0709.3614/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: \usepackage{epsfig}
3: 
4: \shortauthors{Chang, Konopelko, \& Cui} 
5: \shorttitle{PWN Associations with Unidentified TeV $\gamma- $Ray Sources}
6: 
7: \begin{document}
8: 
9: \title{Search for Pulsar Wind Nebula Association with Unidentified TeV $\gamma- $Ray Sources}
10: 
11: \author{Chulhoon~Chang, Alexander~Konopelko, and Wei~Cui}
12: \affil{Department of Physics, Purdue University, West Lafayette, IN 47907}
13: \altaffiltext{1}{Email: chang40@physics.purdue.edu, akonopel@purdue.edu, cui@physics.purdue.edu}
14: 
15: \begin{abstract}
16: 
17: Many of the recently discovered TeV $\gamma- $ray sources are associated with pulsar wind nebulae
18: (PWNe). In fact, they represent the most populous class of Galactic sources at TeV energies.
19: In addition, HESS has also discovered, in a survey of the Galactic plane, a population of TeV 
20: $\gamma- $ray sources that are still without definitive counterparts at longer wavelengths. 
21: For a number of these sources, a pulsar is an evident association, which is often located within 
22: an extended region of the TeV $\gamma- $ray emission. These particular HESS sources are promising
23: candidates for yet not resolved pulsar wind nebulae. Here we have undertaken a systematic search 
24: for X-ray counterparts of the sources, using the archival {\it Chandra} data, within the spatial 
25: bounds of the unidentified HESS sources. A number of X-ray sources have been detected in the 
26: {\it Chandra} fields. Two of them,  CXOU~J161729.3-505512 and CXOU~J170252.4-412848, are of a 
27: special interest because of their excellent positional coincidence with the pulsars 
28: PSR~J1617-5055 and PSR~J1702-4128, respectively. The first source is extended, with a bright 
29: core of $2.6\arcsec$ (FWHM) in radius but the emission can be seen up to roughly $20\arcsec$. 
30: The second one is much fainter and detected only with marginal 
31: significance ($4.6 \sigma$). It might also be slightly extended, although the situation 
32: is quite uncertain due to very limited statistics. The analysis of the archival {\it Chandra} 
33: data for a middle-aged pulsar (PSR~J1913+1011) does not reveal any statistically significant 
34: excess at and around the position of the pulsar or the center of gravity of its plausible
35: TeV $\gamma- $ray counterpart (HESS~J1912+101). We discuss the implications of the results.
36: 
37: \end{abstract}
38: 
39: \keywords{acceleration of particles ---  radiation mechanisms: non-thermal --- pulsars: 
40: individual (PSR J1617-5055, PSR J1702-4128, PSR J1913+1011) --- X-rays: general}
41: 
42: \section{Introduction}
43: 
44: The charged particles accelerated in the vicinity of a pulsar flow out into the supernova (SN) 
45: ejecta and form a shock. The shock may further accelerate particles to relativistic speeds. 
46: These particles interact with the surrounding medium to produce a pulsar wind nebula (PWN),
47: which is often observable at radio and X-ray wavelengths. Around the youngest, most energetic 
48: pulsars, the radio emitting parts of these nebulae are rather amorphous, whereas the X-ray 
49: emitting regions can be highly structured \citep{GaenslerSlane2006}. The unprecedented spatial 
50: resolution of {\it Chandra} has made it possible to resolve these structures. The spatial 
51: extent and morphology of X-ray nebulae can vary from a toroidal (or arc-like) nebulae with 
52: perpendicular jets to the prominent tails behind the moving pulsars, indicative of bow 
53: shocks. The review of recent {\it Chandra} observational results on PWNe was given in 
54: \citet{KargaltsevPavlov2008}.
55: 
56: It was widely believed that PWNe could be sources of very high-energy (VHE) $\gamma- $ray 
57: emission. The emission may arise from inverse Compton scattering of low-energy photons by 
58: the relativistic electrons, while the X-ray emission may be associated with the synchrotron 
59: radiation from the same electrons. The best example is the Crab Nebula, which was established 
60: as a source of pulsed GeV $\gamma- $rays by EGRET \citep{Nolan1993} and of steady TeV $\gamma $ 
61: rays by ground-based \v{C}erenkov detectors \citep{Crab2006}. The TeV emission is thought to 
62: originate in its PWN \citep{deJagerHarding1992}. Recently, the HESS array of four imaging 
63: \v{C}erenkov telescopes have detected many TeV sources, a number of which are evidently 
64: associated with PWNe. 
65: Such associations usually rest on positional and morphological match of VHE $\gamma- $ray 
66: sources with known PWNe. It is worth noting that in many cases the pulsar and associated PWN 
67: are significantly offset from the center of the TeV source. The offset could be attributed to 
68: the interaction between the PWN and the SNR \citep{Blondinetal2001}.
69: 
70: A survey of the inner part of the Galactic Plane was performed with HESS between $\pm$30$^\circ$ 
71: in longitude and $\pm$3$^\circ$ in latitude relative to the Galactic Center. Fourteen previously
72: unknown, extended sources were detected with high significance \citep{aha2005,aha2006b}. 
73: Recently, the survey region has been extended to cover the 
74: longitude range out to +60$^\circ$ \citep{aha2008a}, leading to the discovery of five new TeV
75: sources. The HESS sources are likely Galactic in origin, given their concentration around the 
76: Galactic plane. The $\gamma- $ray flux of these sources varies from about 3\% to 25\% of that 
77: of the Crab Nebula. Most of them show rather hard energy spectra and are spatially extended, 
78: with the angular size of a few tenths of a degree. Some of the sources have fairly 
79: well-established counterparts at longer wavelengths, based exclusively on positional 
80: coincidence, but others have none at all. A number of proposals have been made on the nature of 
81: these unidentified TeV sources. At present, PWNe and shell-type SNRs are considered to be most 
82: probable, although other possibilities also exist. 
83: 
84: Observationally, a PWN often manifests itself as extended X-ray emission around a pulsar. The 
85: positional coincidence of such an object with an unidentified TeV source would strongly support 
86: the scenario that X-rays and TeV $\gamma- $rays are both powered by the PWN. For example, 
87: HESS~J1804-21, one of the brightest and most extended sources discovered in the HESS Galactic
88: plane survey, contains the Vela-like pulsar PSR~J1803-2137 at distance of about $3.8\, \rm kpc$ 
89: \citep{aha2005}. The pulsar has a spin-down age of about $16\, \rm kyr$ and a spin-down power 
90: of about $2.25\times 10^{36}\, \rm  erg\, s^{-1}$, which makes it among the top 20 pulsars as 
91: ranked by spin-down flux and thus a candidate to produce an X-ray emitting 
92: PWN \citep{Manchesteretal2005}. Recently, using an 
93: archival 30-ks {\it Chandra} observation of PSR~J1803-2137, \citet{CuiKonopelko2007} discovered
94: significantly extended X-ray emission around the pulsar (see also \citet{Kargaltsevetal2007}), 
95: indicating the presence of a PWN. The emission has a very hard spectrum, which is well 
96: described by a power law with photon index of about 1.2. The spatially-averaged flux is 
97: $\sim$$10^{-13}\, \rm erg\, cm^{-2}\, s^{-1}$ in the 0.3--10 keV band. The PWN is offset from 
98: the center-of-gravity of HESS~J1804-21 by about 10\arcmin. It is elongated, perpendicular to 
99: the pulsar's proper motion, 
100: suggesting that its X-ray emission probably emerges from a torus associated with the termination 
101: shock in the equatorial pulsar wind. Another example is HESS~J1809-193 and its plausible 
102: association with the pulsar PSR J1809-1917. \citet{Kargaltsev2007} has detected extended X-ray 
103: emission around the pulsar with {\it Chandra}, which is attributed to a PWN. As in the case of 
104: HESS~J1804-21, the PWN is also significantly offset from the center-of-gravity of HESS~J1809-193 
105: (by about 8\arcmin). 
106: 
107: There are several more HESS sources that also seem to be associated with pulsars but no 
108: corresponding PWNe have been seen yet. HESS~J1912+101 contains 
109: PSR~J1913+1011 \citep{aha2008b}, which is a rather old pulsar with a spin period of 36 ms 
110: and a spin-down age of $1.7 \times 10^5$~yrs, at a distance of 4.5 kpc. A radio pulsar was 
111: also detected in the Parkes Pulsar Survey, PSR~J1702-4128, near the tip of a tail-like 
112: extension of HESS~J1702-420 \citep{Krameretal2003}. It is a younger pulsar with a spin period 
113: of 182 ms and a spin-down age of about 55000 yrs and is located at a distance of about 5.2 kpc. 
114: Finally, PSR~J1617-5055 is located near the edge of HESS~J1616-508. It is an X-ray emitting, 
115: young pulsar with a spin period of 69~ms and a spin-down age of about 8000 yrs and is at 
116: a distance of about 6.5 kpc \citep{Toriietal1998}. 
117: In this work, we have undertaken a systematic search for extended X-ray emission associated with
118: PSR~J1702-4128, PSR~J1913+1011, and PSR~J1617-5055, using data from the archival {\it Chandra} 
119: observations, to take advantage of the superior spatial resolution of {\it Chandra}. We have 
120: analyzed all {\it Chandra} fields that contain the pulsars. The results from imaging and 
121: spectral analyses are reported here.
122: 
123: \section{Data Analysis and Results}
124: 
125: The {\it Chandra} data were extracted from the archival observations of RCW~103 (ObsID \#970), 
126: PSR~J1617-5055 (ObsID \#6684), PSR~J1702-4128 (ObsID \#4603), and PSR J1913+1011 (ObdID \#3854) 
127: with total exposure times of about 19~ks, 57~ks, 10.5~ks, and 19~ks, respectively. The data 
128: were all taken with the ACIS detector. Table~1 summarizes some of the key characteristics of 
129: these observations.
130: 
131: The data were reduced and analyzed with the standard {\it CIAO} analysis package (version 3.4), 
132: along with {\it CALDB 3.4.0}. For ObsID \#907, we followed the CIAO Science Threads\footnote{See 
133: http://asc.harvard.edu/ciao/threads/index.html} to prepare, filter, and reprocess the Level 1 
134: data to produce Level 2 data for subsequent analyses. For all other observations, we simply 
135: started with the Level 2 data that were derived from the archive, because they already 
136: incorporated the updated calibrations.
137: 
138: \subsection{Imaging Analysis}
139: 
140: We carried out a search for discrete sources in the 0.5--10~keV energy band with 
141: {\it celldetect}. This tool utilizes a sliding detection cell algorithm. The detection cell 
142: size was chosen to match the width of the local point spread function (PSF). Here we adopted
143: default values for all key parameters. For example, the size of the detection cell corresponds 
144: to the 80\% encircled energy area of the PSF and the signal-to-noise ratio threshold was set 
145: to 3. We excluded spurious detections in the vicinity of significant exposure variations, such 
146: as the detector edges or chip gaps etc. We estimated the statistical significance of each 
147: detection using the statistical method of \citet{LiMa1983}. For that we used the output of 
148: {\it celldetect}, which includes the sizes of the source and background regions, as well as 
149: the total number of counts extracted from the regions. Table~\ref{table2} summarizes 
150: the sources detected with a statistical significance greater than 4$\sigma$. 
151: 
152: For ObsID \#970, a total of ten sources were detected, including two sources on each of the I2, 
153: I3, and S4 chips and four sources on the S2 chip, respectively. CXOU~J161729.3-505512 coincides 
154: spatially with PSR~J1617-15055, which makes it a plausible counterpart of HESS~J1616-508. The 
155: source shows no apparent extension in this data set. However, a subsequent observation of 
156: longer exposure (ObsID \#6684) enabled us to resolve it as an extended X-ray source. To better 
157: identify the diffuse emission we ran {\it wavdetect} on this data set, and found that the 
158: effective $\sigma$ of the derived count distribution is about 2.36 times that of the PSF at the 
159: location of the source, indicating that the emission is significantly extended. 
160: Figure~\ref{fig2} shows an expanded view of the vicinity of CXOU~J161729.3-505512. From the 
161: raw count images, we constructed the radial profile of the source. Fitting the profile with a 
162: Gaussian function (plus a constant background) that is convolved with the local PSF resulted 
163: in a source extension of about 5.28$\pm$0.05 pixels (FWHM), which corresponds to about 
164: $2.6\arcsec$. However, the image also seems to show a more extended component of the emission.
165: To minimize the effects of asymmetry, we made a linear profile of the emission along right 
166: ascension, as shown in Figure~\ref{fig1}, with the PSF overlaid for comparison. We can clearly 
167: see that the emission extends up to roughly $20\arcsec$ in both directions. 
168: 
169: In ObsID \#4603, we detected one source on each of the I2 and S3 chips but observed no apparent
170: X-ray emission around the position of PSR~J1702-4128. Only after we lowered the signal-to-noise 
171: threshold used by {\it celldetect} down to 1.5, an X-ray source (CXOU~J170252.4-412848) emerged. 
172: To evaluate the statistical significance of the detection, we extracted source counts from a 
173: circular region of 7 pixels in radius and background counts from another circular region of 20 
174: pixels in radius. Based on the Li \& Ma method, we derived a significance of about 4.6$\sigma$. 
175: We also ran {\it wavdetect} on this {\it Chandra} field and found that the 
176: effective $\sigma$ of the count distribution is about 1.46 times that of the PSF at the location 
177: of the source, suggesting that CXOU~J170252.4-412848 might also be a slightly extended source.
178: There are too few counts to meaningfully quantify the extension. The image is also shown in 
179: Figure~\ref{fig2}. The source is located almost exactly at the position of the pulsar, which 
180: makes it a plausible X-ray counterpart of the latter. 
181: 
182: For ObsID \#3854, a few sources have been detected one of them on the I2 chip, three others
183: on each of the S2 and S3 chips, and two more on the S4 chip. No apparent X-ray emission can be 
184: seen around the position of PSR J1913+1011 nor near the center-of-gravity of HESS J1912+101. 
185: We also carried out more sensitive searches, with {\it wavdetect}, for faint extended emission 
186: at both locations but failed to detect any. To derive flux upper limits, we used a circular 
187: source region of a 35$\arcsec$ radius around each of the positions and a similar circular 
188: background region nearby. The measured count rates were converted into the corresponding fluxes 
189: by adopting the line-of-sight hydrogen column densities \citep{Dickey1990} and assuming a 
190: power-law spectrum of photon index 2. The results are summarized in Table~\ref{table3}.
191: 
192: We searched through the SIMBAD and NED databases for the counterparts of all detected sources 
193: within an angular offset of less than $30\arcsec$ from each source. Besides the pulsars of 
194: interest, two of the sources, CXOU~J161723.7-505150 and CXOU~J191316.1+100902, appear to be 
195: associated with the stars, HD~146184 and HD~179712, respectively. In addition, two other 
196: sources, CXOU~J161727.9-505549 and CXOU~J191338.4+101200, are about $24\arcsec$ and $25\arcsec$ 
197: away from IRAS~16137-5048 and IRAS~19112+1007, respectively.
198: 
199: \subsection{Spectral Analysis}
200: 
201: The CIAO tool {\it specextract} was used to extract the X-ray spectrum of CXOU~J161729.3-505512, 
202: which is possibly a PWN associated with the pulsar PSR~J1617-5055. We used a circular region of
203: a 50 pixel radius ($\sim$25$\arcsec$) centered on the best-fit position of the source 
204: to extract source counts and a concentric annulus with an inner radius of 60 pixels and an 
205: outer radius of 100 pixels to extract background counts. This tool enabled to generate both the 
206: overall (source+background) and background spectra, as well as the corresponding response matrix 
207: files and auxiliary response files for subsequent spectral modeling. 
208: 
209: We modeled the spectra with {\it XSPEC} \citep{Arnaud1996}. The data points below 0.3~keV 
210: or above 10~keV were excluded an the remaining data rebinned to achieve at least 15 counts in 
211: each energy bin. The obtained spectrum can be fitted very well with an absorbed power-law model, 
212: as shown in Figure~\ref{fig3}, with a reduced $\chi^2$ about 0.94 for 660 degrees of freedom.
213: The best-fit parameters are shown in Table~\ref{table3}. We should note that the pile-up
214: effects are quite small ($\sim$3\%), thanks to the 1/4--subarray readout mode adopted.
215: 
216: We attempted the same analyses for CXOU~J170252.4-412848 (but with the source and background 
217: regions used in the imaging analysis). Only about 11 net source counts 
218: were obtained, which is too few to warrant any reliable spectral modeling. Purely for the
219: purpose of estimating the flux of the source, we fitted the data to an absorbed  
220: power-law spectrum with the photon index fixed at $\Gamma=2$. The flux is  
221: $(0.2^{+0.2}_{-0.1}) \times 10^{-13}$ erg cm$^{-2}$ s$^{-1}$ in 0.3 -- 10 keV energy band. These 
222: results are also shown in Table~\ref{table3}.
223: 
224: \section{Discussion}
225: 
226: One of the most significant recent developments in high energy astrophysics is the detection
227: of a wide variety of $\gamma- $ray sources at the very high-energy energies (above 100~GeV; 
228: see, e.g., \citet{Cui2006}, for a review). Identifying counterparts of these sources at other 
229: longer wavelengths can help unveiling the physical mechanism responsible for the VHE 
230: $\gamma- $ray emission. Among the Galactic VHE $\gamma- $ray sources PWNe represent the most
231: populous class. In many cases the association of a VHE $\gamma- $ray source with a PWN is
232: established by combining the positional and morphological similarities observed in various 
233: wavelength ranges. Further support comes from the successful modeling of their broadband 
234: multi-wavelength energy spectra. In some cases the evidence for an association is less 
235: compelling or non-existent and thus requires further observations. 
236: A strong argument in favor of plausible association of a PWN with the a VHE $\gamma- $ray 
237: source is the relatively low integral efficiency needed to convert some of the pulsar's 
238: spin-down luminosity to VHE $\gamma- $rays (e.g., see, \citet{Gallant2006}). If one defines 
239: the efficiency as $\epsilon = (4 \pi d^2 F_{\gamma})/\dot{E}$, where $F_\gamma$ is the 
240: integral VHE $\gamma- $ray flux above 300~GeV measured with the HESS instrument, it ranges
241: from roughly 1\% to 10\% for most of the known TeV sources that are associated with PWNe. 
242: Two caveats should be pointed out. First of all, the pulsar distance is often poorly 
243: determined. Secondly, this approach largely elides any substantial change in pulsar's 
244: spin-down luminosity during early evolution of the pulsar and its PWN.                 
245: 
246: In this work, we used archival {\it Chandra} data to search for X-ray counterparts of three 
247: unidentified TeV $\gamma- $ray sources that appear to be associated with known radio pulsars. 
248: We have likely detected an X-ray emitting PWN associated with PSR~J1617-5055, which could 
249: also be the underlying engine of HESS~J1616-508. Figure~3 shows all the X-ray sources 
250: detected in the vicinity of HESS~J1616-508. Among them only one, CXOU~J161729.3-505512, 
251: is spatially extended. This, coupled with its spatial coincidence with PSR~J1617-5055, 
252: is strongly in favor of a PWN origin of the X-ray emission, which also makes it a promising
253: candidate for the X-ray counterpart of HESS~J1616-508. \citet{Gallant2006} has estimated the 
254: integrated $\gamma- $ray flux in the 0.3--30 TeV band, about 
255: $3.7 \times 10^{-11} \rm \, erg \, cm^{-2} \, s^{-1}$ or about 1.3\% of the spin-down flux 
256: of the pulsar, which is quite achievable. We should note that the X-ray emission associated
257: with PSR~J1617-5055 has been seen before, most clearly with the {\it XMM-Newton} 
258: data \citep{Becker2002,Landietal2007}. However, for the first time, we have spatially 
259: resolved the emission, thanks to the superior resolution of {\it Chandra}. Combining data from 
260: {\it XMM-Newton}, {\it INTEGRAL}, and {\it BeppoSAX}, \citet{Landietal2007} derived the
261: following spectral parameters from a simple power-law fit: 
262: $N_H = 3.87^{+0.36}_{-0.28} \times 10^{22}$ $cm^{-2}$, $\Gamma = 1.42^{+0.12}_{-0.10}$,
263: and the unabsorbed 2-10 keV flux $F_X = 4.2 \times 10^{-12}\rm  \, erg\, cm^{-2} \, s^{-1}$. 
264: Their spectrum thus seems a bit softer than the one that we derived here from the 
265: {\it Chandra} observation, although the hydrogen column density and flux are in general 
266: agreement (the unabsorbed 2-10 keV flux is about 
267: $4.0 \times 10^{-12}\rm  \, erg\, cm^{-2} \, s^{-1}$ in our case). A possible explanation
268: for the discrepancy is that they might have included more diffuse emission in the surrounding
269: region, given the lower spatial resolution of the instruments used (especially {\it INTEGRAL} 
270: and {\it BeppoSAX}). \citet{Becker2002} found that about 53\% of the X-ray flux is pulsed and 
271: thus associated with the pulsar. Therefore, the PWN contribution is no more than 47\% of the
272: measured flux (or $1.6 \times 10^{-12}\rm  \, erg\, cm^{-2} \, s^{-1}$ in the 0.3--10 keV band). 
273: A lower limit on the PWN contribution can be obtained from the fraction of the X-ray emission 
274: outside the PSF. From Fig.~\ref{fig1}, we estimated that at least $\sim$10\% of the observed 
275: flux can be attributed to the PWN.
276: 
277: We also see evidence for X-ray emission associated with PSR~J1702-4128. The emission might be 
278: slightly extended. There are also other sources detected in the vicinity of HESS~J1702-420, as 
279: shown in Fig.~4, but the pulsar/PWN connection makes CXOU~J170252.4-412848 a more likely 
280: counterpart of the X-ray source. If HESS~J1702-420 is powered by the pulsar, its measured
281: VHE $\gamma- $ray flux ($1.4 \times 10^{-11}\rm \, erg\, cm^{-2} \, s^{-1}$) would correspond
282: to about 11\% of the spin-down flux of the pulsar \citep{Gallant2006}, which is rather high 
283: but not implausible. We failed to detect any X-ray emission at either the position of 
284: PSR~J1913+1011 or the center-of-gravity (CoG) of HESS~J1912+101 (although CXOU J191247.0+100948 
285: is only about 52\arcsec\ from the CoG), down to a flux level of roughly
286: $1 \times 10^{-13}$ $erg$ $cm^{-2}$ $s^{-1}$ or lower (see Table~3). If HESS~J1912+101 is
287: powered by the pulsar, the measured VHE $\gamma- $ray flux accounts for only about 1\% of the
288: spin-down flux of the pulsar \citep{aha2008b}. The non-detection of X-ray emission 
289: associated with the 
290: pulsar does not necesarily contradict a possible association between the TeV $\gamma$-ray 
291: source and the pulsar. For instance, different populations of relativistic electrons could be 
292: responsible for the synchrotron X-ray emission and the TeV $\gamma$ rays seen from this source
293: (see, e.g. \citet{Funk2007}). 
294: %Alternatively, the VHE $\gamma- $ray emission might be
295: %associated with relativistic protons, given the presence of the target gas in nearby molecular 
296: %clouds. Such a scenario does not require a high X-ray flux (which is presumably associated
297: %with synchrotron emission from relativistic electrons).
298: 
299: \section*{Acknowledgments}
300: 
301: We thank the anonymous referee for bringing to our attention the long observation of 
302: PSR~J1617-5055 (ObsID \#6684) and for many detailed and constructive comments, which have 
303: helped improve the manuscript significantly. 
304: This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by 
305: the Jet Propulsion Laboratory, California Institute of Technology, under contract with the 
306: National Aeronautics and Space Administration (NASA), and of the SIMBAD Database. We gratefully 
307: acknowledge financial support from the Department of Energy and NASA. 
308: 
309: \begin{thebibliography}{99}
310: 
311: \bibitem[Aharonian et al.(2005)]{aha2005} Aharonian, F., et al. 2005, Science, 307, 1938 
312: \bibitem[Aharonian et al.(2006a)]{Crab2006} Aharonian, F., et al. 2006a, A\&A, 457, 899
313: \bibitem[Aharonian et al.(2006b)]{aha2006b} Aharonian, F., et al. 2006b, ApJ, 636, 777
314: \bibitem[Aharonian et al.(2008a)]{aha2008a} Aharonian, F., et al. 2008a, A\&A, 477, 481
315: \bibitem[Aharonian et al.(2008b)]{aha2008b} Aharonian, F., et al. 2008b, A\&A, in Press
316: \bibitem[Arnaud(1996)]{Arnaud1996} Arnaud, K. A., 1996, Astronomical Data Analysis Software and System V, eds Jacoby G and Barnes J., ASP Conf. Series, vol. 101, p.17
317: \bibitem[Becker \& Aschenbach(2002)]{Becker2002} Becker, W., \& Aschenbach,~B., 2002, Proc. 270 WE-Heraeus Seminar, Bad Honnef, Germany, eds. W. Becker, H. Lesch, \& J. Trumper, MPE Report 278, p. 64    
318: \bibitem[Blondin et al.(2001)]{Blondinetal2001} Blondin, J. M., et al. 2001, ApJ, 563, 806
319: \bibitem[Cui(2006)]{Cui2006}Cui,~W. 2006, in Frontier Objects in Astrophysics and Particle Physics, eds. F. Giovannelli \& G. Mannocchi (Bologna: Editrice Compositori, Bologna), p.207 (astro-ph/0608042)
320: \bibitem[Cui \& Konopelko(2006,2007)]{CuiKonopelko2007} Cui,~W., \& Konopelko,~A., 2006, ApJ, 652, L109; Cui,~W., \& Konopelko,~A., 2007, ApJ, 665, L83
321: \bibitem[de Jager \& Harding(1992)]{deJagerHarding1992} De Jager, O.C., 
322: \& Harding, A.K., 1992, ApJ, 396, 161
323: \bibitem[Dickey \& Lockman(1990)]{Dickey1990} Dickey,~J.~M., \& Lockman,~F.~J., 1990, ARA\&A, 28, 215
324: \bibitem[Funk(2007)]{Funk2007} Funk,~S., 2007, Ap\&SS, 309, 11
325: \bibitem[Gaensler \& Slane(2006)]{GaenslerSlane2006} Gaensler,~B., \& Slane,~P., 2006, ARA\&A, 44, 17
326: \bibitem[Gallant(2006)]{Gallant2006} Gallant,~Y.~A., 2006, Ap\&SS, 309, 197
327: \bibitem[Kargaltsev et al.(2007)]{Kargaltsevetal2007} Kargaltsev, O., et al. 2007, ApJ, 660, 1413
328: \bibitem[Kargaltsev \& Pavlov(2007)]{Kargaltsev2007} Kargaltsev,~O., \& Pavlov,~G.~G., 2007, ApJ, 670, 655
329: \bibitem[Kargaltsev \& Pavlov(2008)]{KargaltsevPavlov2008}  Kargaltsev,~O., \& Pavlov,~G.~G., 2008, astro-ph/0801.2602 
330: \bibitem[Kramer et al.(2003)]{Krameretal2003} Kramer,~M., et al. 2003, MNRAS, 342, 1299
331: \bibitem[Landi et al.(2007)]{Landietal2007} Landi, R., et al. 2007, MNRAS, 380, 926
332: \bibitem[Li \& Ma(1983)]{LiMa1983} Li,~T.~P., \& Ma,~Y.~Q., 1983, ApJ, 272, 317
333: \bibitem[Manchester et al.(2005)]{Manchesteretal2005} Manchester,~R.~N., Hobbs,~G.~B., 
334: Teoh,~A., \& Hobbs,~M. 2005, AJ, 129, 1993
335: \bibitem[Nolan et al.(1993)]{Nolan1993} Nolan, P.~L., et al. 1993, ApJ, 409, 697
336: \bibitem[Torii et al.(1998)]{Toriietal1998} Torii, K., et al. 1998, ApJ, 494, L207
337: \end{thebibliography}
338: 
339: 
340: %Table.1
341: \begin{deluxetable}{ccccc}
342: \tablecolumns{5}
343: \tablewidth{0pt}
344: \tablecaption{Observations.}
345: \tablehead{
346: \colhead{ObsID}&\colhead{Pulsar}&\colhead{Exposure}&\colhead{Offset\tablenotemark{a}}&\colhead{ACIS Chips}\\
347: \colhead{}&\colhead{}&\colhead{Time(ks)}&\colhead{($\arcmin$)}&\colhead{}}
348: \startdata
349: \#970&PSR J1617-5055&19&6.36& I2, I3, S2, S3, S4\\
350: \#6684&PSR J1617-5055&57&0.80& I3\\
351: \#4603&PSR J1702-4128&10.5&0.67& I0, I1, I2, I3, S2, S3\\
352: \#3854&PSR J1913+1011&19&0.60& I2, I3, S1, S2, S3, S4\\
353: \enddata
354: \tablenotetext{a}{From the aim point.}
355: \end{deluxetable}
356: 
357: %Table.2
358: \begin{table}
359: \caption{Detected X-ray Sources$^{\dag}$ \label{table2}}
360: \setlength{\tabcolsep}{1mm}
361: \begin{tabular}{cllll}\hline\hline
362: ObsID&Source&Right ascension&Declination&Count rate\\
363: & &J2000&J2000&10$^{-3}$ cts s$^{-1}$\\\hline
364: &CXOU J161723.7-505150&16:17:23.73(4)&-50:51:50.3(4)&4.5$\pm$0.6\\
365: &CXOU J161727.9-505549&16:17:27.970(3)&-50:55:49.7(4)&1.5$\pm$0.1\\
366: &CXOU J161729.3-505512&16:17:29.353(1)&-50:55:12.77(1)&69$\pm$1\\
367: &CXOU J161734.3-511227&16:17:34.3(1)&-51:12:27.8(8)&1.6$\pm$0.4\\
368: &CXOU J161741.9-511006&16:17:41.90(4)&-51:10:06.3(4)&2.6$\pm$0.4\\
369: \raisebox{1.5ex}[0cm][0cm]{970/6648}&CXOU J161747.2-505709&16:17:47.28(2)&-50:57:09.1(2)&0.9$\pm$0.2\\
370: &CXOU J161820.7-510737&16:18:20.79(3)&-51:07:37.1(2)&7.4$\pm$0.6\\
371: &CXOU J161844.0-505728&16:18:44.1(1)&-50:57:29(1)&1.2$\pm$0.3\\
372: &CXOU J161849.2-510424&16:18:49.29(8)&-51:04:24.2(7)&2.4$\pm$0.5\\
373: &CXOU J161908.4-505507&16:19:08.5(1)&-50:55:08(1)&1.9$\pm$0.4\\ \hline
374: &CXOU J170238.5-413311&17:02:38.52(2)&-41:33:11.6(2)&4.5$\pm$0.6\\
375: 4603&CXOU J170252.4-412848&17:02:52.48(1)&-41:28:48.2(1)&1.1$\pm$0.2\\
376: &CXOU J170333.5-413055&17:03:33.55(5)&-41:30:55.2(6)&1.7$\pm$0.4\\ \hline
377: &CXOU J191238.0+101043&19:12:38.01(2)&+10:10:43.4(3)&8.9$\pm$0.8\\
378: &CXOU J191240.6+101755&19:12:40.63(4)&+10:17:55.2(7)&2.2$\pm$0.4\\
379: &CXOU J191245.2+100656&19:12:45.28(6)&+10:06:56.9(7)&1.3$\pm$0.3\\
380: &CXOU J191247.0+100948&19:12:47.00(3)&+10:09:48.4(4)&1.9$\pm$0.4\\
381: 3854&CXOU J191316.1+100902&19:13:16.192(4)&+10:09:02.06(4)&4.6$\pm$0.5\\
382: &CXOU J191331.9+101231&19:13:31.93(1)&+10:12:31.5(2)&1.2$\pm$0.2\\
383: &CXOU J191338.4+101200&19:13:38.41(2)&+10:12:00.2(3)&1.2$\pm$0.3\\
384: &CXOU J191351.1+101152&19:13:51.16(3)&+10:11:52.1(5)&1.9$\pm$0.4\\
385: &CXOU J191400.8+101403&19:14:00.88(7)&+10:14:03.5(9)&1.5$\pm$0.3\\
386: \hline 
387: \end{tabular}
388: 
389: \vspace*{2mm}
390: $^{\dag}$ The numbers in parentheses indicate uncertainty in the last digit. Note that only 
391: statistical uncertainties are shown. \\
392: \end{table}
393: 
394: 
395: 
396: %Table.3
397: \begin{table}
398: \caption{Properties of X-ray Emission$^{\dag}$ \label{table3}}
399: \setlength{\tabcolsep}{1mm}
400: \begin{tabular}{lccc}\hline\hline
401: Name&N$_{\rm H}$&$\Gamma$&Flux (0.3--10 keV) \\
402: &(10$^{22}$ cm$^{-2})$&&$(10^{-13} \, \rm erg\, cm^{-2} \, s^{-1}$) \\ \hline
403: PSR J1617-5055&3.3$\pm$0.3&1.1$\pm$0.1&34$^{+4}_{-7}$ \\
404: PSR J1702-4128&1.8$^{+0.6}_{-0.5}$&2.0 (fixed) &0.2$^{+0.2}_{-0.1}$ \\
405: PSR J1913+1011&1.79$^c$&2.0 (fixed) &$<$0.74$^b$ \\
406: HESS CoG$^a$&1.81$^c$ &2.0 (fixed) &$<$0.7$^b$ \\
407: \hline\hline
408: \end{tabular}
409: 
410: \vspace*{2mm}
411: $^{\dag}$ The columns are: source name, hydrogen column density, photon index, and flux. The 
412: errors shown represent 90\% confidence intervals, unless otherwise noted. \\ 
413: $^a$ The center of gravity of HESS~J1912+101 (see Figure~\ref{fig2}). \\
414: $^b$ $3\sigma$ upper limits. \\
415: $^c$ Taken from \citet {Dickey1990}\\
416: \end{table}
417: 
418: \clearpage
419: \pagebreak
420: 
421: 
422: %Fig.1
423: \begin{figure}
424: \centering
425: \includegraphics[width=0.45\linewidth]{f1a.eps} \\
426: \includegraphics[width=0.45\linewidth]{f1b.eps} \\
427: \includegraphics[width=0.45\linewidth]{f1c.eps}
428: \caption{{\it Chandra} X-ray images of the pulsar fields: PSR~J1617-5055 
429: (upper panel), PSR~J1702-4128 (middle panel), and PSR~J1913+1011 (low panel).
430: The crosses indicate the positions of the pulsars. All images were obtained 
431: from data in the 0.5--10~keV energy band. The contours correspond to the 
432: brightness levels of 0.4, 3, 25 and 200 counts (upper panel); 0.2, 0.35, and 
433: 0.5 counts (middle panel). In the lower panel, the circles (in white) show the 
434: source regions used to derive flux upper limits. All images (in logarithmic 
435: scale) have been smoothed with a Gaussian kernel of 3 pixels in radius.
436: \label{fig2}}
437: \end{figure}
438: 
439: %Fig.2
440: \begin{figure}
441: \psfig{figure=f2.eps,width=5in}
442: \caption{Linear profile of CXOU~J161729.3-505512 along right ascension (in solid histogram). 
443: It was made by projecting counts inside a horizontal $100\arcsec \times 30\arcsec$  
444: rectangular box that the source is centered on onto the axis of right ascension. Note that the 
445: data have been adaptively rebinned to obtain sufficient statistics. The background level is 
446: indicated by the long-dashed line. For comparison, the point spread function is overlaid 
447: (in dashed histogram).
448: \label{fig1}}
449: \end{figure}
450: 
451: %Fig.3
452: \begin{figure}
453: \psfig{figure=f3.eps,width=3in,angle=-90}
454: \caption{X-ray spectrum of CXOU~J161729.3-505512. The solid histogram shows 
455: the best power-law fit to the data. The residuals of the fit are shown in 
456: the bottom panel. }
457: \label{fig3}
458: \end{figure}
459: 
460: %Fig.4
461: \begin{figure}
462: \psfig{figure=f4.eps,width=3in}
463: \caption{X-ray sources in the vicinity of HESS~J1616-508. The TeV 
464: $\gamma- $ray image was adapted from \citet{aha2006b}. The positions of the 
465: detected X-ray sources are indicated by green filled circles. The red filled 
466: circle indicates the position of CXOU~J161729.3-505512, 
467: which is likely associated with PSR~J1617-5055 and its PWN. \label{fig4}}
468: \end{figure}
469: 
470: %Fig.5
471: \begin{figure}
472: \psfig{figure=f5.eps,width=3in}
473: \caption{Same as Fig.~3, but for HESS~J1702-420. The red filled circle shows 
474: the position of CXOU~170252.4-412848, which might be associated with 
475: PSR~J1702-4128 and its PWN. \label{fig5}} 
476: \end{figure}
477: 
478: \end{document}
479: