1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass[apj]{emulateapj}
3:
4: \usepackage{graphicx}
5:
6: \shorttitle{Verifying the Cosmological Utility of SNe~Ia}
7: \shortauthors{Ellis {\it et al.}}
8:
9: \begin{document}
10:
11: \newcommand{\bq}{\begin{equation}}
12: \newcommand{\eq}{\end{equation}}
13:
14: \newcommand\omatter{\ifmmode \Omega_{\mathrm{M}}\else $\Omega_{\mathrm{M}}$\fi}
15: \newcommand\ok{\ifmmode \Omega_{\mathrm{k}}\else $\Omega_{\mathrm{k}}$\fi}
16: \newcommand\olambda{\ifmmode \Omega_{\Lambda}\else $\Omega_{\Lambda}$\fi}
17: \newcommand\dmB{\ifmmode \Delta m_{15}(B) \else $\Delta m_{15}(B)$\fi}
18: \newcommand\zspec{\ifmmode z_{\mathrm{spec}}\else $z_{\mathrm{spec}}$\fi}
19: \newcommand\zphot{\ifmmode z_{\mathrm{phot}}\else $z_{\mathrm{phot}}$\fi}
20: \newcommand\ebmvmw{\ifmmode E_{B-V}^{\small \mathrm{mw}}\else $E_{B-V}^{\small \mathrm{mw}}$\fi}
21: \newcommand\ebmvhost{\ifmmode E_{B-V}^{\small \mathrm{host}}\else $E_{B-V}^{\small \mathrm{host}}$\fi}
22: \newcommand\aperpix{\ifmmode \mathrm{\AA}\,\mathrm{pix}^{-1}\else \AA\,$\mathrm{pix}^{-1}$\fi}
23:
24: \title{Verifying the Cosmological Utility of Type Ia Supernovae:
25: Implications of a Dispersion in the Ultraviolet Spectra}
26:
27: \author{
28: R.~S.~Ellis\altaffilmark{1},
29: M.~Sullivan\altaffilmark{2,3},
30: P.~E.~Nugent\altaffilmark{4},
31: D.~A.~Howell\altaffilmark{2},
32: A.~Gal-Yam\altaffilmark{1},
33: P.~Astier\altaffilmark{5},
34: D.~Balam\altaffilmark{6},
35: C.~Balland\altaffilmark{5},
36: S.~Basa\altaffilmark{7},
37: R.~G.~Carlberg\altaffilmark{2},
38: A.~Conley\altaffilmark{2},
39: D.~Fouchez\altaffilmark{8},
40: J.~Guy\altaffilmark{5},
41: D.~Hardin\altaffilmark{5},
42: I.~Hook\altaffilmark{3},
43: R.~Pain\altaffilmark{5},
44: K.~Perrett\altaffilmark{2},
45: C.~J.~Pritchet\altaffilmark{6},
46: N.~Regnault\altaffilmark{5}
47: }
48:
49: \altaffiltext{1}{California Institute of Technology, E. California Blvd, Pasadena CA 91125, USA}
50: \altaffiltext{2}{Department of Physics and Astronomy, University of Toronto, 50 St. George Street, Toronto, ON M5S 3H4, Canada}
51: \altaffiltext{3}{Department of Physics (Astrophysics), University of Oxford, Keble Road, Oxford OX1 3RH, UK}
52: \altaffiltext{4}{Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA}
53: \altaffiltext{5}{LPHNE, CNRS-IN2P3 and Universit\'{e}s Paris VI \& VII, 4 Place Jussieu, 75252 Paris Cedex 05, France}
54: \altaffiltext{6}{Department of Physics and Astronomy, University of Victoria, PO Box 2055 STN CSC, Victoria BC V8T1M8, Canada}
55: \altaffiltext{7}{LAM, CNRS, BP8, Traverse du Siphon, 13376 Marseille Cedex 12, France}
56: \altaffiltext{8}{CPPM, CNRS-IN2P3 and Universit\'{e} Aix-Marseille II, Case 907, 13288 Marseille Cedex 9, France}
57:
58:
59: \email{rse@astro.caltech.edu}
60:
61: \begin{abstract}
62:
63: We analyze the mean rest-frame ultraviolet (UV) spectrum of Type Ia
64: Supernovae (SNe) and its dispersion using high signal-to-noise
65: Keck-I/LRIS-B spectroscopy for a sample of 36 events at intermediate
66: redshift ($\overline{z}$=0.5) discovered by the Canada-France-Hawaii
67: Telescope Supernova Legacy Survey (SNLS). We introduce a new method
68: for removing host galaxy contamination in our spectra, exploiting
69: the comprehensive photometric coverage of the SNLS SNe and their
70: host galaxies, thereby providing the first quantitative view of the UV
71: spectral properties of a large sample of distant SNe~Ia. Although
72: the mean SN~Ia spectrum has not evolved significantly
73: over the past 40\% of cosmic history, precise evolutionary
74: constraints are limited by the absence of a comparable sample of
75: high quality local spectra. The mean UV spectrum of our
76: $z\simeq$0.5 SNe~Ia and its dispersion is tabulated for use in
77: future applications. Within the high-redshift sample, we discover
78: significant UV spectral variations and exclude dust extinction as
79: the primary cause by examining trends with the optical SN color.
80: Although progenitor metallicity may drive some of these trends, the
81: variations we see are much larger than predicted in recent models
82: and do not follow expected patterns. An interesting new result is a
83: variation seen in the wavelength of selected UV features with phase.
84: We also demonstrate systematic differences in the SN~Ia spectral features
85: with SN lightcurve width in both the UV and the optical. We show
86: that these intrinsic variations could represent a statistical
87: limitation in the future use of high-redshift SNe~Ia for precision
88: cosmology. We conclude that further detailed studies are needed, both
89: locally and at moderate redshift where the rest-frame UV can be
90: studied precisely, in order that future missions can confidently be
91: planned to fully exploit SNe~Ia as cosmological probes.
92:
93: \end{abstract}
94:
95: \keywords{surveys -- supernovae: general -- cosmological parameters}
96:
97: \section{Introduction}
98: \label{sec:introduction}
99:
100: Supernovae of Type Ia (SNe~Ia) are now well-established as
101: cosmological distance indicators. In addition to the original surveys
102: by the Supernova Cosmology Project
103: \citep[SCP;][]{1997ApJ...483..565P,1999ApJ...517..565P} and the High-Z
104: Supernova Search Team \citep{1998ApJ...507...46S,1998AJ....116.1009R},
105: a new generation of SN~Ia surveys is underway both locally
106: \citep{2002SPIE.4836...61A,2005coex.conf..525L,2006PASP..118....2H}
107: and at higher redshifts
108: \citep{2006A&A...447...31A,2007ApJ...659...98R,2007ApJ...666..694W}.
109: Despite the availability of independent probes of the presence and
110: properties of dark energy from studies of the cosmic microwave
111: background \citep{2007ApJS..170..377S} and galaxy redshift surveys
112: \citep{2002MNRAS.330L..29E,2005MNRAS.362..505C,2005ApJ...633..560E},
113: the luminosity distance--redshift relation for SNe~Ia provides the
114: only {\it direct} evidence for a cosmic acceleration.
115:
116: The detection and characterization of dark energy, via measurements of
117: the average cosmic equation of state parameter $<$$w$$>$, requires the
118: precision measurement of SNe~Ia to redshifts $z\simeq$0.5--1, sampling
119: the epoch of cosmic acceleration \citep{2006A&A...447...31A}. However,
120: more precise constraints on the nature of dark energy, for example
121: evidence for any variation in $w$ with redshift, requires extending
122: these studies to redshift $z$$>$1
123: \citep{2004ApJ...607..665R,2007ApJ...659...98R} where the early
124: effects of deceleration may be detectable. As projects are developed
125: which plan to probe SNe~Ia beyond $z$=1 for this purpose
126: \citep[e.g][]{2005NewAR..49..346A,2006SPIE.6265E..67B}, it becomes
127: important to understand the possible limitations of using SNe~Ia as
128: distance probes. Key issues relating to the diversity of SNe~Ia and
129: their possible evolution with redshift as a population, together with
130: the limiting effects of dust and/or color corrections to their
131: photometric properties, are particularly crucial to understand.
132:
133:
134: Several local studies \citep{1995AJ....109....1H,1999AJ....117..707R,
135: 2000AJ....120.1479H,2001ApJ...554L.193H,
136: 2005A&A...433..807M,2005ApJ...634..210G} have already indicated
137: correlations between SN~Ia properties and host galaxy morphologies.
138: More recently, \citet{2006ApJ...648..868S} have shown that the
139: properties of distant SNe~Ia appear to be a direct function of their
140: local stellar population, with the distribution of light curve widths
141: and hence peak luminosities correlating with the host galaxy specific
142: star-formation rate. This work also determined that the rate of
143: SNe~Ia per unit stellar mass of their host galaxies is larger in
144: actively star-forming galaxies, suggesting many must be produced quite
145: rapidly in recently formed stellar populations, perhaps suggestive of
146: more than one progenitor mechanism. The authors conclude that SNe~Ia
147: may well be a bimodal or a more complex population of events
148: \citep[see also][]{2005ApJ...629L..85S,2006MNRAS.370..773M}. Such
149: diversity in the properties of SNe~Ia could have far-reaching
150: implications, particularly if the {\it mix} of mechanisms or delay
151: times within the broad population gradually changes with look-back
152: time
153: \citep[e.g.][]{2006MNRAS.370..773M,2006ApJ...648..868S,2007astro.ph..1912H}.
154:
155: These recent developments, which illustrate how improved precision
156: reveals new physical correlations in the SN~Ia population, raise the
157: broader question of whether future SN~Ia experiments might be limited
158: in precision by variations of a systematic nature within the
159: population, for example with redshift, which cannot be removed via
160: empirical correlations. Detailed local surveys such as the LOSS/KAIT
161: \citep{2005coex.conf..525L} and CfA surveys
162: \citep{1999AJ....117..707R,2006AJ....131..527J} have presented
163: valuable data on the homogeneity and trends in the SNe~Ia population.
164: Further promising work is being undertaken via the Supernova Factory
165: \citep{2002SPIE.4836...61A} and the Carnegie Supernova Project
166: \citep{2006PASP..118....2H}. Important though these continued programs
167: will be, they are insufficient to address all possible concerns about
168: the use of SNe~Ia as precision tools in cosmology. Comparable studies
169: at intermediate redshift\footnote{Defined here to represent the range
170: 0.2$<$$z$$<$0.7.} will be particularly important in order to address
171: questions relating to possible evolutionary effects and environmental
172: dependencies. In addition, it is not always practical at low redshift
173: to cover the full wavelength range necessary to test for systematic
174: trends.
175:
176: In this paper we analyze high signal-to-noise ratio rest-frame
177: ultraviolet (UV) spectra of a large sample of intermediate redshift
178: SNe~Ia drawn from the Canada-France-Hawaii Telescope Supernova Legacy
179: Survey \citep[SNLS;][]{2006A&A...447...31A}, a rolling search which is
180: particularly effective for locating and studying events prior to their
181: maximum light. Our aim is obtain a substantially higher
182: signal-to-noise in the spectra than that typically obtained during
183: spectroscopic programs to type SNe and measure redshifts. We target
184: the UV wavelength region because in this wavelength region the SN
185: spectrum is thought to provide the most sensitive probe of {\it
186: progenitor metallicity}
187: \citep[e.g.][]{1998ApJ...495..617H,2000ApJ...530..966L}, a variable
188: which may shed light on the possibility of progenitor evolution. The
189: time-dependent UV spectrum is also needed in estimating ``cross-band''
190: $k$-corrections, particularly at redshifts $z>1$ where optical
191: bandpasses probe the rest-frame near-UV
192: \citep{2004ApJ...607..665R,2007ApJ...659...98R}. Little is known about
193: the properties and homogeneity of the UV spectra of SNe~Ia, largely
194: because of the absence of suitable instruments for studying this
195: wavelength range in local events. Although some local SN~Ia UV spectra
196: are available from International Ultraviolet Explorer (IUE) or Hubble
197: Space Telescope (HST) satellite data
198: \citep[e.g.][]{1991ApJ...371L..23L,1993ApJ...415..589K,1995ESASP1189.....C},
199: the bulk of the progress now possible in this area can be provided
200: from optical studies of intermediate-redshift events with large
201: ground-based telescopes.
202:
203: The goals of this paper are thus to address the question of both the
204: diversity and possible physical evolution in the intermediate redshift
205: SN~Ia family. We compare the rest-frame UV behavior of local SNe~Ia
206: with that derived for intermediate-redshift ($z\simeq$0.5) events
207: where the rest-frame UV enters the region of high efficiency of the
208: Keck LRIS-B spectrograph. We also study the degree to which the UV
209: spectra at intermediate redshift represent a homogeneous population,
210: independent of other variables such as the physical environment and
211: light curve stretch.
212:
213: A plan of the paper follows. In $\S$~\ref{sec:selection-cfhtls-sne} we
214: introduce the salient features of the SNLS and our method for
215: selecting SNe~Ia for detailed study. In
216: $\S$~\ref{sec:keck-observations} we discuss the Keck spectroscopic
217: observations and their reduction, including the treatment of host
218: galaxy subtraction and flux calibration. In $\S$~\ref{sec:analyses} we
219: consider our sample with respect to the broader set of SNe found by
220: SNLS, ensuring it is a representative subset in terms of various
221: observables, and discuss existing local UV spectra. In
222: $\S$~\ref{sec:results}, we undertake the detailed analysis. First we
223: compare the UV spectra found in our sample with those found locally.
224: We then examine the diversity of intermediate-redshift SNe~Ia in
225: various ways and correlate the UV variations with the light curves of
226: the SNe and the properties of the host galaxies. We discuss these
227: trends in terms of progenitor mechanisms in $\S$~\ref{sec:discussion}
228: and examine the implications in terms of possible long term
229: limitations of SNe~Ia as probes of dark energy. We also present the
230: mean phase-dependent SN Ia spectrum and its uncertainties for use in
231: future work.
232:
233: \section{Selection of SNLS SNe~Ia}
234: \label{sec:selection-cfhtls-sne}
235:
236: Our SNe~Ia are taken from the Supernova Legacy Survey (SNLS), a
237: ``rolling'' search for distant SNe with a primary science goal of
238: determining the average equation-of-state parameter of dark energy,
239: $<$$w$$>$ \citep[see][]{2006A&A...447...31A}. SNLS exploits the
240: square-degree Megacam camera \citep{2003SPIE.4841...72B} on the
241: Canada-France-Hawaii Telescope (CFHT), and comprises repeat imaging in
242: 4 filters, $g'r'i'z'$, of four deep 1$\times$1 degree fields (see
243: \citeauthor{2006AJ....131..960S} \citeyear{2006AJ....131..960S} for
244: the field coordinates), plus further $u^{\star}$ imaging which is not
245: time-sequenced. Each field is imaged several ($\sim5$) times per
246: lunation for 5-6 lunations per year. A description of the real-time
247: search operations and the criteria for following SN candidates
248: spectroscopically can be found in \citet{2006AJ....131..960S}.
249: Spectroscopic follow-up time for the essential work of basic redshift
250: measurement and SN type determination for cosmological analyses comes
251: from major long-term programs at the European Southern Observatory
252: Very Large Telescopes (Basa et al., in prep; PI: Pain), the Keck observatory
253: (PI: Perlmutter) and the Gemini North and South telescopes \citep[][PI:
254: Hook]{2005ApJ...634.1190H,2007bronder}.
255:
256: Observing time with the Keck-I telescope for the detailed
257: spectroscopic study of individual SNe presented in this paper was
258: scheduled from 2003 through 2005. The goal was to obtain substantially
259: higher signal-to-noise ratio spectra than those required for SN typing
260: (particularly in the rest-frame UV), and hence we targeted SNe~Ia with
261: a mean redshift lower than that of the SNLS as a whole
262: ($\overline{z}$=0.45 versus $\overline{z}$=0.6) and used integration
263: times $\sim$ 3--4 times longer for that redshift.
264:
265: To avoid spectroscopic screening of every candidate (results from the
266: mainstream VLT and Gemini programs were typically not available at the
267: time of the Keck observations), a code was developed to constrain the
268: redshift, SN type and epoch from the available photometric data
269: \citep[see][for details]{2006AJ....131..960S}. In selecting candidates
270: for this program, the following criteria were adopted:
271: \begin{enumerate}
272: \item{}The SN photometric redshift lay at $z$$<$0.75,
273: \item{}The predicted phase was such that the Keck observations would
274: be conducted prior to or close to maximum light.
275: \item{}The light curve and color were consistent with a SN~Ia,
276: \end{enumerate}
277: In practise, this last criteria is conservatively applied to minimize
278: the risk of selecting against SNe~Ia which may differ from the
279: template used in our selection code, though this reduces the fraction
280: of SNe~Ia observed: By the end of 2005, after 20 nights of Keck time,
281: 58 SN candidates were observed of which 36 were confirmed as SNe~Ia
282: ($\sim$60\%). A higher fraction of SNe~Ia could have been obtained by
283: a more rigorous application of the pre-selection technique; this
284: argues that any bias introduced by our selection is small.
285:
286: Table~\ref{tab:sninfo} summarizes the key parameters for the SNe~Ia
287: sampled in the Keck campaign (an independent survey of SNe~IIP was
288: also undertaken during this period \citep{2006ApJ...645..841N}.) We
289: restrict these 36 SNe to a high-quality subset, removing 3 as
290: spectroscopically peculiar (see $\S$~\ref{sec:mean-uv-spectrum}), 2
291: with a low-S/N Keck spectrum, and 5 with poor $g'$ light curve
292: coverage which precludes an accurate subtraction of the host galaxy
293: from the SN spectrum (see Table~\ref{tab:sninfo}). These later 5 were
294: all observed during the SNLS ``pre-survey'' when the light curve
295: coverage in $g'$ was less dense, and hence our photometrically
296: calibrated SN spectrum will suffer a greater systematic uncertainty
297: (see $\S$~\ref{sec:spectr-calibr-host}). Analysis of the remaining 26
298: SNe~Ia forms the basis of this paper.
299: Fig.~\ref{fig:redshiftphasehist} displays the redshift distribution of
300: these high-quality SNe~Ia and the breakdown in phase at which the Keck
301: spectra were obtained.
302:
303: For comparison purposes, we will also make use of an additional sample
304: of 160 spectroscopically-confirmed SNLS SNe~Ia (without high quality
305: Keck coverage) within the above redshift range for which good
306: broad-band photometric data is available. This will be used as a
307: control to demonstrate our SN~Ia selection is unbiased relative to the
308: larger SNLS population in terms of fundamental properties such as
309: light curve stretch, host galaxy type and location within each host.
310:
311: \section{Spectroscopic Data}
312: \label{sec:keck-observations}
313:
314: \subsection{Observations}
315: \label{sec:observations}
316:
317: All spectroscopic observations of SNLS SNe presented here were
318: conducted with the 10 meter Keck-I telescope using the two-channel Low
319: Resolution Imaging Spectrograph \citep[LRIS;][]{1995PASP..107..375O}.
320: Although SNe~Ia have broad absorption features, the need to monitor
321: expansion velocities to reasonable precision as well as to locate
322: narrower diagnostic lines which would otherwise be blended in broader
323: features defined the need for intermediate (600 l mm$^{-1}$)
324: dispersion gratings. Access to the rest-frame UV over 0.2$<$z$<$0.7
325: was a prime goal given the sensitivity of features in this region to
326: metallicity, temperature and photospheric expansion velocity.
327:
328: Taking these considerations into account, we adopted a spectroscopic
329: set up using a 400 line\,mm$^{-1}$ grating blazed at 8500\AA\ (giving
330: a dispersion of $\simeq1.9$\,\aperpix) on the red side and a 600 line
331: grism blazed at 4000\AA\ (dispersion $\simeq0.6$\,\aperpix) on the
332: blue side. With a 5600\AA\ dichroic, this gave a contiguous
333: wavelength coverage from the atmospheric limit on the blue side
334: through to 9400\AA\ on the red side. On some of the earliest runs in
335: 2003, we experimented with a 300 line grism blazed at 5000\AA\
336: (dispersion $\simeq1.4$\,\aperpix) on the blue side and a 600 line
337: grating blazed at 7500\AA\ (dispersion $\simeq1.3$\,\aperpix) on the
338: red side with a 6800\AA\ dichroic. This provided a lower resolution
339: on the blue side, but a more optimal set up on the red side for the
340: higher redshift SNe. However, we found the 6800\AA\ dichroic did
341: not permit accurate flux calibration on the blue side, because of
342: residual contamination in the second order of diffraction. In very
343: good seeing conditions, we used a 0.7\arcsec\ slit, otherwise a
344: 1\arcsec\ slit was used.
345:
346: Observations were performed at the parallactic angle
347: \citep[e.g.][]{1982PASP...94..715F}. Each individual exposure was
348: typically between 1200-1800s, and the SN was dithered along the slit
349: by 3-5\arcsec\ between exposures to aid with fringe removal in the
350: data reduction. The telescope was refocused periodically during each
351: night from which an estimate of the seeing was measured.
352: Table~\ref{tab:observinglog} summarizes these spectroscopic
353: observations.
354:
355: \subsection{Data Reduction}
356: \label{sec:data-reduction}
357:
358: All spectroscopic data were processed using a pipeline developed by
359: one of us (MS). This pipeline uses both standard Image Reduction and
360: Analysis Facility (IRAF\footnote{IRAF is distributed by the National
361: Optical Astronomy Observatories, which are operated by the
362: Association of Universities for Research in Astronomy, Inc., under
363: cooperative agreement with the National Science Foundation.})
364: software as well as our own custom-written routines. LRIS is a two-arm
365: spectrograph and different reduction techniques are required on each
366: side as the red side suffers significant fringing above 7000\AA.
367:
368: Our first step is to remove the overscan level on each amplifier of
369: each CCD, and then subtract a master zero-frame constructed on the
370: afternoon of each night of observing to remove any bias pattern. Both
371: sides are then divided by a normalized internal flat-field which
372: removes pixel-to-pixel sensitivity variations. This division also
373: corrects for the different gains of the LRIS CCD amplifiers. The
374: 0.7\arcsec\ slit additionally requires division by a ``slit-flat'' or
375: ``illumination'' flat-field as the slit illumination is quite uneven,
376: presumably due to milling defects in the slit. An additional
377: complication is that the spatial position of the slit on the CCD can
378: drift by several pixels over the course of the night, so the slit-flat
379: is shifted to best match each observation. Cosmic-rays are then
380: identified using LACOSMIC \citep{2001PASP..113.1420V} and removed via
381: interpolation from neighboring pixels.
382:
383: We perform our sky subtraction following the technique of
384: \citet{2003PASP..115..688K}, which subtracts a two-dimensional sky
385: frame constructed from sub-pixel sampling of the background spectrum
386: and a knowledge of the wavelength distortions as determined from
387: two-dimensional arc comparison frames. For the blue side, we proceed
388: directly to the spectral extraction after sky subtraction; on the red
389: side we perform an additional fringe correction before extraction.
390:
391: To generate the master fringe frame, we first divide every
392: sky-subtracted frame by the two-dimensional sky frame that was
393: subtracted and that caused the fringing, resulting in a
394: two-dimensional map of the fringe strengths per photon for each
395: exposure. All individual fringe maps taken in a given configuration
396: are averaged with sigma-clipping to form a master fringe frame,
397: masking out regions in each exposure that contain object flux, and
398: weighting each frame by the exposure time of the observation. As the
399: observations are dithered, the objects are located on a different part
400: of the CCD in each frame, and the fringe strength at every pixel is
401: therefore sampled across an entire night. Finally, we boxcar smooth
402: the master fringe frame to increase the signal-to-noise ratio in the
403: fringes, which typically extend over several pixels. This smoothed,
404: master fringe frame is then scaled and subtracted from each red-side
405: exposure at the stage prior to sky-subtraction, and the
406: sky-subtraction then repeated on the fringe-subtracted images.
407:
408: The two-dimensional frames are transformed to a constant dispersion
409: using comparison arc lamp images. Because the objects under study
410: typically have a reasonable signal-to-noise ratio, the spectral
411: extraction was performed by tracing the object position on the CCD and
412: using a variance-weighted extraction in a seeing matched aperture,
413: equal to 1.25 times the seeing. An error spectrum from the statistics
414: of the photon noise is also extracted at the same time. The wavelength
415: calibration of each extracted spectrum is then tweaked using the
416: position of the night-sky lines to account for any drift in the
417: wavelength solution during a given night. We then perform an
418: approximate telluric and relative flux calibration using
419: spectrophotometric standard stars taken from
420: \citet{1988ApJ...328..315M}, and correcting for typical Mauna Kea
421: atmospheric extinction using the extinction law of
422: \citet{1987PASP...99..887K} and the effective airmass of each
423: exposure. We note that flux calibration errors introduced during this
424: process that vary smoothly as a function of wavelength will be
425: corrected by our calibration to the SN lightcurves described in the
426: next section.
427:
428: Following optimal combination of the different exposures and relative
429: flux calibration of the blue and red components, our final step is to
430: scale the two sides into a single calibrated spectrum. We first match
431: the flux across the dichroic by defining narrow box filters either
432: side of the dichroic and scale each spectrum appropriately. We then
433: use a weighted mean to combine the spectra, in the process re-binning
434: to a constant 2\AA\ per pixel resolution. The result is a contiguous
435: science spectrum, together with an error spectrum representing the
436: statistical uncertainties in the flux in each binned pixel.
437:
438: Where possible, we measure redshifts from lines in the host galaxy
439: spectrum. In some cases the host galaxy was too faint, or the SN too
440: isolated from the host galaxy, to measure a redshift from the galaxy
441: lines, and the redshift was determined from template fits to the SN
442: spectrum \citep[e.g.][]{2002AAS...201.9103H}. As a final step prior to
443: scientific analysis we interpolate across the positions of host galaxy
444: emission features (e.g., H$\alpha$, H$\beta$, \ion{O}{2}, etc.) to
445: remove these from the spectra.
446:
447: The SNe~Ia themselves were identified by a lack of hydrogen in their
448: spectra, combined with broad (several thousand km\,s$^{-1}$) P Cygni
449: lines of elements such as \ion{Si}{2}, \ion{S}{2}, \ion{Ca}{2},
450: \ion{Mg}{2}, and blends of Fe peak lines. The classification of SNe is
451: discussed extensively in the literature; see
452: \citet{1997ARA&A..35..309F} for a review, and
453: \citet{2005AJ....130.2788H}, \citet{2005A&A...430..843L},
454: \citet{2005AJ....129.2352M}, and \citet{2005ApJ...634.1190H} for
455: issues associated with high-redshift SN classification. SNLS as a
456: whole adopts the numerical classification scheme presented in
457: \citet{2005ApJ...634.1190H} for typing SNe: a confidence index (CI)
458: that a given spectrum is of a SN Ia. Three classifications denote
459: SNe~Ia: ``Certain SN Ia'' (CI:5; type denoted as SN~Ia), ``Highly
460: probable SN Ia'' (CI:4; type denoted as SN~Ia) and ``Probable SN~Ia''
461: (CI:3; type denoted as SN~Ia$^{\ast}$). In all of the spectra in this
462: paper, the relatively low-redshift compared to the SNLS sample as a
463: whole, coupled with the long integration times, results in high
464: signal-to-noise ratios ensuring the classifications are unambiguous.
465: All SNe~Ia presented here are CI=5 on the SNLS scheme.
466:
467: \subsection{Spectrophotometric Calibration and Host Galaxy Subtraction}
468: \label{sec:spectr-calibr-host}
469:
470: Although standard spectrophotometric stars were observed periodically
471: on each observing night, these provide only an approximation to a flux
472: calibrated spectrum. Complexities such as differential slit-losses
473: (as the seeing is wavelength dependent) and uncertain atmospheric
474: extinction corrections require a more rigorous treatment.
475: Furthermore, a key issue in the interpretation of SN spectra is the
476: removal of underlying contaminating flux from the host galaxy, which
477: for some high-redshift spectra can be significant. We adopted a
478: procedure for removing the host galaxy spectrum that takes advantage
479: of the photometric $u^{\star}g'r'i'z'$ host-galaxy data available from
480: the SNLS, as well as the near-simultaneous SNLS $g'r'i'z'$ SN
481: light-curve photometry, ensuring that each spectrum is carefully
482: corrected to be consistent with the appropriate broad-band SN colors.
483:
484: We begin by measuring the surface brightness of each galaxy (in flux
485: units per square arcsecond) at the position of the SN in a series of
486: small apertures (with radii of 1-8 pixels) for each of the
487: $u^{\star}g'r'i'z'$ filters. These fluxes were measured from deep
488: stacks with no SN light present, constructed for the purpose of
489: studying the properties of the SN host galaxies
490: \citep{2006ApJ...648..868S}. The image quality of these stacks was
491: around 0.8\arcsec\ in each filter, similar to the seeing for the
492: majority of our LRIS spectral observations. The amount of host galaxy
493: light present in each SN spectrum is then estimated by interpolating
494: the galaxy surface brightnesses at the seeing of the observation, and
495: then multiplying by the product of the slitwidth ($S_W$) and the
496: spectral extraction aperture (1.25$\times$seeing;
497: $\S$~\ref{sec:data-reduction}), giving a series of $u^{\star}g'r'i'z'$
498: fluxes representing the host galaxy contamination in our spectra. We
499: then fit a series of smooth galaxy spectral energy distribution (SED)
500: templates to these flux data to estimate the contaminating host galaxy
501: spectrum. These templates are generated by the galaxy spectral
502: synthesis code P\'EGASE.2
503: \citep{1997A&A...326..950F,1999astro.ph.12179F} rebinned to the
504: resolution of our SN spectra. We interpolate between neighboring
505: galaxy spectra to find the best-fitting template. The resulting
506: spectrum represents the estimate of the contaminating host galaxy
507: continuum in every SN spectrum.
508:
509: We next use the $g'r'i'z'$ SNLS light curves and their resulting
510: light-curve parameterization (using the lightcurve fitter SiFTO;
511: Conley et al., in prep) to interpolate the SN flux at the time of each
512: spectral observation from nearby light-curve observations. Due to the
513: rolling-search nature of the SNLS and the dense light curve coverage,
514: most of the spectra in our main sample have $g'$ photometric data
515: within 2 rest-frame days of the spectroscopic observation, and all
516: have data within 4 days (Table~\ref{tab:sninfo}), hence the
517: interpolation required is small.
518:
519: We next estimate the amount of SN flux that went through the LRIS slit
520: and was inside the spectral extraction window of 1.25$\times$seeing,
521: and hence is present in the reduced spectrum. We model the SN PSF as
522: a two-dimensional Gaussian with $\sigma=\textrm{seeing}/2.355$ and
523: integrate this Gaussian through an rectangular aperture of
524: $S_W$$\times$(seeing$\times$1.25), where $\sigma$ is calculated at the
525: effective wavelengths of the $g'r'i'z'$ filters adjusting the seeing
526: as $\lambda^{-0.2}$. This estimates the fraction of the total SN flux
527: that passed through the slit, mimicking the inclusion of differential
528: slit-losses. These adjusted SN fluxes are added to the host galaxy
529: fluxes to generate combined ``SN+host'' $g'r'i'z'$ fluxes, the
530: estimate of the amount of flux that passed through the slit.
531:
532: We correct our combined, contiguous and flux-calibrated spectrum to
533: have the same colors and absolute flux level as dictated by these
534: ``SN+host'' fluxes using a smooth interpolating multiplicative spline
535: function \citep[see][]{2007ApJ...663.1187H}. The size of the
536: correction made to the observed spectrum is usually $<$10\% and is
537: invariably a monotonic function of wavelength. We then subtract our
538: best-fitting host galaxy template, and adjust the resulting spectrum
539: to have the correct colors of the SN on the night of observation
540: (again using a spline function), giving a SN spectrum with the correct
541: relative and absolute flux calibration. This final multiplicative
542: adjustment is small and essentially corrects the subtracted spectrum
543: for differential slit losses. We note that this technique would not be
544: possible without either the $u^{\star}g'r'i'z'$ host galaxy fluxes, or
545: the densely-sampled $g'r'i'z'$ SN~Ia light curves.
546:
547: We estimate the uncertainty in the host subtraction process using a
548: Monte-Carlo simulation. We repeat the subtraction for each SN 500
549: times, but adjusting the different variables (seeing, host-galaxy
550: fluxes, SN fluxes) according to their uncertainties and assuming
551: normal distributions. We assume a 15\% uncertainty in the seeing, and
552: also simulate errors in the centering of the SN in the slit,
553: introduced during target acquisition or by telescope tracking errors,
554: of $\pm0.15$\arcsec\ (1$\sigma$). Where appropriate, we use these 500
555: simulated spectra to estimate errors due to host galaxy subtraction in
556: derived quantities (such as spectral colors) in later sections. We
557: also carry forward in our analysis the statistical error arising from
558: the photon statistics in our spectra ($\S$~\ref{sec:data-reduction}).
559:
560: Our final task is to correct the SN~Ia spectrum for the effects of
561: extinction in the Milky-Way. To correct for Galactic extinction, we
562: use the dust-maps of \citet*{1998ApJ...500..525S} and a
563: \citet*{1989ApJ...345..245C} extinction law. We defer the more complex
564: discussion of extinction in the SN host galaxy and intrinsic SN color
565: variations until later.
566:
567: We illustrate the results of this process for all 26 of our high quality
568: SNe~Ia in Fig.~\ref{fig:eghostsub}\footnote{The online version contains
569: all spectra; the printed version contains two examples.}
570:
571: \section{Methodology}
572: \label{sec:analyses}
573:
574: We now turn to the two key questions our dataset is designed to address:
575:
576: \begin{enumerate}
577:
578: \item{} Has there been significant evolution in the mean SN~Ia
579: spectrum since $z\simeq$0.5 - a period of 5 Gyr corresponding to the
580: past 37\% of cosmic history?
581:
582: \item{} To what extent are there intrinsic spectral variations in our
583: sample? Do these give indications that SNe~Ia are a more complex
584: population than assumed in cosmological studies, and might these
585: variations limit the use of SNe~Ia in future cosmology experiments?
586:
587: \end{enumerate}
588:
589: \subsection{Theoretical Predictions}
590: \label{sec:theory}
591:
592: We will begin to address these questions by exploring the role that
593: our rest-frame UV data offers as a proxy for {\it progenitor
594: metallicity}. The possible effects of progenitor metallicity, either
595: as a function of redshift through galaxy evolution, or as a function
596: of environment or host-galaxy luminosity, can be explored through
597: blanketing and wavelength dependent features in the rest-frame UV
598: corresponding to $\lambda\lambda$ 2900-3500\AA. At z$>$0.2, this
599: portion of the rest-frame spectrum is well-sampled in our LRIS data.
600: In making the assumption that our UV data are indicative of
601: metallicity effects, it should be realized we are relying largely on
602: theoretical studies; the UV spectral region is poorly explored
603: observationally. First we review the various theoretical
604: expectations.
605:
606: \citet*{1998ApJ...495..617H} argue that direct traces of the progenitor
607: metallicity can best be seen in the unburned SN layers which are only
608: observable significantly before maximum light. However, they also
609: predict that an increase in progenitor metallicity will cause an
610: increase in the amount of $^{54}$Fe synthesized in the explosion, and
611: this will result in an increase in line opacity in the UV region which
612: may be observable at maximum light. The degree of mixing in the
613: explosion complicates the interpretation, however. The net effect
614: predicted by H\"oflich et al. is that an increased metallicity will
615: result in an {\it increase} in the UV pseudo-continuum at maximum
616: light.
617:
618: \citet{2000ApJ...530..966L} start with the findings of
619: \citet{1998ApJ...495..617H} and examine the spectroscopic implications
620: in greater detail. Using the starting isotopic distribution of the W7
621: model (Nomoto et al. 1984), they simultaneously change the progenitor
622: metallicity in the unburned C+O region and increase the amount of
623: $^{54}$Fe in the partially burned region. They find two effects -- a
624: shift in the wavelength of UV features redward with decreasing
625: metallicity, and a simultaneous increase in the level of the UV
626: pseudo-continuum. They argue that as metallicity decreases so the line
627: opacity decreases with the result that lines form deeper in the
628: atmosphere, and therefore from a lower velocity region. A lower
629: metallicity also decreases the cooling producing a higher temperature
630: and bluer colors. We note that such a color difference is {\it
631: opposite} to the effect predicted by H\"oflich et al. However,
632: Lentz et al. caution that the overall UV flux level is not necessarily
633: a good indicator of metallicity, as it is dependent on many variables
634: such as the temperature, density and velocity of the C+O layer.
635:
636: \citet*{2003ApJ...590L..83T} argue for much more dramatic changes in
637: SN~Ia physics with metallicity. They argue that synthesized $^{56}$Ni
638: mass should be linearly proportional to progenitor metallicity. Since
639: the decay of $^{56}$Ni drives the luminosity, SNe~Ia in high
640: metallicity environments should be less luminous. This is because
641: stars from higher metallicity environments will end up with larger
642: mass fractions of $^{22}$Ne and $^{56}$Fe after helium burning. Since
643: these isotopes have excess neutrons, the authors argue that in these
644: cases fewer radioactive elements are produced during the process of
645: burning to nuclear statistical equilibrium during a SN~Ia. Note that
646: these results are in sharp contrast to other studies which found no
647: significant increase in $^{56}$Ni with increasing metallicity
648: \citep[e.g.][]{1998ApJ...495..617H,1999ApJS..125..439I}
649:
650: Finally, we note that various authors predict that the SN~Ia rate
651: should be affected by metallicity, although they do not make explicit
652: predictions about the resulting effects on SN~Ia properties.
653: \citet{1998ApJ...503L.155K} argue that in very low metallicity
654: environments ($[Fe/H]< -1$) the white dwarf wind that they believe is
655: essential for producing SNe Ia will be inhibited, thus leading to
656: fewer SNe~Ia. \citet{2000A&A...362.1046L} find that metallicity
657: differences should alter the range of progenitor masses that produce
658: SNe~Ia.
659:
660: In summary therefore, theory cannot yet offer us a clear consensus as
661: to the effects of metallicity on the UV properties of SNe~Ia. Indeed,
662: there is disagreement not only about which effects are the most
663: important, but also about the sign of any possible effect. This is a
664: very challenging theoretical problem hindered by correlations between
665: the wanted effect of metallicity and other correlations such as the
666: viability of certain progenitor systems, the explosion mechanism and
667: radiative transfer in an atmosphere under a variety of mixing
668: conditions. Full simulations of all these effects may soon become
669: feasible, but substantial campaigns will still be needed to track
670: statistical shifts with metallicity. We therefore undertake an
671: empirically-motivated study of the properties of supernovae in the UV,
672: and note where the observations agree or disagree with certain
673: theoretical studies. In this respect, it is helpful to have some
674: theoretical basis for making the empirical measures.
675:
676: Fig.~\ref{fig:col_lentz} illustrates, for the range of metallicities
677: sampled by \citet{2000ApJ...530..966L}, that the bulk of the spectral
678: changes would clearly be detectable within our rest-frame wavelength
679: range. The blanketing and wavelength shifts apparent in
680: Fig.~\ref{fig:col_lentz} suggest the use of both photometric and
681: spectroscopic diagnostics. As high quality UV spectra are in scarce
682: supply at low redshift, we can best make meaningful comparisons using
683: rest-frame $U-B$ measurements. As our spectra do not always extend to
684: the full redward end of the standard $B$ filter, we will also adopt a
685: top-hat filter extending from 4000 to 4800\AA\ which we will refer to
686: as $b$. Pseudo-photometric ultraviolet measures can be constructed
687: from our spectral data using two top-hat filters, UV1 and UV2,
688: centered around regions where the strongest trends are predicted.
689: Although these various photometric measures degrade the information
690: content of our spectra, they have the considerable benefit of being
691: consistent over a wide range in redshift without making any
692: assumptions about the k-correction. The metallicity-induced spectral
693: shifts discussed above can likewise be tracked by measuring the rest
694: wavelengths of two UV features at $\lambda\lambda$2920 and 3180 \AA\ .
695: We will refer to these indicators as $\lambda_1$ and $\lambda_2$.
696: Both the photometric and wavelength features introduced above are
697: marked in Fig.~\ref{fig:col_lentz}.
698:
699: \subsection{Suitability of the Keck SNe~Ia Sub-sample for Detailed Studies}
700: \label{sec:suitability-keck-sne}
701:
702: If our conclusions concerning the properties of $z\simeq$0.5 SNe~Ia
703: are to be relevant for the role of SNe~Ia as cosmological probes, our
704: sample of SNe~Ia should be a representative subset of those being used
705: for probing dark energy. Accordingly, first we will verify whether
706: this is the case by comparing the distribution of properties with
707: those for the parent SNLS SN~Ia sample. This parent sample (N=160)
708: includes those presented by \citet{2006A&A...447...31A}, as well as
709: further events up to May 2006, restricted to lie within our chosen
710: redshift range 0.15$<z<$0.7. This is the redshift range over which
711: SNLS suffers least from Malmquist-type effects in the properties of
712: SNe chosen for spectroscopic follow-up, and the broad SNLS population
713: is reasonably unbiased over this redshift range
714: \citep[e.g.][]{2006A&A...447...31A,2006AJ....132.1126N}.
715:
716: Fig.~\ref{fig:samplecomparison} compares the distribution of SN~Ia
717: stretch, location and host galaxy properties for the two samples.
718: There is no evidence that our selection criteria ($\S$2) have
719: significantly biased the SN~Ia population chosen for intensive study
720: relative to the larger SNLS population. Kolmogorov-Smirnov (K-S)
721: tests indicate the Keck and SNLS sample distributions in all four
722: parameters are consistent with being drawn from the same distribution;
723: the exception is the stretch distribution where the K-S test indicates
724: the distributions are possibly different due to a marginal deficiency
725: of low stretch events in the Keck sample. This can be easily
726: understood as these events are both fainter (making the Keck
727: spectroscopic observations more demanding) and have faster light
728: curves (making a pre-maximum or at maximum light observation less
729: likely, a criterion when selecting SNe for Keck followup;
730: $\S$~\ref{sec:selection-cfhtls-sne}) than higher-stretch events.
731: Otherwise, the larger scale environs in which our SNe~Ia occurred seem
732: quite representative of the parent population, for example in the
733: specific star formation rate, the separation from the host galaxy, and
734: (not shown in Fig.~\ref{fig:samplecomparison}) the absolute magnitude
735: of the host galaxy.
736:
737: \subsection{Local UV Spectra}
738: \label{sec:local-data}
739:
740: In order to examine possible evolutionary effects we require a local
741: baseline UV spectrum. For this, we turn to the analysis of
742: \citet*{2002PASP..114..803N} which provides a phase-dependent template
743: SN~Ia spectrum which has become the standard reference used
744: extensively, for example, in the calculation of cross-color
745: $k$-corrections in deriving SNe~Ia luminosities at high redshift.
746: Further details of its construction can be found in the original
747: article. Note that though improved spectral templates now exist
748: \citep[e.g.][]{2007A&A...466...11G,2007ApJ...663.1187H}, these
749: typically exploit high-$z$ spectra in their construction, whereas
750: Nugent et al. use solely local SN~Ia data.
751:
752: The UV portion of this mean local spectrum is based on very few local
753: events studied by either the IUE or HST (see Table~1 in
754: \citet{2002PASP..114..803N}), and significant uncertainties remain at
755: the shortest wavelengths. Unfortunately, unless STIS is resuscitated
756: on the upcoming HST Servicing Mission, there is no immediate prospect
757: of improving this situation. A particular concern is the fact that
758: within a week of peak brightness only three SN~Ia (comprising of a
759: total of five spectra) contribute to the UV template. They are
760: SNe~1981B, 1990N and 1992A. These SNe have stretches of 0.929, 1.074
761: and 0.819 measured using SiFTO, i.e. within the range sampled here
762: (Fig.~\ref{fig:samplecomparison}). However, their hosts, classified as
763: SAB(rs)bc (NGC~4536), SAB(rs)bc (NGC~4639) and SA0 (NGC~1380)
764: respectively, are possibly biased towards large early type systems.
765:
766: \section{Results}
767: \label{sec:results}
768:
769: \subsection{Tests for Progenitor Evolution: the Mean UV SN~Ia Spectrum}
770: \label{sec:mean-uv-spectrum}
771:
772: We begin by comparing the mean spectrum of the intermediate-redshift
773: Keck SN~Ia sample with the local \citet{2002PASP..114..803N} template.
774: Earlier work comparing the properties of low and high-redshift SN~Ia
775: spectra
776: \citep[e.g.][]{2000ApJ...544L.111C,2005AJ....130.2788H,2006AJ....131.1648B,2007A&A...470..411G,2007bronder}
777: has concentrated largely on comparisons of optical spectral features
778: involving lower-S/N spectra. In general, these studies find that
779: equivalent widths and ejection velocities of optical spectroscopic
780: features measured on high-redshift SNe~Ia have similar distributions
781: to those observed locally. No study has yet found evidence for
782: evolution in the optical spectroscopic properties of SNe~Ia with
783: redshift.
784:
785: In this paper, we concentrate on the properties of the mean UV
786: spectra. As discussed earlier, models suggest metallicity variations will
787: have greatest effect in this wavelength region and thus our sensitivity
788: to evolutionary effects should be maximized. We separate the distant
789: SNLS events into three groups
790: (Fig.~\ref{fig:redshiftphasehist}): those observed prior to maximum
791: light ($t<$-4 rest-frame days), those observed close to maximum light
792: (-4$<t<$+4 days), and those observed after maximum light ($t>$+4
793: days). We exclude 3 spectra, labeled (f) in Table~\ref{tab:sninfo},
794: which we classify as peculiar. One is the super-Chandrasekhar event
795: SNLS-03D3bb \citep{2006Natur.443..308H} and the other two (SNLS-05D1hk
796: and SNLS-03D4cj) have peculiar spectra similar to that of SN1991T-type
797: \citep{1992ApJ...384L..15F}. (A further 7 events, labeled (d-e) in
798: Table~\ref{tab:sninfo}, were discarded as described in
799: $\S$~\ref{sec:selection-cfhtls-sne}.) At this stage, our comparisons
800: will involve spectra uncorrected for host galaxy extinction and/or
801: intrinsic color variations.
802:
803: Our 26 high-quality SNe~Ia consist of 8 pre-max, 15 maximum light, and
804: 3 post-max spectra. In creating a composite spectrum, flux-calibrated
805: spectra are normalized using the top-hat $b$ filter introduced earlier
806: which runs from $\lambda\lambda$4000-4800\AA\ (see
807: Fig.~\ref{fig:col_lentz}). We use a clipped mean and determine the
808: error on the mean by bootstrap resampling. To facilitate a meaningful
809: comparison we color-adjust the local spectrum so that its $U-B$ color
810: matches that of the mean high redshift sample.
811: Figs.~\ref{fig:mean_max} and \ref{fig:mean_early} show the
812: maximum-light and pre-maximum high-$z$ mean spectra compared against
813: the local template at similar phases. For convenience, we present in
814: Tables~\ref{tab:mean_spectra_early} and \ref{tab:mean_spectra_max},
815: the early and maximum light spectral energy distributions with upper
816: and lower 90\% confidence limits. For the local template, due to the
817: very small number of SNe~Ia with rest-frame UV spectra
818: ($\S$~\ref{sec:local-data}), a similar bootstrap resampling technique
819: to asses the uncertainty in the local template is not meaningful.
820:
821: Examining Fig.~\ref{fig:mean_max}, it is clear that the
822: spectrum at $\overline{z}\simeq$0.5 is reasonably similar to its local
823: equivalent. The agreement, particularly longward of 4000\AA, is
824: reassuring, indicating that there has been no significant evolution in
825: the mean SN~Ia spectroscopic properties over the past 5~Gyr. However,
826: some systematic departures are seen, for example, in the
827: \ion{Si}{2}/\ion{Co}{2} doublets at $\simeq$4150\AA\ and in the
828: positions of the UV features around 2900-3200\AA.
829:
830: Do these UV changes represent systematic differences in the progenitor
831: properties between $z\simeq$0.5 and 0? In this respect it is helpful
832: to consider the error in the $z\simeq$0.5 mean spectrum (shaded in
833: Fig.~\ref{fig:mean_max}). Here we see that the local
834: spectrum lies within the 90\% confidence boundary in the distribution
835: of $z\simeq$0.5 spectra. Thus it is quite possible that the UV
836: differences seen between high and low redshift arise as a result of
837: statistical variations occurring within a non-evolving population,
838: especially considering the number of local SNe~Ia with rest-frame UV
839: coverage is very small.
840:
841: Turning to the pre-maximum light spectra
842: (Fig.~\ref{fig:mean_early}), both \citet{1998ApJ...495..617H} and
843: \citet{2000ApJ...530..966L} argue that differences in the UV spectrum
844: due to changes in the progenitor metallicity should be larger at
845: earlier times, where the unburned outer layers of the white dwarf play
846: a larger role in shaping the appearance of the spectrum. Although the
847: agreement does indeed seem to be poorer, the local template is
848: considerably uncertain at this phase; we conclude that the most
849: likely explanation for the differences we see is that the local
850: template is unrepresentative in the far UV.
851:
852: However, the differences that we do see in the UV data, whether due to
853: evolution or simply as a result of intrinsic scatter in the
854: population, are much larger than the effects predicted by
855: \citet{2000ApJ...530..966L} for quite substantial changes in
856: progenitor metallicity. It should be noted that in these models,
857: luminosity and kinetic energy were fixed, whereas our observations
858: presumably span a considerable range in both of these variables.
859: Lentz et al. also predict systematic wavelength shifts of the
860: diagnostic features identified as $\lambda_1$ and $\lambda_2$ in
861: Fig.~\ref{fig:col_lentz}, the study of which should be independent of
862: reddening. We will defer consideration of these spectroscopic features
863: until $\S$~\ref{sec:spectroscopic_features}.
864:
865: At longer wavelengths, the local data is more reliable. Here, several
866: features of the pre-maximum spectrum deserve comment
867: (Fig.~\ref{fig:mean_early}). Compared to the low-z template, the
868: high-z average spectrum has a shallower \ion{S}{2} 5400\AA\ feature,
869: and a shallower \ion{Si}{2} 4130\AA\ feature. In addition the
870: 3700\AA\ Ca+Si feature is narrower and shows a greater degree of
871: splitting into the separate Ca and Si components in the high-z
872: composite spectrum. More high-z SNe~Ia belong to what
873: \citet{2006PASP..118..560B} call the ``shallow Si'' group --- SNe such
874: as SN~1991T, which have a narrower zone of intermediate mass elements.
875: These results agree with the findings of \citet{2001ApJ...546..734L}
876: who note that, while a significant fraction of SNe~Ia show such
877: behavior at early times, they become more normal at maximum light.
878: This supports the claims of \citet{2006ApJ...648..868S}: at high
879: redshift, where the fraction of star forming hosts is higher, more
880: ``prompt'' SNe are found and these tend to have broader light curves.
881:
882: The lack of a comprehensive local sample with comparable quality to
883: the Keck data extending into the UV clearly limits any precise tests
884: for spectral evolution. This impasse raises the question of whether it
885: might be more profitable to search for evolutionary trends within the
886: Keck sample itself. Fig.~\ref{fig:mean_max_z} shows the mean maximum
887: light spectra compiled for two Keck sub-samples, split into two
888: redshift ranges at $z$=0.50. The mean redshifts of the two subsamples
889: are $z$=0.36 (N=7) and $z$=0.58 (N=8). The agreement is striking, with
890: only marginal evidence for differences in blanketing at the shortest
891: wavelengths. Further progress in this respect may be possible with
892: the ongoing SDSS SN~Ia survey (which samples $\overline{z}\simeq$0.2)
893: and the higher-redshift SNLS SN~Ia spectra at $z\simeq0.6-1.0$.
894:
895: In summary, no convincing evidence is available for a {\it systematic}
896: change with redshift in the UV spectra of SNe~Ia. The bulk of the
897: differences seen probably arise from a natural dispersion within the
898: sample and it seems reasonable to assume this dispersion is present at
899: all epochs. In the next section we explore the possible origin of this
900: UV dispersion.
901:
902: \subsection{Reality of the UV Dispersion}
903: \label{sec:disp-uv-reality}
904:
905: In this section we first consider the physical reality of the
906: dispersion seen in the UV spectra at maximum light. The
907: bootstrap-resampled variations (Figs.~\ref{fig:mean_max} and
908: ~\ref{fig:mean_early}) indicate a greater degree of scatter at
909: $\lambda<3700$\AA\ than in the optical around $\lambda\sim5000$\AA.
910: However, both the effects of dust extinction and the size of the
911: calibration uncertainties are unaccounted for in these comparisons.
912: Could the increased dispersion simply arise as a result of one or
913: both of these effects and not represent a true physical variation
914: amongst our SNe?
915:
916: We first discuss the effects of dust extinction.
917:
918: \subsubsection{Effects of Dust Extinction}
919: \label{sec:dust}
920:
921: An explanation for the UV variations is the presence of varying amount
922: of interstellar dust, either in each host galaxy or along the line of
923: sight to the observer. Correcting the properties of SNe~Ia for
924: extinction is currently one of the most pernicious problems in their
925: use as cosmological probes \citep[e.g., see discussion
926: in][]{2007arXiv0705.0367C}. Earlier analyses
927: \citep{1996ApJ...473...88R,1998AJ....116.1009R, 1999ApJ...517..565P,
928: 2000ApJ...536...62R, 2003MNRAS.340.1057S, 2003ApJ...598..102K},
929: based on the first generation of SN~Ia surveys, examined the range of
930: likely extinction using both SN colors and host-galaxy morphologies as
931: markers. Generally only modest levels of extinction ($A_V<0^m.2$)
932: were seen at high-redshift across the full range of host types, with
933: no evidence of any systematic change with redshift in the samples used
934: for cosmological analyses.
935:
936: In the case of the present SNLS sample, the key diagnostic is the
937: rest-frame $B-V$ color estimator ``$c$''
938: \citep{2005A&A...443..781G,2007A&A...466...11G}, essentially the $B-V$
939: color measured at maximum light. We wish to test whether dust is the
940: primary cause of the UV variations seen in the Keck spectra. This can
941: be done by considering, statistically, correlations between the UV
942: spectra and $c$. Reddening measures for each SN are not directly
943: output by the SNLS analysis since $c$ is treated as an empirical
944: variable when performing the cosmological analysis
945: \citep{2006A&A...447...31A}; no physical model is assumed for the
946: color variations. However, we can readily test the hypothesis that
947: the bulk of the UV dispersion arises from reddening that is linked to
948: the $B-V$ variations.
949:
950: We investigate two methods for color-correcting our SNe~Ia spectra.
951: The first is to use a standard reddening law
952: (\citet{1989ApJ...345..245C}, CCM) and extinction-correct each
953: individual Keck spectrum using that SN's $c$ measurement and a value
954: of $R_B=4.1$, suitable for standard Milky-Way type dust. The results
955: from this process are not particularly sensitive to either the exact
956: $B-V$ color used as the fiducial/zeropoint SN~Ia color, nor to the
957: exact value of $R_B$ used. Our second approach is similar but uses the
958: SALT \citep{2005A&A...443..781G} color law together with the $c$
959: estimated from the light curve fits. As this color law was trained on
960: local SN spectra and colors, it includes not only the effect of dust
961: extinction but may also partially account for intrinsic variations
962: between SN optical colors and UV fluxes unrelated to dust. The CCM
963: and SALT laws have a similar form in the optical, but differ in the
964: near-UV, where the SALT law implies more ``extinction'' in the UV than
965: the CCM law \citep{2007A&A...466...11G}. To test the validity of
966: these approaches we split those Keck spectra studied at maximum light
967: into two sub-samples according to whether the rest-frame maximum light
968: optical color is redder or bluer than a $c$=-0.057.
969:
970: Examining the mean UV spectra for the two subsets
971: (Fig.~\ref{fig:mean_max_col}), we find that, before any correction,
972: optically bluer SNe~Ia do indeed have more flux in the UV as expected
973: if they are less extincted. When the spectra are individually
974: color-corrected using a CCM law and then combined by subset, the mean
975: UV spectra agree better though the entire color difference is not
976: corrected shortward of 3500\AA. The approach using the SALT color law
977: performs better in this UV region, and provides a good agreement
978: across our entire wavelength range, but is unable to correct all the
979: differences seen. Given the slightly superior performance of the SALT
980: color law, we adopt this approach for color-correcting our spectra in
981: the remainder of this paper.
982:
983: \subsubsection{Effects of Calibration Uncertainties}
984: \label{sec:effect-calibr-uncert}
985:
986: Our host galaxy subtraction and spectral calibration procedures are
987: described in $\S$~\ref{sec:spectr-calibr-host}. For each
988: host-subtracted and photometrically-calibrated SN spectrum we also
989: record a series of 500 Monte-Carlo simulated spectra generated by
990: randomly varying the SN fluxes, host galaxy fluxes, seeing and SN
991: position in the slit according to the observational errors.
992:
993: For each of the SN studied, we estimate the 1-$\sigma$ uncertainty as
994: follows. In each wavelength bin for a given SN, we order the 500
995: random spectra and find the range in flux that encompasses 68\% of the
996: population and assign this flux range as the 1-$\sigma$ error in this
997: wavelength bin. This is repeated in every wavelength bin for all the
998: spectra, providing the error from the observational uncertainties for
999: each SN spectrum as a function of wavelength.
1000:
1001: We can then assess whether the scatter that we see about the mean
1002: spectrum is significant. For every spectrum, we calculate the residual
1003: from the mean spectrum and divide by the appropriate error spectrum.
1004: The result is the deviation from the mean for each spectrum in units
1005: of sigma. Smoothed versions of these are plotted in
1006: Fig.~\ref{fig:meandisp}; the top panel shows the deviation for the
1007: uncorrected spectra, the middle and lower panels show the deviations
1008: for spectra corrected using the CCM and SALT laws as described in
1009: $\S$~\ref{sec:dust}.
1010:
1011: In general, the scatter is markedly larger in the UV
1012: ($\lambda<4000$\AA) than in the optical region. In the optical, the
1013: spectra appear well-behaved, with most of the deviations $<$2$\sigma$,
1014: consistent with a mean spectrum showing little intrinsic dispersion
1015: over the broad wavelength range.
1016:
1017: These tests and those of the previous section demonstrate that most of
1018: the difference in the \textit{average} continuum levels seen between
1019: the mean UV spectra of optically blue and red SNe~Ia can be corrected
1020: using existing techniques such as the CCM dust extinction law or the
1021: SALT SN~Ia color law. However, these techniques can only reduce, and
1022: not eliminate, the scatter observed, which is significantly larger
1023: than that expected from a consideration of the observational
1024: uncertainties. We thus conclude that the bulk of the dispersion
1025: is intrinsic to our SNe and investigate its origin in the next sections.
1026:
1027: \subsection{Physical Correlations within the UV Dispersion}
1028: \label{sec:uv-disp-behav}
1029:
1030: \subsubsection{Photometric Comparisons}
1031: \label{sec:photometric-comparisons}
1032:
1033: Given we have established that the UV variations are larger than those
1034: expected from photometric errors and are still present after attempts
1035: to correct for color differences, it is important to consider the
1036: extent to which the intrinsic dispersion we see is consistent or
1037: otherwise with that observed locally. Although no comparable local
1038: spectroscopic dataset exists, \citet{2006AJ....131..527J} have
1039: analyzed a homogeneous photometric database of 44 local SNe~Ia and
1040: examined the $U-B$ color dispersion at maximum light in terms of both
1041: the light curve stretch factor $s$ and redder colors such as $B-V$.
1042: As in the higher redshift data, there is still the complication of
1043: separating the effects of host galaxy dust extinction and intrinsic
1044: color effects. In the Jha et al. study, host galaxy extinction
1045: estimates were available for a subset of the data yielding intrinsic
1046: $U-B$ and $B-V$ colors. The increased intrinsic dispersion seen in the
1047: $U-B$ vs. stretch relationship with respect to that in $B-V$ can thus
1048: be attributed primarily to an increased dispersion in the $U$ band of
1049: $\sigma_U\simeq$0.12 ($\simeq$12\%).
1050:
1051: \citet{2006A&A...447...31A} also address the question of the intrinsic
1052: dispersion in $U$ by comparing the photometric properties of that
1053: subset of SNLS SNe~Ia for which some of the $g'r'i'z'$ bands map
1054: conveniently onto rest-frame $UBV$. They estimate the quantity
1055: $\Delta\,U_3$ which is the difference between the $U$ band flux at
1056: maximum light predicted from a suitably-chosen triplet of observed
1057: filters drawn from $g'r'i'z'$ and the actual observed $U$ magnitude. As
1058: the machinery adopted by Astier et al. includes provision for a
1059: color-stretch relation, this is not quite the same test as that
1060: adopted by Jha et al., although the local scatter found in
1061: $\Delta\,U_3$ is comparable (0.12).
1062:
1063: We compare our intermediate redshift $U-b$-stretch relation with that
1064: derived from the local sample in Fig.~\ref{fig:col_stretch}. In this
1065: figure all colors are color-corrected using a SALT color law and the
1066: method of $\S$~\ref{sec:dust}. To derive the rest-frame $U-b$ and $c$
1067: measures for the low-$z$ sample, we re-perform the light curve fits to
1068: the SN photometry using the same fitter used for the SNLS sample
1069: (SiFTO; Conley et al. in prep). Clearly, at maximum light the trends
1070: are very similar. As discussed in
1071: $\S$~\ref{sec:selection-cfhtls-sne}, the Keck sample (and the SNLS
1072: sample in general) is slightly under-represented in low-stretch SNe~Ia
1073: compared to the local data (Fig.~\ref{fig:samplecomparison}), so the
1074: color-stretch relation found in previous studies is less evident.
1075: Crucially, there is no evidence that the colors derived directly from
1076: the Keck spectra show a larger photometric dispersion. For events
1077: with stretch $s>$0.85 we find $(U-b)$=-0.452 with a standard deviation
1078: of 0.132 for 15 Keck SNe~Ia c.f. -0.436 (0.167) for 41 local events.
1079: The small difference in mean color probably arises from the different
1080: selection criteria, as Jha et al. are able to include events with
1081: greater extinction. Note there are small differences in the definition
1082: of ``maximum light''; the Keck spectra range over -4 to +4 days which
1083: will add an additional scatter to this sample in the presence of color
1084: evolution with phase.
1085:
1086: A similar behavior is seen in the $(UV2-b)$\,-\,stretch relation; a
1087: modest dispersion together with a weak correlation with stretch and a
1088: slight phase dependence (the definitions of UV1 and UV2 are shown in
1089: Fig.\ref{fig:col_lentz}). However, a marked increase in scatter is
1090: seen in the $(UV1-b)$\,-\,stretch relation (as expected if the UV1
1091: filter is the more sensitive to variations in progenitor composition)
1092: together with a stronger evolution with phase. Although the UV1
1093: filter is the shortest wavelength probed by the Keck spectra, this
1094: dispersion is considerable even within a phase bin, and is unlikely to
1095: arise from spectrophotometric errors given the tight dispersion
1096: observed in the $UV2-b$ relation. Fig.~\ref{fig:col_stretch} also
1097: shows weak correlations between color and stretch in the sense that
1098: the high stretch (brighter) events are bluer than those of low stretch
1099: (fainter)
1100: \citep[e.g.][]{1999AJ....118.1766P,2003ApJ...598..102K,2004ApJ...613.1120G}.
1101:
1102: Clearly as the SNe~Ia develop they display an increased dispersion in
1103: the wavelength region where metal-dependent features are expected to
1104: be most prominent. This may indicate that the metallicity of the
1105: outer, unburned layers, seen only at early times, is not as important
1106: as expected by \citet{1998ApJ...495..617H} and
1107: \citet{2000ApJ...530..966L}. Instead line blanketing from iron-peak
1108: elements may play the dominant role in affecting the UV opacity. This
1109: may affect maximum-light and post-maximum spectra more than for
1110: pre-maximum spectra. At early times the photosphere has not yet
1111: receded into the bulk of the iron-peak elements, and, as time
1112: progresses, $^{56}$Ni synthesized in the explosion decays into
1113: $^{56}$Co and subsequently into $^{56}$Fe. These additional isotopes
1114: may provide the increased line blanketing seen in the spectra.
1115:
1116: \subsubsection{Spectroscopic Features}
1117: \label{sec:spectroscopic_features}
1118:
1119: Next we consider trends in the spectroscopic features introduced by
1120: \citet{2000ApJ...530..966L} as possible metallicity diagnostics; these
1121: are marked as $\lambda_1$ and $\lambda_2$ in Fig.~\ref{fig:col_lentz}.
1122: The wavelengths of these features have the advantage of being
1123: independent of any dust correction. If progenitor evolution is largely
1124: driven by metallicity effects, we might expect significant shifts to occur
1125: in these features between the local and high-redshift data.
1126:
1127: In fact, a cursory examination of our mean spectra
1128: (Figs.~\ref{fig:mean_max} and ~\ref{fig:mean_early}) shows
1129: that the features do not appear to shift significantly in the mean.
1130: However this could be due to an inadequate or unrepresentative local
1131: data. We have seen that this is a significant limitation.
1132:
1133: Accordingly, we measured the wavelength of both features for each of
1134: our spectra individually. For each feature we fit a Gaussian on a
1135: linear continuum background, returning the central wavelength. The
1136: error is estimated by varying the continuum definition either side of
1137: the feature, performing the fit 100 times, and taking the standard
1138: deviation of the resulting central wavelength distribution.
1139:
1140: The precision is typically $\pm$2-5\AA, adequate to detect the shifts
1141: of $\simeq$10\AA\ predicted by \citet{2000ApJ...530..966L}. We found a
1142: much greater SN to SN scatter of $\pm$60\AA, with no obvious trend
1143: with stretch as might be expected from
1144: Fig.~\ref{fig:col_stretch}. However, as shown in
1145: Fig.~\ref{fig:bumps_phase}, for both features there is a possible
1146: shift with phase in the sense both move gradually to longer wavelength
1147: as the SN expands. Moreover, there is a considerable dispersion at a
1148: particular phase, particularly before maximum light.
1149:
1150: We find that $\lambda_2$ correlates reasonably well with the
1151: wavelength of the \ion{Si}{2} 4130\AA\ - a line commonly used to
1152: measure the photospheric expansion (Fig.~\ref{fig:siII_lambda2}).
1153: Thus, to first order, the scatter seen in $\lambda_2$ at a particular
1154: phase appears to arise from variations in kinetic energy within the
1155: sample. However the phase-dependent trend suggests a further process
1156: is at work. Fig.~\ref{fig:ratio_phase} shows the wavelength ratio
1157: versus phase, coded by stretch, where there is a noticeable shift at
1158: maximum light (corresponding to a redward shift of 50\AA\ in
1159: $\lambda_2$). This shift presumably arises as a result of the effects
1160: of temperature and ionization changes which are more pronounced in the
1161: UV feature.
1162:
1163: In summary, therefore, the above discussion suggests it will be very
1164: difficult to use these wavelength features as diagnostics of
1165: progenitor properties, unless perhaps larger samples are available.
1166: Clearly the analysis in Lentz et al. is oversimplified in several
1167: respects. Their models considered metallicity variations for a fixed
1168: luminosity and epoch, whereas it seems a wide range will be needed to
1169: isolate the effects of metallicity on this part of the spectrum.
1170: Recent work by \citet{2007ApJ...656..661K}, in which a grid of
1171: light curve models that span a large range in the production of
1172: $^{56}$Ni and intermediate mass material (and hence overall kinetic
1173: energy), should provide an excellent starting place to address these
1174: factors in future work.
1175:
1176: \subsubsection{Host Galaxy and Stretch Dependences}
1177: \label{sec: host_dependences}
1178:
1179: In view of recent work using the SNLS \citep{2006ApJ...648..868S} where
1180: convincing correlations have been found between the properties of
1181: distant SNe~Ia and their host galaxy environment, it is also natural to
1182: question whether the dispersion implied by the
1183: increased scatter in Fig.~\ref{fig:col_stretch} is
1184: dependent on the stellar population and total stellar mass.
1185:
1186: The Keck SN~Ia dataset samples a wide range of passive and
1187: star-forming host galaxies (Fig.~\ref{fig:samplecomparison}) for
1188: which stellar masses and luminosities are available \citep[see][for
1189: details]{2006ApJ...648..868S}. Specific star formation rates and
1190: masses have been measured for the hosts using SED fits to the
1191: extracted colors with the Z-PEG photometric redshift code
1192: \citep{2002A&A...386..446L}.
1193:
1194: In Fig.~\ref{fig:col_mass} we plot the color-corrected $UV1-b$ color
1195: against the host-galaxy stellar mass coding the data points according
1196: to whether the hosts are primarily passive or active. The
1197: discriminating specific star-formation rate was chosen to be 10$^{-12}$ M$_{\odot}$
1198: yr$^{-1}$ per unit stellar mass (Fig.~\ref{fig:samplecomparison}).
1199: Gas phase metallicities in star forming galaxies are well-known to
1200: correlate with galaxy mass
1201: \citep{2004ApJ...613..898T,2003ApJ...599.1006K} and recent studies
1202: have revealed these correlations were already in place at
1203: $z\simeq$1-1.5 \citep{2005ApJ...635.1006S}. Over the 3~dex range in
1204: stellar mass seen in our host galaxies, we would expect a change in
1205: metallicity of over 1~dex.
1206:
1207: If, as Sullivan et al. surmise, the higher SN~Ia rate in active hosts
1208: indicate the bulk of those SNe formed relatively recently, one might
1209: expect a tighter or different correlation between metallicity and
1210: galaxy mass than for those in passive galaxies. Unfortunately,
1211: Fig.~\ref{fig:col_mass} reveals an unavoidable difficulty with the
1212: modest Keck sample; our four passive galaxies are considerably more
1213: massive than most of the star-forming galaxies. Thus it is not
1214: straightforward to separate the effects of galaxy mass and specific
1215: star formation rate in understanding the implied metallicity
1216: dispersion. Nonetheless, to the extent we can examine any
1217: relationships, it seems there are much stronger trends between $UV1-b$
1218: and inferred metallicity for SNe~Ia in star-forming hosts than for
1219: those with passive hosts. At face value this would imply the most
1220: metal-rich progenitors have {\it less} UV blanketing. Clearly a larger
1221: sample will be needed to verify these trends. It may not be
1222: straightforward to sample lower-mass passive hosts, given fewer
1223: passive hosts are expected at high redshift in the hypothesis proposed
1224: by Sullivan et al..
1225:
1226: A more informative comparison of the UV trends can be obtained by
1227: examining the spectra for those SNe~Ia with high and low stretch.
1228: Although the host galaxy spectral class may contain events with a
1229: range of stretch (c.f. fig.~11 of \citet{2006ApJ...648..868S}), such a
1230: categorization of events may be a more direct way of examining the
1231: spectral properties of events in old and young stellar populations
1232: \citep[see also][]{2007bronder}.
1233:
1234: Fig.~\ref{fig:mean_max_s} shows the mean maximum-light SALT-law
1235: color-corrected spectra for those events with stretch s$>$1.03 and
1236: s$\leq$1.03. We also show the mean spectra split by $s$ in the same
1237: way when the spectra are color-matched (rather than color-corrected)
1238: to have the same $U-b$ color. These comparisons confirm in detail the
1239: photometric trend noted in the rightmost panel of
1240: Fig.~\ref{fig:col_stretch} -- a strong and systematic difference in
1241: the far UV maximum-light spectra of low and high-$s$ SNe~Ia.
1242:
1243: As stretch (or light curve shape) maps closely onto both
1244: star-formation activity in the host galaxy \citep{2006ApJ...648..868S}
1245: and the morphology of the host galaxy
1246: \citep[e.g.][]{2000AJ....120.1479H}, this comparison provides a clear
1247: example of the physical differences between SNe~Ia in star forming and
1248: passive hosts and hence SNe~Ia drawn from young and old stellar
1249: populations.
1250:
1251: Clearly the mean spectrum of a SN~Ia in a passive host with an old
1252: stellar population has a depressed far-UV continuum compared to one in
1253: a more active host with a younger stellar population even following
1254: color corrections based on optical colors. A broadly similar trend is
1255: observed by \citet{2007A&A...466...11G}. This is in the opposite sense
1256: than that expected from dust extinction, where the higher-stretch
1257: SNe~Ia residing in spiral galaxies should suffer more extinction and
1258: therefore appear redder in the UV. Furthermore, the far-UV colors
1259: appear more sensitive to stretch than the optical colors.
1260:
1261: Fig.~\ref{fig:mean_max_s} also reveals interesting differences between
1262: the optical spectral regions of high and low stretch supernovae. The
1263: majority of these differences originate due to the underlying
1264: temperature difference, with the high stretch supernovae showing
1265: stronger high excitation lines and ones from doubly-ionized elements.
1266: At $\simeq$4400 \AA, a line due to \ion{Si}{3} 4552 \AA\ is found in
1267: the high stretch supernovae but is absent in the lower stretch
1268: examples. In addition, the 3700 \AA\ Ca+Si feature is narrower and
1269: shows a greater degree of splitting into the separate Ca and Si
1270: components in the high stretch composite spectrum. Here one can also
1271: clearly see the difference between $\cal{R}$(\ion{Ca}{2}) as defined
1272: in \citet{nugseq95} where the ratio of the emission features on either
1273: side of this trough are greater in the low stretch composite. Finally,
1274: the \ion{Si}{2} feature at 4130\AA\ is shallower in high-stretch SNe
1275: consistent with trends identified by \citet{2007bronder}.
1276:
1277: \section{Discussion}
1278: \label{sec:discussion}
1279:
1280: One of the original motivations for this study was to explore both
1281: evolution in the mean properties of SNe~Ia over 0$<z<$0.5 and the
1282: dispersion in their UV spectra. We have seen remarkably little
1283: evidence of evolution in the mean spectroscopic properties over this
1284: interval, confirming, with higher precision, earlier suggestions based
1285: largely on lower-S/N data. The primary limitation in this aspect of
1286: our paper is the paucity of high quality UV data at low redshift.
1287:
1288: Most of the earlier constraints on the possible evolution of SNe~Ia
1289: were largely based on consideration of photometric measures. For
1290: example, \citet{2004ApJ...607..665R} derived the rest-frame $U-B$ and
1291: $B-V$ distributions of a heterogeneous sample of SNe~Ia at maximum
1292: light over z$<$1.5 as a proxy for the more precise constraints
1293: possible with spectroscopic data. They ruled out any color evolution
1294: greater than 0.02~mag in the mean in $U-B$.
1295:
1296: Spectroscopic data is substantially more precise in its tracking of
1297: evolution and dispersion than broad-band photometry for two reasons.
1298: Foremost, detailed differences in key line diagnostics are lost in
1299: photometric data. Secondly, no assumptions need be made about
1300: $k$-corrections in gathering and comparing data over a broad redshift
1301: range.
1302:
1303: At the time this Keck survey was underway, there was no comprehensive
1304: spectroscopic dataset against which our UV data could be compared,
1305: other than the mean template published by \citet{2002PASP..114..803N}
1306: and discussed in $\S$~\ref{sec:local-data}. However, very recently, as
1307: a result of a survey of high-redshift SNe~Ia with the ACS grism
1308: onboard Hubble Space Telescope, \citet{2007ApJ...659...98R} have
1309: published the mean rest-frame UV spectrum of 13 SNe~Ia more distant
1310: ($z>1$) than those studied here; the average redshift of this sample
1311: is $\overline{z}$=1.3. In this study, used to demonstrate the presence
1312: of dark energy at $z>$1, it is claimed that the sample-averaged
1313: spectral energy distributions observed at $z>$1 are consistent with
1314: that observed locally and that any spectral evolution is still
1315: undetected. However, no quantitative statement is made to support
1316: this claim.
1317:
1318: We believe that the spectral comparison undertaken by
1319: \citet{2007ApJ...659...98R} is less precise than that undertaken in
1320: this study, and hence less valuable as a means of justifying the
1321: continued use of local relations in constructing SN~Ia Hubble
1322: diagrams. Foremost, the ACS grism data samples SNe~Ia over a much
1323: wider range of phase than the comparisons undertaken here. Although
1324: the individual spectra comprising the mean $z$=1.3 ACS spectrum are
1325: not tabulated by \citet{2007ApJ...659...98R}, their phases range over
1326: at least 15-20 days and almost all are post-maximum light, compared
1327: with the narrow phase range maximum-light (-4 to +4 days) and
1328: pre-maximum light ($<$-4 days) comparisons presented here. A second
1329: consideration is the inevitable poor signal-to-noise ratio of these
1330: very high redshift spectra which precludes detailed comparisons.
1331:
1332: As we have shown in Figs.~\ref{fig:mean_max} and
1333: ~\ref{fig:mean_early}, the local template
1334: \citep{2002PASP..114..803N} is a poor basis from which to make such
1335: comparisons. More representative datasets are needed for reliable
1336: claims. Internally within our own data, Fig.~\ref{fig:mean_max_z}
1337: reveals little evolution although the redshift baseline is small. It
1338: will be important to examine the case for evolution from 0.2$<z<$1.5
1339: by combining in a consistent manner both the
1340: \citet{2007ApJ...659...98R} dataset with that presented here, as well
1341: as with higher-redshift SNLS SNe~Ia (out to $z\sim1$) observed during
1342: the routine survey spectroscopic screening
1343: \citep[e.g.][]{2007bronder}.
1344:
1345: The second significant finding in this study is the marked increase in
1346: the dispersion among our SNe~Ia shortward of 3300\AA\
1347: (Fig.~\ref{fig:col_stretch}). Although the scatter in $U$ is
1348: comparable to that seen in local data, it increases significantly at
1349: shorter wavelengths. Even allowing for differing amounts of dust
1350: extinction within each host galaxy, this amounts to more than a factor
1351: of 2 variation in continuum flux at maximum light. Although
1352: theoretical models predict a strong sensitivity to metallicity
1353: variations at this wavelength, our variations are also considerably
1354: larger than those predicted. We have demonstrated that a color
1355: correction based on the $B-V$ color of the SN and either a CCM Milky
1356: Way extinction law or a SALT color law, can only marginally reduce, and
1357: not eliminate, this UV dispersion.
1358:
1359: We have found it hard to isolate the physical causes of this
1360: significant UV dispersion. We confirm earlier work at longer
1361: wavelengths ($\lambda \simeq$3500-4000\AA) that shows that stretch
1362: (or equivalently host galaxy class) is partially responsible. However
1363: at the shortest wavelengths ($\lambda\simeq$3000-3300\AA\
1364: corresponding to our UV1 diagnostic), additional effects are clearly
1365: important that are not accounted for by the color-correction
1366: techniques in use in current cosmological programs. In addition to the
1367: newly-found intrinsic scatter at short wavelengths, new trends with
1368: phase are also seen in the wavelengths of diagnostic features in this
1369: region.
1370:
1371: Redward of 4000 \AA\ , corrections for stretch or color work well in
1372: normalizing SNe Ia, however blueward of 4000A, significant scatter
1373: remains even after such corrections are made. This may be related to
1374: a change in the dominant source of opacity in SNe\,Ia. Redward of
1375: 3500-4000 \AA\, electron scattering opacity dominates, but at UV
1376: wavelengths a forest of overlapping lines is the dominant source of
1377: opacity (see Fig. 1 of \cite{2000ARA&A...38..191H}).
1378: Electron scattering is a continuous process involving well-understood
1379: physics, but line opacity depends sensitively on abundances,
1380: ionization states, and possibly non-LTE effects.
1381:
1382: Fig. ~\ref{fig:opac_max} illustrates this point by comparing the wavelength
1383: dependence of the line and electron scattering opacity at maximum light
1384: for a model by \cite{2007ApJ...656..661K} that provides a good match
1385: for a normal SN~Ia (Kason, priv. commun.). The data in question refers
1386: to that at a depth of 7000 km\,s$^{-1}$ where intermediate mass and
1387: Fe peak material are well-mixed. A common feature in these models is the
1388: drop in line opacity compared to the electron scattering opacity near
1389: 4000 \AA\ as seen here. Thus it is understandable that the emerging
1390: UV flux is highly susceptible to changes in the line opacity (due to initial
1391: conditions and/or material synthesized during the explosion) while the
1392: optical and near-IR spectral behavior are dominated by electron
1393: scattering opacity at this phase.
1394:
1395: What are the possible consequences of the above variations in terms of
1396: the use of SNe~Ia as probes of the expansion history? In the highest
1397: redshift surveys, including those proposed with future facilities,
1398: cross-color $k$-corrections are needed to estimate rest-frame light
1399: curves from the observations, typically undertaken in the far-red and
1400: near-infrared for $z>$1.5 \citep{2004ApJ...607..665R,
1401: 2007ApJ...659...98R}. This necessitates the adoption of a suitable
1402: template whose SED is reliable in the 3000-4000\AA\ region. At the
1403: most fundamental level, the dispersion in our UV spectra will
1404: contribute a statistical uncertainty at the 0.05-0.1 magnitude level
1405: depending on how precisely particular observed filters match to the
1406: chosen rest-frame bandpasses.
1407:
1408: A more worrying bias would follow the adoption of an incorrect
1409: template. Possibly the most significant finding in our work is the
1410: demonstration that the mean UV spectrum is different for high and low
1411: stretch SNe~Ia (Fig.~\ref{fig:mean_max_s}) and thus, presumably, for
1412: those that occur in passive and actively star-forming galaxies
1413: \citep{2006ApJ...648..868S}. Adopting a single template would lead to
1414: a systematic bias which could become increasingly serious at high
1415: redshift where the mix between the two populations changes and
1416: high-stretch SNe, typically originating from younger progenitor
1417: systems, become increasingly common. Recently,
1418: \citet{2007astro.ph..1912H} have demonstrated the existence of an
1419: increased fraction of high-stretch SNe~Ia at higher redshifts above
1420: and beyond the selection effect that high stretch, brighter SNe~Ia are
1421: easier to detect.
1422:
1423: Using our individual maximum light spectra, we can evaluate the impact of
1424: the UV dispersion on sample cross-color $k$-corrections, addressing
1425: both the statistical error arising from the intrinsic scatter and the
1426: systematic difference arising from the stretch-based spectral
1427: differences we have found. Fig.~\ref{fig:kcorrections} shows the
1428: result.
1429:
1430: For a typical future dark energy experiment, based on securing
1431: the equation of state parameter $w$ to 5\% using $z>$1 SNe\,Ia,
1432: photometric corrections of better than $\pm$0.02 mag are required
1433: \footnote{e.g. White Papers submitted to the
1434: US Dark Energy Task Force Committee:
1435: {\rm http://www.nsf.gov/mps/ast/detf.jsp}}. This requirement is
1436: indicated on Fig.~\ref{fig:kcorrections}.
1437:
1438: In terms of the {\em statistical} error, the observed dispersion
1439: is typically $\pm$0.05-0.10 mag, larger than that required by a factor of
1440: several. However, such a dispersion, if randomly distributed, need not
1441: present a fundamental obstacle to progress with a survey spanning a
1442: wide redshift range utilizing a large number of SNe\, Ia. As
1443: Fig.~\ref{fig:kcorrections} shows, at certain redshifts where the rest-frame
1444: and observed filters closely match, the dispersion has negligible effect
1445: and SNe\,Ia at these redshifts could be more heavily weighted.
1446:
1447: The more worrying trend, particularly given the recent demonstration
1448: of stretch bias by \citet{2007astro.ph..1912H}, is the likelihood of a
1449: {\em systematic} error introduced by adopting an incorrect template for
1450: the redshifts where the rest-frame and observed filters do not overlap.
1451: Fig.~\ref{fig:kcorrections} shows the potential of this error via a
1452: comparison of the dispersion independently for high and low stretch events.
1453: The differences between these two categories are comparably large and
1454: indicate the importance of securing a physical understanding of the
1455: UV variations seen in our survey.
1456:
1457: \section{Conclusions}
1458: \label{sec: conclusions}
1459:
1460: We summarize our findings as follows:
1461:
1462: \begin{enumerate}
1463:
1464: \item{} We have secured high signal-to-noise ratio Keck spectra for a
1465: sample of 36 intermediate redshift SNe~Ia, observed at various
1466: phases, spanning the redshift range 0.15$<z<$0.7, and drawn from the
1467: Supernova Legacy Survey (SNLS). We demonstrate via inspection of the
1468: SN properties that our Keck sample is a reasonably fair subset of
1469: the larger sample of distant SNe~Ia being studied by the SNLS.
1470:
1471: \item{} We develop a new method for removing host galaxy contamination
1472: from our spectra based on measures of the galaxy and SN photometry.
1473: These refinements to traditional spectral reduction techniques allow
1474: us to achieve host-galaxy subtracted and flux-calibrated rest-frame
1475: spectra of high quality, extending down to rest-frame wavelengths of
1476: 2900\AA.
1477:
1478: \item{} Although no strong evidence is found for spectral evolution in
1479: the mean early-phase and maximum light spectra, when compared to
1480: local data, such evolutionary tests are hampered by the paucity of
1481: quality data at low redshift and a significant scatter in the
1482: spectra shortward of 4000\AA. We find no evidence for evolution
1483: internal to our data. We argue that the well-used local UV spectral
1484: template \citep{2002PASP..114..803N} is likely to be less
1485: representative than the mean spectrum compiled from the Keck data
1486: which we tabulate with the measured dispersion for use in future
1487: cosmological applications.
1488:
1489: \item{} Our principal finding is a large scatter from one SN to the
1490: next in the rest-frame UV spectrum even after standard dust
1491: corrections are made. By constructing various photometric bandpasses
1492: that avoid uncertainties arising from differential $k$-corrections
1493: associated with the range of redshifts in our sample, we show that
1494: while we can reproduce the stretch-dependent trends seen locally at
1495: 3500-4000\AA, the scatter at 3000-3400\AA\ is 3-5 times larger.
1496:
1497: \item{} Although progenitor metallicity may drive some of the trends
1498: seen in the Keck data, the UV variations are much larger than in
1499: contemporary models which span the expected metallicity range.
1500: Moreover, the UV spectrum also changes with phase in a manner which
1501: is not consistent with models. We conclude there are significant
1502: variations in the UV properties of SNe~Ia which are not accounted for
1503: by either the presently-employed empirical trends or the available
1504: SN~Ia models.
1505:
1506: \item{} As an illustration of the importance of understanding these
1507: new results, we calculate the error arising from the use of a single
1508: UV spectral template for calculating the cross-color $k$ correction,
1509: a correction essential for constructing the SN~Ia Hubble diagram as
1510: a probe of the expansion history. The dispersion arising from our UV
1511: spectra, if not randomly distributed along the Hubble diagram, presents
1512: an uncertainty 2-3 times larger than than would be
1513: necessary for recovering the equation of state parameter $w$ to 5\%
1514: using SNe~Ia at $z\simeq$1. We conclude that further detailed
1515: studies are essential if SNe~Ia are to be useful for precision
1516: measures of dark energy.
1517:
1518:
1519: \end{enumerate}
1520:
1521: \acknowledgments
1522:
1523: The spectroscopic data presented herein were obtained at the W.M. Keck
1524: Observatory, which is operated as a scientific partnership among the
1525: California Institute of Technology, the University of California and
1526: the National Aeronautics and Space Administration. The Observatory was
1527: made possible by the generous financial support of the W.M. Keck
1528: Foundation. The authors wish to recognize and acknowledge the very
1529: significant cultural role and reverence that the summit of Mauna Kea
1530: has always had within the indigenous Hawaiian community. We are most
1531: fortunate to have the opportunity to conduct observations from this
1532: mountain. Additional observations were obtained with
1533: MegaPrime/MegaCam, a joint project of CFHT and CEA/DAPNIA, at the
1534: Canada-France-Hawaii Telescope which is operated by the National
1535: Research Council (NRC) of Canada, the Institut National des Sciences de
1536: l'Univers of the Centre National de la Recherche Scientifique (CNRS)
1537: of France, and the University of Hawaii. Canadian collaboration
1538: members acknowledge support from NSERC and CIAR; French collaboration
1539: members from CNRS/IN2P3, CNRS/INSU and CEA. PEN acknowledges support
1540: from the US Department of Energy Scientific Discovery through Advanced
1541: Computing program under contract DE-FG02-06ER06-04. A.G. acknowledges
1542: support by NASA through Hubble Fellowship grant \#HST-HF-01158.01-A
1543: awarded by STScI, which is operated by AURA, Inc., for NASA, under
1544: contract NAS 5-26555. RSE acknowledges financial support from DOE
1545: under contract DE-FG02-04ER41316. This research used resources of the
1546: National Energy Research Scientific Computing Center, which is
1547: supported by the Office of Science of the U.S. Department of Energy
1548: under Contract No. DE-AC03-76SF00098. We thank them for a generous
1549: allocation of computing time. We acknowledge useful discussions with
1550: Adam Riess, Mario Livio, and Ken Nomoto.
1551:
1552: {\it Facilities:} \facility{Keck:I}, \facility{CFHT}.
1553:
1554: \begin{thebibliography}{82}
1555: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
1556:
1557: \bibitem[{{Aldering}(2005)}]{2005NewAR..49..346A}
1558: {Aldering}, G. 2005, New Astronomy Review, 49, 346
1559:
1560: \bibitem[{{Aldering} {et~al.}(2002){Aldering}, {Adam}, {Antilogus}, {Astier},
1561: {Bacon}, {Bongard}, {Bonnaud}, {Copin}, {Hardin}, {Henault}, {Howell},
1562: {Lemonnier}, {Levy}, {Loken}, {Nugent}, {Pain}, {Pecontal}, {Pecontal},
1563: {Perlmutter}, {Quimby}, {Schahmaneche}, {Smadja}, \&
1564: {Wood-Vasey}}]{2002SPIE.4836...61A}
1565: {Aldering}, G., et al.
1566: 2002, in Survey and Other Telescope Technologies and Discoveries. Edited by
1567: Tyson, J. Anthony; Wolff, Sidney. Proceedings of the SPIE, Volume 4836, pp.
1568: 61-72 (2002)., 61--72
1569:
1570: \bibitem[{{Astier} {et~al.}(2006){Astier}, {Guy}, {Regnault}, {Pain},
1571: {Aubourg}, {Balam}, {Basa}, {Carlberg}, {Fabbro}, {Fouchez}, {Hook},
1572: {Howell}, {Lafoux}, {Neill}, {Palanque-Delabrouille}, {Perrett}, {Pritchet},
1573: {Rich}, {Sullivan}, {Taillet}, {Aldering}, {Antilogus}, {Arsenijevic},
1574: {Balland}, {Baumont}, {Bronder}, {Courtois}, {Ellis}, {Filiol}, {Gon{\c
1575: c}alves}, {Goobar}, {Guide}, {Hardin}, {Lusset}, {Lidman}, {McMahon},
1576: {Mouchet}, {Mourao}, {Perlmutter}, {Ripoche}, {Tao}, \&
1577: {Walton}}]{2006A&A...447...31A}
1578: {Astier}, P., et al. 2006, \aap, 447, 31
1579:
1580: \bibitem[{{Benford} \& {Lauer}(2006)}]{2006SPIE.6265E..67B}
1581: {Benford}, D.~J., \& {Lauer}, T.~R. 2006, in Presented at the Society of
1582: Photo-Optical Instrumentation Engineers (SPIE) Conference, Vol. 6265, Space
1583: Telescopes and Instrumentation I: Optical, Infrared, and Millimeter. Edited
1584: by Mather, John C.; MacEwen, Howard A.; de Graauw, Mattheus W. M..
1585: Proceedings of the SPIE, Volume 6265, pp. 626528 (2006).
1586:
1587: \bibitem[{{Bessell}(1990)}]{1990PASP..102.1181B}
1588: {Bessell}, M.~S. 1990, \pasp, 102, 1181
1589:
1590: \bibitem[{{Blondin} {et~al.}(2006){Blondin}, {Dessart}, {Leibundgut}, {Branch},
1591: {H{\"o}flich}, {Tonry}, {Matheson}, {Foley}, {Chornock}, {Filippenko},
1592: {Sollerman}, {Spyromilio}, {Kirshner}, {Wood-Vasey}, {Clocchiatti},
1593: {Aguilera}, {Barris}, {Becker}, {Challis}, {Covarrubias}, {Davis},
1594: {Garnavich}, {Hicken}, {Jha}, {Krisciunas}, {Li}, {Miceli}, {Miknaitis},
1595: {Pignata}, {Prieto}, {Rest}, {Riess}, {Salvo}, {Schmidt}, {Smith}, {Stubbs},
1596: \& {Suntzeff}}]{2006AJ....131.1648B}
1597: {Blondin}, S., et al. 2006, \aj, 131, 1648
1598:
1599: \bibitem[{{Boulade} {et~al.}(2003){Boulade}, {Charlot}, {Abbon}, {Aune},
1600: {Borgeaud}, {Carton}, {Carty}, {Da Costa}, {Deschamps}, {Desforge},
1601: {Eppell{\'e}}, {Gallais}, {Gosset}, {Granelli}, {Gros}, {de Kat}, {Loiseau},
1602: {Ritou}, {Rouss{\'e}}, {Starzynski}, {Vignal}, \&
1603: {Vigroux}}]{2003SPIE.4841...72B}
1604: {Boulade}, O., et al. 2003, in Instrument
1605: Design and Performance for Optical/Infrared Ground-based Telescopes. Edited
1606: by Iye, Masanori; Moorwood, Alan F. M. Proceedings of the SPIE, Volume 4841,
1607: pp. 72-81 (2003)., 72--81
1608:
1609: \bibitem[{{Branch} {et~al.}(2006){Branch}, {Dang}, {Hall}, {Ketchum},
1610: {Melakayil}, {Parrent}, {Troxel}, {Casebeer}, {Jeffery}, \&
1611: {Baron}}]{2006PASP..118..560B}
1612: {Branch}, D., et al. 2006, \pasp, 118, 560
1613:
1614: \bibitem[{{Bronder} {et~al.}(2007){Bronder}, {Hook}, {Astier}, {Balam}, {Basa},
1615: {Carlberg}, {Conley}, {Fouchez}, {Guy}, \& {Howell}}]{2007bronder}
1616: {Bronder}, J.~B., et al.
1617: 2007, in \aap, accepted
1618:
1619: \bibitem[{{Cappellaro} {et~al.}(1995){Cappellaro}, {Turatto}, \&
1620: {Fernley}}]{1995ESASP1189.....C}
1621: {Cappellaro}, E., {Turatto}, M., \& {Fernley}, J. 1995, ESA SCIENTIFIC
1622: PUBLICATION ESA-SP 1189, 1189
1623:
1624: \bibitem[{{Cardelli} {et~al.}(1989){Cardelli}, {Clayton}, \&
1625: {Mathis}}]{1989ApJ...345..245C}
1626: {Cardelli}, J.~A., {Clayton}, G.~C., \& {Mathis}, J.~S. 1989, \apj, 345, 245
1627:
1628: \bibitem[{{Coil} {et~al.}(2000){Coil}, {Matheson}, {Filippenko}, {Leonard},
1629: {Tonry}, {Riess}, {Challis}, {Clocchiatti}, {Garnavich}, {Hogan}, {Jha},
1630: {Kirshner}, {Leibundgut}, {Phillips}, {Schmidt}, {Schommer}, {Smith},
1631: {Soderberg}, {Spyromilio}, {Stubbs}, {Suntzeff}, \&
1632: {Woudt}}]{2000ApJ...544L.111C}
1633: {Coil}, A.~L., et al.
1634: 2000, \apjl, 544, L111
1635:
1636: \bibitem[{{Cole} {et~al.}(2005){Cole}, {Percival}, {Peacock}, {Norberg},
1637: {Baugh}, {Frenk}, {Baldry}, {Bland-Hawthorn}, {Bridges}, {Cannon}, {Colless},
1638: {Collins}, {Couch}, {Cross}, {Dalton}, {Eke}, {De Propris}, {Driver},
1639: {Efstathiou}, {Ellis}, {Glazebrook}, {Jackson}, {Jenkins}, {Lahav}, {Lewis},
1640: {Lumsden}, {Maddox}, {Madgwick}, {Peterson}, {Sutherland}, \&
1641: {Taylor}}]{2005MNRAS.362..505C}
1642: {Cole}, S., et al. 2005, \mnras, 362, 505
1643:
1644: \bibitem[{{Conley} {et~al.}(2007){Conley}, {Carlberg}, {Guy}, {Howell}, {Jha},
1645: {Riess}, \& {Sullivan}}]{2007arXiv0705.0367C}
1646: {Conley}, A., {Carlberg}, R.~G., {Guy}, J., {Howell}, D.~A., {Jha}, S.,
1647: {Riess}, A.~G., \& {Sullivan}, M. 2007, ArXiv e-prints, 705
1648:
1649: \bibitem[{{Efstathiou} {et~al.}(2002){Efstathiou}, {Moody}, {Peacock},
1650: {Percival}, {Baugh}, {Bland-Hawthorn}, {Bridges}, {Cannon}, {Cole},
1651: {Colless}, {Collins}, {Couch}, {Dalton}, {de Propris}, {Driver}, {Ellis},
1652: {Frenk}, {Glazebrook}, {Jackson}, {Lahav}, {Lewis}, {Lumsden}, {Maddox},
1653: {Norberg}, {Peterson}, {Sutherland}, \& {Taylor}}]{2002MNRAS.330L..29E}
1654: {Efstathiou}, G., et al. 2002, \mnras, 330, L29
1655:
1656: \bibitem[{{Eisenstein} {et~al.}(2005){Eisenstein}, {Zehavi}, {Hogg},
1657: {Scoccimarro}, {Blanton}, {Nichol}, {Scranton}, {Seo}, {Tegmark}, {Zheng},
1658: {Anderson}, {Annis}, {Bahcall}, {Brinkmann}, {Burles}, {Castander},
1659: {Connolly}, {Csabai}, {Doi}, {Fukugita}, {Frieman}, {Glazebrook}, {Gunn},
1660: {Hendry}, {Hennessy}, {Ivezi{\'c}}, {Kent}, {Knapp}, {Lin}, {Loh}, {Lupton},
1661: {Margon}, {McKay}, {Meiksin}, {Munn}, {Pope}, {Richmond}, {Schlegel},
1662: {Schneider}, {Shimasaku}, {Stoughton}, {Strauss}, {SubbaRao}, {Szalay},
1663: {Szapudi}, {Tucker}, {Yanny}, \& {York}}]{2005ApJ...633..560E}
1664: {Eisenstein}, D.~J., et al. 2005, \apj, 633, 560
1665:
1666: \bibitem[{{Filippenko}(1982)}]{1982PASP...94..715F}
1667: {Filippenko}, A.~V. 1982, \pasp, 94, 715
1668:
1669: \bibitem[{{Filippenko}(1997)}]{1997ARA&A..35..309F}
1670: ---. 1997, \araa, 35, 309
1671:
1672: \bibitem[{{Filippenko} {et~al.}(1992){Filippenko}, {Richmond}, {Matheson},
1673: {Shields}, {Burbidge}, {Cohen}, {Dickinson}, {Malkan}, {Nelson}, {Pietz},
1674: {Schlegel}, {Schmeer}, {Spinrad}, {Steidel}, {Tran}, \&
1675: {Wren}}]{1992ApJ...384L..15F}
1676: {Filippenko}, A.~V., et al. 1992, \apjl, 384, L15
1677:
1678: \bibitem[{{Fioc} \& {Rocca-Volmerange}(1997)}]{1997A&A...326..950F}
1679: {Fioc}, M., \& {Rocca-Volmerange}, B. 1997, \aap, 326, 950
1680:
1681: \bibitem[{{Fioc} \& {Rocca-Volmerange}(1999)}]{1999astro.ph.12179F}
1682: ---. 1999, ArXiv Astrophysics e-prints
1683:
1684: \bibitem[{{Gallagher} {et~al.}(2005){Gallagher}, {Garnavich}, {Berlind},
1685: {Challis}, {Jha}, \& {Kirshner}}]{2005ApJ...634..210G}
1686: {Gallagher}, J.~S., {Garnavich}, P.~M., {Berlind}, P., {Challis}, P., {Jha},
1687: S., \& {Kirshner}, R.~P. 2005, \apj, 634, 210
1688:
1689: \bibitem[{{Garavini} {et~al.}(2007){Garavini}, {Folatelli}, {Nobili},
1690: {Aldering}, {Amanullah}, {Antilogus}, {Astier}, {Blanc}, {Bronder}, {Burns},
1691: {Conley}, {Deustua}, {Doi}, {Fabbro}, {Fadeyev}, {Gibbons}, {Goldhaber},
1692: {Goobar}, {Groom}, {Hook}, {Howell}, {Kashikawa}, {Kim}, {Kowalski},
1693: {Kuznetsova}, {Lee}, {Lidman}, {Mendez}, {Morokuma}, {Motohara}, {Nugent},
1694: {Pain}, {Perlmutter}, {Quimby}, {Raux}, {Regnault}, {Ruiz-Lapuente},
1695: {Sainton}, {Schahmaneche}, {Smith}, {Spadafora}, {Stanishev}, {Thomas},
1696: {Walton}, {Wang}, {Wood-Vasey}, \& {Yasuda}}]{2007A&A...470..411G}
1697: {Garavini}, G., et al. 2007, \aap, 470, 411
1698:
1699: \bibitem[{{Garnavich} {et~al.}(2004){Garnavich}, {Bonanos}, {Krisciunas},
1700: {Jha}, {Kirshner}, {Schlegel}, {Challis}, {Macri}, {Hatano}, {Branch},
1701: {Bothun}, \& {Freedman}}]{2004ApJ...613.1120G}
1702: {Garnavich}, P.~M., et al. 2004, \apj, 613, 1120
1703:
1704: \bibitem[{{Guy} {et~al.}(2007){Guy}, {Astier}, {Baumont}, {Hardin}, {Pain},
1705: {Regnault}, {Basa}, {Carlberg}, {Conley}, {Fabbro}, {Fouchez}, {Hook},
1706: {Howell}, {Perrett}, {Pritchet}, {Rich}, {Sullivan}, {Antilogus}, {Aubourg},
1707: {Bazin}, {Bronder}, {Filiol}, {Palanque-Delabrouille}, {Ripoche}, \&
1708: {Ruhlmann-Kleider}}]{2007A&A...466...11G}
1709: {Guy}, J., et al. 2007, \aap, 466, 11
1710:
1711: \bibitem[{{Guy} {et~al.}(2005){Guy}, {Astier}, {Nobili}, {Regnault}, \&
1712: {Pain}}]{2005A&A...443..781G}
1713: {Guy}, J., {Astier}, P., {Nobili}, S., {Regnault}, N., \& {Pain}, R. 2005,
1714: \aap, 443, 781
1715:
1716: \bibitem[{{H{\" o}flich} {et~al.}(1998){H{\" o}flich}, {Wheeler}, \&
1717: {Thielemann}}]{1998ApJ...495..617H}
1718: {H{\" o}flich}, P., {Wheeler}, J.~C., \& {Thielemann}, F.~K. 1998, \apj, 495,
1719: 617
1720:
1721: \bibitem[{{Hamuy} {et~al.}(2006){Hamuy}, {Folatelli}, {Morrell}, {Phillips},
1722: {Suntzeff}, {Persson}, {Roth}, {Gonzalez}, {Krzeminski}, {Contreras},
1723: {Freedman}, {Murphy}, {Madore}, {Wyatt}, {Maza}, {Filippenko}, {Li}, \&
1724: {Pinto}}]{2006PASP..118....2H}
1725: {Hamuy}, M., et al.
1726: 2006, \pasp, 118, 2
1727:
1728: \bibitem[{{Hamuy} {et~al.}(1995){Hamuy}, {Phillips}, {Maza}, {Suntzeff},
1729: {Schommer}, \& {Aviles}}]{1995AJ....109....1H}
1730: {Hamuy}, M., {Phillips}, M.~M., {Maza}, J., {Suntzeff}, N.~B., {Schommer},
1731: R.~A., \& {Aviles}, R. 1995, \aj, 109, 1
1732:
1733: \bibitem[{{Hamuy} {et~al.}(2000){Hamuy}, {Trager}, {Pinto}, {Phillips},
1734: {Schommer}, {Ivanov}, \& {Suntzeff}}]{2000AJ....120.1479H}
1735: {Hamuy}, M., {Trager}, S.~C., {Pinto}, P.~A., {Phillips}, M.~M., {Schommer},
1736: R.~A., {Ivanov}, V., \& {Suntzeff}, N.~B. 2000, \aj, 120, 1479
1737:
1738: \bibitem[{{Hillebrandt} \& {Niemeyer}(2000){Hillebrandt} \& {Niemeyer}}]
1739: {2000ARA&A...38..191H}{Hillebrandt}, W. \& {Niemeyer}, J.-C. 2000,
1740: \araa, 38, 191
1741:
1742: \bibitem[{{Hook} {et~al.}(2005){Hook}, {Howell}, {Aldering}, {Amanullah},
1743: {Burns}, {Conley}, {Deustua}, {Ellis}, {Fabbro}, {Fadeyev}, {Folatelli},
1744: {Garavini}, {Gibbons}, {Goldhaber}, {Goobar}, {Groom}, {Kim}, {Knop},
1745: {Kowalski}, {Lidman}, {Nobili}, {Nugent}, {Pain}, {Pennypacker},
1746: {Perlmutter}, {Ruiz-Lapuente}, {Sainton}, {Schaefer}, {Smith}, {Spadafora},
1747: {Stanishev}, {Thomas}, {Walton}, {Wang}, \&
1748: {Wood-Vasey}}]{2005AJ....130.2788H}
1749: {Hook}, I.~M., et al. 2005, \aj, 130, 2788
1750:
1751: \bibitem[{{Howell}(2001)}]{2001ApJ...554L.193H}
1752: {Howell}, D.~A. 2001, \apjl, 554, L193
1753:
1754: \bibitem[{{Howell} {et~al.}(2007){Howell}, {Sullivan}, {Conley}, \&
1755: {Carlberg}}]{2007astro.ph..1912H}
1756: {Howell}, D.~A., {Sullivan}, M., {Conley}, A., \& {Carlberg}, R. 2007, ArXiv
1757: Astrophysics e-prints
1758:
1759: \bibitem[{{Howell} {et~al.}(2006){Howell}, {Sullivan}, {Nugent}, {Ellis},
1760: {Conley}, {Le Borgne}, {Carlberg}, {Guy}, {Balam}, {Basa}, {Fouchez}, {Hook},
1761: {Hsiao}, {Neill}, {Pain}, {Perrett}, \& {Pritchet}}]{2006Natur.443..308H}
1762: {Howell}, D.~A., et al. 2006, \nat, 443, 308
1763:
1764: \bibitem[{{Howell} {et~al.}(2005){Howell}, {Sullivan}, {Perrett}, {Bronder},
1765: {Hook}, {Astier}, {Aubourg}, {Balam}, {Basa}, {Carlberg}, {Fabbro},
1766: {Fouchez}, {Guy}, {Lafoux}, {Neill}, {Pain}, {Palanque-Delabrouille},
1767: {Pritchet}, {Regnault}, {Rich}, {Taillet}, {Knop}, {McMahon}, {Perlmutter},
1768: \& {Walton}}]{2005ApJ...634.1190H}
1769: {Howell}, D.~A., et al. 2005, \apj, 634, 1190
1770:
1771: \bibitem[{{Howell} {et~al.}(2002){Howell}, {Wang}, {Supernova Cosmology Project
1772: Collaboration}, \& {Supernova Factory Collaboration}}]{2002AAS...201.9103H}
1773: {Howell}, D.~A., {Wang}, L., {Supernova Cosmology Project Collaboration}, \&
1774: {Supernova Factory Collaboration}. 2002, Bulletin of the American
1775: Astronomical Society, 34, 1256
1776:
1777: \bibitem[{{Hsiao} {et~al.}(2007){Hsiao}, {Conley}, {Howell}, {Sullivan},
1778: {Pritchet}, {Carlberg}, {Nugent}, \& {Phillips}}]{2007ApJ...663.1187H}
1779: {Hsiao}, E.~Y., {Conley}, A., {Howell}, D.~A., {Sullivan}, M., {Pritchet},
1780: C.~J., {Carlberg}, R.~G., {Nugent}, P.~E., \& {Phillips}, M.~M. 2007, \apj,
1781: 663, 1187
1782:
1783: \bibitem[{{Iwamoto} {et~al.}(1999){Iwamoto}, {Brachwitz}, {Nomoto},
1784: {Kishimoto}, {Umeda}, {Hix}, \& {Thielemann}}]{1999ApJS..125..439I}
1785: {Iwamoto}, K., {Brachwitz}, F., {Nomoto}, K., {Kishimoto}, N., {Umeda}, H.,
1786: {Hix}, W.~R., \& {Thielemann}, F.-K. 1999, \apjs, 125, 439
1787:
1788: \bibitem[{{Jha} {et~al.}(2006){Jha}, {Kirshner}, {Challis}, {Garnavich},
1789: {Matheson}, {Soderberg}, {Graves}, {Hicken}, {Alves}, {Arce}, {Balog},
1790: {Barmby}, {Barton}, {Berlind}, {Bragg}, {Brice{\~n}o}, {Brown}, {Buckley},
1791: {Caldwell}, {Calkins}, {Carter}, {Concannon}, {Donnelly}, {Eriksen},
1792: {Fabricant}, {Falco}, {Fiore}, {Garcia}, {G{\'o}mez}, {Grogin}, {Groner},
1793: {Groot}, {Haisch}, {Hartmann}, {Hergenrother}, {Holman}, {Huchra},
1794: {Jayawardhana}, {Jerius}, {Kannappan}, {Kim}, {Kleyna}, {Kochanek},
1795: {Koranyi}, {Krockenberger}, {Lada}, {Luhman}, {Luu}, {Macri}, {Mader},
1796: {Mahdavi}, {Marengo}, {Marsden}, {McLeod}, {McNamara}, {Megeath}, {Moraru},
1797: {Mossman}, {Muench}, {Mu{\~n}oz}, {Muzerolle}, {Naranjo}, {Nelson-Patel},
1798: {Pahre}, {Patten}, {Peters}, {Peters}, {Raymond}, {Rines}, {Schild},
1799: {Sobczak}, {Spahr}, {Stauffer}, {Stefanik}, {Szentgyorgyi}, {Tollestrup},
1800: {V{\"a}is{\"a}nen}, {Vikhlinin}, {Wang}, {Willner}, {Wolk}, {Zajac}, {Zhao},
1801: \& {Stanek}}]{2006AJ....131..527J}
1802: {Jha}, S., et al. 2006, \aj, 131, 527
1803:
1804: \bibitem[{{Kasen} \& {Woosley}(2007)}]{2007ApJ...656..661K}
1805: {Kasen}, D., \& {Woosley}, S.~E. 2007, \apj, 656, 661
1806:
1807: \bibitem[{{Kelson}(2003)}]{2003PASP..115..688K}
1808: {Kelson}, D.~D. 2003, \pasp, 115, 688
1809:
1810: \bibitem[{{Kirshner} {et~al.}(1993){Kirshner}, {Jeffery}, {Leibundgut},
1811: {Challis}, {Sonneborn}, {Phillips}, {Suntzeff}, {Smith}, {Winkler}, {Winge},
1812: {Hamuy}, {Hunter}, {Roth}, {Blades}, {Branch}, {Chevalier}, {Fransson},
1813: {Panagia}, {Wagoner}, {Wheeler}, \& {Harkness}}]{1993ApJ...415..589K}
1814: {Kirshner}, R.~P., et al. 1993, \apj, 415,
1815: 589
1816:
1817: \bibitem[{{Knop} {et~al.}(2003){Knop}, {Aldering}, {Amanullah}, {Astier},
1818: {Blanc}, {Burns}, {Conley}, {Deustua}, {Doi}, {Ellis}, {Fabbro}, {Folatelli},
1819: {Fruchter}, {Garavini}, {Garmond}, {Garton}, {Gibbons}, {Goldhaber},
1820: {Goobar}, {Groom}, {Hardin}, {Hook}, {Howell}, {Kim}, {Lee}, {Lidman},
1821: {Mendez}, {Nobili}, {Nugent}, {Pain}, {Panagia}, {Pennypacker}, {Perlmutter},
1822: {Quimby}, {Raux}, {Regnault}, {Ruiz-Lapuente}, {Sainton}, {Schaefer},
1823: {Schahmaneche}, {Smith}, {Spadafora}, {Stanishev}, {Sullivan}, {Walton},
1824: {Wang}, {Wood-Vasey}, \& {Yasuda}}]{2003ApJ...598..102K}
1825: {Knop}, R.~A., et al. 2003, \apj, 598, 102
1826:
1827: \bibitem[{{Kobayashi} {et~al.}(1998){Kobayashi}, {Tsujimoto}, {Nomoto},
1828: {Hachisu}, \& {Kato}}]{1998ApJ...503L.155K}
1829: {Kobayashi}, C., {Tsujimoto}, T., {Nomoto}, K., {Hachisu}, I., \& {Kato}, M.
1830: 1998, \apjl, 503, L155
1831:
1832: \bibitem[{{Kobulnicky} {et~al.}(2003){Kobulnicky}, {Willmer}, {Phillips},
1833: {Koo}, {Faber}, {Weiner}, {Sarajedini}, {Simard}, \&
1834: {Vogt}}]{2003ApJ...599.1006K}
1835: {Kobulnicky}, H.~A., et al. 2003, \apj, 599, 1006
1836:
1837: \bibitem[{{Krisciunas} {et~al.}(1987){Krisciunas}, {Sinton}, {Tholen},
1838: {Tokunaga}, {Golisch}, {Griep}, {Kaminski}, {Impey}, \&
1839: {Christian}}]{1987PASP...99..887K}
1840: {Krisciunas}, K., et al. 1987, \pasp, 99,
1841: 887+
1842:
1843: \bibitem[{{Langer} {et~al.}(2000){Langer}, {Deutschmann}, {Wellstein}, \& {H{\"
1844: o}flich}}]{2000A&A...362.1046L}
1845: {Langer}, N., {Deutschmann}, A., {Wellstein}, S., \& {H{\" o}flich}, P. 2000,
1846: \aap, 362, 1046
1847:
1848: \bibitem[{{Le Borgne} \& {Rocca-Volmerange}(2002)}]{2002A&A...386..446L}
1849: {Le Borgne}, D., \& {Rocca-Volmerange}, B. 2002, \aap, 386, 446
1850:
1851: \bibitem[{{Leibundgut} {et~al.}(1991){Leibundgut}, {Kirshner}, {Filippenko},
1852: {Shields}, {Foltz}, {Phillips}, \& {Sonneborn}}]{1991ApJ...371L..23L}
1853: {Leibundgut}, B., {Kirshner}, R.~P., {Filippenko}, A.~V., {Shields}, J.~C.,
1854: {Foltz}, C.~B., {Phillips}, M.~M., \& {Sonneborn}, G. 1991, \apjl, 371, L23
1855:
1856: \bibitem[{{Lentz} {et~al.}(2000){Lentz}, {Baron}, {Branch}, {Hauschildt}, \&
1857: {Nugent}}]{2000ApJ...530..966L}
1858: {Lentz}, E.~J., {Baron}, E., {Branch}, D., {Hauschildt}, P.~H., \& {Nugent},
1859: P.~E. 2000, \apj, 530, 966
1860:
1861: \bibitem[{{Li} \& {Filippenko}(2005)}]{2005coex.conf..525L}
1862: {Li}, W., \& {Filippenko}, A.~V. 2005, in IAU Colloq. 192: Cosmic Explosions,
1863: On the 10th Anniversary of SN1993J, ed. J.-M. {Marcaide} \& K.~W. {Weiler},
1864: 525--+
1865:
1866: \bibitem[{{Li} {et~al.}(2001){Li}, {Filippenko}, {Treffers}, {Riess}, {Hu}, \&
1867: {Qiu}}]{2001ApJ...546..734L}
1868: {Li}, W., {Filippenko}, A.~V., {Treffers}, R.~R., {Riess}, A.~G., {Hu}, J., \&
1869: {Qiu}, Y. 2001, \apj, 546, 734
1870:
1871: \bibitem[{{Lidman} {et~al.}(2005){Lidman}, {Howell}, {Folatelli}, {Garavini},
1872: {Nobili}, {Aldering}, {Amanullah}, {Antilogus}, {Astier}, {Blanc}, {Burns},
1873: {Conley}, {Deustua}, {Doi}, {Ellis}, {Fabbro}, {Fadeyev}, {Gibbons},
1874: {Goldhaber}, {Goobar}, {Groom}, {Hook}, {Kashikawa}, {Kim}, {Knop}, {Lee},
1875: {Mendez}, {Morokuma}, {Motohara}, {Nugent}, {Pain}, {Perlmutter}, {Prasad},
1876: {Quimby}, {Raux}, {Regnault}, {Ruiz-Lapuente}, {Sainton}, {Schaefer},
1877: {Schahmaneche}, {Smith}, {Spadafora}, {Stanishev}, {Walton}, {Wang},
1878: {Wood-Vasey}, {Yasuda}, \& {The Supernova Cosmology
1879: Project}}]{2005A&A...430..843L}
1880: {Lidman}, C., et al. 2005, \aap, 430, 843
1881:
1882: \bibitem[{{Mannucci} {et~al.}(2006){Mannucci}, {Della Valle}, \&
1883: {Panagia}}]{2006MNRAS.370..773M}
1884: {Mannucci}, F., {Della Valle}, M., \& {Panagia}, N. 2006, \mnras, 370, 773
1885:
1886: \bibitem[{{Mannucci} {et~al.}(2005){Mannucci}, {della Valle}, {Panagia},
1887: {Cappellaro}, {Cresci}, {Maiolino}, {Petrosian}, \&
1888: {Turatto}}]{2005A&A...433..807M}
1889: {Mannucci}, F., {della Valle}, M., {Panagia}, N., {Cappellaro}, E., {Cresci},
1890: G., {Maiolino}, R., {Petrosian}, A., \& {Turatto}, M. 2005, \aap, 433, 807
1891:
1892: \bibitem[{{Massey} {et~al.}(1988){Massey}, {Strobel}, {Barnes}, \&
1893: {Anderson}}]{1988ApJ...328..315M}
1894: {Massey}, P., {Strobel}, K., {Barnes}, J.~V., \& {Anderson}, E. 1988, \apj,
1895: 328, 315
1896:
1897: \bibitem[{{Matheson} {et~al.}(2005){Matheson}, {Blondin}, {Foley}, {Chornock},
1898: {Filippenko}, {Leibundgut}, {Smith}, {Sollerman}, {Spyromilio}, {Kirshner},
1899: {Clocchiatti}, {Aguilera}, {Barris}, {Becker}, {Challis}, {Covarrubias},
1900: {Garnavich}, {Hicken}, {Jha}, {Krisciunas}, {Li}, {Miceli}, {Miknaitis},
1901: {Prieto}, {Rest}, {Riess}, {Salvo}, {Schmidt}, {Stubbs}, {Suntzeff}, \&
1902: {Tonry}}]{2005AJ....129.2352M}
1903: {Matheson}, T., et al. 2005, \aj, 129,
1904: 2352
1905:
1906: \bibitem[{{Neill} {et~al.}(2006){Neill}, {Sullivan}, {Balam}, {Pritchet},
1907: {Howell}, {Perrett}, {Astier}, {Aubourg}, {Basa}, {Carlberg}, {Conley},
1908: {Fabbro}, {Fouchez}, {Guy}, {Hook}, {Pain}, {Palanque-Delabrouille},
1909: {Regnault}, {Rich}, {Taillet}, {Aldering}, {Antilogus}, {Arsenijevic},
1910: {Balland}, {Baumont}, {Bronder}, {Ellis}, {Filiol}, {Gon{\c c}alves},
1911: {Hardin}, {Kowalski}, {Lidman}, {Lusset}, {Mouchet}, {Mourao}, {Perlmutter},
1912: {Ripoche}, {Schlegel}, \& {Tao}}]{2006AJ....132.1126N}
1913: {Neill}, J.~D., et al. 2006, \aj, 132, 1126
1914:
1915: \bibitem[{Nugent {et~al.}(1995)Nugent, Phillips, Baron, Branch, \&
1916: Hauschildt}]{nugseq95} Nugent, P., Phillips, M., Baron, E., Branch,
1917: D., \& Hauschildt, P. 1995, \apj, 455, L147
1918:
1919: \bibitem[{{Nugent} {et~al.}(2002){Nugent}, {Kim}, \&
1920: {Perlmutter}}]{2002PASP..114..803N}
1921: {Nugent}, P., {Kim}, A., \& {Perlmutter}, S. 2002, \pasp, 114, 803
1922:
1923: \bibitem[{{Nugent} {et~al.}(2006){Nugent}, {Sullivan}, {Ellis}, {Gal-Yam},
1924: {Leonard}, {Howell}, {Astier}, {Carlberg}, {Conley}, {Fabbro}, {Fouchez},
1925: {Neill}, {Pain}, {Perrett}, {Pritchet}, \& {Regnault}}]{2006ApJ...645..841N}
1926: {Nugent}, P., et al. 2006, \apj, 645, 841
1927:
1928: \bibitem[{{Oke} {et~al.}(1995){Oke}, {Cohen}, {Carr}, {Cromer}, {Dingizian},
1929: {Harris}, {Labrecque}, {Lucinio}, {Schaal}, {Epps}, \&
1930: {Miller}}]{1995PASP..107..375O}
1931: {Oke}, J.~B., et al. 1995, \pasp, 107, 375
1932:
1933: \bibitem[{{Perlmutter} {et~al.}(1999){Perlmutter}, {Aldering}, {Goldhaber},
1934: {Knop}, {Nugent}, {Castro}, {Deustua}, {Fabbro}, {Goobar}, {Groom}, {Hook},
1935: {Kim}, {Kim}, {Lee}, {Nunes}, {Pain}, {Pennypacker}, {Quimby}, {Lidman},
1936: {Ellis}, {Irwin}, {McMahon}, {Ruiz-Lapuente}, {Walton}, {Schaefer}, {Boyle},
1937: {Filippenko}, {Matheson}, {Fruchter}, {Panagia}, {Newberg}, {Couch}, \& {The
1938: Supernova Cosmology Project}}]{1999ApJ...517..565P}
1939: {Perlmutter}, S., et al. 1999, \apj, 517, 565
1940:
1941: \bibitem[{{Perlmutter} {et~al.}(1997){Perlmutter}, {Gabi}, {Goldhaber},
1942: {Goobar}, {Groom}, {Hook}, {Kim}, {Kim}, {Lee}, {Pain}, {Pennypacker},
1943: {Small}, {Ellis}, {McMahon}, {Boyle}, {Bunclark}, {Carter}, {Irwin},
1944: {Glazebrook}, {Newberg}, {Filippenko}, {Matheson}, {Dopita}, {Couch}, \& {The
1945: Supernova Cosmology Project}}]{1997ApJ...483..565P}
1946: {Perlmutter}, S., et al. 1997,
1947: \apj, 483, 565
1948:
1949: \bibitem[{{Phillips} {et~al.}(1999){Phillips}, {Lira}, {Suntzeff}, {Schommer},
1950: {Hamuy}, \& {Maza}}]{1999AJ....118.1766P}
1951: {Phillips}, M.~M., {Lira}, P., {Suntzeff}, N.~B., {Schommer}, R.~A., {Hamuy},
1952: M., \& {Maza}, J.~. 1999, \aj, 118, 1766
1953:
1954: \bibitem[{{Riess} {et~al.}(1998){Riess}, {Filippenko}, {Challis},
1955: {Clocchiatti}, {Diercks}, {Garnavich}, {Gilliland}, {Hogan}, {Jha},
1956: {Kirshner}, {Leibundgut}, {Phillips}, {Reiss}, {Schmidt}, {Schommer},
1957: {Smith}, {Spyromilio}, {Stubbs}, {Suntzeff}, \&
1958: {Tonry}}]{1998AJ....116.1009R}
1959: {Riess}, A.~G., et al. 1998, \aj, 116, 1009
1960:
1961: \bibitem[{{Riess} {et~al.}(2000){Riess}, {Filippenko}, {Liu}, {Challis},
1962: {Clocchiatti}, {Diercks}, {Garnavich}, {Hogan}, {Jha}, {Kirshner},
1963: {Leibundgut}, {Phillips}, {Reiss}, {Schmidt}, {Schommer}, {Smith},
1964: {Spyromilio}, {Stubbs}, {Suntzeff}, {Tonry}, {Woudt}, {Brunner}, {Dey},
1965: {Gal}, {Graham}, {Larkin}, {Odewahn}, \& {Oppenheimer}}]{2000ApJ...536...62R}
1966: {Riess}, A.~G., et al. 2000, \apj, 536, 62
1967:
1968: \bibitem[{{Riess} {et~al.}(1999){Riess}, {Kirshner}, {Schmidt}, {Jha},
1969: {Challis}, {Garnavich}, {Esin}, {Carpenter}, {Grashius}, {Schild}, {Berlind},
1970: {Huchra}, {Prosser}, {Falco}, {Benson}, {Brice{\~ n}o}, {Brown}, {Caldwell},
1971: {dell'Antonio}, {Filippenko}, {Goodman}, {Grogin}, {Groner}, {Hughes},
1972: {Green}, {Jansen}, {Kleyna}, {Luu}, {Macri}, {McLeod}, {McLeod}, {McNamara},
1973: {McLean}, {Milone}, {Mohr}, {Moraru}, {Peng}, {Peters}, {Prestwich},
1974: {Stanek}, {Szentgyorgyi}, \& {Zhao}}]{1999AJ....117..707R}
1975: {Riess}, A.~G., et al. 1999, \aj, 117, 707
1976:
1977: \bibitem[{{Riess} {et~al.}(1996){Riess}, {Press}, \&
1978: {Kirshner}}]{1996ApJ...473...88R}
1979: {Riess}, A.~G., {Press}, W.~H., \& {Kirshner}, R.~P. 1996, \apj, 473, 88
1980:
1981: \bibitem[{{Riess} {et~al.}(2004){Riess}, {Strolger}, {Tonry}, {Casertano},
1982: {Ferguson}, {Mobasher}, {Challis}, {Filippenko}, {Jha}, {Li}, {Chornock},
1983: {Kirshner}, {Leibundgut}, {Dickinson}, {Livio}, {Giavalisco}, {Steidel},
1984: {Ben{\'{\i}}tez}, \& {Tsvetanov}}]{2004ApJ...607..665R}
1985: {Riess}, A.~G., et al. 2004, \apj, 607, 665
1986:
1987: \bibitem[{{Riess} {et~al.}(2007){Riess}, {Strolger}, {Casertano}, {Ferguson},
1988: {Mobasher}, {Gold}, {Challis}, {Filippenko}, {Jha}, {Li}, {Tonry}, {Foley},
1989: {Kirshner}, {Dickinson}, {MacDonald}, {Eisenstein}, {Livio}, {Younger}, {Xu},
1990: {Dahl{\'e}n}, \& {Stern}}]{2007ApJ...659...98R}
1991: {Riess}, A.~G., et al. 2007, \apj, 659, 98
1992:
1993: \bibitem[{{Scannapieco} \& {Bildsten}(2005)}]{2005ApJ...629L..85S}
1994: {Scannapieco}, E., \& {Bildsten}, L. 2005, \apjl, 629, L85
1995:
1996: \bibitem[{{Schlegel} {et~al.}(1998){Schlegel}, {Finkbeiner}, \&
1997: {Davis}}]{1998ApJ...500..525S}
1998: {Schlegel}, D.~J., {Finkbeiner}, D.~P., \& {Davis}, M. 1998, \apj, 500, 525
1999:
2000: \bibitem[{{Schmidt} {et~al.}(1998){Schmidt}, {Suntzeff}, {Phillips},
2001: {Schommer}, {Clocchiatti}, {Kirshner}, {Garnavich}, {Challis}, {Leibundgut},
2002: {Spyromilio}, {Riess}, {Filippenko}, {Hamuy}, {Smith}, {Hogan}, {Stubbs},
2003: {Diercks}, {Reiss}, {Gilliland}, {Tonry}, {Maza}, {Dressler}, {Walsh}, \&
2004: {Ciardullo}}]{1998ApJ...507...46S}
2005: {Schmidt}, B.~P., et al. 1998, \apj, 507, 46
2006:
2007: \bibitem[{{Shapley} {et~al.}(2005){Shapley}, {Coil}, {Ma}, \&
2008: {Bundy}}]{2005ApJ...635.1006S}
2009: {Shapley}, A.~E., {Coil}, A.~L., {Ma}, C.-P., \& {Bundy}, K. 2005, \apj, 635,
2010: 1006
2011:
2012: \bibitem[{{Spergel} {et~al.}(2007){Spergel}, {Bean}, {Dore'}, {Nolta},
2013: {Bennett}, {Hinshaw}, {Jarosik}, {Komatsu}, {Page}, {Peiris}, {Verde},
2014: {Barnes}, {Halpern}, {Hill}, {Kogut}, {Limon}, {Meyer}, {Odegard}, {Tucker},
2015: {Weiland}, {Wollack}, \& {Wright}}]{2007ApJS..170..377S}
2016: {Spergel}, D.~N., et al. 2007, \apjs, 170, 377
2017:
2018: \bibitem[{{Sullivan} {et~al.}(2003){Sullivan}, {Ellis}, {Aldering},
2019: {Amanullah}, {Astier}, {Blanc}, {Burns}, {Conley}, {Deustua}, {Doi},
2020: {Fabbro}, {Folatelli}, {Fruchter}, {Garavini}, {Gibbons}, {Goldhaber},
2021: {Goobar}, {Groom}, {Hardin}, {Hook}, {Howell}, {Irwin}, {Kim}, {Knop},
2022: {Lidman}, {McMahon}, {Mendez}, {Nobili}, {Nugent}, {Pain}, {Panagia},
2023: {Pennypacker}, {Perlmutter}, {Quimby}, {Raux}, {Regnault}, {Ruiz-Lapuente},
2024: {Schaefer}, {Schahmaneche}, {Spadafora}, {Walton}, {Wang}, {Wood-Vasey}, \&
2025: {Yasuda}}]{2003MNRAS.340.1057S}
2026: {Sullivan}, M., et al. 2003, \mnras, 340, 1057
2027:
2028: \bibitem[{{Sullivan} {et~al.}(2006{\natexlab{a}}){Sullivan}, {Howell},
2029: {Perrett}, {Nugent}, {Astier}, {Aubourg}, {Balam}, {Basa}, {Carlberg},
2030: {Conley}, {Fabbro}, {Fouchez}, {Guy}, {Hook}, {Lafoux}, {Neill}, {Pain},
2031: {Palanque-Delabrouille}, {Pritchet}, {Regnault}, {Rich}, {Taillet},
2032: {Aldering}, {Baumont}, {Bronder}, {Filiol}, {Knop}, {Perlmutter}, \&
2033: {Tao}}]{2006AJ....131..960S}
2034: {Sullivan}, M., et al.
2035: 2006{\natexlab{a}}, \aj, 131, 960
2036:
2037: \bibitem[{{Sullivan} {et~al.}(2006{\natexlab{b}}){Sullivan}, {Le Borgne},
2038: {Pritchet}, {Hodsman}, {Neill}, {Howell}, {Carlberg}, {Astier}, {Aubourg},
2039: {Balam}, {Basa}, {Conley}, {Fabbro}, {Fouchez}, {Guy}, {Hook}, {Pain},
2040: {Palanque-Delabrouille}, {Perrett}, {Regnault}, {Rich}, {Taillet}, {Baumont},
2041: {Bronder}, {Ellis}, {Filiol}, {Lusset}, {Perlmutter}, {Ripoche}, \&
2042: {Tao}}]{2006ApJ...648..868S}
2043: {Sullivan}, M., et al. 2006{\natexlab{b}}, \apj, 648, 868
2044:
2045: \bibitem[{{Timmes} {et~al.}(2003){Timmes}, {Brown}, \&
2046: {Truran}}]{2003ApJ...590L..83T}
2047: {Timmes}, F.~X., {Brown}, E.~F., \& {Truran}, J.~W. 2003, \apjl, 590, L83
2048:
2049: \bibitem[{{Tremonti} {et~al.}(2004){Tremonti}, {Heckman}, {Kauffmann},
2050: {Brinchmann}, {Charlot}, {White}, {Seibert}, {Peng}, {Schlegel}, {Uomoto},
2051: {Fukugita}, \& {Brinkmann}}]{2004ApJ...613..898T}
2052: {Tremonti}, C.~A., et al. 2004, \apj, 613, 898
2053:
2054: \bibitem[{{van Dokkum}(2001)}]{2001PASP..113.1420V}
2055: {van Dokkum}, P.~G. 2001, \pasp, 113, 1420
2056:
2057: \bibitem[{{Wood-Vasey} {et~al.}(2007){Wood-Vasey}, {Miknaitis}, {Stubbs},
2058: {Jha}, {Riess}, {Garnavich}, {Kirshner}, {Aguilera}, {Becker}, {Blackman},
2059: {Blondin}, {Challis}, {Clocchiatti}, {Conley}, {Covarrubias}, {Davis},
2060: {Filippenko}, {Foley}, {Garg}, {Hicken}, {Krisciunas}, {Leibundgut}, {Li},
2061: {Matheson}, {Miceli}, {Narayan}, {Pignata}, {Prieto}, {Rest}, {Salvo},
2062: {Schmidt}, {Smith}, {Sollerman}, {Spyromilio}, {Tonry}, {Suntzeff}, \&
2063: {Zenteno}}]{2007ApJ...666..694W}
2064: {Wood-Vasey}, W.~M., et al. 2007, \apj, 666, 694
2065:
2066: \end{thebibliography}
2067:
2068:
2069: \clearpage
2070:
2071: \begin{deluxetable}{lllccccll}
2072: \tablecaption{Distant Supernova Sample}
2073: \tabletypesize{\scriptsize}
2074: \tablehead{\colhead{SN Name$^a$} & \colhead{RA} & \colhead{DEC} & \colhead{Redshift} & \colhead{Phase} & \colhead{Stretch} & \colhead{Nearest $g'$} & \colhead{\%} & \colhead{Redshift}\\ \colhead{} & \colhead{(J2000)} & \colhead{(J2000)} & \colhead{} & \colhead{(Days)$^b$} & \colhead{} & \colhead{phot (Days)$^b$} & \colhead{Inc.$^c$} & \colhead{Source}}
2075: \startdata
2076: 03D1au & 02:24:10.38 & -04:02:14.9 & 0.5043\,$\pm$\,0.0005 & -1.6\,$\pm$\,0.3 & 1.14\,$\pm$\,0.02 & +0.6 & 432 & Galaxy\\
2077: 03D1aw & 02:24:14.78 & -04:31:01.6 & 0.582\,$\pm$\,0.005 & +2.4\,$\pm$\,0.3 & 1.08\,$\pm$\,0.02 & +0.6 & 856 & Galaxy\\
2078: 03D1co & 02:26:16.23 & -04:56:05.7 & 0.679\,$\pm$\,0.001 & +7.1\,$\pm$\,0.5 & 1.07\,$\pm$\,0.04 & +0.7 & 321 & Galaxy\\
2079: 03D1dj & 02:26:19.08 & -04:07:09.3 & 0.40\,$\pm$\,0.01 & -1.8\,$\pm$\,2.3 & 1.25\,$\pm$\,0.61 & +0.1 & 4199 & Template Fit\\
2080: 03D3af & 14:21:14.92 & +52:32:15.3 & 0.5320\,$\pm$\,0.0005 & +2.8\,$\pm$\,0.4 & 1.02\,$\pm$\,0.03 & +3.8 & 368 & Galaxy\\
2081: 03D3aw\tablenotemark{e} & 14:20:53.61 & +52:36:20.6 & 0.449\,$\pm$\,0.001 & -0.8\,$\pm$\,0.2 & 0.94\,$\pm$\,0.03 & -13.7 & 52 & Galaxy\\
2082: 03D3ay\tablenotemark{e} & 14:17:58.43 & +52:28:57.4 & 0.3709\,$\pm$\,0.0003 & -1.0\,$\pm$\,0.2 & 1.00\,$\pm$\,0.03 & -14.6 & 72 & Galaxy\\
2083: 03D3ba\tablenotemark{e} & 14:16:33.44 & +52:20:32.1 & 0.2912\,$\pm$\,0.0003 & +12.8\,$\pm$\,0.4 & 1.04\,$\pm$\,0.03 & -15.5 & 259 & Galaxy\\
2084: 03D3bb\tablenotemark{f} & 14:16:18.78 & +52:14:55.3 & 0.2437\,$\pm$\,0.0003 & +2.2\,$\pm$\,0.3 & 1.18\,$\pm$\,0.03 & -16.2 & 2323 & Galaxy\\
2085: 03D3bh\tablenotemark{e} & 14:21:35.89 & +52:31:37.6 & 0.2486\,$\pm$\,0.0003 & -3.5\,$\pm$\,0.2 & 1.02\,$\pm$\,0.03 & -16.1 & 140 & Galaxy\\
2086: 03D3bl & 14:19:55.90 & +53:05:51.0 & 0.3553\,$\pm$\,0.0005 & +3.7\,$\pm$\,0.5 & 0.98\,$\pm$\,0.03 & -0.1 & 270 & Galaxy\\
2087: 03D3cc & 14:19:45.25 & +52:32:25.3 & 0.4627\,$\pm$\,0.0003 & +8.5\,$\pm$\,0.4 & 1.04\,$\pm$\,0.02 & -0.1 & 204 & Galaxy\\
2088: 03D3cd & 14:18:39.95 & +52:36:43.8 & 0.4607\,$\pm$\,0.0005 & -5.6\,$\pm$\,0.1 & 1.15\,$\pm$\,0.02 & -0.1 & 625 & Galaxy\\
2089: 03D4ag & 22:14:45.79 & -17:44:23.0 & 0.2847\,$\pm$\,0.0003 & -5.0\,$\pm$\,0.1 & 1.08\,$\pm$\,0.02 & -1.6 & 141 & Galaxy\\
2090: 03D4cj\tablenotemark{f} & 22:16:06.66 & -17:42:16.7 & 0.27\,$\pm$\,0.01 & -7.4\,$\pm$\,0.1 & 1.09\,$\pm$\,0.01 & +1.5 & 5000 & Template Fit\\
2091: 03D4dh & 22:17:31.04 & -17:37:46.9 & 0.6268\,$\pm$\,0.0004 & +0.5\,$\pm$\,0.3 & 1.08\,$\pm$\,0.02 & +0.6 & 444 & Galaxy\\
2092: 03D4gl\tablenotemark{e} & 22:14:44.17 & -17:31:44.4 & 0.571\,$\pm$\,0.001 & +7.7\,$\pm$\,0.5 & 1.00\,$\pm$\,0.03 & +14.6 & 347 & Galaxy\\
2093: 04D1hd & 02:26:08.85 & -04:06:35.2 & 0.3688\,$\pm$\,0.0004 & +1.6\,$\pm$\,0.1 & 1.06\,$\pm$\,0.01 & +1.4 & 1998 & Galaxy\\
2094: 04D1jg & 02:26:12.56 & -04:08:05.3 & 0.5842\,$\pm$\,0.0005 & -2.0\,$\pm$\,0.2 & 1.05\,$\pm$\,0.02 & +1.3 & 128 & Galaxy\\
2095: 04D1oh\tablenotemark{d} & 02:25:02.37 & -04:14:10.5 & 0.59\,$\pm$\,0.01 & -6.0\,$\pm$\,0.2 & 1.03\,$\pm$\,0.02 & +0.1 & 512 & Template Fit\\
2096: 04D1rh & 02:27:47.16 & -04:15:13.6 & 0.4349\,$\pm$\,0.0005 & +2.6\,$\pm$\,0.3 & 1.02\,$\pm$\,0.03 & -0.1 & 393 & Galaxy\\
2097: 04D1sk & 02:24:22.77 & -04:21:13.3 & 0.6634\,$\pm$\,0.0005 & -0.4\,$\pm$\,0.6 & 0.86\,$\pm$\,0.04 & +0.0 & 139 & Galaxy\\
2098: 04D2gc & 10:01:39.26 & +01:52:59.5 & 0.5216\,$\pm$\,0.0005 & -0.2\,$\pm$\,0.3 & 1.14\,$\pm$\,0.03 & -0.9 & 367 & Galaxy\\
2099: 04D2kr & 10:00:37.32 & +01:42:43.1 & 0.7441\,$\pm$\,0.0005 & -2.5\,$\pm$\,0.6 & 1.06\,$\pm$\,0.04 & +0.6 & 91 & Galaxy\\
2100: 04D3cp\tablenotemark{d} & 14:20:23.95 & +52:49:15.5 & 0.83\,$\pm$\,0.02 & +5.2\,$\pm$\,0.4 & 1.07\,$\pm$\,0.02 & +0.5 & 557 & Galaxy\\
2101: 04D3ez & 14:19:07.91 & +53:04:18.8 & 0.2630\,$\pm$\,0.0005 & +1.6\,$\pm$\,0.1 & 0.92\,$\pm$\,0.02 & -0.4 & 78 & Galaxy\\
2102: 04D3fk & 14:18:26.21 & +52:31:42.7 & 0.3578\,$\pm$\,0.0007 & -6.9\,$\pm$\,0.1 & 0.97\,$\pm$\,0.01 & -0.4 & 136 & Galaxy\\
2103: 04D4in & 22:15:08.58 & -17:15:39.8 & 0.5160\,$\pm$\,0.0005 & -4.6\,$\pm$\,0.1 & 1.13\,$\pm$\,0.01 & +2.0 & 443 & Galaxy\\
2104: 04D4jr & 22:14:14.33 & -17:21:00.9 & 0.482\,$\pm$\,0.007 & -0.2\,$\pm$\,0.2 & 1.16\,$\pm$\,0.02 & +2.7 & 5000 & Template Fit\\
2105: 05D1hk\tablenotemark{f} & 02:24:39.16 & -04:38:03.0 & 0.2631\,$\pm$\,0.0002 & -8.6\,$\pm$\,0.1 & 1.16\,$\pm$\,0.01 & -0.8 & 337 & Galaxy\\
2106: 05D1hn & 02:24:36.25 & -04:10:54.9 & 0.1489\,$\pm$\,0.0006 & -7.2\,$\pm$\,0.1 & 1.06\,$\pm$\,0.02 & -0.8 & 232 & Galaxy\\
2107: 05D1if & 02:24:29.71 & -04:34:13.0 & 0.763\,$\pm$\,0.001 & -5.0\,$\pm$\,0.3 & 1.03\,$\pm$\,0.03 & -0.0 & 5000 & Galaxy\\
2108: 05D1ix & 02:24:19.95 & -04:40:11.7 & 0.49\,$\pm$\,0.01 & -8.7\,$\pm$\,0.2 & 1.05\,$\pm$\,0.01 & -0.7 & 5000 & Template Fit\\
2109: 05D1iy & 02:27:39.97 & -04:25:21.3 & 0.2478\,$\pm$\,0.0003 & -9.2\,$\pm$\,0.1 & 0.88\,$\pm$\,0.01 & -0.7 & 52 & Galaxy\\
2110: 05D2le & 10:01:54.85 & +02:05:34.7 & 0.7002\,$\pm$\,0.0005 & +6.3\,$\pm$\,0.4 & 1.10\,$\pm$\,0.04 & -0.0 & 2647 & Galaxy\\
2111: 05D2mp & 09:59:08.61 & +02:12:14.6 & 0.3537\,$\pm$\,0.0004 & -3.9\,$\pm$\,0.2 & 1.11\,$\pm$\,0.02 & +0.7 & 831 & Galaxy\\
2112: \enddata
2113: \tablenotetext{a}{The SNLS ID format is explained in \citet{2005ApJ...634.1190H}; briefly, it relates to
2114: the year of discovery (03, 04 etc.), the relevant SNLS field (D1
2115: through D4), and a running ID (e.g. aa, ab, ac etc.).}
2116: \tablenotetext{b}{``Effective'' rest-frame phase relative to maximum light i.e. $(T_{\mathrm{obs}}-T_{\mathrm{max}})/(s\times(1+z))$}
2117: \tablenotetext{c}{Estimated through the spectral extraction aperture; increases greater than 5000\% are listed as 5000\%.}
2118: \tablenotetext{d}{Light curve deemed inadequate in temporal coverage.}
2119: \tablenotetext{e}{Spectrum deemed inadequate in signal-to-noise ratio.}
2120: \tablenotetext{f}{Spectroscopically peculiar; excluded from primary analysis sample.}
2121:
2122: \label{tab:sninfo}
2123: \end{deluxetable}
2124:
2125: \clearpage
2126:
2127: \begin{deluxetable}{lccccccc}
2128: \tabletypesize{\small}
2129: \tablecaption{Observing log}
2130: \tablehead{\colhead{SN Name} & \colhead{Date Obs} & \colhead{MJD} & \colhead{Seeing} & \colhead{Slit} & \colhead{Dichroic} & \colhead{Exptime}& \colhead{SN apparent}\\ \colhead{} & \colhead{} & \colhead{} & \colhead{(\arcsec)} & \colhead{(\arcsec)} & \colhead{} & \colhead{(s)} & \colhead{i' mag.}}
2131: \startdata
2132: 03D1au & 2003-09-23 & 52905.51 & 0.8 & 1.0 & 560 & 5400 & 22.4\\
2133: 03D1aw & 2003-09-23 & 52905.58 & 0.8 & 1.0 & 560 & 5400 & 22.7\\
2134: 03D1co & 2003-11-22 & 52965.48 & 0.9 & 1.0 & 680 & 9000 & 23.5\\
2135: 03D1dj & 2003-11-21 & 52964.47 & 0.8 & 1.0 & 560 & 6000 & 22.0\\
2136: 03D3af & 2003-04-07 & 52736.41 & 0.9 & 1.0 & 680 & 5400 & 22.7\\
2137: 03D3aw & 2003-05-06 & 52765.53 & 0.9 & 1.0 & 680 & 4800 & 22.2\\
2138: 03D3ay & 2003-05-06 & 52765.47 & 0.9 & 1.0 & 680 & 3600 & 21.7\\
2139: 03D3ba & 2003-05-06 & 52765.38 & 0.9 & 1.0 & 680 & 3600 & 21.7\\
2140: 03D3bb & 2003-05-06 & 52765.29 & 0.9 & 1.0 & 560 & 2000 & 19.8\\
2141: 03D3bh & 2003-05-06 & 52765.32 & 0.9 & 1.0 & 560 & 3600 & 20.9\\
2142: 03D3bl & 2003-06-01 & 52791.33 & 0.9 & 1.0 & 680 & 3600 & 22.0\\
2143: 03D3cc & 2003-06-02 & 52792.27 & 1.0 & 1.0 & 680 & 5400 & 22.1\\
2144: 03D3cd & 2003-06-01 & 52791.27 & 1.0 & 1.0 & 680 & 9777 & 22.2\\
2145: 03D4ag & 2003-07-02 & 52822.54 & 0.9 & 1.0 & 560 & 5400 & 21.1\\
2146: 03D4cj & 2003-08-27 & 52878.32 & 0.9 & 1.0 & 560 & 4000 & 21.2\\
2147: 03D4dh & 2003-09-23 & 52905.43 & 0.8 & 1.0 & 560 & 3600 & 22.6\\
2148: 03D4gl & 2003-11-22 & 52965.26 & 0.9 & 1.0 & 680 & 7200 & 22.6\\
2149: 04D1hd & 2004-09-21 & 53269.47 & 0.9 & 1.0 & 560 & 6000 & 21.6\\
2150: 04D1jg & 2004-09-21 & 53269.56 & 0.9 & 1.0 & 560 & 6000 & 22.5\\
2151: 04D1oh & 2004-10-19 & 53297.54 & 1.0 & 1.0 & 560 & 7200 & 22.9\\
2152: 04D1rh & 2004-12-14 & 53353.24 & 0.9 & 0.7 & 560 & 6000 & 21.9\\
2153: 04D1sk & 2004-12-14 & 53353.36 & 0.9 & 1.0 & 560 & 9000 & 23.2\\
2154: 04D2gc & 2004-04-22 & 53117.00 & 0.7 & 0.7 & 560 & 6600 & 22.5\\
2155: 04D2kr & 2004-12-14 & 53353.58 & 0.9 & 1.0 & 560 & 9600 & 22.8\\
2156: 04D3cp & 2004-04-23 & 53118.45 & 0.8 & 0.7 & 560 & 2700 & 23.1\\
2157: 04D3ez & 2004-04-22 & 53117.00 & 0.8 & 0.7 & 560 & 3600 & 21.0\\
2158: 04D3fk & 2004-04-22 & 53117.00 & 0.8 & 0.7 & 560 & 4500 & 22.3\\
2159: 04D4in & 2004-09-21 & 53269.31 & 0.9 & 1.0 & 560 & 7200 & 22.4\\
2160: 04D4jr & 2004-10-19 & 53297.28 & 1.0 & 1.0 & 560 & 7200 & 22.0\\
2161: 05D1hk & 2005-11-30 & 53704.37 & 0.9 & 1.0 & 560 & 2400 & 21.5\\
2162: 05D1hn & 2005-11-30 & 53704.40 & 0.9 & 1.0 & 560 & 2100 & 20.8\\
2163: 05D1if & 2005-12-01 & 53705.31 & 0.9 & 1.0 & 560 & 9000 & 23.2\\
2164: 05D1ix & 2005-11-30 & 53704.28 & 0.9 & 1.0 & 560 & 7200 & 22.9\\
2165: 05D1iy & 2005-11-30 & 53704.43 & 0.9 & 1.0 & 560 & 4800 & 22.1\\
2166: 05D2le & 2005-12-01 & 53705.57 & 0.9 & 1.0 & 560 & 7200 & 23.2\\
2167: 05D2mp & 2005-11-30 & 53704.53 & 0.9 & 1.0 & 560 & 6600 & 21.9\\
2168: \enddata
2169: \label{tab:observinglog}
2170: \end{deluxetable}
2171:
2172: \clearpage
2173:
2174: \begin{deluxetable}{cccc}
2175: \tablecaption{Mean SN~Ia spectrum (Early)}
2176: \tablehead{\colhead{Wavelength (\AA)} & \colhead{Flux} & \colhead{Flux lower} & \colhead{Flux upper}}
2177: \startdata
2178: 2805 & 0.506 & 0.302 & 0.629\\
2179: 2810 & 0.579 & 0.322 & 0.643\\
2180: 2815 & 0.417 & 0.364 & 0.662\\
2181: 2820 & 0.477 & 0.390 & 0.682\\
2182: 2825 & 0.482 & 0.431 & 0.708\\
2183: 2830 & 0.529 & 0.449 & 0.724\\
2184: 2835 & 0.747 & 0.472 & 0.757\\
2185: 2840 & 0.723 & 0.483 & 0.792\\
2186: 2845 & 0.676 & 0.481 & 0.813\\
2187: 2850 & 0.719 & 0.495 & 0.830\\
2188: \enddata
2189: \tablecomments{The complete version of this table is in the electronic edition of
2190: the Journal. The printed edition contains only a sample.}
2191: \label{tab:mean_spectra_early}
2192: \end{deluxetable}
2193:
2194: \newpage
2195: \begin{deluxetable}{cccc}
2196: \tablecaption{Mean SN~Ia spectrum (Max)}
2197: \tablehead{\colhead{Wavelength (\AA)} & \colhead{Flux} & \colhead{Flux lower} & \colhead{Flux upper}}
2198: \startdata
2199: 2805 & 0.287 & 0.262 & 0.411\\
2200: 2810 & 0.360 & 0.282 & 0.422\\
2201: 2815 & 0.372 & 0.299 & 0.435\\
2202: 2820 & 0.438 & 0.317 & 0.448\\
2203: 2825 & 0.355 & 0.334 & 0.456\\
2204: 2830 & 0.477 & 0.349 & 0.470\\
2205: 2835 & 0.433 & 0.357 & 0.472\\
2206: 2840 & 0.378 & 0.367 & 0.483\\
2207: 2845 & 0.442 & 0.375 & 0.496\\
2208: 2850 & 0.428 & 0.377 & 0.506\\
2209: \enddata
2210: \tablecomments{The complete version of this table is in the electronic edition of
2211: the Journal. The printed edition contains only a sample.}
2212: \label{tab:mean_spectra_max}
2213: \end{deluxetable}
2214:
2215:
2216: \clearpage
2217:
2218: \begin{figure}
2219:
2220: \includegraphics[width=3.5in]{f1a.eps}
2221: \includegraphics[width=3.5in]{f1b.eps}
2222:
2223: \caption{
2224: Redshift (left) and phase (right) distributions of the high-redshift
2225: SNLS SNe~Ia studied in this paper. The vertical dashed lines in the
2226: phase distributions show how the SN~Ia sample is divided into ``early''
2227: and ``maximum-light'' spectra. Numbers (and the open histograms) refer to the primary sample of
2228: 36 high signal-to-noise ratio spectra whereas those in parentheses (and the shaded histograms) refer to
2229: the sample of 26 used to construct the mean UV spectra (see text).
2230: \label{fig:redshiftphasehist}}
2231:
2232: \end{figure}
2233:
2234: \begin{figure}
2235:
2236: \centerline{\includegraphics[width=3.7in]{f2a.eps} \includegraphics[width=3.7in]{f2b.eps}}
2237:
2238: \caption{Examples of the host galaxy subtraction and flux calibration
2239: techniques for two SNe~Ia suffering host contamination. The blue
2240: spectrum represents the observed data, the red spectrum the
2241: estimated host galaxy spectrum from fits to the broad-band galaxy
2242: $u^{\star}g'r'i'z'$ photometry (red circles), and the black spectrum
2243: is the host-galaxy subtracted final spectrum color-corrected to the
2244: $g'r'i'z'$ SN photometry (black circles). A smoothed version of the
2245: final spectrum is over-plotted, the heavy line denoting the spectral
2246: region deemed to have the most reliable flux calibration. The
2247: photometric points are plotted at the effective wavelengths of the
2248: filters on the appropriate spectrum, and the total system response
2249: through the $g'r'i'z'$ filters overplotted at the observed
2250: wavelengths. The position of the dichroic between the two arms of
2251: the spectrograph is also marked. Left (a): SNLS-04D3ez at $z=0.263$.
2252: This SN had an increase in $i'$ of $\sim80$\% on the day of
2253: observation. Right (b): SNLS-03D3cc at $z=0.463$, with an increase
2254: in $i'$ of $\sim200$\%. Note that due to the color correction
2255: applied to the final spectrum, as plotted the black spectrum differs
2256: slightly from the simple subtraction of the red from the blue. [See
2257: the electronic edition of the journal for
2258: Figs.~\ref{fig:eghostsub}.3--\ref{fig:eghostsub}.26]\label{fig:eghostsub}}
2259:
2260: \end{figure}
2261:
2262: \begin{figure}
2263:
2264: \centerline{\includegraphics[width=4in]{f3.eps}}
2265: \caption{Maximum light SN~Ia spectra from \citet{2000ApJ...530..966L}
2266: at a variety of different metallicities (top to bottom: 0.1, 0.3,
2267: 1.0, 3.0 and 10.0 times solar metallicity). Over-plotted are the
2268: standard \citet{1990PASP..102.1181B} $U$ and $B$ filter responses,
2269: as well as the UV box filters (``UV1'' and ``UV2'') and
2270: spectroscopic normalizing filter (``b'') used in the analysis.
2271: Features thought to represent possible metallicity diagnostics
2272: ($\lambda_1$, $\lambda_2$) are also marked, as well as the position
2273: of the blue-shifted \ion{Si}{2} 4130\AA\ absorption
2274: feature.\label{fig:col_lentz}}
2275: \end{figure}
2276:
2277: \begin{figure}
2278:
2279: \centerline{\includegraphics[width=4in]{f4.eps}}
2280: \caption{Comparison of SN stretch, SN B-V color, host galaxy specific
2281: star formation rate and SN/galaxy projected separation for the Keck
2282: SN~Ia sample (black histogram; present paper) and the entire SNLS
2283: sample within the appropriate redshift range discovered over the
2284: same period (grey histogram). Galaxies with undetected star
2285: formation were placed at 10$^{-12}\, M_{\odot}$ yr$^{-1}$ per unit
2286: stellar mass.\label{fig:samplecomparison}}
2287: \end{figure}
2288:
2289: \begin{figure}
2290:
2291: \plotone{f5.eps}
2292: \caption{
2293: The mean high-redshift maximum light (effective day $< \pm$ 4 days)
2294: rest-frame UV SN~Ia spectrum compared to the local average template of
2295: \citet{2002PASP..114..803N}. Over-plotted in light grey are 100
2296: bootstrap-resampled mean spectra drawn from the high-redshift
2297: population; the dotted lines show the region containing 90\% of
2298: this distribution. The local template has been color-adjusted to
2299: match the high redshift data.
2300: \label{fig:mean_max}}
2301:
2302: \end{figure}
2303:
2304: \begin{figure}
2305:
2306: \plotone{f6.eps}
2307: \caption{ As Fig.~\ref{fig:mean_max}, but showing the pre-maximum
2308: (effective day $<$-4 days) high-redshift spectrum versus the local template.
2309: \label{fig:mean_early}}
2310:
2311: \end{figure}
2312:
2313:
2314:
2315: \begin{figure}
2316:
2317: \plotone{f7.eps}
2318: \caption{The mean high-redshift maximum light rest-frame UV SN
2319: spectrum split into two redshift intervals below (black) and above
2320: (red) $z$=0.5.\label{fig:mean_max_z}}
2321: \end{figure}
2322:
2323: \begin{figure}
2324:
2325: \centerline{\includegraphics[width=4.25in]{f8.eps}}
2326:
2327: \caption{A test of the validity of applying a color correction to
2328: those Keck spectra sampled at maximum light. The top panel shows the
2329: mean {\it observed} spectrum for two subsamples split according to
2330: the rest-frame $B-V$ color at maximum light; red corresponds to SNe
2331: with $B-V>-0.057$, blue to those with $B-V<$-0.057. The middle panel
2332: shows the same comparison after applying a
2333: \citet{1989ApJ...345..245C} reddening correction to each individual
2334: spectrum using an $E(B-V)$ estimated from the light curve fits, and
2335: $R_B=4.1$. Varying $R_B$ has a negligible effect on this comparison.
2336: The lower panel shows the comparison after correcting individual
2337: spectra using the SALT color law
2338: \citep{2005A&A...443..781G}.\label{fig:mean_max_col}}
2339: \end{figure}
2340:
2341: \begin{figure}
2342: \plotone{f9.eps}
2343: \caption{The deviation of the 15 maximum-light spectra from the mean
2344: spectrum in $\sigma$-units derived from their error spectra
2345: $\S$~\ref{sec:effect-calibr-uncert}. The top panel shows the
2346: uncorrected spectra, the middle panel the spectra corrected with the
2347: CCM dust law, and the lower panel spectra corrected with the SALT
2348: color law. The horizontal lines denote $\pm$2-$\sigma$
2349: deviations.\label{fig:meandisp}}
2350: \end{figure}
2351:
2352: \begin{figure}
2353: \plotone{f10.eps}
2354: \caption{Various SN~Ia stretch-color relations, with colors measured
2355: directly from the SALT color law corrected Keck spectra. The colors
2356: shown are $U-b$, $UV1-b$ and $UV2-b$ -- see Fig.~\ref{fig:col_lentz}
2357: for filter definitions. SNe are color-coded according to the phase
2358: of the spectrum as defined in Fig.~\ref{fig:redshiftphasehist}. The
2359: black points in the $U-b$ distribution refers to the $U-b$
2360: distribution of local SNe~Ia at maximum light from local SN surveys
2361: \citep[e.g.][]{2006AJ....131..527J}. The error-bars for the SNLS SNe
2362: include propagated uncertainties from host galaxy subtraction
2363: ($\S$~\ref{sec:spectr-calibr-host}).\label{fig:col_stretch}}
2364: \end{figure}
2365:
2366: \begin{figure}
2367:
2368: \includegraphics[width=3.5in]{f11a.eps}
2369: \includegraphics[width=3.5in]{f11b.eps}
2370: \caption{Variation of wavelength with phase for the two UV diagnostic
2371: features marked in Fig.~\ref{fig:col_lentz}; a clear trend is
2372: apparent. The vertical line refers to the full extent of the
2373: metallicity variation predicted by
2374: \citet{2000ApJ...530..966L}.\label{fig:bumps_phase}}
2375: \end{figure}
2376:
2377: \begin{figure}
2378: \plotone{f12.eps}
2379: \caption{Comparison between the wavelength of the \ion{Si}{2} 4130\AA\
2380: feature commonly used as a measure of the photospheric expansion
2381: velocity and that of the UV diagnostic $\lambda_2$ (see
2382: Fig.~\ref{fig:col_lentz} for definition). Spectra taken before
2383: maximum light are indicated with blue symbols; those after maximum
2384: light with red symbols. Numbers indicate the phase in rest-frame
2385: days.\label{fig:siII_lambda2}}
2386: \end{figure}
2387:
2388: \begin{figure}
2389: \plotone{f13.eps}
2390: \caption{Ratio of the wavelength of the \ion{Si}{2} 4130\AA\ feature
2391: to the UV diagnostic $\lambda_2$ versus phase with data points sized
2392: according to the stretch. There is a clear discontinuity in the
2393: behavior after maximum light.\label{fig:ratio_phase}}
2394: \end{figure}
2395:
2396:
2397: \begin{figure}
2398: \plotone{f14.eps}
2399: \caption{The variation of the Keck SN~Ia $UV1-b$ spectral color as a
2400: function of the stellar mass of the SN host galaxy color coded
2401: according to the specific star formation rate. The division is made
2402: at a rate of 10$^{-12} M_{\odot}$ yr$^{-1}$ per unit stellar mass.
2403: The color error-bars include propagated uncertainties from host
2404: galaxy subtraction. The conversion from stellar mass to gas-phase
2405: metallicity using the relation of \citet{2004ApJ...613..898T} is
2406: also shown.\label{fig:col_mass}}
2407: \end{figure}
2408:
2409: \begin{figure}
2410: \includegraphics[width=3.5in]{f15a.eps}
2411: \includegraphics[width=3.5in]{f15b.eps}
2412: \caption{The mean SALT-law color-corrected (left) and color-matched
2413: (right) Keck spectra for SNe~Ia split according to the SN light
2414: curve stretch, $s$. The red spectrum refers to events with
2415: $s\leq1.03$, the blue spectrum to those with
2416: $s>1.03$.\label{fig:mean_max_s}}
2417: \end{figure}
2418:
2419: \begin{figure}
2420: \centering
2421: \includegraphics[width=5.5in,angle=270]{f16.eps}
2422: \caption{Comparison of the wavelength-dependent line and electron
2423: scattering opacity for a typical model in
2424: \citet{2007ApJ...656..661K} at peak SN\,Ia brightness. The data
2425: refers to a depth of 7000 km\,s$^{-1}$. Note the drop in line
2426: opacity with respect to the electron scattering opacity near 4000
2427: \AA\ . This behavior makes the emergent UV flux highly sensitive to
2428: changes in the line opacity whereas the optical and near-IR spectral
2429: regions are largely dominated by electron scattering
2430: opacity.\label{fig:opac_max}}
2431: \end{figure}
2432:
2433: \begin{figure}
2434: \plotone{f17.eps}
2435: \caption{Redshift dependence of various maximum light cross-filter
2436: $k$-corrections differenced to that based on the local template.
2437: Lines represent deviations observed for individual Keck spectra
2438: color-coded according to stretch: red lines refer to events with
2439: $s\leq1.03$, while blue to those with $s>1.03$. The dashed lines
2440: indicates the approximate precision necessary to secure an equation
2441: of state parameter $w$ to a precision of 5\% using $z>$1 SNe\,Ia
2442: (see text for discussion).\label{fig:kcorrections}}
2443: \end{figure}
2444:
2445:
2446:
2447: \end{document}
2448:
2449: