1: \documentclass{aastex}
2: %\documentstyle[emulateapj,epsf]{article}
3: \usepackage{emulateapj5,natbib,makeidx,epsfig,amsmath,epic,tabularx,subfigure}
4:
5: \begin{document}
6:
7: \title{A PERSPECTIVE ON THE CMB ACOUSTIC PEAK}
8: \author{T.A. Marriage\altaffilmark{1,2}}
9: \altaffiltext{1}{University of Cambridge, Department of Applied Mathematics
10: and Theoretical Physics, Centre for Mathematical Sciences, Wilberforce Road,
11: Cambridge CB3 0WA}
12: \altaffiltext{2}{Current address: Princeton University, Department of Physics,
13: Jadwin Hall, Princeton, NJ 08544}
14:
15: \begin{abstract}
16: CMB angular spectrum measurements suggest a flat universe.
17: This paper clarifies the relation between geometry and the spherical
18: harmonic index of the first acoustic peak ($\ell_{peak}$).
19: Numerical and analytic calculations show that $\ell_{peak}$
20: is approximately a function of $\Omega_K/\Omega_M$ where $\Omega_K$ and
21: $\Omega_M$ are the curvature ($\Omega_K > 0$ implies an open geometry)
22: and mass density today in units of critical density.
23: Assuming $\Omega_K/\Omega_M \ll 1$, one obtains
24: $\ell_{peak} \approx \frac{11\sqrt{3}}{9(\sqrt{a_*+a_{eq}}-\sqrt{a_{eq}})}
25: \left(2+\frac{\Omega_K}{\Omega_M}\right)$ where $a_*$ and $a_{eq}$ are
26: the scale factor at decoupling and radiation-matter equality.
27: The derivation of $\ell_{peak}$ gives another perspective on the
28: widely-recognized $\Omega_M$-$\Omega_\Lambda$ degeneracy in flat models.
29: This formula for near-flat cosmogonies together with current angular
30: spectrum data yields familiar parameter constraints.
31: \end{abstract}
32: \keywords{cosmic microwave background --- cosmological parameters}
33:
34: \section{INTRODUCTION}
35:
36: It has long been recognized that the first peak in the CMB angular
37: spectrum provides information about the curvature of the universe
38: \citep{doroshkevich, bond, kamionkowski_1, efstathiou, cornish, weinberg_1}.
39: The data are now in. TOCO, BOOMERanG, MAXIMA, and DASI
40: have measured the peak position
41: \citep{amber,netterfield,lee,halverson}.
42: Analyses of the data strongly suggest a flat universe
43: \citep{hu_3,jaffe_1,stompor,pryke,dodelson}.
44: The MAP satellite, cosmic variance limited through $\ell \approx 600$,
45: should make the definitive measurement in the near future
46: \citep{page_1}.
47:
48: As new data resolve higher order peaks, attention shifts to the physics
49: at angular scales beyond that of the first maximum. The object of
50: the present analysis is to clarify the physics derived
51: from the position of the first peak, hereafter called ``the peak index'' or
52: $\ell_{peak}$.
53: The size of the sound horizon $r_{s*}$ at an angular
54: diameter distance $D_{a*}$ to decoupling
55: determines the peak index.\footnote{An expression subscripted by `*'
56: or `eq' is evaluated at decoupling or matter-radiation equality
57: respectively.}
58: This is widely recognized and serves as a starting point.
59: In Sections \ref{sec:da}, \ref{sec:snd}, and \ref{sec:angleandindex},
60: numerical and analytic calculations yield the peak index as a
61: function of $\Omega_M$ and $\Omega_K$. All results are checked
62: with CMBFAST \citep{seljak_CMBFAST}.
63: In Section \ref{sec:dscsn}, a simple formula for $\ell_{peak}$, applicable to
64: low $\Omega_K/\Omega_M$ universes, is developed alongside geometric and
65: classical interpretations of the formal results. It is found that the
66: peak index approximates a function of $\Omega_K/\Omega_M$.
67: Although our analysis grounds itself in the familiar
68: concepts of $D_{a*}$ and $r_{s*}$, the
69: results are not widely recognized. While the physical effects
70: responsible for $\ell_{peak}$ are understood, the interplay that
71: gives the $\Omega_K/\Omega_M$ dependence is not intuitive and holds
72: a number of surprises. We end the investigation by using the functional form
73: of $\ell_{peak}$ and current angular spectrum data to
74: obtain parameter constraints resembling those of more
75: sophisticated treatments.
76:
77: Throughout this work, the ($\Omega_M$,$\Omega_K$) plane serves
78: as the parameter space. Other quantities enter, though the relation
79: between $\ell_{peak}$ and $\Omega_K/\Omega_M$ is relatively independent
80: of them. To be concrete, we take $\Omega_Bh^2=0.02$,
81: consistent with nucleosynthesis \citep{burles}.
82: The redshift of decoupling, $z_*$, is taken as 1400 from the Saha equation.
83: The Hubble constant assumes 72 km/s/Mpc in accord with
84: the HST Key Project results \citep{freedman_1}.
85:
86: \begin{figure*}
87: \begin{center}
88: \centering
89: \epsscale{1.8}
90: \plotone{f1.eps}
91: \caption{\footnotesize
92: The angle subtended by the sound horizon at decoupling.
93: $\Delta\Theta_S = r_{s*}/D_{a*}$. Data favors $\Delta\Theta_s = 0.60^\circ$
94: \citep{knox_2}.
95: In both numerical and analytic plots, equation (\ref{eqn:hrzn_slvd})
96: gives the sound horizon $r_{s*}$. In the numerical plot,
97: the angular diameter distance to last scatter $D_{a*}$
98: is computed using the cosmography code \citep{hogg_1}.
99: In the analytic result, the classical equations for $D_{a*}$
100: are integrated with $\Omega_\Lambda,a_{eq},a_*=0$, and the resulting formulae
101: are used without assuming $\Omega_M + \Omega_K = 1$
102: (see eq. [\ref{eqn:daa}]). $\Delta\Theta_S$ from either method
103: approximates a function of $\Omega_K/\Omega_M$.
104: The bending of the contours in the analytic result
105: arises from radiation-dominated dynamics before last scatter.
106: In the numerical result, explicit inclusion of $\Omega_\Lambda$
107: in the Friedmann equation balances the radiation effect:
108: $\Delta\Theta_S$ contours are straight
109: lines converging on zero in the full calculation. At far right is
110: illustrated the comoving coordinates of a closed FRW universe with
111: time and azimuthal angular coordinates suppressed.
112: Inspection of the drawing shows
113: $\Delta\Theta_s=r_{s*}/R\sin\chi_*$ where the comoving sound
114: horizon at decoupling $r_{s*}$ and
115: the comoving distance to decompling $R\chi_*$ are calculated
116: using Friedmann dynamics.}
117: \label{fig:dths}
118: \end{center}
119: \end{figure*}
120:
121:
122: \section{ANGULAR DIAMETER DISTANCE}
123: \label{sec:da}
124:
125: The comoving Friedmann-Robertson-Walker (FRW) metric for
126: a closed universe ($\Omega_K < 0$) may be written as
127: \begin{equation}
128: ds^2 = d\eta^2 - R^2 \left( d\chi^2 + \sin^2 \chi d\Sigma^2 \right)
129: \label{eqn:frw}
130: \end{equation}
131: where $\eta$ denotes conformal time, $R^2=-1/\Omega_K$, and
132: $d\Sigma^2$ is the line element of a unit two-sphere.
133: All spacetime intervals are given in units of $H_0^{-1}$.
134: The angular diameter distance
135: $D_a$ to an object of proper length $ds$ at
136: comoving distance $R\chi$ ($d\eta = d\chi = 0$) is defined
137: so that $d\Sigma = \mid ds \mid /D_a$.
138: \begin{equation}
139: D_a = R \sin \chi(a)
140: \label{eqn:rawda}
141: \end{equation}
142: where $\chi^2$ is the solution to the Friedmann equation:
143: \begin{equation}
144: d \chi ^2 = \frac{-da^2}{\Omega_M(a/\Omega_K + a_{eq} /\Omega_K)
145: + \Omega_\Lambda a^4/\Omega_K + a^2}.
146: \label{eqn:friedmann}
147: \end{equation}
148: Flat and open geometries are treated analogously.
149:
150: {\bf Numerical Result} $D_{a*}$ can be evaluated numerically. The distance
151: to redshift 1400 is computed for models across the
152: $(\Omega_M,\Omega_K)$ plane using cosmography routines from
153: David Hogg \citep{hogg_1}.
154:
155: {\bf Analytic Result} To complement the numerical result,
156: $D_{a*}$ can be estimated analytically by setting
157: $\Omega_\Lambda, a_{eq} , a_*$ to zero in equation (\ref{eqn:friedmann}),
158: and then using Mattig's solution (cf. \citep{peebles}).
159: \begin{equation}
160: D_{a*} = \frac{2}{\sqrt{\Omega_M}}
161: \times S(\gamma)
162: \label{eqn:daa}
163: \end{equation}
164: where $\gamma=2 \mid\Omega_K\mid/\Omega_M$, and S=1
165: for a flat universe. If $\Omega_K$ is non-zero, then S takes the form
166: \begin{equation}
167: \begin{array}{c@{\quad}c@{\quad}l}
168: S(\gamma) & = & \sqrt{1/2\gamma} \times \left\{
169: \begin{array}{l@{\quad:\quad}r}
170: \frac{1}{2}(\beta - 1/\beta) & \Omega_K > 0 \\
171: \sqrt{1-(\gamma-1)^2} & - \Omega_M < \Omega_K < 0 \\
172: \end{array}
173: \right. \\
174: && \\
175: \beta & = & \gamma+1+\sqrt{(\gamma+1)^2-1}.
176: \end{array}
177: \label{eqn:R}
178: \end{equation}
179:
180:
181: \section{THE SOUND HORIZON}
182: \label{sec:snd}
183:
184: Sound travels through a tightly coupled photon-baryon
185: system with speed $c_s$
186: \begin{equation}
187: c_s = \frac{1}{\sqrt{3(1+Q)}}
188: \label{eqn:snd_spd}
189: \end{equation}
190: where Q is 3$\rho_B$/4$\rho_\gamma$, and $\rho_B$ and
191: $\rho_\gamma$ are the energy densities of baryons and radiation,
192: respectively.
193: The sound horizon at decoupling $r_{s*}$ in comoving coordinates is then
194: \begin{eqnarray}
195: r_{s*} & = & \int_{a=0}^{a_*} c_s \frac{d\eta}{da}da \nonumber \\
196: & = & \frac{2}{\sqrt{3\Omega_M}}\sqrt{\frac{a_{eq}}{Q_{eq}}}
197: \ln\frac{\sqrt{1+Q_*} + \sqrt{Q_*+Q_{eq}}}{1+\sqrt{Q_{eq}}}
198: \label{eqn:hrzn_slvd}
199: \end{eqnarray}
200: \citep{hu_2}. Note that curvature does not affect sound dynamics
201: before decoupling. In any geometry, $r_{s*}$ will be the same.
202:
203: \section{THE HORIZON ANGLE AND THE PEAK INDEX}
204: \label{sec:angleandindex}
205:
206: $D_{a*}$ and $r_{s*}$ give the angular size of the horizon:
207: \begin{equation}
208: \Delta\Theta_{s}=\frac{r_{s*}}{D_{a*}}.
209: \label{eqn:dths}
210: \end{equation}
211: Values for $\Delta\Theta_s$ corresponding to the numerical and
212: analytic results for $D_{a*}$ (see Section \ref{sec:da}) are plotted
213: in Figure \ref{fig:dths}. Curves of constant $\Delta\Theta_{s}$
214: approximate straight lines, which intersect
215: the origin of the ($\Omega_M,\Omega_K$) plane.
216: \emph{The angle subtended by the sound horizon, and so the position
217: of the first peak, approximates a function of
218: $\Omega_K/\Omega_M$.}
219: A well known corollary to this general statement is that a peak
220: corresponding to $\Omega_K=0$ should be insensitive to
221: variations in $\Omega_M$ \citep{bond,hu_3}.
222:
223:
224: To check whether this simple analysis agrees with the standard model, we
225: compare the above results to those from CMBFAST \citep{seljak_CMBFAST}.
226: CMBFAST calculates the CMB angular
227: spectrum.\footnote{CMBFAST inputs are
228: $(\Omega_B=0.04,\Omega_\nu =0,H_0=72,T_{cmb}=2.726,Y_{He}=0.24,
229: N_{\nu}(massless)=3.04,N_{\nu}(massive)=0$,recfast,no reion,
230: scalar only,primordial index 1,adiabatic).}
231: Then, $\ell_{peak}$ from the spectrum multiplies
232: $\Delta\Theta_s$ from Figure \ref{fig:dths} to give the constant of
233: proportionality $\alpha=\ell_{peak}\times\Delta\Theta_s$. As shown in
234: Figure \ref{fig:cmbfast}, the numerically derived
235: $\alpha$ increases from 110 ($\Omega_M=0.14$)
236: to 125 ($\Omega_M=0.74$) and has weak if any dependence on curvature.
237: The near constant graphs of $\alpha$ suggests a well-defined agreement
238: between the peak indices from CMBFAST and those derived from this work's
239: numerical and analytic calculations (see middle frame of
240: Figure \ref{fig:cmbfast}).
241:
242: \begin{figure*}
243: \begin{center}
244: \centering
245: \epsscale{1.8}
246: \plotone{f2.eps}
247: \caption{\footnotesize
248: The relation $\alpha\;=\;\Delta\Theta_S \times \ell_{peak}$
249: and parameter constraints.
250: The peak index from CMBFAST times $\Delta\Theta_S$ of numerical and analytic
251: calculations gives $\alpha$ (left frame). The curves for $\alpha$ are
252: averaged and reinserted into the equation $\ell_{peak}=\Delta\Theta_S/\alpha$
253: to generate the corresponding graphs of peak index (middle frame).
254: Contours of $\ell_{peak}$ in the ($\Omega_M$,$\Omega_K$) plane
255: are calculated using equations (\ref{eqn:rsapprox}) and
256: (\ref{eqn:daapprox}) (right frame).
257: To coincide with the values of $\Delta\Theta_s$ and $\ell_{peak}$ derived from
258: current data, the contour plot assumes
259: $\alpha = 125 (\approx 0.6^\circ \times 210)$ throughout the plane
260: \citep{knox_2, knox_1, hu_3}. The dashed contours show constraints from
261: current estimates of $\ell_{peak}$ and from the projected MAP satellite
262: measurement. }
263: \label{fig:cmbfast}
264: \end{center}
265: \end{figure*}
266:
267: \section{DISCUSSION}
268: \label{sec:dscsn}
269:
270: To gain intuition about $\Delta\Theta_s$, consider the closed comoving
271: universe (eq. [\ref{eqn:frw}]) illustrated in Figure \ref{fig:dths},
272: where time and azimuthal angular coordinates have been supressed
273: to produce the familiar two-sphere geometry. In this picture,
274: CMB photons follow great circles, and, assuming a small angle,
275: $\Delta\Theta_s$ derives from inspection:
276: \begin{math}
277: \Delta\Theta_s = r_{s*}/R \sin \chi_*,
278: \end{math}
279: which is exactly equation (\ref{eqn:dths}).
280:
281: To better understand the parameter dependence of $r_{s*}$,
282: take the sound speed $c_s$ (eq. [\ref{eqn:snd_spd}]) to be constant at
283: $9/10\sqrt{3}$. (With $\Omega_Bh^2=0.02$ and radiation density
284: derived from the COBE FIRAS measurement, the sound speed decreases at
285: a near constant rate ($dc_s/da \approx constant$) from $c/\sqrt{3}$ at a=0
286: to four-fifths that value at decoupling \citep{fixen}.)
287: The sound horizon is then
288: \begin{equation}
289: r_{s*} = c_s \int_{0}^{a_*}\frac{d\eta}{da}da =
290: \frac{9}{5\sqrt{3\Omega_M}} (\sqrt{a_*+a_{eq}} - \sqrt{a_{eq}})
291: \label{eqn:rsapprox}
292: \end{equation}
293: where the final equality follows from the Friedmann equation
294: (\ref{eqn:friedmann}) with $\Omega_K=\Omega_\Lambda=0$.
295:
296: To obtain a simple formula for $\ell_{peak}$, expand $D_{a*}$
297: to first order in $\gamma = 2\mid\Omega_K\mid/\Omega_M$:
298: \begin{equation}
299: D_{a*} = \frac{1}{\sqrt{\Omega_M}} \left( 2 +
300: \frac{\Omega_K}{\Omega_M} \right).
301: \label{eqn:daapprox}
302: \end{equation}
303: The expansion applies to models with
304: a low curvature-to-matter ratio. Coincidentally, such models are
305: favored by experiment, so that the resulting formula for $\ell_{peak}$
306: is useful. Given equations (\ref{eqn:rsapprox}) and (\ref{eqn:daapprox}) and
307: a value of $\alpha$ from Figure \ref{fig:cmbfast},
308: \begin{equation}
309: \ell_{peak} = \frac{\alpha D_{a*}}{r_{s*}} \approx
310: \frac{11\sqrt{3}}{9(\sqrt{a_*+a_{eq}}-\sqrt{a_{eq}})}
311: \left(2+\frac{\Omega_K}{\Omega_M}\right).
312: \label{eqn:roughell}
313: \end{equation}
314: This near-flat approximation is plotted in Figure \ref{fig:cmbfast}
315: where $a_{eq} \approx 2.4 \times 10^{-5} (\Omega_M h^2)^{-1}$.
316: For flat models, the distance to last scatter
317: (eq. [\ref{eqn:daapprox}]) scales as $\Omega_M^{-1/2}$.
318: One may intuit that self-gravitation leads to
319: smaller cosmological separations. This same ``gravitational''
320: effect decreases the distance between the big bang and last
321: scatter, and therefore $r_{s*}$ (eq. [\ref{eqn:rsapprox}]) also
322: scales with an overall factor of $\Omega_M^{-1/2}$.
323: In equation (\ref{eqn:roughell}), the $\Omega_M^{-1/2}$ dependence of
324: $D_{a*}$ cancels that of $r_{s*}$. This cancellation
325: helps explain the $\Omega_M$-$\Omega_\Lambda$ degeneracy of
326: flat cosmogonies. Furthermore, the $\Omega_K/\Omega_M$ term in equation
327: (\ref{eqn:roughell}) is proportional to the curvature times the
328: area between light rays in a two-dimensional representation of
329: an $\Omega_K \approx 0$ universe (e.g. Figure \ref{fig:dths}).
330: This suggests that one may think of ($\Omega_M$,$\Omega_K$) dependence of
331: $\ell_{peak}$ in near-flat spacetimes as resulting from light curving
332: like the geodesics of a two-dimensional space of constant curvature. Finally,
333: the effect of radiation on early universe dynamics and, in particular,
334: on $r_{s*}$ is made explicit by the appearance of $a_{eq}$ in
335: equation (\ref{eqn:rsapprox}). Radiation-dominated
336: cosmological growth per expansion scale ($d\eta/da$) is less than
337: that of matter-dominated dynamics. In low $\Omega_M$ universes,
338: radiation brings last scattering even closer to the big bang and so
339: shortens $r_{s*}$. Thus, the effect of radiation in the early
340: universe is to spoil the pure $\Omega_K/\Omega_M$ dependence of
341: $\ell_{peak}$ as manifest by the curved contours in the plot of
342: equation (\ref{eqn:roughell}) in Figure \ref{fig:cmbfast}.
343: The straightness of the $\ell_{peak}$ contours is restored in the
344: numerical result shown in Figure \ref{fig:dths}.
345: This suggests that explicit inclusion of $\Omega_\Lambda$ in
346: the computation of $D_{a*}$ balances the $a_{eq}$ dependence of
347: $r_{s*}$.
348:
349: \section{CONCLUSION}
350:
351: The peak index has long been recognized as an
352: indicator of geometry. It is hoped that the present
353: analysis sheds new light on $\ell_{peak}$.
354: The peak index does not determine the magnitude of curvature,
355: but rather the ratio of curvature to matter. A measurement of the peak's
356: angular scale gives the precise geometry only if $\Omega_K \approx 0$,
357: otherwise $\ell_{peak}$ is a function of $\Omega_K/\Omega_M$.
358: Furthermore, in deriving the
359: $\Omega_K/\Omega_M$ dependence of $\ell_{peak}$, unexpected
360: cosmological cancellings were discovered. Particularly useful is
361: the balance of overall matter dependencies in $r_{s*}$ and $D_{a*}$
362: which helps account for the $\Omega_M$-$\Omega_\Lambda$ degeneracy in
363: flat models. At the same time, however, it is remarkable that the
364: admittedly simple arguments of this work yield such a decisive
365: cosmological indicator.
366: Within the next few years, NASA's MAP satellite data should give $\ell_{peak}$
367: to cosmic-variance levels. This measurement will burn a sharp line of
368: possible worlds across the ($\Omega_M,\Omega_K$) plane.
369:
370: \acknowledgments
371: This letter was begun at Cambridge as part of DAMTP's
372: tripos. TM thanks Ofer Lahav and Daniel Wesley for early
373: guidance, David Hogg for cosmography computer code, and
374: Lyman Page and James Peebles for suggestions regarding the final draft. TM
375: is an NSF Graduate Research Fellow and is supported by NSF grant
376: PHY-0099493.
377:
378:
379: \begin{thebibliography}{99}
380: \bibitem[Bond et al. (1994)]{bond} Bond, J.R., Crittenden, R., Davis, R.L.,
381: Efstathiou, G., \& Steinhardt, P.J. 1994, Phys. Rev. Lett., 72, 13
382: \bibitem[Burles, Nollett, \& Turner (2001)]{burles}Burles, S., Nollett, K.M.,
383: \& Turner, M.S. 2001, ApJ, 552, L1
384: \bibitem[Cornish (2000)]{cornish}Cornish, N.J. 2000, preprint(astro-ph/0005261)
385: \bibitem[Dodelson \& Knox\ (2000)]{dodelson}Dodelson, S., \& Knox, L.
386: 2000, Phys. Rev. Lett., 84, 3523
387: \bibitem[Doroshkevich, Zel'dovich, \& Syunyaev (1978)]{doroshkevich}
388: Doroshkevich, A.G., Zel'dovich, Ya.B., \& Syunyaev, R.A. 1978, Sov. Astron.,
389: 22, 523
390: \bibitem[Efstathiou \& Bond (1999)]{efstathiou}Efstathiou, G., \& Bond, J.R.
391: 1999, MNRAS, 304, 75
392: \bibitem[Fixen et al. (1997)]{fixen} Fixen, D.J., Hinshaw, G., Bennet, C.L.,
393: and Mather, J.C. 1997, ApJ, 486, 623
394: \bibitem[Freedman et al. (2001)]{freedman_1} Freedman, W.L., et al.
395: 2001, ApJ, 553, 47
396: \bibitem[Halverson et al. (2001)]{halverson} Halverson, N.W., et al.
397: 2001, ApJ, submitted (astro-ph/0104489)
398: \bibitem[Hogg (1999)]{hogg_1} Hogg, D.W. 1999, preprint(astro-ph/9905116)
399: \bibitem[Hu \& Sugiyama\ (1996)]{hu_2} Hu, W., \& Sugiyama, N. 1996,
400: ApJ, 471, 542
401: \bibitem[Hu et al. (2001)]{hu_3} Hu, W., Fukugita, M., Zaldarriaga, M., \&
402: Tegmark, M. 2001, ApJ, 549, 669
403: \bibitem[Jaffe et al. (2000)]{jaffe_1} Jaffe, A.H., et al. 2000,
404: Phys. Rev. Lett., 86, 3475
405: \bibitem[Kamionkowski, Spergel, \& Sugiyama (1994)]{kamionkowski_1}
406: Kamionkowski, M., Spergel, D.N., \& Sugiyama, N. 1994, ApJ, 426, L57
407: \bibitem[Knox \& Page\ (2000)]{knox_1} Knox, L., \& Page, L.A. 2000,
408: Phys. Rev. Lett., 85, 1366
409: \bibitem[Knox, Christensen, \& Skordis (2001)]{knox_2} Knox, L.,
410: Christensen, N., \& Skordis, C. 2001, ApJ, 563, L95
411: \bibitem[Lee et al. (2001)]{lee} Lee, A.T., et al. 2001, ApJ, 561, L1
412: \bibitem[Miller et al. (1999)]{amber}Miller, A.D., et al. 1999, ApJ,
413: 524, L1
414: \bibitem[Netterfield et al. (2001)]{netterfield}Netterfield, C.B.,
415: et al. 2001, ApJ, in press (astro-ph/0104460)
416: \bibitem[Page (2000)]{page_1}Page, L.A. 2000, in Proc. IAU Symposium 201,
417: ed. A. Lasenby \& A. Wilkinson, in press (astro-ph/0012214)
418: \bibitem[Peebles (1993)]{peebles} Peebles, P.J.E. 1993, Principles of
419: Physical Cosmology, (Princeton University Press)
420: \bibitem[Pryke et al. (2001)]{pryke}Pryke, C., Halverson, N.W., Leitch,
421: E.M., Kovac, J., Carlstrom, J.E., Holzapfel, W.L., \& Dragovan, M. 2001,
422: ApJ, submitted (astro-ph/0104490)
423: \bibitem[Seljak \& Zaldarriaga\ (1996)]{seljak_CMBFAST}Seljak, U., \&
424: Zaldarriaga, M. 1996, ApJ, 469, 437
425: \bibitem[Stompor et al. (2001)]{stompor}Stompor, R., et al.
426: 2001, ApJ, 561, L7
427: \bibitem[Weinberg (2000)]{weinberg_1}Weinberg, S. 2000, Phys. Rev. D,
428: 62, 127302
429: \end{thebibliography}
430:
431:
432: \end{document}
433:
434: