astro-ph0402208/ms.tex
1: 
2: 
3: \documentclass[12pt,preprint]{aastex}
4: %\documentstyle[aps,preprint,epsfig]{revtex}
5: %\documentclass{article}
6: \usepackage{emulateapj5,pstricks}
7: 
8: \newcommand{\be}{\begin{equation}}
9: \newcommand{\ee}{\end{equation}}
10: \newcommand{\ba}{\begin{eqnarray}}
11: \newcommand{\ea}{\end{eqnarray}}
12: 
13: \shorttitle{}
14: \shortauthors{Wang \& Freese}
15: 
16: %\def\baselinestretch{2}
17: \begin{document}
18: 
19: \title{Probing Dark Energy Using Its Density \\
20: Instead of Its Equation of State}
21: 
22: \author{Yun~Wang$^{1}$, and Katherine Freese$^{2}$}
23: \altaffiltext{1}{Department of Physics \& Astronomy, Univ. of
24:   Oklahoma, 440 W Brooks St., Norman, OK 73019; email:
25:   wang@nhn.ou.edu} \altaffiltext{2}{Michigan Center for Theoretical
26:   Physics, Univ. of Michigan, Ann Arbor, MI 48109; email:
27:   ktfreese@umich.edu}
28: \begin{abstract}
29:   
30:   The variation of dark energy density with redshift, $\rho_X(z)$,
31:   provides a critical clue to the nature of dark energy.  Since
32:   $\rho_X(z)$ depends on the dark energy equation of state $w_X(z)$
33:   through an integral, $\rho_X(z)$ can be constrained more tightly than
34:   $w_X(z)$ given the same observational data.  We demonstrate this
35:   explicitly using current type Ia supernova (SN Ia) data [the
36:   Tonry/Barris sample], together with the Cosmic Microwave Background
37:   (CMB) shift parameter from CMB data (WMAP, CBI, and ACBAR), and the
38:   large scale structure (LSS) growth factor from 2dF galaxy survey
39:   data. We assume a flat universe, and use Markov Chain Monte Carlo 
40:   (MCMC) technique in our analysis.  
41:   We find that, while $w_X(z)$ extracted from current data is
42:   consistent with a cosmological constant at 68\% C.L., $\rho_X(z)$
43:   (which has far smaller uncertainties) is not.  Our results clearly
44:   show the advantage of using $\rho_X(z)$, instead of $w_X(z)$, to
45:   probe dark energy. 
46: 
47: \end{abstract}
48: 
49: %\end{document}
50: 
51: \keywords{cosmology:observations -- distance scale -- supernovae:general}
52: 
53: 
54: 
55: \section{Introduction}
56: 
57: 
58: Recent observations of type Ia Supernovae \citep{Riess98,Perl99} indicate that
59: the universe is accelerating.  A fundamental quest in physics and
60: cosmology is to identify the nature of the ``dark energy'' 
61: driving this acceleration.
62: Possibilities include: (1) a cosmological constant, (2) a time dependent
63: vacuum energy, or a scalar field known as ``quintessence'' that
64: evolves dynamically with time 
65: \citep{fafm,peebles88,Wett88,frieman,caldwell98,Zlatev99}
66: \footnote{See \cite{Pad03} and 
67: \cite{peebles03} for reviews with more complete lists of references).} 
68: or (3) modified Friedmann equation, e.g. the Cardassian models 
69: \citep{freeselewis,freese03,mpcard,Wang03}, that
70: could result as a consequence of our observable universe living as a
71: 3-dimensional brane in a higher dimensional universe.  Other proposed
72: modifications to the Friedmann equation include 
73: \cite{Parker99,ddg,Bilic02,Ahmed02,Capo03,Carroll03b,Meng03,Puet04}.  The
74: various dark energy models produce dark energy densities $\rho_X(z)$
75: with different redshift dependences.  Hence, in order to differentiate
76: between dark energy models, it is important that we allow the
77: dark energy density to be an arbitrary function of redshift $z$
78: \citep{Wang01a,Wang01b,Wang03}.
79: 
80: 
81: A powerful probe of dark energy is type Ia supernovae (SNe Ia), which
82: can be used as cosmological standard candles to measure how distance
83: depends on redshift in our universe.  The luminosity distance $d_L(z)
84: = (1+z) r(z)$, with the comoving distance $r(z)$ given by 
85: \begin{equation} 
86: r(z)= cH_0^{-1} \int_0^z \frac{dz'}{E(z')}, 
87: \end{equation} 
88: with 
89: \begin{equation}
90: \label{eq:E(z)}
91: E(z) \equiv \left[ \Omega_m(1+z)^3 + \Omega_k (1+z)^2 + \Omega_X
92:   \rho_X(z)/\rho_X(0) \right]^{1/2}, 
93: \end{equation} 
94: where $\Omega_k \equiv
95: 1-\Omega_m-\Omega_X$, and $\rho_X(z)$ is the dark energy density.  
96: 
97: The dark energy equation of state, $w_X(z)$, is related to $\rho_X(z)$
98: as follows \citep{Wang01a}: 
99: \begin{equation}
100: \label{eq:wrhoprime}
101: w_X(z) =\frac{1}{3}(1+z)
102: \frac{\rho'_X(z)}{\rho_X(z)} -1,  
103: \end{equation} 
104: so that
105: \begin{equation}
106: \label{eq:rhoprimew}
107: \frac{\rho_X(z)}{\rho_X(0)} = \exp\left\{ \int_0^z \frac{3
108:     [1+w_X(z)]}{1+z} \right\}.  
109: \end{equation} 
110: One can see that it is easier to extract $\rho_X(z)$ from the data
111: than to extract $w_X(z)$.  To obtain the dark energy density directly,
112: one need only take a single derivative of the luminosity distance,
113: whereas to extract $w_X(z)$, one needs to take a second derivative as
114: well; from Eq.(\ref{eq:wrhoprime}) one can see that $w_X(z)$ is on the
115: same footing as $\rho_X'(z)$.  Specifically, \cite{Wang01a} argued
116: that $\rho_X(z)$ should be preferred since it suffers less from the
117: smearing effect (due to the multiple integrals that relate $w_X(z)$
118: to $d_L(z)$) that makes constraining $w_X(z)$ extremely difficult
119: \citep{Maor01,barger01}.  \cite{Tegmark02} came to the same conclusion.
120: However, researchers have generally chosen to parametrize dark energy
121: using its equation of state $w_X(z)$. Some have used
122:   $H(z)=H_0 E(z)$ (for example, see \cite{Kujat02,Daly03,peri}, and
123:   references therein), which is similar to $\rho_X(z)$, but
124:   measurements of which are not as straightforward to interpret, since
125:   $E(z)$ depends on $\Omega_m$ (see Eq.(\ref{eq:E(z)})).
126: 
127: In this paper, we explicitly demonstrate 
128: the advantage of using $\rho_X(z)$, instead 
129: of $w_X(z)$, to probe dark energy. 
130: Sec.2 contains a comparison of $w_X(z)$ and $\rho_X(z)$
131: parametrizations using current SN Ia, CMB, and LSS data.
132: We give a recipe for parametrizing dark energy using $\rho_X(z)$
133: in Sec.3.
134: Sec.4 contains a summary and discussions.
135: 
136: \section{Dark energy equation of state versus \\
137: dark energy density}
138: %\section{\mbox{$w_X(z)$} versus \mbox{$\rho_X(z)$}}
139: 
140: >From SN, CMB, and LSS data, we independently reconstruct
141: first the dark energy equation of state and then the dark 
142: energy density directly.
143: We use (1) current SN Ia data from Tonry et al. 2003
144: and Barris et al. 2003 [the Tonry/Barris sample], 
145: (2) the Cosmic Microwave 
146: Background (CMB) shift parameter \citep{Bond97}
147: from CMB data [WMAP \citep{Bennett03,Spergel03}, CBI \citep{Pearson03}, 
148: and ACBAR \citep{Kuo02}], and 
149: (3) the large scale structure (LSS) growth factor 
150: from 2dF \citep{Percival02,Verde02,Hawkins03} galaxy survey data.\citep{Wang04}
151: 
152: We parametrize the data first in terms of the dark energy equation of state,
153: and then in terms of the dark energy density, to see which
154: parametrization produces the reconstruction with the least
155: uncertainty. 
156: 
157: To find the function $w_X(z)$ which best fits the data, we consider a
158: five-dimensional parameter space: the function evaluated at three
159: discrete redshift intervals, the value of $\Omega_m$, and the value of
160: the Hubble constant.  We assume a flat universe.  It is our goal to
161: find a set of values for these five parameters that fits the
162: data.  We use the Markov Chain Monte Carlo (MCMC) technique
163: \citep{neal,LB02}, which selects randomly from the 5D parameter space,
164: evaluates $\chi^2$, and create a large number of sets of parameter
165: values (each set is a MCMC sample); 
166: we use $10^6$ MCMC samples \footnote{   
167: Here $\chi^2$ is only used to move around efficiently in the 
168: entire parameter space (based on entropy considerations), 
169: such that for sufficent sampling, the resultant parameter
170: distributions converge to the true probability distribution functions (pdf's).
171: This leads to smooth pdf's since they receive contributions
172: from all MCMC samples. }.  For the current data, we consider
173: the function $w_X(z)$ at three redshift values: $z$=0, $z_{max}/2$,
174: and $z_{max}$ (where $z_{max}$ is the maxmimum redshift of SNe
175: Ia)\footnote{This is the largest number of values one can get out of
176:   the current (sparse) data.}. We find the values (that fit the
177: data) at these points and interpolate at all intermediate redshifts.
178: The parameters estimated from data are $w_X(0)$, $w_X(z_{max}/2)$,
179: $w_X(z_{max})$, $\Omega_m$, and a dimensionless Hubble constant $h$.
180: 
181: Fig.1a shows the $w_X(z)$ reconstructed from
182: 192 SNe Ia from the Tonry/Barris sample, 
183: \footnote{Flux averaging has been performed to reduce the bias
184: in estimated parameters due to weak gravitational lensing.
185: See \cite{Wang00b} and \cite{Wang04} for details.}
186: combined with CMB (shift parameter 
187: ${\cal R}_0 = 1.716 \pm 0.062$)
188: and LSS data (growth parameter $f_0 \equiv 
189: f(z=0.15) =0.51\pm 0.11$)\citep{Wang04}.
190: The regions inside the solid and dashed lines correspond
191: to 68.3\% and 95\% confidence levels 
192: respectively;
193: the 68.3\% confidence level (C.L.) region is also shaded.
194: The circles indicate the mean values 
195: at the three redshift points, $z$=0, $z_{max}/2$, and $z_{max}$.
196: The other simultaneously estimated parameters 
197: (mean, 68.3\% and 95\% confidence ranges) are:
198: $\Omega_m=.39 [.29, .50] [.21, .57]$ and
199: $h= .658 [.642, .674] [.627, .689]$.\footnote{The uncertainty 
200: on $h$ are statistical error only, not including the contribution
201: from the much larger SN Ia absolute magnitude error of 
202: $\sigma_h^{int}\simeq 0.05$.\citep{Wang04}} 
203: Clearly, the equation of state is consistent with a constant
204: $w_X(z)=-1$ for all redshifts at 95\% confidence level (C.L.).  At
205: 68.3\% C.L., it is consistent with a constant for $0 \la z \la 0.5$
206: and marginally consistent with $w_X(z)=-1$ for $0.5 \la z \la 1$.
207: 
208: Fig.1b shows the $\rho_X(z)$ directly reconstructed from the same data
209: as Fig. 1.  The same technique of discretizing the function $\rho_X(z)$
210: has been used.  The solid lines and dashed lines indicate the 68.3\% and
211: 95\% confidence levels respectively 
212: \citep{Wang04}. The other simultaneously estimated
213: parameters (mean, 68.3\% and 95\% confidence ranges) are:
214: $\Omega_m=.33 [.27, .39] [.22, .46]$, $h= .660 [.644, .673] [.630,
215: .688]$.  One can see that the uncertainties in Fig.1b on $\rho_X(z)$
216: obtained from the data
217: are smaller than those on $w_X(z)$ obtained from the data.
218: We see that the time dependence of the dark energy density
219: deviates from a constant at 68.3\% C.L. (a similar statement
220: could not be made from the $w_X(z)$ reconstruction).
221: With more data in the future, the statistical significance of this
222: discrepancy will become more clear.
223: 
224: For comparison, in Fig.2 we have also plotted $\rho_X(z)$ obtained
225: in a more indirect way: by first obtaining $w_X(z)$ from the data, as
226: described above, and then integrating over redshift as in
227: Eq.(\ref{eq:rhoprimew}).  Clearly the uncertainties obtained in this
228: way are far larger than if one obtains the dark energy directly from
229: the data.  While the results from $w_X(z)$ parametrization and
230: $\rho_X(z)$ parametrization \citep{Wang04} are consistent with one
231: another, the $w_X(z)$ parametrization results have uncertainties that
232: are several times larger.  One can also obtain $w_X(z)$ indirectly
233: from $\rho_X(z)$ in Fig.1b (similar to what was done
234: by \cite{alam}). However, doing so would require taking
235: the derivative of the polynomial used in the interpolation,
236: thus making the resultant $w_X(z)$ dependent on the interpolation
237: technique used.
238: 
239: Our main result is that one can learn more information by
240: reconstructing $\rho_X(z)$ rather than $w_X(z)$ from the data.  At
241: 95\% C.L., both the $w_X(z)$ and $\rho_X(z)$ reconstructions are
242: consistent with a cosmological constant.  However at 68.3\% C.L., the
243: $\rho_X(z)$ reconstruction has smaller uncertainties
244: and hence shows more information than the $w_X(z)$
245: reconstruction: the $\rho_X(z)$ reconstruction is {\it not}
246: consistent with a time-independent dark energy.  Even with the
247: $\rho_X(z)$ parametrization, a significant number of SNe Ia at $z>1$
248: from a deep SN survey on a dedicated telescope \citep{Wang00a} will be
249: required to place robust constraints on the time-dependence of
250: $\rho_X(z)$.
251: 
252: 
253: \section{A recipe for parametrizing dark energy \\
254: using its density}
255: %\mbox{$\rho_X(z)$}}
256: 
257: For the convenient application of our methodology by others,
258: we now present a recipe for parametrizing dark energy using $\rho_X(z)$
259: as an arbitrary continuous function.
260: 
261: (1) Choose the number of redshift bins, $N$ (the number of parameters  
262: for $\rho_X(z)$).
263: $N$ needs to be sufficiently large to probe the time-variation
264: of $\rho_X(z)$. However, if $N$ is too large, the uncertainties
265: on all the estimated parameters will increase, leading
266: to less stringent constraints. $N=2$ is appropriate
267: for current (sparse) data.
268: 
269: (2) The values of the dimensionless dark energy density
270: $f_i \equiv \rho(z_i)/\rho_X(0)$ ($i=1,2,...,N)$ are the independent variables
271: to be estimated from data. 
272: Note that $z_N=z_{max}$ (the maximum redshift of SNe Ia in the data).
273: 
274: (3) Parametrize $\rho_X(z)/\rho_X(0)$ as a continuous function,
275: given by interpolating its amplitudes at equally spaced
276: $z$ values in the redshift range covered by SN Ia data
277: ($0 \leq z \leq z_{max}$),
278: and a constant at larger $z$ ($z>z_{max}$,
279: where $\rho_X(z)$ is only weakly constrained by CMB data).
280: The results should not be sensitive to the interpolation
281: method used. 
282: Polynomial interpolation was
283: used in \cite{Wang04}. For $N=2$, this gives
284: \be
285: \label{eq:rho_X}
286: \frac{\rho_X(z)}{\rho_X(0)}=1+ \left(4 f_1 -f_2 -3\right)\, 
287: \frac{z}{z_{max}} + \left(f_2 -2 f_1 +1\right)\, 
288: \frac{2z^2}{z^2_{max}},
289: \ee
290: where $f_1=\rho(z_{max}/2)/\rho_X(0)$, and
291: $f_2=\rho(z_{max})/\rho_X(0)$.
292: 
293: 
294: (4) Use Eq.(\ref{eq:rho_X}) or its equivalent (if the
295: interpolation method or $N$ differs) in
296: all equations where the factor $E(z)$ from
297: Eq.(\ref{eq:E(z)}) appears.
298: 
299: {\it Caution:} It is important to note that we are
300: using a polynomial to interpolate $\rho_X(z)$
301: between equally spaced $z$ values; the independent
302: variables are the values of $\rho_X(z)$
303: at these $z$ values, as in Eq.(\ref{eq:rho_X}).
304: In this case, the errors on the reconstructed $\rho_X(z)$ are 
305: tied to how the quality of the data varies with $z$ (sparse
306: data lead to large errors).
307: Changing the interpolation method from polynomial interpolation
308: to a different method should have negligible effect
309: on the reconstructed $\rho_X(z)$. 
310: On the other hand, if a polynomial is used as a global
311: fit function with its coefficients being the independent variables;
312: the errors on the reconstructed $\rho_X(z)$
313: will not correlate with how the quality of data 
314: varies with $z$.
315: 
316: 
317: \section{Summary and Discussion}
318: 
319: The critical first step in solving the mystery of dark energy 
320: is to determine whether the dark energy density $\rho_X(z)$
321: varies with time.\citep{Wang01a}
322: A definitive answer to this question can have profound
323: implications for particle physics and cosmology.
324: 
325: Our main result is that one can learn more information by
326: reconstructing $\rho_X(z)$ rather than $w_X(z)$ from the data.  The
327: two quantities are related by an integral, which in the case of
328: $w_X(z)$ smears out much of the information one could otherwise learn.
329: We show this explicitly by using a combination of SN Ia data from the
330: Tonry/Barris sample as well as CMB (WMAP, CBI, and ACBAR)
331: and large scale structure (2dF)
332: data.  At 95\% CL, both the $w_X(z)$ and $\rho_X(z)$ reconstructions
333: are consistent with a cosmological constant.  However at 68\% CL, the
334: $\rho_X(z)$ reconstruction has smaller uncertainties and hence shows
335: information that the $w_X(z)$ reconstruction cannot: the $\rho_X(z)$
336: reconstruction is {\it not} consistent with a time-independent dark
337: energy, and the dark energy density appears to be increasing with
338: redshift.  Future data will be required to resolve this question.
339: 
340: We have shown definitively the advantage of the $\rho_X(z)$
341: parametrization over the $w_X(z)$ parametrization in determining the
342: time-variation of $\rho_X(z)$.  To help others apply the $\rho_X(z)$
343: parametrization, we have given a recipe for using the $\rho_X(z)$
344: parametrization in data analysis to probe dark energy (see Sec.3).
345: Our methodology should be very useful in all data analysis aiming at
346: unraveling the nature of dark energy.
347: 
348: 
349: 
350: \acknowledgements 
351: We are grateful to Pia Mukherjee for helping us set up MCMC
352: and for helpful discussions,
353: and thank P. Gondolo for useful discussions.
354: This work is supported in part by NSF CAREER grant
355: AST-0094335 (YW) and by the DOE and the MCTP via grants at the
356: University of Michigan (KF).
357: 
358: \begin{thebibliography}{}
359: 
360: \bibitem[Alam et al.(2003)]{alam}
361: Alam, U., Sahni, V., Saini, T.D., and Starobinsky, A.A. 2003,
362: astro-ph/0311364
363: 
364: \bibitem[Ahmed et al.(2002)]{Ahmed02}
365: Ahmed, M., Dodelson, S., Greene, P.B., and Sorkin, R. 2002, astro-ph/0209227; 
366: 
367: \bibitem[Armendariz-Picon, Mukhanov, \& Steinhardt(2000)]{Armenda00}
368: Armendariz-Picon, C., Mukhanov, V., Steinhardt, P.J. 2000,
369: Phys. Rev. Lett. 85, 4438
370: 
371: \bibitem[Barger \& Marfatia(2001)]{barger01}
372: Barger, V., and Marfatia, D. 2001, Phys.\ Lett.\ B498, 67-73 
373: 
374: \bibitem[Barris et al.(2003)]{Barris03}
375: Barris, B.J., et al. 2003, astro-ph/0310843, ApJ, in press
376: 
377: 
378: \bibitem[Bennett et al.(2003)]{Bennett03}
379: Bennett, C., et al. 2003, ApJ, Suppl. 148, 1
380: 
381: \bibitem[Bilic, Tupper, \& Viollier(2002)]{Bilic02}
382: Bilic, N., Tupper, G.B, and Viollier, R., Phys.Lett. {\bf B535} 17 (2002);
383: 
384: \bibitem[Bond, Efstathiou, \& Tegmark(1997)]{Bond97}
385: Bond, J.R.; Efstathiou, G.; \& Tegmark, M. 1997, MNRAS, 291, L33
386: 
387: \bibitem[Caldwell, Dave, \& Steinhardt(1998)]{caldwell98} Caldwell, R., Dave, R.,
388: Steinhardt, P.J. 1998, Phys. Rev. Lett., 80,  1582 
389: 
390: \bibitem[Capozziello, Carloni, \& Troisi]{Capo03}
391: Capozziello, S., Carloni, S.,  and Troisi, A., astro-ph/0303041;
392: 
393: \bibitem[Carroll, Hoffman, \& Trodden(2003)]{Carroll03}
394: Carroll, S.M.; Hoffman, M., \& Trodden, M. 2003,
395: Phys. Rev. D68, 023509
396: 
397: \bibitem[Carroll et al.(2003)]{Carroll03b}
398: Carroll,  S., Duvvuri, V., Trodden, M., and Turner, M., astro-ph/0306438.
399: 
400: \bibitem[Daly \& Djorgovski(2003)]{Daly03}
401: Daly, R.A., \& Djorgovski, S.G. 2003, ApJ, 597, 9
402: 
403: \bibitem[Deffayet(2001)]{ddg} Deffayet, C. 2001, Phys. Lett. B502, 199
404: 
405: \bibitem[Freese et al.(1987)]{fafm} Freese, K., Adams, F.C., Frieman, J.A.,
406:  and Mottola, E. 1987, Nucl. Phys. B287, 797
407: 
408:   
409: \bibitem[Freese \& Lewis(2002)]{freeselewis} Freese, K., and Lewis, M.,
410: 2002, Phys. Lett., B540, 1
411: 
412: \bibitem[Freese(2003)]{freese03}
413: Freese, K., Nuclear Physics B (Proc. Suppl.) {\bf 124}, 50 (2003)
414:   
415: \bibitem[Frieman et al.(1995)]{frieman} Frieman, J.,  Hill, J., Stebbins, A.,
416: and Waga, I. 1995, Phys. Rev. Lett., 75, 2077
417: 
418: \bibitem[Gondolo \& Freese(2003)]{mpcard} Gondolo, P.,  and Freese, K. 2003,
419: Phys.Rev. {\bf D68} 063509
420:  
421: \bibitem[Hawkins et al.(2003)]{Hawkins03}
422: Hawkins, E. et al. 2003, astro-ph/0212375, MNRAS in press
423: 
424: 
425: \bibitem[Knop et al.(2003)]{Knop03}
426: Knop, R. A., et al. 2003, astro-ph/0309368, ApJ, in press
427: 
428: \bibitem[Kujat et al.(2002)]{Kujat02}
429: Kujat, J.; Linn, A.M.; Scherrer, R.J.; \& Weinberg, D.H. 2002,
430: ApJ, 572, 1
431: 
432: \bibitem[Kuo et al.(2002)]{Kuo02}
433: Kuo, C.L., et al. 2002, submitted to ApJ, astro-ph/0212289
434: 
435: \bibitem[Lewis \& Bridle(2002)]{LB02}
436: Lewis, A., \& Bridle, S. 2002, Phys. Rev. D, 66, 103511, astro-ph/0205436
437: 
438: \bibitem[Maor, Brustein, \& Steinhardt(2001)]{Maor01}
439: Maor, I., Brustein, R., \& Steinhardt, P.J. 2001, 
440: Phys. Rev. Lett., 86, 6; Erratum-ibid. 87 (2001) 049901
441: 
442: \bibitem[Meng \& Wang(2003)]{Meng03}
443: Meng, X., \& Wang, P. (2003), Class.Quant.Grav. 20, 4949
444: 
445: \bibitem[Neal(1993)]{neal}
446: Neal, R.M. 1993, Technical Report CRG-TR-93-1,
447: ftp://ftp.cs.utoronto.ca/pub/~radford/review.ps.gz
448: 
449: \bibitem[Nesseris \& Perivolaropoulos(2004)]{peri} 
450: Nesseris, S., \& Perivolaropoulos, L. 2004, astro-ph/0401556
451: 
452: \bibitem[Padmanabhan(2003)]{Pad03}
453: Padmanabhan, T. 2003, Physics Reports 380, 235-320 
454: 
455: \bibitem[Parker \& Raval(1999)]{Parker99}
456: Parker, L., and Raval, A. 1999, Phys. Rev. D {\bf 60}, 063512
457:   
458:   
459: \bibitem[Peebles \& Ratra(1988)]{peebles88}
460: Peebles, P. J. E.; Ratra, B. 1988, ApJ, 325L, 17
461: 
462:   
463: \bibitem[Peebles \& Ratra(2003)]{peebles03}
464: Peebles, P. J. E.; Ratra, B. 2003, Rev.Mod.Phys. 75, 559-606
465: 
466: \bibitem[Pearson et al.(2003)]{Pearson03} 
467: Pearson, T.J., et al. 2003, ApJ, 591, 556-574
468: 
469: \bibitem[Percival et al.(2002)]{Percival02}
470: Percival, W.J. et al. 2002, MNRAS, 337, 1068
471: 
472: \bibitem[Perlmutter et al.(1999)]{Perl99} Perlmutter, S., et al. 1999,  
473: ApJ,  517,  565
474: 
475: 
476: \bibitem[Phillips(1993)]{Phillips93}
477: Phillips, M.M., ApJ, 413, L105 (1993)
478: 
479: \bibitem[Puetzfeld \& Chen(2004)]{Puet04}
480: Puetzfeld, D., and Chen, X. 2004, gr-qc/0402026
481: 
482: 
483: \bibitem[Riess et al.(1998)]{Riess98} Riess, A.~G., et al. 1998,  
484: AJ, 116, 1009 
485: 
486: 
487: \bibitem[Riess, Press, \& Kirshner(1995)]{Riess95}
488: Riess, A.G., Press, W.H., and Kirshner, R.P., ApJ, 438, L17 (1995)
489: 
490: \bibitem[Riess et al.(1999)]{Riess99}
491: Riess, A.G., et al. 1999, AJ, 118,  2675
492: 
493: \bibitem[Spergel et al.(2003)]{Spergel03}
494: Spergel, D.N., et al. 2003, astro-ph/0302209
495: 
496: \bibitem[Tegmark(2002)]{Tegmark02}
497: Tegmark, M. 2002, Phys. Rev. D66, 103507
498: 
499: 
500: \bibitem[Tonry et al.(2003)]{Tonry03}
501: Tonry, J.L., et al. 2003, ApJ, 594, 1-24
502: 
503: 
504: \bibitem[Verde et al.(2002)]{Verde02}
505: Verde, L. et al. 2002, MNRAS, 335, 432
506: 
507: 
508: \bibitem[Wang(2000a)]{Wang00a} 
509: Wang, Y. 2000a, ApJ, 531, 676 
510: 
511: \bibitem[Wang(2000b)]{Wang00b}
512: Wang, Y. 2000b, ApJ, 536, 531 
513: 
514: \bibitem[Wang \& Garnavich(2001)]{Wang01a}
515: Wang, Y., and Garnavich, P. 2001, ApJ, 552, 445 
516: 
517: \bibitem[Wang \& Lovelace(2001)]{Wang01b} 
518: Wang, Y., and Lovelace, G. 2001, ApJ, 562, L115 
519: 
520: \bibitem[Wang et al.(2003)]{Wang03}
521: Wang, Y.;  Freese, K.; Gondolo, P.; \& Lewis, M. 2003,
522: ApJ, 594, 25
523: 
524: \bibitem[Wang \& Mukherjee(2004)]{Wang04}
525: Wang, Y., \& Mukherjee, P. 2004, astro-ph/0312192, ApJ submitted
526: 
527: \bibitem[Wetterich(1988)]{Wett88}
528: Wetterich, C. 1988,  Nucl. Phys. {\bf B302}, 668 
529:  
530: \bibitem[Zlatev, Wang, \& Steinhardt(1999)]{Zlatev99}
531: Zlatev,  I., Wang, L., and Steinhardt, P.J. 1999,
532:   Phys. Rev. Lett.  {\bf 82}, 896 
533:    
534: \end{thebibliography}
535:   
536: 
537: 
538: %\clearpage
539: \setcounter{figure}{0} \plotone{f1color.eps} \figcaption[f1.eps] {(a) The
540:   $w_X(z)$ reconstructed from 192 SNe Ia from the Tonry/Barris sample,
541:   flux-averaged, and combined with CMB and LSS data.  The regions
542:   inside the solid and dashed lines correspond to 68.3\% and 95\%
543:   confidence levels respectively; the 68.3\% confidence level region
544:   is also shaded. Circles indicate the mean values of the regions.
545:   (b) The $\rho_X(z)$ reconstructed from the same data as Fig.1(a),
546:   with the same shading and line types \citep{Wang04}.  Whereas the
547:   $w_X(z)$ reconstruction is consistent at 68.3\% C.L. with a
548:   cosmological constant, the $\rho_X(z)$ reconstruction is not.  }
549: 
550: \setcounter{figure}{1} \plotone{f2color.eps} \figcaption[f2.eps] { The
551:   $\rho_X(z)$ reconstructed by taking the integral in Eq.(4) of the
552:   $w_X(z)$ plotted in Fig.1(a).  Again, we use the same data, with
553:   the same shading and line types.  Clearly the uncertainties in this
554:   method are much greater than if one obtains $\rho_X(z)$ directly
555:   from the data as in Fig.1(b).  }
556: 
557: 
558: \end{document}
559: 
560: 
561: