1: % move floats to the end, comment out \usepackage{natbib},
2: % use plotone, & use aastex preprint 12pt
3:
4: %\documentclass[preprint2]{aastex}
5: %\documentclass[12pt,preprint]{aastex}
6: \documentclass[apjl,letterpaper]{emulateapj}
7: \usepackage{natbib}
8: \citestyle{aa}
9:
10: \newcommand{\be}{\begin{equation}}
11: \newcommand{\ee}{\end{equation}}
12: \newcommand{\bea}{\begin{eqnarray}}
13: \newcommand{\eea}{\end{eqnarray}}
14: \newcommand{\mpc}{\mbox{$h^{-1}$Mpc}}
15: \newcommand{\mpci}{\mbox{$h$\,Mpc$^{-1}$}}
16: \newcommand{\bit}[1]{\mbox{\textbf{\emph{#1}}}}
17: \newcommand{\msc}[1]{\mbox{\sc #1}}
18: \newcommand{\phz}{photo-\emph{z}}
19: \newcommand{\Phz}{Photo-\emph{z}}
20:
21: \begin{document}
22:
23: \title{Measuring Baryon Acoustic Oscillations with Millions of
24: Supernovae}
25: \shorttitle{Baryon Acoustic Oscillations from Supernovae}
26: \author{Hu Zhan$^1$, Lifan Wang$^{2}$, Philip Pinto$^3$,
27: and J.~Anthony Tyson$^1$}
28: \shortauthors{Zhan et al.}
29: \affil{$^1$ Department of Physics, University of California,
30: Davis, CA 95616; hzhan@ucdavis.edu, tyson@physics.ucdavis.edu \\
31: $^2$ Department of Physics, Texas A\&M University,
32: College Station, TX 77843; wang@physics.tamu.edu\\
33: $^3$ Department of Astronomy and Steward Observatory,
34: University of Arizona, Tucson, AZ 85721; ppinto@as.arizona.edu}
35:
36: \begin{abstract}
37: Since type Ia Supernovae (SNe) explode in galaxies, they can, in
38: principle, be used as the same tracer of the large-scale structure
39: as their hosts to measure baryon acoustic oscillations (BAOs).
40: To realize this, one must obtain
41: a dense integrated sampling of SNe over a large fraction of
42: the sky, which may only be achievable photometrically with
43: future projects such as the Large Synoptic Survey Telescope.
44: The advantage of SN BAOs is that SNe have more
45: uniform luminosities and more accurate photometric redshifts than
46: galaxies, but the disadvantage is that they are transitory and
47: hard to obtain in large number at high redshift.
48: We find that a half-sky photometric SN survey to redshift $z = 0.8$
49: is able to measure the baryon signature in the SN
50: spatial power spectrum.
51: Although dark energy constraints from SN BAOs are weak,
52: they can significantly improve the results from SN luminosity
53: distances of the same data, and the combination of the two is no
54: longer sensitive to cosmic microwave background priors.
55:
56: \end{abstract}
57:
58: \keywords{cosmological parameters --- distance scale ---
59: large-scale structure of universe}
60:
61: \section{Introduction}
62: Type Ia supernovae\footnote{We consider only type Ia SNe in this
63: \emph{Letter}.} (SNe) have become a mature tool for studying
64: the cosmic expansion history
65: \citep[e.g.,][]{phillips93,riess98, perlmutter99a}.
66: A number of SN surveys, such as
67: the Sloan Digital Sky Survey (SDSS) II \citep{frieman08},
68: the Supernova Legacy Survey \citep{astier06},
69: and the ESSENCE Supernova Survey \citep{miknaitis07}
70: are being carried out to improve the statistics and our
71: understanding of the systematics. Moreover, the SN technique
72: will be an integral part of almost every proposed dark
73: energy survey including the Large Synoptic Survey
74: Telescope\footnote{See \url{http://www.lsst.org}.}
75: \citep[LSST, see][]{tyson06} and the Joint Dark Energy Mission.
76:
77: The conventional SN technique, measuring only the relative luminosity
78: distance, $D_{\rm L}$,
79: is subject to degeneracies between cosmological parameters.
80: For example, the SN constraint on the dark energy equation-of-state
81: (EOS, $w$) parameter $w_{\rm a}$, as defined
82: by $w(z) = w_0 + w_{\rm a}z/(1+z)$, is sensitive to
83: the prior on the mean curvature of the universe
84: \citep*{linder05b, knox06c}. The reason is that the
85: response of the relative distance to a variation in $w_{\rm a}$
86: resembles that to a variation in the mean curvature and
87: that the SN technique lacks the calibration of absolute distances
88: \citep{zhan06e}. Even for a flat universe with $w(z)\equiv w_0$,
89: the SN constraint on $w_0$ can be tightened considerably if the
90: matter density is known to high precision \citep{frieman03}.
91: Such priors may come from other techniques, such as the cosmic microwave
92: background (CMB), weak lensing, and baryon acoustic oscillations
93: \citep*[BAOs,][]{eisenstein98, cooray01b, blake03, hu03b, linder03b,
94: seo03}. It has indeed been demonstrated that the latter
95: three techniques are highly complementary to the SN $D_{\rm L}$
96: technique \citep{frieman03, seo03, knox06c}.
97:
98: Since SNe explode in galaxies, their distribution bears the
99: BAO imprint as well. To measure the SN spatial power spectrum, one
100: needs the angular position and redshift of each SN, not its luminosity.
101: Hence, the SN BAO technique does not suffer from uncertainties in
102: the SN standard candle, which constitute the largest unknown in the
103: $D_{\rm L}$ measurements.
104: Nevertheless, the narrow range of the SN intrinsic
105: luminosity reduces the effect of Malmquist-like biases
106: and luminosity evolution.
107: The SN rate traces the mass and star formation of the host galaxies
108: with a time delay \citep{sullivan06}.
109: This means that SNe have a different clustering
110: bias than galaxies that are selected by their luminosity or color.
111: Finally, SNe have rich and time-varying spectral features for
112: accurate estimation of photometric redshifts (\phz{}s)
113: \citep*{pinto04, wang07a, wang07b}, which is helpful for measuring
114: BAOs from a photometric survey.
115:
116: There have been discussions of using the SN weak lensing magnification
117: \citep*{cooray06} and nearby SN peculiar velocities \citep*{hannestad07}
118: to probe the large-scale structure. We focus on photometric SN surveys
119: for BAOs in this \emph{Letter} and note in passing that the SN weak
120: lensing technique is more limited by shot noise than the SN BAO
121: technique \citep{zhan06d, zhan06e} and that the SN peculiar
122: velocity technique requires precise redshift and distance measurements.
123:
124: For the BAO technique to be useful, one must survey a large volume
125: at a sufficient sampling density as uniformly as possible.
126: Although SN events are rare, the spatial density of SNe
127: accumulated over several years will be
128: comparable to the densities targeted for
129: future spectroscopic galaxy BAO surveys.
130:
131: \section{Photometric Supernova Surveys}
132:
133: We assume two \phz{} SN survey models: a shallow one
134: (S20k) that covers 20,000 deg$^2$ to $z = 0.8$ for 10 years,
135: and a deep one (D2k) that covers 2000 deg$^2$ to $z = 1.2$
136: for 5 years. The S20k data may be extracted as one of the many
137: products from the proposed LSST, and a dedicated ground-based
138: SN survey will likely be sufficient to produce the D2k data.
139: Since there is no precedent for such \phz{} SN
140: BAO surveys, we have to make crude estimates for the fiducial
141: survey parameters and vary them to cover a wider range.
142:
143: We calculate the observer-frame SN rate
144: based on the rest-frame SN rates
145: \citep{cappellaro99,hardin00,pain02,madgwick03,tonry03,blanc04,
146: dahlen04,neill06}.
147: The resulting number of SNe per steradian per unit redshift
148: per year (observer frame) is roughly
149: \[
150: \frac{dn}{d\Omega dz dt} \propto
151: \big(e^{3.12 z^{2.1}} - 1\big) \times \left\{
152: \begin{array}{lr}
153: 1 & z \le 0.5 \\
154: e^{-(z - 0.5)^2/2a^2} & z > 0.5
155: \end{array} \right. ,
156: \]
157: where the first term on the r.h.s.~fits the observed SN rate at
158: $z \le 0.55$, and the additional exponential term with $a = 0.125$
159: ($0.21$) cuts off the distribution at $z \sim 0.8$ ($1.2$)
160: for S20k (D2k).
161: We take the efficiency with which S20k (D2k) will produce well-observed
162: SN light-curves suitable for accurate \phz{}s from
163: this distribution as 50\% (100\%).
164: The accumulated SN surface density is then
165: $\Sigma = 370$ and 980 deg$^{-2}$ for S20k (7.4 million SNe in 10
166: years) and D2k (2 million SNe in 5 years), respectively.
167:
168: We adopt a pedagogical convention and model the \phz{} error as a
169: Gaussian with rms $\sigma_z=\sigma_{z{\rm0}}(1+z)$ and zero
170: bias\footnote{Any known \phz{} bias can be calibrated out in advance,
171: so only the uncertainty of the bias matters.} $\delta z = 0$.
172: We assign for S20k $\sigma_{z{\rm 0}}=0.02$ and for D2k
173: $\sigma_{z{\rm 0}}=0.01$, which are achievable with simple
174: \phz{} algorithms \citep{pinto04,wang07b}.
175:
176: Table~\ref{tab:model} summarizes the SN surveys. It includes
177: two additional quantities: the SN clustering bias $b$, and the cut-off
178: wavenumber $k_{\rm max}$ for the BAO analysis,
179: which is set to reduce the impact of
180: nonlinear growth (see \citealt{eisenstein06}
181: for recovering high-$k$ information from spectroscopic surveys).
182:
183: \section{Forecast Method}
184:
185: We use a modified forecast tool {\sc cswab}\footnote{Available
186: at \url{http://hzhan.net/soft/.}}
187: \citep{zhan06d} to assess the cosmological constraints from
188: the clustering of SNe. In summary, the SN power spectrum in the
189: $i$th redshift bin $P_i(\bit{k}_{\rm f})$
190: reads \citep{seo03}
191: \bea \label{eq:psn}
192: P_i(k_{{\rm f}\perp},k_{{\rm f}\parallel})
193: =&&\frac{D_{{\rm f},i}^2 H_i}{D_i^2 H_{{\rm f},i}}
194: \left(1+\beta_i\mu^2\right)^2 b_i^2 G_i^2 \mathcal{P}(k) \\ && \times
195: \exp\left[-(c \sigma_{z,i} k_{\parallel}/H_i)^2\right] + s_i,
196: \nonumber
197: \eea
198: where the subscript f denotes quantities in a reference
199: cosmological model (the same as the fiducial model in the
200: forecast), $D$ is the angular diameter distance,
201: $H$ is the Hubble parameter, $\beta$ is the redshift distortion
202: parameter, $\mu = k_\parallel/k$,
203: $G$ is the linear growth factor, $\mathcal{P}(k)$ is the matter power
204: spectrum at $z = 0$, and $s=n^{-1}$ is the shot noise.
205: The true wavenumbers $k_\perp$ and $k_\parallel$ are related to
206: the references by
207: $k_\perp = k_{\rm f\perp} D_{{\rm f}} / D$ and
208: $k_\parallel = k_{\rm f\parallel} H / H_{{\rm f}}$.
209: Note that for a \phz{} survey with $\sigma_z \gtrsim 0.01(1+z)$,
210: the radial BAO information is essentially lost, e.g., the power
211: spectrum (excluding the shot noise) at the fundamental mode of BAOs
212: $k_\parallel\sim 2\pi/150\,\mbox{Mpc}^{-1}$ is suppressed by a factor
213: of 80 or more at $z = 0.6$, though a \phz{} rms of
214: $0.003(1+z)$ would recover that information.
215:
216: {\sc cswab} uses the Fisher matrix to estimate the error bounds
217: of the parameters \citep{tegmark97b}:
218: \be \label{eq:FBAO}
219: F_{\alpha\beta}^{\rm BAO} = \sum_{i} \frac{V_{{\rm f},i}}{2} \int
220: \frac{\partial \ln P_i(\bit{k}_{\rm f})}{\partial q_\alpha}
221: \frac{\partial \ln P_i(\bit{k}_{\rm f})}{\partial q_\beta}
222: \frac{d \bit{k}_{\rm f}}{(2 \pi)^3},
223: \ee
224: where $V_{\rm f}$
225: is the comoving survey volume and
226: $\{q_\alpha\}$ is the parameter set, which includes $b_i$,
227: $\sigma_{z,i}$, $\delta z_i$, and $s_i$ of each bin and 8
228: cosmological parameters: $w_0$, $w_{\rm a}$, $\omega_{\rm m}$
229: (the matter density), $\omega_{\rm b}$ (the baryon density),
230: $\theta_{\rm s}$ (the angular size of the sound horizon at the last
231: scattering surface), $\Omega_{\rm k}$ (the curvature parameter),
232: $n_{\rm s}$ (the scalar spectral index),
233: and $\Delta_{\rm R}^2$ (the
234: normalization of the primordial curvature power spectrum).
235: The minimum marginalized error of $q_\alpha$ is $\sigma(q_\alpha) =
236: (F^{-1})_{\alpha\alpha}^{1/2}$. Independent Fisher matrices are
237: additive; a prior on $q_\alpha$, $\sigma_{\rm P}(q_\alpha)$,
238: can be introduced via $F_{\alpha\alpha}^{\rm new} = F_{\alpha\alpha}
239: +\sigma_{\rm P}^{-2}(q_\alpha)$.
240: The fiducial values of $b_i$, $\sigma_{z,i}$, and $s_i^{-1}$
241: are listed in Table~\ref{tab:model}, $\delta z_i = 0$,
242: and ($w_0$, $w_{\rm a}$, $\omega_{\rm m}$,
243: $\omega_{\rm b}$, $\theta_{\rm s}$, $\Omega_{\rm k}$, $n_{\rm s}$,
244: $\Delta_{\rm R}^2) = (-1$, 0, 0.127, 0.0223,
245: 0.596$^\circ$, 0, 0.951, $2.0\times 10^{-9}$) from the \emph{WMAP}
246: 3-year results \citep{spergel07}.
247: Unless stated otherwise, we always include fairly weak
248: priors $\sigma_{\rm P}(\ln b_i) = 0.3$,
249: $\sigma_{\rm P}(\ln \Delta_{\rm R}^2) = 0.2$,
250: $\sigma_{\rm P}(\ln \theta_{\rm s}) =
251: \sigma_{\rm P}(n_{\rm s}) = 0.05$, and
252: $\sigma_{\rm P}(\delta z_i) = 2^{-1/2} \sigma_{\rm P}(\sigma_{z,i}) =
253: 0.25\sigma_{z,i}$.
254:
255: \begin{deluxetable}{c c c c c c c}
256: \tablewidth{0pt}
257: \tablecaption{Supernova Survey Parameters
258: \label{tab:model}}
259: \tablehead{\colhead{} & \colhead{Area} & & \colhead{$n$} & &
260: & \colhead{$k_{\rm max}$} \\
261: \colhead{Survey} & \colhead{deg$^2$} & \colhead{$z$} &
262: \colhead{$h^3\mbox{Mpc}^{-3}$} & \colhead{$\sigma_z$} &
263: \colhead{$b$} & \colhead{\mpci}}
264: \startdata
265: & & 0.3 & $4.2\times 10^{-4}$ & 0.026 & 1.18 & 0.17 \\
266: S20k & 20000 & 0.5 & $6.3\times 10^{-4}$ & 0.030 & 1.30 & 0.20 \\
267: & & 0.7 & $3.4\times 10^{-4}$ & 0.034 & 1.42 & 0.24 \\
268: \cline{1-7} \\[-2ex]
269: & & 0.3 & $4.2\times 10^{-4}$ & 0.013 & 1.18 & 0.17 \\
270: & & 0.5 & $6.6\times 10^{-4}$ & 0.015 & 1.30 & 0.20 \\
271: D2k & 2000 & 0.7 & $7.8\times 10^{-4}$ & 0.017 & 1.42 & 0.24 \\
272: & & 0.9 & $5.0\times 10^{-4}$ & 0.019 & 1.54 & 0.29 \\
273: & & 1.1 & $1.8\times 10^{-4}$ & 0.021 & 1.66 & 0.34 \\[-2ex]
274: \enddata
275: \tablecomments{The redshift is central to each bin, and the width of
276: each bin is $\Delta z = 0.2$.
277: }
278: \end{deluxetable}
279:
280: To show the complementarity between the SN BAO and $D_{\rm L}$
281: techniques, we include the
282: $D_{\rm L}$ constraints at the end of Section~\ref{sec:cons}.
283: The Fisher matrix for SN $D_{\rm L}$ is
284: \bea \label{eq:FDL}
285: F_{\alpha\beta}^{\rm D_L} &=& \frac{1}{\sigma_{\rm m}^2}\int
286: n_{\rm p}(z_{\rm p})
287: \frac{\partial \bar{m}_{\rm p}(z_{\rm p})}{\partial q_\alpha}
288: \frac{\partial \bar{m}_{\rm p}(z_{\rm p})}{\partial q_\beta}
289: dz_{\rm p} \\ \nonumber
290: \bar{m}_{\rm p} &=& \int
291: \left[5 \log D_{\rm L}(z) + M + e_1 z + e_2 z^2\right]
292: p(z | z_{\rm p}) dz,
293: \eea
294: where $\sigma_{\rm m} = 0.15$ is the scatter of the SN apparent
295: magnitude, the subscript p signifies \phz{} space, $M$ is the
296: SN absolute magnitude, $e_1$ and $e_2$ account for possible
297: SN evolution, and $p(z|z_{\rm p})$ is the probability density
298: of a SN at $z$ given its \phz{} $z_{\rm p}$.
299: Following \citet{albrecht06b},
300: we impose a prior of $0.015$ on $e_1$ and $e_2$ and let $M$ float.
301: We also include $z < 0.2$ SNe in $F^{\rm D_L}$, because they are
302: important for cosmology with $D_{\rm L}$
303: \citep{linder06b}.
304:
305: \begin{deluxetable}{l c c c c c c}
306: \tablewidth{0pt}
307: \tablecaption{Marginalized $1\sigma$ Errors on Selected Cosmological
308: Parameters from SN BAOs \label{tab:cons}}
309: \tablehead{ & & & & \colhead{$\ln \omega_{\rm m}$} &
310: \colhead{$\ln \omega_{\rm b}$} & \colhead{$\Omega_{\rm k}$}
311: \\[.4ex] \cline{5-7} \\[-1.2ex]
312: \colhead{Surveys} & \colhead{$w_0$} & \colhead{$w_{\rm a}$} &
313: \colhead{$w_{\rm p}$} & \multicolumn{3}{c}{$\times10^{-2}$}}
314: \startdata
315: S20k & 0.85 & 2.9\phn & 0.22\phn & \phn\phd23 & \phn\phd38 & 1.8\phn \\
316: S20k+\emph{Planck} & 0.66 & 1.8\phn & 0.12\phn & 0.83 & 0.91 & 0.55 \\
317: \cline{1-7} \\[-2ex]
318: D2k & 0.96 & 3.0\phn & 0.19\phn & \phn\phd25 & \phn\phd42 & 2.2\phn \\
319: D2k+\emph{Planck} & 0.74 & 2.0\phn & 0.13\phn & 0.85 & 0.91 & 0.60 \\[-2ex]
320: \enddata
321: \tablecomments{The error of the pivot EOS $w_{\rm p}$ equals
322: that of a constant EOS, and the Dark Energy Task Force figure of
323: merit \citep{albrecht06b} equals
324: $0.052[\sigma(w_{\rm a}) \times \sigma(w_{\rm p})]^{-1}$.
325: }
326: \end{deluxetable}
327:
328: \section{Cosmological Constraints} \label{sec:cons}
329: The marginalized $1\,\sigma$ errors of a subset of the cosmological
330: parameters are given in Table~\ref{tab:cons}.
331: The two \phz{} SN BAO surveys place rather weak constraints
332: on the dark energy EOS even with \emph{Planck} priors.
333: Nevertheless, they provide moderate constraints on the matter
334: density $\omega_{\rm m}$, baryon density $\omega_{\rm b}$, and
335: curvature parameter $\Omega_{\rm k}$, which are helpful for the SN
336: $D_{\rm L}$ technique. The smaller area of D2k is compensated by its
337: greater depth and better \phz{}s, so that D2k performs nearly as
338: well as S20k. Spectroscopic BAO surveys of similar characteristics
339: can reduce the error on $\omega_{\rm m}$, $\omega_{\rm b}$, and
340: $\Omega_{\rm k}$ by a factor of 2 and significantly more on $w_0$
341: and $w_{\rm a}$.
342:
343: The baryon signature has been detected at $\sim 3\,\sigma$ level [and
344: $\sigma(\ln\omega_{\rm m})\sim 10\%$] from SDSS
345: Luminous Red Galaxies, both spectroscopically
346: \citep{eisenstein05} and photometrically \citep{blake07,padmanabhan07}.
347: These detections assume a flat universe with a cosmological constant
348: and a fixed scalar spectral index $n_{\rm s}$.
349: Under the same assumptions, S20k BAO can constrain $\omega_{\rm m}$ to
350: 8\% and $\omega_{\rm b}$ to 15\%. If $\omega_{\rm b}$ is fixed as well
351: \citep[as in][]{eisenstein05}, S20k BAO can achieve
352: $\sigma(\ln\omega_{\rm m})=1.5\%$.
353:
354: Theoretical uncertainties in the redshift distortion, nonlinear
355: evolution, and scale-dependent clustering bias can be important
356: to BAOs \citep{seo05, white05, guzik07}.
357: For a simple test, we replace the linear redshift
358: distortion factor $(1 + \beta \mu^2)^2$ \citep{kaiser87}
359: in equation (\ref{eq:psn}) with
360: $1 + 2 e \beta \mu^2 + f \beta^2 \mu^4$,
361: where $e=f=1$ are parameters accounting for our uncertain knowledge
362: of the redshift distortion \citep{scoccimarro04}.
363: We take priors $\sigma_{\rm P}(e) = 0.05$ and
364: $\sigma_{\rm P}(f) = 0.1$ and find a less than 1\% change to
365: the results in Table~\ref{tab:cons}.
366: Spectroscopic (galaxy) surveys with the same redshift distribution
367: and coverage will see $w_0$ and $w_{\rm a}$ errors doubled in
368: this test, because they have more information to lose.
369:
370: \begin{figure}
371: \centering
372: %\epsscale{0.46}
373: \plotone{f1}
374: \caption[f1]{The EP $\sigma(w_{\rm p}) \times \sigma(w_{\rm a})$
375: of S20k (thick lines) and D2k (thin lines) as a function of the rms
376: \phz{} error $\sigma_z$. The priors on the \phz{} biases are
377: taken to be $0.5\sigma_z$ (solid lines) and $0.05\sigma_z$ (dashed
378: lines), which correspond to calibrations with 4 and 400 spectra
379: per redshift bin, respectively, in the Gaussian case.
380: To reduce the dimensions, we peg the prior on the \phz{} rms
381: to that on the \phz{} bias: $\sigma_{\rm P}(\sigma_z) =
382: \sqrt{2} \sigma_{\rm P}(\delta z)$.
383: For comparison, LSST weak lensing, galaxy BAOs, and the two combined
384: will achieve EP $\sim 0.01$, $0.02$, and $0.002$, respectively
385: \citep{zhan06d}.
386: \label{fig:conz}}
387: \end{figure}
388:
389: \begin{figure}
390: \centering
391: \epsscale{0.93}
392: \plotone{f2}
393: \caption[f2]{The EP $\sigma(w_{\rm p}) \times \sigma(w_{\rm a})$
394: (solid lines) as a function of the survey area and SN surface
395: density $\Sigma$ (accumulated over 10 years).
396: The EP contours are spaced logarithmically. The SN distribution is
397: scaled from S20k (solid circle, 7.4 million SNe) by $\Sigma$.
398: \Phz{} parameters are the same as those of S20k, and the \emph{Planck}
399: priors are assumed.
400: The total number of SNe is held constant
401: along each dotted line, which is spaced at factors of $\sqrt{2}$,
402: \label{fig:conn}}
403: \end{figure}
404:
405: Figure~\ref{fig:conz} explores the dependence of the SN BAO error
406: product $\sigma(w_{\rm a}) \times \sigma(w_{\rm p})$ (EP) on
407: the \phz{} rms $\sigma_z$ and prior $\sigma_{\rm P}(\sigma_{z})$.
408: At a large $\sigma_z$, there is little radial BAO information,
409: so that the Fisher matrix
410: $F^{\rm BAO}$ scales roughly with the $k$-space volume, which
411: is proportional to $H/c\sigma_z$. This leads to a scaling
412: $\mbox{EP}\propto \sigma_z$ when $\sigma_z$ and $\delta z$ are
413: known accurately (dashed lines), consistent with \citet{seo03}
414: and \citet{zhan06d}. When $\sigma_z \lesssim 0.008(1+z)$,
415: the radial BAOs become available, and,
416: hence, the EP slope steepens at smaller $\sigma_z$.
417:
418: Figure~\ref{fig:conn} shows the EP contours (solid lines)
419: of the shallow survey as the projected SN number
420: density and survey area vary.
421: Over-plotted in dotted lines are contours of the total number of SNe.
422: If the number of SNe is taken as a crude proxy for effort, one can
423: optimize the survey (in terms of dark energy constraints) by searching
424: the minimum EP on the constant-effort curve. This means that for
425: $\Sigma \gtrsim 200\,\mbox{deg}^{-2}$ and area less than
426: $30,000\,\mbox{deg}^2$, one would always choose the maximum survey
427: area possible for SN BAO as opposed to accumulating more SNe in a
428: smaller area.
429:
430: \begin{figure}
431: \centering
432: %\epsscale{0.45}
433: \plotone{f3}
434: \caption[f3]{Marginalized $1\,\sigma$ error contours of the dark energy
435: EOS parameters $w_0$ and $w_{\rm a}$ from S20k SN BAOs (dashed line,
436: $\mbox{EP}=0.22$ with \emph{Planck}), luminosity distances (dotted line,
437: $\mbox{EP}=0.064$ with \emph{Planck}), and the two combined with
438: (shaded area, $\mbox{EP}=0.012$) and without (solid line,
439: $\mbox{EP}=0.018$) \emph{Planck}.
440: \label{fig:snb}}
441: \end{figure}
442:
443: Finally, we illustrate the complementarity between the SN BAO
444: and $D_{\rm L}$ techniques in Figure~\ref{fig:snb}.
445: Although the S20k BAO technique (dashed line) does not place
446: useful constraints on dark energy, its combination (shaded area)
447: with the $D_{\rm L}$ technique (dotted line) reduces the EP by a
448: factor of 5.5 over the $D_{\rm L}$-alone EP. Moreover, the combined
449: result (solid line) is not very sensitive to the CMB priors, because
450: the BAO technique can provide adequate constraints on cosmological
451: parameters, such as $\omega_{\rm m}$ and $\Omega_{\rm k}$, for the
452: $D_{\rm L}$ technique.
453:
454: \section{Discussion and Conclusions}
455:
456: We have demonstrated that \phz{} SN data can be used to measure
457: BAOs and to constrain cosmological parameters.
458: The BAO constraints on the matter density and the baryon
459: density are sensitive to the priors on the curvature and the scalar
460: spectral index but not to the dark energy parameters.
461: The dark energy constraints from the SN BAO technique alone are not
462: meaningful. However, a combination of the SN BAO and $D_{\rm L}$
463: techniques reduces the EP considerably, and the \emph{Planck} priors
464: are no longer crucial. The SN BAO results in Section~\ref{sec:cons}
465: are also applicable to \phz{} galaxy BAOs.
466: We note that
467: \phz{} errors are a large uncertainty for \phz{} SN cosmology.
468: Although our assumption about them is conservative
469: compared to the results in \citet{pinto04}, further studies are
470: needed to make realistic forecasts.
471:
472: Long and non-uniform cadence may result in uneven sampling of
473: the SN spatial distribution.
474: Fortunately, LSST will be likely to always catch the SNe
475: (especially high-$z$ ones) at
476: their maximum owing to its fast sky coverage and rapid sampling.
477: Furthermore, the effect of the cadence on the SN
478: depth can be simulated and determined, and methods of correcting
479: for uneven depths in galaxy surveys can be applied to the SN data.
480:
481: A spectroscopic SN BAO survey will be impractical, because one would
482: have to revisit the sky many times spectroscopically over thousands of
483: square degrees to catch the SNe that occur at different times.
484: However,
485: LSST will be able to obtain SNe in the millions over half the sky
486: photometrically.
487: This opens a window for applying the BAO technique
488: to SNe and achieving more robust constraints with \phz{} SN data.
489: Since this SN BAO analysis requires no additional observations than
490: doing the SN $D_{\rm L}$ analysis alone, it should be a feature of
491: all large-area SN cosmology analyses.
492: Moreover, the SN data can also help calibrate the host galaxy
493: \phz{}s and the \phz{} error distribution of other
494: galaxies through the cross-correlation method
495: \citep{schneider06,zhan06d,newman08}.
496:
497: \acknowledgments
498: We thank J.~Frieman, L.~Knox, and M.~Wood-Vasey for useful
499: conversations. This work was partially supported by a UC Davis
500: Academic Federation Innovative Developmental Award.
501:
502: \begin{thebibliography}{46}
503: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
504:
505: \bibitem[{{Albrecht} {et~al.}(2006)}]{albrecht06b}
506: {Albrecht}, A. {et~al.} 2006, ArXiv e-prints, astro-ph/0609591
507:
508: \bibitem[{{Astier} {et~al.}(2006)}]{astier06}
509: {Astier}, P. {et~al.} 2006, \aap, 447, 31
510:
511: \bibitem[{{Blake} {et~al.}(2007)}]{blake07}
512: {Blake}, C., {Collister}, A., {Bridle}, S., \& {Lahav}, O. 2007, \mnras, 374,
513: 1527
514:
515: \bibitem[{{Blake} \& {Glazebrook}(2003)}]{blake03}
516: {Blake}, C., \& {Glazebrook}, K. 2003, \apj, 594, 665
517:
518: \bibitem[{{Blanc} {et~al.}(2004)}]{blanc04}
519: {Blanc}, G. {et~al.} 2004, \aap, 423, 881
520:
521: \bibitem[{{Cappellaro} {et~al.}(1999){Cappellaro}, {Evans}, \&
522: {Turatto}}]{cappellaro99}
523: {Cappellaro}, E., {Evans}, R., \& {Turatto}, M. 1999, \aap, 351, 459
524:
525: \bibitem[{{Cooray} {et~al.}(2006){Cooray}, {Holz}, \& {Huterer}}]{cooray06}
526: {Cooray}, A., {Holz}, D.~E., \& {Huterer}, D. 2006, \apjl, 637, L77
527:
528: \bibitem[{{Cooray} {et~al.}(2001)}]{cooray01b}
529: {Cooray}, A., {Hu}, W., {Huterer}, D., \& {Joffre}, M. 2001, \apjl, 557, L7
530:
531: \bibitem[{{Dahlen} {et~al.}(2004)}]{dahlen04}
532: {Dahlen}, T. {et~al.} 2004, \apj, 613, 189
533:
534: \bibitem[{{Eisenstein} {et~al.}(1998){Eisenstein}, {Hu}, \&
535: {Tegmark}}]{eisenstein98}
536: {Eisenstein}, D.~J., {Hu}, W., \& {Tegmark}, M. 1998, \apjl, 504, L57
537:
538: \bibitem[{{Eisenstein} {et~al.}(2007)}]{eisenstein06}
539: {Eisenstein}, D.~J., {Seo}, H.-J., {Sirko}, E., \& {Spergel}, D.~N. 2007, \apj,
540: 664, 675
541:
542: \bibitem[{{Eisenstein} {et~al.}(2005)}]{eisenstein05}
543: {Eisenstein}, D.~J. {et~al.} 2005, \apj, 633, 560
544:
545: \bibitem[{{Frieman} {et~al.}(20078)}]{frieman08}
546: {Frieman}, J.~A. {et~al.} 2008, \aj, 138, 338
547:
548: \bibitem[{{Frieman} {et~al.}(2003)}]{frieman03}
549: {Frieman}, J.~A., {Huterer}, D., {Linder}, E.~V., \& {Turner}, M.~S. 2003,
550: \prd, 67, 083505
551:
552: \bibitem[{{Guzik} {et~al.}(2007){Guzik}, {Bernstein}, \& {Smith}}]{guzik07}
553: {Guzik}, J., {Bernstein}, G., \& {Smith}, R.~E. 2007, \mnras, 375, 1329
554:
555: \bibitem[{{Hannestad} {et~al.}(2007){Hannestad}, {Haugboelle}, \&
556: {Thomsen}}]{hannestad07}
557: {Hannestad}, S., {Haugboelle}, T., \& {Thomsen}, B. 2007, submitted to JCAP,
558: astro-ph/0705.0979
559:
560: \bibitem[{{Hardin} {et~al.}(2000)}]{hardin00}
561: {Hardin}, D. {et~al.} 2000, \aap, 362, 419
562:
563: \bibitem[{{Hu} \& {Haiman}(2003)}]{hu03b}
564: {Hu}, W., \& {Haiman}, Z. 2003, \prd, 68, 063004
565:
566: \bibitem[{{Kaiser}(1987)}]{kaiser87}
567: {Kaiser}, N. 1987, \mnras, 227, 1
568:
569: \bibitem[{{Knox} {et~al.}(2006){Knox}, {Song}, \& {Zhan}}]{knox06c}
570: {Knox}, L., {Song}, Y.-S., \& {Zhan}, H. 2006, \apj, 652, 857
571:
572: \bibitem[{{Linder}(2003)}]{linder03b}
573: {Linder}, E.~V. 2003, \prd, 68, 083504
574:
575: \bibitem[{{Linder}(2005)}]{linder05b}
576: ---. 2005, Astroparticle Physics, 24, 391
577:
578: \bibitem[{{Linder}(2006)}]{linder06b}
579: ---. 2006, \prd, 74, 103518
580:
581: \bibitem[{{Madgwick} {et~al.}(2003)}]{madgwick03}
582: {Madgwick}, D.~S., {Hewett}, P.~C., {Mortlock}, D.~J., \& {Wang}, L. 2003,
583: \apjl, 599, L33
584:
585: \bibitem[{{Miknaitis} {et~al.}(2007)}]{miknaitis07}
586: {Miknaitis}, G. {et~al.} 2007, \apj, 666, 674
587:
588: \bibitem[{{Neill} {et~al.}(2006)}]{neill06}
589: {Neill}, J.~D. {et~al.} 2006, \aj, 132, 1126
590:
591: \bibitem[{{Newman}(2008)}]{newman08}
592: {Newman}, J. 2008, submitted to \apj
593:
594: \bibitem[{{Padmanabhan} {et~al.}(2007)}]{padmanabhan07}
595: {Padmanabhan}, N. {et~al.} 2007, \mnras, 378, 852
596:
597: \bibitem[{{Pain} {et~al.}(2002)}]{pain02}
598: {Pain}, R. {et~al.} 2002, \apj, 577, 120
599:
600: \bibitem[{{Perlmutter} {et~al.}(1999)}]{perlmutter99a}
601: {Perlmutter}, S. {et~al.} 1999, \apj, 517, 565
602:
603: \bibitem[{{Phillips}(1993)}]{phillips93}
604: {Phillips}, M.~M. 1993, \apjl, 413, L105
605:
606: \bibitem[{{Pinto} {et~al.}(2004){Pinto}, {Smith}, \& {Garnavich}}]{pinto04}
607: {Pinto}, P.~A., {Smith}, C.~R., \& {Garnavich}, P.~M. 2004, in Bulletin of the
608: American Astronomical Society, Vol.~36, 1530--+
609:
610: \bibitem[{{Riess} {et~al.}(1998)}]{riess98}
611: {Riess}, A.~G. {et~al.} 1998, \aj, 116, 1009
612:
613: \bibitem[{{Schneider} {et~al.}(2006)}]{schneider06}
614: {Schneider}, M., {Knox}, L., {Zhan}, H., \& {Connolly}, A. 2006, \apj, 651, 14
615:
616: \bibitem[{{Scoccimarro}(2004)}]{scoccimarro04}
617: {Scoccimarro}, R. 2004, \prd, 70, 083007
618:
619: \bibitem[{{Seo} \& {Eisenstein}(2003)}]{seo03}
620: {Seo}, H., \& {Eisenstein}, D.~J. 2003, \apj, 598, 720
621:
622: \bibitem[{{Seo} \& {Eisenstein}(2005)}]{seo05}
623: {Seo}, H.-J., \& {Eisenstein}, D.~J. 2005, \apj, 633, 575
624:
625: \bibitem[{{Spergel} {et~al.}(2007)}]{spergel07}
626: {Spergel}, D.~N. {et~al.} 2007, \apjs, 170, 377
627:
628: \bibitem[{{Sullivan} {et~al.}(2006)}]{sullivan06}
629: {Sullivan}, M. {et~al.} 2006, \apj, 648, 868
630:
631: \bibitem[{{Tegmark}(1997)}]{tegmark97b}
632: {Tegmark}, M. 1997, Physical Review Letters, 79, 3806
633:
634: \bibitem[{{Tonry} {et~al.}(2003)}]{tonry03}
635: {Tonry}, J.~L. {et~al.} 2003, \apj, 594, 1
636:
637: \bibitem[{{Tyson}(2006)}]{tyson06}
638: {Tyson}, J.~A. 2006, in AIP Conf. Ser., Vol. 870, Intersections of Particle and
639: Nuclear Physics, 44--52
640:
641: \bibitem[{{Wang}(2007)}]{wang07a}
642: {Wang}, Y. 2007, \apjl, 654, L123
643:
644: \bibitem[{{Wang} {et~al.}(2007){Wang}, {Narayan}, \& {Wood-Vasey}}]{wang07b}
645: {Wang}, Y., {Narayan}, G., \& {Wood-Vasey}, M. 2007, \mnras, 382, 377
646:
647: \bibitem[{{White}(2005)}]{white05}
648: {White}, M. 2005, Astroparticle Physics, 24, 334
649:
650: \bibitem[{{Zhan}(2006)}]{zhan06d}
651: {Zhan}, H. 2006, J. of Cosmo. and Astro-Part. Phys., 8, 8
652: (astro-ph/0605696)
653:
654: \bibitem[{{Zhan} \& {Knox}(2006)}]{zhan06e}
655: {Zhan}, H., \& {Knox}, L. 2006, ArXiv e-prints, astro-ph/0611159
656:
657: \end{thebibliography}
658:
659: \end{document}
660: