1: %\documentclass[manuscript]{aastex}
2: %\documentclass[12pt,preprint]{aastex}
3: \documentclass[apjl]{emulateapj}
4:
5: %\usepackage{amsmath}
6: %\usepackage{amssymb}
7: %\usepackage{latexsym}
8: %\usepackage{xspace}
9: %\usepackage{graphicx}
10: %\usepackage{txfonts}
11: %\usepackage{natbib}
12: %\usepackage{epsf}
13: %\usepackage{longtable}
14: %\usepackage{psfig}
15:
16: \newcommand{\PWN}{\rm{PWN}}
17: \newcommand{\PWNe}{\rm{PWNe}}
18: \newcommand{\psr}{\rm{PSR}}
19: \newcommand{\etal}{{\rm et~al.\ }}
20: \newcommand{\hess}{\textit{H.E.S.S.}}
21: \newcommand{\integral}{\textit{INTEGRAL}}
22: \newcommand{\glast}{\textit{GLAST}}
23: \newcommand{\xmm}{\textit{XMM-Newton}}
24: \newcommand{\cxo}{\textit{Chandra}}
25: \newcommand{\dede}[2]{\frac{\partial{#1}}{\partial{#2}}}
26: \newcommand{\myemail}{fabio.mattana@apc.univ-paris7.fr.}
27: \newcommand{\MT}[1]{{\sf {\bf #1}}}
28:
29: \shorttitle{Evolution of the $\gamma$- and X-ray luminosities of PWNe}
30: \shortauthors{Mattana et al.}
31:
32: \begin{document}
33: \title{On the evolution of the $\gamma$- and X-ray luminosities of Pulsar Wind Nebulae}
34:
35: \author{F. Mattana,\altaffilmark{1,2}
36: M. Falanga,\altaffilmark{3} D. G\"otz,\altaffilmark{3} R. Terrier,\altaffilmark{1} P. Esposito,\altaffilmark{2,4,5}
37: \\ A. Pellizzoni,\altaffilmark{2} A. De Luca,\altaffilmark{2,6,5} V. Marandon,\altaffilmark{1}
38: A. Goldwurm,\altaffilmark{1,3} P. A. Caraveo\altaffilmark{2}
39: }
40:
41: \altaffiltext{1}{AstroParticule et Cosmologie (APC), CNRS, Universit\`e Paris 7 Denis Diderot, F-75205 Paris, France; \myemail}
42: \altaffiltext{2}{INAF --Istituto di Astrofisica Spaziale e Fisica Cosmica, via Bassini 15, I-20133 Milano, Italy}
43: \altaffiltext{3}{CEA Saclay, DSM/IRFU/Service d'Astrophysique, F-91191 Gif-sur-Yvette, France}
44: \altaffiltext{4}{Universit\`a degli Studi di Pavia, Dipartimento di Fisica Nucleare e Teorica, via Bassi 6, I-27100 Pavia, Italy}
45: \altaffiltext{5}{Istituto Nazionale di Fisica Nucleare, sezione di Pavia, via Bassi 6, I-27100 Pavia, Italy}
46: \altaffiltext{6}{Istituto Universitario di Studi Superiori, v.le Lungo Ticino Sforza 56, I-27100 Pavia, Italy}
47:
48:
49:
50:
51:
52:
53: \begin{abstract}
54: Pulsar wind nebulae are a prominent class of very high energy ($E >
55: 0.1$ TeV) Galactic sources. Their $\gamma$-ray spectra are interpreted as due to inverse
56: Compton scattering of ultrarelativistic electrons on the ambient photons, whereas
57: the X-ray spectra are due to synchrotron emission. We investigate the
58: relation between the $\gamma$- and-X-ray emission and the
59: pulsars' spin-down luminosity and characteristic age. We find that the distance-independent
60: $\gamma$- to X-ray flux ratio of the nebulae is inversely proportional to the spin-down
61: luminosity, ($\propto \dot{E}^{-1.9}$), while it appears proportional to the characteristic age,
62: ($\propto \tau_c^{2.2}$), of the parent pulsar.
63: We interpret these results as due to the evolution of the electron
64: energy distribution and the nebular dynamics, supporting the idea
65: of so-called relic pulsar wind nebulae.
66: These empirical relations provide a new tool to classify unidentified
67: diffuse $\gamma$-ray sources and to estimate the spin-down
68: luminosity and characteristic age of rotation powered pulsars with no detected pulsation
69: from the X- and $\gamma-$ray properties of the associated pulsar wind nebulae.
70: We apply these relations to predict the spin-down luminosity
71: and characteristic age of four (so far unpulsing) candidate pulsars associated to wind nebulae.
72: \end{abstract}
73:
74: \keywords{pulsars : general --- radiation mechanisms: non-thermal --- supernova remnants --- X-rays : stars --- gamma rays: observations}
75:
76: \section{Introduction}
77: Pulsar Wind Nebulae (PWNe) arise when the wind ejected from a rotation
78: powered pulsar is confined by the pressure of the surrounding
79: medium, be it their supernova remnant or compressed interstellar gas
80: \citep[see][for a review]{Gaensler06}.
81: The Galactic survey performed by the \hess\ experiment
82: \citep[\textit{High Energy Stereoscopic System},][]{Hinton04} has
83: detected several \PWNe\ making them a prominent class of Very High
84: Energy Galactic sources \citep{HESSscanII,Gallant07,Funk07a}. In
85: addition to the classical investigations through radio and X-ray astronomies, Very High Energy
86: $\gamma$-rays (VHE $\gamma$-rays, $E > 0.1$ TeV) provide a new probe of
87: the physical conditions in PWNe \citep[e.g.,][]{deJager08}.
88:
89: The \PWN\ broad-band radiation is most likely due to electron-positron
90: pairs of the pulsar wind generated close to the magnetosphere. The wind
91: flow is ultrarelativistic \citep[bulk Lorentz factor $\Gamma_W \sim
92: 10^6$ in the Crab Nebula;][]{kc84a,kc84b}, until it experiences a
93: strong shock, where electrons are accelerated. After the shock,
94: the flow speed is sub-relativistic at the outer edge of
95: the \PWN. Depending on the radiation mechanisms at work,
96: the electrons can produce photons in different energy ranges: while synchrotron radiation
97: yields photons with energies up to several MeV, inverse Compton
98: scattering of the ambient photon field can produce
99: high energy photons, up to tens of TeV.
100:
101: The electrons responsible for the \PWNe\ $\gamma$-ray emission (here
102: after $\gamma$-ray electrons) are likely less energetic than those generating
103: the X-ray one (X-ray electrons), their synchrotron radiation lying at
104: infrared, optical, or ultraviolet frequencies. For typical nebular
105: magnetic field intensities ($B \sim 1$--100 $\mu$G), synchrotron
106: photons with energy $\sim$1 keV are produced by electrons with
107: Lorentz factor $\sim$0.3--3 $\times 10^8$. The Cosmic
108: Background Radiation, the dust-scattered light, and the starlight
109: provide the target photons for inverse Compton scattering, with
110: typical photon energies around $10^{-3}$ eV, $10^{-2}$ eV, and 1 eV,
111: respectively. In the Thomson regime, photons with energy $\sim$1 TeV
112: are produced by electrons with Lorentz factor $\sim$0.1--3 $\times
113: 10^7$. Due to their different energies, the cooling time of the X-ray
114: electrons is smaller than the one of the $\gamma$-ray
115: electrons. Therefore, the X-ray emission traces the recent history of
116: the nebula, whereas the $\gamma$-ray emission traces a longer history,
117: possibly up to the pulsar birth.
118: The different lifetime of the electrons, together with the interaction with the ambient
119: medium, can lead to the significant projected angular separation, sometimes measured between
120: the peaks of the $\gamma$- and X-ray brightness profiles \citep[e.g. G18.0--0.7,][]{HESSJ1825b}.
121: Since the source of injected electrons, the
122: pulsar rotational energy loss rate dubbed spin-down luminosity,
123: decreases as time goes by, we expect a different
124: evolution of the $\gamma$- and X-ray luminosities, following the
125: particle aging and the pulsar spin-down.
126:
127: In this paper we address first the correlations between the
128: \PWN\ VHE $\gamma$-ray luminosities (1--30 TeV) and their X-ray
129: luminosities (2--10 keV) with the spin-down
130: luminosities, $\dot{E}$, and the characteristic ages, $\tau_c$, of
131: their pulsars. Next we consider the behaviour of the ratio between
132: the gamma and X-ray luminosity as a function of the pulsar spin-down power and age.
133: These relations are discussed in the frame of an evolving
134: electron energy population.
135:
136: \section{Observed correlations}
137: In Table \ref{tab:tevpwne} we report a sample of the identified
138: \PWNe\ observed by the \hess\ experiment. We further included six
139: candidate \PWNe, selecting unidentified \hess\ diffuse sources located
140: near young and energetic pulsars, with $\tau_c \lesssim 100$ kyr and
141: $\dot{E} > 10^{35}$ erg s$^{-1}$. These parameters are defined as $\dot
142: E \equiv 4\pi^2 I \dot P/P^3$ and $\tau_c \equiv\ P/2\dot P$, where
143: $P$ is the pulsar spin period, $\dot{P}$ its derivative, and $I \equiv
144: 10^{45}$ gm cm$^{2}$ the moment of inertia. We calculated $\dot{E}$
145: and $\tau_c$ using the $P$ and $\dot{P}$ values
146: reported in the Australia Telescope National Facility (ATNF) pulsar catalogue\footnote{{\tt http://www.atnf.csiro.au/research/pulsar/psrcat}}
147: \citep{ATNF}. The $\gamma$-ray fluxes, $F_\gamma$, are derived from literature and computed in the 1--30 TeV energy band, with
148: statistical errors estimated with standard Montecarlo propagation technique.
149: The lower energy value corresponds to the highest observed
150: detection threshold. The upper value of 30 TeV reduces the bias of
151: possible unmeasured high-energy cut-offs. The unabsorbed X-ray fluxes, $F_X$,
152: have been derived from literature based on X-ray imaging
153: observatories, and converted in the 2--10 keV energy band. The lower
154: energy is chosen in order to minimize the contamination by possible
155: thermal components due to the pulsar or supernova remnant. When it was possible to
156: single out the PWN from the pulsar component, only the PWN flux is reported.
157:
158: \begin{deluxetable*}{lllclcll}[!h]
159: \tabletypesize{\scriptsize}
160: \tablewidth{\textwidth}
161: \tablecolumns{7}
162: \tablecaption{\label{tab:tevpwne} Properties of Wind Nebulae observed with \hess\ and associated Pulsars}
163: \tablehead{
164: \colhead{Source} & \colhead{Associated} &
165: \colhead{$F_{\gamma}^a$ (1--30 TeV)} & \colhead{ $F_X^b$ (2--10 keV)} &
166: \colhead{ {$\tau_c$} } & \colhead{ {$\dot E$} } & \colhead{Distance}&
167: \colhead{References} \\
168: \colhead{Name} & \colhead{Pulsar} &
169: \colhead{10$^{-12}$ erg cm$^{-2}$ s$^{-1}$} & \colhead{erg cm$^{-2}$ s$^{-1}$} &
170: \colhead{kyr} & \colhead{erg s$^{-1}$} & \colhead{kpc}
171: & \colhead{}
172: }
173: \startdata
174: Crab & PSR~B0531$+$21 & 80\phd\phn\ (4)\phd\phn\ (16) & 2.10 $\times$ 10$^{-8\phn}$ & 1.2 & 4.6 $\times$ 10$^{38}$ & 1.93$^{+0.11}_{-0.11}$ & 1,2,3 \\
175: Vela & PSR~B0833$-$45 & 79\phd\phn\ (15)\phd\ (16) & 5.39 $\times$ 10$^{-11}$ & 11 & 6.9 $\times$ 10$^{36}$ & 0.287$^{+0.019}_{-0.017}$ & 4,5,6 \\
176: K3 in Kookaburra & PSR~J1420$-$6048 & 14.5 (1.6) (2.9) & 1.3\phn\ $\times$ 10$^{-12}$ & 13 & 1.0 $\times$ 10$^{37}$ & 5.6$^{+0.9}_{-0.8}$ & 7,8,9 \\
177: MSH~15--52 & PSR~B1509$-$58 & 20.3 (1.1) (4.1) & 2.86 $\times$ 10$^{-11}$ & 1.6 & 1.8 $\times$ 10$^{37}$ & 5.2$^{+1.4}_{-1.4}$ & 10,11,12 \\
178: G18.0--0.7 & PSR~B1823$-$13 & 61\phd\phn\ (7)\phd\phn\ (12) & 4.4\phn\ $\times$ 10$^{-13}$ & 21 & 2.8 $\times$ 10$^{36}$ & 3.9$^{+0.4}_{-0.4}$ & 13,14,9 \\
179: G21.5--0.9 & PSR~J1833$-$1034 & 2.4\phn\ (1.1) (0.5) & 4.0\phn\ $\times$ 10$^{-11}$ & 4.9 & 3.4 $\times$ 10$^{37}$ & 3.3$^{+0.4}_{-0.5}$ & 15,16,9 \\
180: AX J1838.0-0655 & PSR~J1838$-$0655 & 18.0 (2.7) (3.6) & 1.0\phn\ $\times$ 10$^{-12}$ & 23 & 5.5 $\times$ 10$^{36}$ & 6.6$^{+0.9}_{-0.9}$ & 17,18,19 \\
181: Kes 75 & PSR~J1846$-$0258 & 2.3\phn\ (0.6) (0.5) & 2.27 $\times$ 10$^{-11}$ & 0.73 & 8.1 $\times$ 10$^{36}$ & 6.3$^{+1.2}_{-1.2}$ & 15,20,21 \\
182: HESS~J1303$-$631$^{\dagger}$ & PSR~J1301$-$6305 & 12\phd\phn\ (1.2) (2.4) & 6.2\phn\ $\times$ 10$^{-14}$ & 11 & 1.7 $\times$ 10$^{36}$ & 6.6$^{+1.2}_{-1.1}$ & 22,23,9 \\
183: HESS~J1616$-$508$^{\dagger}$ & PSR~J1617$-$5055 & 21\phd\phn\ (3)\phd\phn\ (4) & 4.2\phn\ $\times$ 10$^{-12}$ & 8.1 & 1.6 $\times$ 10$^{37}$ & 6.7$^{+0.7}_{-0.7}$ & 17,24,9 \\
184: HESS~J1702$-$420$^{\dagger}$ & PSR~J1702$-$4128 & 9.1\phn\ (3.4) (1.8) & 6.0\phn\ $\times$ 10$^{-15}$ & 55 & 3.4 $\times$ 10$^{35}$ & 4.7$^{+0.5}_{-0.5}$ & 17,25,9 \\
185: HESS~J1718$-$385$^{\dagger}$ & PSR~J1718$-$3825 & 4.3\phn\ (1.3) (0.9) & 1.4\phn\ $\times$ 10$^{-13}$ & 90 & 1.3 $\times$ 10$^{36}$ & 3.6$^{+0.4}_{-0.4}$ & 26,27,9 \\
186: HESS~J1804$-$216$^{\dagger}$ & PSR~B1800$-$21 & 11.8 (1.6) (2.4) & 6.8\phn\ $\times$ 10$^{-14}$ & 16 & 2.2 $\times$ 10$^{36}$ & 3.8$^{+0.4}_{-0.5}$ & 17,28,9 \\
187: HESS~J1809$-$193$^{\dagger}$ & PSR~J1809$-$1917 & 19\phd\phn\ (4)\phd\phn\ (4) & 2.3\phn\ $\times$ 10$^{-13}$ & 51 & 1.8 $\times$ 10$^{36}$ & 3.5$^{+0.4}_{-0.5}$ & 26,29,9
188: \vspace{0.1cm}
189: \enddata
190: \tablenotetext{ }{$\dagger$Candidate sources. $^a\gamma$-ray
191: fluxes, statistical, and systematical errors. When not stated in the
192: original papers, the systematic errors were assumed at the typical
193: value of 20\% as in \citet{HESSscanII}. $^b$Errors are conservatively estimated at 20\%.}
194: \tablenotetext{ }{References.--
195: (1) \citealt{HESSCrab}; (2) \citealt{Willingale01}; (3) \citealt{Trimble73}; %Crab
196: (4) \citealt{HESSVela}; (5) \citealt{Manzali07}; (6) \citealt{Dodson03}; %Vela
197: (7) \citealt{HESSKooka}; (8) \citealt{Ng05}; (9) \citealt{ATNF}; %K3
198: (10) \citealt{HESSB1509}; (11) \citealt{Gaensler02}; (12) \citealt{Gaensler99}; %B1509
199: (13) \citealt{HESSJ1825b}; (14) \citealt{Gaensler03}; %(9) \citealt{ATNF}; %B1823
200: (15) \citealt{HESSG21Kes75}; (16) \citealt{Slane00}; %(9) \citealt{ATNF}; %G21
201: (17) \citealt{HESSscanII}; (18) \citealt{Gotthelf08}; (19) \citealt{Davies08} %AX J1838/HESS J1837
202: (20) \citealt{Helfand03}; (21) \citealt{Leahy08}; %(15) \citealt{HESSG21Kes75}; %Kes75
203: (22) \citealt{HESSJ1303}; (23) XMM public data archive; %(9) \citealt{ATNF}; %J1303
204: (24) \citealt{Kargaltsev08b}; %(17) \citealt{HESSscanII}; %(9) \citealt{ATNF}; %J1617
205: (25) \citealt{Chang08}; %(9) \citealt{ATNF}; (17) \citealt{HESSscanII}; %J1702
206: (26) \citealt{HESSJ1809_J1718}; (27) \citealt{Hinton07}; %(9) \citealt{ATNF}; %J1718
207: (28) \citealt{Kargaltsev07a}; %(17) \citealt{HESSscanII}; (9) \citealt{ATNF}; %J1804
208: (29) \citealt{Kargaltsev07b}. }%(25) \citealt{HESSJ1809_J1718}; (9) \citealt{ATNF}; %J1809
209: \end{deluxetable*}
210:
211: We investigated the relations between the different luminosities and
212: the pulsar parameters, using the data collected in Table
213: \ref{tab:tevpwne}. The $\gamma$-ray luminosities, $L_\gamma$, do not
214: appear correlated with the pulsar spin-down luminosities $\dot{E}$, nor
215: they do with the characteristic ages $\tau_c$, as shown in
216: Fig. \ref{fig:fig1} (top panels). This is at variance with the observed
217: PWNe X-ray luminosities, for which a scaling relation is apparent with
218: both $\dot{E}$ and $\tau_c$ (Fig. \ref{fig:fig1}, middle panels). The
219: weighted least square fit on the whole dataset yields
220: \begin{equation}
221: \label{eq:lxedot}
222: \log_{10} L_X = (33.8 \pm 0.04) + (1.87 \pm 0.04) \log_{10} \dot{E}_{37}.
223: \end{equation}
224: All the uncertainties are at 1$\sigma$ level, and $\dot{E} = \dot{E}_{37} \times 10^{37}$
225: erg s$^{-1}$. The $L_X-\dot{E}$ scaling is known for the pulsars as
226: well as for the \PWNe. This scaling was firstly noted
227: by \citet{Seward88}; further, \citet{Becker97} investigate a sample of 27 pulsars with \textit{ROSAT},
228: yielding the simple scaling $L_{X (0.1-2.4 \, \mathrm{keV})} \simeq 10^{-3} \dot{E}$.
229: A re-analysis was performed by \citet{Possenti02}, who studied a sample of 39 pulsars
230: observed by several X-ray observatories, accounting for the statistical and systematic errors. They found
231: $\log_{10} L_X = (-14.36 \pm 0.01) + (1.34 \pm 0.03) \log_{10} \dot{E}$, a relation harder than Eq. (\ref{eq:lxedot}).
232: However, they could not separate the PWN from the pulsar contribution. A better comparison can be done with the results
233: from \citet{Kargaltsev08a}, who recently used high-resolution \textit{Chandra} data in order to decouple the PWN and the
234: pulsar fluxes. Indeed, taking $\dot{E}$, $\tau_c$, and $L_{PWN}$ in the 0.5--8 keV energy band\footnote{The X-ray luminosity
235: reported in \citet{Kargaltsev08a} for Kes 75 was corrected according to the distance measured by \citet{Leahy08}.} from their Tables 1 and 2,
236: we obtained as fitted values
237: $\log_{10} L_{X (0.5-8 \, \mathrm{keV})} = (34.02 \pm 0.05) + (1.46 \pm 0.04) \log_{10} \dot{E}_{37}$ for their whole sample, and
238: $\log_{10} L_{X (0.5-8 \, \mathrm{keV})} = (34.26 \pm 0.03) + (1.87 \pm 0.01) \log_{10} \dot{E}_{37}$ restricting the fit
239: only to the sources also present in our sample.
240: The latter is compatible in the terms of slope with Eq. (\ref{eq:lxedot}), and the slight difference in normalization can
241: be due to the different energy band.
242:
243: \begin{figure}[!thb]
244: \includegraphics[angle=0,width=0.48\textwidth]{./f1.eps}
245: \caption{$\gamma$-ray luminosity, X-ray luminosity, and $\gamma$- to X-ray flux ratio
246: versus pulsar spin-down luminosity, $\dot{E}$ (left column),
247: and characteristic age, $\tau_c$ (right column). Filled and open
248: circles stand for identified and candidate \PWNe, respectively. The
249: upper-limit for the flux ratio of PSR B1706-44 \citep{HESSB1706, Romani05}
250: %$F_\gamma < 1.4 \times 10^{-12}$ erg cm$^{-2}$ s$^{-1}$ >350 GeV
251: %$F_X = 2.52 \times 10^{-13}$ erg cm$^{-2}$ s$^{-1}$ 2-10 KeV equatorial PWN unabsorbed
252: is reported with an arrow. Also shown are the best-fit curves for identified
253: \PWNe\ (dotted lines), and for the whole sample (dashed lines).
254: \label{fig:fig1}
255: }
256: \end{figure}
257:
258: X-ray sources of our whole dataset also show a dependence of
259: $L_X$ on $\tau_c$, with a best-fit relation
260: \begin{equation}
261: \label{eq:lxtau}
262: \log_{10} L_X = (33.7 \pm 0.04) - (2.49 \pm 0.06) \log_{10} \tau_4,
263: \end{equation}
264: where $\tau_c$ is in units of years. The $L_X-\tau_c$ scaling was already noted by
265: \citet{Becker97} and \citet{Possenti02}. Also in this case we compared our fit to the one derived
266: using the whole \citet{Kargaltsev08a} dataset, which results in
267: $\log_{10} L_{X (0.5-8 \, \mathrm{keV})} = (34.29 \pm 0.01) - (2.03 \pm 0.01) \log_{10} \tau_4$ for their whole sample, and
268: $\log_{10} L_{X (0.5-8 \, \mathrm{keV})} = (34.23 \pm 0.02) - (2.60 \pm 0.02) \log_{10} \tau_4$ restricting the fit
269: only to the sources also present in our sample.
270:
271: \begin{deluxetable*}{lccrr}
272: \tabletypesize{\footnotesize}
273: \tablewidth{0.7\textwidth}
274: \tablecolumns{5}
275: \tablecaption{\label{tab:new} PWNe hosting a neutron star without detected pulsations}
276: \tablehead{
277: \colhead{Source} & \colhead{$F_{\gamma}$ (1--30 TeV)} & \colhead{
278: $F_X$ (2--10 keV)} & \colhead{ {$\tau_c^\ast$} } & \colhead{ {$\dot{E}^\ast$} } \\
279: \colhead{Name} & \colhead{erg cm$^{-2}$ s$^{-1}$} &
280: \colhead{erg cm$^{-2}$ s$^{-1}$} & \colhead{kyr} & \colhead{erg s$^{-1}$}
281: }
282: \startdata
283: G313.+0.1 Rabbit & $1.0 \times 10^{-11}$ & $7.3 \times 10^{-12}$ & $\sim$6 & $\sim1.5 \times 10^{37}$ \\
284: G0.9+0.1 & $3.3 \times 10^{-12}$ & $5.8 \times 10^{-12}$ & $\sim$4 & $\sim2 \times 10^{37}$ \\
285: G12.8-0.0$^{\dagger}$ & $1.3 \times 10^{-11}$ & $9.2 \times 10^{-12}$ & $\sim$6 & $\sim1.5 \times 10^{37}$ \\
286: HESS~J1640--465$^{\dagger}$ & $9.3 \times 10^{-12}$ & $1.0 \times 10^{-12}$ & $\sim$13 & $\sim5 \times 10^{36}$ \\
287: \vspace{-0.3cm}
288: \enddata
289: %\tablenotetext{a}{pinuccio.}
290: \tablenotetext{ }{$^\dagger$Candidate sources. $^\ast$Predicted values. References.-- HESS~J1418/G313.+0.1 Rabbit: \citet{HESSKooka}, \citet{Ng05}; HESS~J1747--281/G0.9+0.1: \citet{HESSG09}, \citet{Porquet03}; HESS~J1813--178/G12.8--0.0: \citet{HESSscanII}, \citet{Helfand07}; HESS~J1640--465/G338.3--0.0: \citet{HESSscanII}, \citet{Funk07b}.}
291: \end{deluxetable*}
292:
293: The lower panels of Fig. \ref{fig:fig1} refer to the $\gamma$- to X-ray flux ratio
294: $F_\gamma/F_X$. There is a clear anticorrelation between $F_\gamma/F_X$ and $\dot{E}$, spanning
295: over four decades in $\dot{E}$ and seven decades in $F_\gamma/F_X$
296: (Fig. \ref{fig:fig1}, bottom left panel). Considering only the identified \PWNe,
297: the correlation coefficient is $r = -0.7\pm 0.2$; including also the
298: candidate sources, the anticorrelation is more significant, with $r =
299: -0.84\pm 0.09$. The best-fit including only the identified sources yields
300: \begin{equation}
301: \label{eq:fgammafx_edot1}
302: \log_{10} F_\gamma/F_X = (0.47 \pm 0.05) - (1.87 \pm 0.07) \log_{10} \dot{E}_{37}.
303: \end{equation}
304: For all the data points, it results
305: \begin{equation}
306: \label{eq:fgammafx_edot2}
307: \log_{10} F_\gamma/F_X = (0.57 \pm 0.04) - (1.88 \pm 0.05) \log_{10} \dot{E}_{37},
308: \end{equation}
309: compatible within the errors with the relation obtained using only the identified sources.
310:
311: The $\gamma$- to X-ray flux ratio is also found to correlate with the
312: characteristic age $\tau_c$ (Fig. \ref{fig:fig1}, bottom right panel), with
313: a correlation coefficient $r = 0.7 \pm 0.2$ for identified \PWNe\ only, and
314: $r = 0.75 \pm 0.13$ for the whole sample. The ordinary weighted least
315: square fit only for the identified \PWNe\ yields
316: \begin{equation}
317: \label{eq:fgammafx_tau1}
318: \log_{10} F_\gamma/F_X = (0.70 \pm 0.06) + (2.21 \pm 0.09) \log_{10} \tau_4,
319: \end{equation}
320: and for all the data points
321: \begin{equation}
322: \label{eq:fgammafx_tau2}
323: \log_{10} F_\gamma/F_X = (0.89 \pm 0.04) + (2.14 \pm 0.07) \log_{10} \tau_4.
324: \end{equation}
325: One should note that these correlations are based on 8 identified sources,
326: and are consistent when the 6 candidate sources are considered.
327: More $\gamma$-ray detections may improve their significance.
328:
329: \section{Discussion}
330: \label{sec:model}
331: We found the $\gamma$- to X- ray luminosity ratio $L_\gamma/L_X =
332: F_\gamma/F_X$ to be anticorrelated with the spin-down luminosity
333: $\dot{E}$ and correlated with the characteristic age $\tau_c$. Formally,
334: such dependencies are driven by the scaling law of the X-ray
335: luminosity $L_X$, which increases with $\dot{E}$ and decreases with
336: $\tau_c$, since the values of $L_\gamma$ were found uncorrelated with
337: the pulsar parameters. However, the $F_\gamma/F_X$ is a
338: distant-independent indicator which relates two electron populations,
339: differing by energy and age. An evolution in the \PWN\ broad-band
340: spectrum is pointed out by Eq. (\ref{eq:fgammafx_tau1}), which implies
341: $L_\gamma > L_X$ after $\sim$5 kyr from pulsar birth. Therefore,
342: the $\gamma$-ray emission remains efficient around
343: $L_\gamma \sim 10^{33}$--10$^{35}$ erg s$^{-1}$, while the X-ray luminosity
344: decreases by a factor $\sim$10$^6$ in 10$^5$ yr following the pulsar
345: spin-down.
346:
347:
348: Such a broad-band spectral evolution can be expected on the basis of
349: the \PWNe\ leptonic model \citep{kc84b, Chevalier00}. In a PWN, the
350: source of the injected electrons is the pulsar spin-down luminosity,
351: $\dot{E}$. The total injection rate of the electrons can be written:
352: \begin{equation}
353: \dot{N} = \frac{\dot{E}}{\Gamma_W \, m_e c^2 \, (1 + \sigma)},
354: \end{equation}
355: where the magnetization parameter $\sigma$ sets the fraction of the spin-down luminosity converted in kinetic energy of the wind. The whole spin-down luminosity is converted in particle kinetic energy for $\sigma \ll 1$, as for the Crab Nebula \citep{kc84a,kc84b}. For sake of simplicity, we assume a constant wind Lorentz factor $\Gamma_W$ upstream the shock. $\dot{E}$ decreases in time as \cite[e.g.,][]{Pacini73}
356: \begin{equation}
357: \label{eq:varying-edot2}
358: \dot{E}(t) = \frac{\dot{E}_0}{ \left( 1 + t/t_{dec} \right)^\beta},
359: \end{equation}
360: where $\dot{E}_0 \sim 10^{38}$--10$^{40}$ erg s$^{-1}$ is the
361: spin-down luminosity at the pulsar birth, $t_{dec} \sim 100$--1000 yr is a
362: characteristic decay time, $t$ is the time elapsed since pulsar birth
363: ($t_0 = 0$), and $\beta = (n+1)/(n-1)$, where $n$ is the braking index.
364: In the following, we assume a pure dipolar magnetic field torque, i.e. $n = 3$.
365: As the braking indices inferred from the measurement of the period and its derivatives
366: are significantly smaller than 3 \citep{Livingstone07}, we dealt with a generic $n$ (see App. \ref{app:a}),
367: and found that the results derived from Eq. (\ref{eq:varying-edot2}) are unaffected by the choice of $n$.
368:
369: Since it depends on $\dot{E}$, also the particle injection rate $\dot{N}$
370: decreases in time. Therefore, the total number of particles
371: \begin{equation}
372: \label{eq:N}
373: N \propto \int_0^t \dot{E}(t') \, dt' = \dot{E}_0 \, t_{dec} \left(
374: \frac{t}{t+t_{dec}} \right),
375: \end{equation}
376: reaches a constant value $N \propto \dot{E}_0 \, t_{dec}$ for $t \gg
377: t_{dec}$, and the particle supply by the pulsar becomes negligible.
378:
379: The electron energy distribution $n(E,t)$ accounting for particle
380: injection and radiative losses evolves according to the kinetic
381: equation \citep[e.g.,][]{Ginzburg64,Blumenthal70}:
382: \begin{equation}
383: \label{eq:kineticeq}
384: \dede{n}{t} = \dede{}{E} \left(n P\right) + Q,
385: \end{equation}
386: where $Q = Q(E,t)$ is the particle distribution injected per unit time, and $P = P(E,t)$
387: is the radiated power per particle with energy $E$. The normalization
388: of $n(E,t)$ is set by $N$ via the injection rate: $\dot{N}(t) = \int
389: Q(E,t) dE$.
390:
391: At energies for which the radiative losses are negligible, the number
392: of particles $n(E,t)$ with energy $E$ at time $t$ has the same profile
393: of the injected distribution $Q(E)$ with a normalization set by
394: $N$. Therefore,
395: \begin{equation}
396: \label{eq:n_uncooled}
397: n_{u}(E,t) \propto \int_0^t \dot{E}(t') \, dt' = \dot{E}_0 \, t_{dec}
398: \left( \frac{t}{t+t_{dec}} \right),
399: \end{equation}
400: where $u$ stands for {\em uncooled}. As in Eq. (\ref{eq:N}), a constant
401: value $n_{u}(E,t) \propto \dot{E}_0 \, t_{dec}$ is reached for $t \gg t_{dec}$.
402:
403: The effect of the radiative losses is to limit the accumulation of
404: particles at a given energy. After an energy-dependent cooling time
405: $t_c(E)$, the particles with initial energy $E$ have radiated a
406: significant fraction of their energy \citep{Chevalier00}. Accounting
407: for pitch-angle averaged synchrotron and inverse Compton in the Thomson regime energy losses,
408: the cooling time can be written as
409: \begin{equation}
410: \label{eq:coolingtime}
411: t_c (E) = \frac{9 \, m_e^3 c^5}{4 \, (1+\xi) \, e^4 \, \gamma_E \, B^2} \simeq
412: 24.5 \, (1+\xi)^{-1}\, \gamma_7^{-1} \, B_5^{-2} \quad \textrm{kyr},
413: \end{equation}
414: where $\gamma_E= E/(m_e c^2)$ is the particle Lorentz factor, and $\xi
415: = U_{ph}/U_B$, with $U_{ph}$ and $U_B$ the photon field and
416: magnetic field energy densities, respectively ($\gamma_E = \gamma_7 \times 10^7$, $B = B_5 \times 10^{-5}$ G).
417: When the photon field is provided by the Cosmic Background Radiation ($U_{ph} = 0.26$ eV cm$^{-3}$), the synchrotron
418: radiation is the main cooling process ($\xi < 1$) if $B > 3$
419: $\mu$G. This condition is generally fulfilled in \PWNe\ as the
420: equipartition magnetic field intensity ranges in $B \sim 1 - 100$
421: $\mu$G.\footnote{In radiation-dominated environment, like the Galactic
422: Center, the inverse Compton can contribute to the cooling. In this
423: case, the Klein-Nishina regime should be taken into account
424: \citep{Manolakou07}.} Eq. (\ref{eq:coolingtime}) shows that
425: the cooling time of $\gamma$-ray radiating particles, $t_{c\gamma}$,
426: is one order of magnitude longer than that of the X-ray radiating
427: particles, $t_{cX}$, e.g., for $B = 10$ $\mu$G, $t_{c\gamma} \sim
428: 8$--250 kyr, and $t_{cX} \sim 0.8$--8 kyr. By comparing $t_{c\gamma}$
429: and $t_{cX}$ with the average characteristic ages of pulsars in TeV
430: \PWNe, the $\gamma$-radiation is produced by long-lived electrons
431: tracing the time-integrated evolution of the nebula, even up to the
432: pulsar birth, whereas the X-ray emission is generated by younger
433: electrons, injected in the last thousands of years.
434:
435: Only the particles injected since the last $t_c(E)$ years will contribute to
436: $n(E,t)$. Eq. (\ref{eq:n_uncooled}) is accordingly modified:
437: \begin{equation}
438: \label{eq:n_cooled}
439: n_{c}(E,t) \propto \int_{t-t_c}^{t} \dot{E}(t') \, dt' =
440: \frac{\dot{E}_0 \, t_{dec}^2 \, t_c}{(t - t_c + t_{dec}) \, (t+t_{dec})},
441: \end{equation}
442: where $c$ stands for {\em cooled}. This implies $n_c(E,t) \propto
443: \dot{E}_0 \, t_{dec}^2 \, t_c t^{-2}$ for $t \gg {\rm
444: max}(t_c,\,t_{dec})$, and hence $n_c(E,t) \propto \dot{E} \, t_c$
445: using Eq. (\ref{eq:varying-edot2}).
446:
447: \section{Conclusions}
448: Eqs. (\ref{eq:n_uncooled}) and (\ref{eq:n_cooled}) describe the
449: time evolution of a particle populations in two regimes, uncooled and
450: cooled. Such an evolution is exemplified in Fig. \ref{fig:fig2} for
451: the populations of particles producing $\gamma$-rays, $n_\gamma$, and
452: X-rays, $n_X$. After the initial rise, both the particle populations
453: reach a plateau ($t > t_{dec}$). The decrease begins when the
454: evolution time is greater than the cooling time. As in general $t_{cX}
455: < t_{c\gamma}$, the X-ray emission fades long before the $\gamma$-ray
456: one.
457:
458: \begin{figure}[htb]
459: % \centering
460: \includegraphics[angle=0,width=0.48\textwidth]{./f2.eps}
461: \caption{Time evolution of the number of particles radiating in VHE
462: $\gamma$-rays, $n_\gamma$, and in X-rays, $n_X$ (solid lines), and of their ratio (dashed line).
463: Pulsar birth is at $t = 0$. Initial conditions for the pulsar spin-down luminosity
464: are $\dot{E}_0 = 10^{39}$ erg s$^{-1}$ and $t_{dec}$ = 100 yr. Both curves are
465: normalized to their maximum value. After the initial rise, both
466: particle populations reach a plateau. The fall begins at t
467: greater than the cooling time, which is assumed to be: $t_{cX}$ = 2.6 kyr for X-rays,
468: $t_{c\gamma}$ = 25 kyr for $\gamma$-rays (for a magnetic field intensity $B = 10 \, \mu$G, and
469: a Lorentz factor of $\gamma$-ray radiating electrons $\gamma = 10^7$).
470: }
471: \label{fig:fig2}
472: \end{figure}
473:
474: As the characteristic ages of the pulsars powering a VHE $\gamma$-ray
475: \PWN\ are in the range 1--20 kyr, likely $t_{cX} < \tau_c <
476: t_{c\gamma}$. Accordingly, the population of the X-ray electrons,
477: $n_X$, is likely to be in the cooling regime, i.e., it decreases. The
478: scaling laws $n_X \propto \tau_c^{-2}$ and $n_X \propto \dot{E}$ of Eq. (\ref{eq:n_cooled}) support the
479: trend observed in the data, see Eq. (\ref{eq:lxedot}). At variance,
480: the population of the $\gamma$-ray electrons, $n_\gamma$, is in the
481: uncooled regime, the asymptotic limit of Eq.~(\ref{eq:n_uncooled});
482: this explains the lack of correlation between $\gamma$-ray luminosity
483: $L_\gamma$ and $\dot{E}$. Finally, Eqs.~(\ref{eq:n_uncooled}) and
484: (\ref{eq:n_cooled}) for $t_{cX} < \tau_c < t_{c\gamma}$ imply
485: a ratio $n_\gamma/n_X \propto t^{2} \propto \dot{E}^{-1}$. Since the luminosities
486: are roughly proportional to the population of radiating particles,
487: finally one gets
488: \[
489: L_\gamma/L_X \propto t^{2} \propto \dot{E}^{-1},
490: \]
491: to compare with the best-fit empirical relations $L_\gamma/L_X \propto
492: \tau_c^{2.2}$ and $L_\gamma/L_X \propto \dot{E}^{-1.9}$, see
493: Eqs. (\ref{eq:fgammafx_tau1}) and (\ref{eq:fgammafx_edot1}). Though
494: the outlined model does not correctly predict the slopes, not surprisingly in being simplified,
495: it highlights the concurrent roles of the evolving pulsar injection and of the radiative losses
496: in producing the observed trends.
497:
498: The scattering around the relations for $F_\gamma/F_X$ reflects the
499: lack of correlation between $L_\gamma$ and $\dot{E}$. Environmental
500: factors can affect the $\gamma$-ray luminosities, like the local
501: energy density of the ambient photon field, or the interaction with
502: the surrounding medium causing an enhancement in the magnetic
503: field. Also, unmeasured pulsar properties such as the magnetic field,
504: its orientation with respect to the spin axis, and the initial spin period
505: might affect the pulsar wind properties.
506:
507: We stress that the relations presented here are derived under several assumptions, the most important of which being that
508: X-ray and $\gamma$-ray emitting particles are in different cooling regimes, cooled for X-rays and uncooled for $\gamma$-rays.
509: However, the Lorentz factors ranges of the two populations get closer, and they can even overlap, if the nebular magnetic field is very high,
510: on the order of $B$=170 $\mu$G. On the other hand, in the case of a young nebula with a very low magnetic field, the X-ray electrons may
511: not have reached the cooling regime, leaving the $\gamma$-ray production to the low-energy freshly injected electrons.
512: Hence, PWNe with a very weak magnetic field, like 3C 58 \citep{Slane08}, or possibly with a unusually strong one, as reported lately by \citet{Arzoumanian08} for DA 495, could represent outliers to our derived relations. These regimes can be properly taken into account through numerical
513: modelling of the kinetic equation (Eq. [\ref{eq:kineticeq}]). Another important assumption is a uniform and constant
514: magnetic field: indeed high resolution imaging observations of several PWNe show
515: a dynamical and structured nebular morphology \citep{Gaensler06}.
516: The evolution of the average magnetic field is complicated by the interaction with the supernova ejecta,
517: which is expected to occur after a few thousands of years since pulsar birth, causing global oscillations of
518: the magnetic field intensity \citep[e.g.,][]{Bucciantini03}.
519: One should note that the cooling time is not well defined if it is comparable to or longer than the time scale
520: of variation of the magnetic field. The cases of patchy or evolving magnetic field are further sources of
521: scattering around our relations.
522:
523: Given the limitations discussed above, the empirical relations in Eqs. (\ref{eq:fgammafx_edot1}) and
524: (\ref{eq:fgammafx_tau1}) provide a new tool to estimate the spin-down
525: luminosity and characteristic age of a pulsar lacking detected pulsation
526: from the $\gamma$- and X-ray properties of the associated \PWN. For
527: the four candidate pulsars in Table 2, we used $F_\gamma/F_X$ to predict
528: $\dot{E}$ and $\tau_c$. Taking into account the average scattering (average absolute ratio) around the
529: best fit relations, Eqs. (\ref{eq:fgammafx_edot1}) and (\ref{eq:fgammafx_tau1}),
530: one should expect an uncertainty of a factor $\sim$2.5 for $\dot{E}$
531: and $\sim$2.3 for $\tau_c$ considering only the eight identified sources.
532: On the other hand, considering Eqs. (\ref{eq:fgammafx_edot2}) and (\ref{eq:fgammafx_tau2}),
533: and including also the candidate sources, the uncertainties are $\sim$2.2 for $\dot{E}$
534: and $\sim$2.6 for $\tau_c$.
535:
536: The correlations for $F_\gamma/F_X$ hold also after including
537: the candidate sources. The pulsars possibly associated to the candidate sources are
538: mostly older Vela-like pulsars, with $8 \times 10^3$ yr $<$ $\tau_c$ $<$ $9 \times 10^4$ yr,
539: and $3.4 \times 10^{35}$ erg s$^{-1}$ $<$ $\dot{E}$ $<$
540: $1.6 \times 10^{37}$ erg s$^{-1}$. Due to the pulsar ages, the
541: electrons had the time to be advected far from the pulsar, producing the
542: offset between the $\gamma$-ray emission centroid and the pulsar
543: position, the process leading to the so-called relic \PWNe\ \citep{deJager08}.
544: The fact that all the confirmed associations contain younger pulsars is hence not surprising, as the
545: positional coincidence is one of the main identification criteria.
546: If the identification of candidate sources with relic \PWNe\ is
547: confirmed, the $\gamma$-ray luminosity would persist up to 10$^{5}$
548: yr, with remarkable time-integrated energy channeled in radiation ($\sim 3 \times 10^{45}$--$3\times 10^{47}$ erg).
549:
550: \acknowledgments
551: FM, MF, and DG acknowledge the French Space Agency (CNES) for financial support.
552: FM is also grateful for support from the Moscow St. NGO.
553: We wish to thank the referee, P. Slane, for his very constructive comments and suggestions that helped to improve the manuscript.
554:
555: \begin{appendix}
556: \section{Particle population injected by a pulsar with generic braking index}
557: \label{app:a}
558:
559: By adopting Eq. (\ref{eq:varying-edot2}) for a generic braking index $n > 1$, Eqs. (\ref{eq:n_uncooled}) and (\ref{eq:n_cooled}) are so modified:
560: \begin{equation}
561: \label{eq:An_uncooled}
562: n_{u} (E,t) \propto \int_0^t \dot{E}(t') \, dt' = \frac{\dot{E}_0 \, t_{dec}}{\beta-1}\left[ 1 - \left(\frac{t_{dec}}{t+t_{dec}} \right)^{\beta -1} \right],
563: \end{equation}
564: and
565: \begin{equation}
566: \label{eq:An_cooled}
567: n_{c}(E,t) \propto \int_{t-t_c}^{t} \dot{E}(t') \, dt' = \frac{\dot{E}_0 \, t_{dec}^\beta}{\beta-1} \; (t_{dec} + t)^{1 - \beta} \; \left[ \left(1 - \frac{t_c}{t_{dec} + t} \right)^{1-\beta} -1 \right].
568: \end{equation}
569: For $t \gg t_{dec}$ Eq. (\ref{eq:An_uncooled}) yields $n_{u} \propto \dot{E}_0 \, t_{dec} / (\beta -1)$, while for $t \gg {\rm
570: max}(t_c,\,t_{dec})$ Eq. (\ref{eq:An_cooled}) yields $n_{c} \propto \dot{E}(t) \, t_{c}$. As in the case of the dipolar magnetic braking, in the uncooled regime most of the radiating particles has been injected in the early phases, whereas in the cooled regime the particle population is limited by the cooling time and follows more closely the current spin-down rate.
571: \end{appendix}
572: %\clearpage
573:
574:
575: \bibliographystyle{apj}
576: \begin{thebibliography}
577:
578: \bibitem[Aharonian et al.(2005a)]{HESSB1706}
579: Aharonian, F., et al. (HESS Collaboration)\ 2005a, \aap, 432, L9
580:
581: \bibitem[Aharonian et al.(2005b)]{HESSG09}
582: Aharonian, F., et al. (HESS Collaboration)\ 2005b, \aap, 432, L25
583:
584: %\bibitem[Aharonian et al.(2005)]{2005Sci...307.1938A}
585: %Aharonian, F., et al. (HESS Collaboration)\ 2005, Science, 307, 1938
586:
587: \bibitem[Aharonian et al.(2005c)]{HESSB1509}
588: Aharonian, F., et al. (HESS Collaboration)\ 2005c, \aap, 435, L17
589:
590: \bibitem[Aharonian et al.(2005d)]{HESSJ1303}
591: Aharonian, F., et al. (HESS Collaboration)\ 2005d, \aap, 439, 1013
592:
593: %HESS J1825 first paper
594: %\bibitem[Aharonian et
595: %al.(2005)]{2005A&A...442L..25A} Aharonian, F.~A., et al. (HESS Collaboration)\ 2005, \aap, 442, L25
596:
597: \bibitem[Aharonian et al.(2006a)]{HESSVela}
598: Aharonian, F., et al. (HESS Collaboration)\ 2006a, \aap, 448, L43
599:
600: \bibitem[Aharonian et al.(2006b)]{HESSKooka}
601: Aharonian, F., et al. (HESS Collaboration)\ 2006b, \aap, 456, 245
602:
603: \bibitem[Aharonian et al.(2006c)]{HESSCrab}
604: Aharonian, F., et al. (HESS Collaboration)\ 2006c, \aap, 457, 899
605:
606: \bibitem[Aharonian et al.(2006d)]{HESSJ1825b}
607: Aharonian, F., et al. (HESS Collaboration)\ 2006d, \aap, 460, 365
608:
609: \bibitem[Aharonian et al.(2006e)]{HESSscanII}
610: Aharonian, F., et al. (HESS Collaboration)\ 2006e, \apj, 636, 777
611:
612: \bibitem[Aharonian et al.(2007)]{HESSJ1809_J1718}
613: Aharonian, F., et al. (HESS Collaboration)\ 2007, \aap, 472, 489
614:
615: \bibitem[Aharonian et al.(2008)]{HESSJ1913}
616: Aharonian, F., et al. (HESS Collaboration)\ 2008, \aap, 484, 435
617:
618: \bibitem[Arzoumanian et al.(2008)]{Arzoumanian08}
619: Arzoumanian, Z., Safi-Harb, S., Landecker, T.~L., Kothes, R.,
620: \& Camilo, F.\ 2008, \apj, in press (astro-ph/0806.3766)
621:
622: \bibitem[Becker \& Tr\"umper(1997)]{Becker97}
623: Becker, W., \& Tr\"umper, J.\ 1997, \aap, 326, 682
624:
625: \bibitem[Bucciantini et al.(2003)]{Bucciantini03}
626: Bucciantini, N., Blondin, J.~M., Del Zanna, L., \& Amato, E.\ 2003, \aap, 405, 617 \bibitem[Blumenthal \& Gould (1970)]{Blumenthal70}
627: {{Blumenthal}, G.~R. \& {Gould}, R.~J.} 1970, Rev. Mod. Phys., 42, 237
628:
629: % J1702, ApJ accepted
630: \bibitem[Chang et al.(2008)]{Chang08}
631: Chang, C., Konopelko, A., \& Cui, W.\ 2008, \apj, 682, 1177
632:
633: \bibitem[Chevalier(2000)]{Chevalier00}
634: {{Chevalier}, R.~A.} 2000, \apjl, 539, L45
635:
636: \bibitem[Davies et al.(2008)]{Davies08}
637: Davies, B., Figer, D.~F., Law, C.~J., Kudritzki, R.-P., Najarro, F., Herrero, A.,
638: \& MacKenty, J.~W.\ 2008, \apj, 676, 1016
639:
640: \bibitem[Djannati-Ata{\"i} et al.(2007)]{HESSG21Kes75}
641: {{Djannati-Ata{\"i}}, A., {de Jager}, O.~C., {Terrier}, R.,
642: {Gallant}, Y.~A., {Hoppe}, S.} 2007, in Proceedings of the 30th ICRC
643: (Merida, Mexico), in press, (astro-ph/0710.2247)
644:
645: \bibitem[Dodson et al.(2003)]{Dodson03}
646: Dodson, R., Legge, D., Reynolds, J.~E., \& McCulloch, P.~M.\ 2003, \apj, 596, 1137
647:
648: %\bibitem[Esposito(2003)]{EspositoPC}
649: %Esposito, P., private communication, based on the XMM public data archive
650:
651: \bibitem[Funk(2007)]{Funk07a}
652: Funk, S.\ 2007, \apss, 309, 11
653:
654: \bibitem[Funk et al.(2007)]{Funk07b}
655: Funk, S., Hinton, J.~A., P{\"u}hlhofer, G., Aharonian, F.~A., Hofmann, W., Reimer, O., \& Wagner, S.\ 2007, \apj, 662, 517
656:
657: \bibitem[Gaensler et al.(1999)]{Gaensler99}
658: Gaensler, B.~M., Brazier, K.~T.~S., Manchester, R.~N., Johnston, S., \& Green, A.~J.\ 1999, \mnras, 305, 724
659:
660: \bibitem[Gaensler et al.(2002)]{Gaensler02}
661: Gaensler, B.~M., Arons, J., Kaspi, V.~M., Pivovaroff, M.~J., Kawai, N., \& Tamura, K.\ 2002, \apj, 569, 87
662:
663: \bibitem[Gaensler et al.(2003)]{Gaensler03}
664: Gaensler, B.~M., Schulz, N.~S., Kaspi, V.~M., Pivovaroff, M.~J., \& Becker, W.~E.\ 2003, \apj, 588, 441
665:
666: \bibitem[Gaensler \& Slane(2006)]{Gaensler06}
667: Gaensler, B.~M., \& Slane, P.~O.\ 2006, \araa, 44, 17
668:
669: \bibitem[Gallant(2007)]{Gallant07}
670: Gallant, Y.~A.\ 2007, \apss, 309, 197
671:
672: \bibitem[Ginzburg \& Syrovatskii(1964)]{Ginzburg64}
673: Ginzburg, V.~L., \& Syrovatskii, S.~I.\ 1964, The Origin of Cosmic Rays (New York: Macmillan)
674:
675: %\bibitem[Gotthelf(2003)]{Gotthelf03}
676: %Gotthelf, E.~V.\ 2003, \apj, 591,361
677:
678: %AX J1838
679: \bibitem[Gotthelf \& Halpern(2008)]{Gotthelf08}
680: Gotthelf, E.~V., \& Halpern, J.~P.\ 2008, \apj, 681, 515
681:
682: %\bibitem[\protect\citeauthoryear{Gould}{1965}]{Gould65}
683: %{{Gould}, R.~J.} 1965, PRL, 15, 577
684:
685: \bibitem[Helfand et al.(2003)]{Helfand03}
686: Helfand, D.~J., Collins, B.~F., \& Gotthelf, E.~V.\ 2003, \apj, 582, 783
687:
688: \bibitem[Helfand et al.(2007)]{Helfand07}
689: Helfand, D.~J., Gotthelf, E.~V., Halpern, J.~P., Camilo, F., Semler, D.~R., Becker, R.~H., \& White, R.~L.\ 2007, \apj, 665, 1297
690:
691: \bibitem[Hinton(2004)]{Hinton04}
692: Hinton, J.~A.\ 2004, NewA Rev., 48, 331
693:
694: \bibitem[Hinton et al.(2007)]{Hinton07}
695: Hinton, J.~A., Funk, S., Carrigan, S., Gallant, Y.~A., de Jager, O.~C., Kosack, K., Lemi{\`e}re, A., P\"uhlhofer, G.\ 2007, \aap, 476, L25
696:
697: %B1800
698: \bibitem[Kargaltsev et al.(2007)]{Kargaltsev07a} Kargaltsev, O.,
699: Pavlov, G.~G., \& Garmire, G.~P.\ 2007, \apj, 660, 1413
700:
701: %J1809
702: \bibitem[Kargaltsev \& Pavlov(2007)]{Kargaltsev07b}
703: Kargaltsev, O., \& Pavlov, G.~G.\ 2007, \apj, 670, 655
704:
705: %Lx - Edot relation
706: \bibitem[Kargaltsev \& Pavlov(2008)]{Kargaltsev08a}
707: Kargaltsev, O., \& Pavlov, G.~G.\ 2008, in AIP Conf. Proc. 983, 171
708: %in AIP Conf. Proc. 983, 40 Years of Pulsars: Millisecond Pulsars,
709: %Magnetars, and More, ed. C. Bassa, Z. Wang, A. Cumming, & V. M. Kaspi
710: %(New York: AIP), 171
711:
712: %1617
713: \bibitem[Kargaltsev et al.(2008)]{Kargaltsev08b}
714: Kargaltsev, O., Pavlov, G.~G., \& Wong, J.~A.\ 2008, \apj, submitted (astro-ph/0805.1041)
715:
716: \bibitem[Kennel \& Coroniti(1984a)]{kc84a}
717: Kennel, C.~F., \& Coroniti, F.~V.\ 1984a, \apj, 283, 694
718:
719: \bibitem[Kennel \& Coroniti(1984b)]{kc84b}
720: Kennel, C.~F., \& Coroniti, F.~V.\ 1984b, \apj, 283, 710
721:
722: \bibitem[\protect\citeauthoryear{de Jager \& Djannati-Ata{\"i}}{2008}]{deJager08}
723: {{de Jager}, O.~C. and {Djannati-Ata{\"i}}, A.} 2008, (astro-ph/0803.0116)
724:
725: \bibitem[Leahy \& Tian(2008)]{Leahy08}
726: Leahy, D.~A., \& Tian, W.~W.\ 2008, \aap, 480, L25
727:
728: \bibitem[Livingstone et al.(2007)]{Livingstone07}
729: Livingstone, M.~A., Kaspi, V.~M., Gavriil, F.~P., Manchester, R.~N., Gotthelf, E.~V.~G., \& Kuiper, L.\ 2007, \apss, 308, 317
730:
731: \bibitem[\protect\citeauthoryear{Manchester et al.}{2005}]{ATNF}
732: {{Manchester}, R.~N. and {Hobbs}, G.~B. and {Teoh}, A. and {Hobbs}, M.} 2005, \aj, 129, 1993
733:
734: \bibitem[\protect\citeauthoryear{Manolakou et al.}{2007}]{Manolakou07}
735: {{Manolakou}, K., {Horns}, D. \& {Kirk}, J.~G.} 2007, \aap, 474, 689
736:
737: \bibitem[Manzali et al.(2007)]{Manzali07}
738: Manzali, A., De Luca, A., \& Caraveo, P.~A.\ 2007, \apj, 669, 570
739:
740: \bibitem[Ng et al.(2005)]{Ng05}
741: Ng, C.-Y., Roberts, M.~S.~E., \& Romani, R.~W.\ 2005, \apj, 627, 904
742:
743: \bibitem[Porquet et al.(2003)]{Porquet03}
744: Porquet, D., Decourchelle, A., \& Warwick, R.~S.\ 2003, \aap, 401, 197
745:
746: \bibitem[Pacini \& Salvati(1973)]{Pacini73}
747: Pacini, F., \& Salvati, M.\ 1973, \apj, 186, 249
748:
749: %%ref piu' nuova per B1823, non usata
750: %\bibitem[Pavlov et al.(2008)]{Pavlov08}
751: %Pavlov, G.~G., Kargaltsev, O., \& Brisken, W.~F.\ 2008, \apj, 675, 683
752:
753: \bibitem[Possenti et al.(2002)]{Possenti02}
754: Possenti, A., Cerutti, R., Colpi, M., \& Mereghetti, S.\ 2002, \aap, 387, 993
755:
756: \bibitem[Romani et al.(2005)]{Romani05}
757: Romani, R.~W., Ng, C.-Y., Dodson, R., \& Brisken, W.\ 2005, \apj, 631, 480
758:
759: \bibitem[Rybicki \& Lightman(1979)]{Rybicki79}
760: {{Rybicki}, G.~B., \& {Lightman}, A.~P.} 1979, Radiative Processes in Astrophysics (New York: Wiley)
761:
762: \bibitem[Seward \& Wang(1988)]{Seward88}
763: Seward, F.~D., \& Wang, Z.-R.\ 1988, \apj, 332, 199
764:
765: \bibitem[Slane et al.(2000)]{Slane00}
766: Slane, P., Chen, Y., Schulz, N.~S., Seward, F.~D., Hughes, J.~P., \& Gaensler, B.~M.\ 2000, \apjl, 533, L29
767:
768: \bibitem[Slane et al.(2008)]{Slane08}
769: Slane, P., Helfand, D.~J., Reynolds, S.~P., Gaensler, B.~M., Lemiere, A.,
770: \& Wang, Z.\ 2008, \apjl, 676, L33
771:
772: \bibitem[\protect\citeauthoryear{{Trimble}, V.}{1973}]{Trimble73}
773: {{Trimble}, V.} 1973, \pasp, 85, 579
774:
775: %\bibitem[\protect\citeauthoryear{Weekes et al.}{1989}]{Weekes89}
776: %{{Weekes}, T.~C., et al.} 1989, \apj, 342, 379
777:
778: \bibitem[Willingale et al.(2001)]{Willingale01}
779: Willingale, R., Aschenbach, B., Griffiths, R.~G., Sembay, S., Warwick, R.~S., Becker, W., Abbey, A.~F., \& Bonnet-Bidaud, J.-M.\ 2001, \aap, 365, L212
780:
781: \end{thebibliography}
782:
783:
784: \end{document}
785: