0704.1094/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: %\documentclass[draft]{emulateapj}
3: \documentclass{emulateapj}
4: 
5: %\bibliographystyle{apj}
6: 
7: \newcommand{\HI}{\ion{H}{1}\ }
8: \newcommand{\hi}{\ion{H}{1}\ }
9: \newcommand{\halpha}{H$\alpha$}
10: \newcommand{\vlsr}{\ensuremath{v_{\mathrm{LSR}}}}
11: \newcommand{\kms}{\ensuremath{\mathrm{km\,s}^{-1}}}
12: \newcommand{\kmss}{\ensuremath{\mathrm{km\,s}^{-1}}}
13: \newcommand{\Trms}{\ensuremath{\Delta\mathrm{T}_{\mathrm{rms}}}}
14: \newcommand{\co}{\ensuremath{^{12}\mathrm{CO}}}
15: \newcommand{\cmthree}{\ensuremath{\mathrm{cm}^{-3}}}
16: \newcommand{\kpc}{kpc}
17: \newcommand{\paper}{\emph{Letter}}
18: \newcommand{\pip}{PIP}
19: 
20: \newcommand{\hh}{\ensuremath{\mathrm{H}_2}}
21: 
22: \newcommand{\hms}[3]{#1$^{\mathrm h}$ #2$^{\mathrm m}$ #3$^{\mathrm s}$}
23: \newcommand{\hmss}[4]{#1$^{\mathrm h}$ #2$^{\mathrm m}$ #3$\fs$#4}
24: \newcommand{\dms}[3]{#1$\arcdeg$ #2$\arcmin$ #3$\arcsec$}
25: \newcommand{\dmss}[4]{#1$\arcdeg$ #2$\arcmin$ #3$\farcs$#4}
26: 
27: 
28: \newcommand{\knotg}{Knot g}
29: 
30: \newcommand{\gzz}{G00}
31: 
32: \shorttitle{SUBARU HDS Observation of Tycho}
33: \shortauthors{Lee et al.}
34: 
35: \begin{document}
36: 
37: \title{SUBARU HDS Observations of a Balmer-Dominated Shock in Tycho's 
38: Supernova Remnant
39: \footnote{Based on data collected at Subaru Telescope, which is
40: operated by the National Astronomical Observatory of Japan}
41: }
42: 
43: \author{Jae-Joon Lee,\altaffilmark{1,2} Bon-Chul Koo,\altaffilmark{1}
44: John Raymond,\altaffilmark{3} Parviz Ghavamian,\altaffilmark{4}
45: Tae-Soo Pyo,\altaffilmark{5} Akito Tajitsu,\altaffilmark{5} and Masahiko Hayashi\altaffilmark{5}}
46: %\affil{Astronomy Program, Department of Physics and Astronomy, Seoul National University, Seoul 151-742, Korea}%
47: \altaffiltext{1}{Astronomy Program, Department of Physics and Astronomy,
48:   Seoul National University, Seoul 151-742, Korea }
49: 
50: \altaffiltext{2}{jjlee@astro.snu.ac.kr}
51: %\email{jjlee@astro.snu.ac.kr}
52: 
53: 
54: 
55: %\author{}
56: \altaffiltext{3}{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138}
57: 
58: %\author{}
59: \altaffiltext{4}{Department of Physics and Astronomy, Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218}
60: %\email{parviz@pha.jhu.edu}
61: 
62: 
63: %\author{}
64: \altaffiltext{5}{Subaru Telescope, National Astronomical Observatory of Japan, 
65: 650 North A'oh$\bar{\mathrm{o}}$k$\bar{\mathrm{u}}$ Place, Hilo, HI 96720}
66: %\email{pyo@subaru.naoj.org, tajitsu@subaru.naoj.org, masa@subaru.naoj.org}
67: 
68: 
69: 
70: 
71: \begin{abstract}
72: We present an \halpha\ spectral observation of a Balmer-dominated 
73: shock on the eastern side of Tycho's supernova remnant using SUBARU Telescope. 
74: Utilizing the High Dispersion Spectrograph (HDS), we measure the spatial
75: variation of the line profile between preshock and postshock gas.
76: Our observation clearly shows a broadening and centroid shift of the 
77: narrow-component postshock \halpha\ line relative to the \halpha\,
78: emission from the preshock gas. 
79: The observation supports the existence of a thin precursor
80: where gas is heated and accelerated ahead of the shock.
81: Furthermore, the spatial profile of the emission ahead of the Balmer filament
82: shows a gradual gradient in the \halpha\ intensity and line width
83: ahead of the shock.  We propose that this region ($\sim\, 10^{16}$ cm)
84: is likely to be the spatially resolved precursor.  The line width increases from
85: $\sim 30\ \kms$ up to $\sim 45\ \kms$
86: and its central velocity shows a redshift of $\sim5\ \kms$ across the shock front.
87: The characteristics of the precursor are consistent with a cosmic
88: ray precursor, although a possibility of a fast neutral precursor
89: is not ruled out.
90: \end{abstract}
91: 
92: \keywords{ISM:supernova remnants -- ISM: individual (Tycho,
93:   G120.1+1.4) -- Shock Waves -- line: profiles}
94: 
95: 
96: \section{INTRODUCTION}
97: \label{sec:intro}
98: 
99: Balmer-dominated filaments are the signature of non-radiative
100: shocks propagating into partially neutral medium %
101: \citep{1978ApJ...225L..27C}. % Chevalier & Raymond 1978}.
102: The \halpha\ line profile is composed of two distinctive
103: components (narrow and broad)  
104: representing the velocity distribution of preshock and postshock
105: gases, respectively \citep{1980ApJ...235..186C}. % Chevalier et al. 1980}.
106: High resolution spectroscopic observations of several of these
107: shocks have revealed that the width of the narrow component is 
108: unusually large (30 $\sim$ 50 \kms) for ambient neutral hydrogen,
109: and it was proposed that the gas was heated 
110: in a precursor thin enough ($\lesssim 10^{17}$ cm) to avoid complete ionization of
111: hydrogen \citep[see][and references therein]{2001ApJ...547..995G,2003A&A...407..249S}.%
112: \footnote{We refer to this precursor explicitly as a
113: ``thin precursor'', to avoid potential confusion with 
114: a photoionization precursor which will be shortly
115: introduced.}
116: Two likely candidates are cosmic-ray (CR) and fast neutral precursors
117: \citep{1994ApJ...420..286S,1994ApJ...420..721H,2001ApJ...547..995G}.
118: Both scenarios predict significant Doppler shifts of preshock gas.
119: No clear indication of such a shift of the \halpha\ narrow component 
120: is reported \citep[but see][]{2004ApJ...605L.113L}.
121: A careful comparison of postshock and preshock line profiles 
122: with high spectral resolution is crucial for
123: confirming the existence of such a precursor.
124: 
125: 
126: The blast wave of Tycho's SNR, the historical remnant of the 1572
127: supernova, has been known to exhibit 
128: Balmer-dominated emission 
129: \citep{1971ApJ...168...37V% van den Bergh 1971
130: ,1978ApJ...224..851K%, Kamper \& van den Bergh 1978
131: }.
132: The region of the brightest \halpha\ emission %(Knot g from Kamper \&
133: %van den Bergh (1978))
134: \citep[Knot g from][]{1978ApJ...224..851K}
135: is located along the north-eastern edge of the remnant. 
136: Faint, diffuse optical emission extends $\sim$1 pc
137: ahead of the Balmer-dominated filaments
138: \citep[][\gzz\ hereafter]{2000ApJ...535..266G}.  
139: This feature has been identified as a photoionization precursor (\pip)
140: produced by photoionization of the preshock gas by \ion{He}{2} $\lambda
141: 304$\AA\ emission from behind the blast wave
142: (\gzz).
143: The estimated temperature of the \pip\ 
144: %estimated from the observed line ratio
145:   is $\sim 1.2 \times 10^4$ K, which is not
146: high enough to explain the observed width of the Balmer narrow component. 
147: 
148: 
149: The existence of \halpha\ emission from the PIP makes Tycho an unique
150: target for the study of nature of the thin precursor, as we can investigate
151: the change of the line profile across this precursor by comparing the
152: line emission from the PIP (preshock) and that of \knotg\ (postshock).
153: In this \paper, we present 
154: high resolution (Echelle) long slit \halpha\ spectra  of
155: Tycho \knotg\  and its PIP using the
156: SUBARU  High Dispersion Spectrograph \citep[][]{2002PASJ...54..855N}.
157: Our observations reveal the line broadening and the Doppler shift of the 
158: the narrow component, 
159: providing strong evidence for the existence of a cosmic ray or fast
160: neutral precursor. 
161: 
162: \section{OBSERVATIONS AND RESULTS}
163: \label{sec:obs}
164: 
165: 
166: The spectrosopic observation was carried out on 2004 October 1.
167: Longslit Echelle spectroscopy of the \halpha\ line was performed 
168: using order-blocking filters (HDS standard setup
169: ``stdHa''). This gives a spectral coverage of $6540$ \AA $\sim$
170: $6690$ \AA\ over the $60\arcsec$ of slit length. %, which would well fit for our purpose. 
171: The slit was centered at 
172: $\alpha(2000)$, $\delta(2000)$ = (\hmss{00}{25}{56}{5}, \dms{64}{09}{28}) with
173: position-angle of  $76\arcdeg$ (measured E of N), covering both \knotg\  and
174: its \pip\ simultaneously (Fig.~\ref{fig:one}(a)). 
175: A total of 3$\times$30 minutes of source exposure was obtained, and 
176: the same amount of
177: exposure for nearby sky. 
178: The spectrum was binned by 2 along the slit direction and 4 along the
179: dispersion direction before the readout. 
180: The pixel scale after the binning is 0.27\arcsec\ pixel$^{-1}$ 
181: ($9.3\times 10^{15}$ cm pixel$^{-1}$ at a
182: distance of 2.3 kpc \citep{1978ApJ...224..851K}) and %along the slit direction and 
183: $0.08$ \AA\, %along the dispersion direction.
184: respectively.
185: The slit width was $2 \arcsec$, which gives velocity
186: resolution of $17$ \kms.
187: The seeing was $0\farcs 5 \pm 0\farcs 1$. 
188: The processing of the SUBARU data included a typical CCD
189: preprocessing 
190: (including overscan correction and flat fielding) 
191: and two-dimensional spectral extraction.
192: A wavelength calibration solution is obtained from the spectrum of a Th-Ar
193: lamp. % was applied to the object spectrum.
194: The source spectrum was sky-subtracted, and normalized 
195: using the spectra of standard stars. 
196: The uncertainty in the wavelength calibration is estimated to be
197: around $0.2$ \kms\ at the wavelength of \halpha. %around XXX \AA.
198: 
199: 
200: 
201: In Fig.~1(b), we present the fully processed two dimensional
202: spectrum of \halpha\ line. 
203: The bright patch at the bottom with three distinct 
204: local emission peaks corresponds to \knotg, and 
205: the faint emission extending to the top of the image
206: %upper part 
207: corresponds to the PIP.
208: The average \halpha\ spectrum of \knotg\ and that of PIP are
209: shown together in  Fig.~\ref{fig:spec}.
210: The velocity width of the \knotg\ narrow component line is clearly larger than that of the
211: PIP \halpha\ line.  In addition, the velocity centroid of the \knotg\ narrow component is
212: slightly redshifted ($5.5 \pm 0.6\ \kms$) relative to that of the PIP \halpha\ line. And as clearly seen in Fig.~\ref{fig:spec}(b),
213: the \knotg\  spectrum shows a very broad ($\sim$ 1,000 \kms), faint
214: \halpha\ line.
215: The small peak near $900$ \kms\ is the [\ion{N}{2}] $\lambda 6583.4$~\AA\
216: line from the PIP.
217: 
218: The spectrum of the PIP is well fitted by a single Gaussian, and yields a FWHM
219: of $33.8\pm 0.8$ \kms\ and centroid velocity of $-35.8\pm 0.6$ 
220: \kms\ (in LSR frame).\footnote{throughout this paper, we give line widths 
221: corrected for instrumental broadening and velocity in LSR frame.}
222: The measured FWHM of [\ion{N}{2}] $\lambda
223: 6583.4$~\AA\  in the PIP is $\sim 23$ \kms.
224: If the broadening were purely thermal, the widths of the lines whould
225: be inversely proportional to the square root of their atomic masses.
226: The expected width of the [\ion{N}{2}] line in this case is 
227: $1/\sqrt{14}=0.267$ of \halpha. As this is significantly narrower than 
228: what is observed, a significant amount of nonthermal broadening
229: %(which is considered to be independent of atomic mass) 
230: is suggested.
231: If we simply assume that the observed line widths are a
232: convolution of thermal and nonthermal  broadenings, 
233: the estimated thermal temperature is $\sim 13,000$ K,
234: consistent with $12,000$ K of \gzz. 
235: It is also possible that the large line width is due to a residual of
236: Galactic \halpha\ emission.
237: 
238: An adequate fit to the \knotg\ spectrum requires three Gaussian components.
239: They have velocity widths of $45.3\pm 0.9$ \kms\ (narrow), 
240: $108\pm 4$ \kms\ (intermediate) and $931 \pm 55$ \kms\ (broad),
241: with central velocities of $-30.3\pm 0.2$, $-25.8\pm 0.8 $ and
242: $29 \pm 18$ \kms, respectively.
243: This result confirms that the \halpha\ narrow component line of \knotg\  is
244: redshifted and broadened relative to that of the PIP. 
245: The broad component (FWHM$\sim 1,000\ \kms$) should correspond to the
246: previously reported $\sim 2,000\ \kms$ component
247: \citep{2001ApJ...547..995G}.  The relatively
248: narrower width of our observation is probably due to the insensitivity
249: of our spectroscopic configuration to the very broad line, although the
250: possibility of temporal variation (e.g., by crossing a density jump) does exist.
251: The abrupt density increase might be possible if the shock is
252: propagating into the edge outskirts of the dense cloud
253: \citep{2004ApJ...605L.113L}.
254: The existence of the intermediate width component has already been
255: reported by \gzz. %, its origin is still unclear.
256: It may be produced when slow protons picked up by the postshock magnetic
257: field undergo a secondary charge exchange. 
258: Alternatively, it might be an artifact of the assumption of Gaussian
259: distributions, which would not necessarily be appropriate if the
260: motions are non-thermal. 
261: 
262: 
263: The characteristics  of the narrow component \halpha\ line
264: of \knotg\ are consistent with \gzz, except that the velocity centroid of the
265: line in our data is significantly different from theirs
266: ($\vlsr = -30.3\pm 0.2$ from our data vs.\ $-53.9\pm 1.3\ \kms$ from \gzz).
267: We have carefully assessed our wavelength calibration% with several methods
268: , but couldn't find any significant source of error. %First of all, 
269: The wavelengths of night sky lines observed near \halpha\ matched well with
270: those of VLT UVES night-sky emission line catalog
271: \citep{2003A&A...407.1157H}
272: within error of $0.02$ \AA\ ($\sim 1\ \kms$). %We also
273: The \halpha\ spectrum of blank sky is also consistent with 
274: nearby WHAM spectra \citep{2003ApJS..149..405H}.
275: Furthermore, the central velocity of [\ion{N}{2}] $\lambda$6583.4 \AA\ 
276: emission line from the \pip\ is consistent with that of \halpha.
277: Although we believe that our wavelength calibration solution
278: is quite secure, the discrepancy with the previous observation should be
279: confirmed by an independent observation.
280: For the rest of this paper, we mainly concentrate on
281: the line width and the relative variation of central velocity.
282: 
283: 
284: 
285: 
286: 
287: \section{Location of the Shock Front and Discovery of a Thin ($10^{16}$ cm)  Precursor}
288: 
289: Fig.~\ref{fig:two} shows the spatial variation of the \halpha\ line
290: profile along the slit. 
291: In the top panel, we separately plot the integrated intensities of
292: the representative \halpha\ of narrow and broad components, 
293: together with their sum
294: (refer to the figure caption for the velocity range of each component).
295: In the middle and bottom panels, we plot the central
296: velocities and FWHMs at each slit position from the {\em single}
297: Gaussian fit to the full profile.  
298: In all three panels, the x-axis represents the pixel offset from the
299: southwest end of the slit. 
300: Although three Gaussian components are actually required to adequately fit the 
301: \halpha\, profile of \knotg, the non-uniqueness of the
302: multiple-component fitting can complicate the interpretation of the 
303: different emission line components.  Therefore, we plot the fit result
304: from a single Gaussian and describe our interpretation of this fit 
305: in the region of \knotg\ as following:
306: First, the central velocities closely %given in the middle panel
307: match those of the narrow component.
308: Fitting with multiple  components 
309: leaves the fitted line centroids unchanged to within the errors.
310: %doesn't change the result to within
311: %the errors.
312: Second, the FWHMs %given in the bottom %panel 
313: basically trace the spatial variation of the narrow 
314: component width, but the presence %effect
315: of the intermediate and the broad components contribute significantly to the width.
316: When fit with three Gaussian components, the large widths of 
317: the \halpha\ narrow component in \knotg\ from the single profile 
318: models ($60 \sim 80\ \kms$)  are reduced to
319: $40 \sim 50\ \kms$. 
320: 
321: 
322: Inspection of Fig.~\ref{fig:two} reveals that there %the measured narrow 
323: exists only one location (pixel offset 27, marked as a dotted vertical
324: line) where the intensity and width of the \halpha\ line exhibits an abrupt jump.
325: The jump is most noticeable for the broad component, where 
326: the flux is virtually zero toward the direction of ambient medium 
327: (i.e., toward positive pixel offsets). % and suddenly starts to appear at this location.
328: The central velocity of the narrow component also shows rapid change
329: around this location. 
330: The FWHM also steeply increases behind this point, but this 
331: might be largely
332: due to %would not be
333: the sudden appearance of the broad component.  
334: The fact that the intensity of the
335: broad component, which is associated with the postshock gas,
336: shows a significant jump at this location suggests that 
337: it corresponds to the location of the shock front.
338: 
339: The \halpha\ intensity in the PIP region is nearly constant along the
340: slit, but shows an rapid % there exists a steep
341: rise just before the shock front, i.e., from pixel
342: offset 32 to 27 in Fig.~\ref{fig:two}. 
343: The intensity of the \halpha\ line in the PIP region is about 10\%
344: of the observed peak value in \knotg, and it reaches about half 
345: the peak very near the shock. 
346: The line width also increases 
347: from 30 to 45 \kms\ within this region.
348: We propose that the steep increase of the \halpha\ intensity
349: accompanied with line broadening corresponds to %defines 
350: a thin precursor where gas is heated and accelerated.  
351: Unlike the line broadening, the observed velocity centroids shift 
352: slightly behind the shock instead of the precursor region. 
353: This does not necessarily indicate that the broadening and the Doppler
354: shift takes at different region as this 
355: is likely due to geometrical projection effects. 
356: The \halpha\ intensity profile gives 
357: an e-folding thickness %for this precursor 
358: of $1.4 \pm 0.4$ pixel (after accounting for seeing),
359: which corresponds to $(1.4 \pm 0.4) \times 10^{16}$ cm at a distance
360: of 2.3 kpc. % \citep{1980ApJ...235..186C}. 
361: 
362: The observed emission of Knot g shows a few local peaks indicating 
363: possible substructure, e.g., a collection of shock
364: tangencies seen in projection along the line of sight. 
365: This leads to a possibility that the `precursor' is simply the results
366: of geometric projection of fainter Balmer-dominated filament lying just
367: ahead of Knot g.
368: If we assume that this filament has a line profile similar to that of
369: \knotg, i.e., have a same broad-to-narrow ratio, 
370: then a detectable amount of broad component is expected. 
371: The flux profile of broad component plotted in
372: Fig.~\ref{fig:two} would show similar gradual increase in
373: the precursor region, which is not seen. 
374: Also, no hint of a broad component is found in a summed spectra of 
375: precursor region, which is expected to show up.
376: %
377: %However, no such emission is detected in our SUBARU observation.
378: %
379: Examining the archival 
380: {\em Chandra} observation of Tycho's SNR
381: (the analysis is presented by Warren et al. 2005), we did not find evidence
382: of X-ray emission extending ahead of Knot g.  Therefore, there is no
383: strong supporting evidence in favor of a projection, although
384: we cannot rule out the possibility. % of projection is still possible
385: 
386: 
387: 
388: 
389: \begin{figure}
390: \epsscale{1.20}
391: \plotone{f1.eps}
392: \caption{
393: (a) \halpha\ image of Tycho \knotg\  with position of the 2\arcsec\,$\times$60\arcsec\, slit
394: overlaid.
395: (b) The fully-processed \halpha\ 2-d spectrum (only the western part
396: of the slit is shown). The bright emission knot at the bottom is \knotg.
397: \label{fig:one}}
398: \end{figure}
399: 
400: 
401: \begin{figure}
402: \epsscale{.9}
403: \plotone{f2.eps}
404: \caption{
405: (a) \halpha\ spectrum of \knotg\  and the
406: PIP. The spectrum of \knotg\  is fitted with three Gaussian components, 
407: and two narrowest components are displayed (thin dashed) together
408: with their sum (thin solid).
409: (b) Same as (a) except the full observed velocity range is presented and
410: the spectra are binned. The \knotg\  broad component from the above fit
411: is displayed as a thin solid line.
412: \label{fig:spec}
413: }
414: \end{figure}
415: 
416: 
417: \begin{figure}
418: \epsscale{.95}
419: \plotone{f3.eps}
420: \caption{
421: (top) Integrated fluxes along the slit position of \halpha\
422: spectrum for a given velocity range.
423: The thin solid line shows emission in the range $-60 < \vlsr < 10\ \kms$,
424: representative of \halpha\ narrow component,  
425: while the thick solid line shows emission in the range 
426: $-400 < \vlsr < -120\ \kms$ and
427: $+100 < \vlsr < +350\ \kms$, for the broad component. 
428: The thick dashed line shows the summed intensity for both components.
429: ($-400 < \vlsr < +350\ \kms$). Middle and bottom panels: central
430: velocities and FWHMs from fits with a single Gaussian. 
431: The proposed location of shock front is marked as a vertical dotted line.
432: \label{fig:two}
433: }
434: \end{figure}
435: 
436: 
437: \section{Nature of the Thin Precursor}
438: \label{sec:vel-struct}
439: 
440: \label{cr-precursor}
441: 
442: As the shock is nearly tangential to our line of sight, 
443: the actual amount of bulk gas acceleration could be much larger than
444: the observed Doppler shift.  
445: The shock angle can be inferred from the radial velocity shift of
446: the broad component of \knotg\ relative to the narrow component
447: \citep{1980ApJ...235..186C}, which is
448: measured to be $\sim 60\ \kms$.  
449: As our observation could be insensitive to this broad component,
450: using this value is rather conservative.
451: On the other hand, \citet{2001ApJ...547..995G} reported a redshift of
452: $\sim 130\ \kms$, 
453: but their field of view is different from ours.
454: We take these two values as limits and give shock angle of
455: $2\arcdeg-5\arcdeg$ assuming a shock velocity of 2,000 \kms.
456: We consider $5\ \kms$ to be the representative 
457: redshift of the narrow component compared to unperturbed medium, 
458: which gives actual acceleration of $60\sim130\ \kms$.
459: 
460: 
461: The line width of the narrow component at the shock
462: front ($\sim 45\,\kms$) corresponds to
463: a neutral hydrogen temperature of $40,000$ K, if the line broadening is 
464: purely thermal, or lower, if there is a significant non-thermal broadening
465: such as a wave motion.
466: %The line width of ... 40,000 K, if the line broadening is purely
467: %thermal. ...(You may discuss how neutrals are heated.)...
468: Neutral hydrogen atoms and protons may have similar velocity
469: distributions due to their charge exchange interactions. 
470: %Neutrals would have an identical velocity distribution 
471: %as protons due to charge exchange interaction between them.
472: But that of electrons can be different, which would depend on the
473: heating mechanism in the precursor.
474: %We infer the electron temperature from ...
475: %
476: %
477: %It would be meaningful to compare the observed line width of \halpha\
478: %narrow component to the 
479: In the following, we estimate the electron temperature in the 
480: thin precursor (TP)
481: %
482: %which can be estimated 
483: from the observed intensity increase of factor 5.
484: %It is important to know whether the linewidth increase 
485: %within the narrow precursor (TP) region has some non-thermal
486: %origin such as a wave motion. One way to check this is to
487: %estimate a required temperature increase within TP 
488: %needed to explain the observed intensity increase of factor 5. 
489: Since the observed spectrum is an %As we see a 
490: integrated emission along the line of sight where
491: a significant contribution from PIP region is expected, the actual
492: emissivity increase within the precursor should be much greater.
493: We assume that regions of 
494: the PIP and the TP
495:  are represented by two concentric shells with thicknesses of
496: $\delta_{\mathrm{PIP}} \simeq 10^{16}$\,cm and
497: $\delta_{\mathrm{TP}} \simeq 1$\,pc (\gzz), respectively,
498: and that both have an inner diameter of the Tycho
499: (5.4\,pc). Then the ratio of path length through each shell along the
500: tangential direction of the shock front is $\simeq 18$.
501: This implies that the emissivity increase
502: in the TP could be as large as a factor of $90$. 
503: This value should be regarded as an upper limit as it is likely that the
504: filament is nearly flat and tangent to the line of sight.
505: % reducing  $l_{\mathrm{PIP}}/l_{\mathrm{TP}}$ significantly.
506: %% The increase of the 
507: %
508: %% If the electrons in the TP are thermally equilibrated with the neutrals,
509: %% the electron temperature $\mathrm{T}_{\mathrm{e}}$ should be no higher than
510: %% $40,000$ K. 
511: %% At this $\mathrm{T}_{\mathrm{e}}$, 
512: %
513: %If we interprets 
514: The collisional excitation rate of \halpha\ at
515: $\mathrm{T}_{\mathrm{e}} \sim 40,000$\,K is $2,000$ times greater than
516: that of $12,000$\,K which is the temperature of
517: the PIP region (\gzz). This value greatly exceeds the emissivity
518: increase of $90$, and implies a few possibilities. 
519: The electron temperature can be less than $40,000$\,K, either if 
520: %
521: %% First, 
522: %% $\mathrm{T}_{\mathrm{e}}$ can be less than $40,000$\,K. 
523: %% As mentioned earlier, 
524: observed \halpha\ line width has significant nonthermal broadening,
525: or if $\mathrm{T}_{\mathrm{e}}$ is intrinsically less than 
526: $\mathrm{T}_{\mathrm{p}}$. 
527: %% if the heating mechanism in the TP  heats
528: %% ions and neutrals first
529: %% (e.g., dissipation of MHD waves by ion-neutral friction)
530: %% and electrons are slowly equilibrated.
531: %% %As a matter of fact, the temperature of neutrals can be lower
532: %% %than $40,000$\,K 
533: %% %if a significant fraction of line broadening has non-thermal origin,
534: %% %such as a wave motion.
535: On the other hand, the estimated emissivity increase might be
536: explained if 
537: %$\mathrm{T}_{\mathrm{e}}$ is higher than $10^5$\,K and 
538: the emissivity increase due to high $\mathrm{T}_{\mathrm{e}}$
539: is suppressed  by ionization of neutrals. 
540: %Since the
541: %existence of Blamer-dominated filaments require significant fraction
542: %of neutral hydrogens to survive within precursor, this possibility is
543: %less favored. 
544: Since the
545: existence of Blamer-dominated filaments requires a significant fraction
546: of neutral hydrogen atoms to survive within precursor, this possibility is
547: less favored. 
548: 
549: 
550: %Such a high electron temperature is expected if significant heating is
551: %through a damping of lower-hybrid waves (see below).
552: %
553: %energies
554: %are electrons are heated
555: %by damping of 
556: %% But we note that as the Balmer dominated filament
557: %% requires survival of neutrals in the precursor, 
558: %% a significant ionization in the precursor is disfavored.
559: 
560: %
561: %% dissipation of MHD waves in the partially neutral gas, and that is
562: %% dominated
563: %% by ion-neutral charge transfer
564: 
565: %% The increase of the emissivity
566: %% within the TP should be mostly due to the increase of 
567: %% collisional excitation rate which is very sensitive to
568: %% $\mathrm{T}_{\mathrm{e}}$. %the electron temperature. 
569: %% If the line width of the narrow component at the shock
570: %% front ($\sim 40\,\kms$) is purely thermal and if the 
571: %% electrons are thermally equilibrated with neutrals,
572: %% then the electron temperature
573: %% would be $\sim 40,000$\,K.
574: %% The collisional excitation rate at
575: %% 40,000 K is
576: %% 2,000 times greater than that of 12,000 K which is the temperature of
577: %% the PIP region (\gzz). This is more than an order of
578: %% magnitude greater than the emissivity increase that we derived
579: %% above.
580: %% Therefore %, although a detailed model calculation is required, 
581: %% %the above consideration suggests that 
582: %% a significant fraction of the broadening could be of non-thermal origin.
583: %% On the other hand, it is possible that the electrons are not thermally
584: %% equilibrated with the neutrals. If their temperature is higher than 
585: %% $10^5$\,K, the ionization of neutral hydrogen within the TP could be
586: %% significant, compensating the emissivity increase due
587: %% to the increased excitation rate. The existence of 
588: %% the non-thermal broadening is not trivial in this case.
589: %% %a significant fraction of the broadening could be of non-thermal origin.
590: 
591: 
592: 
593: 
594: The two likely  candidates for this
595: precursor are fast neutral and CR precursors.  
596: The momentum and energy carried by upstreaming fast neutrals can be large
597: enough to explain the observed heating and acceleration in the precursor,
598: but the available model calculations 
599: do not predict significant heating by these neutrals
600: \citep{1996MNRAS.280..103L,2005PhDT..........K}. 
601: %% Protons produced by charge exchange of such neutrals 
602: %% can generate wave motion and may amplify the magnetic field in the
603: %% precursor. Details of these processes have not been worked out yet, and
604: %% it is not clear whether they can explain the observed 
605: %% non-thermal broadening.
606: Furthermore, It is difficult to reproduce the small range
607: of FWHMs observed for 
608: narrow component \halpha\ lines from shocks of different velocities %% is
609: \citep{1994ApJ...420..286S,1994ApJ...420..721H}.
610: The observed characteristics of the precursor are consistent 
611: with a CR precursor. CR acceleration in the shock does require a
612: precursor \citep[e.g.,][]{1987PhR...154....1B} and there has been growing
613: evidence of CR acceleration in young SNRs including Tycho itself
614: \citep[e.g.,][]{2005ApJ...634..376W}. 
615: Significant heating and acceleration are expected to
616: happen in the CR precursor \citep{1987PhR...154....1B}.
617: Interaction of CR particles in
618: the upstream generates Alf\'{v}en waves and significant amplification
619: of magnetic field has been suggested \citep{2001MNRAS.321..433B}. 
620: Although a measurement of the preshock magnetic field is hardly available,
621: a high value ($40\,\mu$G) of preshock magnetic field is
622: suggested for Tycho %to match the observed spectrum in radio and X-ray bands
623: \citep{2002A&A...396..649V}, 
624: which may explain the line width of the narrow component.
625: %% The Balmer-dominated filaments
626: %% do not spatially coincide with the sharp nonthermal X-ray filaments
627: %% described by \citet{2005ApJ...634..376W}, 
628: %% suggesting a reduced CR acceleration efficiency at these locations.  
629: %% This is also consistent with relatively small amount of estimated 
630: %% gas acceleration in the precursor. 
631: %% Although the CRs in the Balmer filaments
632: %% are unlikely to reach the
633: %% $\sim$100 TeV energies inferred for the synchrotron filaments 
634: %% %(Warren et al. 2005), 
635: %% \citep{2005ApJ...634..376W},
636: %% they should still be energetic enough to produce 
637: %% some amplification of the preshock magnetic field.
638: 
639: %%is required
640: 
641: %% Recently, \citet{2006astro.ph.11306G}
642: %% suggested that 
643: %% a damping of the lower-hybrid wave excited in the CR precursor
644: %% heats electrons to  T$_{\mathrm e}$ $\gtrsim 10^6$ K.
645: %% %before they enter the shock front, by . 
646: %% Such a high temperature is less favored by 
647: %% our observed `minor' ($\sim5$) increase of \halpha\ intensity,
648: %% but a detailed model is required to assess this issue.
649: 
650: %% in general,
651: %% it might be 
652: %% %It is not clear whether such a high temperature can be 
653: %% consistent with , 
654: %% and detailed modeling is required 
655: 
656: %% Such a high temperature to be consistent with 
657: %% our observed `minor' ($\sim5$) increase of \halpha\ intensity, 
658: %% a very sharp temperature increase is required 
659: %% and the \halpha\ emission from precursor should be dominated by
660: %% low temperature region. % in the precursor.
661: 
662: %% It seems rather difficult (but not impossible)
663: %% for this high temperature to be consistent with
664: %% the observed `minor' ($\sim5$) increase of \halpha\ intensity in this 
665: %% paper.
666: 
667: 
668: %% A significant fraction of H atoms should be ionized within the
669: %% TP to compensate the increase of the excitation rate, 
670: %% while the number of survived neutrals should be enough to form
671: %% the broad component in the postshock area.
672: %% %
673: %% %(which is not impossible because %
674: %% %
675: %% We note that the ionization time
676: %% scale 
677: %% %Although a significantly high excitation rate is expected at this high
678: %% %temperature, the estimated intensity increase of $\sim 90$ might be
679: %% %explained if we consider significant ionizations of neutrals in the TP.
680: %% %The ionization time scale 
681: %% at this
682: %% temperature is comparable to the shock crossing time for the measured 
683: %% width ($\sim 10^{16}$ cm) of the TP.
684: %% %XXXX also suggested the proton temperature is nevertheless same, while
685: %% %they tend to  
686: 
687: To conclude, our SUBARU observation clearly
688: has shown  the line broadening and the Doppler shift 
689: between the preshock gas and postshock gas.
690: This strongly suggests the existence of 
691: the thin precursor. % suggested from the observed
692: Furthermore, the precursor itself is likely to be resolved.
693: Given the lack of observational constraints on CR ion acceleration in SNR
694: shocks, the tentatively measured width ($\sim 10^{16}$ cm) of the thin
695: precursor whose primary candidate is the CR precursor is promising.
696: Follow-up observation, such as a high resolution imaging with {\em HST},
697: would clearly 
698: resolve the structures of the precursor.
699: 
700: 
701: \acknowledgments{We thank to the anonymous referee for valuable 
702: comments. %, which helped us to improve this paper.
703: This work was supported by the 
704: Korea Research Foundation (grant No. R14-2002-058-01003-0).
705: JJL has been supported in part by the BK 21 program.}
706: 
707: %\bibliography{ms}
708: 
709: %\begin{thebibliography}{32}
710: \begin{thebibliography}{19}
711: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
712: 
713: \bibitem[{{Bell} \& {Lucek}(2001)}]{2001MNRAS.321..433B}
714: {Bell}, A.~R. \& {Lucek}, S.~G. 2001, \mnras, 321, 433
715: 
716: \bibitem[{{Blandford} \& {Eichler}(1987)}]{1987PhR...154....1B}
717: {Blandford}, R. \& {Eichler}, D. 1987, \physrep, 154, 1
718: 
719: \bibitem[{{Chevalier} {et~al.}(1980){Chevalier}, {Kirshner}, \&
720:   {Raymond}}]{1980ApJ...235..186C}
721: {Chevalier}, R.~A., {Kirshner}, R.~P., \& {Raymond}, J.~C. 1980, \apj, 235, 186
722: 
723: \bibitem[{{Chevalier} \& {Raymond}(1978)}]{1978ApJ...225L..27C}
724: {Chevalier}, R.~A. \& {Raymond}, J.~C. 1978, \apjl, 225, L27
725: 
726: \bibitem[{{Ghavamian} {et~al.}(2000){Ghavamian}, {Raymond}, {Hartigan}, \&
727:   {Blair}}]{2000ApJ...535..266G}
728: {Ghavamian}, P., {Raymond}, J., {Hartigan}, P., \& {Blair}, W.~P. 2000, \apj,
729:   535, 266
730: 
731: \bibitem[{{Ghavamian} {et~al.}(2001){Ghavamian}, {Raymond}, {Smith}, \&
732:   {Hartigan}}]{2001ApJ...547..995G}
733: {Ghavamian}, P., {Raymond}, J., {Smith}, R.~C., \& {Hartigan}, P. 2001, \apj,
734:   547, 995
735: 
736: \bibitem[{{Haffner} {et~al.}(2003){Haffner}, {Reynolds}, {Tufte}, {Madsen},
737:   {Jaehnig}, \& {Percival}}]{2003ApJS..149..405H}
738: {Haffner}, L.~M., {Reynolds}, R.~J., {Tufte}, S.~L., {Madsen}, G.~J.,
739:   {Jaehnig}, K.~P., \& {Percival}, J.~W. 2003, \apjs, 149, 405
740: 
741: \bibitem[{{Hanuschik}(2003)}]{2003A&A...407.1157H}
742: {Hanuschik}, R.~W. 2003, \aap, 407, 1157
743: 
744: \bibitem[{{Hester} {et~al.}(1994){Hester}, {Raymond}, \&
745:   {Blair}}]{1994ApJ...420..721H}
746: {Hester}, J.~J., {Raymond}, J.~C., \& {Blair}, W.~P. 1994, \apj, 420, 721
747: 
748: \bibitem[{{Kamper} \& {van den Bergh}(1978)}]{1978ApJ...224..851K}
749: {Kamper}, K.~W. \& {van den Bergh}, S. 1978, \apj, 224, 851
750: 
751: \bibitem[{{Korreck}(2005)}]{2005PhDT..........K}
752: {Korreck}, K. 2005, Ph.D.~Thesis
753: 
754: \bibitem[{{Lee} {et~al.}(2004){Lee}, {Koo}, \&
755:   {Tatematsu}}]{2004ApJ...605L.113L}
756: {Lee}, J., {Koo}, B., \& {Tatematsu}, K. 2004, \apjl, 605, L113
757: 
758: \bibitem[{{Lim} \& {Raga}(1996)}]{1996MNRAS.280..103L}
759: {Lim}, A.~J. \& {Raga}, A.~C. 1996, \mnras, 280, 103
760: 
761: \bibitem[{{Noguchi} {et~al.}(2002){Noguchi}, {Aoki}, {Kawanomoto}, {Ando},
762:   {Honda}, {Izumiura}, {Kambe}, {Okita}, {Sadakane}, {Sato}, {Tajitsu},
763:   {Takada-Hidai}, {Tanaka}, {Watanabe}, \& {Yoshida}}]{2002PASJ...54..855N}
764: {Noguchi}, K., {Aoki}, W., {Kawanomoto}, S., {Ando}, H., {Honda}, S.,
765:   {Izumiura}, H., {Kambe}, E., {Okita}, K., {Sadakane}, K., {Sato}, B.,
766:   {Tajitsu}, A., {Takada-Hidai}, T., {Tanaka}, W., {Watanabe}, E., \&
767:   {Yoshida}, M. 2002, \pasj, 54, 855
768: 
769: \bibitem[{{Smith} {et~al.}(1994){Smith}, {Raymond}, \&
770:   {Laming}}]{1994ApJ...420..286S}
771: {Smith}, R.~C., {Raymond}, J.~C., \& {Laming}, J.~M. 1994, \apj, 420, 286
772: 
773: \bibitem[{{Sollerman} {et~al.}(2003){Sollerman}, {Ghavamian}, {Lundqvist}, \&
774:   {Smith}}]{2003A&A...407..249S}
775: {Sollerman}, J., {Ghavamian}, P., {Lundqvist}, P., \& {Smith}, R.~C. 2003,
776:   \aap, 407, 249
777: 
778: \bibitem[{{V{\" o}lk} {et~al.}(2002){V{\" o}lk}, {Berezhko}, {Ksenofontov}, \&
779:   {Rowell}}]{2002A&A...396..649V}
780: {V{\" o}lk}, H.~J., {Berezhko}, E.~G., {Ksenofontov}, L.~T., \& {Rowell}, G.~P.
781:   2002, \aap, 396, 649
782: 
783: \bibitem[{{van den Bergh}(1971)}]{1971ApJ...168...37V}
784: {van den Bergh}, S. 1971, \apj, 168, 37
785: 
786: \bibitem[{{Warren} {et~al.}(2005){Warren}, {Hughes}, {Badenes}, {Ghavamian},
787:   {McKee}, {Moffett}, {Plucinsky}, {Rakowski}, {Reynoso}, \&
788:   {Slane}}]{2005ApJ...634..376W}
789: {Warren}, J.~S., {Hughes}, J.~P., {Badenes}, C., {Ghavamian}, P., {McKee},
790:   C.~F., {Moffett}, D., {Plucinsky}, P.~P., {Rakowski}, C., {Reynoso}, E., \&
791:   {Slane}, P. 2005, \apj, 634, 376
792: 
793: \end{thebibliography}
794: 
795: 
796: \end{document}
797: 
798: 
799: 
800: 
801: %%% Local Variables: ***
802: %%% TeX-master: "ms" ***
803: %%% mode:flyspell ***
804: %%% mode:auto-fill ***
805: %%% End: ***
806: