1: \documentclass[12pt,preprint]{aastex}
2:
3: \newcommand{\psr}{PSR J0538+2817}
4:
5: \shorttitle{Proper motion of \psr}
6: \shortauthors{Kramer et al.}
7:
8: \slugcomment{Revised version of MS17462}
9:
10: \begin{document}
11:
12:
13: \title{Proper motion, age and initial spin period of
14: PSR J0538+2817 in S147}
15:
16: \author{M.~Kramer\altaffilmark{1},
17: A.G. Lyne\altaffilmark{1},
18: G.~Hobbs\altaffilmark{2,1},
19: O.~L\"ohmer\altaffilmark{3},
20: P.~Carr\altaffilmark{1},
21: C.~Jordan\altaffilmark{1},
22: A.~Wolszczan\altaffilmark{4}}
23: \altaffiltext{1}{University of Manchester, Jodrell Bank Observatory, Macclesfield, Cheshire SK11 9DL, UK}
24: \altaffiltext{2}{Australia Telescope National Facility, CSIRO, P.O.~Box~76, Epping, NSW~1710, Australia}
25: \altaffiltext{3}{Max Planck Institut f\"ur Radioastronomie, Auf dem H\"ugel 69, D-53121 Bonn, Germany}
26: \altaffiltext{4}{Department of Astronomy and Astrophysics, Pennsylvania State
27: University, University Park, PA 16802, USA}
28: \email{mkramer@jb.man.ac.uk}
29:
30: \begin{abstract}
31: We present results of timing observations of the 143-ms pulsar
32: J0538+2817 that provide a proper motion measurement which clearly
33: associates the pulsar with the supernova remnant S147. We measure a
34: proper motion of 67$_{-22}^{+48}$ mas yr$^{-1}$,
35: %($\mu_\lambda=-41(3)$ mas yr$^{-1}$, $\mu_\beta=47(57)$ mas yr$^{-1}$),
36: implying a transverse
37: velocity of $v= 385^{+260}_{-130}$ km s$^{-1}$. We derive an age of
38: the pulsar and S147 of only $30\pm4$ kyr which is a factor
39: of 20 times less
40: than the pulsar's characteristic age of $\tau_c = 620$ kyr. This age
41: implies an initial spin period of $P_0=139$ ms, close to the present
42: pulse period and a factor of several larger than what is usually
43: inferred for birth periods. Implications for recent X-ray detections
44: of this pulsar are discussed.
45: \end{abstract}
46:
47: \keywords{pulsars: general; pulsars: individual (J0538+2817);
48: stars: neutron; supernova remnants}
49:
50:
51: \section{Introduction}
52: \label{intro}
53:
54: The determination of the birth properties of pulsars is of crucial
55: importance in understanding both the physics of core-collapse
56: supernovae as well as the population and evolution of a radio
57: pulsar. The initial spin period of pulsars, $P_0$, is particularly
58: difficult to measure as it requires the knowledge of the pulsar's
59: age, $\tau$, and its spin-down behaviour. It is usually assumed that
60: the observed evolution of the spin-frequency $\nu = 1/P$ can be
61: described by a power-law $\dot{\nu} \propto -\nu^n $, where $n$ is
62: the so-called {\em braking index}. The initial spin period can then
63: be calculated from
64: \begin{equation}
65: \label{eqn:p0}
66: P_0 = P \left[ 1 - \frac{n-1}{2}\;\frac{\tau}{\tau_c}
67: \right]^{\frac{1}{n-1}}
68: \end{equation}
69: where $\tau_c=P/2\dot{P}$ is the {\em characteristic age} of the
70: pulsar. The characteristic age is a good estimator for the true
71: age of the pulsar, $\tau_c\approx \tau$,
72: under the assumptions that $P_0\ll P$
73: and that the spin-down is due to magnetic braking
74: for which $n=3$.
75:
76: Since the use of the characteristic age, rather than the true age, can
77: lead to considerable errors, it is desirable to have an independent
78: age measurement. The age of a supernova remnant (SNR) that originated
79: in the same explosion as the pulsar, can serve as such an estimator
80: but the only pulsar for which the age of an associated SNR is clearly
81: known, is the Crab pulsar. From the observation of the explosion in
82: A.D.1054 and a measured braking index of $n=2.51(1)$, the initial spin
83: period is computed to be $P_0=$19 ms (Lyne et al.~1993).\nocite{lps93}
84: It may also be possible to measure a proper motion of a pulsar associated
85: with a SNR. If this transverse motion is directed away from the
86: center of the SNR, then this is strong evidence that the pulsar is
87: genuinely associated with the remnant. It also allows one to
88: determine the age of both the pulsar and the SNR by comparing the
89: present offset from the center with its speed. As SNRs typically fade
90: away after $\sim100,000$ yr, pulsars genuinely associated with SNRs
91: are necessarily young. Such pulsars, however, often show rotational
92: instabilities in the form of glitches and/or timing noise (e.g.~Lyne
93: et al.~1995)\nocite{lps95} which makes the measurement of proper
94: motion via timing observations usually a difficult
95: task. Interferometric measurements offer a solution but so far it has
96: only been possible for PSR B1951+32 in CTB 80 for which an initial
97: spin period of $P_0 =27(6)$ ms was derived (Migliazzo et
98: al.~2002\nocite{mgb+02}).
99:
100: In this work we present a timing proper motion measurement for the
101: 143-ms pulsar J0538+2817 which was found within the boundaries of
102: G180.0$-$1.7, also called S147 (Anderson et al.~1996\nocite{acj+96}).
103: The SNR S147 has a prominent shell structure with a radius of $\theta
104: = 83(3)$ arcmin (Sofue et al.~1980). With an estimated age of $\sim
105: 100$ kyr (e.g.~Kundu et al.~1980\nocite{kafh80}) it is considered to
106: be one of the oldest well-defined SNRs in the Galaxy, although other
107: authors derived much younger ages (e.g.~20 kyr, Sofue et
108: al.~1980). Therefore, S147 has been studied rather extensively at
109: radio frequencies (e.g.~F\"urst et al.~1982\nocite{frb+82}). Very
110: recently, X-ray observations with CHANDRA have revealed a structure
111: interpreted as a pulsar wind nebula (Romani \& Ng 2003\nocite{rn03})
112: while XMM-Newton observations revealed pulsed X-ray emission from the
113: surface (McGowan et al.~2003\nocite{mkz+03}). In the following we
114: describe our timing observations and data
115: analysis which leads to a proper motion measurement which clearly
116: associates the pulsar with SNR S147. We hence obtain accurate
117: estimates of the age of the pulsar and derive its initial spin
118: period. The results are finally compared to the recent X-ray
119: observations.
120:
121:
122: \section{Observations and Data Analysis}
123:
124: The observations were made with the 100-m Effelsberg radiotelescope
125: from April 1994 at 1410 MHz and with the 76-m Lovell telescope at
126: Jodrell Bank from March 1996 at 606 MHz and 1400 MHz. At both
127: telescopes, two circularly polarized signals were mixed down to
128: intermediate frequencies, detected and incoherently de-dispersed in
129: hardware using filterbanks before sub-integrations of 15 sec
130: (Effelsberg) and 60 sec (Jodrell Bank) were written to disk for
131: off-line processing. All data were time-stamped with clock information
132: provided by local H-maser clocks which were later synchronized to UTC
133: by using signals from the Global Positioning System (GPS) satellites.
134: Details of the observing systems can be found in Anderson et
135: al.~(1996) and Hobbs et al.~(2003a).\nocite{hlk+03}
136:
137:
138: The data from both telescopes were first processed to a common time
139: resolution of 238.3 $\mu$s before being subjected to the same template
140: matching procedure which used identical templates to determine the pulse
141: times-of-arrival (TOAs). Since PSR J0538+2817 shows mode-changing,
142: exhibiting two distinct profiles which differ in the relative
143: height of the two prominent components (Anderson et al.~1996), we
144: applied a proven technique that uses two different, carefully created
145: templates for the given modes (see e.g.~Stairs et al.~2000).
146: Any possible remaining effects due to the occurrence of
147: mixed-mode profiles are accounted for by adopting relatively large
148: minimum errors of $\Delta t\ge180\mu$s (Effelsberg) and $\Delta
149: t\ge130\mu$s (Jodrell Bank), respectively. The success of this
150: procedure is notable by the fact that no arbitrary clock offsets
151: between the telescopes or profile modes were needed in the further
152: timing analysis.
153:
154: In a first step of the timing analysis using the DE200 planetary
155: ephemerides, the dispersion measure (DM) was determined by fitting for
156: DM and a simple spin-down model to data obtained with the Lovell
157: telescope at 606 and 1400 MHz over a small period of time. The DM value
158: was then held fixed for the subsequent analysis.
159: Like many
160: other young pulsars, PSR J0538+2817 shows long-term timing noise
161: visible in the timing residuals. Hobbs et al.~(2003a)\nocite{hlk+03}
162: developed a new technique to remove such timing noise, allowing one
163: to reliably separate the signature due to proper motion. Applying
164: this technique, the transverse proper motions have been determined for
165: more than 300 pulsars and excellent agreement is found for those
166: pulsar where interferometric measurements are available (Hobbs et
167: al.~2003a,b)\nocite{hlk03,hlk+03}. We applied this technique,
168: before fitting to a spin-down model which now included proper motion and a
169: second period derivative. Iterative tests were made to check
170: the robustness of the obtained solution which also studied the effects
171: of omitting and including different parts of the data in the analysis.
172: We do not believe that the non-zero second period derivative
173: is due to magnetospheric braking, therefore we do not attempt to calculate
174: a braking index, $n$ (see Hobbs et al.~2003a).
175:
176: As the pulsar lies close to the ecliptic with an ecliptic latitude,
177: $\beta$, of only $\beta=4^\circ.9$,
178: position and proper motion measurements in the
179: latitudinal direction are necessarily much less accurate than those in
180: ecliptic longitude. In order to minimize covariances between the
181: astrometric parameters, fits were made in ecliptic coordinates using
182: the software package {\tt TEMPO}\footnote{\tt
183: http://www.atnf.csiro.au/research/pulsar/timing/tempo}, resulting
184: in post-fit residuals shown in Fig.~1. The final
185: spin and astrometric parameters are presented in
186: Table~\ref{tab:parms}. Quoted uncertainties are derived from twice the
187: formal {\tt TEMPO} error and standard Monte-Carlo simulations.
188: Procedures for the latter are detailed in Lange et
189: al.~(2001)\nocite{lcw+01} and results obtained for proper motion are
190: shown in Fig.~\ref{fig:mc}.
191:
192:
193:
194: \section{Results and Discussion}
195:
196: As can be seen from Table~\ref{tab:parms} and Fig.~\ref{fig:mc}, we
197: have obtained a proper motion measurement for PSR J0538+2817. Whilst
198: the movement in ecliptic longitude is measured to high significance,
199: the uncertainty for the proper motion in latitude is much larger.
200: With a probability of 78\%, the proper motion is positive in
201: latitudinal direction, making the pulsar moving in the right quadrant
202: on the sky to be consistent with a movement away from the center of
203: S147. We measure a position angle of P.A.$={311^\circ}^{+28}_{-56}$. The
204: obtained proper motion is consistent with a comparison of the
205: interferometric position obtained by Anderson et al.~(1996)
206: and our present timing position.
207:
208: Using the measured values, we can compute the pulsar's motion in the
209: past and compare it to the location of the center of the SNR. Fitting
210: a circular shape
211: to the radio contours of S147 over a frequency range from 430
212: MHz to 4750 MHz (Kundu et al.~1980\nocite{kafh80}, Sofue et
213: al.~1980\nocite{sfh80}, Angerhofer \& Kundu 1981\nocite{ak81},
214: F\"urst et
215: al.~1982\nocite{frb+82}, F\"urst \& Reich 1986\nocite{fr86})
216: we determine the SNR center to be at
217: $\lambda_{SNR}=85^\circ.57(1)$ and $\beta_{SNR}=4^\circ.44(1)$ which
218: agrees well with the center determined by Sofue et al.~(1980) at 4750
219: MHz only. We mark the central position on the 2.7-GHz map
220: obtained by F\"urst \& Reich~(1986) shown in Fig.~\ref{fig:map}. We
221: also mark the position of the pulsar 30 kyr ago as computed from the
222: measured proper motion ($\lambda_{30k}=85^\circ.57(3)$,
223: $\beta_{30k}=4^\circ.5(5)$). It is clear that the previous position of
224: the pulsar agrees very well with the center of the SNR, strongly
225: suggesting that the pulsar was born in the same explosion that created
226: S147, about 30,000 years ago.
227:
228: Assuming that the pulsar was born at the center of S147, we use the
229: offset of the pulsar from this position, $\Delta \Theta =
230: 2.18(4)\times 10^6 $ mas and the measured proper motion to determine a
231: true age of pulsar and remnant of $\tau= 33_{-9}^{+17}$ kyr. The
232: uncertainty in this age is dominated by the error in the proper motion
233: measurement in latitudinal direction. In order to derive more
234: accurate estimate, we can use the offset and motion in longitudinal
235: direction only. This results in an age of $\tau_\lambda = 30\pm4$ kyr,
236: confirming that the kinematic age of the pulsar is dramatically smaller
237: than the characteristic age of $\tau_c= 618$ kyr. The only assumption
238: made in deriving this age is that the pulsar was born in the center of
239: the SNR.
240:
241: Anderson et al.~(1996) already discussed the possible
242: association of PSR J0538+2817 with S147 in detail and concluded that
243: an association is plausible. They based their arguments on the
244: proximity of the pulsar to the SNR center and the consistent distance
245: estimates for both pulsar and SNR. The distances estimated for S147
246: ranging from of 0.8 to 1.6 kpc are well consistent with the dispersion
247: measure distance of the pulsar, i.e. $d = 1.2$ kpc as
248: derived from the NE2001 model (Cordes \& Lazio 2002\nocite{cl02}).
249: At this distance, our proper motion measurement yields a
250: transverse speed of $v=385_{-130}^{+260}$ km s$^{-1}$ which is in
251: excellent agreement with mean observed velocity of pulsars (Lyne \&
252: Lorimer 1993). Given the position of the pulsar within the SNR
253: boundaries and a normalized angular distance of only $\delta = \Delta
254: \Theta/ \Theta = 0.43(2)$ away from the center, and
255: its location in the Galactic anti-center region where we
256: find only a rather sparse population of known SNRs and pulsars, an
257: association seemed indeed very likely. This is now confirmed by the
258: pulsar's movement away from the center.
259:
260: Further independent evidence is available that the characteristic age
261: is much larger than the true age of the pulsar.
262: Firstly, while some authors estimate a blast wave age of
263: S147 in a range from 80 kyr to 200 kyr (e.g.~Kundu et al.~1980), Sofue
264: et al.~(1980) estimated an age of only 20 kyr. Even though these age
265: estimates for SNRs depend also on the density of the ambient
266: interstellar medium and are known to be highly uncertain (e.g.~F\"urst
267: \& Reich 1986), all values are lower than $\tau_c$,
268: and the latter estimate agrees indeed very well with our
269: findings. Given this young age of the SNR, the well defined shell
270: structure appears less surprising. If the explosion occurred in a
271: low-density, hot stellar wind cavity blown up by the progenitor star,
272: the expansion will not be describable by a Sedov-phase but will be
273: free until it reaches the cavity boundaries
274: %The optical filaments indicate that this has happened some time ago.
275: (E.~F\"urst, private communication). At a distance of 1.2~kpc, the
276: observed SNR radius corresponds to $\sim 30$ pc, implying an
277: expansion velocity of 1000 km s$^{-1}$ in the free expansion
278: phase. Future CO observations may be able
279: to reveal the wind cavity.
280:
281:
282: Secondly, we have access to polarization information for the radio
283: emission of PSR J0538+2817 obtained by Mitra et al.~(2003), who
284: measured a rotation measure of RM=$-7\pm12$ rad
285: m$^{-2}$. The pulsar exhibits an extremely high degree of
286: polarization of 92(2)\%. While this is not uncommon, it is
287: usually found in young sources (e.g.~Morris et al.~1981),
288: again supporting a young age of the pulsar.
289:
290: Finally, we can compare our results to recent X-rays observations.
291: Romani \& Ng (2003)\nocite{rn03} reported the discovery of a faint
292: nebula surrounding the pulsar. They interpret this
293: nebula as an equatorial torus, supporting the association of pulsar
294: and S147. In their calculation they assumed a pulsar age of 100 to 200
295: kyr, but we can derive consistent results with our smaller derived age
296: of $\tau=30$ kyr with a somewhat smaller ISM density. The orientation
297: of their modelled torus with a symmetry axis located at position angle
298: $\Psi=154(6)^\circ$ (measured N through E) and $\xi=100(6)^\circ$
299: (into plane of the sky) is consistent with the position angle swing
300: measured for our polarization data. Applying a rotating vector model
301: (Radhakrishnan \& Cooke 1969)\nocite{rc69a}, we find a best fit for a
302: magnetic inclination of $\alpha \sim 95^\circ$ and an impact angle of
303: $\sigma\sim 2^\circ$ although the uncertainties are considerable. These
304: findings support Romani \& Ng's interesting conclusion derived from
305: the torus geometry that the velocity direction appears to be nearly
306: aligned with the rotation axis of the pulsar. However, with a pulse
307: width of about $40^\circ$ and such a geometry, one may expect to
308: observe also emission from the opposite magnetic pole. This is not the
309: case, but the emission beam may not be circular and/or the emission
310: beam could be patchy (Lyne \& Manchester 1988). We also note that
311: the bilateral symmetry axis of S147 appears to be very nearly
312: parallel to both the symmetry axis of the torus and the proper motion
313: direction of the pulsar. While the remnant is somewhat asymmetric
314: about the orthogonal axis, the center of the explosion could in
315: principle deviate from the SNR center which is determined by
316: the sharp circular rim. However, it is reasonable to assume that the
317: explosion center is located on the symmetry axis between the
318: determined SNR center and the current pulsar position. In such a
319: case, the age of the pulsar would be even younger, and our conclusions
320: would be essentially unaffected.
321:
322:
323:
324:
325:
326: Equally intriguing is the comparison of our results with those by
327: McGowan et al.~(2003) \nocite{mkz+03} who detected pulsed X-ray
328: emission from PSR J0538+2817. Their data are well explained by
329: blackbody radiation from a heated polar cap. However, using the
330: dispersion measure distance the derived temperature is significantly
331: higher than predicted by standard cooling theories. This discrepancy
332: can be reduced by using atmospheric fits, but with a pure-H
333: non-magnetized atmosphere McGowan et al.'s result still falls above
334: the expected temperature. In their analysis McGowan et al.~used the
335: characteristic age of the pulsar, and the high temperature indeed
336: suggests that the true age of the pulsar is much smaller. Using our
337: derived age of $\tau=30$ kyr, the temperature from the atmospheric fit
338: falls well below the standard cooling curve. However, this fit also
339: produces a distance which is much smaller than the dispersion measure
340: estimate. A proper modelling of the observed X-ray spectra needs to
341: include the surface magnetic field of $B=7.3\times 10^{11}$ G, which
342: may change the results considerably (e.g. Pavlov et al.~2001, Zavlin
343: \& Pavlov 2002).\nocite{zp02}\nocite{pzs+01} This is important
344: since with such a low temperature one would be forced to
345: consider the presence of exotic cooling processes. Interestingly, a
346: similar conclusion has been reached recently by Slane et
347: al.~(2002)\nocite{shm02} for PSR J0205+6449 associated with 3C58 (but
348: see also Yakovlev et al.~2002\nocite{ykhg02}).
349:
350: Pavlov et al.~(2002)\nocite{pzst02} studied CHANDRA data of the
351: neutron star 1E 1207.4$-$5209 which is likely to be associated with
352: SNR PKS 1209$-$51/52. They also find a characteristic age which is
353: much larger than that estimated for the SNR, suggesting a long birth
354: period of the neutron star. Romani \& Ng already already pointed out
355: that the initial spin period of PSR J0538+2817 is likely to be large
356: and close to the present value. Using our age estimate we derive an
357: initial spin period of $P_0 = 139.6$ ms ($n=3$), which is insensitive
358: to the actual choice of the braking index as the ratio of
359: $\tau/\tau_c=0.05$ is very small (e.g. $P_0 = 139.6 $ ms for $n=10$
360: and $P_0 =139.8$ ms for $n=0.5$). It has been attempted to estimate
361: the initial spin period for seven radio pulsars (see Migliazzo et
362: al.~2002 and references therein) but apart from the results for the
363: Crab (based on the known age of the SNR and pulsar) and PSR B1951+32
364: in CTB 80 (based on kinematic age derived from proper motion), all
365: estimates rely on less certain ages estimated for associated SNRs. It
366: appears that the initial period of those pulsars is $P_0\le 60$ ms. An
367: exception is PSR J1124$-$5916 with $P_0\approx 90$ ms (Camilo et
368: al.~2002)\nocite{cmg+02}. All estimated birth periods are already
369: significantly larger than what is expected from core-collapse theory
370: of massive stars and it appears difficult to explain even spin periods
371: of a few tens milliseconds (see Heger et al.~2003\nocite{hwls03} for a
372: recent review). The estimated initial spin period for PSR J0538+2817
373: is much larger still, placing strong constraints on the origin of
374: birth kicks imparted on neutron stars as discussed by Romani \& Ng
375: (2003).
376:
377: In summary, we have measured the proper motion of PSR J0538+2817 which
378: clearly associates the pulsar with the SNR S147. From the separation
379: of the pulsar from the SNR center we determine an age of $\tau
380: =30\pm4$ kyr, making the pulsar significantly younger than is
381: indicated by its characteristic age of $\tau_c=618$ kyr. This implies
382: a large initial spin period of $P_0=139$ ms. The implied possibility
383: of exotic cooling should be revisited after a magnetized atmosphere
384: has been fitted to the recent X-ray data.
385:
386:
387: \section*{Acknowledgements}
388:
389: We thank Ernst F\"urst for useful disucssions and the 2.7-GHz map of
390: S147. We also thank Werner Becker and Slava Zavlin for helpful
391: discussions and are grateful to an anonymous referee for his comments.
392:
393: %\bibliographystyle{apj}
394: %\bibliography{journals,mod_jfb,modrefs,psrrefs,crossrefs}
395: %\bibliography{journals,modrefs,psrrefs,crossrefs}
396:
397: \begin{thebibliography}{}
398:
399: \bibitem[Anderson {\rm et~al.}~{1996}]{acj+96}
400: Anderson~S., Cadwell~B.~J., Jacoby~B.~A., Wolszczan~A., Foster~R.~S.,
401: Kramer~M., 1996, ApJ, 468, L55
402:
403: \bibitem[Angerhofer \& Kundu~{1981}]{ak81}
404: Angerhofer~P.~E., Kundu~M.~R., 1981, AJ, 86, 1003
405:
406: \bibitem[Camilo {\rm et~al.}~{2002}]{cmg+02}
407: Camilo~F., Manchester~R.~N., Gaensler~B.~M., Lorimer~D.~L., Sarkissian~J.,
408: 2002, ApJ, 567, L71
409:
410: \bibitem[{Cordes} \& {Lazio}~{2002}]{cl02}
411: {Cordes}~J.~M., {Lazio}~T.~J.~W., 2002, ApJ, submitted ({astro-ph/0207156})
412:
413: \bibitem[{F\"urst} \& Reich~{1986}]{fr86}
414: {F\"urst}~E., Reich~W., 1986, A\&A, 163, 185
415:
416: \bibitem[{F\"urst} {\rm et~al.}~{1982}]{frb+82}
417: {F\"urst}~E., Reich~W., Beck~R., Hirth~W., Angerhofer~P.~E., 1982, A\&A, 115,
418: 428
419:
420: \bibitem[Heger {\rm et~al.}~{2003}]{hwls03}
421: Heger~A., Woosley~S.~E., Langer~N., Spruit~H.~C., 2003, in Maeder~A.,
422: Eenens~P., eds, Stellar Rotation, Proc.~of IAU Symposium S215.
423: \newblock PASP, San Francisco, in press (astro-ph/0301374)
424:
425: \bibitem[Hobbs {\rm et~al.}~{2003}]{hlk+03}
426: Hobbs~G., Lyne~A.~G., Kramer~M., Martin~C.~E.~J.~C., 2003a, MNRAS, in prep.
427:
428: \bibitem[Hobbs, Lyne \& Kramer~{2003}]{hlk03}
429: Hobbs~G., Lyne~A.~G., Kramer~M., 2003b, in M.~Bailes~D.~Nice~.~S.~T., ed, Radio
430: Pulsars (ASP Conf.~Ser.),
431: \newblock PASP, San Francisco, in press (astro-ph/0211001)
432:
433: \bibitem[Kundu {\rm et~al.}~{1980}]{kafh80}
434: Kundu~M.~R., Angerhofer~P.~E., {F\"urst}~E., Hirth~W., 1980, A\&A, 92, 225
435:
436: \bibitem[Lange {\rm et~al.}~{2001}]{lcw+01}
437: Lange~C., Camilo~F., Wex~N., Kramer~M., Backer~D., Lyne~A., Doroshenko~O.,
438: 2001, MNRAS, 326, 274
439:
440:
441: \bibitem[Lyne \& Manchester~{1988}]{lm88}
442: Lyne~A.~G., Manchester, R.~N., 1988, MNRAS, 234, 477
443:
444: \bibitem[Lyne, Pritchard \& Smith~{1993}]{lps93}
445: Lyne~A.~G., Pritchard~R.~S., Smith~F.~G., 1993, MNRAS, 265, 1003
446:
447: \bibitem[Lyne, Pritchard \& Shemar~{1995}]{lps95}
448: Lyne~A.~G., {Pritchard} R.~S., {Shemar} S.~L., 1995, JApA, 16, 179
449:
450:
451: \bibitem[McGowan {\rm et~al.}~{2003}]{mkz+03}
452: McGowan~K.~E., Kenea~J.~A., Zane~S., Cordova~F.~A., Cropper~M., Ho~C.,
453: Sasseen~T., Vestrand~W.~T., 2003, ApJ, in press (astro-ph/0303380)
454:
455: \bibitem[Migliazzo {\rm et~al.}~{2002}]{mgb+02}
456: Migliazzo~J.~M., Gaensler~B.~M., Backer~D.~C., Stappers~B.~W., van~der
457: Swaluw~E., Strom~R.~G., 2002, ApJ, 567, L141
458:
459: \bibitem[Morris {\rm et~al.}~{1981}]{mgs+81}
460: Morris~D., Graham~D.~A., Seiber~W., Bartel~N., Thomasson~P., 1981, A\&AS, 46,
461: 421
462:
463: \bibitem[{Pavlov} {\rm et~al.}~{2001}]{pzs+01}
464: {Pavlov}~G.~G., {Zavlin}~V.~E., {Sanwal}~D., {Burwitz}~V., {Garmire}~G.~P.,
465: 2001, ApJ, 552, L129
466:
467: \bibitem[{Pavlov} {\rm et~al.}~{2002}]{pzst02}
468: {Pavlov}~G.~G., {Zavlin}~V.~E., {Sanwal}~D., {Tr{\" u}mper}~J., 2002, ApJ, 569, L95
469:
470: \bibitem[Radhakrishnan \& Cooke~{1969}]{rc69a}
471: Radhakrishnan~V., Cooke~D.~J., 1969, Astrophys. Lett., 3, 225
472:
473: \bibitem[{Romani} \& {Ng}~{2003}]{rn03}
474: {Romani}~R.~W., {Ng}~C.-Y., 2003, ApJ, 585, L41
475:
476: \bibitem[Slane, Helfand \& Murray~2002]{shm02}
477: {Slane}~P.~O., {Helfand}~D.~J., {Murray}~S.~S., 2002, ApJ, 571, L45
478:
479: \bibitem[Sofue, F\"urst \& Hirth~1980]{sfh80}
480: {Sofue}~Y., {F\"urst}~E., {Hirth}~W., 1980, PASJ, 32, 1
481:
482: \bibitem[Stairs, Lyne \& Shemar 2000]{sls00}
483: {Stairs} I.~H., {Lyne} A.~G., {Shemar}, S.~L., 2000, Nature,
484: 406, 484
485:
486: \bibitem[Yakolev {\rm et~al.}~{2002}]{ykhg02}
487: Yakolev~D.~G., Kaminker~A.~D., Haensel~P., Gnedin~O.~Y., 2002, A\&A, 389, L24
488:
489: \bibitem[{Zavlin} \& {Pavlov}~{2002}]{zp02}
490: {Zavlin}~V.~E., {Pavlov}~G.~G., 2002, in Neutron Stars, Pulsars, and Supernova
491: Remnants, eds.~W.~Becker, H.~Lesch, J.~Tr\"umper, MPE-Report 278, p.~263
492:
493: \end{thebibliography}
494:
495:
496: \clearpage
497:
498:
499: \begin{deluxetable}{ll}
500: \tablecaption{\label{tab:parms}Timing parameter of PSR J0538+2817}
501: \tablecolumns{2}
502: \tablewidth{0pc}
503: \tablehead{
504: \colhead{Parameter} & \colhead{Value}\\}
505: \startdata
506: Ecliptic longitude, $\lambda$ (deg) & 85.232553(3) \\
507: Ecliptic latitude, $\beta$ (deg) & 4.93582(4) \\
508: R.A.\tablenotemark{a} (J2000) & 05 38 25.0623\\
509: DEC\tablenotemark{a} (J2000) & 28 17 09.1\\
510: Epoch (MJD) & 51086.0 \\
511: Spin frequency, $\nu$ (s$^{-1}$) & 6.985276348019(5) \\
512: First derivative, $\dot{\nu}$ ($10^{-15}$ s$^{-2}$) & $-$179.04753(6) \\
513: Second derivative, $\ddot{\nu}$ ($10^{-24}$ s$^{-3}$) & $-$0.637(2) \\
514: Spin period, $P$ (ms) & 143.1582589118(1) \\
515: First derivative, $\dot{P}$ ($10^{-15}$) & 3.669452(1) \\
516: Dispersion Measure, DM (cm$^{-3}$ pc) & 39.814(6) \\
517: Proper motion, $\mu_\lambda$ (mas yr$^{-1}$) & $-$41(3) \\
518: Proper motion, $\mu_{\beta}$ (mas yr$^{-1}$) & 47(57) \\
519: Proper motion, composite\tablenotemark{b} (mas yr$^{-1}$) & $67^{-22}_{+48}$ \\
520: TOA Span (MJD) & 49453 -- 52712 \\
521: Number of TOAs & 249 \\
522: Timing RMS ($\mu$s) & 144.2 \\
523: %Characteristic age, $\tau$ (kyr) & 618\\
524: %Distance$^{\dagger \dagger}$, $d$ (kpc) & 1.2 \\
525: \enddata
526: \tablecomments{The uncertainties in the last
527: quoted digits are given in parenthesis.}
528: \tablenotetext{a}{calculated from ecliptic coordinates}
529: \tablenotetext{b}{quoted value is median of asymmetric distribution}
530:
531: \end{deluxetable}
532:
533: \clearpage
534:
535: \begin{figure}[ht]
536: \epsscale{0.8}
537: \plotone{f1.eps}
538:
539: \caption{ \label{fig:res}
540: Timing residuals obtained after applying a spin-down model
541: listed in Table 1 with a fit for proper motion (bottom)
542: and with a corresponding proper motion set to zero (top).
543: The solid line in the upper plot shows the expected
544: behaviour for residuals in the latter case.}
545: \end{figure}
546:
547: \clearpage
548:
549:
550: \begin{figure}[ht]
551: \epsscale{0.8}
552: \plotone{f2.eps}
553:
554: \caption{ \label{fig:mc}
555: Results of Monte-Carlo simulations to determine
556: the uncertainties in proper motion measurements.}
557: \end{figure}
558:
559: \clearpage
560:
561:
562: \begin{figure}[ht]
563: \epsscale{0.8}
564: \plotone{f3.eps}
565:
566: \caption{
567: \label{fig:map}Radio image of S147 obtained at
568: 2.7 GHz by F\"urst \& Reich~(1986) convolved to a 5' beam. Contours
569: are in steps of 25 mK $T_B$ beginning at $27.5$ mK $T_B$. The
570: right scale indicates Galactic coordinates. The
571: determined center of the SNR is marked by a circle, the position of
572: PSR J0538+2817 by a 'X'. The position of the pulsar 30,000 years ago
573: (uncertainties are marked) agrees well with the centre of the SNR.}
574: \end{figure}
575:
576:
577: \end{document}
578:
579: