1: \documentclass[12pt,preprint]{aastex}
2: \slugcomment{Accepted for the publication in the {\it Astrophysical Journal}}
3: \pagenumbering{arabic}
4:
5: \shortauthors{Mori}
6: \shorttitle{X-ray Spectral Analysis of The Crab Nebula}
7:
8: \begin{document}
9: \title{Spatial Variation of the X-ray spectrum of the Crab Nebula}
10:
11: \author{Koji Mori and David N.\ Burrows}
12: \affil{Department of Astronomy and Astrophysics, Pennsylvania State
13: University, 525 Davey Laboratory, University Park, PA. 16802, USA}
14: \email{mori@astro.psu.edu}
15:
16: \author{J.\ Jeff Hester}
17: \affil{Department of Physics and Astronomy, Box 871504, Arizona State
18: University, Tyler Mall, Tempa, AZ. 85287-1504, USA}
19:
20: \author{George G.\ Pavlov}
21: \affil{Department of Astronomy and Astrophysics, Pennsylvania State
22: University, 525 Davey Laboratory, University Park, PA. 16802, USA}
23:
24: \author{Shinpei Shibata}
25: \affil{Department of Physics, Yamagata University, Yamagata
26: 990-8560, Japan}
27:
28: \and
29: \author{Hiroshi Tsunemi}
30: \affil{Department of Earth and Space Science, Graduate School of
31: Science, Osaka University, 1-1 Machikaneyama, Toyonaka, Osaka
32: 560-0043, Japan}
33:
34: \begin{abstract}
35:
36: We present spectral analysis of the Crab Nebula obtained with the {\it
37: Chandra} X-ray observatory. The X-ray spectrum is characterized by a
38: power-law whose index varies across the nebula. The variation can be
39: discussed in terms of the particle injection from the pulsar in two
40: different directions: the equatorial plane containing the torus and the
41: symmetry axis along the jet. In the equatorial plane, spectra within the
42: torus are the hardest, with photon index $\alpha \approx 1.9$, and are
43: almost independent of the surface brightness. At the periphery of the
44: torus, the spectrum gradually softens in the outer, lower surface
45: brightness regions, up to $\alpha \approx 3.0$. This indicates that
46: synchrotron losses become significant to X-ray emitting particles at the
47: outer boundary of the torus. We discuss the nature of the torus,
48: incorporating information from observations at other wavelengths.
49: Spectral variations are also seen within the southern jet. The core of
50: the jet is the hardest with $\alpha \approx$ 2.0, and the outer sheath
51: surrounding the core becomes softer with $\alpha$ up to 2.5 at the
52: outermost part. Based on the similarity between the spectra of the jet
53: core and the torus, we suggest that the electron spectra of the
54: particles injected from the pulsar are also similar in these two
55: different directions. The brightness ratio between the near and far
56: sides of the torus can be explained by Doppler boosting and relativistic
57: aberration; however, the observed ratio cannot be derived from the
58: standard weakly magnetized pulsar wind model. We also found a site where
59: an optical filament comprised of supernova ejecta is absorbing the soft
60: X-ray emission ($<$ 2 keV).
61:
62: \end{abstract}
63:
64: \keywords {ISM: individual (Crab Nebula) --- supernova remnants ---
65: X-rays: ISM}
66:
67: \section {\label {sec:intro} INTRODUCTION}
68:
69: The observed pulsating radiation is only a small fraction of a
70: pulsar's spin-down energy. The bulk of this energy is carried off in
71: the form of a magnetized, relativistic pulsar wind. This pulsar wind
72: is randomized by a shock where its pressure balances that of the
73: ambient material (Rees \& Gunn 1974; Kennel \& Coroniti 1984). Diffuse
74: synchrotron radiation is expected to arise from the randomized pulsar
75: wind, forming a {\it pulsar wind nebula} (PWN). Although this general
76: idea is widely accepted, details of the formation are still
77: controversial. For example, it is a long-standing problem that the
78: ratio of the Poynting flux to the particle kinetic flux just upstream
79: of the shock, $\sigma$, is much smaller than 10$^{-2}$, in spite of
80: expectations that the ratio at the light cylinder is larger than unity
81: (e.g., Arons 2002). Additionally, the formation of the torus and the
82: collimation of the pulsar wind to form a jet along the rotational axis
83: are quite curious but unresolved problems.
84:
85: An X-ray observation is a powerful probe to peer into the inner PWN,
86: whose spectral and morphological properties provide essential
87: information to answer such questions. Recent {\it Chandra} and {\it
88: XMM-Newton} observations increased the number of objects where
89: spatially resolved X-ray spectroscopy within the PWN can be
90: performed. In a number of PWNe, the spectrum of the outer edges of the
91: PWN is observed to be softer than the spectrum of the center (e.g.,
92: G21.5$-$0.9, Slane et al.\ 2000, Safi-Harb et al.\ 2001; 3C 58,
93: Bocchino et al.\ 2001; G0.9$+$0.1, Porquet et al.\ 2003; G54.1$+$0.3,
94: Lu et al.\ 2002; Vela, Kargaltsev \& Pavlov 2004). Since the
95: synchrotron energy losses are proportional to the square of the particle
96: energy, the higher energy particles have shorter life times and the
97: spectrum becomes softer as they move down the post-shock flow, which
98: is in general agreement with these observations.
99: Such spatial variation of the spectrum can be used to
100: verify theoretical models and to provide new diagnostics for the study
101: of PWNe.
102:
103: The Crab Nebula has been one of the best examples for investigating a
104: PWN. Its X-ray brightness and large apparent structure including the
105: torus (Aschenbach \& Brinkmann 1975) and jets (Brinkmann, Aschenbach, \&
106: Langmeier 1985) make it an ideal target for spatially resolved X-ray
107: spectroscopy. Previous {\it Chandra} and {\it XMM-Newton} observations
108: have also shown spectral softening of the Crab (Weisskopf et al.\ 2000;
109: Willingale et al.\ 2001). Here, we present spectral analysis results
110: from multiple observations of the Crab Nebula with the Advanced CCD
111: Imaging Spectrometer (ACIS) on board {\it Chandra}, which are derived
112: from the monitoring program described by Hester et al.\ (2002).
113:
114: \section{\label{sec:obs} OBSERVATION AND ANALYSIS}
115:
116: Owing to the X-ray brilliance of the Crab Nebula, observations with
117: the {\it Chandra} ACIS instrument are technically challenging. The high
118: count rate generally causes two problems in the data acquisition:
119: telemetry saturation and event pile-up (caused by more than one photon
120: depositing charge in a given pixel during a CCD exposure, thereby
121: compromising our ability to measure the energy of each individual
122: photon). Although telemetry saturation results in loss of observing
123: efficiency, it does not lead to biased estimates of physical
124: parameters (e.g., count rate) because events are discarded in a unit
125: of frames, not event by event. On the other hand, pile-up makes
126: interpretation of spectral results problematic, since it converts two
127: photons into a single ``event'' which has apparent energy equal to the
128: sum of the photon energies. This results in underestimated count rate
129: and hardens the detected spectrum, as discussed in Weisskopf et al.\
130: (2000).
131:
132: We observed the Crab Nebula eight times (Table~\ref{tbl:log}), using
133: an instrumental mode of ACIS-S different from that of Weisskopf et
134: al.\ (2000) in order to reduce pile-up. We adopted a frame time of 0.2
135: sec instead of the standard frame time of 3.2 seconds, but we did not
136: utilize the grating to reduce count rate. This reduced the counts
137: pixel$^{-1}$ frame$^{-1}$ by a factor of 2.3 with respect to the data
138: of Weisskopf et al.\ (2000), while improving statistics by increasing
139: the total number of counts by one order of magnitude for a similar
140: effective exposure time. Since the Crab Nebula is a diffuse source,
141: the event pile-up was suppressed by more than a factor of 2.3. The
142: observational parameters (e.g., telemetry format of graded mode, no
143: dithering) were the same through the eight observations. A combined
144: image from the last seven observations is shown in
145: Figure~\ref{fig:index_map}a (the first observation did not include the
146: whole remnant due to an instrument setup error). This image is an
147: order of magnitude deeper than the image of Weisskopf et al.\ (2000).
148:
149: In order to study X-ray spectral variations within the Crab Nebula, we
150: divided the whole image into $2.^{\!\!\prime\prime}5 \times
151: 2.^{\!\!\prime\prime}5$ square regions, from which spectra were
152: extracted. The eight data sets were combined in the analysis in order
153: to provide each region with enough statistics. We excluded regions
154: dominated by scattered photons (less than 2000 counts per analysis
155: region) and/or by trailing (or ``out of time'') events. Trailing
156: events, which are caused by photons detected during the CCD readout,
157: can be seen not only as the line emanating from the pulsar along the
158: detector readout direction (which is almost east-west for all of our
159: observations), but also as diffuse emission which dominates the faint
160: regions to the east and west of the nebula due to the enormous
161: brilliance of the Crab PWN. The number of the trailing events in a
162: given pixel can be estimated by multiplying the total number of events
163: integrated along the readout direction by the ratio of the charge
164: transfer row period to the frame exposure time (this ratio is $1.89
165: \times 10^{-4}$). We exclude regions where more than 20\% of the total
166: events recorded in that region are due to the trailing events.
167:
168: We fitted the 2074 spectra with a power-law model in the 0.5--8 keV
169: energy band. We included galactic absorption with a fixed $N_{\rm H}$
170: of $3.2 \times 10^{21} $cm$^{-2}$, which is derived from a fit of the
171: faintest region where pile-up is negligible. The recently discovered
172: quantum efficiency decay of ACIS\footnote{See
173: http://asc.harvard.edu/cal/Acis/Cal\_prods/qeDeg/index.html} is taken
174: into account\footnote{See
175: http://www.astro.psu.edu/users/chartas/xcontdir/xcont.html}.
176: Figure~\ref{fig:correlation_plot}a shows a correlation plot of
177: apparent photon index versus observed surface brightness. The
178: statistical errors of the apparent photon indices are quite small.
179: They can be roughly expressed empirically as $0.01 \times I^{-0.5}$
180: where $I$ is the observed surface brightness in unit of counts
181: s$^{-1}$ arcsecond$^{-2}$ (90\% confidence level).
182:
183: In spite of the reduction in counts per frame, many of the spectra are
184: still distorted by event pile-up. The spectral parameters derived from
185: a fit of the whole nebula spectrum with a power-law plus galactic
186: absorption model are summarized in Table~\ref{tbl:whole_fit} (uncorrected values),
187: as are the canonical values for comparison. The pile-up results in a lower
188: apparent spectral index and lower apparent intensity than the
189: canonical values. The observed intensity normalization is about 84\%
190: of that estimated from the canonical values with the instrumental
191: response. Pile-up has the effect of reducing photon index in regions
192: of increased surface brightness, which is the same general trend seen
193: in the data of Figure~\ref{fig:correlation_plot}a. The pile-up
194: effects must be corrected in order to determine the nature of any real
195: spectral variations across the Crab nebula.
196:
197: In order to estimate how pile-up distorts the incident spectrum, we
198: utilized the spectral simulator tool LYNX (Chartas et al.\ 2000). LYNX
199: traces the propagation of individual photons through the mirror and
200: ACIS via the raytrace tool MARX (Wise et al.\ 1997) and the PSU ACIS
201: Monte Carlo CCD simulator (Townsley et al.\ 2001), respectively. Since
202: LYNX takes into account the possible overlap of the resulting charge
203: clouds within each CCD exposure frame, spectra affected by pile-up can
204: be simulated for both point sources and diffuse sources. We simulated
205: the observed Crab spectrum for several different values of true
206: surface brightness and for true photon indices of 1.5, 2.0, and 2.5.
207: The results are superposed on the data in
208: Figure~\ref{fig:correlation_plot}a. The expected reduction of
209: apparent photon index in higher surface brightness regions is
210: evident. However, it is also clear that the variations in observed
211: photon indices greatly exceed those expected from pile-up, and there
212: are certainly real spectral variations across the Crab Nebula, as
213: described by Weisskopf et al.\ (2000).
214:
215: We used the LYNX simulation results to correct the data points in
216: Figure~\ref{fig:correlation_plot}a for the photon index and surface
217: brightness errors induced by the pile-up effects. For each data
218: point, we interpolated the simulated data points to the observed
219: surface brightness and photon index and calculated the corresponding
220: true surface brightness and photon index.
221: Figure~\ref{fig:correlation_plot}b presents a correlation plot of the
222: corrected data points. This simulation was intended to give
223: corrections on the data points in Figure~\ref{fig:correlation_plot}a
224: and was not adjusted to correct the whole nebula spectrum in order to
225: match to the canonical spectral parameters.
226: However, we used the corrected photon indices
227: and surface brightnesses of each data point in
228: Figure~\ref{fig:correlation_plot}b to calculate a corrected spectrum
229: for the whole nebula and we present the results in Table~\ref{tbl:whole_fit}
230: (corrected values). The photon index and the
231: normalization factor of the corrected spectrum agree well with the
232: canonical values within the errors, which strongly supports the
233: success of our pile-up corrections. Hereafter, we use the terms ``photon
234: index'' and ``surface brightness'' to mean these corrected values.
235:
236: We should note a limitation of our simulation. Our corrections are
237: based on modeling of a uniform diffuse source. We apply them to each
238: $2.^{\!\!\prime\prime}5 \times 2.^{\!\!\prime\prime}5$ square region
239: under the reasonable assumption that the surface brightness within
240: each region is uniform. Therefore, application of this technique to
241: small complicated structures like the inner ring or the
242: relativistically expanding wisps requires more complex modeling which
243: is beyond the scope of this work. The systematic errors of the photon
244: index mainly come from this assumption. Judging from the results of
245: the simulation shown in Figure~\ref{fig:correlation_plot}a and typical
246: surface brightness fluctuations of $\sim 10\%$ within an analysis
247: region, the systematic errors in photon index are $\sim$ 0.05 and
248: dominate the statistical errors.
249:
250: Figure~\ref{fig:index_map}b shows a map of the photon index. The
251: structure in the torus is more symmetrical about the pulsar in the
252: photon index map than in the broad-band image
253: (Fig.~\ref{fig:index_map}a). While Doppler boosting and relativistic
254: aberration brighten the northwestern portion of the torus (Pelling et
255: al.\ 1987) in the Figure~\ref{fig:index_map}a image, by contrast there
256: is relatively little variation of the photon index within the
257: torus. The hardest structures in the nebula, with photons indices as
258: low as 1.8, are the inner ring and portions of the torus, including
259: the circular structures seen at each extremity of the torus discovered
260: in the first {\it Chandra} observation (Weisskopf et al.\
261: 2000). However, these regions contain small complex structures and the
262: hardest photon index should be re-examined using data set free from
263: pile-up. The entire torus is quite hard ($\alpha \approx 1.9$)
264: compared with the outer portions of the nebula. The southern jet is
265: also relatively hard ($\alpha \approx 2.0$), whereas the northern
266: counter jet is significantly softer ($\alpha \approx 2.25$). The
267: bright region around the counter jet to the northwest of the torus,
268: hereafter called the ``umbrella'' because of its shape, has even
269: softer emission ($\alpha
270: \approx 2.5$). Photon indices as large as 3.0 are found in the outer
271: peripheral portions of the nebula.
272:
273: \section{\label{sec:analysis} RESULTS}
274:
275: We studied the spatial variation of the photon index over the nebula.
276: As a first step, we divided the nebula into four prominent regions:
277: the torus, the umbrella, the jet, and the peripheral region. We took
278: the torus region as an ellipse bounding the far side of the torus. We
279: also added two circles at each extremity of the ellipse to include the
280: circular structures in the torus region. The rest of the bright nebula
281: was taken as the umbrella region. The jet region covers only the
282: southern jet. The peripheral region includes the faint emission
283: surrounding the torus and umbrella regions except for the
284: jet. Figure~\ref{fig:color_plot} shows that those four regions occupy
285: separate regions in the plot of photon index against surface
286: brightness, with the umbrella region significantly overlapping the
287: torus and periphery. The color coding is shown in the inset figure.
288:
289: The data points of the region within the torus (red) form a band with
290: nearly constant photon index of $\approx 1.9$ over a wide range of
291: surface brightness (more than a factor of 4). Although the spectral
292: index increases slightly for surface brightness below 0.4 counts
293: s$^{-1}$ arcsecond$^{-2}$, this result shows that our spectral
294: corrections were successful in removing the spectral distortion due to
295: pile-up.
296:
297: The data points of the peripheral region (green) dominate the low
298: surface brightness / high photon index region. The photon indices at
299: the periphery of the bright nebula are strongly anticorrelated to the
300: surface brightness, even after correcting for pile-up effects. This
301: makes a marked contrast to the constancy of the photon index in the
302: torus, especially considering that the variation of the surface brightness in the
303: periphery is 1/3 of that in the torus.
304:
305: The photon indices of the jet (blue) are generally small compared to
306: those of the peripheral regions although their surface brightnesses
307: are comparable. The jet also shows a strong anticorrelation between
308: surface brightness and photon index, again suggesting softening of
309: the underlying electron spectrum at the boundaries of the jet.
310:
311: The remaining population corresponds to the umbrella region
312: (yellow). Although distributed over a similar wide range of surface
313: brightness, the photon indices in the umbrella are generally greater
314: than those of the torus region. In fact, the umbrella and torus merge
315: both spatially and spectrally. The umbrella is seen only at the
316: northern part of the nebula and has no counterpart with similar
317: brightness in the south, where the jet is much more distinct.
318:
319: \section{\label{sec:disc} DISCUSSION}
320:
321: \subsection{Photon Index Variation}
322:
323: We found photon index variations across the nebula on arcsecond scales
324: which are similar to those reported by Weisskopf et al.\ (2000) and
325: Willingale et al.\ (2001) on larger angular scales. The photon index
326: variations should be discussed along the two different directions of
327: particle injection from the pulsar: the direction in the equatorial
328: plane containing the torus, and the direction of the jet along the
329: symmetry axis. We first consider the equatorial plane.
330:
331: Although the standard model of the Crab Nebula given by Kennel and
332: Coroniti (1984) assumes a spherical symmetry, it can be adapted to
333: regions of limited solid angle in the equatorial plane. Across the
334: equatorial plane, the spectrum is almost constant to the outer
335: boundary of the torus, but softens significantly beyond this boundary.
336: In the standard picture, the outer boundary of the torus is
337: interpreted as a synchrotron burn-off boundary where the synchrotron
338: losses become significant to the X-ray emitting particles. Until the
339: outward post-shock flow reaches the torus, the synchrotron critical
340: energy, $E_c$, must be greater than 10 keV and falls outside the {\it
341: Chandra} X-ray band (0.5--8 keV). Here, the synchrotron critical
342: energy is defined as the energy at which a spectral break occurs in
343: the power-law spectrum.
344:
345: Due to the shorter synchrotron life time of higher energy particles,
346: the value of $E_c$ for a given flow element decreases with time, hence
347: with distance from the shock. Kennel and Coroniti (1984) gave $E_c$ as
348: a function of radius from the pulsar and explained the energy
349: dependence of the overall nebular extent from optical to hard X-ray
350: ($\sim$ 60keV). In this picture, if the outer boundary of the X-ray
351: torus is where the synchrotron burn-off takes place in the X-ray band,
352: then, in the same manner, larger tori are expected in the optical and
353: radio bands. However, it is remarkable that the size of the torus
354: does not change significantly over a wide bandpass. Optical images
355: clearly show the torus and related circular structures with the same
356: size as the X-ray images, albeit with less contrast (Hester et al.\
357: 2002). Temporal variations between two radio images taken 2 months
358: apart also revealed a torus-like elliptical structure of the same size
359: (Bietenholz et al.\ 2001). In hard X-ray images (22--43keV and
360: 43--64keV; Pelling et al.\ 1987), again the nebular extent is similar
361: to that of the torus seen in the soft X-ray band, in contradiction to
362: an expectation that they should be smaller. Additionally, Shibata et
363: al.\ (2003) showed that the observed surface brightness decreases as a
364: function of distance beyond the torus much faster than expected from
365: the Kennel \& Coroniti model. Therefore, we suggest that the torus
366: seen from radio to hard X-rays is not a simple result of synchrotron
367: burn-off in a radial post-shock flow. The standard Kennel \&
368: Coroniti flow solution should be modified to include the torus
369: structure observed in common over the wide bandpass. Detailed
370: small-scale comparison of the synchrotron burn-off boundary between
371: the X-ray and optical bands will reveal the true nature of the torus,
372: which is the subject of a subsequent paper.
373:
374: Figure~\ref{fig:jet} shows a close-up view of the photon index map
375: around the southern jet. It shows that the spectral softening occurs
376: from the central core of the jet, with $\approx \alpha$ of 2.0, to the
377: outer fainter sheath regions, with $\approx \alpha$ of 2.5.
378: Considering the cylindrical structure of the jet, the observed
379: spectrum of the jet core is integrated through this structure and
380: includes contributions from the softer outer sheath regions. The true
381: photon index of the jet core is therefore even harder than 2.0, and
382: must be close to that seen within the torus. Willingale et al.\
383: (2001), using {\it XMM-Newton}, reported that the jet has a spectrum
384: steeper than the torus by 0.3, implying that the corresponding
385: electron spectra differ by 0.6. However, this conclusion is based on the spectrum
386: integrated all over the jet. The similarity of the jet core spectrum
387: to that of the torus indicates that the electron spectra are actually rather
388: similar in the two different directions of particle injection from the
389: pulsar. This may give a strong constraint on theoretical models of
390: magnetic collimation.
391:
392: We estimate the fraction of the X-ray luminosity that comes from the
393: jets as $\sim 4$\%, with $\sim$3\% from the southern jet and $\sim$1\%
394: from the counter jet. The bright umbrella makes it difficult to estimate
395: the luminosity from the counter jet and results in a relatively large
396: uncertainty in its luminosity, although the discussion here is little
397: affected. The jet luminosity fraction is close to the ratio of the
398: volumes of the jets and the torus,
399:
400: \begin{equation}
401: \frac{2 \, \int_{0}^{R_{j}} \int_{0}^{\theta_{j}} \int_{0}^{2\pi}
402: r^{2}\,\sin\theta\, dr\, d\theta \, d\phi} {\int_{0}^{R_{t}}
403: \int_{\frac{\pi}{2} - \theta_{t}}^{\frac{\pi}{2} + \theta_{t}} \int_{0}^{2\pi}
404: r^{2} \sin\theta\, dr\, d\theta\, d\phi} =
405: \frac{1-\cos\theta_{j}}{\sin\theta_{t}}\left(\frac{R_{j}}{R_{t}}\right)^{3},
406: \label{eq:ratio}
407: \end{equation}
408:
409: \noindent
410: where $R_{t},\ R_{j},\ \theta_{t}$ and $\theta_{j}$ are the radii and
411: semi-opening angles of the torus and the jet. From
412: Figure~\ref{fig:index_map}a, we see that $R_{t} \approx R_{j}$ and
413: that $\theta_{t}$ and $\theta_{j}$ are 12$^{\circ}$ and 8$^{\circ}$,
414: respectively. Then, Equation (\ref{eq:ratio}) gives a ratio of about
415: 5\% for these volumes. Considering the uncertainties in the opening angles and model
416: assumptions, the fraction of the total X-ray luminosity in the jet is
417: comparable to its volume fraction, indicating that the volume
418: emissivities of the jet and the torus are similar, even though they
419: represent orthogonally different directions of particle injection from
420: the pulsar. Taking $R_{t}=R_{j}=50$ arcseconds ($\simeq$ 0.5 pc
421: assuming the distance of 2 kpc to the Crab Nebula) in the above
422: equation (see Fig.~\ref{fig:index_map}a) and an X-ray luminosity in
423: the 0.5--8 keV band of 1.3 $\times 10^{37}$ erg s$^{-1}$, which is
424: estimated from the canonical spectral parameters, we obtain an average
425: volume emissivity of $\sim$ 4 $\times 10^{-18}$ erg s$^{-1}$
426: cm$^{-3}$.
427:
428: \subsection{Surface Brightness Ratio between Near and Far Sides of the Torus}
429:
430: The north-western (NW) part of the torus, which is the near side, is
431: much brighter than the south-eastern (SE) part. The post-shock flows
432: in the NW and SE parts are relativistically approaching and receding
433: from us, respectively. Therefore, those apparent surface brightnesses
434: are subject to Doppler boosting and relativistic
435: aberration. Considering these effects, the surface brightness ratio of
436: the NW part to the SE part should be given by
437:
438: \begin{equation}
439: \left(\frac{1+\beta \, \cos\theta}{1-\beta \, \cos\theta}
440: \right)^{\alpha+2},
441: \end{equation}
442:
443: \noindent
444: where $\theta$ is the angle measured in the observer's frame between the
445: direction of the flow and the line of sight, $\alpha$ is the photon
446: index, and $\beta$ is the flow speed at the torus normalized to the
447: speed of light (Rybicki \& Lightman 1979).
448:
449: Hester et al.\ (2002) showed that the inner ring, which is inside the
450: torus, is highly variable and that wisps moving at about 0.5$c$ are
451: emerging from the inner ring. This provides strong evidence that the
452: inner ring represents the pulsar wind termination shock, whereas the
453: torus is a downstream structure. Since the post-shock flow is
454: sub-sonic, the flow speed decreases as a function of distance. Hester
455: et al.\ (2002) and Mori (2002) showed that the {\it projected} speed
456: of the far side of the torus is 0.03$c$--0.1$c$, measured directly
457: east of the pulsar. The relatively large uncertainty is due to the
458: small angular displacement of the far side of the torus. Considering
459: relativistic effects and the geometry of the torus, where the
460: rotational-symmetry axis lies about 58$^{\circ}$ west of north and
461: about 25$^{\circ}$ out of the celestial plane, these speeds correspond
462: to {\it true} speeds of 0.07$c$--0.25$c$ (0.15$c$). The value in
463: parentheses is calculated for the central value of the {\it projected}
464: speed range, 0.065$c$. With $\alpha = 1.9$, these values should result
465: in a surface brightness ratio of 1.6--6.3 (2.9). The observed surface
466: brightness ratio in Figure~\ref{fig:index_map}a is about 3.4, which
467: agrees with the speed range inferred from above discussion.
468:
469: Kennel \& Coroniti (1984) deduced the post-shock flow speed as a
470: function of $z$, the distance from the pulsar normalized to the shock
471: radius, for different values of the magnetization parameter $\sigma$,
472: the ratio of the Poynting flux to the particle kinetic flux upstream
473: of the shock. From Figure~\ref{fig:index_map}a, $z$ is about 2.7 at
474: the distance where we measured the speed of the torus. Kennel \&
475: Coroniti suggested that $\sigma$ is about 0.003 based on theoretical
476: arguments, which would correspond to a velocity of about 0.05$c$ at
477: the torus (from Fig.~3 of Kennel \& Coroniti), far below our measured
478: value. Shibata et al.\ (2003) constructed a two-dimensional model
479: image of the Crab Nebula in the framework of the Kennel \& Coroniti
480: model, and found that the model surface brightness ratio between the
481: near and far sides of the torus is about 1.3 for $\sigma = 0.003$,
482: also much lower than the observed value. By contrast, our measured
483: speed of $0.07c$--$0.25c$ (0.15$c$), which predicts the correct
484: surface brightness ratio, corresponds to $\sigma = 0.01$--0.13 (0.05).
485: The theoretical speed and brightness ratio values can be reconciled
486: with the measured ones if we assume a $\sigma$ of about 0.05.
487:
488: It is worth noting that departures from spherical symmetry and proper
489: treatment of radiative losses in the flow can both affect this result.
490: It is likely that the flow is confined to a narrow range of latitudes
491: around the equatorial plane upstream of the shock (the inner ring).
492: After the shock the flow diverges out of the plane more rapidly than a
493: radial flow to form the broad torus. This should result in more rapid
494: deceleration than in a spherical model. Any radiative losses of
495: particle pressure may also lead to greater compression and hence more
496: rapid deceleration. In both cases, however, we might expect flow
497: velocities {\it slower} than those predicted by Kennel \& Coroniti,
498: for a given $\sigma$. As a result, either of these effects could
499: require an even greater $\sigma$ to account for the observations.
500:
501:
502: \subsection{\label{sec:nh} Correlation between Column Density and Ejecta}
503:
504: There are some regions exhibiting unusually small photon index in
505: Figure~\ref{fig:color_plot}, which are enclosed by the red dashed
506: circle. The corresponding positions in the nebula are shown in black
507: in the inset, also enclosed by a red dashed
508: circle. Figure~\ref{fig:spectrum_comparison} presents two spectra
509: extracted from one of the small photon index regions and its
510: neighboring region. The spectra are identical above about 2 keV, and
511: the comparison of the spectra clearly reveals that the small apparent
512: photon index is actually caused by a deficit of soft emission due to
513: increased absorption in this region. Refitting the spectra from those regions with
514: $N_{\rm H}$ treated as a free parameter, we obtained a similar photon index and a
515: higher $N_{\rm H}$, by $\sim 7 \times 10^{20}$ cm$^{-2}$, compared to
516: the neighboring region.
517:
518: The Crab Nebula appears to contain a number of embedded filaments in
519: optical narrow-band emission-line images (e.g., Sankrit \& Hester
520: 1997). The optical filaments are thought to be the result of a
521: Rayleigh-Taylor instability between the light pulsar wind nebula and
522: the dense ejecta of the supernova explosion (Hester et al.\ 1996).
523: The J2000 position of the region showing higher $N_{\rm H}$, $\alpha=$
524: 05$^{\mathrm h}$34$^{\mathrm m}$29\fs6,
525: $\delta=+$22\arcdeg00\arcmin30\farcs4, coincides with knotty structure
526: in the most prominent optical filament, which is often referred to as
527: part of the ``High-Helium Band'' (MacAlpine et al.\ 1989). The
528: emission lines from the High-Helium Band are blueshifted, indicating that this
529: filament is located between the synchrotron nebula and us. In an
530: optical continuum image, the knotty structure is visible as a
531: ``shadow'' because it consists of dense dust which absorbs the optical
532: synchrotron continuum from the PWN (e.g., Blair et al.\
533: 1997). Similarly, the higher $N_{\rm H}$ we find is likely due to
534: absorption of soft X-rays by the dust and gas in this knotty
535: structure. Assuming that the thickness of this filament is
536: 5$^{\prime\prime}$ based on the optical emission line images (Blair et
537: al.\ 1997), which corresponds to $\sim 1.5 \times 10^{17}$ cm, the
538: density of this filament corresponding to the observed X-ray
539: absorption column density is $\sim 5 \times 10^{3}$ cm$^{-3}$. The
540: excess absorption can be seen only around the knotty structures all
541: through this filament.
542:
543: We searched for other examples of X-ray absorption from the ``shadow'' features in the optical
544: continuum image (Hester et al.\ in preparation), from which we can
545: pick up only dense dust core in filaments on the near side of the
546: nebula. There are only a few such shadows within the X-ray
547: nebula. The number of corresponding analysis regions is about 10 out of
548: 2074 (in addition to the regions discussed above). The second noticeable
549: shadows are located to the east of the pulsar, overlapping the edge of the
550: torus. Excess absorption was again observed, but the amount was
551: $\sim 3 \times 10^{20}$ cm$^{-2}$. The excess absorption related to
552: other shadows is similar or less. Therefore, our results are little
553: affected by soft X-ray absorption due to the filaments.
554:
555: It is conceivable that our results are affected by intrinsic column
556: density variation due to not only the optical filaments but also
557: arcsecond scale variation of galactic absorption. We assessed this
558: effect as follows. In Figure~\ref{fig:color_plot}, the photon indices
559: of the data points of the torus (red) having surface brightness higher
560: than 0.5 counts s$^{-1}$ arcsec$^{-2}$ are distributed around the mean
561: of 1.91 with the root mean square of 0.07. The statistical errors are
562: small, so the observed scatter is dominated by systematic errors and
563: by real variations in the power law index. We can obtain an upper
564: limit on possible absorption variations across the nebula if we assume
565: that this scatter is produced by variations in $N_{\rm H}$, leading to
566: an estimate of the variation in $N_{\rm H}$ of $2 \times 10^{20}$
567: cm$^{-2}$.
568:
569: \section{\label{sec:summary} SUMMARY}
570:
571: We have shown spatial variations of the X-ray spectrum of the Crab
572: Nebula in terms of photon index, at an angular scale of
573: arcseconds. The variations can be viewed in two different directions
574: of the particle injection from the pulsar.
575:
576: Across the equatorial plane, the spectrum is almost constant to the
577: outer boundary of the torus, with photon index $\alpha$ $\approx$ 1.9
578: regardless of the surface brightness. It softens significantly, up to
579: $\alpha$ $\approx$ 3.0, in the outer, fainter peripheral region. This
580: seems qualitatively consistent with the previous suggestions that the outer boundary
581: of the torus is interpreted as a synchrotron burn-off boundary where
582: the synchrotron losses become significant to X-ray emitting
583: particles. However, the fact that structures similar to the torus are
584: seen at other wavelengths indicates that the torus is not a simple
585: result of synchrotron burn-off.
586:
587: Within the southern jet, photon index variations are also seen: the
588: spectral softening takes place from the central core to the outer
589: sheath. The photon index at the central core is almost the same as
590: that of the torus. This indicates that the electron spectra are
591: similar in the two different directions of the particle injection from
592: the pulsar. We also found that the volume emissivities of the jet and
593: the torus are similar.
594:
595: Assuming that the brightness difference between the near and far sides
596: of the torus is caused by Doppler boosting and relativistic
597: aberration, the ratio can be explained by the observed speed of the
598: downstream flow at the torus. However it cannot be obtained from the
599: so-called weakly magnetized pulsar wind with $\sigma \approx 0.003$,
600: as suggested by Kennel \& Coroniti (1984).
601:
602: Finally, we found that an optical filament comprised of supernova ejecta
603: surrounding the pulsar wind nebula is absorbing the soft X-ray
604: emission from a small portion of the X-ray nebula.
605:
606: \acknowledgments
607:
608: The authors thank G.\ Chartas for his extensive cooperation on
609: simulation using LYNX. We also thank the anonymous referee for comments
610: which significantly improved the final text of this paper. K.\ M.\
611: acknowledges the support of JSPS through the fellowship for research
612: abroad. The work of G.\ G.\ P.\ was partially supported by NASA grant
613: NAG5-10865. This work was supported by the NASA through Chandra Awards
614: G00-1032B and G01-2076B, issued by the Chandra X-ray Observatory Center,
615: which is operated by the Smithsonian Astrophysical Observatory for and
616: on behalf of NASA under contract NAS8-39073.
617:
618:
619: \begin{thebibliography}{}
620:
621: \bibitem[Aschenbach \& Brinkmann 1975]{} Aschenbach, B., \& Brinkmann,
622: W.\ 1975, \aap, 41, 147
623:
624: \bibitem[Arons 2002]{} Arons, J.\ 2002, in Neutron Stars in Supernova
625: Remnants, ed. P.~O.\ Slane \& B.~M.\ Gaensler, (San Francisco: ASP), 71
626:
627: \bibitem[Blair et al.\ 1997]{} Blair, W.~P., Davidson, K., Fesen, R.~A.,
628: Uomoto, A., MacAlpine, G.~M., \& Henry,
629: R.~B.~C.\ 1997, \apjs, 109, 473
630:
631: \bibitem[Brinkmann, Aschenbach, \& Langmeier 1985]{} Brinkmann, W.,
632: Aschenbach, B., \& Langmeier,
633: A. 1985, Nature, 313, 662
634:
635: \bibitem[Chartas et al.\ 2000]{} Chartas, G., Worrall, D.~M., Birkinshaw,
636: M., et al.\ 2000, \apj, 542, 655
637:
638: \bibitem[Bietenholz et al.\ 2001]{} Bietenholz, M.~F., Frail, D.~A., \&
639: Hester, J.~J.\ 2001, \apj, 560, 254
640:
641: \bibitem[Bocchino et al.\ 2001]{} Bocchino, F., Warwick, R.~S., Marty,
642: P., Lumb, D., Becker, W., \& Pigot, C.\ 2001,
643: \aap, 369, 1078
644:
645: %\bibitem[Fester \& Blair 1990]{} Fesen, R.~A. \& Blair W.~P.\ 1990, \apj, 351, 45
646:
647: \bibitem[Hester et al.\ 1996]{} Hester, J.~J., Stone, J.~M., Scowen, P.~A.,
648: et al.\ 1996, \apj, 456, 225
649:
650: \bibitem[Hester et al.\ 2002]{} Hester, J.~J., Mori, K., Burrows,
651: D.~N. et al.\ 2002, \apjl, 577, L49
652:
653: \bibitem[Kargaltsev \& Pavlov 2004]{} Kargaltsev, O. \& Pavlov, G.~G.
654: 2004, in Young Neutron Stars and Their Environments,
655: IAU Symposium 218, ASP Conference Proceedings,
656: eds F.\ Camilo and B.~M.\ Gaensler, in press (astro-ph/0310767)
657:
658: \bibitem[Kennel \& Coroniti 1984]{} Kennel, C.~F., \& Coroniti,
659: F.~V. 1984, \apj, 283, 694
660:
661: \bibitem[Kuiper et al.\ 2001]{} Kuiper, L., Hermsen, W., Cusumano, G.,
662: Diehl, R., Schnfelder, V., Strong, A., Bennett, K.,
663: McConnell, M.~L.\ 2001, \aap, 378, 918
664:
665:
666: \bibitem[Lu et al.\ 2002]{} Lu, F.~J., Wang, Q.~D., Aschenbach, B., Durouchoux, P., \& Song, L.~M.\ 2002, \apjl, 568, L49
667:
668: \bibitem[MacApline et al.\ 1989]{} MacAlpine, G.~M., McGaugh, S.~S.,
669: Mazzarella, J.~M., \& Uomoto, A.\ 1989,
670: \apj, 342, 364
671:
672: \bibitem[Mori 2002]{} Mori, K.\ 2002, PhD thesis, Osaka University
673:
674: \bibitem[Pelling et al.\ 1987]{} Pelling, R.~M., Paciesas, W.~S.,
675: Peterson, L.~E., Makishima, K., Oda, M.,
676: Ogawara, Y., \& Miyamoto, S.\ 1987, \apj,
677: 319, 416
678:
679: \bibitem[Porquet et al.\ 2003]{} Porquet, D., Decourchelle, A., \&
680: Warwick, R.~S.\ 2003, \aap, 401, 197
681:
682: \bibitem[Rees \& Gunn 1974]{} Rees, M.~J., \& Gunn, J.~E.\ 1974, \mnras,
683: 167, 1
684:
685: \bibitem[Rybicki \& Lightman 1979]{} Rybicki, G.~B., \& Lightman, A.~P.\ 1979, Radiative Process in Astrophysics (New York: Wiley-Interscience)
686:
687: \bibitem[Safi-Harb et al.\ 2001]{} Safi-Harb, S., Harrus, I.~M., Petre,
688: R., Pavlov, G.~G., Koptsevich, A.~B., \& Sanwal, D.\ 2001, \apj, 561, 308
689:
690: \bibitem[Sankrit \& Hester 1997]{} Sankrit, R., \& Hester, J.~J.\ 1997,
691: \apj, 491, 796
692:
693: \bibitem[Shibata et al.\ 2003]{} Shibata, S., Tomatsuri, H., Shimanuki,
694: M., Saito, K., Nakamura, Y. \& Mori, K.\
695: 2003, \mnras, 346, 841
696:
697: \bibitem[Slane et al.\ 2000]{} Slane, P., Chen, Y., Schulz, N.~S.,
698: Seward, F.~D., Hughes, J.~P., \&
699: Gaensler, B.~M.\ 2000, \apjl, 533, L29
700:
701: \bibitem[Toor \& Seward 1974]{} Toor, A., \& Seward, F.~D.\ 1974, \aj,
702: 79, 995
703:
704: \bibitem[Townsley et al.\ 2001]{} Townsley, L.~K., Broos, P.~S., Chartas,
705: G., Moskalenko, E., Nousek, J.~A., \&
706: Pavlov, G.~G.\ 2002, Nuclear Instruments
707: \& Methods in Physics Research Section A,
708: 486, 716
709:
710: \bibitem[Weisskopf et al.\ 2000]{} Weisskopf, M.~C., Hester, J.~J.,
711: Tennant, A.~F., et al.\ 2000, \apj, 536, L81
712:
713: \bibitem[Willingale et al.\ 2001]{} Willingale, R., Aschenbach, B.,
714: Griffiths, R.~G., Sembay, S., Warwick,
715: R.~S., Becker, W., Abbey, A.~F., \&
716: Bonnet-Bidaud, J.-M.\ 2001, \aap, 365, 212
717:
718: \bibitem[Wise et al.\ 1997]{} Wise, M.~W., Davis, J.~E., Huenemoerder,
719: Houck, J.~C., Dewey, D., Flanagan, K.~A.,
720: and Baluta, C.\ 1997, The MARX 2.0 User
721: Guide, CXC Internal Document
722:
723: \end{thebibliography}
724:
725: \begin{figure}
726: \epsscale{1.0}
727: \plotone{f1.eps}
728: \caption{({\it a}) {\it Chandra} ACIS-S combined image
729: of the 2nd$-$8th observations.
730: The center of the white circle ($R=50''$) is located
731: at the pulsar position. The ``hole'' at the pulsar position is
732: caused by severe event pile-up resulting in the rejection of
733: most events at this position. Narrow lines through the pulsar are
734: instrumental artifacts due to trailing events.
735: ({\it b}) Photon index map of the Crab Nebula
736: after correction for pile-up effects. White regions around the eastern,
737: western, and southern edges were excluded from analysis due to the
738: dominance of trailing events and scattered photons. The pulsar
739: position is also white because no photon index could be derived there
740: due to severe pile-up.
741: \label{fig:index_map}}
742: \end{figure}
743:
744: \begin{figure}
745: \epsscale{0.9}
746: \plotone{f2.eps}
747: %\includegraphics[angle=-90, width=16cm]{f2.eps}
748: \caption{({\it a})
749: Plot of apparent photon index against observed surface brightness
750: derived from $2.^{\!\!\prime\prime}5 \times 2.^{\!\!\prime\prime}5$
751: square regions. Results of pile-up simulations for incident photon
752: index of 1.5, 2.0, and 2.5 are superimposed (red lines), which shows
753: that pile-up results in smaller photon index in higher surface
754: brightness. ({\it b}) Same as (a) but corrected for pile-up effects.
755: \label{fig:correlation_plot}}
756: \end{figure}
757:
758: \begin{figure}[]
759: \epsscale{1.0}
760: \plotone{f3.eps}
761: \caption{Plot of photon index against surface brightness, same as
762: Figure~\ref{fig:correlation_plot}b. The data points of the torus, the
763: jet, the umbrella-shaped northwest region, and the faint peripheral
764: region are color-coded as red, blue, yellow, and green, respectively.
765: The inset shows the definitions of the four regions. The black
766: contours represent surface brightness. The red dashed circle encloses
767: data points with unusually small photon index compared to others (see
768: \S \ref{sec:nh}). The corresponding positions in the nebula are shown
769: in black in the inset, also enclosed by a red dashed circle.
770: \label{fig:color_plot}}
771: \end{figure}
772:
773: \begin{figure}[]
774: \epsscale{1.0}
775: \plotone{f4.eps}
776: \caption{Close-up view of photon index map around the southern jet.
777: Note that the range of photon index is different from that in
778: Figure~\ref{fig:index_map}a to show fine variation. The green
779: contours represent surface brightness.
780: \label{fig:jet}}
781: \end{figure}
782:
783: \begin{figure}[]
784: \epsscale{0.8}
785: \plotone{f5.eps}
786: \caption{Spectra extracted from one of the small photon index
787: regions (red) and a neighboring region (black).
788: \label{fig:spectrum_comparison}}
789: \end{figure}
790:
791: \clearpage
792:
793: \begin{deluxetable}{cccc}
794: \footnotesize
795: \tablecaption{Observational log of the Crab Nebula \label{tbl:log}}
796: \tablewidth{0pt}
797: \tablehead{ \colhead{ObsID} & \colhead{Date} &
798: \colhead{Coord.(J2000)\tablenotemark{a}} &
799: \colhead{Exposure (sec)} }
800: \startdata
801: 1994 & 3 November 2000 & 83.637, 22.024 & 3241.8 \\
802: 1996 & 25 November 2000 & 83.632, 22.014 & 2926.8 \\
803: 1998 & 18 December 2000 & 83.631, 22.016 & 2703.8 \\
804: 1999 & 9 January 2001 & 83.631, 22.017 & 2663.4 \\
805: 2001 & 30 January 2001 & 83.631, 22.017 & 2553.0 \\
806: 1995 & 21 February 2001 & 83.631, 22.017 & 2648.0 \\
807: 1997 & 14 March 2001 & 83.631, 22.017 & 2568.6 \\
808: 2000 & 6 April 2001 & 83.631, 22.016 & 2643.0 \\
809: \enddata
810: \tablenotetext{a}{coordinates of the center of the subarray}
811: \end{deluxetable}
812:
813: \clearpage
814:
815: \begin{deluxetable}{lccc}
816: \rotate
817: \footnotesize
818: \tablecaption{Spectral parameters for whole nebula \label{tbl:whole_fit}}
819: \tablewidth{0pt}
820: \tablehead{ \colhead{Parameter} & \colhead{Uncorrected Value} &
821: \colhead{Corrected Value} & \colhead{Canonical Value} }
822: \startdata
823: $N_{\rm H}$ (10$^{22}$ cm$^{-2}$) & 0.27 & 0.32 (fixed) & 0.33--0.36\tablenotemark{a} \\
824: Photon index & 1.92 & 2.10 & 2.10 $\pm$ 0.03\tablenotemark{b} \\
825: Normalization\tablenotemark{c} & 6.57 & 9.65 & 9.7 $\pm$ 0.5\tablenotemark{d} \\
826: \enddata
827: \tablenotetext{a}{Kuiper et al.\ (2001)}
828: \tablenotetext{b}{Toor \& Seward (1974)}
829: \tablenotetext{c}{(Photons cm$^{-2}$ s$^{-1}$ keV$^{-1}$ at 1 keV)}
830: \tablenotetext{d}{Willingale et al.\ (2001)}
831: \end{deluxetable}
832:
833: \end{document}
834:
835: