values of pixels without any obvious structure in the four map corners and subtracted for each SNR to remove the local-background Galactic diffuse emission in I, U, and Q maps. We obtained total intensity and polarization images of these SNRs as shown in Fig. 1.

We integrated the flux density (S6   cm) and polarization intensity (PI6   cm) of SNRs within the boundary defined by the 3σ noise level around each SNR in total intensity (see Table 1), where σ is the rms of the map estimated from the surrounding areas of each SNR without obvious structure. For the SNRs G39.7−2.0 (W50), G65.1+0.6, G78.2+2.1, G82.2+5.3 (W63), G106.3+2.7, G189.1+3.0 (IC443), and G69.0+2.7 (CTB 80), diffuse emission adjacent to the SNR is detected in the λ6 cm map. We thus outline the SNR boundary according to published radio maps at other frequencies for flux density integration (see Fig. 1) to avoid possible confusion. For the complicated structure of SNR G205.5+0.5 (Monoceros Nebula), we outline only the western shell for the integration at λ6 cm, λ11 cm, and λ21 cm to determine its spectrum. The uncertainty of each integrated SNR flux density is calculated from the rms of a map and the pixel number of the integration area. The calibration and base-level uncertainties are assumed to be 10%, and added to the final value of the uncertainty. Compact sources exceeding a flux density of 20 mJy in the NVSS catalogue at 1.4 GHz (Condon et al. 1998) and located within the SNR boundary were discounted from the integrated SNR flux densities. The flux densities of point-like sources at λ6 cm were estimated from the extrapolation of the NVSS flux density at 1.4 GHz (Condon et al. 1998) using a spectral index either taken from Vollmer et al. (2005) or derived from the flux densities at 408 MHz and 1420 MHz from the Canadian Galactic Plane Survey (CGPS) (Taylor et al. 2003). If no spectral index is available for a source, we adopted the mean value of α = −0.9 according to Zhang et al. (2003) obtained from the NVSS and the WSRT source samples.

We obtained integrated flux densities of a number of SNRs (see Table 1) from the Effelsberg λ11 cm survey (Fürst et al. 1990; Reich et al. 1990a), λ21 cm survey (Reich et al. 1990b, 1997), and the λ21 cm Effelsberg Medium Latitude Survey (EMLS) (Uyanıker et al. 1999) within the same boundary as used for the λ6 cm integration.

Using the newly determined flux densities at λ6 cm, λ11 cm, and λ21 cm, together with previously published integrated flux densities at other frequencies (see Table 1), we determined the spectral indices α (here Sν ~ να) of 15 SNRs, except for SNR G205.5+0.5, where ν is the observing frequency. Flux densities and the determined spectra are shown in Fig. 2. The spectra of a few SNRs were verified by using the TT-plot method (Turtle et al. 1962), i.e. the brightness temperatures TB at two frequencies are plotted against each other. The brightness temperature spectral index, β, is defined as TB = νβ. In general, β = α−2.

3.1. G192.8−1.1 is not a SNR

G192.8−1.1 (PKS 0607+17) has been proposed as a SNR with an angular extent of about 80′ (Milne & Hill 1969; Caswell 1970). The non-thermal spectral index derived was α ~ −0.5 (Milne & Dickel 1974; Dickel & Denoyer 1975). Berkhuijsen (1974) suggested that G192.8−1.1 is a small part of an even larger SNR, which she called the “Origem Loop” with a diameter of about 5°, but later Caswell (1985) performed high angular resolution (~1′) observations at λ21 cm using the DRAO synthesis telescopes and argued again that G192.8−1.1 is a discrete SNR with a radio spectral index of α ~ −0.55. However, previous measurements were limited either by low sensitivity or too coarse angular resolution, which could lead to an incorrect conclusion.

thumbnail Fig. 3

Top: total intensity (I) image at λ6 cm for G192.8−1.1 (PKS 0607+17) and its surroundings. The “background filtering” technique was used to remove large-scale emission. Contours start at 3.0   mK  TB (3σ level) and run in steps of 5.0   mK  TB. The big rectangle area includes the G192.8−1.1, which was used for the TT-plot analysis by excluding the two areas marked by white small rectangles. Middle: PI image of G192.8−1.1 area (PKS 0607+17) at λ6 cm. The same total intensity contours are overlaid as in the top plot. Bars are plotted for observed B-vectors (i.e. PA + 90°) at λ6 cm if . Bottom: infrared image at 60   μm of the same area (Cao et al. 1997) overlaid by λ6 cm I contours.

Open with DEXTER

thumbnail Fig. 4

TT-plot for the G192.8−1.1 “plateau” using the Urumqi λ6 cm and the Effelsberg λ21 cm (1408 MHz) survey data (top panel) and the λ6 cm and the Effelsberg λ11 cm (2695 MHz) survey data (bottom panel).

Open with DEXTER

Sensitive survey observations from Urumqi at λ6 cm (Gao et al. 2010), Effelsberg at λ11 cm (Fürst et al. 1990), and λ21 cm (Reich et al. 1997) all covered this object. The λ6 cm total intensity map shows G192.8−1.1 in the vicinity of a number of bright H ii regions (Fig. 3). The H ii region to the southeast, SH 2-261, has a distance of 1 kpc (Chavarria-K et al. 1987); the cluster of H ii regions, SH 2-254 to 258, to the northwest have distances of about 2.5 kpc (Carpenter et al. 1995). Four individual objects, A (), B (), C (), and D () as indicated in Fig. 3 were discussed by Day et al. (1972) based on their 2.7 GHz Galactic survey data obtained with the Parkes telescope. All of them except source A were believed to be parts of the SNR. The point-like source A was listed as an H ii region by Paladini et al. (2003) according to Felli & Churchwell (1972). However, this source also named as 4C 16.15 clearly has a non-thermal spectrum (Vollmer et al. 2005) and is therefore not an H ii region. The elongated source B is located west of source A. Caswell (1985) considered source B as part of the SNR shell of G192.8−1.1. We found that source B has a thermal spectrum according to the TT-plot between the Urumqi λ6 cm and the Effelsberg λ21 cm survey data. The brightness-temperature spectral index is β = −2.06 ± 0.06. In addition, strong infrared emission (see Fig. 3, bottom plot) is visible, which proves its thermal nature.

The enhanced radio emission region where the source C and D reside is regarded as another part of the SNR shell of G192.8−1.1 (Caswell 1985). Source C consists of a few point-like sources including the H ii region SH 2-259. It appears slightly extended towards source D at angular resolution. Source D was resolved into two sources in the NVSS, and its flux density mainly originates in the source NVSS J060951+172532, which has a spectral index of α ~ −0.73 (Vollmer et al. 2005) and is most likely extragalactic. In summary, sources A, B, C, and D do not seem to be related to the proposed SNR.

The diffuse emission plateau was regarded by Caswell (1985) as part of the SNR. We subtracted all point-like sources including source A and discarded the regions where sources B, C, and D are located. Using the TT-plot method, we found a well-defined flat spectrum with a slope of β = −2.00 ± 0.07 between the Urumqi λ6 cm and the Effelsberg λ21 cm survey data, and β = −1.87 ± 0.06 between λ6 cm and Effelsberg λ11 cm survey data (see Fig. 4). Taking the diffuse infrared emission in this area (Cao et al. 1997) into account, we conclude that the extended emission plateau is of thermal nature.

Table 1

Integrated flux densities, polarization intensities and derived spectral indexes for 15 large SNRs (not including G205.5+0.5, see text).

The diffuse polarization patch in Fig. 3 located between G192.8−1.1 and the H ii region SH 2-261 is unrelated to G192.8−1.1.

To summarize, we found a thermal spectrum for the diffuse plateau of G192.8−1.1, which is in contradiction to the non-thermal spectrum reported in previous investigations. The extended source in the southeast, which was regarded as part of the SNR shell, is also identified as a thermal source. No polarized emission from G192.8−1.1 was detected at λ6 cm. This “object”, G192.8−1.1, is evidently not a SNR, because it consists of several unresolved H ii regions, background radio sources, and a thermal emission plateau.

3.2. G39.7−2.0 (W50)

W50 is an extended SNR located at about 5.5 kpc (Blundell & Bowler 2004; Lockman et al. 2007). It consists of a central component with a diameter of about 1° and two “ears” extended to the northwest and southeast (see Fig. 1). The spectacular X-ray binary system SS 433 is located at the center of the SNR (e.g. Brinkmann et al. 1996). Two relativistic jets from SS 433 are precessing around the major axis of W50. According to Begelman et al. (1980), the “ears” of W50 are caused by the pressure of these two jets and interact with the ambient H i (Lockman et al. 2007). Jowett & Spencer (1995) mapped SS 443 at 5 GHz with the MERLIN array and detected ejected radio blobs moving away from SS 443. Downes et al. (1981) and Downes et al. (1986) observed the total intensity and the polarized emission of W50 at λ18 cm, λ11 cm, and λ6 cm with the 100-m Effelsberg radio telescope. The spectral index and RM distributions of this SNR were studied in detail.

The Urumqi λ6 cm survey clearly shows the SNR W50 with strong polarized emission along its northeastern shell of the central part. SS 433 is seen at the center, but the related radio jets (Jowett & Spencer 1995) cannot be resolved. We obtained integrated flux densities of S6   cm = 37.0 ± 3.8 Jy from the Urumqi survey data, S11   cm = 51.5 ± 5.2 Jy and S21   cm = 69.7 ± 7.1 Jy from the Effelsberg λ11 cm and λ21 cm survey data. All of these agree with the values obtained by Downes et al. (1981, 1986) and Dubner et al. (1998). Using these flux densities together with previously published integrated flux densities, we derived an overall spectral index of α = −0.48 ± 0.03, consistent with that found by Dubner et al. (1998). The TT-plot using the Urumqi λ6 cm and Effelsberg λ11 cm survey data gave β = −2.41 ± 0.01 for the central part, β = −2.52 ± 0.02 for the southeast “ear”, and β = −2.57 ± 0.14 for the northwest “ear”, respectively. The central part appears to have a slightly flatter spectrum than the southeast “ear”, confirming the finding of Downes et al. (1986).

3.3. G65.1+0.6

G65.1+0.6 was first identified as a faint shell-type SNR by Landecker et al. (1990) from 408 MHz and 1420 MHz observations with the DRAO Synthesis Telescope. It shows strong southern shell emission, while its emission in the north is rather diffuse. Tian & Leahy (2006b) analyzed CGPS H i data and proposed a SNR distance of 9.2 kpc. Polarization from the southern shell was already observed at λ6 cm with the Effelsberg 100-m telescope (Seiradakis et al. 1985). Despite a smaller beam, no polarized emission was detected with the Effelsberg 100-m telescope at λ11 cm (Duncan et al. 1999) indicating significant depolarization at that wavelength. This is not unexpected in view of the large distance of the SNR.

Our λ6 cm data confirm the oval shape of G65.1+0.6 (see Fig. 1). We obtained an integrated flux density at λ6 cm of S6   cm = 3.2 ± 0.3 Jy. We also calculated the λ21 cm integrated flux density from the Effelsberg λ21 cm survey to be S21   cm = 5.4 ± 0.6 Jy. This value is consistent with those quoted by Landecker et al. (1990) and Tian & Leahy (2006b) from the CGPS, which used the Effelsberg λ21 cm survey for missing zero-spacings. Using these flux densities together with the flux densities at 408 MHz (Landecker et al. 1990; Kothes et al. 2006b; Tian & Leahy 2006b) and other bands (see references in Table 1), we obtained a radio spectral index of α = −0.45 ± 0.04 (see Fig. 2).

We detected very weak polarization at λ6 cm (see Fig. 1), which seems to be related to the SNR shell, as the magnetic field direction is almost tangential.

3.4. G69.0+2.7 (CTB 80)

CTB 80 is a SNR with an enigmatic shape. It consists of a bright pulsar wind nebula (PWN) in its center, a diffuse emission plateau and three major arms (Angerhofer et al. 1981; Mantovani et al. 1985). Its distance is 2 kpc (Koo et al. 1990). A young energetic pulsar, PSR B1951+32, with Ė = 3.7 × 1036 erg s-1 was discovered in the central core region (Kulkarni et al. 1988). Strong polarized emission of CTB 80 was observed, which varies with frequency (Mantovani et al. 1985). The overall radio spectral index is α = −0.45 ± 0.03 (Kothes et al. 2006b). A possible spectral break at a low frequency below 1 GHz was noted by Mantovani et al. (1985) and Castelletti et al. (2003).

The morphology of CTB 80 from the λ6 cm observations (see Fig. 1) generally agrees with previous results. We see faint extended emission from the eastern arm (E-arm) extending to about , in agreement with the observations by Mantovani et al. (1985). This part of CTB 80 is highly polarized at λ6 cm, and seen in the polarization map at 1.4 GHz (Kothes et al. 2006b). The apparent size of this SNR measured from the λ6 cm map is about 128′ from east to west and about 96′ in the south-north direction, which implies that it has a physical size of 74 pc  ×  56 pc at 2 kpc. In the λ6 cm map, we detected diffuse extended emission in the southwest between the W-arm and S-arm.

We investigated the spectral index distribution in the central core region and the three arms using Urumqi λ6 cm and Effelsberg λ21 cm survey data for TT-plots (not shown in this paper). We confirmed the flat spectrum in the central core with a brightness temperature spectral index of β = −1.97 ± 0.20. We derived β = −2.49 ± 0.03 for the western arm (W-arm), β = −2.38 ± 0.04 for the southern arm (S-arm), and β = −2.28 ± 0.01 for the eastern arm (E-arm) and the outer extension.

To get the spectrum of CTB 80 from integrated flux densities, we set the SNR boundary according to the maps at four frequencies above 1 GHz published by Mantovani et al. (1985), and outlined by the dashed line in Fig. 1. After subtracting all background radio sources, we calculated an integrated flux density at λ6 cm of S6   cm = 35.6 ± 3.9 Jy, slightly lower than that quoted by Mantovani et al. (1985). In the same area, we got the λ11 cm and λ21 cm flux densities from the Effelsberg survey data as S11   cm = 47.6 ± 5.0 Jy and S21   cm = 56.7 ± 6.3 Jy, respectively. Using these three well-determined new flux densities, we got the spectral index of α = −0.38 ± 0.13 (see Fig. 2). As noticed by Mantovani et al. (1985), the spectral index of CTB 80 at higher frequencies should be considerably flatter than previously thought (Velusamy & Kundu 1974; Sofue et al. 1983).

Flux densities available from literatures are shown in Fig. 2. Taking the three newly obtained flux densities into account, the spectrum of this SNR seems to be flat below about 1 GHz. However, owing to the scattered data at low frequencies the spectral shape is difficult to assess. Castelletti & Dubner (2005) discussed a possible low-frequency spectral turn-over. However, free-free thermal absorption, which may produce such a spectral break, would be inconsistent with the low DM of the central pulsar, PSR B1951+32 (DM = 45   pc   cm-3, Hobbs et al. 2004).

Intense λ6 cm polarized emission pervades nearly the entire SNR. We found B-field vectors running almost continuously along the southern and the western arm, but vary in the eastern arm. The weaker and discontinuous polarized emission in the middle of the eastern arm is probably due to internal cancelation by the superposition of polarization vectors orientated in different directions. In the outer extension of the eastern arm the percentage polarization is up to 50%. In the western arm, the polarization fraction is about 14%, while in the southern arm, the percentage increases from north to the south with an average value of about 21%.

thumbnail Fig. 5

TT-plot for G78.2+2.1 using the Urumqi λ6 cm and Effelsberg λ11 cm survey data.

Open with DEXTER

3.5. G78.2+2.1

The SNR G78.2+2.1 is located in the direction of the Cygnus X region. It was suggested and confirmed to be a large SNR by Higgs (1977) and Higgs et al. (1977) based on radio maps at λ3 cm and λ21 cm. Its distance is about 1.5 kpc (Landecker et al. 1980). A few H ii regions were detected near its shell. The well-known H ii region γ Cygni Nebula overlaps with the bright southern shell. The H ii region G78.3+2.8 (Sabbadin 1976) is located to the north of the SNR, while the H ii region IC 1318b is seen in the southeast direction. Several methods have been tried to separate non-thermal and thermal emission in the Cygnus X area (Wendker et al. 1991; Zhang et al. 1997; Ladouceur & Pineault 2008) to study the SNR emission without the dominating thermal background. Spectral index variations within the SNR were discussed in detail by Zhang et al. (1997) and later by Ladouceur & Pineault (2008) with a higher angular resolution of .

The Urumqi λ6 cm total intensity image (see Fig. 1) shows two bright unresolved shell sections in the north and south and the central weak extended emission. Not many detailed structures can be recognized because of the angular resolution of . The boundary of the SNR is difficult to determine in the Urumqi λ6 cm map, since the shell sections of the SNR merge with ambient H ii regions. We subtracted the H ii region G78.3+2.8 from the map using Gaussian fitting, and removed point-like sources as described in Sect. 3. After discounting the flux density of the γ Cygni Nebula, S6   cm = 4.3 Jy (Wendker et al. 1991), we obtained an integrated flux density of this SNR as S6   cm = 170 ± 18 Jy, consistent with the value of S6   cm = 150 ± 15 Jy measured by Wendker et al. (1991). We derived S11   cm = 222 ± 23 Jy for the Effelsberg λ11 cm survey data. Combining the flux densities reported at other wavelengths, we obtained an overall spectral index of α = −0.48 ± 0.03, consistent with the result of Kothes et al. (2006b). The TT-plot for the SNR using Urumqi λ6 cm and Effelsberg λ11 cm survey data (Fig. 5) gives β = −2.47 ± 0.04, which agrees with the spectrum derived from integrated flux intensities.

The λ6 cm polarization data in this area are largely confused with residual instrumental effects, e.g., some horizontal stripes from the strong Cygnus X region emission.

thumbnail Fig. 6

TT-plot for the two filaments aside of G82.2+5.3 (W63) indicated in Fig. 1 using the Urumqi λ6 cm and Effelsberg λ21 cm data.

Open with DEXTER

3.6. G82.2+5.3 (W63)

W63 is a shell-type SNR located at a high Galactic latitude north of the Cygnus X complex. In this area, filamentary structures and strong diffuse Galactic background emission exist, which might confuse with emission from this SNR. Some radio filaments coincide with optical emission. They are most probably thermal and unrelated to the SNR (Wendker 1971). Angerhofer et al. (1977) mapped W63 at λ6 cm, but did not present a complete map. They reported a lower flux density limit of  Jy. From the Urumqi λ6 cm observations, we obtained complete total intensity and polarization maps for W63 at λ6 cm (see Fig. 1). We checked the properties of the two major filaments centered at and (marked with “W-fila” and “E-fila” in Fig. 1). The brightness-temperature spectral index from the TT-plot (Fig. 6) using the Urumqi λ6 cm and Effelsberg Medium Latitude Survey (EMLS) λ21 cm data (Uyanıker et al. 1999) is β = −2.05 ± 0.07 for the western filament (W-fila) and β = −2.05 ± 0.01 for the eastern filament (E-fila) (Fig. 6). The Hα map (Fig. 7) clearly shows the two bright filaments and their extensions beyond W63, which proves their thermal nature. Therefore, we discounted these two filaments in the flux density integration and got S6   cm = 48.5 ± 4.9 Jy. We also calculated the integrated flux density at λ21 cm for the same area from the EMLS (Uyanıker et al. 1999), S21   cm = 93.1 ± 9.5 Jy. Combining all available flux densities, we obtained a spectral index of α = −0.44 ± 0.04.

thumbnail Fig. 7

Hα emission (from Finkbeiner 2003) overlaid by Urumqi λ6 cm total intensity contours and B vectors (i.e. PA + 90°) from the SNR G82.2+5.3 (W63). Contours start at 9.0 mK TB and run in steps of 24.0 mK TB. B-field vector lengths are proportional to the polarization intensities.

Open with DEXTER

Strong polarized λ6 cm emission was detected within the SNR, where the strongest polarized emission coincides with the area of minimum total intensity, similar to what was seen at λ11 cm by Velusamy & Kundu (1974). Polarized emission at λ21 cm is visible within the SNR with a percentage polarization of about 4% to 11% in the area, where the λ6 cm polarization is up to 66%. We noticed that the Hα emission (Finkbeiner 2003) is relatively low in an area almost coinciding with the high polarization of W63 (see Fig. 7). This suggests that the observed Hα emission is probably in the foreground of W63 and causes depolarization in other parts of the SNR.

3.7. G89.0+4.7 (HB21)

HB21 is a strong, large, and evolved SNR with an unusual complex morphology (see Fig. 1). It has been observed at many frequencies from 38 MHz (Crowther 1965) to 5 GHz (Kundu 1971). The distance to the SNR was estimated to be about 800 ± 70 pc (Tatematsu et al. 1990). HB21 resides in an area of relatively low Galactic emission and has a well-defined outer boundary, which allows us to obtain integrated flux densities without any confusion problem.

HB21 was partly covered by the 1.4 GHz polarization survey of the Galactic plane by Landecker et al. (2010) with 1′ angular resolution, where data from the DRAO interferometer were combined with Effelsberg 100-m and DRAO 26-m telescope data. At that frequency, HB21 appears almost depolarized with the possible exception of a few discrete patches in this direction.

The λ6 cm total intensity and polarization intensity maps resemble the previous λ6 cm observations presented by Kundu (1971). We obtained an integrated flux density of λ6 cm map as S6   cm = 107 ± 11 Jy. Including other available flux densities for a wide frequency range (see references in Table 1), we fitted a spectral index of α = −0.36 ± 0.03. The λ6 cm polarization map shows that polarization angles are not regularly ordered. Some depolarized “canals” are visible, where the PAs on both sides differ by 90°. At the angular resolution of of the λ6 cm observations, details are smoothed out.

3.8. G93.7−0.2

SNR G93.7−0.2 is a shell-type SNR with two major thick brightened limbs and a few diffuse extensions. The extensions were regarded as possible “break-out” phenomenon (Uyanıker et al. 2002). Polarized emission of G93.7−0.2 is found to be widely spread at 4.75 GHz (Mantovani et al. 1991) and patchy at 1.4 GHz because of depolarization (Uyanıker et al. 2002). The distance to G93.7−0.2 was found to be about 1.5 kpc (Uyanıker et al. 2002).

The λ6 cm I map (Fig. 1) shows all extensions previously reported. The eastern extension at about is not part of the SNR but is an H ii region (Uyanıker et al. 2002). We found another H ii region at , where the TT-plot between the Urumqi λ6 cm and the Effelsberg λ21 cm survey data (not shown in this paper) gave a brightness-temperature spectral index of β = −2.11 ± 0.11. Discarding point-like sources and the thermal extensions, we obtained an integrated flux density of S6   cm = 25.0 ± 2.5 Jy, lower than that reported by Mantovani et al. (1991). We obtained S21   cm = 39.6 ± 4.0 Jy and S11   cm = 35.9 ± 3.6 Jy from the Effelsberg survey data in the same area. Combining these new flux densities with those in the literature, we obtained an overall spectral index for G93.7−0.2 of α = −0.52 ± 0.03.

The polarized emission of SNR G93.7−0.2 we detected at λ6 cm agrees with the observational results by Mantovani et al. (1991).

3.9. G106.3+2.7

G106.3+2.7 is a comet-shaped SNR discovered by Joncas & Higgs (1990) from the DRAO northern Galactic plane survey at 408 MHz, and further studied by Pineault & Joncas (2000). The SNR consists of a compact “head” region containing an off-center PWN G106.6+2.9 in the north, the so-called “Boomerang” and a diffuse “tail” region (see Fig. 1). On the basis of the linearly polarized emission of the SNR, H i, and CO measurements, Kothes et al. (2001) argued that both the PWN and the SNR resulted from the same SN event at a distance of 800 pc. The “head” region is created by the interaction between the expanding blast wave and the ambient dense material, while the “tail” region is an outbreak to the interior of an H i bubble. Kothes et al. (2006a) performed a multi-frequency analysis towards the PWN and found a spectral break at about 4.3 GHz. The overall spectral index of the whole SNR is α ~ −0.6, while the compact “head” region has a flatter spectrum than the “tail” region (Pineault & Joncas 2000).

Observations of the entire SNR at λ6 cm have never been made. The Urumqi λ6 cm map shows a compact “head”, while the elongated “tail” region is confused by diffuse emission in the west. The PWN, as indicated by a star in Fig. 1, cannot be resolved in the λ6 cm map. To determine the integrated flux density of G106.3+2.7 at λ6 cm, we adapt the integration boundary from the Effelsberg λ21 cm survey map, which can be more clearly defined. We obtained S6   cm = 2.0 ± 0.3 Jy from the Urumqi map, and both S11   cm = 3.1 ± 0.3 Jy and S21   cm = 5.0 ± 0.6 Jy from the Effelsberg maps. Using these flux densities, together with published flux densities at other frequencies, we obtained an integrated spectral index of α = −0.64 ± 0.04.

Strong polarized emission of SNR G106.3+2.7 was detected at λ6 cm. The average percentage polarization is about 25% with maxima around 42%.

thumbnail Fig. 8

TT-plot of the SNR G114.3+0.3 using the Urumqi λ6 cm and Effelsberg λ21 cm survey data.

Open with DEXTER

3.10. G114.3+0.3

G114.3+0.3 was identified as a low surface brightness shell-type SNR by Reich & Braunsfurth (1981). It is located in the local arm at about 700 pc distance (Yar-Uyanıker et al. 2004). The pulsar B2334+61, indicated by a star in Fig. 1, is located near the SNR center, and physically associated with the SNR (Fürst et al. 1993; Kulkarni et al. 1993). The spectral index of this SNR cannot be clearly determined because of its low surface brightness. Reich & Braunsfurth (1981) quoted a spectral index α = −0.31 ± 0.1 from flux densities at λ21 cm and λ11 cm. Tian & Leahy (2006a) used the TT-plot method for the 408 MHz and 1420 MHz CGPS data and obtained α = −0.16 ± 0.41.

No other λ6 cm observations of G114.3+0.3 have so far been reported. The map of this elliptical-shaped SNR is presented in Fig. 1. The ridges of strong continuum emission in the southwest and southeast directions are clearly visible. After removing all point-like sources including the bright H ii region, SH 2-165, in the eastern part of SNR (Reich & Braunsfurth 1981), we calculated an integrated flux density of S6   cm = 6.9 ± 0.7 Jy. From the Effelsberg λ21 cm survey, we obtained an integrated flux density S21   cm = 10.8 ± 1.1 Jy, which agrees with that obtained by Tian & Leahy (2006a). Using these flux density data, as well as the value measured at 408 MHz from Tian & Leahy (2006a) despite its large uncertainties, we got a spectral index of α = −0.31 ± 0.10 (see Fig. 2). This spectrum index can be verified by the TT-plot between the Urumqi λ6 cm and Effelsberg λ21 cm survey data as β = −2.30 ± 0.12 (Fig. 8).

Polarized emission of this SNR mainly comes from the central and southern parts, and is very strong along the southern edge. When we compared the PA = −15° at λ11 cm (Reich & Braunsfurth 1981) in a region centered at about with the PA = 45° we observed at λ6 cm, we found that RM = −120 rad m-2 for the SNR, which is consistent with RM of the pulsar PSR B2334+61, RM = −100 ± 18   rad m-2 (Mitra et al. 2003).

thumbnail Fig. 9

The same as shown in Fig. 8, but for the SNR G116.5+1.1.

Open with DEXTER

3.11. G116.5+1.1

G116.5+1.1 is a shell-type SNR discovered by Reich & Braunsfurth (1981). Its distance is 1.6 kpc (Yar-Uyanıker et al. 2004). Tian & Leahy (2006a) derived a spectral index of α = −0.28 ± 0.15 from the 408 MHz and 1420 MHz CGPS data, while Kothes et al. (2006b) calculated α = −0.16 ± 0.11. By supplementing the flux densities measured at λ21 cm and λ11 cm (Reich & Braunsfurth 1981), Kothes et al. (2006b) found a steeper overall spectral index of α = −0.53 ± 0.08.

We obtained the first total intensity and polarization measurements for G116.5+1.1 at λ6 cm. The integrated flux density was found to be S6   cm = 5.7 ± 0.6 Jy, We also obtained S21   cm = 10.3 ± 1.1 Jy from the Effelsberg λ21 cm survey data and calculated an overall radio spectral index of α = −0.45 ± 0.05. The TT-plot was used to verify this result (see Fig. 9).

Intense polarized emission is detected from the western and northern parts of the shell of this SNR. In the western part, B-field vectors (i.e. PA + 90°) are aligned tangentially to the shell at λ6 cm. By comparing them with the PAs observed at λ11 cm by Reich & Braunsfurth (1981, see their Fig. 2b), we estimated the RM. The PA values in the region around at λ11 cm and λ6 cm are PA11   cm ~ 40° and PA6   cm ~ 100°, so the calculated RM ~ −120  rad m-2, is the same as that for SNR G114.3+0.3.

3.12. G160.9+2.6 (HB9)

HB9 is a strong large nearby SNR that has been mapped at radio frequencies from 10 MHz (Caswell 1976) to 5 GHz (DeNoyer 1974). We note that the designation G160.9+2.6 is based on early radio maps. The center of HB9 is at . The SNR has a distance of 0.8 ± 0.4 kpc (Leahy & Tian 2007). HB9 has multiple fragmented shells and internal filamentary structures. Reich et al. (2003) reported a spectral index of α = −0.57 ± 0.03 based on a TT-plot between 865 MHz data and 4750 MHz Effelsberg data. Kothes et al. (2006b) compiled integrated flux densities at many frequencies and obtained a spectral index of α = −0.64 ± 0.02.

From the Urumqi λ6 cm survey we calculated an integrated flux density of S6   cm = 33.9 ± 3.4 Jy, which is consistent with the value of 36 ± 8 Jy previously reported by DeNoyer (1974) at 5 GHz. Using our λ6 cm flux density and those at lower frequencies (see Table 1), we calculated a spectral index of α = −0.59 ± 0.02, in agreement with that of Reich et al. (2003).

The polarized emission of HB9 at λ6 cm was previously mapped with the 100-m Effelsberg telescope to find the intrinsic magnetic field configuration within the SNR, by using in addition polarization data at λ11 cm and λ21 cm (Fürst & Reich 2004). Our polarization map at λ6 cm of HB9 has lower angular resolution. Strong polarization is seen in a few patches near the southern, northern, and eastern periphery of the shell.

3.13. G166.0+4.3 (VRO 42.05.01)

VRO 42.05.01 has an unusual morphology as it consists of two shells with significantly different radii (see Fig. 1). Pineault et al. (1987) interpreted this unusual morphology by modeling the southern shell (wing region) expanding from the dense warm medium surrounding the northern shell into a hot tunnel of very low density (see their Fig. 3). Kothes et al. (2006b) calculated an overall radio spectral index of α = −0.37 ± 0.03. Leahy & Tian (2005) studied the distribution of the spectral indices across the SNR and found that the spectral index in the southern wing is steeper than elsewhere.

We measured an integrated flux density of S6   cm = 3.3 ± 0.3 Jy, and derived an overall spectral index of α = −0.33 ± 0.04 using available data (see references in Table 1), which is consistent with previous results. By comparing the Urumqi λ6 cm and Effelsberg λ21 cm survey data, we confirm that the southern wing has a steeper spectral index. We calculated a brightness-temperature spectral index of β6−21 = −2.33 ± 0.05 for the northern shell (head region) and β6−21 = −2.49 ± 0.09 for the southern wing.

Both the northern head region and the southern wing region of VRO 42.05.01 emit strong polarized emission. B-field vectors in the head region and the eastern part of the wing region run almost tangential to the shell, while significant deviations are seen for the western part of the wing region indicating a significant RM within the SNR. Assuming the initial B-fields run tangential along the wing, the deviation of B-field vectors indicates that |RM| is about 130 rad m-2.

3.14. G179.0+2.6

G179.0+2.6 is a thick shell-type SNR identified and studied by Fürst & Reich (1986) using λ21 cm, λ11 cm, and λ6 cm data observed with the Effelsberg 100-m telescope. A polarized triple source near the center of the SNR was identified from high angular resolution VLA observations and by optical identification as a background double-sided radio galaxy (Fürst et al. 1989). The overall spectral index of the SNR was found to be about α = −0.30 ± 0.15.

After we discounted the flux contribution of 15 point-like sources listed in the NVSS catalogue (Condon et al. 1998), we obtained an integrated flux density of SNR G179.0+2.6 as S6   cm = 3.2 ± 0.3 Jy. The flux densities we obtained from the Effelsberg λ21 cm and λ11 cm survey are S21   cm = 5.4 ± 0.6 Jy and S11   cm = 5.0 ± 0.5 Jy. Because Fürst & Reich (1986) removed only the strongest source 4C 31.21 within the area of the SNR when calculating the integrated flux density, our results with an improved assessment of background source contribution are more accurate and reveal an overall spectral index for SNR G179.0+2.6 of α = −0.45 ± 0.11 (see Fig. 2).

A radial configuration of the magnetic field of SNR G179.0+2.6 was revealed by Fürst & Reich (1986) that is confirmed by the Urumqi λ6 cm polarization observations. Such a magnetic field configuration is usually seen in young SNRs, although the low surface brightness of G179.0+2.6 indicates that it is an evolved object. When the SNR is barrel-shaped, but viewed along its poles as described by Whiteoak & Gardner (1968) the magnetic field appears radial, although the magnetic field lines are tangential to the compressed shell. The observed radio emission is weak. This scenario would require that the regular Galactic magnetic field in the area of G179.0-2.6 is orientated along the line of sight. This is unexpected for the Galactic anti-center, where the regular magnetic field is believed to run almost tangential to the line of sight.

3.15. G189.1+3.0 (IC 443)

IC443 is a “mixed-morphology” type SNR (Rho & Petre 1998), that is distinct from the three well-defined SNR types: shell-like, Crab-like, and composite. It interacts with a molecular cloud (Burton et al. 1988; van Dishoeck et al. 1993; Cesarsky et al. 1999). This SNR has been extensively observed throughout the whole electromagnetic spectrum (e.g. Mufson et al. 1986; Albert et al. 2007). Kundu & Velusamy (1968) collected flux densities at five frequencies from 430 MHz to 5 GHz, and suggested a spectral break around 750 MHz towards the brightened northwestern shell. Erickson & Mahoney (1985) proposed a possible overall spectral break below 20 MHz. According to Green (1986), spectral indices vary across IC 443. Leahy (2004) found that the brightened northwestern shell region has a radio spectral index of α = −0.43 ± 0.02, while in the fainter southern part α varies with values ranging from −0.2 to −0.6.

As in the λ11 cm map by Fürst et al. (1990), the λ6 cm map of IC 443 in Fig. 1 shows strong radio emission from the SNR, as outlined by the dashed line, and also the weaker unpolarized thermal emission from the extended H ii region SH 2-249 north of the remnant. For IC 443, we obtained an integrated flux density from the λ6 cm map of S6   cm = 84.6 ± 9.4 Jy. Together with flux densities at other frequencies (see references in Table 1), we calculate an overall radio spectral index of α = −0.38 ± 0.01.

The Urumqi λ6 cm polarization image (Fig. 1) shows a similar dipole-shaped magnetic field configuration as earlier noted by Kundu & Velusamy (1969) and Dickel & Milne (1976). The radial configuration of the magnetic fields observed at λ6 cm seems to be intrinsic. The question is whether and how the field structure is related to the PWN inside IC 443. Olbert et al. (2001) and Bocchino & Bykov (2001) identified the PWN from hard X-ray observations with a compact head at about near the center of the dipole-shaped B fields (as indicated by a star in Fig. 1). Gaensler et al. (2006) found that inside IC 443 the X-ray source G189.22+2.9 is a thermally emitting neutron star moving through the hot plasma. The distance to the SNR IC 443 was noted to be about 1.5 kpc (Fesen 1984). At this distance, a 1° scale corresponds to the size of 26 pc, much larger than the typical PWN size of a few pc. It is thus unlikely that the PWN is responsible for the whole dipole-shaped B field observed (see Fig. 1). However, our data have insufficient angular resolution to clarify the role of the PWN inside IC 443 and determine the origin of the radial magnetic field configuration seen in IC 443.

From the X-ray image of ROSAT, Asaoka & Aschenbach (1994) detected a large (~1.5° in diameter) and very faint X-ray shell that overlapped with IC 443. They proposed that this shell is another SNR, G189.6+3.3. Near the IC 443 boundary, a filament was seen both in the optical image from the Digital Sky Survey and the radio at 1.4 GHz (Leahy 2004) that might be a part of this proposed SNR. At λ6 cm, only a very short narrow radio rim appears at , i.e. along the optical filament emerging from IC 443. This short rim is also seen in the Effelsberg λ11 cm and λ21 cm surveys. Using the TT-plot, we obtained β6−11 = −2.49 ± 0.25 and β6−21 = −2.35 ± 0.22. No radio emission from other parts of the proposed SNR G189.6+3.3 is detected at λ6 cm and the Effelsberg surveys. Therefore, these observations cannot clarify the existence of the proposed large SNR G189.6+3.3.

3.16. G205.5+0.5 (Monoceros Nebula)

The SNR Monoceros Nebula is located in the area of the bright Rosette Nebula (SH 2-275), a famous H ii region. Odegard (1986) suggested that the SNR is at the same distance as the H ii region of about 1.6 kpc and that both objects interact with each other. On the basis of the radio absorption, he also argued that the SNR is situated behind the H ii region. The SNR was studied at low and intermediate radio frequencies by Holden (1968), Milne & Hill (1969), and Dickel & Denoyer (1975). The radio structures are quite complicated in this region. The Rosette Nebula overlaps with the southern part of the SNR. The southern ridge of the H ii region SH 2-273 is also mixed with the edge of the western shell of the SNR as shown by Graham et al. (1982). The extended nebula 0641+06 with the center at about was identified to be an H ii region (Graham et al. 1982) and be unrelated to the SNR. Therefore, it was difficult to determine the boundary of the SNR, and reliable integrated flux densities could not be obtained. Early λ11 cm polarization observations were made by Milne & Dickel (1974), but no significant polarized emission was detected.

The λ6 cm Urumqi survey provides the shortest wavelength observations of the Monoceros Nebula and shows polarized emission, mainly towards the western shell of the SNR. After subtracting point-like sources, we obtained flux densities of the shell (integration area indicated in Fig. 1) from the Urumqi λ6 cm, Effelsberg λ11 cm, and λ21 cm survey data. These were well fitted by a non-thermal spectrum with α = −0.43 ± 0.12 (see Fig. 10).

3.17. G206.9+2.3 (PKS 0646+06)

G206.9+2.3 is a discrete SNR close to the Monoceros Nebula that has a bright northwestern radio shell. The distance was estimated to be from about 3 kpc to 5 kpc (Graham et al. 1982). In earlier studies, G206.9+2.3 was believed to be a possible extension of the Monoceros Nebula, until Davies et al. (1978) identified it as a distinct SNR based on optical observations. Graham et al. (1982) quoted flux densities at different radio frequencies from 38 MHz to 2.7 GHz and obtained a spectral index of α = −0.45 ± 0.03. On the basis of this result, Odegard (1986) argued for a spectral turnover below 38 MHz because of the lower flux density he measured at 20.6 MHz.

Observations of this SNR at λ6 cm or shorter wavelength bands were not yet available. The Urumqi survey showed strong polarized emission of up to 27% in the northwestern shell of the SNR (see Fig. 1). The B-field vectors are not tangential to the shell. The RM value for the shell of PKS 0646+06 may be high if the intrinsic B-field direction follows the shell.

After subtracting point-like sources including the strong point source NVSS J064935+053823 in the east of the SNR G206.9+2.3, we obtained a flux density of S6   cm = 2.9 ± 0.3 Jy from the Urumqi λ6 cm, and S11   cm = 3.9 ± 0.4 and S21   cm = 5.1 ± 0.6 Jy from the Effelsberg λ11 cm and λ21 cm survey data, respectively. Combining the available flux densities in the literature, we got a spectral index of α = −0.47 ± 0.04, which agrees with that derived by Graham et al. (1982).

thumbnail Fig. 10

Integrated radio flux densities of the western shell of SNR G205.5+0.5 (Monoceros Nebula) (as marked in Fig. 1).

Open with DEXTER

4. Summary

We have studied 16 SNRs larger than 1° based on maps from the Sino-German λ6 cm polarization survey of the Galactic plane carried out with the Urumqi 25-m telescope. We have found that G192.8−1.1 consists of compact sources and thermal components, and conclude that G192.8−1.1 is not a SNR.

For the SNRs, we obtained integrated flux densities when possible, derived their spectral indices, and analyzed the polarization data. Our results for the SNRs W50, G78.2+2.1, HB21, and IC 443, are consistent with previous observations. The Urumqi λ6 cm flux densities of SNRs, G65.1+0.6, W63, G93.7−0.2, G106.3+2.7, G114.3+0.3, G116.5+1.1, HB9, VRO 42.05.01, and G206.9+2.3, are those of the highest frequency available and are valuable for deriving the SNR spectra. All SNRs are polarized by more than 10% with the exception of G78.2+2.1, which is located towards the thermal Cygnus region and IC 443. The polarization images of W63, G106.3+2.7, G114.3+0.3, G116.5+1.1, HB9, and VRO 42.05.01, are those observed at the highest frequency so far, including those of the Monoceros Nebula and G206.9+2.3, which are the first polarization images at a radio band.

At λ6 cm, the polarization observations trace the intrinsic magnetic field configuration of SNRs when Faraday rotation can be neglected, which seems to be the case for most SNRs.

The λ6 cm total intensity and polarization data of the 16 SNRs and G192.8−1.1 are available from the NAOC website1.


Acknowledgments

We like to thank the anonymous referee for constructive comments. X.Y.G. thanks the joint doctoral training plan between CAS and MPG and financial support from CAS and MPIfR. The Sino-German λ6 cm polarization survey was carried out with a receiver system constructed by Mr. Otmar Lochner at MPIfR mounted at the Nanshan 25-m telescope of the Urumqi Observatory of NAOC. The MPG and the NAOC/CAS supported the construction of the receiving system by special funds. We thank Mr. Maozheng Chen and Mr. Jun Ma for qualified maintenance of the receiving system for many years. The Chinese authors are supported by the National Natural Science foundation of China (10773016, 10821061, and 10833003), the National Key Basic Research Science Foundation of China (2007CB815403) and the Partner group of the MPIfR at NAOC in the frame of the exchange program between MPG and CAS for many bilateral visits. X.H.S. acknowledges financial support by the MPG and by Prof. M. Kramer during his stay at MPIfR.

References

All Tables

Table 1

Integrated flux densities, polarization intensities and derived spectral indexes for 15 large SNRs (not including G205.5+0.5, see text).

All Figures

thumbnail Fig. 1

Polarization intensity (PI) in greyscale images overlaid by total intensity contours for 16 SNRs (more in next pages). Four of them, G69.0+2.7 (CTB 80), G89.0+4.7 (HB21), G160.9+2.6 (HB9), and G205.5+0.5 (Monoceros Nebula), are presented in page 5 by slightly larger plots. Observed polarization B-field vectors (i.e. PA of the observed E-vector plus 90°) of SNRs were overlaid every 6′ in case PI exceeds a certain threshold (see below). Their lengths are proportional to PI. The stars indicate the pulsar in G114.3+0.3 and pulsar wind nebula inside G106.3+2.7, G189.1+3.0 (IC 443) and CTB 80. The thick dashed line marks the boundary for flux density integration for G39.7−2.0 (W50), G65.1+0.6, G78.2+2.1, G82.2+5.3 (W63), G106.3+2.7, IC 443, CTB 80, and the western shell of the Monoceros Nebula. All contour lines run from the local 3σ level (see values for each SNR below) in different steps. The contour start level, contour step, and PI threshold for plotted B-vectors (all in mK TB) are for W50: 13.5, 13.5, 3.0; for G65.1+0.6: 3.9, 5.0, 4.0; for G78.2+2.1: 90.0, 120.0, 15.0; for W63: 9.0, 24.0, 8.0; for G93.7−0.2: 4.5, 12.0, 7.0; for G106.3+2.7: 6.0, 4.2, 3.2; for G114.3+0.3: 4.0, 4.0, 3.8; for G116.5+1.1: 3.6, 4.5, 7.0; for G166.0+4.3 (VRO 42.05.01): 2.1, 5.0, 3.5; for G179.0+2.6: 1.5, 3.5, 2.5; for G206.9+2.3 (PKS 0646+06): 3.0, 4.5, 3.0 for HB21: 3.3, 34.0, 7.5; for HB9: 3.0, 7.5, 7.0. For the Monoceros Nebula: 4.0, 6.0, 4.5. For IC 443, contours start at 7.2   mK  TB and run in steps of 2n  −  1 × 4.8   mK   TB (n = 1,2,3,...), the PI threshold is 7.0   mK  TB. For CTB 80, contours start at 3.3   mK  TB and run in steps of , and the PI threshold is 10.0   mK  TB.

Open with DEXTER
In the text
thumbnail Fig. 2

Flux densities at various frequencies were used to derive the integrated radio spectra for 15 SNRs (not including G205.5+0.5). The black dots indicate the newly derived flux density at λ6 cm from the Urumqi observations, and the triangles for the newly derived flux densities from the Effelsberg λ11 cm and λ21 cm surveys. The open circles are integrated flux densities taken from literature (see references in Table 1).

Open with DEXTER
In the text
thumbnail Fig. 3

Top: total intensity (I) image at λ6 cm for G192.8−1.1 (PKS 0607+17) and its surroundings. The “background filtering” technique was used to remove large-scale emission. Contours start at 3.0   mK  TB (3σ level) and run in steps of 5.0   mK  TB. The big rectangle area includes the G192.8−1.1, which was used for the TT-plot analysis by excluding the two areas marked by white small rectangles. Middle: PI image of G192.8−1.1 area (PKS 0607+17) at λ6 cm. The same total intensity contours are overlaid as in the top plot. Bars are plotted for observed B-vectors (i.e. PA + 90°) at λ6 cm if . Bottom: infrared image at 60   μm of the same area (Cao et al. 1997) overlaid by λ6 cm I contours.

Open with DEXTER
In the text
thumbnail Fig. 4

TT-plot for the G192.8−1.1 “plateau” using the Urumqi λ6 cm and the Effelsberg λ21 cm (1408 MHz) survey data (top panel) and the λ6 cm and the Effelsberg λ11 cm (2695 MHz) survey data (bottom panel).

Open with DEXTER
In the text
thumbnail Fig. 5

TT-plot for G78.2+2.1 using the Urumqi λ6 cm and Effelsberg λ11 cm survey data.

Open with DEXTER
In the text
thumbnail Fig. 6

TT-plot for the two filaments aside of G82.2+5.3 (W63) indicated in Fig. 1 using the Urumqi λ6 cm and Effelsberg λ21 cm data.

Open with DEXTER
In the text
thumbnail Fig. 7

Hα emission (from Finkbeiner 2003) overlaid by Urumqi λ6 cm total intensity contours and B vectors (i.e. PA + 90°) from the SNR G82.2+5.3 (W63). Contours start at 9.0 mK TB and run in steps of 24.0 mK TB. B-field vector lengths are proportional to the polarization intensities.

Open with DEXTER
In the text
thumbnail Fig. 8

TT-plot of the SNR G114.3+0.3 using the Urumqi λ6 cm and Effelsberg λ21 cm survey data.

Open with DEXTER
In the text
thumbnail Fig. 9

The same as shown in Fig. 8, but for the SNR G116.5+1.1.

Open with DEXTER
In the text
thumbnail Fig. 10

Integrated radio flux densities of the western shell of SNR G205.5+0.5 (Monoceros Nebula) (as marked in Fig. 1).

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.