1: %\documentclass{emulateapj}

2: %\usepackage{amsmath}

3: %\usepackage{graphicx}

4: %\usepackage{float}

5: %\usepackage{amsmath}

6: %\usepackage{epsfig,floatflt}

7: %\usepackage{subfigure}

8: %\usepackage{apjfonts}

9: 

10: %\documentclass{aastex}

11: 

12: \documentclass[10pt,preprint]{aastex}

13: 

14: %\documentclass{emulateapj}

15: %\usepackage{emulateapj5}

16: %%\usepackage{rotating}

17: 

18: %\usepackage{apjfonts}

19: 

20: %\font\bften=cmbx10 scaled 1000 \font\tenrm=cmr10 scaled 1000

21: %\font\eightrm=cmr8 scaled 1000 \font\sevenrm=cmr7 scaled 1000

22: 

23: \newcommand{\myemail}{wangxf@astro.berkeley.edu}

24: 

25: 

26: \shorttitle{Light Echo for SN 2006X} \shortauthors{Wang et al.}

27: 

28: \def\gsim{\;\lower4pt\hbox{${\buildrel\displaystyle >\over\sim}$}\;}

29: \def\lsim{\;\lower4pt\hbox{${\buildrel\displaystyle <\over\sim}$}\;}

30: \def\grls{\;\lower4pt\hbox{${\buildrel\displaystyle >\over <}$}\;}

31: 

32: \begin{document}

33: 

34: \title{The Detection of a Light Echo from the Type Ia \\ Supernova 2006X in M100}

35: 

36: \author{Xiaofeng Wang\altaffilmark{1,2}, Weidong Li\altaffilmark{1},

37: Alexei V. Filippenko\altaffilmark{1}, Ryan J. Foley\altaffilmark{1},

38: \\ Nathan Smith\altaffilmark{1}, and Lifan Wang\altaffilmark{3}}

39: 

40: \altaffiltext{1}{Department of Astronomy, University of California,

41: Berkeley, CA 94720-3411, USA; wangxf@astro.berkeley.edu .}

42: \altaffiltext{2}{Physics Department and Tsinghua Center for

43: Astrophysics (THCA), Tsinghua University, Beijing, 100084, China;

44: wang\_xf@mail.tsinghua.edu.cn .} \altaffiltext{3}{Physics

45: Department, Texas A\&M University, College Station, TX 77843 .}

46: 

47: \begin{abstract}

48: 

49: We report the discovery of a light echo (LE) from the Type Ia

50: supernova (SN) 2006X in the nearby galaxy M100. The presence of the

51: LE is supported by analysis of both the {\it Hubble Space Telescope

52: (HST)} Advanced Camera for Surveys (ACS) images and the Keck optical

53: spectrum that we obtained at $\sim$300~d after maximum brightness.

54: In the image procedure, both the radial-profile analysis and the

55: point-spread function (PSF) subtraction method resolve significant

56: excess emission at 2--5 ACS pixels ($\sim0.05''-0.13''$) from the

57: center. In particular, the PSF-subtracted ACS images distinctly

58: appear to have an extended, ring-like echo. Due to limitations of

59: the image resolution, we cannot confirm any structure or flux within

60: 2 ACS pixels from the SN. The late-time spectrum of SN 2006X can be

61: reasonably fit with two components: a nebular spectrum of a normal

62: SN~Ia and a synthetic LE spectrum. Both image and spectral analysis

63: show a rather blue color for the emission of the LE, suggestive of a

64: small average grain size for the scattering dust. Using the Cepheid

65: distance to M100 of 15.2 Mpc, we find that the dust illuminated by

66: the resolved LE is $\sim$27--170~pc from the SN. The echo inferred

67: from the nebular spectrum appears to be more luminous than that

68: resolved in the images (at the $\sim$2$\sigma$ level), perhaps

69: suggesting the presence of an inner echo at $<$2 ACS pixels

70: ($\sim0.05''$). It is not clear, however, whether this possible

71: local echo was produced by a distinct dust component (i.e., the

72: local circumstellar dust) or by a continuous, larger distribution of

73: dust as with the outer component. Nevertheless, our detection of a

74: significant echo in SN 2006X confirms that this supernova was

75: produced in a dusty environment having small dust particles.

76: 

77: \end{abstract}

78: 

79: \keywords {circumstellar matter -- dust, extinction -- supernovae:

80: general -- supernovae: individual (SN 2006X)}

81: 

82: \section{Introduction}

83: 

84: Light echoes (LEs) are produced when light emitted by the explosive

85: outburst of some objects is scattered toward the observer by the

86: foreground or surrounding dust, with delayed arrival time due to the

87: longer light path. This phenomenon is rare, having been observed

88: only around a few variable stars in the Galaxy, and around several

89: extragalactic supernovae (SNe). The best-studied events are SN 1987A

90: (Schaefer 1987; Gouiffes et al. 1988; Chevalier \& Emmering 1988;

91: Crotts 1988; Crotts, Kunkel, \& McCarthy 1989; Bond et al. 1990; Xu

92: et al. 1995) and the peculiar star V838 Mon (Bond et al. 2003).

93: Other SNe with LEs include the Type II SNe 1993J (Liu et al. 2002;

94: Sugerman 2003), 2002hh (Meikle et~al. 2006; Welch et al. 2007), and

95: 2003gd (Sugerman 2005; Van Dyk et al. 2006), as well as the Type Ia

96: SNe 1991T (Schmidt et al. 1994; Sparks et al. 1999), 1998bu

97: (Cappellaro et al. 2001; Garnavich et al. 2001), and possibly 1995E

98: (Quinn et al. 2006). Besides their spectacular appearance, LEs offer

99: a unique means to diagnose the composition, distribution, and

100: particle size of the scattering dust. In particular, LEs from the

101: circumstellar environments might provide constraints on SN

102: progenitors.

103: 

104: The Type Ia SN 2006X was discovered on 2006 February 7.10 (UT dates

105: are used throughout this paper) by S. Suzuki and M. Migliardi (IAUC

106: 8667, CBET 393) in the nearby spiral galaxy NGC 4321 (M100).

107: Extensive photometric and spectroscopic coverage is presented by

108: Wang et al. (2007, hereafter W07). They suggest that SN 2006X is

109: highly reddened [$E(B - V)_{\rm host} = 1.42 \pm 0.04$ mag] by

110: abnormal dust with $\Re_{V} = 1.48 \pm 0.06$. Its early-epoch

111: spectra are characterized by strong, high-velocity features of both

112: intermediate-mass and iron-group elements. In addition to the

113: anomalous extinction and the very rapid expansion, SN 2006X exhibits

114: a continuum bluer than that of normal SNe~Ia. Moreover, its

115: late-time decline rate in the $B$ band is slow, $\beta = 0.92 \pm

116: 0.05$ mag (100~{\rm d})$^{-1}$, significantly below the 1.4 mag

117: (100~{\rm d})$^{-1}$ rate observed in normal SNe~Ia and comparable

118: to the decay rate of 1.0 mag (100~{\rm d})$^{-1}$ expected from

119: $^{56}$Co $\rightarrow$ $^{56}$Fe decay. This may suggest additional

120: energy sources besides radioactive decay, such as the interaction of

121: the supernova ejecta with circumstellar material (CSM) and/or a LE.

122: 

123: Attempts to detect the CSM in SNe~Ia in different wavebands were

124: unsuccessful before SN 2006X, and only some upper limits could be

125: placed (see Patat et al. 2007a, and references therein) except for

126: the peculiar SNe~Ia/IIn 2002ic (Hamuy et al. 2003; Deng et al. 2004;

127: Wang et al. 2004; Wood-Vasey et~al. 2004) and 2005gj (Aldering

128: et~al. 2006; Prieto et~al. 2007). Recent progress in this respect

129: was made from high-resolution spectroscopy by Patat et~al. (2007b,

130: hereafter P07), who find time-variable Na~I~D absorption lines in

131: spectra of SN 2006X. This has been interpreted as the detection of

132: CSM within a few 10$^{16}$ cm ($\sim 0.01$ pc) from the explosion

133: site of the supernova. With the inferred velocity, density, and

134: location of the CSM, P07 proposed that the companion star of the

135: progenitor of SN 2006X is most likely to be a red giant (but see

136: Hachisu et~al. 2007, who present a main-sequence star model with

137: mass stripping). Note, however, that SN 2006X exhibited somewhat

138: abnormal features in spectra and photometry; it may not represent a

139: typical SN~Ia. Multi-epoch, high-resolution spectral observations of

140: SN 2007af, a normal SN~Ia, do not reveal any significant signature

141: of CSM absorption (Simon et~al. 2007).

142: 

143: In this paper we report the discovery of an optical LE around SN

144: 2006X, with evidence from {\it Hubble Space Telescope (HST)}

145: Advanced Camera for Surveys (ACS) images and Keck optical spectra.

146: The paper is organized as follows. In \S 2 we briefly describe the

147: late-epoch data available for SN 2006X, while the data analysis and

148: the interpretation are presented in \S 3. We discuss the properties

149: of the light echo and the underlying dust in \S 4. Our conclusions

150: are given in \S 5.

151: 

152: \section{Observations}

153: 

154: \subsection{{\it HST} Archival Images}

155: Several epochs of {\it HST} data covering the site of SN 2006X are

156: publicly available in the MAST archive. The pre-discovery images

157: were taken on 1993 December 31 (Proposal ID 5195: PI, Sparks) by the

158: Wide Field Planetary Camera 2 (WFPC2) in the F439W and F555W

159: filters, with the same integration time of 1800~s. The images taken

160: prior to the SN explosion allow us to examine the immediate

161: environment of the progenitor, whereas the post-explosion

162: observations enable us to search for a possible LE. The most recent,

163: post-discovery images of SN 2006X were obtained with the High

164: Resolution Channel (HRC, with a mean spatial resolution of $0.026''$

165: pixel$^{-1}$) of {\it HST}/ACS on 2006 May 21 (90~d after $B$

166: maximum) and on 2006 December 25 (308~d after $B$ maximum),

167: respectively (GO--10991; PI, Arlin Crotts). At $t = 90$~d, the SN

168: was imaged in F435W (1480~s), F555W (1080~s), and F775W (1080~s),

169: while at $t = 308$~d, the SN was again observed in the same three

170: bandpasses, with the exposure times of 920~s, 520~s, and 520~s,

171: respectively.

172: 

173: The standard {\it HST} pipeline was employed to pre-process the

174: images and remove cosmic-ray hits. In Figure 1 we show the pre- and

175: post-explosion {\it HST} images of SN 2006X in the F555W filter.

176: This pre-discovery image does not reveal any significant source

177: brighter than 24.0 mag in F555W, excluding the possibility of a

178: significant star cluster at the location of SN 2006X. Neither of the

179: two post-discovery images exhibits any resolved LE arcs or rings

180: around SN 2006X. The magnitudes of SN 2006X were measured from the

181: {\it HST} ACS images with both the Dolphot method (Dolphin 2000) and

182: the Sirianni procedure (Sirianni et~al. 2005), and the mean

183: photometry is given in Table 1.

184: 

185: %The left panel shows the field of SN 2006X and neighboring brighter

186: %field stars, as imaged with WFPC2 on 1993 December 31. The middle

187: %panel presents same field observed on day 90 after maximum SN

188: %brightness. In the right panel, we show the image obtained on day

189: %308 after maximum.

190: 

191: %Neither of these two images exhibits any resolved LE arcs or rings

192: %around SN 2006X. In Table 1 we give the magnitudes of SN 2006X

193: %measured from the {\it HST}/ACS archival images.

194: 

195: \subsection{Keck Optical Spectrum}

196: Observations of nebular-phase spectra provide an alternative way to

197: explore the possibility of an LE around SNe, as the scattered,

198: early-phase light will leave a noticeable imprint on the nebular

199: spectra (e.g., Schmidt et~al. 1994; Cappellaro et~al. 2001) when the

200: SN becomes dimmer. Two very late-time spectra of SN 2006X taken at t

201: $\approx$ 277 and 307 days after $B$ maximum were published by W07

202: (see their Fig. 19), which were obtained by the Keck telescopes at

203: the W. M. Keck Observatory: one with the Low Resolution Imaging

204: Spectrometer (LRIS; Oke et al. 1995) mounted on the 10~m Keck I

205: telescope, and the other with the Deep Extragalactic Imaging

206: Multi-Object Spectrograph (DEIMOS) mounted on the 10~m Keck II

207: telescope. In the following analysis we focus on the LRIS spectrum

208: taken at $t \approx 277$~d because of its wider wavelength coverage.

209: 

210: \section{Data Analysis}

211: \subsection{Late-Time Light Curves}

212: Figure 2 shows the absolute $B$, $V$, and $I$ light curves of SN

213: 2006X and SN~Ia 1996X (Salvo et~al. 2001). The former were obtained

214: using the Cepheid distance $\mu = 30.91 \pm 0.14$ mag (Freedman

215: et~al. 2001), and corrected for extinction in the Milky Way

216: ($A_{V}({\rm MW})$ = 0.08 mag; Schlegel et~al. 1998) and in the host

217: galaxy ($A_{V}({\rm host})$ = 2.10 mag; W07). The distance modulus

218: and host-galaxy extinction for SN 1996X are derived by Wang et~al.

219: (2006) and Jha et al. (2007) using independent methods, and we

220: adopted the mean values $\mu = 32.11 \pm 0.15$ mag ($H_{0} = 72$ km

221: s$^{-1}$ Mpc$^{-1}$ and $A_{V}({\rm host}) = 0.08$ mag are assumed

222: throughout this paper). The absolute magnitudes of these two SNe are

223: similar near maximum, except in the $I$ band where SN 2006X is

224: $\sim$0.4 mag fainter than SN 1996X.

225: 

226: Noticeable differences between the two SNe emerge in $B$ one month

227: after maximum light, when SN 2006X begins to decline slowly at a

228: rate of $0.92 \pm 0.05$ mag (100~d)$^{-1}$. The discrepancy reaches

229: about 0.9 mag in $B$ at $t = 308$~d, while it is $\sim$0.7 mag in

230: $V$ and $\sim$0.2 mag in $I$. This suggests that the emission of SN

231: 2006X is 130\% $\pm$ 60\% higher in $B$, 90\% $\pm$ 40\% higher in

232: $V$, and 20\% $\pm$ 20\% higher in $I$ with respect to SN 1996X. The

233: large error bars primarily reflect the uncertainty in the distances.

234: 

235: The apparently overluminous behavior seen in SN 2006X in the tail

236: phase is possibly linked to the light scattering of the surrounding

237: dust, though the interaction of the SN ejecta with the CSM produced

238: by the progenitor system and/or the excess trapping of photos and

239: positrons (created in $^{56}$Co $\longrightarrow$ $^{56}$Fe decays

240: within the ejecta) cannot be ruled out. The resultant LE, if

241: present, may not be directly resolved even in the {\it HST}/ACS

242: images at a distance of $\sim$15 Mpc due to the limited angular

243: resolution. To examine this conjecture, in \S 3.2 we compare the

244: PSFs of the SN and local stars, and in \S 3.3 we apply the

245: image-subtraction technique to analyze the SN images.

246: 

247: %In principle, there are several possible explanations for the

248: %apparent overluminosity of SN 2006X during the tail phase: excess

249: %trapping of photos and positrons (created in $^{56}$Co

250: %$\longrightarrow$ $^{56}$Fe decays) within the ejecta, the

251: %interaction of SN ejecta with the CSM produced by the progenitor

252: %system, and a LE produced by the surrounding dust. However, the

253: %strong wavelength dependence of the overluminous components in SN

254: %2006X may favor the dust-scattering interpretation. The resultant

255: %LE, if present, may not be directly resolved even in the {\it

256: %HST}/ACS images at a distance of $\sim$15 Mpc due to the limited

257: %angular resolution. To examine this conjecture, in \S 2.2 we compare

258: %the PSFs of the SN and local stars, and in \S 2.3 we apply the

259: %image-subtraction technique to analyze the SN images.

260: 

261: \subsection{Radial Brightness Profile}

262: 

263: The radial brightness profiles of the images of the SN and local

264: stars in the same field (see Fig. 1) are compared in Figure 3. These

265: were obtained by extracting the flux using different apertures,

266: ranging from 0.1 to 10 pixels with a resolution of 0.1 pixel. The

267: fluxes of the four local stars labeled in Figure 1 are scaled so

268: that the integrated flux within the 10-pixel aperture is the same.

269: Based on the distribution of the radial profiles of these four

270: stars, we derived a mean radial profile with the same integrated

271: flux through Monte Carlo simulations. The radial profile of the four

272: stars was thus normalized by the peak flux of the simulated radial

273: profile.

274: 

275: One can see that the star profiles are uniform at large radius but

276: show noticeable scatter within 2 pixels from the center. For

277: comparison, the central flux of SN 2006X is scaled to be 1.0, with

278: the assumption that the central region of the SN image was not

279: affected by any LE. At $t = 90$~d the SN profile does not show a

280: significant difference from that of the local stars. At $t = 308$~d

281: the SN profiles appear distinctly broader at radii of 2--4 pixels,

282: especially in the F435W and F555W images. Note that the SN data are

283: quite steep at around 1 pixel, probably due to noise.

284: 

285: The inset plot of Figure 3 shows the residual of the radial profile

286: between SN 2006X and the local stars. This was obtained by

287: subtracting the simulated radial profile of the stars from that of

288: the SN. Also plotted is the scatter of the simulated profile of the

289: local stars. At $t = 308$~d, the SN shows significant extra flux in

290: the F435W, F555W, and F775W filters at radii of $\sim$2 to 5 pixels,

291: suggesting the presence of a LE. Such a residual flux was not

292: present at $t = 90$~d.

293: 

294: One can see some structure (peaks and valleys) at $<$2 pixels in the

295: inset residual plot of Figure 3. These alternating negative/positive

296: residuals clearly show that the substructure within the inner 2

297: pixels cannot be trusted, and could result from the misalignment of

298: the peak surface brightness of the images. Of course, it is possible

299: that part of the LE is so close to the SN, but the above analysis

300: cannot definitively reveal it.

301: 

302: Integrating the overall residual emission in the range 2--10 pixels,

303: we find that the observed LE brightness is $\sim$22.8 mag in F435W,

304: $\sim$22.0 mag in F555W, and $\sim$22.1 mag in F775W. Its

305: contribution to the total flux of the SN + LE is $\sim$29\% in

306: F435W, $\sim$27\% in F555W, and $\sim$11\% in F775W. In view of

307: potential additional LE emission at radii $<$2 pixels, these values

308: are probably lower limits to the true brightness of the LE.

309: 

310: %\begin{figure}[htbp]

311: %\figurenum{4}\vspace{-0.5cm}\hspace{-0.5cm}

312: %\includegraphics[angle=0,width=100mm]{06Xint.eps}

313: %\plotone{06Xint.eps}

314: %\includegraphics[angle=0,width=95mm]{s96sed.eps}

315: %\vspace{-0.0cm} \caption{The ratio of the integrated net flux of the

316: %residual emission for SN 2006X and the local star, plotted as a

317: %function of the radius. The black curve represents the SN, color

318: %curves represent the field stars as labeled in Fig. 1.}

319: %\vspace{-0.0cm} \label{fig:four}

320: %\end{figure}

321: 

322: 

323: Although Star 1 and SN 2006X show some diffraction spikes in Figure

324: 1, the spikes have the same shape and orientation for all stars in

325: the field. Thus, (a) they should affect the radial surface

326: brightness profiles of all stars in the same way, and not affect the

327: excess light from an echo, and (b) they should be adequately removed

328: by the image subtraction procedure (\S 3.3).

329: 

330: The difference between the radial profile of SN 2006X and other

331: stars can also be demonstrated by their measured full width at

332: half-maximum intensity (FWHM). Table 2 lists the FWHM of SN 2006X

333: and the average value of several local stars, obtained by running

334: the IRAF\footnote{IRAF, the Image Reduction and Analysis Facility,

335: is distributed by the National Optical Astronomy Observatory, which

336: is operated by the Association of Universities for Research in

337: Astronomy, Inc. (AURA) under cooperative agreement with the National

338: Science Foundation (NSF).} ``imexamine'' task in three modes: $r$

339: (radial profile Gaussian fit), $j$ (line 1D Gaussian fit), and $k$

340: (column 1D Gaussian fit). At $t = 90$~d, the PSF of SN 2006X is

341: comparable to that of the average values of the local stars, while

342: at $t = 308$~d, the SN exhibits a significantly broader profile. The

343: FWHM increases by about 0.3 pixel in the $r$-profile and by

344: $\sim$1.0 pixel in the $j$-profile and $k$-profile with respect to

345: the local stars. The reasonable interpretation is that the PSF is

346: broadened by scattered radiation (that is, the LE).

347: 

348: \subsection{Light Echo Images}

349: 

350: The radial-profile study suggests the presence of a LE in SN 2006X.

351: In this section, we apply image subtraction to provide further

352: evidence for the LE, and study its two-dimensional (2D) structure.

353: 

354: We extract a small section ($20 \times 20$ pixels) centered on SN

355: 2006X and Star 1 (the brightest star in the field), and align their

356: peak pixels to high precision (0.01 pixel). We then scale Star 1 so

357: that its peak has the same counts as that of SN 2006X, and subtract

358: it from the SN 2006X image. The underlying assumption is the same as

359: in our radial-profile study: the central peak of SN 2006X is not

360: affected by any LE.

361: 

362: Figure 4 shows the PSF-subtracted images of SN 2006X. The left panel

363: shows the subtracted images at the original {\it HST}/ACS

364: resolution. To bring out more details, the middle panel shows

365: subsampled images by using a cubic spline function to interpolate

366: one pixel into $8 \times 8$ pixels. The right panel has three

367: circles (with radii of 2, 4, and 6 pixels, respectively)

368: overplotted. The residual images all show an extended, bright,

369: ring-like feature around the supernova, consistent with the general

370: expected appearance of a LE. These features emerge primarily at

371: radii of 2--4 pixels (or $0.05''$--$0.11''$) in the images,

372: consistent with those derived above from the radial profiles. The

373: central structure seen within a circle of radius 2 pixels (e.g., the

374: asymmetric feature in F435W, the double features in F775W, and the

375: arc in F555W) are not to be trusted; due to the limited spatial

376: resolution, the images used for the image subtraction may not be

377: perfectly aligned (in terms of the geometry and/or the flux of the

378: central regions), and some artifacts could be introduced at the

379: center of the subtracted images. Similarly, the apparent clumps

380: within the echo ring are not reliable, generally being only a few

381: pixels in size.

382: 

383: The integrated flux, measured from the PSF-subtracted images at

384: 2--10 pixels from the SN site, contributes to the total flux of SN +

385: LE by $\sim$33\% in F435W, $\sim$29\% in F555W, and $\sim$9\% in

386: F775W. This is fully consistent with the above estimate from the

387: radial-profile analysis, taking into account the uncertainty in the

388: PSF subtraction. The brightness of the LE component is estimated to

389: be $\sim$22.7 mag in F435W, $\sim$21.9 mag in F555W, and $\sim$22.3

390: mag in F775W.

391: 

392: As with the radial-profile analysis, the PSF-subtraction method

393: might remove some fraction of flux from the LE itself; it had been

394: assumed that none of the flux in the central 2-pixel radius is

395: produced by the LE, but this might be incorrect. Thus, our estimate

396: of the echo flux from the image analysis may be only a lower limit

397: of the true LE emission. In view of the image analysis, we cannot

398: verify or rule out that the LE may be distributed continuously from

399: the SN site to an angular radius of $\sim$6 pixels ($0.15''$).

400: 

401: \subsection{Light Echo Spectrum}

402: 

403: A consistency check for the existence of a LE around a source can

404: also be obtained by comparing the observed supernova spectrum and

405: the synthetic spectrum using an echo model. The observed spectrum

406: should be a combination of the intrinsic late-time SN spectrum and

407: the early-time scattered SN spectrum. Inspection of the late-epoch

408: Keck spectrum (see Fig. 19 of W07) clearly reveals that SN 2006X

409: behaves unlike a normal SN~Ia, showing a rather blue continuum at

410: short wavelengths and a broad absorption feature near 6100~\AA\

411: (probably due to Si~II $\lambda$6355).

412: 

413: To construct the composite spectrum containing the echo component,

414: we use the nebular-phase spectrum of SN 1996X to approximate that of

415: SN 2006X. SN 1996X is a normal SN~Ia in the elliptical galaxy NGC

416: 5061 (Salvo et~al. 2001), with $\Delta m_{15} = 1.30 \pm 0.05$ mag,

417: similar to that of SN 2006X (W07). Late-time optical spectra with

418: wide wavelength coverage and high signal-to-noise ratio (S/N) are

419: available on day 298 for SN 1996X (Salvo et~al. 2001;

420: http://bruford.nhn.ou.edu/$^{\thicksim}$suspect/) and on day 277 for

421: SN 2006X (W07). Comparing the spectrum of SN 2006X obtained at $t =

422: 277$~d with that taken at $t = 307$~d, we found that the overall

423: spectral slope changed little during this period. We thus could

424: extrapolate the original nebular spectra $t = 308$~d, a phase when

425: both SNe have relatively good multicolor photometry. To completely

426: match the spectrum of SN 2006X, the spectral flux of SN 1996X was

427: multiplied by a factor of 3.0 caused by the difference in distances.

428: Extinction corrections have also been applied to the nebular spectra

429: of these two SNe (W07; Wang et~al. 2006).

430: 

431: We considered the cases of both SN 2006X and SN 1996X as the central

432: pulse source when deriving the echo spectrum. The observed spectra

433: of SN 2006X are available at eleven different epochs from about

434: $-$1~d to 75~d after $B$ maximum, while 14 spectra of SN 1996X are

435: available from about $-$4~d to 87~d after $B$ maximum (Salvo et~al.

436: 2001). The above spectra were properly dereddened\footnote{Here we

437: assume that the dust surrounding SN 2006X is a plane-parallel slab

438: and/or shell, so that both the SN and the LE were affected by

439: roughly the same amount of extinction.} and interpolated to achieve

440: uniform phase coverage. Regardless of the original flux calibration,

441: all of the input spectra have been recalibrated according to their

442: light curves at comparable phases (W07; Salvo et~al. 2001) and

443: corrected for the effects of scattering using a similar function,

444: $S(\lambda) \propto \lambda^{-\alpha}$ (e.g., Suntzeff et~al. 1988;

445: Cappellaro et~al. 2001). These corrected spectra were then coadded

446: and scaled, together with the nebular spectrum of SN 1996X, to match

447: the nebular spectrum of SN 2006X.

448: 

449: The best-fit $\alpha$ values obtained for the combinations of SN

450: 2006X (near $B$ maximum) + SN 1996X (nebular) and SN 1996X (near $B$

451: maximum) + SN 1996X (nebular) are $3.0 \pm 0.3$ and $3.3 \pm 0.5$,

452: respectively. One can see that the combination of SN 2006X + SN

453: 1996X gives a somewhat better fit to the observed spectrum of SN

454: 2006X. This is not surprising; the spectrum of SN 2006X differs from

455: that of a normal SN~Ia at early times, showing extremely broad and

456: blueshifted absorption minima (W07). The large value of $\alpha$ may

457: indicate a small grain size for the scattering dust. The composite

458: nebular spectrum and the underlying echo spectrum are compared with

459: the observed spectrum of SN 2006X in Figure 5 (upper and middle

460: panels). Given the simple assumption of the scattering function,

461: incomplete spectral coverage, and intrinsic spectral difference

462: between SN 1996X and SN 2006X, the agreement between the observation

463: and the model is satisfactory, with major features in the spectrum

464: well matched. This provides independent, strong evidence for the LE

465: scenario. However, the broad emission peak seen at $\lambda \approx

466: 4300$--4500~\AA\ cannot be reasonably fit by the echo model (see

467: Fig. 5); this mismatch is probably produced by intrinsic features in

468: the nebular spectrum of SN 2006X.

469: 

470: \subsection{Light-Echo Luminosity and Color}

471: We can constrain the properties of the LE and the underlying dust

472: through the luminosity and colors of the LE. The LE luminosity of SN

473: 2006X has been estimated by analyzing the {\it HST} SN images; it

474: can also be obtained by integrating the echo spectrum shown in

475: Figure 5. The magnitudes of the echo given by different methods are

476: listed in Table 3. For the image-based measurement, the error

477: accounts only for the scatter of the stellar PSF. On the other hand,

478: for the spectrum-based measurement, the error primarily consists of

479: the uncertainties in extinction correction (i.e., $\sim$0.2 mag for

480: SN 2006X and $\sim$0.1 mag for SN 1996X in the $B$ band) and

481: distance modulus (i.e., $\sim$0.14 mag for SN 2006X and $\sim$0.15

482: mag for SN 1996X).

483: 

484: We note that the echo inferred from the spectral fitting seems

485: somewhat brighter than that revealed by the image analysis: $\delta

486: m_{F435W} = -0.6 \pm 0.3$ mag. This difference is also demonstrated

487: in the bottom panel of Figure 5, where the flux ratio of the

488: inferred echo spectrum and the observed spectrum of SN 2006X is

489: plotted as a function of wavelength. Overplotted are the ratios

490: yielded for the photometry of the echo image (circles) and the

491: spectrophotometry of the echo spectrum (squares) in F435W and F555W,

492: respectively. Such a discrepancy, at a confidence level of only

493: $\sim2\sigma$, may suggest that there is some echo emission within a

494: radius of 2 pixels (21\% $\pm$ 12\% of the total flux of SN + LE in

495: F435W and 17\% $\pm$ 11\% in F555W) that was not resolved by the

496: image analysis. Despite this possibility, we must point out that the

497: echo luminosity derived from the echo spectrum may have an error

498: that is actually larger than our estimate, since we did not consider

499: possible uncertainties associated with the spectrum itself and the

500: simple scattering model adopted in our analysis (see \S 3.4).

501: 

502: Assuming that all of the observed differences between the light

503: curves and spectra of SN 2006X and SN 1996X at $t = 308$~d are

504: entirely due to the LE around SN 2006X, we can place an upper limit

505: on the LE brightness as $21.9 \pm 0.3$ mag in F435W and $21.3 \pm

506: 0.3$ mag in F555W. The magnitudes and the resulting color are not

507: inconsistent with those presented in Table 3, especially in the case

508: of the spectral fit which likely takes into account most of the echo

509: emission. This leaves little room for other possible mechanisms for

510: the extra emission, suggesting that the echo is the primary cause of

511: the abnormal overluminosity of SN 2006X at $t = 308$~d.

512: 

513: We find, from analysis of both the {\it HST} images and the nebular

514: spectrum (see Table 3), that the LE has an average color

515: (F435W--F555W)$_{\rm echo}$ = $0.8 \pm 0.3$ mag (this roughly equals

516: $(B -V)_{\rm echo}$ = $0.8 \pm 0.3$ mag), which is much bluer than

517: the SN color at maximum brightness. The LE is clearly brighter in

518: bluer passbands than at redder wavelengths (see Fig. 5). Comparing

519: the colors of the echo and the underlying SN light helps us

520: interpret the dust, as the color shift depends on the scattering

521: coefficient and hence on the dimensions of the dust grains (Sugerman

522: 2003a).

523: 

524: Integrating over the entire SN light curve (W07) from about $-$11~d

525: to 116~d after $B$ maximum yields $(B - V)_{\rm SN} = 1.70$ mag for

526: the overall emission of SN 2006X. The observed change in color,

527: $\Delta(B - V) = -0.9 \pm 0.3$ mag, is much larger than the color

528: shift derived for Galactic dust but is comparable to the change

529: derived for Rayleighan dust\footnote{The Rayleighan dust consists of

530: only small particles with grain size $<0.01~\mu$m, and hence has a

531: scattering efficiency proportional to $\lambda^{-4}$ (Sugerman

532: 2003a).}, $\Delta(B - V)_{\rm max} = -0.96$ mag (Sugerman et~al.

533: 2003b). This is consistent with constraints from the direct spectral

534: fit, which suggests that the dust has a scattering efficiency

535: proportional to $\lambda^{-3.0}$. We thus propose that the dust

536: surrounding SN 2006X is different from that of the Galaxy and may

537: have small-size grains, perhaps with diameter $\lesssim 0.01~\mu$m,

538: reflecting the shorter wavelengths of light more effectively.

539: Smaller dust particles are also consistent with the low value of

540: $\Re_{V} \approx 1.5$ derived by W07.

541: 

542: \subsection{Dust Distance}

543: 

544: Of interest is the distribution of the dust producing the echo; for

545: example, it may be a plane-parallel dust slab or a spherical dust

546: shell. Couderc (1939) was the first to correctly interpret the LE

547: ring observed around Nova Persei 1901. Detailed descriptions of LE

548: geometries can also be found in more recent papers (e.g., Sugerman

549: 2003a; Tylenda et~al. 2004; Patat 2005). In general, the analytical

550: treatment shows that both a dust slab and a dust shell could produce

551: an echo that is a circular ring containing the source. Assuming that

552: the SN light is an instantaneous pulse, then the geometry of an LE

553: is straightforward: the distance of the illuminated dust material

554: lying on the paraboloid can be approximated as

555: \begin{equation}

556: R \approx \frac{{}D^{2}\theta^{2} \mp (ct)^{2}}{2ct},

557: \end{equation}

558: where $D$ is the distance from the SN to the observer, $\theta$ is

559: the angular radius of the echo, $c$ is the speed of light, and $t$

560: is the time since the outburst. The equation with a minus sign

561: corresponds to the single dust slab, while the plus sign represents

562: the case for a dust shell.

563: 

564: As suggested by the analysis of the radial profile and the

565: PSF-subtracted image of the SN, there is a confirmed LE ring

566: $\sim0.08''$ away from the SN, with a possible width of

567: $\sim0.03''$. For this echo of SN 2006X, $ct$ = 0.27 pc, which leads

568: to $R$ $\approx$ 27--120 pc, consistent with the scale of the ISM

569: dust cloud. As the dust cloud in front of SN 2006X seems to be very

570: extended, we do not give the thickness of the dust along the line of

571: sight. Considering the possible echo emission within 2.0 pixel

572: ($\sim0.05''$) inferred from the echo luminosity (see discussion in

573: \S 4.1) and that extending up to 5 pixels ($\sim0.13''$; see Fig.

574: 3), the actual distribution of the dust may be from $<$27~pc to

575: $\sim$170~pc from the SN.

576: 

577: In principle, one can also estimate the distance of the dust itself

578: from the SN through a fit to the observed echo luminosity using the

579: light-echo model (e.g., Cappellaro et~al. 2001), as the actual echo

580: flux is related to the light emitted by the SN, the physical nature

581: of the dust, and the dust geometry. However, current analytical

582: treatments for the LE model must assume some idealized

583: configuration, which may not apply to the dust surrounding SN 2006X

584: that is found to probably have smaller dust grains with $\Re_{V}

585: \approx 1.5$ and a relatively extended distribution. Moreover,

586: multiple scattering processes rather than a single scattering should

587: be considered in the echo model due to the large optical depth

588: measured from the dust: $\tau^{V}_{d} \approx 2.0$ for SN 2006X.

589: Detailed modelling of the LE emission seen in SN 2006X is beyond the

590: scope of this paper.

591: 

592: \section{Discussion}

593: Analysis of both the late-time {\it HST} images and the late-time

594: Keck optical spectrum favors the presence of a LE in SN 2006X, the

595: fourth non-historical SN~Ia with a detection of echo emission.

596: Comparison of the SN 2006X echo with the other three known events,

597: SNe 1991T, 1995E, and 1998bu, shows that the Type Ia echoes may have

598: a wide range of dust distances from $\lesssim$ 10 pc to $\sim$ 210

599: pc. The echo detected in SN 1991T is consistent with being a dust

600: cloud of radius 50 pc (Sparks et~al. 1999), while the echo

601: speculated from SN 1995E probably corresponds to a dust sheet at a

602: distance of $207 \pm 35$ pc (Quinn et~al. 2006). Garnavich et~al.

603: (2001) proposed from the {\it HST} WFPC2 imaging that SN 1998bu may

604: have two echoes, caused by dust at $120 \pm 15$ pc and $<10$ pc away

605: from the SN; the outer echo is consistent with an ISM dust sheet,

606: while the inner component is likely from the CSM dust. On the other

607: hand, the resolved echo image of SN 2006X appears quite extended in

608: the direction perpendicular to the line of sight. This yields a dust

609: distance spanning from $\sim$ 27 pc to $\sim$ 170 pc away from the

610: site of the SN, indicating that the dust causing the LE may not be a

611: thin dust sheet but could be a cloud or shell distribution of the

612: dust around the progenitor or a more complicated dust system.

613: 

614: The echo from SN 2006X is found to be brighter than that of the

615: other three Type Ia echo events. Assuming the echo magnitude listed

616: in Table 3 and the SN peak magnitude derived in W07, one can find

617: that the echo flux with respect to the extinction-corrected peak

618: magnitude of SN 2006X is $\sim$9.6 mag in $V$. Quinn et~al. (2006)

619: proposed that all of the other three Type Ia echoes (SNe 1991T,

620: 1995E, 1998bu) show a striking similarity in their echo brightness

621: relative to the extinction-corrected peak SN brightness, $\Delta V

622: \approx 10.7$ mag. According to the analytical expression of the

623: dust scattering (e.g., Patat 2005), the excess echo brightness from

624: SN 2006X by $\sim$1 mag perhaps suggests a dust distribution closer

625: to the SN, given the similar optical depth for SNe 2006X and 1995E.

626: The SN 2006X echo emission also shows a prominent wavelength

627: dependence, with more light from the shorter wavelengths, suggestive

628: of smaller-size dust around SN 2006X. This is also demonstrated by

629: the difference of the scattering coefficient $\alpha$ required to

630: fit the observed nebular spectrum, which is $\sim$3.0 for SN 2006X,

631: $\sim$2.0 for 1991T (Schmidt et~al. 1994), and $\sim$1.0 for SN

632: 1998bu (Cappellaro et~al. 2001).

633: 

634: In fitting the nebular spectrum, the echo brightness is found to be

635: $\sim$ 60\% brighter than that from the echo image at the

636: $\sim$2$\sigma$ level, likely suggesting the presence of a local

637: echo that was not resolved at the regions close to the SN site.

638: Regarding the location of the echo emission in SN 2006X, one may

639: naturally tie the distribution of the dust underlying the echo to a

640: combination of local CSM dust and distant ISM dust, given the quite

641: extended dust distribution and the small dust grains that were not

642: typical for the ISM dust. Detection of the CSM dust is of particular

643: importance for understanding SN~Ia progenitor models. P07 recently

644: reported the detection of CSM in SN 2006X from variable Na~I~D

645: lines, and they estimate that the absorbing dust is a few $10^{16}$

646: cm from the SN. It is hence expected that an echo very close to the

647: SN ($<$0.01~pc away) should be produced, although the SN UV

648: radiation field could destroy or change the distribution of the

649: surrounding dust particles out to a radius of a few $10^{17}$ cm

650: (Dwek 1983). However, it is not possible for us to detect the

651: emission of such a close CSM echo at $t = 308$~d, since the maximum

652: delayed travel time of the light for this echo is $<$0.1~yr and the

653: SN radiation decreases with time.

654: 

655: As noted by W07, the spectrum of SN 2006X probably showed a UV

656: excess at $t \approx 30$~d. This may be a signature of the nearby

657: CSM claimed by P07, but the S/N of the spectrum is quite low below

658: 4000~\AA. In this case, the possible echo emission at $<$27~pc

659: inferred from the nebular spectrum at $t = 308$~d could result from

660: a dust shell that is farther out than that claimed by P07. This is

661: possible if the CSM dust around SN 2006X has multiple shells, such

662: as the dust ring (or shell) of a planetary nebula (Wang 2005) and

663: nova-like shells.

664: 

665: %Another possible signature for the presence of CSM

666: %dust in SN 2006X is the poor fit to the emission feature of the

667: %observed spectrum at 4300--4500~\AA\ with the LE model. This

668: %mismatch could be produced by intrinsic features in the nebular

669: %spectrum of SN 2006X, or it perhaps suggests a more complicated

670: %scattering process by the dust, such as multiple scattering

671: %(Chevalier 1986; Patat 2005) or composite scattering of the dust

672: %(CSM + ISM). However, the multiple scattering caused by higher

673: %optical depth is less likely to account for the same emission

674: %feature seen in the nebular-phase spectrum of SN 2004dt (Altavilla

675: %et~al. 2007), which, like SN 2006X, displayed features of

676: %high-velocity intermediate-mass elements in the spectra and a slower

677: %decline rate in the $B$-band light curve but with relatively lower

678: %optical depth (Wang et~al. 2008, in prep.). Modeling the observed

679: %nebular spectrum under various scattering/interaction circumstances

680: %may provide a better understanding of the production of this

681: %emission feature.

682: 

683: The presence of a local echo helps explain the slow decline of the

684: $B$-band light curve of SN 2006X at early phases. Nevertheless, the

685: local echo (if present) is not necessarily from the CSM dust, as

686: forward scattering from the distant dust cloud in front of the SN

687: could also produce an echo of very small angular size. To further

688: distinguish between the two possible cases of distant ISM plus local

689: dust and single ISM dust, future {\it HST} observations of SN 2006X

690: are necessary. More late-phase {\it HST}/ACS images would help

691: constrain the evolution of the LE. Using equation (1), we can

692: predict the evolution of the echo ring with time. If the dust formed

693: as a result of past mass loss from the central source, the echo will

694: be more symmetric and the expansion will slow down after the

695: initially rapid phase; with time, its size will eventually shrink to

696: zero. On the other hand, if the dust is of interstellar origin, the

697: echo should expand continuously with slowly decreasing brightness as

698: more-distant regions are illuminated. Assuming that the inner

699: component of the echo within 2 pixels is caused by a CSM dust shell

700: $\sim$1~pc from the SN, then the emission within 2 pixels will

701: finally decrease to zero at $t \approx 6.5$~yr. In contrast, the

702: local echo from distant ISM dust should remain nearly constant for a

703: longer time.

704: 

705: It is worth pointing out that the recent nearby SNe~Ia, SN 2007gi

706: (CBET 1017, CBET 1021), SN 2007le (CBET 1100, CBET 1101), and

707: probably SN 2007sr (CBET 1172,1174, ATEL 1343), may exhibit

708: high-velocity features in their spectra similar to those of SN

709: 2006X. If the SN 2006X-like events preferentially occur in

710: environments with abundant ISM dust or CSM dust (Wang et~al. 2008,

711: in prep.), then we might expect to detect late-time echo emission in

712: the above three SNe~Ia. Thus, it would be interesting to obtain

713: future high-resolution {\it HST}/ACS images of these SNe.

714: 

715: \section{Conclusions}

716: The emergence of a LE in SN 2006X has been confirmed with

717: PSF-subtracted {\it HST} ACS images which show a ring-like, but

718: rather extended, echo 2--5 pixels ($0.05''$--$0.13''$) from the SN

719: site at $t = 308$~d past maximum brightness. A Keck nebular spectrum

720: of the SN taken at a similar phase provides additional evidence for

721: the LE scenario; it can be decomposed into a nebular spectrum of a

722: normal SN~Ia and a reflection spectrum consisting of the SN light

723: emitted at early phases.

724: 

725: From the resolved echo image, we derive that the intervening dust is

726: $\sim$27--170~pc from the supernova. Based on the quite blue color

727: of the echo, we suggest that the mean grain size of the scattering

728: dust is substantially smaller than Galactic dust. Smaller dust

729: particles are also consistent with the low $\Re_{V}$ value obtained

730: from the SN photometry. Our detection of a LE in SN 2006X confirms

731: that this SN~Ia occurred in a dusty environment with atypical dust

732: properties, as suggested by the photometry (W07).

733: 

734: Analysis of the nebular spectrum might also suggest a local echo at

735: $<$27 pc (or at $<$2 pixels) that is not resolved in the

736: PSF-subtracted image. This possible local echo is likely associated

737: with the CSM dust produced by the progenitors, though detailed

738: modeling of the echo spectrum and/or further high-resolution imaging

739: are required to test for the other possibilities, such as very

740: forward scattering by a distant cloud or CSM-ejecta interaction.

741: 

742: %In the case of a cool RG wind, the dust might be the ring or shell

743: %of the planetary nebulae (Wang 2005) which is found to distribute in

744: %patchy clumps (O'Dell et~al. 2004) and have abnormal extinction

745: %towards the center, e.g. $\Re_{V}\sim$ 2.0 (Ruffle 2004). The

746: %type-Ia/IIn SNe 2002ic and 2005gj having strong signature of

747: %CSM-ejecta interaction (Wang et~al. 2004, Kotak et~al. 2004,

748: %Aldering et~al. 2006, Prieto et~al. 2007) may be of the extreme

749: %cases in terms of this nebular model, which probably exploded at the

750: %proto-planetary nebular stage when the dust and gas may still not

751: %disperse into the interstellar medium (Wood-Vasey et~al. 2004).

752: 

753: 

754: \acknowledgments Some of the data presented herein were obtained at

755: the W. M. Keck Observatory, which is operated as a scientific

756: partnership among the California Institute of Technology, the

757: University of California, and the National Aeronautics and Space

758: Administration (NASA). The Observatory was made possible by the

759: generous financial support of the W. M. Keck Foundation. This

760: research was supported by NASA/{\it HST} grants AR--10952 and

761: AR--11248 from the Space Telescope Science Institute, which is

762: operated by the Association of Universities for Research in

763: Astronomy, Inc., under NASA contract NAS5--26555. We also received

764: financial assistance from NSF grant AST--0607485, the TABASGO

765: Foundation, the National Natural Science Foundation of China (grant

766: 10673007), and the Basic Research Funding at Tsinghua University

767: (JCqn2005036).

768: 

769: \begin{thebibliography}{}

770: \bibitem [Aldering et~al. (2006)]{ald06} Aldering, G., et~al. 2006, \apj, 650, 510

771: %\bibitem [Altavilla et~al. (2007)]{alta07} Altavilla, G., et~al. 2007, \aap,

772: %submitted (arXiv:0708.4119)

773: \bibitem [Bloom et~al. (2007)]{bloo07} Bloom, J. S., et~al. 2007, ATEL, 1343

774: \bibitem [Bond et~al. (1990)]{bond90} Bond, H. E., Gilmozzi, R., Meakes, M. G.,

775: \& Panagia, N. 1990, \apj, 354, L49

776: \bibitem [Bond et~al. (2003)]{bond03} Bond, H. E., et~al. 2003, Nature, 422, 405

777: \bibitem [Cappellaro et~al. (2001)]{capp01} Cappellaro, E., et~al. 2001, \apj, 549,

778: L215

779: %\bibitem [Chevalier (1986)]{che86} Chevalier, R. A. 1986, \apj, 308, 225

780: \bibitem [Chevalier \& Emmering (1988)]{che88} Chevalier, R. A., \& Emmering,  R. T.

781: 1988, \apj, 338, 388

782: \bibitem [Couderc (1939)]{coud39} Couderc, P. 1939, Ann. Astrophys., 2, 271

783: \bibitem [Crotts (1988)]{crot88} Crotts, A. 1988, \apj, 333, L51

784: %\bibitem [De Marco et~al. (2004)]{dem04} De Marco, O., Bond, H. E., Harmer, D.,

785: %\& Fleming, A. J. 2004, \apj, 602, L93

786: \bibitem [Crotts et~al. (1989)]{crot89} Crotts, A. P. A., Kunkel, W. E.,

787: \& McCarthy, P. J. 1989, \apj, 347, L61

788: \bibitem [Deng et~al. (2004)]{deng04} Deng, J. S., et~al. 2004, \apj, 605, L37

789: %\bibitem [Draine (2003)]{drain03} Draine, B. T., et~al. 2003, \apj, 598, 1017

790: \bibitem [Dolphin (2000)]{dol00} Dolphin, A. E. 2000, \pasp, 112, 1383

791: \bibitem [Drake et~al. (2007)]{drak07} Drake, A. J., et~al. 2007, CBET, 1172

792: \bibitem [Dwek (1983)]{dwek83} Dwek, E. 1983, \apj, 274, 175

793: %\bibitem [Hachisu \& Kato (2001)]{hk01} Hachisu, I., \& Kato, M. 2001,

794: \apj, 558, 323

795: \bibitem [Hachisu et~al. (2007)]{hach07} Hachisu, I., Kato, M., \&

796: Nomoto, K. 2007, ApJ, submitted (arXiv:0710.0319)

797: \bibitem [Harutyunyan et~al. (2007)]{har07} Harutyunyan, A., Benetti, S., \&

798: Cappellaro, E. 2007, CBET 1021

799: %\bibitem [Henyey \& Greenstein (1941)]{HG41} Henyey, L. C., \& Greenstein, J. L.

800: 1941, \aj, 93, 70

801: \bibitem [Garnavich et~al. (2001)]{gar01} Garnavich, P., et~al. 2001, AAS, 199, 4701

802: \bibitem [Gouiffes et~al. (1988)]{gou88} Gouiffes, C., et~al. 1988, \aap, 198, L9

803: \bibitem [Filippenko et~al. (2007)]{fili07} Filippenko, A. V., Silverman, J. M.,

804: Foley, R. J., \& Modjaz, M. 2007, CBET 1101

805: \bibitem [Freedman et~al. (2001)]{fre01} Freedman, W. L., et~al. 2001, \apj, 553, 47

806: \bibitem [Jha et~al. (2007)]{jha07} Jha, S., Riess, A. G., \& Kirshner, R. P. 2007,

807: \apj, 659, 122

808: \bibitem [Kotak et~al. (2004)]{kota04} Kotak, R., et~al. 2004, \mnras, 354, L13

809: \bibitem [Liu et~al. (2003)]{liu03} Liu, J. F., Bregman, J. N., \& Seitzer, P. 2003,

810: \apj, 582, 919

811: \bibitem [Mathis et~al. (1977)]{math77} Mathis, J. S., Rumpl, W., \& Nordsieck, K.

812: H. 1977, \apj, 217, 425

813: \bibitem [Meikle et~al. (2006)]{meik06} Meikle, W. P. S., et~al. 2006, \apj, 649, 332

814: \bibitem [Monard (2007)]{mon07} Monard, L. A. G. 2007, CBET 1100

815: \bibitem [Nakano (2007)]{nank07} Nakano, S. 2007, CBET 1017

816: %\bibitem [O'Dell et~al. (2004)]{odell04} O'Dell, C. R., McCullough, P. R., \&

817: Meixner, M. 2004, \aj, 128, 2339

818: \bibitem [Patat (2005)]{pat05} Patat, F. 2005, \mnras, 357, 1161

819: %\bibitem [Patat et~al. (2006)]{pat06} Patat, F., Benetti, S., Cappellaro, E., \&

820: %M. 2006, \mnras, 369, 1949

821: \bibitem [Patat et~al. (2007a)]{pat07a} Patat, F., et~al. 2007a, A\&A, 474, 931

822: \bibitem [Patat et~al. (2007b)]{pat07b} Patat, F. 2007b, Science, 317, 924 (P07)

823: \bibitem [Prieto et~al. (2007)]{prieto07} Prieto, J. L., et~al. 2007, ApJ,

824: submitted (arXiv:0706.4088)

825: \bibitem [Quinn et~al. (2006)]{quinn06} Quinn, J. L., et~al. 2006, \apj, 652, 512

826: %\bibitem [Ruffle et~al. (2004)]{ruff04} Ruffle, P. M. E., et~al. 2004, \mnras, 353,

827: \bibitem [Salvo et~al. (2001)]{sal01} Salvo, M. E., et~al. 2001, \mnras, 321, 254

828: %\bibitem [Schaefer (1987)]{sch87} Schaefer, B. E. 1987, \apj, 323, L47

829: \bibitem [Schlegel et~al. (1998)]{sfd98} Schlegel, D. J., Finkbeiner, D. P., \&

830: Davis, M. 1998, \apj, 500, 525

831: \bibitem [Schaefer (1987)]{sch87} Schaefer, B. E. 1987, \apj, 323, L47

832: \bibitem [Schmidt et~al. (1994)]{sch94} Schmidt, B. P., et~al. 1994, \apj, 434, L19

833: \bibitem [Sirianni et~al. (2005)]{siria05} Sirianni, M., et~al. 2005, \pasp, 117, 1049

834: \bibitem [Simon et~al. (2007)]{sim07} Simon, J. D., et~al. 2007, \apj, 671, L25

835: \bibitem [Suzuki \&  Migliardi]{SM06} Suzuki, S., \& Migliardi, M.  2006, IAUC 8667

836: \bibitem [Sparks et~al. (1999)]{sparks99} Sparks, W. B., Macchetto, F., Panagia, N.,

837: Boffi, F. R., Branch, D., Hazen, M. L., \& Della Valle, M. 1999,

838: \apj, 523, 585

839: %\bibitem [Spyromilio et~al.(2004)]{spy04} Spyromilio, J., Gilmozzi R., Sollerman

840: %Fransson C., Cuby J. G. 2004, \aap, 426, 547

841: \bibitem [Suntzeff et~al. (1988)]{suntz88} Suntzeff, N. B., et~al. 1988, Nature,

842: 334, 135

843: \bibitem [Sugerman \& Crotts (2003a)]{sc03a} Sugerman, B. E. K., \& Crotts A. P. S.

844: 2003a, \apj, 632, L17

845: \bibitem [Sugerman (2003b)]{sc03b} Sugerman, B. E. K. 2003b, \aj, 126, 1939

846: \bibitem [Sugerman (2005)]{sug05} Sugerman, B. E. K. 2005, \apj, 632, L17

847: \bibitem [Tylenda (2004)]{tylen04} Tylenda, R. 2004, \aap, 414, 223

848: \bibitem [Umbriaco et~al. (2007)]{umb07} Umbriaco, G., et~al. 2007, CBET, 1174

849: \bibitem [Van Dyk et~al. (2006)]{Van06} Van Dyk, S. D., Li, W., \& Filippenko, A. V.

850: 2006, \pasp, 118, 351

851: \bibitem [Wang L et~al. (2004)]{wan04} Wang, L. F., et~al. 2004, \apj, 604, L53

852: \bibitem [Wang L (2005)]{wan05} Wang, L. F. 2005, \apj,  635, L33

853: \bibitem [Wang et~al. (2006)]{wang06} Wang, X. F., et~al. 2006, \apj, 645, 488

854: \bibitem [Wang et~al. (2007)]{wang07} Wang, X. F., et~al. 2007, \apj, in press

855: (arXiv:0708.0140)(W07)

856: \bibitem [Welch et~al. (2007)]{welch07} Welch, D. L., et~al. 2007, \apj, 669, 525

857: %\bibitem [White (1979)]{white79} White, R. L. 1979, \apj, 229, 54

858: \bibitem [Wood-Vasey et~al. (2004)]{wood04} Wood-Vasey, W. M., et~al. 2004, \apj,

859: 616, 339

860: \bibitem [Xu et~al. (1995)]{xu95} Xu, J., et~al. 1994, \apj, 435, 274

861: \end{thebibliography}

862: 

863: \clearpage

864: %%%%%%%%%%%%

865: %%% Table 1.

866: %%%%%%%%%%%%

867: \begin{table}

868: \begin{center}

869: \caption{Late-Time {\it HST} Photometry of SN 2006X.$^a$} {\small

870: \begin{tabular}{lllccc}

871: \tableline\tableline

872: UT Date& JD$-$2,450,000 & Phase (d) &F435W & F555W & F775W \\

873: \tableline

874: 05/21/2006&3876.0 & +90.0 &18.71(04) &17.36(02) &16.46(02) \\

875: 12/25/2006&4094.0 & +308.0&21.48(06) &20.56(09) &19.69(02)\\

876: \tableline

877: \end{tabular}}

878: \tablenotetext{a}{Uncertainties in hundredths of a magnitude are

879: given in parentheses.}

880: \end{center}

881: \end{table}

882: 

883: 

884: %%%%%%%%%%%%

885: %%% Table 2.

886: %%%%%%%%%%%%

887: \begin{table}

888: \begin{center}

889: \caption{FWHM of SN 2006X and Local Stars in {\it HST} Images.}

890: {\small

891: \begin{tabular}{lllllc}

892: \tableline \tableline

893:  Object  &      $r$(pixel)    &     $j$(pixel)    &   $k$(pixel)   & bandpass\\

894: \tableline

895: & &$t = 90$~d & & \\

896: \tableline

897:  SN      &    2.21            & 2.61      &  2.51       & F435W  \\

898:  star    &    2.17$\pm$0.02   & 2.60$\pm$0.03 & 2.60$\pm$0.05 & F435W \\

899:  SN      &    2.38            & 2.70       & 2.68       & F555W \\

900:  star    &    2.39$\pm$0.05   & 2.81$\pm$0.04 &2.78$\pm$0.03& F555W \\

901:  SN      &    2.79            & 2.84       &2.85        &F775W  \\

902:  star    &    2.82$\pm$0.05   & 2.89$\pm$0.08& 2.89$\pm$0.03&F775W\\

903: \tableline

904: & &$t = 308$~d & & \\

905: \tableline

906: SN  &        2.50          &     3.85         & 3.36 &  F435W \\

907: star&        2.21$\pm$0.03 &     2.81$\pm$0.03& 2.51$\pm$0.02 & F435W \\

908: SN  &        2.62          &     4.00         & 3.69  & F555W \\

909: star&        2.33$\pm$0.03 &     2.74$\pm$0.02& 2.64$\pm$0.05&F555W\\

910: SN  &        2.95          &     3.37         & 3.33  & F775W \\

911: star&        2.79$\pm$0.01 &     2.94$\pm$0.02 &2.83$\pm$0.02 & F775W \\

912: \tableline

913: \end{tabular}}

914: \end{center}

915: \end{table}

916: 

917: %%%%%%%%%%%%

918: %%% Table 3.

919: %%%%%%%%%%%%

920: \begin{table}

921: \begin{center}

922: \caption{Light Echo of SN 2006X at t = 308~d} {\small

923: \begin{tabular}{lclc}

924: \tableline\tableline

925: Method & F435W (mag) & F555W (mag) & F775W (mag)\\

926: \tableline

927: Residual radial profile (2--10 pixels) &22.8$\pm$0.1  & 22.0$\pm$0.3 & 22.1$\pm$0.7 \\

928: PSF-subtracted image (2--10 pixels)    &22.7$\pm$0.1  & 21.9$\pm$0.3 & 22.3$\pm$0.9 \\

929: Synthetic echo spectrum (SN 1996X) &22.2$\pm$0.3  & 21.4$\pm$0.3 & \nodata  \\

930: Synthetic echo spectrum (SN 2006X) &22.1$\pm$0.3&21.5$\pm$0.3 & \nodata\\

931: \tableline

932: \end{tabular}}

933: \end{center}

934: \end{table}

935: 

936: \clearpage

937: \begin{figure}

938: \figurenum{1} \hspace{-0.2cm} {\plotone{f1.eps}} \hspace{-0.0cm}

939: %\includegraphics[angle=0,width=90mm]{f1.eps}

940: %\includegraphics[angle=0,width=110mm]{f33.eps}

941: \caption{(Left) An {\it HST} image of SN 2006X in the F555W band,

942: taken on 1993 December 31; the circle marks the position of the SN.

943: (Middle) The same field was imaged on 2006 May 21 (90~d after $B$

944: maximum). (Right) The same field was observed on 2006 December 25

945: (308~d after $B$ maximum). The supernova and the reference stars are

946: labeled.} \label{fig-1}

947: \end{figure}

948: 

949: \begin{figure}[htbp]

950: \figurenum{2} \vspace{-0.5cm} \hspace{-1.0cm} \plotone {f2.eps}

951: %\includegraphics[angle=0,width=100mm]{f2.eps}

952: \caption{Light curve of SN 2006X (symbols) compared with that of

953: SN~Ia 1996X (dashed lines). The open circles represent the $B$, $V$,

954: and $I$ data from Wang et~al. (2007). The filled circles denote the

955: {\it HST} magnitudes that were transformed from the F435W, F555W,

956: and F775W bands (see Table 1) to the $B$, $V$, and $I$ bands

957: (respectively) through an empirical correlation (Sirianni et~al.

958: 2005). Proper extinction corrections have been applied to all of the

959: magnitudes (see text for details).} \label{fig:two} \vspace{-0.5cm}

960: \end{figure}

961: 

962: \begin{figure}[htbp]

963: \figurenum{3}\vspace{-0.5cm}\hspace{-1.5cm}

964: %\includegraphics[angle=0,width=110mm]{f3.eps}

965: \plotone{f3.eps} \vspace{-1.0cm} \caption{Radial brightness profile

966: for SN 2006X in F435W, F555W, and F775W. The black curves show the

967: SN, while the color curves represent local stars in the same field,

968: marked in Figure 1 (S1, red; S2, green; S3, blue; S4, orange). The

969: inset panel shows the residual flux between the SN and the average

970: of the local stars.} \vspace{-0.0cm} \label{fig:three}

971: \end{figure}

972: 

973: 

974: \begin{figure}[htbp]

975: \figurenum{4}\vspace{-0.0cm}\hspace{-0.0cm} \plotone{f4.eps}

976: %\includegraphics[angle=0,width=85mm]{f4.eps}

977: \vspace{0.0cm} \caption{The PSF-subtracted {\it HST}/ACS images of

978: SN 2006X (taken on 2006 December 24) with a $0.53'' \times 0.53''$

979: field surrounding SN 2006X. The supernova is at the center of each

980: frame. Column (1) shows the residual image of SN 2006X obtained by

981: subtracting the local bright Star 1 whose central flux is scaled to

982: that of the supernova; column (2) displays the residual image after

983: resampling from 1 pixel to $8 \times 8$ pixels; and in column (3)

984: there are circles of radius 2, 4, and 6 pixels to guide the eye.}

985: \label{fig:four}

986: \end{figure}

987: 

988: \begin{figure}[htbp]

989: \figurenum{5}\vspace{-0.0cm}\hspace{-0.3cm} \plotone{f5.eps}

990: %\includegraphics[angle=0,width=100mm]{f5.eps}

991: \vspace{-2.0cm} \caption{Top panel: Comparison of the observed

992: spectrum of SN 2006X at 308~d with the composite spectrum containing

993: the nebular emission of SN 1996X and the emission of the echo

994: computed as described in the text. Middle panel: Residual of the

995: spectral flux between SN 2006X and SN 1996X at $t = 308$~d,

996: overlapped with the time-integrated echo spectra. Bottom panel: The

997: fraction of the total light contributed by the echo as a function of

998: wavelength. The fractions inferred from the PSF-subtracted image

999: (open circles) and the spectrophotometry (filled squares) are also

1000: shown with error bars (vertical ones for uncertainties and

1001: horizontal ones for the FWHM of the F435W and F555W filters).}

1002: \label{fig:five}

1003: \end{figure}

1004: 

1005: \end{document}

1006: