1: %\documentstyle[12pt,aasms4]{article}
2: %\documentclass{aastex}
3: \documentclass[12pt,preprint]{aastex}
4: %\documentclass[preprint2]{aastex}
5:
6: %\documentclass{emulateapj}
7:
8: %\input epsf.sty
9:
10: \newcommand{\etal}{et~al.\ }
11: \newcommand{\eg}{e.g.,\ }
12: \newcommand{\ie}{i.e.,\ }
13: \newcommand{\Msun}{M_{\odot}}
14: \newcommand{\Rsun}{R_{\odot}}
15: \newcommand{\Lsun}{L_{\odot}}
16: \newcommand{\kms}{km~s$^{-1}$}
17: \newcommand{\ergs}{erg s$^{-1}$}
18: \newcommand{\Ha}{H$\alpha$}
19: \newcommand{\Hb}{H$\beta$}
20: \newcommand{\HI}{H~{\sc i}}
21: \newcommand{\HeI}{He~{\sc i}}
22: \newcommand{\OI}{O~{\sc i}}
23: \newcommand{\OII}{O~{\sc ii}}
24: \newcommand{\CII}{C~{\sc ii}}
25: \newcommand{\NaI}{Na~{\sc i}}
26: \newcommand{\MgII}{Mg~{\sc ii}}
27: \newcommand{\MgI}{Mg~{\sc i}}
28: \newcommand{\SiI}{Si~{\sc i}}
29: \newcommand{\SiII}{Si~{\sc ii}}
30: \newcommand{\SiIII}{Si~{\sc iii}}
31: \newcommand{\SI}{S~{\sc i}}
32: \newcommand{\CaII}{Ca~{\sc ii}}
33: \newcommand{\TiII}{Ti~{\sc ii}}
34: \newcommand{\CrII}{Cr~{\sc ii}}
35: \newcommand{\FeI}{Fe~{\sc i}}
36: \newcommand{\FeII}{Fe~{\sc ii}}
37: \newcommand{\FeIII}{Fe~{\sc iii}}
38: \newcommand{\CoII}{Co~{\sc ii}}
39: \newcommand{\CoIII}{Co~{\sc iii}}
40: \newcommand{\NiII}{Ni~{\sc ii}}
41: \newcommand{\Fefs}{$^{56}$Fe}
42: \newcommand{\Cofs}{$^{56}$Co}
43: \newcommand{\Nifs}{$^{56}$Ni}
44: \newcommand{\Mej}{$M_{\rm ej}$}
45: \newcommand{\KE}{$E_{\rm kin}$}
46:
47:
48: \begin{document}
49:
50: \title{The Aspherical Properties of the Energetic Type Ic SN\,2002ap
51: as Inferred from its Nebular Spectra$^1$}
52:
53:
54: \author{P. A. Mazzali\altaffilmark{2,3,4,5},
55: K.~S.~Kawabata\altaffilmark{6},
56: K.~Maeda\altaffilmark{2,7},
57: R.~J.~Foley\altaffilmark{8},
58: K.~Nomoto\altaffilmark{4,5},
59: J.~Deng\altaffilmark{9},
60: T.~Suzuki\altaffilmark{4},
61: M.~Iye\altaffilmark{10,11},
62: N.~Kashikawa\altaffilmark{10},
63: Y.~Ohyama\altaffilmark{12},
64: A.~V. Filippenko\altaffilmark{8},
65: Y.~Qiu\altaffilmark{9}, and
66: J.~Wei\altaffilmark{9}}
67:
68: \altaffiltext{1}{Based in part on data obtained at the Subaru Telescope,
69: which is operated by the National Astronomical Observatory of
70: Japan (NAOJ). Also based in part on data obtained at the University
71: of California's Lick Observatory.}
72: \altaffiltext{2}{Max-Planck Institut f\"ur Astrophysik,
73: Karl-Schwarzschild-Str.\ 1, 85748 Garching, Germany.}
74: \altaffiltext{3}{National Institute for Astrophysics--OATs, Via G.B. Tiepolo, 11,
75: 34143 Trieste, Italy.}
76: \altaffiltext{4}{Department of Astronomy, School of Science,
77: University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan.}
78: \altaffiltext{5}{Research Center for the Early Universe, School of Science,
79: University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan.}
80: \altaffiltext{6}{Hiroshima Astrophysical Science Center,
81: Hiroshima University, Hiroshima 739-8526, Japan.}
82: \altaffiltext{7}{Department of Earth Science and Astronomy,
83: College of Arts and Sciences, University of Tokyo, Meguro-ku,
84: Tokyo 153-8902, Japan.}
85: \altaffiltext{8}{Department of Astronomy, University of California, Berkeley,
86: CA 94720-3411.}
87: \altaffiltext{9}{National Astronomical Observatories, CAS, 20A Datun Road,
88: Chaoyang District, Beijing 100012, China.}
89: \altaffiltext{10}{National Astronomical Observatory of Japan, Mitaka, Tokyo
90: 181-8588, Japan.}
91: \altaffiltext{11}{Department of Astronomical Science, Graduate University for
92: Advanced Studies, Mitaka, Tokyo 181-8588, Japan.}
93: \altaffiltext{12}{Subaru Telescope, National Astronomical Observatory of Japan,
94: 650 North A'ohoku Place, Hilo, HW 96720.}
95:
96: \begin{abstract}
97:
98: The nebular spectra of the broad-lined, SN\,1998bw-like Type Ic SN\,2002ap are
99: studied by means of synthetic spectra. Two different modelling techniques are
100: employed. In one technique, the SN ejecta are treated as a single zone, while in
101: the other a density and abundance distribution in velocity is used from an
102: explosion model. In both cases, heating caused by $\gamma$-ray and positron
103: deposition is computed (in the latter case using a Monte Carlo technique to
104: describe the propagation of $\gamma$-rays and positrons), as is cooling via
105: forbidden-line emission. The results are compared, and although general agreement
106: is found, the stratified models are shown to reproduce the observed line profiles
107: much more accurately than the single-zone model. The explosion produced $\sim
108: 0.1\, \Msun$ of \Nifs. The distribution in velocity of the various elements is in
109: agreement with that obtained from the early-time models, which indicated an
110: ejected mass of $\sim 2.5\, \Msun$ with a kinetic energy of $4 \times
111: 10^{51}$\,erg. Nebular spectroscopy confirms that most of the ejected mass ($\sim
112: 1.2\, \Msun$) was oxygen. The presence of an oxygen-rich inner core, combined with
113: that of \Nifs\ at high velocities as deduced from early-time models, suggests that
114: the explosion was asymmetric, especially in the inner part.
115:
116: \end{abstract}
117:
118:
119: \keywords{supernovae: general ---supernovae: individual (SN\,2002ap) ---
120: nucleosynthesis --- gamma rays: bursts }
121:
122:
123: \section{Introduction}
124:
125:
126: One of the most exciting recent discoveries in astrophysics is that the
127: nearest observed long-duration gamma-ray bursts (GRBs) are associated with
128: supernovae (SNe) (\eg GRB980425/SN\,1998bw, \citealt{gal98};
129: GRB030329/SN\,2003dh, \citealt{sta03}; GRB031203/SN\,2003lw, \citealt{mal04}).
130: Adding to the excitement, it turned out that these SNe are not at all ordinary.
131: Spectroscopically, they are of Type Ic (no H, no He, weak Si lines; see
132: \citealt{fil97} for a review of SN classification), but their distinguishing
133: feature is that unlike normal SNe~Ic, they show extremely broad absorption
134: lines dominated by Fe, Ca, and O. Very high expansion velocities, reaching $0.1
135: c$, are easily observed at early times.
136:
137: Broad lines suggest a very energetic explosion, or at least a large amount of
138: energy per unit mass. Estimates of the explosion kinetic energy assuming
139: spherical symmetry for SNe 1998bw, 2003dh, and 2003lw are \KE\,$\approx (3 - 6)
140: \times 10^{52}$\,erg (\eg \citealt{iwa98,nak01,maz03, den04,maz06a}). This is
141: more than one order of magnitude larger than in typical core-collapse SNe,
142: which have \KE\,$\approx 10^{51}$\,erg. Accordingly, these exceptionally
143: powerful SNe have sometimes been called ``hypernovae'' (\eg \citealt{nom05}).
144: In the presence of significant deviations from spherical symmetry, these
145: estimates may be reduced, but even then the energies remain large (\eg
146: \citealt{mae06}). The mechanism behind these events is thought to be the
147: collapse of the stripped core of a massive star ($\sim 25 - 60\, \Msun$) to a
148: black hole \citep{mac99}.
149:
150: Additionally, evidence has been obtained that X-ray flashes (XRFs), the soft
151: analogues of GRBs \citep{heise01}, are also linked to SNe
152: \citep[e.g.,][]{fyn03,mod06,pian06}. The SNe associated with XRFs also seem to
153: be overenergetic, but not as much so as GRB-SNe \citep{tom03,maz06b}. They have
154: been suggested to be the result of the collapse of the stripped carbon-oxygen
155: core of stars of originally $\sim 20\, \Msun$ to a neutron star
156: \citep{maz06b,mae07}. If the neutron star is born highly magnetized and rapidly
157: spinning --- a ``magnetar'' \citep{dun92,tho04} --- it may cause the explosion
158: to be overenergetic and produce the XRF \citep{maz06b,mae07}.
159:
160: One difference between SNe in GRBs and XRFs may be the degree of asphericity.
161: GRB-SNe are thought to be highly aspherical, as GRBs are generally believed to
162: be highly asymmetric phenomena (see \citealt{woo06} for a recent review of the
163: SN-GRB connection). For XRF-SNe, the degree of asphericity may be smaller
164: \citep{maz07}.
165:
166: Evidence for asphericity is not easy to glean from early-time data. The
167: significant mixing outward of \Nifs\ required to reproduce the early rise of the
168: light curve is a general feature of hypernovae \citep{mae03} and suggests an
169: aspherical explosion, as does the connection of these SNe with GRBs. The best time
170: to look for signatures of asphericity, however, is starting a few months after the
171: explosion, when the SN becomes nebular, exposing the deepest parts of the ejecta.
172: In that phase, both the spectrum and the light curve have a characteristic
173: behavior. The light curve of SN\,1998bw showed a phase of strictly exponential
174: decline, but at a rate steeper than that of \Cofs\ \citep{pat01}. In this phase
175: the luminosity is larger than what is predicted by spherically symmetric models
176: that fit the peak of the light curve \citep{nak01}. \citet{mae03} showed that this
177: behavior, which is actually not unusual in SNe~Ic, can be explained using a
178: modified spherically symmetric model where a relatively massive but slowly
179: expanding inner core is placed in the center of the expanding ejecta. Although
180: this is technically still a spherically symmetric model, one-dimensional explosion
181: simulations do not predict such a density distribution, which suggests that we are
182: observing an aspherical explosion, similar perhaps to the collapsar model of
183: \citet{mac99}.
184:
185: The nebular spectra provide more direct evidence. In SN\,1998bw the [\FeII] lines
186: are broader than the [\OI] $\lambda\lambda$6300, 6363 doublet \citep{maz01}. This
187: is also something that cannot be explained in the context of a spherically
188: symmetric model of the collapse and explosion of a CO core, as in such a model
189: \Nifs\ (which decays into \Cofs\, and then into \Fefs) is synthesized near the
190: compact remnant and therefore is always located at smaller velocities than O,
191: which is left from the progenitor in the unprocessed outer layers. \citet{mae02}
192: showed that this configuration can be obtained in an axisymmetric explosion. In
193: such an explosion, most of the \KE\ is released along the ``jet'' axis, which is
194: probably linked to the launching of the GRB, and \Nifs\ is synthesized along that
195: direction and ejected at a high velocity. Away from the jet axis, however, less
196: kinetic energy is deposited and burning proceeds much less efficiently. Therefore,
197: in these directions large amounts of unburned O are ejected at low velocity. For a
198: near-polar viewing angle, this scenario naturally leads to broad [\FeII] lines and
199: a narrow [\OI] line at late times. For an equatorial view, on the other hand, the
200: [\OI] line should have a characteristic double-peaked profile \citep{mae02}, as
201: was indeed observed in the SN~Ic 2003jd \citep{maz05}. Further evidence that
202: SNe~Ic are aspherical also comes from their high degree of polarization
203: \citep{wang01,leo02,fil04,leo05}.
204:
205: For XRF/SNe, the case for asphericity is much weaker. The early-time spectra of
206: SN\,2006aj did not show very broad features \citep{pian06}. Radio observations
207: \citep{sod06} suggest a very broad opening angle ($> 60^{\circ}$). The broad
208: nebular lines are not inconsistent with spherically symmetric ejecta
209: \citep{maz07}. Nevertheless, the presence of emission at velocities below
210: 2000\,\kms\ is at odds with the prediction of one-dimensional explosion models,
211: that a density hole is present at the lowest velocities \citep{maz07,mae07}.
212:
213: Besides those clearly associated with GRBs or XRFs, other energetic SNe Ic have
214: been observed. Energetic SNe~Ic, with or without a GRB, are recognized from the
215: extreme width (up to $\sim 0.1 c$) of their spectral lines at early times, and
216: have sometimes also been called hypernovae (although their total energy is not
217: always much higher than that of normal SNe~Ic). The nearest such object ever
218: observed was SN\,2002ap in M74. SN\,2002ap was immediately recognized as a
219: broad-lined event, similar spectroscopically to SN\,1998bw \citep{kinu02}, and
220: was therefore intensively observed. SN\,2002ap remained much less luminous
221: than SN\,1998bw \citep{maz02,galyam02}, and it was not significantly more
222: luminous than the average SN~Ic. Also, SN\,2002ap was not seen in association
223: with a GRB \citep{hurley02,galyam02}.
224:
225: \citet{maz02} modelled the early-time light curve and spectral evolution of
226: SN\,2002ap and derived values of the ejected mass (\Mej\,$= 2.5-5\, \Msun$), the
227: kinetic energy of the explosion (\KE\,$= 4-10 \times 10^{51}$\,erg), and the
228: mass of \Nifs\ synthesised in the explosion ($M$(\Nifs)\,$=0.07\Msun$). While
229: the mass of \Nifs\ is similar to that of ``normal'' SNe~Ic \citep{sauer06}, and
230: much smaller than that of hypernovae linked to GRBs, both \Mej\ and \KE\ are
231: intermediate between those of normal SNe~Ib/c and GRB-SNe. The uncertainty in
232: the values of \Mej\ and \KE\ follows from assuming either the absence of any He
233: envelope (lower bound) or the presence of a maximal, $2.5 \Msun$ helium envelope,
234: the presence of which would not affect the light curve significantly but is at
235: the same time not supported by the spectral appearance of SN\,2002ap.
236: Interestingly, the light-curve behavior at advanced phases was very similar to
237: that of SN\,1998bw \citep{tom05}, suggesting a common nature for the two
238: events.
239:
240: \citet{maz02} argued that SN\,2002ap was the $\sim 5~\Msun$ carbon-oxygen core
241: of a star of initially $\sim 25~\Msun$ that collapsed to a black hole. On the
242: other hand, no GRB was detected, only weak radio emission was detected from the
243: SN \citep{berger02}, and the X-ray signal was very weak \citep{soria03},
244: suggesting that little or no relativistic ejecta were produced. This may be
245: related to the relatively small mass of the collapsing star, which may have
246: been close to the minimum required to form a black hole. It is therefore
247: interesting to explore further the nature of this object. As in the case of
248: SNe\,1998bw and SN\,2006aj, this can be done by modelling the nebular spectra.
249:
250: \citet{fol03} published a series of late-time spectra of SN\,2002ap, while
251: \citet{kaw02} published a single spectrum. The nebular phase in SN\,2002ap
252: started to develop rather early, as in SN\,1998bw but unlike other hypernovae
253: such as SN\,1997ef \citep{maz04}. At an age of $\sim 4$ months, the SN spectra
254: were already fully nebular \citep{fol03}. The optical spectra were dominated by
255: very strong [\OI] $\lambda\lambda$6300, 6363 emission, and showed strong \CaII\
256: emission lines, similar to SN\,1998bw. Unlike the case of SN\,1998bw, however, the
257: [\FeII]-dominated blend near 5200~\AA\ is rather weak. This is not surprising,
258: since SN\,1998bw is thought to have produced $\sim 4$ times as much \Nifs\ as
259: SN\,2002ap \citep{maz06b}. Still, the [\FeII] lines in SN\,2002ap are
260: sufficiently pronounced that modelling can be meaningfully attempted. One
261: peculiarity of SN\,2002ap is the great strength of the \MgI] $\lambda\lambda$4571
262: line, which may suggest that the envelope of SN\,2002ap was more thoroughly
263: stripped than that of SN\,1998bw. This would favor the lower bound of the mass
264: and energy estimates of \citet{maz02}, which were obtained assuming the absence
265: of any significant He layer.
266:
267: \section{The Spectral Dataset}
268:
269: In this paper, we model all the spectra published by \citet{fol03}.
270: Additionally, we present and model three additional late-time spectra.
271: Two of these spectra were obtained with the Subaru telescope.
272: The other spectrum was obtained at the National Astronomical Observatory of China
273: (NAOC, formerly Beijing Astronomical Observatory, BAO).
274:
275: The Subaru spectra were obtained on 2002 Jun 7.6 (UT dates are used
276: throughout this paper) and Sep 15.6 with FOCAS.
277: For both spectra, the exposure time was 240 s, and grism B300 and Y47 order-cut
278: filters were used. A $0\farcs 8$ wide slit was used for the June spectrum and
279: the resulting spectral resolution ($\lambda/\Delta\lambda$) is $\sim 650$. For
280: the September spectrum, a $2\farcs 0$ slit was used, but the spectral resolution
281: is still similar to that of the June spectrum because the seeing size was
282: $0\farcs 7$--$0\farcs 8$. The flux was calibrated using observations of
283: either Feige 110 or G191B2B obtained on the same night as the target spectra.
284:
285: The BAO spectrum was obtained on 2002 July 11.8 with the NAOC
286: 2.16~m telescope at Xinlong Observatory (then BAO).
287: The observations were carried out with an OMR (Optomechanics Research, Inc.)
288: spectrograph, using a Tektronix $1024\times 1024$ pixel CCD as the detector.
289: A grating of 300 g mm$^{-1}$ was used, which provided a spectral resolution
290: of $\sim 10-11$~\AA. The exposure time was $\sim 40$ min.
291:
292: The newly published spectra are shown in Figure 1.
293: The NAOC spectrum is reasonably consistent with the almost contemporaneous Lick
294: spectrum published by \citet{fol03}, and therefore it was modelled but is not
295: explicitly shown in the Figures in the following sections.
296:
297: The 11 spectra we used for modelling cover a time span of almost 9 months, from
298: June 2002 through February 2003, corresponding to SN ages of 4--13 months given
299: that the SN exploded on 29 January 2002 \citep{maz02}. Unfortunately, the
300: wavelength coverage is not uniform. In particular, the Subaru spectra extend over
301: a shorter range (4700--9000\,\AA) than the other spectra, missing the important
302: \MgI] line. Models based on the Subaru spectra are therefore somewhat less
303: reliable than those based on the Lick spectra. All spectra were
304: calibrated by using the available photometry
305: \citep{fol03} with the exception of the last three, for which an extrapolation of
306: the light curve was used.
307:
308:
309: \section{Modelling Technique}
310:
311: In order to model the nebular spectra of SN\,2002ap, we used our non-local
312: thermodynamic equilibrium (non-LTE) code \citep{maz01}. The code computes the
313: heating of the gas following the deposition of $\gamma$-rays and positrons emitted
314: by the decay of \Cofs\ into \Fefs. Heating is balanced by cooling via nebular
315: line emission. The emission rate in each line is computed solving a non-LTE matrix
316: of level populations \citep{axe80}. In the original version \citep{RLL92}, a
317: homologously expanding nebula of finite extent, uniform density, and uniform
318: composition is assumed, and the emission spectrum is obtained assigning to all
319: lines a parabolic profile, bounded by the velocity of the outer edge of the
320: nebula.
321:
322: Together with this classical version, a more advanced --- although more
323: model-dependent --- version has been developed and is used in this work. In
324: this new version, stratification in density and abundance is adopted. The
325: density profile is taken from explosion models, and $\gamma$-rays and positrons
326: are emitted at various depths according to the distribution of \Nifs. Their
327: propagation and deposition is followed using a Monte Carlo scheme similar to
328: that discussed by \citet{cap97} for their light-curve models. A constant
329: $\gamma$-ray opacity ($\kappa_{\gamma} = 0.027$\,cm$^2$\,g$^{-1}$) and a
330: constant positron opacity ($\kappa_{e^+} = 7$\,cm$^2$\,g$^{-1}$) are assumed.
331: The heating and cooling of the gas are then computed in non-LTE in each radial
332: shell, and so is the line emissivity. The line profiles from each shell are
333: assumed to be truncated parabolas, the inner truncation point corresponding to
334: the inner boundary of the shell considered. These truncated parabolas are then
335: summed to produce the emerging spectrum. Line profiles thus depend on the
336: density and abundance distributions. Therefore, this approach constitutes a
337: test of the explosion models which are used to simulate the light curves and
338: the early-time spectra, especially for what concerns the innermost part, which
339: is only visible directly in the nebular phase.
340:
341: Because the stratified code depends more directly on the adopted explosion
342: model than the one-zone version, it is useful to compare the results of the
343: two approaches. The next two sections deal with both cases in turn.
344:
345: One important ingredient for the modelling is the assumed distance and reddening
346: to the SN. In the case of SN\,2002ap, the distance is highly uncertain, as
347: discussed by \citet{vinko04}. \citet{sha96}, using the brightest blue
348: supergiant stars, obtained a distance modulus of $\mu = 29.32$ mag
349: for M74, and $\mu = 29.50$ mag for the M74 group, confirming previous
350: results by \citet{sohn96}, who preferred to use red supergiants.
351: As \citet{vinko04} comment, this value was
352: derived using the Galactic absorption maps of \citet{burstein82}, which give a
353: value of the total absorption in the $B$ band of $A_B = 0.13$ mag. On the other
354: hand, the more recent maps published by \citet{schlegel98} predict a larger
355: absorption, $A_B = 0.301$ mag, leading to a reduced distance, $\mu = 29.15$ mag.
356:
357: As for the reddening, combining the Galactic extinction of \citet{schlegel98},
358: $E(B-V)_G = 0.075$ mag, with the small reddening within M74
359: \citep[$E(B-V)_H = 0.020$ mag]{tak02}, a total value of $E(B-V) = 0.09$ mag
360: is used.
361:
362: Clearly, the uncertainty associated with the distance is the largest
363: contribution to the overall uncertainty on the luminosity of SN\,2002ap.
364: Therefore, in this paper we have chosen to adopt the same values used in previous
365: modelling and analysis papers
366: \citep[($\mu = 29.50$ mag, $E(B-V) = 0.09$ mag)]{maz02,yosh03} in order to make the
367: results immediately comparable. The effect of adopting a different value for the
368: distance is discussed below.
369:
370:
371: \section{One-Zone Models}
372:
373: The driving parameters for the one-zone fits are the line width, the mass of
374: \Nifs, and the masses of the other elements. Fitting the widths of the complex
375: [\FeII] blend near 5200~\AA\ gives a measure of the distribution of \Nifs, since
376: most Fe is the product of the decay of \Nifs. This blend can be fitted for a
377: nebular velocity of between 6200 and 5200 \kms. The line width slowly decreases
378: over the period considered, which is in good agreement with what was found for
379: SN\,1998bw \citep{maz01}, although in that case the velocities were significantly
380: larger. This behavior indicates that the outer parts of the ejecta are becoming
381: progressively more difficult to excite as the density decreases owing to the
382: expansion. Unlike the case of SN\,1998bw, however, the velocity that is required
383: to fit the [\FeII] lines is also appropriate for the [\OI] line. This is typical
384: of lower-energy SNe~Ic \citep[\eg SN\,1994I,][]{sauer06}, and indicates that the
385: degree of asphericity of the explosion is small. The only apparent deviation from
386: sphericity is the presence of narrow emission spikes in the cores of [\OI]
387: $\lambda\lambda$6300, 6363 and \MgI] $\lambda$4571, at velocities below
388: 2000\,\kms \citep[][Figure 17]{fol03}. Such a feature was also observed in
389: SN\,2006aj \citep{maz07}.
390:
391: The masses of the other elements are determined by fitting the various emission
392: lines. One caveat is that two intermediate-mass elements that are expected to be
393: abundant in the SN ejecta are only diagnosed by a single line. The only line of
394: sulfur is [\SI] $\lambda\lambda$4069, which is blended with various Fe lines.
395: Silicon, a highly abundant element, also has no strong or isolated line at optical
396: wavelenghts. A shoulder on the red side of [\OI] $\lambda\lambda$6300, 6363 may be
397: due to [\SiI] $\lambda$6527. We have determined the masses of Si and S by
398: trying to reproduce those two features, but the uncertainty involved is large, affecting
399: also the estimate of the mass in the ejecta. In fact, both silicon and sulfur
400: have strong lines in the infrared ([\SiI] 1.61, 1.65 $\mu$m, [\SI] 1.08, 1.13
401: $\mu$m), which act as efficient coolants. Thus, increasing the mass of these two
402: elements requires all other masses, including that of \Nifs, to be increased.
403: Additionally, since in the later spectra it is not possible to reproduce
404: accurately the red shoulder of the [\OI] line, there may be some doubt as to the
405: actual contribution of silicon to that feature. More accurate statements about the
406: masses of these elements would require the availability of infrared spectra.
407:
408: As in SN\,1998bw, the spectra of SN\,2002ap show lines of [\FeII], but not
409: [\FeIII]. This implies a rather low degree of ionization, which can be
410: reproduced assuming significant clumping to favor recombination. As in
411: \citet{maz01}, a volume filling factor of 0.1 was adopted in all models to
412: achieve this. The models with stratified density and composition discussed in
413: the next section are useful to address this issue.
414:
415: The results are summarized in Table 2, and the time series of the spectral fits
416: is shown in Figures 1 and 2. The electron temperature and density in the nebula
417: decrease with time. The inferred value of $M$(\Nifs) is $\sim 0.10\, \Msun$.
418: Small oscillations around this value in the various fits are probably due to a
419: combination of inconsistent flux calibration and incomplete wavelength
420: coverage, in particular for spectra that do not include the \MgI] line. Only in
421: the two very late-time spectra does the \Nifs\ mass increase significantly, but these
422: spectra are calibrated in flux using an extrapolation of the observed
423: photometry, making the results less reliable.
424:
425: In order to fit the spectra, we
426: require an ejected mass of $\sim 1.6-1.9\, \Msun$ below a velocity of
427: 5500\,\kms. Of this mass, $\sim 0.7\,\Msun$ is oxygen. The density profile
428: (CO100/4) used by \citet{maz02} to fit the early-time light curve and spectra
429: only contained $\sim 0.8\,\Msun$ below the same velocity. Interestingly, the
430: larger mass is in good agreement with the results of \citet{mae03}, who used a
431: two-component, one-dimensional density distribution to reproduce the light
432: curve of SN\,2002ap and obtained a \Nifs\ mass of $0.08\, \Msun$ and an ejected
433: mass of $1.6\, \Msun$ below 5750\,\kms\ (see also \citealt{tom05}). Although
434: the mass at low velocity may be somewhat overestimated by the one-zone models,
435: which overfit the flux at the lowest velocities where the observed [\OI] line
436: has a narrow core, the need for additional mass at low velocity is clear.
437: Similar results were also obtained for other SNe~Ic
438: \citep[e.g.,][]{sauer06,maz07}. They suggest the presence of a dense,
439: oxygen-rich core in the ejecta, which would be most naturally explained as the
440: result of an aspherical explosion.
441:
442: One place where our models fail is the emission near 7700~\AA. This is likely to be
443: O~I $\lambda$7774. The excitation temperature of this line is significantly
444: higher than the nebular temperatures obtained in our models, suggesting that the
445: flux in the line may be caused by recombination or non-thermal excitation by fast
446: electrons.
447:
448: Therefore, while the results from the one-zone models generally confirm
449: findings from the light curve and spectral study of \citet{maz02}, they suggest
450: that the improvements introduced by \citet{mae03} are realistic, and that a
451: simple one-dimensional explosion model is not sufficient to explain the
452: behavior of SN\,2002ap.
453:
454:
455: \section{Multi-Zone Models}
456:
457: We have modelled the same 11 spectra of SN\,2002ap with the multi-zone code
458: described in Section 2. As a model of the explosion, we selected CO100/4, which
459: was used by \citet{maz02} to fit the early-time light curve and spectra of
460: SN\,2002ap.
461:
462: Our multi-zone code is still one-dimensional, but by allowing us to investigate
463: the element distribution in velocity space at an epoch when the ejecta are
464: fully transparent, it can give us indirect clues of possible asymmetries
465: through unusual abundance distributions. Therefore, although we started our
466: models using the density profile and the abundances of CO100/4, we allowed the
467: abundances to vary so as to achieve the best possible fit to the data.
468: Repeating this exercise over several epochs ensures the validity of the
469: results.
470:
471: Also, because the ejecta are assumed to be fully transparent, once the abundance
472: distribution has been defined from fitting one spectrum, it should not need to
473: be changed to reproduce the other epochs. Ideally, this should be done with the
474: earliest nebular spectrum, since this is likely to show emission from a broader
475: range of velocities than later spectra, as discussed in the previous section.
476:
477: Since the first spectrum of our series does not cover the \MgI] $\lambda$4571
478: line, which is an important coolant, we used as reference the Lick Observatory 8
479: June 2002 spectrum, which offers very broad wavelength coverage at a high
480: signal-to-noise ratio. The model used to fit that spectrum gave good results for
481: the other spectra, with only small changes in composition, which may be attributed
482: to the variable observing conditions and flux calibration.
483:
484: The series of our synthetic spectra is shown in Figures 4 and 5. In order to
485: reproduce the width of the [\FeII] lines, the mass fraction of \Nifs\ must be
486: small ($\sim 1$\%) at $9000 < v < 17,000$\,\kms, and then slowly increase inward,
487: reaching $\sim 10$\% at $3000 < v < 6000$\,\kms. At the lowest velocities, below
488: 3000\,\kms, the abundance drops again to $\sim 2$\%. For the highest velocities
489: ($v > 17,000$\,\kms), the value of the abundance derived from the early-time
490: modelling ($\sim 10$\%) was used, but this does not affect the nebular spectra.
491: The oxygen mass fraction, on the other hand, does not change significantly with
492: velocity. It is $\sim 65$\% at $v > 10,000$\,\kms, decreasing to $\sim 40$\% at
493: $2000< v < 7000$\,\kms, a region that is dominated by \Nifs, Si, and S. The
494: abundance of S was determined from the weak [\SI] $\lambda$4069 line, and the Si
495: abundance was assumed to have a constant ratio of 3:1 with respect to that of S.
496: With this assumption, the emission on the red side of the [\OI] line is not fully
497: reproduced, suggesting that [\SiII] is not the only line contributing to it, or
498: that a larger mass, or a larger Si/S ratio should be adopted. Other elements
499: behave as expected. The Mg and Ca abundances also increase inward, while C
500: decreases with O.
501:
502: One problem using model CO100/4 is that the strong, narrow cores of [\OI]
503: $\lambda\lambda$6300, 6363 and \MgI] $\lambda$4571 cannot be reproduced. The
504: characteristic velocities of these cores are in fact $< 2000$\,\kms, but
505: CO100/4 has a density ``hole'' at $v < 3000$\,\kms. Therefore, following
506: \citet{mae03}, we added a dense inner region at $v < 3000$\,\kms. The
507: composition of this region must be dominated by O, with C, Mg, and other
508: intermediate-mass elements also present so that the line cores can be
509: reproduced. However, only little \Fefs\ must be present there, or the [\FeII]
510: feature would be negatively affected. This inner zone contains only $\sim 0.2\,
511: \Msun$ of material, mostly oxygen. The total \Nifs\ mass below 3000\,\kms\ is
512: $\sim 0.01\, \Msun$, in agreement with the result of \citet{mae03}. The oxygen
513: mass below 5750\,\kms\ in these models is $0.35\, \Msun$, which is
514: significantly less than in the one-zone models. The total \Nifs\ mass is,
515: however, slightly larger ($\sim 0.11\, \Msun$). The total ejected mass is
516: \Mej\,$\approx 2.5\, \Msun$, of which $\sim 1.3\, \Msun$ is oxygen.
517:
518: The main results obtained from the multi-zone models are recapped in Table 2.
519: The last two spectra again require a somewhat larger \Nifs\ mass. Because of
520: the uncertain flux calibration of these spectra, we do not regard this as an
521: inconsistency of our modelling procedure. The oxygen mass is, however, much more
522: stable than in the one-zone models, showing that the multi-zone approach is
523: more accurate, and that the explosion model that we used is a good
524: representation of the ejecta of SN\,2002ap in the late phase as well as the early
525: phase.
526:
527: One further point concerns clumping. Introducing a density gradient should
528: reduce the ionization degree at low velocity. Thus, it may be expected that in
529: the stratified models clumping would not be necessary. However, even in these
530: models, if clumping is not used the Fe ionization is too high at all
531: velocities. A filling factor of 0.1 was therefore used in all zones with $v >
532: 2000$\,\kms --- that is, wherever \Nifs\ is present. No clumping is necessary
533: in the O-dominated inner zone, below 2000\,\kms. There are a number of possible
534: explanations for this. One is that clumping does indeed exist. The lack of
535: direct evidence (\eg narrow emission features at different velocities) suggests
536: that any clumps must be small and evenly distributed in velocity space. Another
537: possibility is that \Nifs\ is located mostly near the jet direction in an
538: axisymmetric explosion. In this case, the actual density of Fe in the
539: effectively emitting volume in the nebular phase would be higher than what is
540: suggested by a spherically symmetric approach like the one used here. Clumping
541: would then be a proxy for this aspherical distribution. The total mass estimate
542: would not be affected, since in the nebular phase each emitting ion contributes
543: to the line profile, independent of its position. An axisymmetric
544: configuration was indeed suggested for the outer part of the ejecta as a
545: possible interpretation of the detection of significant line polarization in
546: the early-time spectra of SN\,2002ap \citep{kaw02,leo02}. If significant
547: asymmetry affected the inner part of the ejecta, the profiles of the nebular
548: emission lines would depend sensitively on viewing angle, as demonstrated by
549: \citet{mae02} for SN\,1998bw. However, in this case the lack of a coincident
550: GRB and of dramatic signatures of asphericity in the nebular spectra suggest
551: that any deep asphericity was most likely weak.
552:
553:
554: \section{The Geometry of the Ejecta of SN\,2002ap}
555:
556: We have modelled a series of nebular spectra of the broad-lined SN\,2002ap,
557: using a one-zone non-LTE code and a multi-zone code where $\gamma$-ray
558: deposition and line emission are computed more accurately based on a
559: pre-defined explosion model. Both simulations yield a good reproduction of the
560: [\FeII] lines, but the narrow cores observed in [\OI] $\lambda\lambda$6300,
561: 6363 can only be reproduced assuming the presence of a dense, O-dominated inner
562: zone. In the one-zone synthetic spectra, a mass excess of $\sim 0.7\, \Msun$ is
563: needed to reproduce qualitatively the [\OI] emission line. In the multi-zone
564: model, which reproduces the line profiles much more accurately, the mass excess
565: is smaller, $\sim 0.2 \Msun$, but the presence of oxygen-dominated material
566: moving at low velocity is confirmed. This material is not predicted by
567: one-dimensional explosion models, but its presence was suggested in various
568: hypernovae based on light-curve studies \citep{mae03}.
569:
570: Similar results were also obtained for the normal SN\,Ic 1994I \citep{sauer06} as
571: well as for the XRF/SN 2006aj \citep{maz07}. SN\,1994I shows emission in the [\OI]
572: $\lambda\lambda$6300, 6363 line at velocities below $2000$\,\kms\ \citep{fil95},
573: while the corresponding one-dimensional explosion model predicts the absence of
574: material at low velocities. The rounded profile suggests a smooth distribution of
575: oxygen, which does not by itself indicate an aspherical explosion, but the very
576: presence of material does, and it can be modelled adding $0.2 \Msun$ of
577: oxygen-dominated mass at low velocity. SN\,2006aj shows a peak in [\OI]
578: $\lambda\lambda$6300, 6363 at velocities below 2000\,\kms, although not quite as
579: sharp as in SN\,2002ap (Figure 4). This was modelled for a mass excess of $\sim
580: 0.7 \Msun$ over the one-dimensional explosion model \citep{maz07}. The presence of
581: a dense inner zone in all SNe~Ic suggests that these explosions are intrinsically
582: aspherical, as is also deduced from polarization studies
583: \citep{wang01,leo02,fil04,leo05}.
584:
585: Figure~6 shows a comparison of the [\OI] $\lambda\lambda$6300, 6363 line in
586: SN\,2002ap, and in SNe\,2006aj, 1998bw, and 2003jd (Mazzali et al. 2007, 2001,
587: and 2005, respectively). The line in SN\,2006aj has a full-width at
588: half-maximum intensity (FWHM) of $\sim 8000$\,\kms, while for the other three
589: SNe the FWHM is $\sim 6000$\,\kms. The line profile in SN\,2006aj is not very
590: sharp, indicating that the nebula is to a good approximation spherically
591: symmetric \citep{maz07}. In this case the [\OI] line velocity should correspond
592: to the real expansion velocity. The sharp profile in SN\,1998bw was interpreted
593: as a disk-like distribution of oxygen viewed from a near-polar direction.
594: Although the low observed velocity is partly the result of a projection effect,
595: the actual velocity of the oxygen ($\sim 8000$\,\kms) is smaller than the
596: velocity of Fe \citep{mae02}. The [\OI] line of SN\,2003jd has a width similar
597: to that of SN\,1998bw, but the double-peaked profile suggests that we are
598: viewing the oxygen-rich disk from near its plane \citep{maz05}. The case of
599: SN\,2002ap seems intermediate between those of SNe~1998bw and 2006aj. The line
600: can be separated into two components. A broad base, of width similar to that in
601: SNe 1998bw and 2003jd, is present down to velocities of $\sim 3000$\,\kms\ and
602: is well fitted by a one-zone model. Its profile is reminiscent of that of
603: SN\,2003jd, but the presence of the sharp core makes it impossible to
604: distinguish between a flat-topped profile, characteristic of a shell-like
605: spherical distribution, and a double-peaked profile as in a disk viewed
606: edge-on. This core is narrower than even the corresponding profile of
607: SN\,1998bw.
608:
609: There may be two explanations for this type of profile. One is that in
610: SN\,2002ap the outer part of the oxygen, from $v \approx 3000$ out to $v
611: \approx 8000$\,\kms, has a distribution that is not far from spherical, giving
612: rise to a flat-topped profile. Below $\sim 3000$\,\kms, on the other hand, we
613: may be seeing a situation qualitatively similar to that of SN\,1998bw, with the
614: oxygen distributed in a disk-like region oriented not far from face-on. The
615: size and mass of this disk-like region are, however, much smaller in SN\,2002ap
616: than in SN\,1998bw. The other possibility is that the outer part reflects a
617: disk-like distribution of most of the oxygen viewed not far from its plane,
618: like in SN\,2003jd. The broad component of the profile would then be
619: double-peaked rather than flat-topped, and the inner component would contain
620: more mass, so that its emission could fill the valley between the two peaks.
621: The sharpness of this inner-component emission may again indicate a disk-like
622: distribution viewed face-on, but in this case two main orientations would exist
623: in the ejecta, which may not be easy to explain in terms of the collapse
624: mechanism. Alternatively, the inner region could simply be sharply peaked in
625: density, as reproduced by our multi-zone model, but not grossly aspherical.
626:
627: In both scenarios the lack of a GRB may be justified. If the inner explosion was
628: aspherical, but the asphericity did not propagate to the intermediate part of the
629: ejecta (3000 to $8000$\,\kms), it is unlikely that a beam of material moving at
630: relativistic velocities was produced, yet the overenergetic explosion could have
631: caused enough material to move at high velocities ($\sim 30000$\,\kms), giving
632: rise to the broad absorption features in the early spectra. The distribution of
633: this material may not have been very far from spherical: polarization studies
634: \citep{kaw02,leo02} suggest a degree of asphericity of $\sim 10$\%. On the other hand,
635: if the outer explosion was aspherical, a relativistic jet may have been produced,
636: but it would not have been pointing to us. This latter possibility is, however,
637: not supported by the lack of an X-ray detection \citep{soria03} and the weakness
638: of the radio signal \citep{berger02}, so we tend to favor an explosion that was
639: not far from spherical, except in the innermost part.
640:
641: Both simulations agree that the \Nifs\ mass ejected by SN\,2002ap is $\sim 0.10\,
642: \Msun$, a slightly larger value than that predicted by studies of the early-time light
643: curve \citep{maz02}, but in agreement with the late-time light curve
644: \citep{mae03,tom05}. The multi-zone model confirms that only a small fraction of
645: this ($\sim 0.01\, \Msun$) is located in the innermost zone, a clear indication
646: that the explosion was asymmetric to some degree.
647:
648: Other global values that can be derived rather accurately from a
649: one-dimensional study of the nebular spectra ($M$(\Nifs)\,$\approx 0.11\,
650: \Msun$, \Mej\,$\approx 2.5\, \Msun$) are small compared to those of other
651: broad-lined SNe, confirming previous results based on the early-time light
652: curve and spectra. These values point to a progenitor less massive than for
653: objects such as SN 1998bw or SN 2003dh, although still a massive star ($M
654: \approx 22-25\, \Msun$) that could have produced a black hole. This also
655: suggests that the ``hypernova'' properties of SN\,2002ap were less extreme,
656: including asymmetry and the launching of a GRB.
657:
658: The estimated mass of \Nifs\ would be smaller if we adopted a shorter distance.
659: If the distance modulus was $\mu = 29.15$ mag, the \Nifs\ mass would be reduced by
660: $\sim 15$\%. The mass and energy of the ejecta would also be similarly reduced,
661: but the change is sufficiently small that the shape of the light curve would
662: not be affected.
663:
664: Indirect indications of asymmetry can be obtained in our one-dimensional study,
665: but the details of the exact properties are not. The development of axisymmetric
666: explosion models suitable for SN\,2002ap, and the study of the nebular spectra
667: using three-dimensional spectrum synthesis codes as in \citet{mae02}, are
668: currently planned.
669:
670: \bigskip
671:
672: The work of A.V.F.'s group at UC Berkeley is supported by National
673: Science Foundation grant AST--0607485. We thank the Subaru and Lick
674: Observatory staffs for their assistance with the observations.
675:
676: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
677:
678: \begin{thebibliography}{}
679:
680: \bibitem[Axelrod(1980)]{axe80} Axelrod, T. S. 1980, Ph.D. thesis,
681: Univ. of California, Santa Cruz
682:
683: \bibitem[Berger, Kulkarni, \& Chevalier(2002)]{berger02} Berger, E.,
684: Kulkarni, S. R., \& Chevalier, R. A., 2002, \apj, 577, L5
685:
686: \bibitem[Burstein \& Heiles(1982)]{burstein82} Burstein, D., \& Heiles, C.,
687: 1982, \aj, 87, 1165
688:
689: \bibitem[Cappellaro et al.(1997)]{cap97} Cappellaro, E., Mazzali, P. A.,
690: Benetti, S., Danziger, I. J., Turatto, M., Della Valle, M., \&
691: Patat, F. 1997, \aap, 328, 203
692:
693: \bibitem[Deng et al.(2004)]{den04} Deng, J., Tominaga, N., Mazzali, P. A.,
694: Maeda, K., \& Nomoto, K. 2004, \apj, 624, 898
695:
696: \bibitem[Duncan \& Thompson(1992)]{dun92} Duncan, R. C., \& Thompson, C. 1992,
697: \apj, 392, L9
698:
699: \bibitem[Filippenko(1997)]{fil97} Filippenko, A. V. 1997, ARAA, 35, 309
700:
701: \bibitem[Filippenko \& Leonard(2004)]{fil04} Filippenko, A. V., \& Leonard,
702: D. C. 2004, in Cosmic Explosions in Three Dimensions, ed. P. H\"{o}flich,
703: P. Kumar, \& J. C. Wheeler (Cambridge: Cambridge Univ. Press), 30
704:
705: \bibitem[Filippenko et al.(1995)]{fil95} Filippenko, A. V., et al.
706: 1995, ApJ, 450, L11
707:
708: \bibitem[Foley et al.(2003)]{fol03} Foley, R. J., et al. 2003, \pasp, 115, 1220
709:
710: \bibitem[Fynbo et al.(2003)]{fyn03} Fynbo, J. P. U., et al. 2003, \apj, 609, 962
711:
712: \bibitem[Galama et al.(1998)]{gal98} Galama, T. J., et al. 1998, \nat, 395, 670
713:
714: \bibitem[Gal-Yam, Ofek, \& Shemmer(2002)]{galyam02} Gal-Yam, A., Ofek, E. O.,
715: \& Shemmer, O. 2002, \mnras, 332, L73
716:
717: \bibitem[Hurley et al.(2002)]{hurley02} Hurley, K., et al. 2002, GCN Circ. 1252
718:
719: \bibitem[Heise et al.(2001)]{heise01} Heise, J., in't Zand, J., Kippen, R. M.,
720: \& Woods, P. M. 2001, in Gamma-ray Bursts in the Afterglow Era,
721: ed. E. Costa, F. Frontera, \& J. Hjorth (Berlin: Springer), 16
722:
723: \bibitem[Iwamoto et al.(1998)]{iwa98} Iwamoto, K., et al. 1998, \nat, 395, 672
724:
725: \bibitem[Kawabata et al.(2002)]{kaw02} Kawabata, K., et al. 2002, \apj, 580, L39
726:
727: \bibitem[Kinugasa et al.(2002)]{kinu02} Kinugasa, K., et al. 2002, \apj, 577, L97
728:
729: \bibitem[Leonard \& Filippenko(2005)]{leo05} Leonard, D. C., \&
730: Filippenko, A. V. 2005, in 1604--2004, Supernovae as Cosmological
731: Lighthouses, ed. M. Turatto, et al. (San Francisco: ASP), 330
732:
733: \bibitem[Leonard et al.(2002)]{leo02} Leonard, D. C., Filippenko, A. V.,
734: Chornock, R., \& Foley, R. J. 2002, \pasp, 1333, 1348
735:
736: \bibitem[MacFadyen \& Woosley(1999)]{mac99} MacFadyen, A. I., \&
737: Woosley, S. E. 1999, \apj, 524, 262
738:
739: \bibitem[Maeda et al.(2002)]{mae02} Maeda, K., Nakamura, T., Nomoto, K.,
740: Mazzali, P. A., Patat, F., \& Hachisu, I., \apj, 565, 405
741:
742: \bibitem[Maeda et al.(2003)]{mae03} Maeda, K., et al. 2003, \apj, 593, 931
743:
744: %\bibitem[Maeda et al.(2006)]{mae06} Maeda, K., Nomoto, K., Mazzali, P. A., \&
745: % Deng, J. 2006, \apj, 640, 854
746:
747: \bibitem[Maeda et al.(2006)]{mae06} Maeda, K., Mazzali, P. A., \& Nomoto, K.,
748: \apj, 645, 1331
749:
750: \bibitem[Maeda et al.(2007)]{mae07} Maeda, K., et al. 2007, \apj, 658, L5
751:
752: \bibitem[Malesani et al.(2004)]{mal04} Malesani, D., et al. 2004, \apj, 609, L5
753:
754: \bibitem[Mazzali et al.(2004)]{maz04} Mazzali, P. A., Deng, J.,
755: Maeda, K., Nomoto, K., Filippenko, A. V., \& Matheson, T. 2004, \apj, 614, 858
756:
757: \bibitem[Mazzali et al.(2001)]{maz01} Mazzali, P. A., Nomoto, K.,
758: Patat, F., \& Maeda, K., et al. 2001, 559, 1047
759:
760: \bibitem[Mazzali et al.(2002)]{maz02} Mazzali, P. A., et al. 2002, \apj, 572, L61
761:
762: \bibitem[Mazzali et al.(2003)]{maz03} Mazzali, P. A., et al. 2003, \apj, 599, L95
763:
764: \bibitem[Mazzali et al.(2005)]{maz05} Mazzali, P. A., et al. 2005, Science, 308,
765: 1284
766:
767: \bibitem[Mazzali et al.(2006)]{maz06a} Mazzali, P. A., et al. 2006a, \apj, 645, 1323
768:
769: \bibitem[Mazzali et al.(2006)]{maz06b} Mazzali, P. A., et al. 2006b, \nat, 442, 1018
770:
771: \bibitem[Mazzali et al.(2007)]{maz07} Mazzali, P. A., et al. 2007, \apj, in press
772:
773: \bibitem[Modjaz et al.(2006)]{mod06} Modjaz, M., et al. 2006, \apj, 645, L21
774:
775: \bibitem[Nakamura et al.(2001)]{nak01} Nakamura, T., Mazzali, P. A.,
776: Nomoto, K., \& Iwamoto, K. 2001, \apj, 550, 991
777:
778: \bibitem[Nomoto et al.(2005)]{nom05} Nomoto, K., Maeda, K, Tominaga, N.,
779: Ohkubo, T., Deng, J., \& Mazzali, P. A. 2005, \apss, 298, 81
780:
781: \bibitem[Patat et al.(2001)]{pat01} Patat, F., et al. 2001, \apj, 555, 900
782:
783: \bibitem[Pian et al.(2006)]{pian06} Pian, E., et al. 2006, \nat, 442, 1011
784:
785: \bibitem[Ruiz-Lapuente \& Lucy(1992)]{RLL92} Ruiz-Lapuente, P., \& Lucy, L. B.
786: 1992, \apj, 400, 127
787:
788: \bibitem[Sauer et al.(2006)]{sauer06} Sauer, D., et al. 2006, \mnras, 369, 1939
789:
790: \bibitem[Schlegel, Finkbeiner, \& Davis(1998)]{schlegel98} Schlegel, D.,
791: Finkbeiner, D., \& Davis, M., 1998, \apj 500, 525
792:
793: \bibitem[Sharina, Karachentsev, \& Tikhonov(1996)]{sha96} Sharina, M. E.,
794: Karachentsev, I. D., \& Tikhonov, N. A. 1996, \aaps, 119, 499
795:
796: \bibitem[Soderberg et al.(2006)]{sod06} Soderberg, A. M., et al. 2006, \nat, 442,
797: 1014
798:
799: \bibitem[Sohn \& Davidge(1996)]{sohn96} Sohn, Y.-J., \& Davidge, T.J., 1996,
800: \aj, 111, 2280
801:
802: \bibitem[Soria, Pian, \& Mazzali(2003)]{soria03} Soria, R., Pian, E., \& Mazzali,
803: P. A., 2003, \aaps, 413, 107
804:
805: \bibitem[Stanek et al.(2003)]{sta03} Stanek, K. Z., et al. 2003, \apj, 591, L17
806:
807: \bibitem[Takada-Hidai, Aoki, \& Zhao(2002)]{tak02} Takada-Hidai, M.,
808: Aoki, W., \& Zhao, G. 2002, \pasj, 54, 899
809:
810: \bibitem[Thompson, Chang, \& Quataert(2004)]{tho04} Thompson, T. A., Chang, P.,
811: \& Quataert, E. 2004, \apj, 611, 380
812:
813: \bibitem[Tominaga et al.(2003)]{tom03} Tominaga, N., Deng, J., Mazzali, P. A.,
814: Maeda, K., Nomoto, K., Pian, E., Hjorth, J., \& Fynbo, J. P. U. 2003,
815: \apj, 612, L105
816:
817: \bibitem[Tomita et al.(2005)]{tom05} Tomita, H., et al. 2005, \apj, 644, 400
818:
819: \bibitem[Vinko et al.(2004)]{vinko04} Vinko, J., et al. 2004, \aaps, 427, 453
820:
821: \bibitem[Wang et al.(2001)]{wang01} Wang, L., Howell, D. A., H\"oflich, P.,
822: \& Wheeler, J. C., 2001, \apj, 550, 1030
823:
824: \bibitem[Woosley \& Bloom(2006)]{woo06} Woosley, S. E., \& Bloom, J. S. 2006,
825: \araa, 44, 507
826:
827: \bibitem[Yoshii et al.(2003)]{yosh03} Yoshii, Y, et al. 2003, \apj, 592, 467
828:
829: \end{thebibliography}
830:
831: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
832:
833: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
834:
835: \begin{figure}
836: %\plotone{neb02ap_newspec_f1.eps}
837: \plotone{f1.eps}
838: \figcaption[]{The nebular spectra of SN\,2002ap obtained at Subaru and NAOC. The
839: two earlier spectra have been shifted upward by adding a constant. The zero
840: level of the flux is shown by a thin horizontal line.}
841: \end{figure}
842:
843: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
844:
845: \begin{figure}
846: \epsscale{0.90}
847: %\plotone{neb02ap_all_1z_1_f2.eps}
848: \plotone{f2.eps}
849: \figcaption[]{The nebular spectra of SN\,2002ap obtained from 2002 June to 2002
850: September, compared to the synthetic spectra computed with the one-zone model
851: (dotted/red). Except for the lowest one,
852: all spectra have been shifted upward by arbitrary amounts. Long tick marks on
853: the left ordinate axis show the zero flux level for each spectrum. [{\em See
854: the electronic edition for a color version of this figure.}]}
855: \end{figure}
856:
857: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 3 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
858:
859: \begin{figure}
860: \epsscale{0.90}
861: %\plotone{neb02ap_all_1z_2_f3.eps}
862: \plotone{f3.eps}
863: \figcaption[]{The nebular spectra of SN\,2002ap obtained from 2002 October to
864: 2003 February, compared to the synthetic spectra computed with the one-zone
865: model (dotted/red). Except for the lowest one, all spectra have been shifted
866: upward by arbitrary amounts. Both the lowest spectrum and the corresponding
867: model have been multiplied by a factor of 2 for display purposes.
868: Long tick marks on
869: the left ordinate axis show the zero flux level for each spectrum. [{\em See
870: the electronic edition for a color version of this figure.}]}
871: \end{figure}
872:
873: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 4 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
874:
875: \begin{figure}
876: \epsscale{0.90}
877: %\plotone{neb02ap_all_sh_1_f4.eps}
878: \plotone{f4.eps}
879: \figcaption[]{The nebular spectra of SN\,2002ap obtained from 2002 June to 2002
880: September, compared to the synthetic spectra obtained with the multi-shell
881: model (dotted/red). Except for the lowest one,
882: all spectra have been shifted upward by arbitrary amounts. Long tick marks on
883: the left ordinate axis show the zero flux level for each spectrum. [{\em See
884: the electronic edition for a color version of this figure.}]}
885: \end{figure}
886:
887: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 5 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
888:
889: \begin{figure}
890: \epsscale{0.90}
891: %\plotone{neb02ap_all_sh_2_f5.eps}
892: \plotone{f5.eps}
893: \figcaption[]{The nebular spectra of SN\,2002ap obtained from 2002 October to
894: 2003 February, compared to the synthetic spectra computed with the multi-shell
895: model (dotted/red). Except for the lowest one, all spectra have been shifted
896: upward by arbitrary amounts. Both the lowest spectrum and the corresponding
897: model have been multiplied by a factor of 2 for display purposes.
898: Long tick marks on
899: the left ordinate axis show the zero flux level for each spectrum. [{\em See
900: the electronic edition for a color version of this figure.}]}
901: \end{figure}
902:
903: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE 6 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
904:
905:
906: \begin{figure}
907: \epsscale{0.90}
908: %\plotone{sne02ap98bw06ajOI_f6.eps}
909: \plotone{f6.eps}
910: \figcaption[]{A comparison of the [\OI] $\lambda\lambda$6300, 6363 line of
911: SNe 2002ap (solid grey/black line), 2006aj (long dashed/blue line),
912: 1998bw \citep[dotted/red line,][]{pat01}, and
913: 2003jd \citep[short dashed/green line,][]{maz05}.
914: %The profile of the line in
915: %SN\,2006aj is less peaked than that of SN\,1998bw, although the average
916: %expansion velocity of other elements is larger in SN\,1998bw than in
917: %SN\,2006aj, indicating a smaller degree of asphericity in SN\,2006aj.
918: Spectra have been scaled by various amounts to facilitate comparison.
919: [{\em See the electronic edition for a color version of this figure.}]}
920: \end{figure}
921:
922: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
923:
924:
925: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% TABLES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
926:
927:
928: %%%%%%%%%%%%%%%%%%%%%%%% Table 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
929:
930: \begin{deluxetable}{rccccccccc}
931: \tablewidth{0pt}
932: \tabletypesize{\scriptsize}
933: \tablenum{1}
934: \tablecaption{Parameters of the One-Zone Synthetic Spectra}
935: \tablehead{\colhead{UT Date} &
936: \colhead{SN epoch} &
937: \colhead{Telescope} &
938: \colhead{velocity} &
939: \colhead{$M$(\Nifs)} &
940: \colhead{$M$(O)} &
941: \colhead{$M$(Si)} &
942: \colhead{$M$(tot)} &
943: \colhead{$T_e$} &
944: \colhead{log ($n_e$/} \\
945: \colhead{~} &
946: \colhead{[days]\tablenotemark{a}} &
947: \colhead{~} &
948: \colhead{[km s$^{-1}$]} &
949: \colhead{[$\Msun$]} &
950: \colhead{[$\Msun$]} &
951: \colhead{[$\Msun$]} &
952: \colhead{[$\Msun$]} &
953: \colhead{[K]} &
954: \colhead{(g~cm$^{-3}$))} }
955: \startdata
956: 7 June 2002 & 129 & Subaru & 6200 & 0.090 & 0.71 & 0.60 & 1.78 & 5511 & 7.58 \\
957: 8 June 2002 & 130 & Lick 3~m & 6200 & 0.092 & 0.62 & 0.50 & 1.58 & 5488 & 7.56 \\
958: 11 July 2002 & 163 & Lick 3~m & 5750 & 0.088 & 0.71 & 0.57 & 1.73 & 5051 & 7.34 \\
959: 11 July 2002 & 163 & NAOC & 5750 & 0.090 & 0.71 & 0.58 & 1.73 & 5082 & 7.35 \\
960: 9 Aug 2002 & 192 & Lick 3~m & 5550 & 0.093 & 0.66 & 0.60 & 1.69 & 4751 & 7.17 \\
961: 15 Sept 2002 & 229 & Subaru & 5550 & 0.098 & 0.70 & 0.74 & 1.90 & 4360 & 6.95 \\
962: 1 Oct 2002 & 245 & Lick 3~m & 5550 & 0.102 & 0.67 & 0.78 & 1.86 & 4209 & 6.86 \\
963: 8 Oct 2002 & 253 & Lick 3~m & 5550 & 0.103 & 0.66 & 0.70 & 1.88 & 4160 & 6.82 \\
964: 6 Nov 2002 & 281 & Lick 3~m & 5550 & 0.110 & 0.71 & 0.80 & 2.04 & 3903 & 6.69 \\
965: 7 Jan 2003 & 343 & Lick 3~m & 5450 & 0.145 & 0.61 & 0.80 & 1.93 & 3422 & 6.49 \\
966: 27 Feb 2003 & 394 & Lick 3~m & 5200 & 0.170 & 0.62 & 0.80 & 1.92 & 3092 & 6.37 \\
967: \enddata
968: \tablenotetext{a}{The epoch is given from the putative date of explosion, 29 Jan
969: 2002 \citep{maz02}.}
970: \end{deluxetable}
971: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
972:
973: %%%%%%%%%%%%%%%%%%%%%%%% Table 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
974:
975: \begin{deluxetable}{rccccc}
976: \tablewidth{0pt}
977: \tabletypesize{\scriptsize}
978: \tablenum{2}
979: \tablecaption{Parameters of the Multi-Zone Synthetic Spectra}
980: \tablehead{\colhead{UT Date} &
981: \colhead{SN epoch} &
982: \colhead{Telescope} &
983: \colhead{$M$(\Nifs)} &
984: \colhead{$M$(O)} &
985: \colhead{$M$(Si)} \\
986: \colhead{~} &
987: \colhead{[days]\tablenotemark{a}} &
988: \colhead{~} &
989: \colhead{[$\Msun$]} &
990: \colhead{[$\Msun$]} &
991: \colhead{[$\Msun$]} }
992: \startdata
993: 7 June 2002 & 129 & Subaru & 0.11 & 1.21 & 0.44 \\
994: 8 June 2002 & 130 & Lick 3~m & 0.11 & 1.13 & 0.58 \\
995: 11 July 2002 & 163 & Lick 3~m & 0.11 & 1.33 & 0.31 \\
996: 11 July 2002 & 163 & NAOC & 0.11 & 1.33 & 0.31 \\
997: 9 Aug 2002 & 192 & Lick 3~m & 0.12 & 1.28 & 0.36 \\
998: 15 Sept 2002 & 229 & Subaru & 0.11 & 1.34 & 0.29 \\
999: 1 Oct 2002 & 245 & Lick 3~m & 0.11 & 1.29 & 0.36 \\
1000: 8 Oct 2002 & 253 & Lick 3~m & 0.11 & 1.29 & 0.33 \\
1001: 6 Nov 2002 & 281 & Lick 3~m & 0.11 & 1.29 & 0.31 \\
1002: 7 Jan 2003 & 343 & Lick 3~m & 0.13 & 1.28 & 0.36 \\
1003: 27 Feb 2003 & 394 & Lick 3~m & 0.13 & 1.35 & 0.30 \\
1004: \enddata
1005: \tablenotetext{a}{The epoch is given from the putative date of explosion, 29 Jan
1006: 2002 \citep{maz02}.}
1007: \end{deluxetable}
1008: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1009:
1010: \end{document}
1011: