1:
2: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3: %
4: % Optimizing future imaging and spectroscopic surveys of galaxies
5: % to confront dark energy and modified gravity models
6: %
7: % Time-stamp: <07/04/19 17:24:00 suto>
8: % Time-stamp: <07/04/03 15:07:00 yamamoto>
9: % Time-stamp: <07/04/17 15:36:00 drp>
10: % Time-stamp: <07/04/19 17:07:00 yamamoto>
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: %\documentclass[useAMS,usenatbib]{mn2e}
13: %\documentclass[12pt,preprint]{aastex}
14: %\documentclass[11pt,preprint]{aastex}
15: %\documentclass[11pt,preprint]{aastex}
16: %\documentclass[11pt,preprint]{revtex4}
17: \documentclass[tightenlines,eqsecnum,floats,aps,amsmath,amssymb,nofootinbib,prd,showpacs]{revtex4}
18: \usepackage{amsmath,amssymb,amsfonts}
19: \usepackage{graphicx}
20: \usepackage{enumerate} % advanced enumerate environment
21: \usepackage{colordvi} % for color text
22: \usepackage{bm}% bold math
23: \usepackage{subfigure}
24: \usepackage{epsf}
25: \newcommand{\simgt}{\lower.5ex\hbox{$\; \buildrel > \over \sim \;$}}
26: \newcommand{\simlt}{\lower.5ex\hbox{$\; \buildrel < \over \sim \;$}}
27: \newcommand{\calD}{{\cal D}}
28: \def\comment#1{\par\noindent\llap{$\Rightarrow$\enskip}{\bf #1}\par}
29: \def\texp{{t_{\rm exp}}}
30: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31: \begin{document}
32: \title{Optimizing future imaging
33: survey
34: of galaxies to confront dark energy and modified gravity models}
35:
36: \author{Kazuhiro Yamamoto}
37: \affiliation{Graduate School of Science,
38: Hiroshima University,
39: Higashi-Hiroshima, 735-8526, Japan}
40: %\email{kazuhiro@hiroshima-u.ac.jp}
41: \author{David Parkinson}
42: \affiliation{Astronomy Centre, University of Sussex,
43: Brighton BN1 9QH, United Kingdom}
44: \author{Takashi Hamana}
45: \affiliation{National Astronomical Observatory of Japan
46: 2-21-1 Osawa, Mitaka, Tokyo, 181-8588, JAPAN}
47: \author{Robert C. Nichol}
48: \affiliation{ICG, University of Portsmouth, Portsmouth, PO1~2EG,
49: United Kingdom}
50: \author{Yasushi Suto}
51: \affiliation{Department of Physics and Research Center for the Early
52: Universe, The University of Tokyo, Tokyo 113-0033, Japan}
53: %%%%%%%%%%%%%%%%%%%%%%%%%%%
54: \begin{abstract}
55: We consider the extent to which future imaging surveys of galaxies can
56: distinguish between dark energy and modified gravity models for the
57: origin of the cosmic acceleration. Dynamical dark energy models may
58: have similar expansion rates as models of modified gravity, yet predict
59: different growth of structure histories. We parameterize the cosmic
60: expansion by the two parameters, $w_0$ and $w_a$, and the linear growth
61: rate of density fluctuations by Linder's $\gamma$, independently. Dark
62: energy models generically predict $\gamma \approx 0.55$, while the DGP
63: model $\gamma \approx 0.68$. To determine if future imaging surveys can
64: constrain $\gamma$ within 20 percent (or $\Delta\gamma<0.1$), we perform
65: the Fisher matrix analysis for a weak lensing survey such as the
66: on-going Hyper Suprime-Cam (HSC) project. Under the condition that the
67: total observation time is fixed, we compute the Figure of Merit (FoM) as
68: a function of the exposure time $\texp$. We find that the tomography
69: technique effectively improves the FoM, which has a broad peak around
70: $\texp\simeq {\rm several}\sim 10$ minutes; a shallow and wide survey is
71: preferred to constrain the $\gamma$ parameter. While $\Delta\gamma <
72: 0.1$ cannot be achieved by the HSC weak-lensing survey
73: alone, one can improve the constraints by combining with a follow-up
74: spectroscopic survey like WFMOS and/or future CMB observations.
75: \end{abstract}
76:
77: %\keywords{
78: %methods: analytical -- galaxies: general -- weak lensing survey --
79: %large-scale structure of Universe }
80:
81: \maketitle
82:
83: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
84: \section{Introduction}
85: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
86:
87: % Decided to split the first paragraph into two (DRP)
88:
89: The existence of the mysterious cosmic acceleration is usually ascribed to
90: the presence of an extra component of the universe with a negative
91: pressure, known as dark energy. However, modification of the law of gravity remains as another
92: interesting and equally valid possibility. One of the most elaborated examples is the DGP
93: cosmological model that incorporates the self-acceleration mechanism
94: \cite{DGP,GG} without dark energy.
95: A fundamental question in this context is whether it is possible to
96: distinguish between the modified gravity and dark energy models that
97: have an (almost) identical cosmic expansion history \cite{YBRSY,Heavens}.
98: The answer to the
99: question is inevitably dependent on the specific model of dark energy or
100: modified gravity\cite{Kunz}. Thus we focus on the DGP model, and consider if it
101: has any observational signature that can be distinguished from dark
102: energy models with future galactic surveys. While it is pointed out
103: that the DGP model has some theoretical inconsistency at a fundamental
104: level\cite{Luty,Nicolis,ghostkoyama}, it is still useful as an empirical
105: prototype of modified gravity models, and its observational consequences
106: are discussed \cite{MM,SSH,KM,YBRSY}.
107:
108:
109: The important key is the growth rate of cosmological density
110: perturbations, which should be different in the two models even
111: if they have an identical cosmic expansion history.
112: The weak lensing power spectrum can be sensitive to the growth rate,
113: while the uncertainty of the clustering bias will be the bottleneck
114: that makes the galaxy power spectrum insensitive to the growth rate.
115:
116: Currently several imaging and spectroscopic surveys of galaxies are
117: planned to unveil the origin of cosmic acceleration via weak lensing
118: and baryon acoustic oscillation methods. The Hyper Suprime-Cam (HSC)
119: project is a fully-funded imaging survey at the Subaru telescope,
120: which is expected to commission in 2011. An associated spectroscopic
121: survey possibility, Wide-field Fiber-fed Multi-Object Spectrograph
122: (WFMOS) project, is under serious discussion between Subaru and Gemini
123: observatories (see e.g. \cite{DETF,RESA} and references therein for
124: other projects).
125:
126: %
127: In the present paper, we consider the extent to which future imaging and
128: spectroscopic surveys of galaxies can distinguish between the DGP and
129: dark energy models. More specifically, we empirically characterize the
130: growth rate of density fluctuations adopting Linder's $\gamma$
131: parameter. By optimizing imaging surveys and the combination with
132: redshift survey following the previous
133: literature\cite{OptimalLensingTom,David}, we consider how we can
134: constrain the value of $\gamma$ from HSC weak lensing survey and/or
135: WFMOS baryon acoustic oscillation (BAO) survey.
136:
137: The present paper is organized as follows: In section 2, we explain our
138: theoretical modeling: the parameterization of the background expansion
139: and the modified gravity, the Fisher matrix analysis of the weak lensing
140: power spectrum, and the modeling of the galaxy sample. A demonstration
141: with the DGP model and dark energy model is also presented. In section
142: 3, our result of the Fisher matrix analysis is presented. Section 4 is
143: devoted to summary and conclusions. Throughout the paper, we use the
144: units in which the speed of light is unity.
145:
146:
147: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
148: \section{Theoretical Modeling}
149: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
150:
151: In this analysis we consider a spatially--flat universe for simplicity,
152: consisting of baryons, cold dark matter, and dark energy. We ignore the
153: dark energy clustering, and assume that the spatial fluctuations
154: entirely originate from the matter component (i.e., baryons and dark
155: matter). We further model that the cosmic expansion history {\it
156: effectively} follows the universe with the matter density parameter
157: $\Omega_{\rm m}$ and the dark energy parameter $1-\Omega_{\rm m}$:
158: %%%%%%%%%%%%%%%%%%%%%%%%%
159: \begin{eqnarray}
160: H(a)^2=H_0^2\left[
161: \Omega_m a^{-3}+(1-\Omega_m)a^{-3(1+w_0+w_a)}e^{3w_a(a-1)}
162: \right],
163: \label{friedman}
164: \end{eqnarray}
165: %%%%%%%%%%%%%%%%%%%%%%%%%
166: where $H_0=100h~{\rm km~s^{-1}~Mpc^{-1}}$ is the Hubble constant, $a$ is
167: the cosmic scale factor, and $w_0$ and $w_a$ are constants parameterizing
168: the equation of state of dark energy\cite{CP,Linder2003,CMP}:
169: %%%%%%%%%%%%%%%%%%%%%%%%%
170: \begin{eqnarray}
171: \label{eq:eos}
172: p/\rho \equiv w(a)=w_0+w_a{(1-a)}.
173: \end{eqnarray}
174: %%%%%%%%%%%%%%%%%%%%%%%%%
175: Note that we use equation (\ref{friedman}) even in the DGP model that
176: does not have dark energy at all by approximating its cosmic expansion
177: law with the two parameters $w_0$ and $w_a$. In this case, they do not
178: have any relations to dark energy in reality, but it is already shown
179: that such an empirical description provides a reasonable approximation
180: to the cosmic expansion in the DGP model. For definiteness, the expansion
181: in the DGP has the {\it effective} equation of state (e.g., \cite{Linder2005})
182: %%%%%%%%%%%%%%%%%%%%%%%%%
183: \begin{eqnarray}
184: \label{eq:eosdgp}
185: w(a)=-{1\over 1+\Omega_m(a)},
186: \end{eqnarray}
187: %%%%%%%%%%%%%%%%%%%%%%%%%
188: where
189: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
190: \begin{eqnarray}
191: \label{eq:omega_a}
192: \Omega_m(a)={H_0^2\Omega_m a^{-3}\over H(a)^2}.
193: \end{eqnarray}
194: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
195: The cosmic expansion in the DGP model is well approximated
196: by the dark energy model with {\it effective} equation of state with
197: $w_0=-0.78$ and $w_a=0.32$ as long as $\Omega_m\sim0.27$.
198: The parameterization gives the distance redshift relation
199: within $0.5$ \% out to the redshift $2$ \cite{Linder2005}.
200:
201:
202: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
203: \subsection{Linder's $\gamma$ parameter}
204: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
205:
206: According to refs. \cite{Linder2005,HL,LC}, the linear growth factor
207: in the DGP and dark energy models is well approximately expressed by
208: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
209: \begin{eqnarray}
210: \label{eq:lineargrowth}
211: {D_1(a)\over a }\propto\exp\left[{\int_0^a {da'\over a'}\left(
212: \Omega_m(a')^\gamma-1\right)}\right].
213: \end{eqnarray}
214: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
215: In this description, the constant parameter $\gamma$ characterizes
216: the gravity force model, i.e., the Poisson equation.
217:
218: The dark energy models with the effective
219: equation of state (\ref{eq:eos}) within the
220: general relativity are well approximated by
221: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
222: \begin{eqnarray}
223: && \gamma=0.55+0.05[1+w(z=1)] ~~~~~(w>-1),
224: \\
225: && \gamma=0.55+0.02[1+w(z=1)] ~~~~~(w<-1).
226: \end{eqnarray}
227: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
228: This formula reproduce the exact linear growth factor within $0.3$\%
229: ($0.5$\%) for $-1.2<w<-0.8$ $(-1.5<w<-0.5)$. Therefore $\gamma$ in dark
230: energy models takes the value $\gamma=0.54-0.56$ for $-1.2<w<-0.8$
231: \cite{Linder2005,HL,LC}.
232:
233:
234: On the other hand, in the DGP model, the Poisson equation is modified in
235: the linear regime. Then $\gamma$ takes a different value from that of
236: the dark energy model even if the background expansion is same (i.e. if
237: $w_0$ and $w_a$ are same). Ref. \cite{LC} found that in the DGP model
238: $\gamma = 0.68$ is an excellent approximation for the evolution of the
239: growth factor and that $\gamma$ varies by only $2$ \% into the past.
240:
241:
242: The point here is that a dark energy model mimicking the cosmic expansion
243: history of the DGP model predicts a different linear growth rate by
244: $\Delta \gamma \sim 0.1$. In what follows, therefore, we employ equations
245: (\ref{friedman}) to (\ref{eq:omega_a}) to describe the expansion history
246: and the growth of density fluctuations, which empirically describe both
247: the DGP and dark energy models, and ask if it is possible to achieve the
248: accuracy of $\Delta \gamma \sim 0.1$ by optimizing future surveys of
249: galaxies.
250:
251:
252: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
253: \subsection{Weak lensing power spectrum and Fisher matrix}
254: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
255:
256: The optimization of imaging surveys is based on the weak lensing
257: tomography method (see e.g., \cite{TW,tomoHu,HTBJ,DJT}). In this
258: methodology, one divides the entire galaxy samples in several different
259: redshift bins according to the weight factor $W_i(z(\chi))$ for the
260: $i$-th redshift bin:
261: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
262: \begin{eqnarray}
263: W_i(z)={1\over \bar N_i} \int_{\max(z_i,z)}^{z_{i+1}} dz'
264: {dN(z')\over dz'}\left(1-{\chi(z)\over \chi(z')}\right),
265: \end{eqnarray}
266: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
267: where $dN/dz$ denotes the differential number count of galaxies with
268: respect to redshift per unit solid angle (see below for details),
269: $\chi(z)$ is the radial comoving distance at $z$,
270: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
271: \begin{eqnarray}
272: \chi(z)=\int_0^z{dz'\over H(z)}={1\over H_0}\int_0^z{dz'\over
273: \sqrt{\Omega_m (1+z')^{3}+(1-\Omega_m)(1+z')^{3(1+w_0+w_a)}
274: e^{-3w_az'/(1+z')}}},
275: \label{chidefi}
276: \end{eqnarray}
277: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
278: and
279: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
280: \begin{eqnarray}
281: \bar N_i=\int_{z_i}^{z_{i+1}} dz'
282: {dN(z')\over dz'}
283: \end{eqnarray}
284: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
285: is the total number of galaxies in the $i$-th redshift bin. While
286: imaging surveys provide photometric redshifts alone from the multi-band
287: photometry, instead of spectroscopic redshifts, for galaxies, it is
288: known that the lensing tomography works even with relatively crude
289: redshift information.
290:
291: Assuming that the anisotropic stress is negligible, the cosmic shear
292: power spectrum is given as:
293: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
294: \begin{eqnarray}
295: P_{(ij)}(l)=\int d\chi W_i(z(\chi)) W_j(z(\chi))
296: \left({3H_0^2 \Omega_m\over 2a}\right)^2
297: P_{\rm mass}^{\rm Nonlinear}
298: \left(k\rightarrow {l\over\chi},z(\chi)\right),
299: \end{eqnarray}
300: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
301: where $P_{\rm mass}^{\rm Nonlinear} \left(k,z\right)$ is the nonlinear
302: mass power spectrum at the redshift $z$, $k$ is the wave number of the
303: three dimensional coordinates, $l$ is the wave number of the two
304: dimension corresponding to the angular coordinates, $a$ is the scale
305: factor normalized to unit at the redshift $z=0$. We compute $P_{\rm
306: mass}^{\rm Nonlinear} \left(k,z\right)$ adopting the Peacock and Dodds
307: formula \cite{PD}.
308:
309: The covariance matrix for $P_{(ij)}(l)$ is approximately given by
310: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
311: \begin{eqnarray}
312: {\rm Cov}\bigl[P_{(ij)}(l),P_{(mn)}(l')\bigr]
313: &=&{\delta_{ll'}\over (2l+1)\Delta l f_{\rm sky}}
314: \bigl[P_{(im)}^{\rm obs}(l)P_{(jn)}^{\rm obs}(l)+
315: P_{(in)}^{\rm obs}(l)P_{(jm)}^{\rm obs}(l)\bigr]
316: \nonumber
317: \\
318: &\equiv&\delta_{ll'}{\rm Cov}_{(ij)(mn)}(l),
319: \end{eqnarray}
320: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
321: where we define
322: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
323: \begin{eqnarray}
324: {P^{\rm obs}_{(ij)}(l)}=P_{(ij)}(l)+\delta_{ij}
325: {\sigma_\varepsilon^2\over \bar N_i},
326: \end{eqnarray}
327: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
328: $f_{\rm sky}$ is the fraction of the survey area, and
329: $\sigma_\varepsilon$ is the rms value of the intrinsic ellipticity of
330: randomly oriented galaxies, for which we adopt $\sigma_\varepsilon=0.4$
331: (see e.g., \cite{TW,tomoHu,HTBJ}).
332:
333: Finally the Fisher matrix is estimated as
334: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
335: \begin{eqnarray}
336: F_{\alpha\beta}=\sum_l\sum_{(ij)(mn)}
337: {\partial P_{(ij)}(l)\over \partial\theta^\alpha}
338: {\rm Cov}_{(ij)(mn)}^{-1}(l)
339: {\partial P_{(mn)}(l)\over \partial\theta^\beta},
340: \end{eqnarray}
341: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
342: where $\theta^\alpha$ denote a set of parameters in the theoretical
343: modeling. To be more specific, we consider 7 parameters, $\gamma$,
344: $w_0$, $w_a$, $\Omega_m$, $\sigma_8$ (the fluctuation amplitude at
345: $8h^{-1}$Mpc), $h$, and $n_s$ (the primordial spectral index of matter
346: power spectrum), assuming the other cosmological parameters are
347: determined from independent cosmological data analysis.
348:
349: We adopt the range of $10\leq l\leq 10^4\times(N_g/35/n_b)^{1/2}$ for the
350: sum of $l$, where $N_g$ is the number density of galaxy per unit solid
351: angle (see next subsection). We define the 3 dimensional Figure of
352: Merit by the reciprocal of the volume of the error ellipsoid enclosing
353: the 1 sigma confidence limit in the $\{\gamma,w_0, w_a\}$ space,
354: marginalizing the Fisher matrix over the other parameters. Similarly,
355: the 2 dimensional Figure of Merit is the reciprocal of the surface of
356: the error ellipse enclosing the 1 sigma confidence limit in the $\{w_0,
357: w_a\}$ plane with $\gamma$ fixed.
358:
359: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
360: \subsection{Modeling galaxy sample}
361: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
362:
363: %Here let us explain the galaxy sample used in
364: %the computation of the lensing power spectrum.
365:
366: We assume the following form of the redshift distribution of the galaxy sample per
367: unit solid angle
368: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
369: \begin{eqnarray}
370: {dN\over dz}={N_g\beta\over z_0^{\alpha+1} \Gamma((\alpha+1)/\beta)}z^
371: \alpha
372: \exp\left[-\left({z\over z_0}\right)^\beta\right],
373: \label{dndzdef}
374: \end{eqnarray}
375: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
376: where $\alpha$, $\beta$, and $z_0$ are the parameters, and
377: $N_g=\int dz dN/dz$.
378: The mean redshift may be determined by
379: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
380: \begin{eqnarray}
381: z_m= {1\over N_g}\int dz z{dN\over dz}
382: ={z_0\Gamma((\alpha+2)/\beta)\over \Gamma((\alpha+1)/\beta)}.
383: \end{eqnarray}
384: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
385:
386: We assume that $N_g$ and $z_m$ is related to the exposure
387: time $\texp$ as, following the reference \cite{OptimalLensingTom},
388: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
389: \begin{eqnarray}
390: &&z_m= 0.9\left({\texp\over 30 ~{\rm min.}}\right)^{0.067},
391: \label{scalezm}
392: \\
393: &&N_g=35 \left({\texp\over 30 ~{\rm min.}}\right)^{0.44} {\rm arcmin.}
394: ^{-2}
395: \label{scaleng}
396: \end{eqnarray}
397: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
398: The mean redshift $z_m$ changes from 0.72 to 1.1, and $N_g$ does from
399: $7.8$ to $163$, as the exposure time $\texp$ changes from $1$ minute to
400: $10^3$ minutes. In the reference \cite{OptimalLensingTom}, $\alpha=2$
401: and $\beta=1.5$ are adopted. However, in the present paper, we adopt
402: $\alpha=0.5$ and $\beta=3$.
403:
404: In order to check the validity of our mock galaxy samples, we show in
405: Figure \ref{figurea} the two cases of $\alpha=0.5$ and $\beta=3$ (dotted curve), and
406: $\alpha=2$ and $\beta=1.5$ (dashed curve), for exposure times of $t_{\rm
407: exp}=1,~5,~10,~30,~45$ minutes (from bottom to top respectively). The
408: solid curves show the real redshift histograms, for the corresponding
409: $i$band magnitude limits, taken from the CFHT photometric redshift data
410: of \cite{CFHT}. These photo-z's were calibrated using the VVDS
411: spectroscopy and are reliable to $i\simeq25$ which is sufficient for
412: this study (see \cite{CFHT}). The relationship between magnitude limit
413: and exposure time was scaled from the published Subaru Suprime-Cam data
414: of \cite{Miyazaki}. These data are shown in Table I for the $i,~g,~r,~z$
415: passbands. Denoting the exposure time for the $i$ band by $\texp$, the
416: exposure time for $g$ band is about $\texp_g=3\times\texp$. Similarly,
417: $\texp_r=1.2\times\texp$ for $r$ band, and $\texp_z=0.3\times\texp$ for
418: $z$ band, respectively.
419:
420: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
421: \begin{table}
422: \begin{center}
423: \begin{tabular}{ccccc}
424: \hline
425: ~~~~~$i_{\rm AB~limit}$~~~~~ & ~~~~~~ $i(S/N=10)$ ~~~~~~ & ~~~~~~ $g(S/N=5)$ ~~~~~~ & ~~~~~~ $r(S/N=5)$ ~~~~~~ &
426: ~~~~~~ $z(S/N=5)$ ~~~~~~ \\
427: \hline
428: $22.97$ & $1$~mins. & $3$~mins. & $1.1$~mins. & $0.3$~mins. \\
429: $23.84$ & $5$~mins. & $15$~mins. & $7$~mins. & $1.4$~mins. \\
430: $24.22$ & $10$~mins. & $30$~mins. & $12$~mins. & $3.5$~mins. \\
431: $24.81$ & $30$~mins. & $90$~mins. & $34$~mins. & $8.1$~mins. \\
432: $25.04$ & $45$~mins. & $130$~mins. & $50$~mins. & $13$~mins. \\
433: \hline
434: \end{tabular}
435: \end{center}
436: \caption{Exposure time for the bands, $i, g, r, z$.}
437: \end{table}
438: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
439:
440: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
441: \begin{table}
442: \begin{center}
443: \begin{tabular}{c|c|c|c|c}
444: \hline
445: ~~~~~${\rm Sub-sample}$~~~~~ & ~~~~~~~ $n_b=1$ ~~~~~~~ & ~~~~~~~ $n_b=2$ ~~~~~~~ & ~~~~~~~ $n_b=3$ ~~~~~~~ &
446: ~~~~~~~ $n_b=4$ ~~~~~~~~~~~ \\
447: \hline
448: choice of band & $i$ & $i,~r$ & $g,r,i,z$ & $g,r,i,z$ \\
449: \hline
450: $\sum_j\texp_j$ & $\texp$ & $2.2\times \texp$ & $5.5\times\texp$ & $5.5\times\texp$ \\
451: \hline
452: redshift bins & $0.05<z<2.5$ & $0.05<z<z_m$ & $0.05<z<3z_m/4$ & $0.05<z<0.6\times z_m$\\
453: ~ & ~ & $z_m<z<2.5$ & $3z_m/4<z<5z_m/4$ & $0.6\times z_m<z<z_m$\\
454: ~ & ~ & ~ & $5z_m/4<z<2.5$ & $z_m<z<1.4\times z_m$\\
455: ~ & ~ & ~ & ~ & $1.4\times z_m<z<2.5$\\
456: \hline
457: \end{tabular}
458: \end{center}
459: \caption{Assumption on the subsample and measurement}
460: \end{table}
461: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
462:
463: The total survey area can be expressed as
464: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
465: \begin{eqnarray}
466: {\rm Area}= \pi \left({{\rm Field~of~View}\over 2}\right)^2
467: {T_{\rm total} \over 1.1\times \sum_{j} \texp_{\rm j}+t_{\rm op}},
468: \label{area}
469: \end{eqnarray}
470: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
471: where we assume that the Field of View of $1.5$ degree, the total
472: observation time $T_{\rm total}$ is fixed as $800$ hours, and the
473: overhead time is modeled by a constant, $t_{\rm op}=5$ minutes, plus a fraction ($10\%$) of the exposure time $\sum_{j} \texp_{\rm j}$ for one field
474: of view.
475:
476: We consider the cases the tomography is used, which we denote by
477: $n_b=2$, $n_b=3$ and $n_b=4$. Here $n_b$ denotes the number of the
478: redshift bin. In the case $n_b=2$, the sample is divided into the two
479: subsamples in the range $0.05<z<z_m$ and $z_m<z<2.5$, while in the case
480: $n_b=3$, we consider the three subsample $0.05<z<3z_m/4$,
481: $3z_m/4<z<5z_m/4$ and $5z_m/4<z<2.5$. In the case $n_b=4$, we consider
482: the four subsample $0.05<z<0.6\times z_m$,~$0.6\times z_m<z<z_m$,
483: ~$z_m<z<1.4\times z_m$, and $1.4\times z_m<z<2.5$ (see also Table II).
484: We also consider the case the tomography is not used, which we denote by
485: $n_b=1$, for which we don't take into account how to obtain $dN/dz$, instead
486: assuming that $dN/dz$ is obtained by some method.
487:
488: We assume that the subsample of $n_b=2$ is constructed by the two band,
489: $r$ and $i$, observation, given that the strategy proposed in
490: \cite{CT} is successful. The cases $n_b=3$ and $n_b=4$ are constructed
491: by the 4 band $g,~i,~r,~z$, observation, assuming that the conventional
492: photo-z is successful. The case $n_b=1$ is based on the $i$ band
493: observation. We assume that $90$\% galaxies of $i$ band measurements
494: $dN/dz$ can be used as the subsample, in the case $n_b=2,~3,~4$.
495:
496: We use $\texp$ to represent the $i$ band exposure time for one field of
497: view, then we assume $\sum_{j} \texp_{\rm j}=5.5\times\texp$ for the
498: cases $n_b=3,4$, $\sum_{j} \texp_{\rm j}=2.2\times\texp$ for the case
499: $n_b=2$, and $\sum_{j} \texp_{\rm j}=\texp$ for the case $n_b=1$,
500: respectively.
501:
502:
503: Figure \ref{figureb} shows the resultant total survey area, and the total number of
504: galaxies as function of the $i$ band exposure time $\texp$, for the
505: cases, $n_b=1,~2,~3~{\rm and}~4$.
506:
507:
508: \subsection{DGP model}
509:
510: Here we demonstrate the weak lens power spectrum of the dark energy
511: model and the DGP model with the same cosmic expansion. The linear
512: perturbation theory in the DGP model has been extensively worked out by
513: \cite{KM}. While more recently Koyama and Silva studied nonlinear
514: evolution of density fluctuations in the DGP model \cite{KS}, the
515: nonlinear nature of the gravity in the DGP model is still an unsolved
516: problem. Therefore we adopt an empirical modeling of the nonlinear
517: growth combining the Peacock-Dodds nonlinear fitting formula \cite{PD}
518: and the linear growth rate in the DGP model \cite{KM}. As a result, our
519: predictions below may be inaccurate on nonlinear scales, but our main
520: conclusions concerning the optimization strategy would be unlikely to be
521: sensitive to this approximation.
522:
523:
524:
525: Figure \ref{figurec} shows the weak shear power spectrum of the spatially flat DGP
526: model and the dark energy model with the same background expansion. The
527: cosmological parameters of both of the models are the same ($\Omega_m=0.27$,
528: $\Omega_b=0.044$, $h=0.72$, $\sigma_8=0.8$, and the spectral index
529: $n_s=0.95$). To realize the same cosmic expansion history, the {\it
530: effective} equation of state parameter of the dark energy is chosen as
531: $w(z)=-0.78+0.32 z/(1+z)$, as mentioned in Section 2.1. A similar
532: computation has been already considered by \cite{Ishak}, but our present
533: work differs in that we use the Peacock \& Dodds formula and
534: that we assume a rather shallow sample of galaxies. Because the Poisson
535: equation of the DGP model is modified, then the difference comes from
536: the growth rate. In this figure we assume $30$ minutes exposure time of
537: $n_b=1$. The theoretical curves and the errors bar depend on the survey
538: sample, but we might expect that the two curves could be distinguished.
539: In the next section, we examine the capability of the differentiation.
540:
541:
542:
543: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
544: \section{Results}
545: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
546:
547: In this section, we present our optimization analyses for the HSC weak lensing survey.
548: Specifically, we fix the total observation time of the HSC survey, $T_{\rm tot}$
549: as $800$ hours, and adopt a model
550: of the background galaxy sample described in Sec. IIC for the HSC survey;
551: in particular the mean redshift of galaxies $z_{\rm m}$ and their surface
552: number density $N_{\rm g}$ are given by eqs. (\ref{scalezm}) and
553: (\ref{scaleng}) as a function of the exposure time $\texp$.
554: In this section , we also present results in combination with
555: a spectroscopic survey, the WFMOS BAO survey,
556: which will be limited by a total observation time
557: (see \cite{David} for discussion of the optimization under
558: this condition).
559: Note that we also assume that the WFMOS survey is limited
560: by the total survey area of the HSC imaging survey.
561: Namely, the survey area of the WFMOS survey must be
562: less than or equal to that of the HSC survey, as the HSC survey is
563: acting as a photometric source catalogue for the WFMOS spectroscopic survey.
564: So, for the WFMOS survey, we fix the same survey area as the HSC
565: imaging survey equation (\ref{area}), and the redshift range
566: of galaxies $0.8\leq z\leq 1.4$ with the number density $\bar
567: n=4\times10^{-4}$ ${h^3}$Mpc${}^{-3}$ \cite{David}, which
568: is a set of optimized survey parameters for the spectroscopic survey.
569: %when the total survey area is limited.
570:
571: %\comment{‚È‚ºWFMOS‚Ìparameters‚ª“¯‚¶•K—v‚ª‚ ‚é‚Ì‚©B‚Æ‚‚É$T_{|rm tot}$
572: %‚Æ$\texp$‚Í?}
573:
574: Figure \ref{figured} shows the Figure of Merit (FoM) of the 3 dimension (3D) of
575: $\{\gamma,w_0, w_a\}$, as function of the exposure time, $\texp$.
576: The 3D FoM, the reciprocal of the volume of the 1$\sigma$ error
577: ellipsoid in the $\{\gamma,w_0, w_a\}$ space, is computed by
578: marginalizing the Fisher matrix of the 7 parameters over
579: $\Omega_m, \sigma_8, h$, and $n_s$, with a fixed value for the
580: baryon density, $\Omega_b=0.044$.
581:
582: The lensing tomography method with $n_b=2$ significantly improves the 3D
583: FoM, and continues to do so with increasing $n_b$ for $\texp \simlt
584: 10$mins. The peak of the FoM systematically shifts to the shorter
585: exposure time with larger $n_b$, while the peak profile is fairly broad.
586: With increasing $n_b$, more information of redshift evolution of structure
587: can be obtained. Similarly, as $\texp$ increases, more information of
588: smaller structure can be obtained. However, these are offset by decrease
589: in total survey area. Namely, observation of more bands and longer
590: exposures consume observation time, and the total survey area
591: becomes smaller. This decreases the FoM.
592:
593:
594: For comparison, we plot in Figure \ref{figuree} the 2D FoM, the reciprocal of the
595: area of the 1$\sigma$ error ellipse in the $\{w_0, w_a\}$ plane,
596: evaluated by marginalizing the Fisher matrix of the 6 parameters ($w_0,
597: w_a, \Omega_m, \sigma_8, h$, and $n_s$) with $\Omega_b=0.044$ and
598: $\gamma=0.55$ fixed. One can find the similar features as those of the 3D
599: FoM. This figure suggests the three redshift bin is enough to constrain
600: $w_0$ and $w_a$ and that the peak of FoM is located around $\texp
601: \approx$ $10$ minutes, and the peak profile is very broad. The FoM of
602: the case $n_b=2$ is larger than that of $n_b=3,4$. This indicates that
603: observation of larger survey area with small number of bands ($n_b=2$)
604: can be useful for the dark energy constraints, though an accurate-photo-z
605: strategy is required.
606:
607:
608: Figure \ref{figuref} shows the 1$\sigma$ error on $\gamma$ as a function of $\texp$,
609: which is estimated by marginalizing the Fisher matrix of the 7
610: parameters, $\gamma,w_0, w_a, \Omega_m, \sigma_8, h$ and $n_s$, over the
611: parameters other than $\gamma$. The curve shows the error from the weak
612: lensing power spectrum adopting a proposed survey with HSC; $\Delta
613: \gamma \approx 0.3 (1)$ can be achieved with (without) tomography. The
614: result indicates that the weak lensing survey alone cannot reach the
615: accuracy of $\Delta \gamma =0.1$ that is required to distinguish between
616: the DGP and dark energy models.
617:
618: The uncertainty in $\gamma$ can be significantly (more than a factor of
619: three) reduced by combining the baryon oscillation features from the
620: WFMOS survey (Figure \ref{figureg}). In modeling the galaxy power spectrum of the
621: redshift survey, we simply considered the linear theory specified by the
622: 9 parameters $\gamma, w_0, w_a, \Omega_m, \sigma_8, h, n_s, b_0$ and
623: $p_0$, where $b_0$ and $p_0$ are the parameters for the bias model, for
624: which we adopted the scale independent bias model with the form
625: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
626: \begin{eqnarray}
627: b(z)=1+(b_0-1)(1+z)^{p_0}.
628: \label{biasmodel}
629: \end{eqnarray}
630: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
631: Here we assumed the target parameters $b_0=1.38$ and $p_0=1$. For the
632: theoretical modeling of the galaxy power spectrum and the computation of
633: the Fisher matrix, the range of the wavenumber $0.01~h$Mpc${}^{-1}$$\leq
634: k\leq 0.2~h$Mpc${}^{-1}$ is included, (see Appendix for details).
635:
636: {}From Figures \ref{figuref} and \ref{figureg}, the error of $\gamma$ has a minimum of $\texp$
637: between several minutes and $100$ minutes, depending on the strategy.
638: For the weak lensing survey (HSC) alone, the tomography technique is
639: very effective in reducing the error, and the result is fairly
640: insensitive to the the choice of $\texp$. An additional spectroscopic
641: survey (WFMOS) significantly reduces the error. In this case, shallow
642: surveys with $\texp <10$ minutes provide the minimum error for $\gamma$.
643: Especially, the case $n_b=1$ and $n_b=2$ is significantly improved by
644: the combination. This behaviour is understood as follows. We assume
645: the total observation time of the WFMOS survey is not fixed, while
646: adopting the same survey area as the HSC survey. Then, in these
647: figures, the cases $n_b=1$ and $n_b=2$ assumes larger survey area for
648: the redshift survey than that of the cases $n_b=3$ and $n_b=4$.
649: However, note that the minimum is located around the several minutes of
650: the exposure time even for the case $n_b=3$ and $4$. Therefore, when
651: considering the combination with the redshift survey,
652: wider and shallower surveys are indeed prefered.
653:
654:
655: Now we are in a position to answer the question: is it possible to
656: distinguish between the DGP and dark energy models ? For that purpose,
657: $\Delta \gamma\simlt 0.1$ is required. Figure \ref{figureh} plots the 1 sigma error
658: as a function of the total observation time $T_{\rm total}$, where we
659: adopt $t_{\rm exp}=10$ minutes and $n_b=4$ (dash-dotted curve) and
660: $n_b=2$ (dashed curve). The thin curve is the result of the weak
661: lensing survey alone, while the thick curve is the result combined with
662: the redshift survey. Note that $\Delta \gamma$ is in proportion to
663: $T_{\rm total}^{-1/2}$. Figure \ref{figureh} suggests that the HSC survey alone
664: may reach $\Delta \gamma < 0.1$ with $T_{\rm tot}=10^4$hours,
665: the combination with the WFMOS survey may do so with
666: $T_{\rm tot}=10^3$hours if we put a prior constraint on $\Omega_b$.
667:
668:
669: Finally in this section, let us consider other impact that the HSC
670: survey may present as a test of modified gravity models. The dash-dotted
671: curves in Figure \ref{figurei}(a) show the $1$, $2$ and $3$-sigma confidence contours (going from the innermost outward) in the $w_0-w_a$ plane, by
672: marginalizing the
673: Fisher matrix of the 7 parameters, $\gamma,w_0, w_a, \Omega_m, \sigma_8,
674: h$ and $n_s$, over the parameters other than $w_0$ and $w_a$.
675: %\comment
676: %{‚±‚ê‚à‚¢‚¢‚Ì‚Å‚·‚ªA‚±‚̘_•¶‚ÌŽïŽ|‚Æ‚µ‚Ä‚Í–{“–‚Í$w_0$, $w_a$, $\gamma$
677: %‚Ì3ŽŸŒ³ƒpƒ‰ƒ[ƒ^‹óŠÔ‚Ål‚¦‚é‚ׂ«‚Å‚Í? ‚Æ‚·‚ê‚ÎA‚±‚Ì}‚ɉÁ‚¦‚ėႦ‚Î
678: %$w_0--\gamma$•½–Êã‚ÅA‚±‚Ì2‚‚̃‚ƒfƒ‹‚ª‹æ•Ê‚Å‚«‚é‚©‚àŽ¦‚·‚ׂ«‚Å‚Í?}
679: Here the constraint from future Planck survey is taken into account by
680: including the prior constraints $\Delta
681: \Omega_m=0.035,~\Delta\sigma_8=0.04,~ \Delta w_0=0.32,~\Delta
682: w_a=1,~\Delta n_s=0.0035$ \cite{Wang}.
683: Here the target parameters are
684: same as those of the $\Lambda$CDM model in Figure \ref{figured}, and we fixed
685: $n_b=4$ and $\texp=10$ minutes. Note that the point of the DGP model
686: $(w_0,w_a)=(-0.78,0.32)$ is marked, and is almost near the $2$ sigma
687: curve. This means that the HSC can distinguish between the DGP model and
688: the $\Lambda$CDM model at the $2$ sigma level by including future
689: constraint by the observation of the cosmic microwave background
690: anisotropy. Here, we fixed the total observation time as $800$ hours,
691: then the constraint can be improved when the total observation time is
692: longer. The solid curve is the combination with the WFMOS survey, which
693: also shows the significant improvement of the constraint.
694: Similarly, figure \ref{figurei}(b) show the$1$, $2$ and $3$-sigma confidence
695: contours in the $w_0-\gamma$ plane, by marginalizing the
696: Fisher matrix over the parameters other than $w_0$ and $\gamma$.
697: The point of the DGP model $(w_0,\gamma)=(-0.78,0.68)$ is marked.
698: With this figure, the constraint is at the 1 sigma level. Then
699: we can not clearly distinguish between the DGP model and the
700: $\Lambda$CDM model with this plot.
701: These features reflect how the shear power spectrum is sensitive to
702: the parameters. This suggests the choice of a projection is
703: important for distinguishing between these models.
704:
705:
706:
707:
708:
709: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
710: \section{Summary and Conclusions}
711: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
712:
713: In this paper, we investigated optimization of a weak lensing survey for
714: the dark energy, and how such a survey might be used for testing
715: modification of the theory of gravity. By introducing a simple model of
716: the survey sample as a function of the exposure time for one band of one
717: field of view, we investigated how the FoM and the constraint on
718: Linder's $\gamma$ parameter depend on the exposure time and the number
719: of passbands. To optimize the survey to probe probe modifications of
720: gravity, we considered a Figure of Merit in the space
721: $\{\gamma,w_0,w_a\}$ as well as in the familiar 2D plane
722: $\{w_0,w_a\}$. We obtained the following results: 1)~The peak of the FoM
723: is located at $\texp\simeq {\rm several}\sim 10$ minutes for
724: $n_b=2,3,4$, though the peak profile is very broad. 2)~The tomography
725: technique improves the FoM effectively when including the parameter
726: $\gamma$. 3)~The combination with the redshift survey like the WFMOS
727: BAO survey improves the error on the parameter $\gamma$. 4)~The shallow
728: and wide survey is advantageous for the tomography, and has potential
729: when taking combination with the redshift survey into account. 5)~The
730: HSC weak lensing survey by itself is not sufficient for distinguishing
731: between the DGP model and a dark energy model with the same background
732: expansion, but it will be able to distinguish between the DGP and
733: $\Lambda$CDM at the 2 sigma level by including the prior constraint from
734: future CMB observation.
735:
736: We assumed a very simplified model of the survey galaxy sample, and
737: the error in the photometric redshift measurement is not taken into
738: account. Also we assumed that the weak lensing power spectrum of the
739: $10\leq l\leq10^4 (N_g/35/n_b)^{1/2}$ can be used. Further
740: investigation is needed including the modeling of the galaxy
741: sample and the error in measuring the photometric redshift.
742: In the present paper, we assumed the spatially flat universe.
743: In general, since the lensing power spectrum is not very sensitive to
744: the curvature of the universe, then the inclusion of the other
745: parameter will degrade the constraint \cite{B}.
746:
747:
748: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
749: \vspace{2mm}
750: \begin{acknowledgments}
751: This work is supported in part by Grant-in-Aid for Scientific research
752: of Japanese Ministry of Education, Culture, Sports, Science and
753: Technology (Nos.~18540277, 18654047, 18072002, 17740116, and 19035007), and by JSPS (Japan
754: Society for Promotion of Science) Core-to-Core Program ``International
755: Research Network for Dark Energy''. We thank M. Takada, S. Miyazaki,
756: H. Furusawa, K. Koyama, B. M. Schaefer, R. Maartens, B. A. Bassett, and
757: M. Meneghetti for useful comments related to the topic in the present
758: paper. We are also grateful to A. Taruya, T. Nishimichi, H. Ohmuro,
759: K. Yahata, A. Shirata, S. Saito, M. Nakamichi and H. Nomura
760: for useful discussions related to the topic in the present paper.
761: K.Y. is grateful to the people at Institute of Cosmology and
762: Gravitation of Portsmouth University for their hospitality and useful
763: discussions during his stay.
764: \end{acknowledgments}
765:
766: \newpage
767: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
768: \begin{appendix}
769: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
770:
771: \section{Modeling of the redshift survey power spectrum}
772:
773: Here we briefly review the power spectrum and the Fisher matrix formula
774: for a galaxy redshift survey \cite{p2m,YBN}, adopted in the present
775: paper. Here we assume a measurement of the multipole power spectrum
776: ${\cal P}_l(k)$ $~(l=0,2)$ from the galaxy redshift survey, which we
777: theoretically model as
778: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
779: \begin{eqnarray}
780: {\cal P}_{l}(k)=
781: {1\over 2}\int d{{\mu}}
782: {\int d{{\bf s}} \bar n({{\bf s}})^2{{\psi}}({{\bf s}},{{k}},\mu)^2
783: P({{k,\mu}},{{s}})
784: {\cal L}_{l}(\mu)
785: \over
786: \int d{{\bf s}}' \bar n^2({{\bf s}}') {{\psi}}({{\bf s}}',{{k}},\mu)^2},
787: \label{expll}
788: \end{eqnarray}
789: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
790: where $\bf s$ is the coordinate of the redshift space, $\bar n({\bf s})$
791: is the mean number density per unit volume, $\psi({{\bf s}},{{k}},\mu)$
792: is the weight factor, ${\cal L}_{l}({{\mu}})$ is the Lenegdre
793: polynomial, $\mu$ is the directional cosine between ${\bf k}$ and ${{\bf
794: s}}$, and $P({{k}},\mu,{{s[z]}})$ is the power spectrum at the redshift
795: $z$, which is modeled as
796: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
797: \begin{eqnarray}
798: P(k,\mu,s[z])={s(z)^2\over \chi(z)^2}{ds(z)\over d\chi(z)}P_{gal}\left(
799: {q_{\scriptscriptstyle \|}}\rightarrow k\mu{ds(z)\over d\chi(z)},
800: {q_{\scriptscriptstyle \bot}}\rightarrow k\sqrt{1-\mu^2}{s(z)\over \chi(z)},z
801: \right)
802: \nonumber
803: \\
804: \end{eqnarray}
805: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
806: with
807: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
808: \begin{eqnarray}
809: P_{gal.}({q_{\scriptscriptstyle \|}},{q_{\scriptscriptstyle \bot}},z)
810: = b(z)^2\left[1+{d\ln D_1(z)/d\ln a(z)\over b(z)}{{q_{\scriptscriptstyle \|}}^2
811: \over q^2}\right]^2 P_{\rm mass}^{\rm Linear}(q,z)
812: \label{lin}
813: \end{eqnarray}
814: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
815: where $q^2={q_{\scriptscriptstyle \|}}^2+{q_{\scriptscriptstyle
816: \bot}}^2$, $P_{\rm mass}^{\rm Linear}(q,z)$ is the linear mass power
817: spectrum at the redshift $z$. The comoving distance $\chi[z]$ is given by
818: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
819: \begin{eqnarray}
820: \chi(z,\Omega_m,w_0,w_a)
821: %\int_0^z{dz'\over H(z)}
822: ={1\over H_0}\int_0^z{dz'\over
823: \sqrt{\Omega_m (1+z')^{3}+(1-\Omega_m)(1+z')^{3(1+w_0+w_a)}
824: e^{-3w_az'/(1+z')}}},
825: \label{defsz}
826: \end{eqnarray}
827: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
828: as given in equation (\ref{chidefi}). For our fiducial model we
829: adopt the flat $\Lambda$CDM model with $\Omega_m=0.27$. Thus, our
830: fiducial model is $s(z)=\chi(z,0.27,-1,0)$. In the modeling of the bias,
831: we consider the scale independent bias model in the form,
832: Eq.(\ref{biasmodel}).
833:
834: The variance of ${\cal P}_{l}(k)$ is given by
835: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
836: \begin{eqnarray}
837: \Delta {\cal P}_{l}(k)^2 &=&
838: %\nonumber
839: %\\
840: % && \times2
841: 2{(2\pi)^3\over \Delta V_{k}}
842: {\cal Q}^2_{l}({{\bf s}},k),
843: \label{defPPF}
844: \end{eqnarray}
845: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
846: where $\Delta V_{k}$ denotes the volume of the shell in the Fourier
847: space, and we have defined
848: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
849: \begin{eqnarray}
850: &&{\cal Q}^2_{l}(k) = {1\over 2}
851: \int d\mu {
852: \int d{{\bf s}} \bar n({{\bf s}})^4
853: {{\psi}}({{\bf s}},{{k}},\mu)^4
854: \bigl[P\bigl({{{k}}},\mu,{{s}}\bigr)
855: +{1/\bar n({{\bf s}})}\bigr]^2
856: [{\cal L}_{l}(\mu)]^2
857: \over
858: [\int d{{\bf s}}' \bar n({{\bf s}}')^2 {{\psi}}({{\bf s}}',{{k}},\mu)^2]^2}.
859: \label{defPPG}
860: \end{eqnarray}
861: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
862: Then, we may evaluate the fisher matrix by
863: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
864: \begin{eqnarray}
865: F_{\alpha\beta}\simeq\sum_{l=0,2}
866: {1\over 4\pi^2} \int_{k_{\rm min}}^{k_{\rm max}}
867: \left[{\cal Q}^2_{l}(k) \right]^{-1}
868: {\partial {\cal P}_{l}(k)\over \partial \theta^\alpha}
869: {\partial {\cal P}_{l}(k)\over \partial \theta^\beta}
870: k^2 dk.
871: \end{eqnarray}
872: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
873: In the present paper, we adopt $\bar n(s[z])=4\times10^{-4}$
874: ${h^3}$Mpc${}^{-3}$ and $\psi({{\bf s}},{{k}},\mu)=1$.
875:
876:
877: %where we define
878: %\begin{eqnarray}
879: % \kappa_{2l}(k)^{-1}=
880: % {1\over 2}
881: % \int_{-1}^1 d\mu
882: % { \int d{\bf s} \bar n({\bf s})^4 \psi(k,\mu,{\bf s})^{4}
883: % \bigl[P(k,\mu,s)+1/\bar n({\bf s})\bigr]^2
884: % [{\cal L}_{2l}(\mu)]^2
885: % \over
886: % \Bigl[\int d{\bf s}' \bar n({\bf s}')^2 \psi({\bf s}',k,\mu)^{2}\Bigr]^2},
887: %\label{optimal}
888: %\end{eqnarray}
889: \end{appendix}
890: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%555
891:
892: \newpage
893: \def\and{{and }}
894: \begin{thebibliography}{99}
895: \bibitem{DGP}
896: G. Dvali, G. Gabadadze, \and M. Porrati, Phys. Lett. B {\bf 485}, 208 (2000)
897: \bibitem{GG}
898: G. Gabadadze, Astrophys. J. {\bf 597}, 566 (2003)
899: \bibitem{YBRSY}
900: K. Yamamoto, B. A. Bassett, R. C. Nichol, Y. Suto, \and K. Yahata,
901: Phys. Rev. D {\bf 74}, 063525 (2006)
902: \bibitem{Heavens}
903: A. F. Heavens, T. D. Kitching, \and L. Verde, arXiv:astro-ph/0703191
904: \bibitem{Kunz}
905: M. Kunz\and D. Sapone, arXiv:astro-ph/0612452
906: \bibitem{Luty}
907: M. A. Luty, M. Porrati, \and R. Rattazzi, JHEP {\bf 0309}, 029 (2003)
908: \bibitem{Nicolis}
909: A. Nicolis, \and R. Rattazzi, JHEP {\bf 0406}, 059 (2004)
910: \bibitem{ghostkoyama}
911: K. Koyama, Phys. Rev. D {\bf 72}, 123511 (2005)
912: \bibitem{MM}
913: R. Maartens \and E. Majerotto, Phys. Rev. D {\bf 74}, 023004 (2006)
914: \bibitem{SSH}
915: Y-S. Song, I. Sawicki, \and W. Hu, arXiv:astro-ph/0606286
916: \bibitem{KM}
917: K. Koyama \and R. Maartens, JCAP {\bf 01}, 016 (2006)
918: \bibitem{DETF}
919: A. Albrecht et al., arXiv:astro-ph/0609591
920: \bibitem{RESA}
921: J. A. Peacock et al., arXiv:astro-ph/0610906
922: \bibitem{OptimalLensingTom}
923: A. Amara \and A. Refegier, arXiv:astro-ph/0610127
924: \bibitem{David}
925: D. Parkinson, C. Blake, M. Kunz, B. A. Bassett, R. C. Nichol, \and K. Glazebrook,
926: arXiv:astro-ph/0702040
927: \bibitem{CP}
928: M. Chevallier \and D. Polarski, Int. J. Mod. Phys. D {\bf 10}, 213 (2001)
929: \bibitem{Linder2003}
930: E. V. Linder, Phys. Rev. Lett. {\bf 90}, 091301 (2003)
931: \bibitem{CMP}
932: R. Crittenden, E. Majerotto, \and F. Piazza, arXiv:astro-ph/0702003
933: \bibitem{Linder2005}
934: E. V. Linder, Phys. Rev. D {\bf 72}, 043529 (2005)
935: \bibitem{HL}
936: D. Huterer \and E. V. Linder, arXiv:astro-ph/060868
937: \bibitem{LC}
938: E. V. Linder \and R. N. Cahn, arXiv:astro-ph/0701317
939: \bibitem{TW}
940: M. Takada \and M. White, Astrophys.J. {\bf 601}, L1 (2004)
941: \bibitem{tomoHu}
942: W. Hu, Astrophys. J. 522, L21 (1999)
943: \bibitem{HTBJ}
944: D. Huterer, M. Takada, G. Bernstein, \and B. Jain, MNRAS {\bf 366}, 101 (2006)
945: \bibitem{DJT}
946: D. Dolney, B. Jain, \and M. Takada, MNRAS {\bf 366}, 884 (2006)
947: \bibitem{PD}
948: J. A. Peacock \and S. J. Dodds, MNRAS {\bf 280}, L19 (1996)
949: \bibitem{CFHT}
950: O. Ilbert, et al., A\&A {\bf 457}, 841 (2006)
951: \bibitem{Miyazaki}
952: S. Miyazaki, et al., Publ.~Astron.~Soc.~Jap. {\bf 54}, 833 (2002)
953: \bibitem{CT}
954: B. Jain, A. Connolly, \and M. Takada, arXiv:astro-ph/0609338
955: \bibitem{KS}
956: K. Koyama \and F. P. Silva, arXiv:hep-th/0702169
957: \bibitem{Ishak}
958: M. Ishak, A. Upadhye \and D. N. Spergel, Phys. Rev. D {\bf 74} 043513 (2006)
959: \bibitem{Wang}
960: S. Wang, J. Khoury, Z. Haiman, \and M. May, Phys. Rev. D {\bf 70}, l23008 (2004)
961: \bibitem{B}
962: C. Clarkson, M. Cortes, \and B. A. Bassett, arXiv:astro-ph/0702670
963: \bibitem{p2m}
964: K. Yamamoto, M. Nakamichi, A. Kamino, B. A. Bassett, \and H. Nishioka,
965: Publ.~Astron.~Soc.~Jap. {\bf 58}, 93 (2006)
966: \bibitem{YBN}
967: K. Yamamoto, B. A. Bassett, \and H. Nishioka,
968: Phys.~Rev.~Lett. {\bf 94}, 051301 (2005)
969: \end{thebibliography}
970:
971:
972: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
973: %%
974: \begin{figure}
975: \begin{center}
976: \leavevmode
977: \epsfxsize=16cm
978: % \epsfbox[20 50 600 720]{vvds1.eps}
979: \epsfbox[20 50 600 720]{f1.eps}
980: \end{center}
981: \caption{$dN/dz$ as function of the exposure time,
982: $\alpha=2,~\beta=1.5$ with $z_0=z_m/1.41$ (dashed curve), and
983: $\alpha=0.5,~\beta=3$ with $z_0=z_m/0.64$ (dotted curve),
984: respectively, for the fitting function of the form (\ref{dndzdef}),
985: for the exposure time $\texp=1,~5,~10,~30,~45$ minutes
986: from bottom to top. The solid curve shows the corresponding
987: CFHT LS photo-z $i$ band data.
988: }
989: \label{figurea}
990: %}
991: \end{figure}
992: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
993:
994: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
995: %%
996: \begin{figure}
997: \begin{center}
998: \leavevmode
999: \epsfxsize=16cm
1000: % \epsfbox[20 50 600 720]{gammawlnc.eps}
1001: \epsfbox[20 50 600 720]{f2.eps}
1002: \end{center}
1003: \caption{The total survey area (thick),
1004: and the total number of the galaxies (thin)
1005: as function of the $i$ band exposure time $\texp$,
1006: for the case $n_b=1,~2,~3~{\rm and}~4$.}
1007: \label{figureb}
1008: %}
1009: \end{figure}
1010: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1011:
1012: \begin{figure}
1013: \begin{center}
1014: \leavevmode
1015: \epsfxsize=16cm
1016: % \epsfbox[20 50 600 720]{gammawldgpH.eps}
1017: \epsfbox[20 50 600 720]{f3.eps}
1018: \end{center}
1019: \caption{
1020: The dark (black) curve is the weak lensing power spectrum of
1021: the dark energy model with the cosmological parameter,
1022: $\Omega_m=0.27$, $\Omega_b=0.044$, $h=0.72$,
1023: $\sigma_8=0.8$, $n_s=0.95$, and the equation of state
1024: parameter of the dark energy $w_0=-0.78$, $w_a=0.32$,
1025: while the bright (red) curve is the flat DGP model of the
1026: same cosmological parameters. Here we assume the
1027: HSC like survey with $t_{\rm exp}=30$ minutes of the case
1028: $n_b=1$(see section 2 for details).}
1029: \label{figurec}
1030: %}
1031: \end{figure}
1032: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1033:
1034:
1035: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1036: \begin{figure}
1037: \begin{center}
1038: \leavevmode
1039: \epsfxsize=16cm
1040: % \epsfbox[20 50 600 720]{gammawl53dNN.eps}
1041: \epsfbox[20 50 600 720]{f4.eps}
1042: \end{center}
1043: \caption{
1044: Three dimensional (3D) FoM in $\{\gamma,w_0, w_a\}$ as function of the
1045: $i$ band exposure time, which is obtained from the Fisher matrix of the 7
1046: parameters $\gamma, w_0, w_a, \Omega_m, \sigma_8, h$, and $n_s$,
1047: Here the target parameter is
1048: $\gamma=0.55$, $w_0=-1$, $w_a=0$, $\Omega_m=0.27$, $\sigma_8=0.8$,
1049: $h=0.72$, $n_s=0.95$. The other parameter, $\Omega_b=0.044$ is fixed. }
1050: \label{figured}
1051: %}
1052: \end{figure}
1053: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1054: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1055: \begin{figure}
1056: \begin{center}
1057: \leavevmode
1058: \epsfxsize=16cm
1059: % \epsfbox[20 50 600 720]{gammawl52dNN.eps}
1060: \epsfbox[20 50 600 720]{f5.eps}
1061: \end{center}
1062: \caption{
1063: Two dimensional FoM in $\{w_0, w_a\}$ from the Fisher matrix of the
1064: 6 parameters $w_0, w_a, \Omega_m, \sigma_8, h$.
1065: Here the fiducial model is $\Lambda$CDM, with
1066: $w_0=-1$, $w_a=0$, $\Omega_m=0.27$, $\sigma_8=0.8$,
1067: $h=0.72$, $n_s=0.95$. The other parameters, $\gamma=0.55$ and $\Omega_b=0.044$ are fixed. }
1068: \label{figuree}
1069: %}
1070: \end{figure}
1071: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1072:
1073: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1074: %\begin{figure}
1075: %\begin{center}
1076: % \leavevmode
1077: % \epsfxsize=16cm
1078: % \epsfbox[20 50 600 720]{gammawl3H.eps}
1079: %\end{center}
1080: %\caption{
1081: %Same as figure 3, but with fixed the parameter $n_s=0.95$.
1082: %3D FoM (thick curve) is from the Fisher matrix of the 6 parameters
1083: %$\gamma, w_0, w_a, \Omega_m, \sigma_8$, and $h$, while
1084: %2D FoM (thin curve) is from the Fisher matrix of the 5 parameters
1085: %$w_0, w_a, \Omega_m, \sigma_8$, and $h$.}
1086: %\label{figurewl3}
1087: %}
1088: %\end{figure}
1089: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1090: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1091: %\begin{figure}
1092: %\begin{center}
1093: % \leavevmode
1094: % \epsfxsize=16cm
1095: % \epsfbox[20 50 600 720]{gammawl2H.eps}
1096: %\end{center}
1097: %\caption{
1098: %Same as figure 3, but with fixing the Hubble parameter $h=0.72$ and $n_s=0.95$.
1099: %3D FoM (thick curve) is the result with the Fisher matrix of the 5 parameters
1100: %$\gamma, w_0, w_a, \Omega_m, \sigma_8$.
1101: %2D FoM (thin curve) is from the Fisher matrix of the 4 parameters
1102: %$w_0, w_a, \Omega_m, \sigma_8$.}
1103: %\label{figurewl2}
1104: %%}
1105: %\end{figure}
1106: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1107: \begin{figure}
1108: \begin{center}
1109: \leavevmode
1110: \epsfxsize=16cm
1111: % \epsfbox[20 50 600 720]{gammapk5HNN.eps}
1112: \epsfbox[20 50 600 720]{f6.eps}
1113: \end{center}
1114: \caption{
1115: $1$ sigma error in measuring $\gamma$ as function of
1116: the exposure time,
1117: obtained by marginalizing the Fisher matrix of the $7$ parameters
1118: $\gamma, w_0, w_a, \Omega_m, \sigma_8, h$, and $n_s$, over the
1119: parameters other than $\gamma$. The target parameters
1120: is the same as those of Figure \ref{figured}.
1121: }\label{figuref}
1122: %}
1123: \end{figure}
1124: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1125: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1126: \begin{figure}
1127: \begin{center}
1128: \leavevmode
1129: \epsfxsize=16cm
1130: % \epsfbox[20 50 600 720]{gammapk5HWNN.eps}
1131: \epsfbox[20 50 600 720]{f7.eps}
1132: \end{center}
1133: \caption{
1134: Same as figure \ref{figuref}, but the considering the case of the weak lensing
1135: power spectrum combined with the galaxy power spectrum of the
1136: redshift survey.
1137: }
1138: \label{figureg}
1139: %}
1140: \end{figure}
1141: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1142: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1143: %\begin{figure}
1144: %\begin{center}
1145: % \leavevmode
1146: % \epsfxsize=16cm
1147: % \epsfbox[20 50 600 720]{gammapk3H.eps}
1148: %\end{center}
1149: %\caption{Same as figure 6 but with $n_s$ fixed.
1150: %For the lensing power spectrum, the Fisher matrix of the $6$
1151: %parameters $\gamma, w_0, w_a, \Omega_m, \sigma_8$ and $h$ is used.}
1152: %\label{figurepk3}
1153: %%}
1154: %\end{figure}
1155: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1156: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1157: %\begin{figure}
1158: %\begin{center}
1159: % \leavevmode
1160: % \epsfxsize=16cm
1161: % \epsfbox[20 50 600 720]{gammapk2H.eps}
1162: %\end{center}
1163: %\caption{
1164: %Same as figure 6 but with $h$ and $n_s$ fixed.
1165: %We used the Fisher matrix of the $5$ parameters
1166: %$\gamma, w_0, w_a, \Omega_m, \sigma_8$ for the lensing
1167: %power spectrum.}
1168: %\label{figurepk3}
1169: %}
1170: %\end{figure}
1171: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1172:
1173: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1174: \begin{figure}
1175: \begin{center}
1176: \leavevmode
1177: \epsfxsize=16cm
1178: % \epsfbox[20 50 600 720]{gammawl5aNN.eps}
1179: \epsfbox[20 50 600 720]{f8.eps}
1180: \end{center}
1181: \caption{
1182: 1 sigma error on $\gamma$ as function of the total observation time.
1183: Here we fixed $\texp=10$ minutes and $n_b=4$(dash-dotted curve) and $n_b=2$
1184: (dashed curve). The thin curve is the result with the 7 parameters
1185: of the Fisher matrix for the lensing power spectrum, but
1186: the thick curve is the constraint from the combined weak lensing
1187: power spectrum and galaxy power spectrum (from a redshift survey).
1188: }
1189: \label{figureh}
1190: %}
1191: \end{figure}
1192: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1193: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1194: \begin{figure}
1195: \leavevmode
1196: \begin{center}
1197: \begin{tabular}{ c c }
1198: % \includegraphics[width=2.5in,angle=0]{gammawl5b.eps}
1199: \includegraphics[width=2.5in,angle=0]{f9.eps}
1200: &
1201: % \includegraphics[width=2.5in,angle=0]{gammawl5d.eps}
1202: \includegraphics[width=2.5in,angle=0]{f10.eps}
1203: % \epsfxsize=16cm
1204: % \epsfbox[20 50 600 720]{gammawl5b.eps}
1205: \end{tabular}
1206: \caption{ (a,~Left) The $1$, $2$ and $3$-sigma contours
1207: in the $w_0-w_a$ plane. The dash-dotted curve is the result
1208: with the 7 parameters of the Fisher matrix for the lensing
1209: power spectrum and the Planck prior constraint, and
1210: the solid curve is these constraints combined with the galaxy power
1211: spectrum from a redshift survey.
1212: The target model is the $\Lambda$ CDM model, then $(w_0,w_a)=(-1,0)$,
1213: and the mark $(w_0,w_a)=(-0.78,0.32)$ is the DGP model.
1214: Here we fixed $n_b=4$, $\texp=10$ minutes, and the total observation time,
1215: $800$ hours.
1216: (b,~Right) Same as (a), but with the contours in the $w_0-\gamma$ plane from
1217: marginalizing the Fisher matrix of the 7 parameters over all other
1218: parameters.
1219: The target model is the $\Lambda$CDM model, then $(w_0,\gamma)=(-1,0.55)$,
1220: and the mark $(w_0,\gamma)=(-0.78,0.68)$ is the DGP model.}
1221: \label{figurei}
1222: %}
1223: \end{center}
1224: \end{figure}
1225: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1226: \end{document}
1227:
1228: