1: %\documentclass[12pt,preprint]{aastex}

2: %\usepackage{emulateapj5,mathptmx,psfig,pstricks}

3: \documentclass[apj]{emulateapj}

4: \usepackage{pstricks,amsmath,psfig}

5: \usepackage{apjfonts}

6: 

7: 

8: \makeatletter

9: \newenvironment{inlinetable}{%

10: \def\@captype{table}%

11: \noindent\begin{minipage}{0.999\linewidth}\begin{center}\footnotesize}

12: {\end{center}\end{minipage}\smallskip}

13: 

14: \newenvironment{inlinefigure}{%

15: \def\@captype{figure}%

16: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}

17: {\end{center}\end{minipage}\smallskip}

18: \makeatother

19: 

20: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

21: 

22: \newcommand{\snr}{CTA~1}

23: \newcommand{\psr}{RX~J0007.0+7302}

24: \newcommand{\asca}{{\em ASCA}}

25: \newcommand{\rosat}{{\em ROSAT}}

26: \newcommand{\einstein}{{\em Einstein}}

27: \newcommand{\xmm}{{\em XMM-Newton}}

28: \newcommand{\chandra}{{\em Chandra}}

29: \newcommand{\gam}{2EG~J0008+7307}

30: \newcommand{\gamm}{3EG~J0010+7309}

31: 

32: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

33: 

34: \begin{document}

35: 

36: \submitted{Accepted for Publication in ApJ}

37: 

38: \lefthead{COMPACT SOURCE IN CTA 1}

39: \righthead{SLANE ET AL.}

40: 

41: \title{X-ray Observations of the Compact Source in CTA~1}

42: 

43: 

44: \author{Patrick Slane\altaffilmark{1}, Erik R. Zimmerman\altaffilmark{2},

45: John P. Hughes\altaffilmark{3}, Frederick D. Seward\altaffilmark{1},

46: Bryan M. Gaensler\altaffilmark{1}, and Melanie J. Clarke\altaffilmark{1}}

47: 

48: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics,

49:     60 Garden Street, Cambridge, MA 02138}

50: \altaffiltext{2}{Harvard College, Cambridge, MA 02138}

51: \altaffiltext{2}{Department of Physics and Astronomy, Rutgers, The State

52: University of New Jersey, Piscataway, NJ 08854-8019}

53: 

54: \begin{abstract}

55: The point source \psr, at the center of supernova remnant \snr,

56: was studied using the X-Ray Multi-mirror Mission (\xmm).  

57: The X-ray spectrum of the source is consistent with a neutron star

58: interpretation, and is well described by a power law

59: with the addition of a soft thermal component that may correspond

60: to emission from hot polar cap regions or to cooling emission from

61: a light element atmosphere over the entire star. There is evidence

62: of extended emission on small spatial scales which may correspond to

63: structure in the underlying synchrotron nebula. No pulsations are

64: observed.  Extrapolation of the nonthermal spectrum of \psr\ to gamma-ray

65: energies yields a flux consistent with that of EGRET source 

66: \gamm, supporting the proposition that there is a gamma-ray 

67: emitting pulsar at the center of \snr. Observations of the outer regions

68: of \snr\ with the Advanced Satellite for Cosmology and Astrophysics confirm

69: earlier detections of thermal emission from the remnant and show that

70: the synchrotron nebula extends to the outermost reaches of the SNR.

71: 

72: \end{abstract}

73: 

74: \keywords{

75:  stars: neutron ---

76:  stars: individual (RX~J0007.0+7302, 3EG~J0010+7309)  ---

77:  ISM: individual (CTA~1) ---

78:  supernova remnants ---

79:  X-rays: ISM}

80: 

81: 

82: \section{Introduction}

83: 

84: \snr\ (G119.5+10.2) is one of the class of composite supernova remnants 

85: (SNRs) characterized

86: by the presence of pulsar wind nebulae (PWNe) at their centers.

87: The SNR has a large-diameter ($\sim 107$~arcmin) with low X-ray surface 

88: brightness (Seward, Schmidt, \& Slane 1995) and a

89: partial shell morphology in the radio with an apparent breakout into lower 

90: density material in the north (Sieber, Salter, \& Mayer 1981).

91: A kinematic distance of $1.4 \pm 0.3$~kpc has been derived based on the 

92: association of an HI shell with the SNR (Pineault et al. 1993).

93: 

94: \rosat\ observations of \snr\ (Seward et al. 1995) 

95: reveal a center-filled morphology as well

96: as a faint compact source \psr\ located at the peak of the central 

97: brightness distribution (Figure 1). \asca\ observations show 

98: that the diffuse central

99: emission is nonthermal, presumably corresponding to a wind nebula

100: driven by an active neutron star for which \psr\ may be the

101: counterpart (Slane et al. 1997 -- hereafter S97). The power law index 

102: of this nonthermal

103: emission increases with distance from the center, consistent with synchrotron

104: losses of particles injected from a central source, and there is also

105: weak evidence for a thermal component with $kT \sim 0.2$~keV,

106: presumably corresponding to emission from the SNR shell. 

107: No radio counterpart to \psr\ is 

108: identified in a list of compact sources by Pineault et al. (1993). 

109: 

110: The EGRET source \gamm\ (earlier designated \gam)

111: lies in the direction of \snr\

112: (Brazier et al. 1998), and the 95\% confidence contour for the location 

113: is consistent with the position of \psr. The two best-established classes

114: of EGRET sources are blazars and pulsars, and the lack of variability

115: in 10 distinct observations of \gamm\ (Brazier et al. 1998)

116: argues in favor of the latter interpretation (although Tompkins 1999 presents 

117: mild evidence for variability). 

118: 

119: In this paper we report on new X-ray observations of \snr\ 

120: from the \asca\ and \xmm\ observatories in an effort to better constrain

121: the nature of \psr\ as a candidate pulsar in this SNR. The observations

122: and data reduction are described in Section 2, and the results of the

123: analysis are detailed in Section 3. We conclude with a discussion of

124: new constraints on the nature of \psr, its relationship to \snr, and

125: the extension of its spectrum to $\gamma$-ray energies for comparison

126: with \gamm.

127: 

128: %%%%%%%% Figures 1 & 2 %%%%%%%%%%%%%%%%

129: \begin{figure*}[tb]

130: %\pspicture(0,10.8)(18.5,21)

131: \pspicture(0,10.1)(18.5,21)

132: 

133: \rput[tl]{0}(0.0,21.5){\epsfxsize=8.5cm

134: \epsffile{f1.eps}}

135: 

136: \rput[tl]{0}(9.75,20.7){\epsfxsize=8.0cm

137: \epsffile{f2.ps}}

138: 

139: \rput[tl]{0}(0,12.7){

140: \begin{minipage}{8.75cm}

141: \small\parindent=3.5mm

142: {\sc Fig.}~1.---

143: Radio image of \snr\ (from Pineault et al. 1997) with PSPC contours.

144: The location of the \asca\ observation described here is indicated by the

145: GIS field of view in the southern portion of the remnant, and the \xmm\

146: pointing is indicated by the approximate MOS field of view in the center.

147: The position of \psr\ is indicated by a cross. The position of \gamm\ is

148: indicated by an X, with the approximate 95\% error region shown as a dashed

149: circle. (Note that the actual error region is slightly irregular and overlaps

150: with the X-ray source position.)

151: \end{minipage}

152: }

153: 

154: \rput[tl]{0}(9.7,12.7){

155: \begin{minipage}{8.75cm}

156: \small\parindent=3.5mm

157: {\sc Fig.}~2.---

158: \asca\ GIS2 spectrum from the southern rim of \snr. The top histogram

159: corresponds

160: to the best-fit two component model described in the text. The bottom histogram

161: represents only the power law component.

162: \end{minipage}

163: }

164: 

165: \endpspicture

166: \end{figure*}

167: 

168: \section{Observations}

169: 

170: \subsection{\asca\ Observations}

171: As an extension of our initial \asca\ studies that covered only the

172: central portions of \snr, we observed the southern shell of the SNR

173: to investigate the spectrum of the nonthermal emission far from the

174: remnant center, and to search for additional evidence of thermal emission

175: from the shell. The 60~ks observation was carried out on 10 February 1997. The 

176: approximate field of view of the GIS detectors is illustrated in Figure 1.

177: The SIS detectors were operated in 2-CCD mode, and cover a smaller portion

178: of the SNR with a lower effective exposure. The associated SIS spectra contain

179: relatively fewer counts, and we concentrate on the GIS data exclusively.

180: 

181: We used standard screening procedures to generate a cleaned set of events

182: for each detector, and extracted spectra from the region indicated

183: in Figure 1. Spectra were grouped to contain a minimum of 25 counts in each

184: spectral bin. Background spectra were extracted from the same regions of

185: each detector using data from the available \asca\ ``blank sky'' fields.

186: While these fields are at high Galactic latitude, and thus free of diffuse

187: emission from the Galactic Plane, they are adequate for \snr\ which lies

188: $\sim 10^\circ$ above the Plane.  A weighted effective area

189: file was generated for each extended spectral region using the 

190: {\tt ascaarf} task in the {\tt ftools} analysis package.

191: The number of counts in each

192: spectrum, after background subtraction, was $\sim 3000.$ The spectrum 

193: from GIS 2 is plotted in Figure 2 along with model fits described below.

194: 

195: 

196: \subsection{\xmm\ Observations}

197: CTA~1 was observed for 40~ks with \xmm\ on February 21, 2002 with a pointing

198: direction centered on \psr.  The EPIC cameras were operated in

199: imaging mode, with the medium and thin filters used for the two MOS detectors

200: and the pn detector, respectively. The small window mode with

201: 6~ms time resolution was used for the pn detector in order to 

202: search for pulsed emission from the source.

203: Data were filtered to eliminate intervals of high background from soft

204: proton flares, using the Scientific Analysis Software (SAS) package, and 

205: events were restricted to the energy range between 100 and 15000 eV.

206: The resulting exposure times were $\sim 30$~ks ($\sim 26$~ks) for

207: the MOS (pn) detectors, and the observed count rates for \psr\ were

208: $1.3 \times 10^{-2} (2.9 \times 10^{-2}) {\rm\ cnt\ s}^{-1}$ in the

209: 0.5--10~keV band.

210: The image of \psr\ obtained with the MOS 1 detector is shown as an inset

211: to Figure 4.  

212: 

213: Spectra of \psr\ were extracted from each EPIC detector and

214: regrouped to contain a minimum of 25 counts in each bin. Background spectra

215: were taken from regions adjacent to \psr\ in order to include similar

216: contributions from the diffuse emission from the synchrotron nebula.

217: Spectral response (RMF) and effective area (ARF) files were created for

218: each spectrum using the SAS tasks {\tt rmfgen} and {\tt arfgen}.

219: The spectrum from the MOS~1 detector is shown in Figure 3.

220: 

221: \section{Analysis}

222: \subsection{Spectral Analysis}

223: 

224: The \asca\ GIS spectra were modeled initially with an absorbed 

225: power law spectrum. The 

226: fit results in significant residual emission below $\sim 1$~keV (Figure 2)

227: which is adequately fit with an additional

228: component for a thermal plasma in collisional ionization equilibrium

229: (Raymond \& Smith 1977).

230: While the best-fit column density is consistent with results

231: from earlier \asca\ measurements (S97), 

232: the uncertainty is large. We thus fix this at the previously determined 

233: value of $N_H \sim 2.8 \times 10^{21}{\rm\ cm}^{-2}$.

234: The best-fit results, summarized in Table 1, yield a spectral

235: index of $2.3^{+0.3}_{-0.4}$ and a plasma temperature of

236: $kT = 0.28^{+0.3}_{-0.08}$~keV. The temperature is consistent with 

237: that inferred from earlier \asca\ observations. 

238: 

239: We first attempted to fit the EPIC spectra of \psr\ using an absorbed 

240: power law model. This model provides a good fit, but yields a column

241: density of $\sim 10^{20}{\rm\ cm}^{-2}$, which is much lower than that

242: for \snr. Fixing the absorption at the SNR value yields residual soft

243: emission which is well fit by a blackbody of $kT \sim 130$~eV with an

244: emitting area of radius less than 1~km. The power law index 

245: $\Gamma \sim 1.5$ is similar to that of the Crab Pulsar. The best-fit 

246: spectral parameters are listed in Table 2. The blackbody flux comprises 

247: roughly 20\% of the unabsorbed flux from the source. 

248: Replacing the blackbody component with a neutron star atmosphere model

249: in a $\sim 10^{12}$~G magnetic field (Pavlov et al. 1995; model kindly

250: provided by Slava Zavlin) reduces the temperature of the thermal component

251: to $T \sim 8\times 10^{5}$~K. 

252: 

253: %%%%%%%% Figures 3 & 4 %%%%%%%%%%%%%%%%

254: \begin{figure*}[t]

255: %\pspicture(0,10.7)(18.5,21)

256: \pspicture(0,10.5)(18.5,21)

257: 

258: \rput[tl]{0}(0.0,21.0){\epsfxsize=8.3cm

259: \epsffile{f3.ps}}

260: 

261: \rput[tl]{0}(9.75,21.0){\epsfxsize=8.0cm

262: \epsffile{f4.eps}}

263: 

264: \rput[tl]{0}(0,12.7){

265: \begin{minipage}{8.75cm}

266: \small\parindent=3.5mm

267: {\sc Fig.}~3.---

268: \xmm\ pn spectrum of \psr\ in \snr. The upper histogram corresponds to the

269: power law plus neutron star atmosphere fit described in the text, while the

270: lower histogram represents only the power law component.

271: \end{minipage}

272: }

273: 

274: \rput[tl]{0}(9.7,12.7){

275: \begin{minipage}{8.75cm}

276: \small\parindent=3.5mm

277: {\sc Fig.}~4.---

278: \xmm\ MOS 1 brightness profile for \psr. Also shown is the profile for a

279: similar compact source in the supernova remnant G347.3--0.5 (histogram)

280: as well as the predicted profile for a point source (dashed curve).

281: There appears to be evidence for extended emission from \psr. The

282: MOS~1 + MOS~2 image is shown as an inset, with a scale bar at the lower

283: left. \psr\ is the brightest source, near the center of the image.

284: \end{minipage}

285: }

286: 

287: \endpspicture

288: \end{figure*}

289: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

290: 

291: 

292: \subsection{Spatial Analysis}

293: 

294: The nominal position of \psr\ derived from the EPIC observations is 

295: ${\rm RA}_{2000} = 00^{\rm h}07^{\rm m}02.2^{\rm s}$, 

296: ${\rm Dec}_{2000} = +73^\circ 03^\prime 07^{\prime \prime}$, with an

297: uncertainty of $\sim 4^{\prime \prime}$. 

298: We find evidence that the source may be slightly extended. In Figure 4 

299: we plot the radial brightness profile extracted from the MOS~1 data, and

300: compare it to that for the point source in G347.3--0.5, obtained from 

301: archival \xmm\ data. The point sources were located

302: at the center of their respective fields, and the two profiles have 

303: been normalized to yield the same number of counts within a 15~arcsec 

304: radius. Also plotted is the radial count distribution for the on-axis 

305: point spread function (PSF) of the MOS1 telescope in the 0.75-2.25~keV 

306: energy range profile, from the \xmm\ Users' Handbook (Issue 2.1), normalized 

307: in the same manner.

308: The profile for \psr\ appears to exceed that from the comparison profiles

309: beyond a radius of $\sim 10$~arcsec. This may

310: be an indication of substructure in the inner portions of the extended

311: nebula. Such structure is now known to be commonplace in PWNe, as dramatically

312: illustrated by high resolution X-ray images of the Crab and Vela pulsars,

313: for example (see Slane 2003). 

314: We note that the spectrum for \psr\ is harder than that for

315: the point source in G347.3--0.5 (Lazendic et al. 2003), while the 

316: energy-dependent point spread function (PSF) of the \xmm\ mirrors is smaller

317: for higher energies. The apparent extent is thus not an artifact of the

318: telescope PSF.

319: 

320: To further investigate this issue, we extracted spectra from smaller regions

321: surrounding \psr\ and repeated the spectral analysis. We find that the ratio

322: of the blackbody flux to the powerlaw flux increases for the smaller regions,

323: as would be expected if the powerlaw component is extended while the blackbody

324: component is pointlike. This result is not highly significant, however, and

325: observations with higher spatial resolution will be required to confirm such 

326: structure in \snr. Such observations have recently been carried out with

327: \chandra\ (PI J. Halpern).

328: 

329: \subsection{Timing Analysis}

330: 

331: We used the EPIC pn data to search for pulsed emission from \psr. The 

332: SAS task {\tt epchain} was used to reprocess events to correct a timing

333: error inherent in some \xmm\ data, and events were then corrected 

334: to the solar system barycenter with the task {\tt barycen}, using the 

335: coordinates derived above. A total of 1055 counts in the energy band

336: 0.3-10~keV were extracted from within a radius of 9.6~arcsec, centered on 

337: \psr, with an estimated $483 \pm 34$ originating from the source 

338: (the rest being background events). A Fast-Fourier Transform was applied to the 

339: binned light curve using $\sim 8.4 \times 10^6$ bins spanning frequencies 

340: up to 83.3~Hz. No significant evidence of pulsations is observed in the 

341: resulting power spectrum. The maximum power in the FFT (aside from values 

342: corresponding to the $\sim 40$~ks duration of the observation) was 28.4, from 

343: which we derive an upper limit of 61\% for the pulsed fraction of a sinusoidal 

344: profile in the quoted frequency range (Vaughan et al. 1994).

345: 

346: \section{Discussion}

347: 

348: Under the Sedov solution (Sedov 1959), 

349: assuming a shell-like morphology similar to the radio emission,

350: with a nominal radius of 40~arcmin, and 

351: extrapolating our spectral results to the entire SNR, we infer an age of

352: $1.3 \times 10^4 D_{1.4}$~yr (where $D_{1.4}$ is the distance in units of

353: 1.4~kpc), and an ambient density of $1.7 \times 10^{-2}

354: D_{1.4}^{-1/2}{\rm\ cm}^{-3}$ assuming that the GIS field covers roughly

355: 15\% of the emission from the shell. These values, while rather uncertain

356: given the small region observed, are in good agreement with estimates based 

357: on earlier \asca\ observations (S97) and indicate a typical SNR that has 

358: evolved in a somewhat low density environment. We note that the density

359: value corresponds to the southern region covered by the ASCA observation;

360: the northern region, which extends beyond the nominal radius used here,

361: presumably corresponds to a lower density.

362: 

363: The new \asca\ observations show that the diffuse synchrotron emission in

364: \snr\ extends to the outer boundaries of the SNR, at least in the

365: southern shell region. Although the telescope

366: resolution does not allow us to identify the outer extent with precision,

367: it is clear that the radius of this component is larger than the $\sim 18$

368: arcmin ($7.2 D_{1.4}$ pc) extent estimated in S97. 

369: The PWN is quite large relative to the size of the surrounding shell,

370: although larger PWNe are known (Gaensler, Dickel, \& Green 2000).

371: The spectral (photon) index 

372: in the outer regions of the nebula may be slightly steeper than the central 

373: value of $2.1 \pm 0.1$, but not significantly so. This suggests little 

374: synchrotron

375: aging of the X-ray emitting electrons, which is indicative of a low

376: magnetic field. This is similar to what is observed for the PWN powered

377: by PSR~B1509--58 (Gaensler et al. 2002).

378: 

379: The spectral results derived for \psr\ are consistent with the interpretation

380: that this source is the pulsar that powers the surrounding synchrotron

381: nebula. The luminosity of the power law component from the compact

382: source is $L_x = 4.7

383: \times 10^{31} d_{1.4}^2 {\rm\ erg\ s}^{-1}$. This is low for a pulsar

384: young enough to have an X-ray luminous SNR, but not exceedingly so: 

385: PSR~B1853+01 is only a factor of two brighter, and is also associated

386: with an observed SNR (W44). The luminosity of the PWN in \snr\ 

387: is $5.6 \times 10^{33} d_{1.4}^2 {\rm\ erg\ s}^{-1}$ (S97)

388: which means the point source comprises roughly 1\% of the nonthermal

389: emission, a value similar to that for pulsars in 3C~58, G54.1+0.3,

390: and G292.0+1.8. Assuming a spin-down power $\dot E \sim 10^3 L_x

391: \approx 5 \times 10^{34} d_{1.4}^2 {\rm\ erg\ s}^{-1}$ from 

392: the crude correlation derived from X-ray studies of other pulsars 

393: (Seward \& Wang 1988, Becker \& Tr\"umper 1999), \psr\ apparently has 

394: sufficient energy to power the nebula.

395: The derived spectral index for \psr\ is similar to that of the Crab Pulsar.

396: If the source is indeed a pulsar with the inferred $\dot E$, this would

397: differ considerably from the empirical relationship between spectral 

398: index and spin-down power suggested by Gotthelf (2003).

399: 

400: The position of \psr\ is offset from the geometrical center of the SNR (defined

401: here by the circular portion of the radio shell, ignoring the blowout region in

402: the north). Using the age above, this offset corresponds to a transverse

403: velocity of $\sim 450 {\rm\ km\ s}^{-1}$ which is reasonable for a pulsar.

404: We note, however, that the apparent nonuniform density of the surrounding ISM

405: introduces considerable uncertainty in estimating the center of the SNR.

406: 

407: 

408: %%%%%%%%%%%%%%%%%% TABLE 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%

409: \begin{center}

410: Table 1: Spectral Parameters

411: \scriptsize

412: \begin{tabular}{cccc}\\ \hline

413: & \snr & \multicolumn{2}{c}{\psr} \\

414: & \asca\ GIS &

415: \multicolumn{2}{c}{\xmm\ EPIC} \\

416: 

417: & PL + Thermal & PL + BB & PL + NSA \\ \hline

418: 

419: $N_H ({\rm\ cm}^{-2}$) &&

420: $2.8 \times 10^{21}$ (fixed) \\ \hline

421: 

422: $\Gamma$ (photon) & $2.3^{+0.2}_{-0.4}$ &

423: $1.5 \pm 0.2$ & $1.6 \pm 0.2$\\

424: 

425: $F_x ({\rm\ erg\ cm}^{-2}{\rm\ s}^{-1})^a$ & $4.8 \times 10^{-12}$ &

426: $2.0 \times 10^{-13}$ & $2.0 \times 10^{-13}$ \\ \hline

427: 

428: $kT$ (keV) & $0.28^{+0.3}_{-0.08}$ &

429: $0.136 \pm 0.012$ & $0.053 \pm 0.004$ \\

430: 

431: $F_x ({\rm\ erg\ cm}^{-2}{\rm\ s}^{-1})^a$ & $1.1 \times 10^{-11}$ &

432: $3.3 \times 10^{-14}$ & $3.5 \times 10^{-14}$ \\

433: 

434: $R$(km) & & $0.63 D_{1.4}$ & 10 (fixed) \\ \hline

435: 

436: $\chi^2/dof$ & 437.2/474 & 117.5/116 & 112.2/116 \\ \hline

437: \\

438: \multicolumn{3}{l}{a) Unabsorbed flux (0.5-10 keV)} \\

439: \end{tabular}

440: \end{center}

441: \normalsize

442: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

443: 

444: The X-ray spectral index of $\sim 1.5$ is similar to that for

445: \gamm\ in $\gamma$-rays ($1.6 \pm 0.2$). We find that extrapolation

446: of the spectrum for \psr\ using $\Gamma = 1.4$ predicts a flux of

447: $4 \times 10^{-7}{\rm\ photons\ cm}^{-2}{\rm\ s}^{-1}$ for $E > 100$~MeV, in

448: good agreement with the value measured by EGRET

449: ($4.6 \pm 0.6 \times 10^{-7}{\rm\ photons\ cm}^{-2}{\rm\ s}^{-1}$). Thus,

450: the spectrum of \psr\ is capable of powering the EGRET flux of \gamm\

451: with no spectral break between the two energy bands. This, combined with the

452: fact that there is no brighter X-ray source in the EGRET error box, and that

453: both \psr\ and \gamm\ possess properties indicative of a pulsar,

454: strongly suggests that these sources are one and the same.

455: 

456: The temperature of the blackbody component in the spectrum of \psr\ is too

457: high, and the emitting area is too low, for this to be interpreted as

458: cooling from the surface of a young neutron star

459: (e.g. Kaminker et al. 2002). Rather, this component

460: could be associated with the heating of polar cap regions of

461: the neutron star by instreaming particles accelerated in the magnetosphere

462: (e.g. Pavlov et al. 2000, Zavlin et al. 2002).

463: The lack of observed pulsations from such an emission structure is not

464: worrisome at the upper limit levels derived here. Strong

465: gravitational bending of the emission from near the surface of the NS

466: can easily reduce the pulsed fraction to such levels (e.g. Psaltis, 

467: \"Ozel, \& DeDeo 2000).

468: More sensitive timing measurements are obviously of considerable

469: interest in an effort to detect pulsations that could potentially secure

470: the identification between \psr\ and \gamm, as well as provide

471: the spindown properties of the pulsar. Deep radio observations have yet

472: to reveal pulsations, with a flux density upper limit of 0.8~mJy at

473: 606~MHz (Lorimer et al. 1998), although recent observations have uncovered

474: much fainter pulsars in the young SNRs G292.0+1.8 (Camilo et al. 2002a)

475: and 3C~58 (Camilo et al. 2002b), both of which are also faint pulsed

476: X-ray sources (Hughes et al. 2003, Murray et al. 2002).

477: Deep X-ray timing observations of \psr\ are currently planned with the 

478: \chandra\ HRC (PI S. Murray).

479: 

480: Alternatively, the thermal emission may be the result of cooling from the

481: entire NS surface, viewed through a light element atmosphere. The inferred

482: temperature for such a model, described above, is below that expected for

483: a NS whose age is 10-20~kyr (as estimated for the SNR) assuming cooling

484: by the modified Urca process, but consistent with models in which the

485: direct Urca process becomes active in sufficiently massive stars

486: (Kaminker et al. 2002, Yakovlev et al. 2002). Such models would predict

487: a mass of $\sim 1.44 M_\odot$ for \psr.

488: 

489: 

490: \section{Conclusions}

491: We have used \asca\ and \xmm\ data to study \snr\ and the apparently associated

492: compact source at its center, \psr. The diffuse nonthermal emission extending

493: from the central regions of the SNR is consistent with previous 

494: interpretations of \snr\ as a composite SNR containing a large PWN.

495: The faint thermal emission detected along the southern shell 

496: is consistent with expectations for a moderate age SNR expanding into

497: a low density environment, as might be encountered at the high Galactic

498: latitude of \snr.

499: 

500: The nonthermal nature of the spectrum from \psr\ strongly suggests

501: that this source is a NS with an actively emitting magnetosphere,

502: and the existence of a blackbody-like component in the spectrum indicates

503: the presence of either hot polar caps or cooling of the neutron star surface

504: through an atmosphere.

505: The X-ray luminosity of the source is reasonable for an active pulsar,

506: though rather low for one of such a young age. We do not detect pulsations

507: from the source, and place an upper limit of 61\% on the pulsed fraction,

508: which is lower than that for some known X-ray emitting pulsars, but

509: is not overly restrictive on the interpretation of this as the

510: pulsar powering the synchrotron nebula in \snr.

511: 

512: We find evidence that the power law component from \psr\ may be 

513: extended, suggesting that there could be inner structure to the 

514: known large PWN, and indicating that the pulsar is still supplying energy

515: to the larger nebula.  This inner structure could be similar to 

516: the jets or toroidal structures that have been recently identified for several 

517: pulsars using high-resolution \chandra\ observations.  

518: 

519: Finally, we find additional evidence for a connection between \psr\ and the 

520: EGRET source \gamm\ due to the similarities in the spectra of the two 

521: sources. This strongly, though not conclusively, suggests that

522: \psr\ is a pulsar that is emitting both X-rays and $\gamma$-rays.

523: More sensitive searches for pulsed emission from the source, through deeper

524: X-ray observations as well as radio observations, are of considerable

525: interest in an effort to confirm this picture.

526: 

527: \acknowledgements{We thank Jan Vrtilek and Dima Yakovlev for helpful 

528: discussions related to this study, Jasmina Lazendic for help with processing

529: the \xmm\ data, and Slava Zavlin and George Pavlov for providing the NS

530: atmosphere model.  POS acknowledges support from NASA

531: contract NAS8-39073 and grants NAG5-4803 and NAG5-10017.}

532: 

533: 

534: \begin{thebibliography}{}

535: \bibitem{bt}

536: Becker, W. \& Tr\"umper, J. 1999, A\&A, v. 341, 803

537: \bibitem{brazier}

538: Brazier, K. T. S., Reimer, O., Kanbach, G., \& Carraminana, A. 1998, MNRAS, 295,

539: 819

540: \bibitem{camilo}

541: Camilo, F. et al. 2002, ApJ, 571, L41

542: \bibitem{bmg}

543: Gaensler, B. M. et al. 2002, ApJ, 569, 878

544: \bibitem{evg}

545: Gotthelf, E. V. 2003, ApJ, 591, 36

546: \bibitem{jph}

547: Hughes, J. P., Slane, P. O., Park, S., Roming, P. W. A.,

548: \& Burrows, D. N. 2003, ApJ, 591, L139

549: \bibitem{kam}

550: Kaminker, A. D., Yakovlev, D. G., \& Gnedin, O. Y. 2002, A\&A, 383, 1076

551: \bibitem{laz}

552: Lazendic, J. et al. 2003, ApJ (in press)

553: \bibitem{ssm}

554: Murray, S. S., Slane, P. O., Seward, F. D., Ransom, S. M.,

555: \& Gaensler, B. M. 2002, ApJ, 568, 226

556: \bibitem{pav1}

557: Pavlov, G. G., Shibanov, Y. A., Zavlin, V. E., \& Meyer, R. D. 1995, 

558: in: The Lives of the Neutron Stars, eds. M. A. Alpar, U. Kiziloglu, \& 

559: J. van Paradijs (Kluwer: Dordrecht) p.71.

560: \bibitem{pav2}

561: Pavlov, G. G., Zavlin, V. E., Achenbach, J., Trumper, J., \& Sanwal, D. 2000,

562: ApJ, 531, L53

563: \bibitem{pine1}

564: Pineault, S., Landecker, T. L., Madore, B., \& Gaumont-Guay 1993, AJ, 105, 1060

565: \bibitem{pine2}

566: Pineault, S., Landecker, T. L., Swerdlyk, C. M., \& Reich, W. 1997, A\&A,

567: 324, 1152

568: \bibitem{psaltis}

569: Psaltis, D., \"Ozel, F., \& DeDeo, S. 2000, ApJ, 544, 390

570: \bibitem{raymond}

571: Raymond, J. C. \& Smith, B. W. 1977, ApJS, 35, 419

572: \bibitem{sedov}

573: Sedov, L. I. 1959, Similarity and Dimensional Methods in Mechanics (New York:

574: Academic)

575: \bibitem{sss}

576: Seward, F. D., Schmidt, B., and Slane, P. 1995, ApJ, 453, 284

577: \bibitem{sw}

578: Seward, F. D. \& Wang, Z.-R. 1988, ApJ, 332, 199

579: \bibitem{sieber}

580: Sieber, W., Salter, C. J., \& Mayer, C. J. 1981, A\&A, 74, 361

581: \bibitem{slane}

582: Slane, P. 2003, in The Universe Viewed in Gamma-Rays, eds. R. Enomoto, 

583: M. Mori, and S. Yanagita (Universal Academy Press: Tokyo), p.3. 

584: \bibitem{s97}

585: Slane, P. O., Seward, F. D., Bandiera, R., Torii, K., \& Tsunemi, H. 1997, ApJ,

586: 485, 221

587: \bibitem{tompkins}

588: Tompkins, W. 1999, Ph.D. thesis, Stanford Univ

589: \bibitem{vaughan}

590: Vaughan, B. A. et al. 1994, ApJ, 435, 362

591: \bibitem{yak}

592: Yakovlev, D. G., Kaminker, A. D., Haensel, P., \& Gnedin, O. Y. 2002, A\&A, 389,

593: L24

594: \bibitem{zavlin}

595: Zavlin, V. E., Pavlov, G. G., Sanwal, D., \& Manchester, R. N. 2002, ApJ, 569,

596: 894

597: 

598: \end{thebibliography}

599: 

600: \end{document}

601: 

602: