I.V. Gosachinskiy " Interaction of the Supernova Remnant S 147 with the Ambient Interstellar Gas ".
ABSTRACT

Based on RATAN-600 21-cm HI line observations with an angular resolution of 2.4', we study the neutral-hydrogen distribution in the region of the supernova remnant (SNR) S 147 (G180.0-1.7).

We detected a rotating shell of neutral gas immediately adjacent to the SNR that expands at a velocity of 20 km/s. The HI shell is less distinct in the southeastern part and at negative radial velocities. The outer shell diameter is 90 pc; the HI mass in the shell is 2.2x10⁴M of the Sun.

These data allowed us to estimate the SNR age, 6.5x10⁵ yr, and the initial explosion energy, 2.2x10⁵¹ erg. (2002, MAIK "Nauka/Interperiodica")

INTRODUCTION

The supernova remnant ( SNR ) S 147 located in the Galactic-anticenter region is believed to be the most evolved SNR in the Galaxy. Except for the socalled spurs whose nature is also associated with nearby SNRs, S 147 has the largest angular size among SNRs in the radio and optical ranges, more than 3 dg.

In the radio range, it appears as an almost regular shell; in contrast to other SNRs, the spectral index of its radio emission clearly varies over the disk [see Furst and Reich ( 1986 ) and references in Green ( 1996 )].

In the optical range, the SNR appears as a tangle of filaments ( Van den Bergh et al.1973 ), with bright radio features corresponding to the brightest filaments. Radial-velocity measurements of optical absorption lines in the spectra of stars projected onto this object revealed large radial velocities of the absorbing gas, which are assumed to be attributable to the expansion of nebular filaments ( Silk and Wallerstein 1973; Phillips et al. 1981 ).

Direct measurements of the H_alpha emission from the nebula filaments ( Lozinskaya 1976 ) confirmed the presence of expansion velocities up to 100 km/s. The interstellar-gas distribution around this SNR was investigated in the HI radio line by Assousa et al. ( 1973 ), who found a shell at negative radial velocities ( from -47 to -24 km/s ).

Since this result was reported in a brief summary to the paper and since the complete paper has never been published, it is rather difficult to get an idea of these results.

In the observatory report for 1975, it was mentioned that the expansion velocity of this shell is 25 km/s. In the only HI survey by Koo and Heiles ( 1991 ) aimed at searching for HI brightenings in SNR regions, this object has the lowest "reliability rank" and was not included in the list of objects with excess HI emission.

The molecular-gas distribution was not studied either, except for Gondhalekar and Phillips ( 1980 ) who reported the detection of an ultraviolet CO absorption line.

Such a chilly attitude of observers to this object appears to be explained by its large sizes, so the background HI emission is difficult to separate from the features that could be associated with the SNR.

The high sensitivity in brightness temperature and high ( though one-dimensional ) angular resolution of RATAN-600 allow such observational problems to be effectively solved. In this paper, we present the results of our RATAN-600 HI 21 cm study for the SNR S 147 region carried out in 1999-2000.

INSTRUMENTATION AND TECHNIQUES

To investigate the distribution of interstellar neutral hydrogen in the S 147 region, we obtained ten drift curves in right ascension at 0.6 dg. intervals in declination over the declination range +25.0 dg. to +31.5 dg. In this elevation range, the RATAN-600 antenna has an angular resolution of 2.4' x 20' and an effective area of about 950 m² (Esepkina et al. 1979). An uncooled HEMT amplifier was used at the input (Il'in et al. 1997).

The system noise temperature was 70 K; the 39-channel filter spectrum analyzer had a channel bandwidth of 30 kHz (6.3 km/s) and the separation between channels was also 30 kHz (Venger et al. 1982). The data acquisition and preprocessing were made using an IBM PC (Alferova et al. 1986).

The drift curve at each declination consisted of two series of three observations each obtained by shifting the receiver tuning frequency by half the channel bandwidth. As a result, each drift curve had 78 spectral channels that followed at 3.15 km/s intervals. This technique also allowed the noises to be effectively cleaned. The rms of the antenna temperature fluctuations in spectral channels on an averaged record was 0.2 K. The antenna and equipment parameters were checked in each series of observations by measuring a set of reference sources (Venger et al. 1979).

Subsequently, we subtracted the extended background obtained by spline interpolation at the lower brightness-distribution level from the drift curves in each spectral channel and reduced the drift curves containing only features of small angular sizes.

The subtracted background component of the drift curves apparently includes the following:

• (1) large-scale features of the interstellar-gas distribution, such as spiral arms or giant complexes;
• (2) emission from the intercloud medium, if present;
• (3) features of small angular sizes unresolved by the RATAN-600 beam; and
• (4) the spurious large-scale background produced by distant side lobes and the RATAN-600 stray field.

It should also be noted that subtracting the background component by the above method can result in an underestimation of the brightness and angular sizes of the remaining small-scale features. We determined the parameters of the latter in each channel by Gaussian analysis and then attempted to establish a relationship between HI features at different radial velocities and at different declinations. Note that this is the only procedure where a significant subjective factor could be introduced.

The errors of the measured parameters have the following values. The radial velocity of an isolated medium-brightness HI feature is measured with an accuracy of at least +-1 km/s. In some cases, the accuracy deteriorates because of the difficulties in separating the object from the background or from adjacent features.

The measurement error of the HI line brightness temperature is 0.5 K, including the antenna calibration errors, and the estimation error of the angular size in right ascension is 0.1 dg. In declination, the antenna resolution is much lower and, accordingly, the accuracy of measuring the angular sizes is lower.

The accuracy of estimating the distances depends on the method of their determination and must be considered separately in each case. As a result, the accuracy of estimating the HI mass in an isolated cloud is no higher than 0.5-1 order of magnitude.

RESULTS OF THE OBSERVATIONS AND DISCUSSION

The distribution of 21 cm HI radio line emission in the SNR S 147 region after the subtraction of the extended background in the right ascensionradial velocity ( a-V ) plane is shown in Fig. 1. All of the zero-level remnant is located between 5^h30^m and 5^h43^m ( 1950.0 ).

The thin ring contours schematically indicate HI clouds in the a-V plane, which may represent an inhomogeneous shell immediately adjacent to the radio remnant in spatial coordinates. The isophotal pattern of this ring structure corresponds to the presence of large-scale motions in the shell-expansion and rotation. The ring-shaped pattern of isophotes in the a-V plane suggests the presence of radial motion; in our case, this can be only expansion.

The inclination of the isophotes of the presumed shell in the a-V plane corresponds to a radial velocity gradient across the object disk of 11 km/s per degree. Note that S 147 lies almost exactly at the Galactic anticenter, where the normal Galactic rotation gives a very small apparent effect: 0.4 km/s per kpc in the radial direction and 0.25 km/s per degree in the direction of Galactic longitude; in a, it is even smaller. Consequently, the most plausible explanation of the observed radial-velocity gradient across the disk is rotation. The observed parameters of the HI ring structure we identified are:

• coordinates of the center: a( 1950.0 ) = 5^h37.7^m, d( 1950.0 ) = +26.0 dg;
• angular size: 3.5 dg. (outer) and 2.2 dg. (inner);
• mean line brightness temperature: 10 +- 0.5 K;
• mean radial velocity: -10 +- 5 km/s;
• large-scale expansion velocity: 20 km/s;
• radial-velocity gradient: 11 km/s per degree.

To check whether the structure we identified is real, it would be appropriate to use data from other HI surveys. The survey by Westerhout and Wendlandt (1982), now accessible in the Strasbourg observational data archive ( cdsarc.u-strasbg.fr/cats /VIII/47 ), is most suitable in resolution. Having a high radial-velocity resolution ( 2 km/s ) and a sufficient angular resolution (13' ), this survey covers a region of only +-2 dg. in Galactic latitude, which is too small for the source S 147. However, we were able to use data from this survey for the drift curve in longitude at b = -1.5 dg. (virtually across the shell center) by transforming them from the FITS format to the common format we use, in order that we could also subtract the background HI line emission component.

The result of our reduction is presented in Fig. 2, where the thin lines delineate the shell structure that is similar in its observed parameterssurvey by Westerhout and Wendlandt (1982) confirms that the HI shell we detected is real.

Figure 1. The distribution of H I emission features in the vicinity of S 147 after the subtraction of the extended background. The radial velocities are given relative to the Local Standard of Rest; the declination is indicated at the top. The antenna temperature scale is given on the left. The thin lines schematically indicate the H I shell around the SNR. The thick vertical lines mark the SNR radio center in right ascension.

The molecular-gas distribution in the vicinity of the SNR S 147 can be checked by using data from the CO surveys by Dame et al. ( 1987, 2001 ), which are also accessible in the Strasbourg database. In this region, two isolated CO clouds are clearly seen at radial velocities of -10 and +5 km/s. Unfortunately, we have no reasons to suggest an association of these molecular clouds with the HI shell or SNR.

To calculate the physical parameters of the shell, we must assume an SNR distance if we are sure that previous distance estimates for this SNR based on the d-D relation was given by Lozinskaya (1976).

This distance is, on average, 1 kpc. After the detection of the pulsar that is apparently associated with this SNR ( Anderson et al. 1996 ), Case and Bhattacharya ( 1998 ) revised its distance to 1.5 kpc; they included S 147 in a small group of reference SNRs whose distances are considered to be known.

Given this distance, the physical parameters of the HI shell around the SNR S 147 are the following:

• outer diameter 90 pc;
• inner diameter 57 pc;
• mean gas density 3.25 cm^-3;
• HI mass in the shell 2.2 x 10⁴ M of the Sun;
• angular velocity -1.4 x 10^-14 rad/s.

We determine the density of the ambient interstellar medium, n₀ = 2.35 cm^-3, by assuming that the gas of the HI shell was initially spread over its entire present volume. Then, using the formulas from Wheeler et.al. ( 1980 ), we can estimate the initial explosion energy and the SNR age:

E₀ = 1.02 x 10⁴⁴V_s^1.27n₀^1.18R_s^3.16 erg,

t = t_rad+13R_rad^5/2( n₀ / E₅₁ )^1/2[( R_s / R_rad )^7/2-1] yr,
where E₅₁ = E₀ / 10⁵¹ erg; R_rad and t_rad are, respectively, the radius and initial time of the radiative SNR expansion phase:

R_rad = 15.2E₅₁^0.29n₀^-0.43 = 13.0 pc;

t_rad = 1.2 x 10⁴E₅₁^0.22n₀^-0.54 = 0.9 x 10⁴ yr.
As a result, we obtain E₀ = 2.2 x 10⁵¹ erg and t = 6.5 x 10⁵ yr, which are typical of the old shell-type SNRs.

CONCLUSIONS

As was mentioned above, we cannot compare the parameters of the HI shell revealed by the RATAN-600 observations with the only mention of a similar shell around the SNR S 147 in Assousa et al. ( 1973, 1975 ), because, in effect, the results of the latter were not published. We can only confirm that there are no detectable H I features at negative radial velocities ( <-50 km/s ) in our data that could be related to the SNR S 147.

Our data show that the shell is highly inhomogeneous in structure and is clearly seen only at low Galactic latitudes ( the northern part of the SNR) and at positive velocities of its expansion. In general, such a structure comes as no surprise, because the size of the SNR itself is comparable to the thickness of the Galactic-gas layer ( 100 pc) and the SNR is located almost 2 dg. to the south from the maximum of the neutral-gas layer and, most likely, in front of it.

It appears that the inhomogeneity of the ambient medium can also account for the variation in radio spectral index across the SNR disk found by Furst and Reich ( 1986 ). As for the shell rotation, it is probably the residual rotation of the gas cloud in which the supernova exploded. According to Phillips ( 1999 ) and Gosachinskij and Morozova ( 2000 ), rotation is typical of almost all molecular and atomic clouds of interstellar gas.

The total supernova explosion energy we obtained is close to the results by Lozinskaya ( 1976 ). However, the SNR age was found to be almost an order of magnitude larger than that in this paper and it is much closer to the estimates based on the radio observations by Sofue et al. ( 1980 ) or Kundu et al. ( 1980 ).

Figure 2. The distribution of H I emission features, as constructed from the survey data by Westerhout and Wendlandt ( 1982 ). The drift curve in l-V coordinates at the Galactic latitude b = -1.5 dg. is shown. The extended background of the H I emission was subtracted; the thin lines trace the shell structure.

By the parameters of the ambient gas, the SNR S 147 clearly belongs to the less than twenty SNRs whose interaction with the ambient interstellar medium is beyond question; i.e., expanding shells of neutral hydrogen were detected around them [see, e.g., Gosachinskij and Khersonskij ( 1985 )]. At present, we make searches for HI shells around more than 90 SNRs with angular diameters larger than 10-15' using RATAN-600. The results presented here are only a small part of this work.

REFERENCES

•  1.  Z. A. Alferova, I. V. Gosachinskij, S. R. Zhelenkov, and A. S. Morozov, Izv. SAO 23, 89 (1986).
•  2.  S. B. Anderson, B. J. Cadwell, D. A. Jacoby, et al., Astrophys. J. 319, 885 (1996).
•  3.  G. E. Assousa, B. Balick, and J. W. Erkes, Bull. Am. Astron. Soc. 5, 410 (1973).
•  4.  G. E. Assousa, B. Balick, and J. W. Erkes, Bull. Am. Astron. Soc. 7, 35 (1975).
•  5.  G. L. Case and D. Bhattacharya, Astrophys. J. 504, 761 (1998).
•  6.  T. M. Dame, H. Ungerechts, R. S. Cohen, et al., Astrophys. J. 322, 706 (1987).
•  7.  T. M. Dame, D. Hartmann, and P. Thaddeus, Astro- phys. J. 547, 792 (2001).
•  8.  N. A. Esepkina, N. S. Bakhvalov, B. A. Vasil'ev, et al., Izv. SAO 11, 182 (1979).
•  9.  E. Furst and W. Reich, Astron. Astrophys. 163, 185 (1986).
•  10.  P. M. Gondhalekar and A. P. Phillips, Mon. Not. R. Astron. Soc. 191, 13P (1980).
•  11.  I. V. Gosachinskij and V. K. Khorsonskij, Astrophys. Space Sci. 108, 303 (1985).
•  12.  I. V. Gosachinskij and V. V. Morozova, Astron. Zh. 76, 883 (1999) [Astron. Rep. 43, 777 (1999)].
•  13.  D. A. Green, in Proceedings of the IAU Colloquium No. 145 "Supernovae and Supermova Remmants", Xian, 1993, Ed. by R. McCray and Z. Wang (Cam- bridge Univ. Press, Cambridge, 1996), p. 419.
•  14.  G. N. Il'in, A. M. Pilipenko, and V. A. Prozorov, in Proceedings of the XXVII Radio Astronomical Conference "Problems of Modern Radio Astron- omy" (IPA RAN, St. Petersburg, 1997), p. 128.
•  15.  B.-C. Koo and C. Heiles, Astrophys. J. 382, 204 (1991).
•  16.  M. R. Kundu, P. E. Angerhofer, E. F ╗urst, and W. Hirth, Astron. Astrophys. 92, 225 (1980).
•   17.  T. A. Lozinskaya, Astron. Zh. 53, 38 (1976) [Sov. Astron. 20, 19 (1976)].
•  18.  A. P. Phillips, Astrophys. J., Suppl. Ser. 134, 241 (1999).
•  19.  A. P. Phillips, P. M. Gondhalekar, and J. C. Blades, Mon. Not. R. Astron. Soc. 195, 485 (1981).
•  20.  J. Silk and G. Wallerstein, Astrophys. J. 181, 799 (1973).
•  21.  Y. Sofue, E. Furst, and W. Hirth, Publ. Astron. Soc. Jpn. 31, 1 (1980).
•  22.  S. van den Berg, A. P. Marscher, and Y. Terzian, Astrophys. J. 26, 19 (1973).
•  23.  A. P. Venger, I. V. Gosachinskij, B. G. Grachev, and N. F. Ryzhkov, Izv. SAO 14, 118 (1981).
•  24.  A. P. Venger, V. G. Grachev, T. M. Egorova, et al., Soobshch. SAO, No. 35, 5 (1982).
•  25.  G. Westerhout and H. U. Wendlandt, Astron. Astro- phys., Suppl. Ser. 49, 137 (1982).
•  26.  J. C. Wheeler, T. J. Masurek, and A. Sivaramakrishnan, Astrophys. J. 237, 781 (1980).

Translated by G. Rudnitskii