1: % First draft of NGC 507 paper.
2: % Second draft after 2003 HEAD meeting.
3: % Third draft after comments from WRF - 7APR03
4: % Modified 11APR03 after comments from CJF
5:
6: \documentclass[12pt,preprint]{aastex}
7:
8: %\usepackage{amsmath}
9: %\usepackage{emulateapj5}
10: %\usepackage{apjfonts}
11:
12: %\makeatletter
13: %\newenvironment{inlinetable}{%
14: %\def\@captype{table}%
15: %\noindent\begin{minipage}{0.999\linewidth}\begin{center}\footnotesize}
16: %{\end{center}\end{minipage}\smallskip}
17:
18: %\newenvironment{inlinefigure}{%
19: %\def\@captype{figure}%
20: %\noindent\begin{minipage}{0.999\linewidth}\begin{center}}
21: %{\end{center}\end{minipage}\smallskip}
22: %\makeatother
23:
24: \newcommand{\etal}{{\em et al.}} % Defines et al.
25: \newcommand{\mic}{$\mu$m}
26: \newcommand{\nh}{N$_H$}
27: \newcommand{\xmm}{{XMM-Newton}}
28: \newcommand{\chdr}{{\em Chandra}}
29: \newcommand{\abs}{{5$\times$10$^{20}$ cm$^{-2}$}}
30:
31: \renewcommand{\floatpagefraction}{1.0} % redefines float page fraction
32: \renewcommand{\textfraction}{0.0}
33: \renewcommand{\topfraction}{1.0}
34: \renewcommand{\bottomfraction}{1.0}
35:
36: \shorttitle{X-rays from Cen A ISM and Radio Lobes}
37: \shortauthors{Kraft \etal}
38:
39: \begin{document}
40:
41: \title{An Unusual Discontinuity in the X-ray Surface Brightness Profile of NGC 507: Evidence of an Abundance Gradient?}
42: \author{R. P. Kraft, W. R. Forman}
43: \affil{Harvard/Smithsonian Center for Astrophysics, 60 Garden St., MS-67, Cambridge, MA 02138}
44: \author{E. Churazov}
45: \affil{Max Planck Institute f\"{u}r Astrophysik, 85740 Garching bei M\"{u}nchen, Germany and Space Research Institute (IKI), Moscow 117810, Russia}
46: \author{N. Laslo, C. Jones, D.-W. Kim, M. Markevitch, S. S. Murray, A. Vikhlinin}
47: \affil{Harvard/Smithsonian Center for Astrophysics, 60 Garden St., MS-67, Cambridge, MA 02138}
48:
49: \begin{abstract}
50:
51: We present results from a 45 ks Chandra/ACIS-I
52: observation of the nearby ($z=0.01646$) NGC 507 group. The X-ray surface
53: brightness profile of the outer region of the group is well described
54: by an isothermal $\beta$-profile consistent with earlier ROSAT observations.
55: We find a sharp edge or discontinuity in the radial surface brightness profile
56: 55 kpc E and SE of NGC 507 covering a~$\sim$125$^\circ$ arc.
57: At the sharpest part of the discontinuity,
58: the surface brightness declines by a factor of $\sim$2
59: over a distance of 6.9 kpc ($\sim 15''$).
60: The inner and outer gas temperatures across the
61: discontinuity differ by only about 0.2 keV
62: (interior and exterior gas temperatures are 1.2 and 1.3 keV, respectively).
63: Spectral analysis indicates that there is a large gradient in the elemental
64: abundance across the discontinuity, and
65: comparison with the low resolution NVSS radio maps suggests that
66: the discontinuity is aligned with a low surface brightness radio lobe.
67: We conclude that the appearance of this discontinuity is the result
68: of the transport of high abundance material from the center of the galaxy
69: by the expanding radio lobe,
70: similar to structures seen in {\em Chandra} observations of Perseus A and
71: Hydra A.
72:
73: \end{abstract}
74:
75: \keywords{galaxies: individual (NGC 507) - X-rays: galaxies - galaxies: ISM}
76:
77: \section{Introduction}
78:
79: Chandra observations of early galaxies, groups of galaxies, and clusters
80: of galaxies have demonstrated that the hot, X-ray emitting
81: coronal gas of these systems is
82: often not resting idly in their gravitating dark matter halos and
83: that many of these systems are far from relaxed.
84: Several different structures have been identified and studied in detail.
85: Chandra observations of shocks and "cold fronts" associated with sub-cluster mergers
86: have contributed to our knowledge of the formation of structure.
87: Evidence for large scale, coherent, non-hydrostatic motions of
88: the central regions of clusters has also been reported \citep{vik01,mar01,maz01}.
89: One of the most interesting, and to some degree unexpected, results
90: from Chandra has been the study of diverse interactions between
91: jets and radio lobes of radio galaxies and the ambient
92: ISM/IGM/ICM \citep{fab00,fin00,mcn00,jon02,kra03}.
93: The study of these various motions and hydrodynamical interactions
94: is of fundamental importance to several astrophysical and cosmological
95: problems. These observations
96: give us insight into the merger process and subsequent
97: relaxation. By studying merger shocks, thermal histories, etc.,
98: we learn about the process of the formation of clusters.
99: In addition, understanding the radio lobe/ICM interactions may be critical in
100: understanding the life cycle of clusters as there may be a cyclical relationship
101: between cooling flows, nuclear activity, and nuclear outflow.
102:
103: NGC 507 is the dominant, massive elliptical galaxy of a nearby ($z=0.01646$) group/poor
104: cluster (the so-called Pisces cluster).
105: Previous ROSAT observations have shown that NGC 507 is one of the most
106: X-ray luminous early type galaxies ($L_x\sim$ 10$^{43}$ ergs s$^{-1}$) in the local
107: universe with evidence of a large cooling flow (cooling rate of 30-40
108: $M_\odot$ yr$^{-1}$) at the center \citep{kim95,pao03}.
109: This group has an unusually large mass-to-light
110: ratio ($\sim$125), similar to that observed in the X-ray overluminous
111: elliptical galaxies (OLEGs) \citep{vik99}.
112: NGC 507 is also known to be an FR I radio galaxy \citep{fan86}.
113:
114: In this letter, we present results from a {\em Chandra}/ACIS-I
115: observation of the hot ISM of this group.
116: We detect a sharp discontinuity in the surface brightness profile
117: 55 kpc E and SE of the center of the galaxy spanning a 125$^\circ$ arc.
118: This discontinuity has been seen in XMM-Newton observations of NGC 507 as
119: well \citep{fab02}. A more detailed discussion of the larger scale emission
120: and the central cooling flow peak based on ROSAT/HRI and {\em Chandra}/ACIS-S observations
121: is presented elsewhere \citep{pao03}.
122: Initially it appears that this discontinuity was another example of
123: a "cold-front" as seen in clusters of galaxies.
124: Such discontinuities have recently been detected in clusters of galaxies
125: and are commonly attributed to either infalling sub-clumps or non-hydrostatic
126: motions of the central core (e.g. A2142 \citep{mar00},
127: A3667 \citep{vik01}, A1795 \citep{mar01}). However, there are significant problems with
128: this interpretation of the discontinuity observed in NGC 507. In particular,
129: the temperature jump across the discontinuity is small (0.1-0.2 keV), so that the
130: discontinuity should not appear sharp. In addition,
131: if the sharp boundary is maintained by motion of the galaxy, a large,
132: nearly supersonic velocity of the galaxy with respect to the larger scale
133: halo is required. Spectral analysis indicates that there is
134: a significant jump in elemental abundance across the discontinuity.
135: The X-ray discontinuity is coincident with a radio lobe detected in the NVSS
136: survey. We suggest that the discontinuity in the surface brightness is
137: due to a combination of the near-supersonic expansion of
138: the radio lobe and to turbulent entrainment of high abundance material from the
139: center of the galaxy to more distant regions caused by this expansion.
140:
141: Throughout this paper, we use $H_0$=75 km s$^{-1}$ Mpc$^{-1}$, $\Omega$=0.3 and
142: $\Lambda$=0.7. The observed redshift of NGC 507 then corresponds to a distance of 65.5 Mpc,
143: and $1'$=18.8 kpc.
144:
145: \section{Observations}
146:
147: The central region of the nearby (z=0.01646, d=65.5 Mpc) NGC 507 group
148: was observed twice by {\em Chandra}. The first observation, taken Oct 11, 2000,
149: was made with the ACIS-S instrument, and the second, taken Jan 8, 2002,
150: with the ACIS-I in very faint mode. The particle backgrounds in both
151: observations were examined for periods of flares,
152: and the average levels compared with the nominal ACIS background
153: rates. We found that the first observation was contaminated
154: by a steady but anomalously high background, so we have used only data from the second
155: observation in all analyses presented in this letter.
156: No flares or periods of high background were found in the
157: second dataset. Very faint mode filtering was applied to this
158: second event file to reduce the background to the lowest possible
159: level. A background event file was created from the standard ACIS
160: background data appropriate for the date of the observation, and very
161: faint mode event filtering applied to these data as well.
162: We also have obtained archival data from the XMM-Newton observation,
163: and have used these data to constrain spectral parameters as described below.
164:
165: A raw X-ray image in the 0.5 to 2.0 keV band is shown in Figure~\ref{rawimg}.
166: An adaptively smoothed, background subtracted, exposure corrected
167: image of NGC 507 is shown in the Figure~\ref{adaptimg}.
168: The complex azimuthal structure in the X-ray emission can be seen
169: in both figures. The large scale coronal emission can be seen in Figure~\ref{adaptimg} with the
170: X-ray peak centered on NGC 507. X-ray emission from the less massive NGC 508
171: is detected to the N. Some structure in the emission is seen to the west of
172: NGC 507, and is most likely related to interaction between the radio lobe and
173: the ISM.
174:
175: \section{Analysis}
176:
177: One of the most striking features of Figure~\ref{rawimg}
178: is the sharp discontinuity in the surface brightness
179: approximately $120''$ to the east of the central peak spanning an angle
180: of approximately $\sim$125$^\circ$ from the north through the east to the south.
181: The radial surface brightness profile in the 0.5 to 2.0 keV
182: band of the 60$^\circ$ wedge shown in Figure~\ref{rawimg}
183: is plotted in Figure~\ref{sbprof}.
184: The discontinuity appears to be sharpest to the SE, but in creating
185: the surface brightness profile in Figure~\ref{sbprof}, we chose
186: the wedge to the E and ENE to avoid uncertainties associated with the CCD chip gap
187: toward the SE and NGC 508 to the N.
188:
189: The temperature of the gas around NGC 507 was determined by two independent methods.
190: First, we divided a 120$^\circ$ wedge (twice the angle of the sector
191: shown in Figure~\ref{rawimg} to improve statistics) into eight regions,
192: three to the west of the discontinuity, one straddling the
193: discontinuity, and four to the east. A single temperature
194: absorbed APEC model was fit to both the {\em Chandra} and
195: the XMM-Newton data (independently) with the abundance as a free parameter.
196: The absorption was held fixed at the galactic value ($N_H$=5.5$\times$10$^{20}$ cm$^{-2}$).
197: The lower limit of the spectral fits was restricted to 0.8 keV to avoid uncertainties
198: in the ACIS-I response due the build up of contaminants, although a
199: correction also was applied to the response to account for this effect.
200:
201: The best fit Chandra and XMM-Newton/MOS temperatures and abundances
202: with 90\% confidence intervals as a function of distance
203: from the center of NGC 507 are plotted in Figure~\ref{tempfig}.
204: Figure~\ref{tempfig} shows only a small
205: (0.1-0.2 keV) difference in the temperature between the region
206: interior and exterior of the discontinuity.
207: The abundance increase from 0.3$\pm$0.1 to 0.6$\pm$0.1
208: towards the center is statistically
209: significant at the 90\% confidence level.
210: If the abundance is held constant, the temperature difference between
211: the interior regions and the exterior regions increases by
212: a only small ($\sim$0.15 keV) amount.
213:
214: We also created a temperature maps from both data sets
215: using the algorithm described in \citet{chu96}.
216: The map created from the XMM-Newton/MOS data
217: is shown in Figure~\ref{tmapfig} and
218: supports the conclusion of the spectral fitting
219: that there is little change in temperature just across
220: the discontinuity, but a small temperature decrease toward the center.
221: Interestingly the temperature is lowest in the
222: region to the S/SE where the discontinuity appears sharpest.
223:
224: \section{Interpretation}
225:
226: If the abundance is roughly constant across the discontinuity,
227: there must be a large pressure jump due to the large density discontinuity because
228: the observed temperature jump across the boundary is small.
229: We consider three possible interpretations to explain the observed
230: surface brightness feature. In the first
231: case, we consider the possibility that the edge is caused by motion
232: of NGC 507 relative to the larger scale group dark matter potential.
233: Second, we consider a scenario where the pressure across the boundary
234: is balanced by an unseen relic radio lobe left over from an
235: earlier (and more powerful) epoch of nuclear activity. Finally, we investigate
236: the possibility that the edge is due to the subsonic expansion of
237: the relatively weak lobe currently observed lying interior to the
238: discontinuity. This expansion has pushed higher elemental abundance gas
239: out from the central region. The appearance of the discontinuity is
240: then a consequence both of the compression of the gas due to
241: expansion of the lobe and to the increase in the abundance across the
242: discontinuity.
243:
244: One interpretation commonly invoked
245: to explain the sharp discontinuities observed in
246: galaxy clusters is that the central object is moving with a significant velocity
247: relative to the larger halo \citep{vik01,mar01}. These are the so-called
248: cluster "cold fronts". There are several problems with this scenario
249: for NGC 507. First, the implied velocity must be very large.
250: If the pressure difference between the interior and exterior regions is balanced
251: by ram pressure, the implied velocity is approximately 450 km s$^{-1}$
252: (Mach 0.7) relative to the ambient medium. Such a large velocity
253: relative to the dark matter halo is not reasonable since
254: NGC 507 is the largest galaxy in the group and lies at
255: or near the center defined by the outer X-ray isophotes.
256: The large scale gravitating mass
257: of the NGC 507 group can be estimated from the parameters of the $\beta$-profile
258: assuming the larger scale gas is in hydrostatic equilibrium.
259: If we consider NGC 507 to be an undamped harmonic oscillator in this dark matter potential
260: moving with a velocity of 450 km s$^{-1}$,
261: it should reach a distance of $\sim$75 kpc from the center before falling
262: back. Since there are no other known instances of the largest galaxy in a group
263: or cluster being so far from the center of the gravitational
264: potential, we consider this possibility to be unlikely.
265:
266: A second significant complication with this interpretation arises from
267: hydrodynamic considerations. The pressure gradient
268: is due largely to a density difference, not a temperature difference.
269: There is no evidence for supersonic motions (i.e. there is no
270: hot, shock heated shell in front of the gas which would
271: be moving supersonically). This implies that the motion is subsonic (consistent with the above
272: analysis), but since the pressure difference is due only to differences in
273: density, the boundary should not appear sharp, as there must be a
274: continuous and gradual pressure and density transition between the two regions.
275: If the pressure difference were due to a temperature discontinuity (such
276: as in A3667 \citep{vik01}), a sharp boundary could be observable because
277: the surface brightness is proportional to $n^2$.
278: It is conceivable that a much smaller region of the
279: galaxy is actually moving, and the discontinuity is material that is piled
280: up in front of the stagnation point.
281: A large velocity would still be required to contain this material.
282:
283: For the sharp boundary to exist in the absence of motion,
284: additional pressure support could exist beyond the discontinuity.
285: An unseen relic radio lobe could provide the needed pressure support beyond the
286: discontinuity. NGC 507 is a known radio source and has been
287: classified as an FR I radio galaxy (Fanti \etal~1986). Data from
288: two radio observations are shown in Figure~\ref{nvss} overlaid
289: onto an adaptively smoothed Chandra image.
290: These radio observations detected a double lobe structure lying along the
291: east-west axis of the galaxy. The eastern (weaker) radio lobe is contained
292: inside the discontinuity. The western lobe is contained within some of
293: the complex emission to the west and may be responsible for
294: the complex morphology \citep{for01}. No evidence for radio emission beyond the X-ray
295: discontinuity has been reported.
296:
297: Is it feasible that such an unobserved, relic lobe is providing pressure
298: support beyond the observed discontinuity?
299: Relativistic electrons with $\gamma\sim$5$\times$10$^2$ in a magnetic field of 50 $\mu$G
300: (a typical value for the lobes of FR II radio galaxies)
301: would radiate at $\sim$10 MHz, which is unobservable. The density of
302: such electrons can be estimated by requiring the pressure of
303: the lobe to balance the pressure difference across the discontinuity.
304: Assuming that the lobe has a $\sim$ 50 kpc radius to cover the observed
305: discontinuity, the total energy
306: in the lobe would be about 5$\times$10$^{60}$ ergs. This number is
307: a factor of a few larger than the total energy in one of the lobes
308: of the FR II radio galaxy Cyg A \citep{car91}, and would imply an energy deposition rate
309: of approximately 10$^{46}$ ergs s$^{-1}$ for 10$^7$ years.
310: The luminosity of IC scattered CMB photons would be considerable
311: ($\sim$10$^{9}$ L$_\odot$), but peaking in the EUV ($\sim$30 eV),
312: again difficult to observe.
313: The existence of such an energetic but otherwise unobservable radio
314: relic is not ruled out by this simple energy argument, but is
315: highly unlikely.
316:
317: As a final possibility, we consider a scenario where the inflation/expansion of
318: the known radio lobe is creating the discontinuity. Such a scenario can
319: explain all the observed features, and has recently
320: been investigated theoretically by \citet{rey00} to understand
321: {\em Chandra} observations of cavities and cool shells seen around the lobes
322: of several other FR I radio galaxies. We suggest that there is a low
323: surface brightness radio lobe interior to the discontinuity, expanding to the S/SE.
324: Such a lobe is hinted at in the NVSS radio map (see Figure~\ref{nvss}).
325: The relationship between the discontinuity and the smaller scale
326: radio lobe presented in \citet{fan86} is not clear.
327: In this scenario, the subsonic expansion of the radio lobe, whether
328: due to ongoing nuclear activity or by buoyancy if the central
329: source has shut off, gently pushes
330: the lower temperature, higher abundance material from the cooling center
331: of the galaxy out to more distant regions \citep{chu01,nul02}.
332:
333: We hypothesize that the sharp appearance of the discontinuity is due
334: to the combination of the motion of the radio lobe and
335: to a significant abundance discontinuity across the interface.
336: At the observed $\sim$ 1.2-1.3 keV temperature of NGC 507,
337: the emissivity of the gas is a strong function of
338: the heavy element abundance, and there is an abundance gradient
339: from the halo to the core (see Figure~\ref{tempfig}).
340: Material entrained from the central regions of
341: NGC 507 would have higher abundance than that in the
342: halo and would naturally give a sharp boundary if the emission were due
343: to a cap surrounding a radio lobe.
344: That is, the sharp abundance discontinuity (shown in
345: Figure~\ref{tempfig}) would naturally produce a discontinuity in the surface
346: brightness, and at the same time maintain pressure equilibrium without the
347: need for a large temperature jump.
348:
349: To demonstrate the effects of the enhanced elemental abundance
350: interior to the discontinuity, we consider a model for the surface brightness where
351: the emissivity of the gas is determined by
352: the extreme values of temperature and abundance in
353: Figure~\ref{tempfig}.
354: That is, the gas interior to the discontinuity has an abundance more than twice that
355: of the material beyond the discontinuity, but with a 20\% lower temperature.
356: We assume that the gas density is described by a single $\beta$-profile,
357: and that there is no pressure discontinuity across the
358: surface brightness discontinuity.
359: The emissivity of the interior material is enhanced by a factor of 2.16
360: due to its higher abundance and lower temperature.
361: We have overplotted the surface brightness profile from this
362: model onto Figure~\ref{sbprof}.
363:
364: Such an idealized model closely approximates the observed surface brightness
365: profile and can account for most, but not all, of the
366: emission interior to the discontinuity.
367: The remaining difference in the surface brightness (between
368: the data and the model) can be attributed to compression of
369: the gas by the inflation of the radio lobe.
370: We have increased the density, and therefore pressure, of the interior region to match the
371: surface brightness profile, and estimate that
372: an expansion velocity of the lobe on the order of 100 km s$^{-1}$ is
373: required to match the profile in detail and balance the increased pressure.
374: We therefore conclude that a model combining the
375: effect of the enhanced abundances and compression from subsonic
376: motion can easily match the observed surface brightness profile.
377: Further detailed radio observations are required to constrain the relationship
378: between the lobe and this discontinuity.
379:
380: {\em Chandra} has now detected more than a dozen examples of active
381: galaxies "blowing bubbles" into the ICM (\citet{bla01, mcn00} among
382: many others). In most cases, there is no evidence of shocks being
383: driven into the IGM/ICM. This is clearly the case with NGC 507
384: as there is no evidence of hotter gas associated with either the
385: discontinuity or any of the radio components.
386: The inflation of the lobe has however disrupted the gas in the central
387: region and may provide an efficient mechanism to mix the higher abundance material at the center
388: with the lower abundance material (presumably much older) in the halo.
389:
390: \section{Conclusions}
391:
392: We have observed a sharp discontinuity in the X-ray surface brightness profile in
393: a Chandra observation of NGC 507.
394: We conclude that this discontinuity is most likely not a "cold front" as has been
395: observed in many clusters, and this is not due to non-hydrostatic motion of the group core.
396: It is also highly improbable that this discontinuity is supported by
397: an unseen, relic radio lobe from an earlier epoch of galaxy activity.
398: The appearance of this discontinuity is probably due to two factors. First,
399: the subsonic expansion/inflation the radio lobe is compressing the ambient ISM.
400: Second, there is an abundance gradient across the boundary that creates
401: a discontinuity in the surface brightness. We suggest that the inflation of
402: the radio lobe has pushed up and/or entrained higher abundance material from the
403: central regions within NGC 507 into the group halo.
404: It is unclear whether this is a common phenomenon or not because it is
405: most readily observable only in relatively cool ($k_B T <$ 2 keV) galaxies and clusters
406: where the emissivity of the gas is a strong function of the abundance.
407: This type of disruption/mixing of the central, high
408: abundance region of galaxy groups and clusters by jets/lobes may be common, but would be
409: difficult to detect in high temperature systems where abundance variations
410: would not produce changes in surface brightness but instead would require detailed
411: abundance maps for their detection.
412: There is at least one other object, A576, that shows evidence for
413: a surface brightness discontinuity that is at least partially
414: attributable to a gradient in the elemental abundance (Kempner and David, in preparation).
415: If this is indeed a common phenomenon, the infall velocity of cool subclusters or
416: the expansion velocity of radio lobe/ICM interactions would be overestimated, perhaps
417: by as much as a factor of a few.
418:
419: \acknowledgements
420:
421: This work was supported by NASA contracts NAS8-38248, NAS8-39073, the
422: Chandra X-ray Center, and the Smithsonian Institution.
423:
424: \clearpage
425:
426: \begin{thebibliography}{}
427:
428: \bibitem[Blanton \etal~(2001)]{bla01} Blanton, E. L., Sarazin, C. L., McNamara, B. R., \& Wise, M. W. 2001, \apj, {\bf 558}, L15.
429:
430: \bibitem[Carilli \etal~(1991)]{car91} Carilli, C. L., Perley, R. A., Dreher, J. W., \& Leahy, J. P. 1991, {\bf 383}, 554.
431:
432: \bibitem[Churazov \etal~(1996)]{chu96} Churazov, E., Gilfanov, M., Forman, W., \& Jones, C. 1996, \apj, {\bf 471}, 673.
433:
434: \bibitem[Churazov \etal~(2001)]{chu01} Churazov, E., \etal~2001, \apj, {\bf 554}, 261.
435:
436: \bibitem[Fabbiano \etal~(2002)]{fab02} Fabbiano, G., Kim, D.-W., \& Brickhouse, N. 2002, Amer. Astron. Soc. Meeting, {\bf 201}, 14.04.
437:
438: \bibitem[Fabian \etal~(2000)]{fab00} Fabian, A. C. \etal~2000, \mnras, {\bf 318}, L65.
439:
440: \bibitem[Fanti \etal~(1986)]{fan86} Fanti, C., Fanti, R., De Ruiter, H. R., \& Parma, P. 1986, A. \& A. Suppl., {\bf 65}, 145.
441:
442: \bibitem[Finoguenov and Jones (2000)]{fin00} Finoguenov, A., and Jones, C. 2000, \apj, {\bf 547}, L107.
443:
444: \bibitem[Forman \etal~(2001)]{for01} Forman, W., \etal~2001, Clusters of Galaxies and the High Redshift Universe Observed in X-rays, ed. by D. M. Neumann \& J. T. T. Van.
445:
446: \bibitem[Jones \etal~(2002)]{jon02} Jones, C. \etal~2002, \apj, {\bf 567}, L115.
447:
448: \bibitem[Kim and Fabbiano~(1995)]{kim95} Kim, D.-W. \& Fabbiano, G. 1995, \apj, {\bf 441}, 182.
449:
450: \bibitem[Kraft \etal~(2003)]{kra03} Kraft, R. P. \etal~2003, \apj, in press.
451:
452: \bibitem[Markevitch \etal~(2000)]{mar00} Markevitch, M., \etal~2000, \apj, {\bf 541}, 542.
453:
454: \bibitem[Markevitch \etal~(2001)]{mar01} Markevitch, M., Vikhlinin, A., \& Mazzotta, P. 2001, \apj, {\bf 562}, L153.
455:
456: \bibitem[Mazzotta \etal~(2001)]{maz01} Mazzotta, P., \etal~2001, \apj, {\bf 555}, 205.
457:
458: \bibitem[McNamara \etal~(2000)]{mcn00} McNamara, B. R. \etal~2000, \apj, {\bf 534}, L135.
459:
460: \bibitem[Nulsen \etal~(2002)]{nul02} Nulsen, P. E. J. \etal~2002, \apj, {\bf 568}, 163.
461:
462: \bibitem[Paolillo \etal~(2003)]{pao03} Paolillo, M. \etal~2003, \apj, {\bf 586}, 850.
463:
464: \bibitem[Reynolds, Heinz, and Begelman~(2000)]{rey00} Reynolds, C. S., Heinz, S., \& Begelman, M. C. 2000, \apj, {\bf 549}, L179.
465:
466: \bibitem[Vikhlinin \etal~(1999)]{vik99} Vikhlinin, A., McNamara, B. R., Hornstrup, A., Quintana, H, Forman, W., Jones, C., \& Way, M. 1999, \apj, {\bf 520}, L1.
467:
468: \bibitem[Vikhlinin \etal~(2001)]{vik01} Vikhlinin, A., Markevitch, M., \& Murray, S. S. 2001, \apj, {\bf 551}, 160.
469:
470: \end{thebibliography}
471:
472: \clearpage
473:
474: \begin{figure}
475: \plotone{./f1.ps}
476: \caption{Raw Chandra/ACIS-I X-ray image of NGC 507 in the 0.5 to 2.0
477: keV band. The green wedge indicates the sector used to create
478: the radial surface brightness profile in Figure~\ref{sbprof}. The white
479: arrows indicate the approximate position of the surface brightness
480: discontinuity.}\label{rawimg}
481: \end{figure}
482:
483: \clearpage
484:
485: \begin{figure}
486: \plotone{./f2.ps}
487: \caption{An adaptively smoothed, exposure corrected image
488: of NGC 507 in the 0.5 to 2.0 keV bandpass. The peak
489: of the emission is centered on NGC 507.}\label{adaptimg}
490: \end{figure}
491:
492: \clearpage
493:
494: \begin{figure}
495: \plotone{./f3.ps}
496: \caption{Radial surface brightness profile of the X-ray emission in a 60$^\circ$
497: sector (shown in Figure~\ref{rawimg}) to the E and NE of NGC 507 in the
498: 0.5-2.0 keV band. The continuous curve labeled "Model \#1" is computed from a model
499: of the emission that consists of two components, an isothermal beta model beyond
500: the discontinuity and a uniform density sphere inside the discontinuity.
501: While reasonably consistent with the data, the model is physically unrealistic since
502: it attributes the discontinuity to a change in density and therefore, since the temperature
503: is roughly constant, to a large pressure discontinuity.
504: The curve labeled "Model \#2" consists of emission from a $\beta$-model
505: density profile but with the
506: emissivity of the gas interior to the discontinuity increased by a factor of 2.16
507: to account for the enhanced abundance and lower temperature
508: derived from spectral fitting (see Figure~\ref{tempfig} below).
509: This model can account for most of the observed structure of the discontinuity.
510: The expansion of the radio lobe may also be important.}\label{sbprof}
511: \end{figure}
512:
513: \clearpage
514:
515: \begin{figure}
516: \plotone{./f4.ps}
517: \caption{The best-fit temperature and elemental abundances in a 120$^\circ$ sector
518: to the SE, E, and NE of NGC 507. The Xs and the triangles correspond to the Chandra/ACIS-I
519: and XMM-Newton/MOS fits, respectively. The top set of points is the temperature (keV), and the
520: bottom set is the elemental abundance (fractional relative to solar). All error bars
521: are 90\% confidence intervals for one interesting parameter.}\label{tempfig}
522: \end{figure}
523:
524: \clearpage
525:
526: \begin{figure}
527: \plotone{./f5.ps}
528: \caption{XMM-Newton/MOS temperature map of NGC 507. The difference in temperature
529: between the blue/green and the red/yellow is approximately 0.2 keV.}\label{tmapfig}
530: \end{figure}
531:
532: \clearpage
533:
534: \begin{figure}
535: \plotone{./f6.ps}
536: \caption{NVSS (white - 1.4 GHz - Condon \etal~1998) and earlier, higher
537: resolution VLA (black - 1.4 GHz - de Ruiter \etal~1986) radio contours overlaid onto an
538: adaptively smoothed Chandra image. Note the close correspondence between
539: the boundaries of the NVSS radio emission and the X-ray surface brightness
540: discontinuity to the S/SE of the nucleus. The spatial resolution ($45''$ FWHM restoring beam) and
541: sensitivity (0.45 mJy at 1.4 GHz) of the NVSS data is rather poor, but this
542: data suggests that there is, in fact, a relationship between the lobe
543: and the discontinuity. The higher resolution map ($\sim 15''$ beam FWHM) indicates
544: a radio lobe inside the discontinuity.}\label{nvss}
545: \end{figure}
546:
547: \end{document}
548: