1:
2: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3: % Paper on SZE and gas fraction evolution %
4: % (draft3) %
5: % S . Majumdar %
6: % %
7: % First draft on 19 feb 2001 %
8: % Revised on 21 feb 2001 %
9: % Revised on 22 feb 2001 %
10: % Revised on 24 feb 2001 (after comments) %
11: % Revised and Submitted on 25 feb 2001 %
12: % %
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14:
15: \documentclass{article}
16: \usepackage{graphics}
17: \usepackage{emulateapj}
18:
19: %\documentclass[12pt,preprint]{aastex}
20: %\documentclass[manuscript]{aastex}
21:
22:
23: \def\etal{{et al.}\thinspace}
24: \def\beb{\begin{thebibliography}{999}}
25: \def\eeb{\end{thebibliography}}
26: \def\bi{\bibitem[]{}}
27: \def\be{\begin{equation}}
28: \def\ee{\end{equation}}
29: \def\bea{\begin{eqnarray}}
30: \def\eea{\end{eqnarray}}
31: \def\o{\over}
32: \newcommand{\gsim}{\raisebox{-0.3ex}{\mbox{$\stackrel{>}{_\sim} \,$}}}
33: \newcommand{\lsim}{\raisebox{-0.3ex}{\mbox{$\stackrel{<}{_\sim} \,$}}}
34: \newcommand{\gmf}{\fg}
35: \newcommand{\fg}{$f_g$ }
36:
37: \begin{document}
38:
39: \slugcomment{submitted to ApJL}
40: %\shorttitle{Gas fraction and SZ effect}
41: %\shortauthors{S. Majumdar}
42:
43:
44: \title{Probing the Evolution of Gas Mass Fraction with Sunyaev-Zel'dovich Effect}
45:
46:
47: \author{Subhabrata Majumdar\altaffilmark{1,2}}
48: \affil{$^1$Joint Astronomy Programme, Physics Department,
49: Indian Institute of Science, Bangalore 560012, India}
50: \affil{$^2$ Indian Institute of Astrophysics, Bangalore 560034, India\\
51: email: sum@physics.iisc.ernet.in}
52: %\email{sum@physics.iisc.ernet.in}
53:
54:
55:
56:
57: \begin{abstract}
58: Study of the primary anisotropies of the Cosmic Microwave Background (CMB) can be used to determine the
59: cosmological parameters to a very high precision.
60: The power spectrum of the secondary CMB anisotropies due to the thermal
61: Sunyaev-Zel'dovich Effect (SZE) by clusters of galaxies, can then be studied, to constrain more cluster specific
62: properties (like gas mass).
63: We show the SZE power spectrum from clusters to be a sensitive probe of any possible
64: evolution (or constancy) of the gas mass fraction. The position of the peak of the SZE power spectrum
65: is a strong discriminatory signature of different
66: gas mass fraction evolution models.
67: For example, for a flat universe, there can be a difference in the $l$ values (of the peak) of {\it as much
68: as 3000} between a
69: constant gas mass fraction model and an evolutionary one.
70: Moreover, observational determination of power spectrum, from blank sky surveys, is
71: devoid of any selection effects that can possibly affect targeted X-ray or radio
72: studies of gas mass fractions in galaxy clusters.
73: \end{abstract}
74:
75:
76:
77: \keywords{Cosmology: observations, cosmic microwave background; clusters : intracluster medium, gas fraction}
78:
79:
80:
81: \section{Introduction}
82: Clusters of galaxies, being perhaps the largest gravitationally bound structures in the universe,
83: are expected to contain a significant amount of baryons of the universe. Moreover, due to their large angular
84: sizes, observational estimates of their total mass $M_T$, the gas mass $M_g$ and hence the gas mass fraction
85: (\fg$={M_g \o M_T}$) are easier.
86: These estimates can be
87: used as probes of large scale structure and underlying cosmological models. For example,
88: the cluster \fg would give a lower limit to the universal baryon fraction $\Omega_b/\Omega_m$.
89: Determination of \fg has been done by numerous people (White \& Fabian, 1995; Mohr \etal 1999; Sadat
90: \& Blanchard, 2001) and the values are in agreement within the observational scatter. A point to be noted here is that
91: the estimated \gmf depends on the distance to the cluster (i.e \fg $\propto d_{ang}^{3/2}$ ). Hence, if
92: \gmf is assumed to be constant, then in principle, one can use the `apparent' evolution of \gmf over a large redshift
93: range to constrain cosmological models (Sasaki 1996 ).
94:
95: The question as to whether there is any evolution (or constancy) of gas mass fraction, however, is still
96: debatable, with
97: claims made either way. For example, Schindler (1999) has investigated a sample of distant clusters
98: with redshifts between 0.3 to 1 and conclude that there is no evolution of the gas mass fraction. Similar
99: conclusion has been drawn by Grego \etal (2000).
100: On the contrary, Ettori and Fabian (1999) have looked at 36 high-luminosity clusters, and
101: find evolution in their gas mass fraction (in both sCDM and $\Lambda$CDM universes). See also David \etal,
102: 1995; Tsuru \etal 1997; Allen \& Fabian, 1998; Mohr \etal 1999).
103: Observations suggest that, though \gmf of massive clusters ($T_e \gsim 5KeV$) appears to be constant, low mass
104: clusters with shallower potential wells may have lost gas due to preheating and/or post-collapse
105: energy input (David \etal 1990, 1995; Ponman \etal 1996; Bialek \etal 2000). It is also well known that ICM
106: is not entirely primordial and there is probably continuous infall of gas, thereby, increasing \fg with time.
107: Thus, there is considerable debate regarding the evolution of gas mass fraction.
108:
109:
110: The intracluster medium (ICM) has been probed mainly through X-ray observations, but also through the so called
111: Sunyaev-Zeldovich effect (Zel'dovich \& Sunyaev, 1969) in the last
112: decade (see Birkinshaw 1999 for a review). The SZE effect from
113: clusters is a spectral distortion of the CMB photons due to inverse Compton scattering by the hot ICM
114: electrons, with its magnitude proportional to the Compton $y$-parameter, given by
115: $y={{k_B \sigma_T}\over{m_e c^2}} \int n_e T_e dl$. Here, $k_B$ is the Boltzman constant, $\sigma_T$ is the
116: Thomson scattering cross section, $m_e$ is the electron mass,
117: and $n_e$ and $T_e$ are the ICM electron density and
118: the temperature.
119: Using bolometric SZE measurements, the \fg has been obtained for a number of clusters (see Birkinshaw 1999).
120: Recently, Grego \etal (2000) have made interferometric observations of SZE from a sample of 18
121: clusters. A major advantage of SZE over X-ray
122: measurements is,
123: SZE does not suffer from the ${(1+z)}^{-4}$ `cosmological dimming', which makes SZE an useful probe of evolution of
124: cluster gas mass fraction.
125:
126: Other than the targeted SZE observations, non-targeted `blank sky'
127: surveys of SZE are one of the main aims of future satellite and ground based small angular scale
128: observations (Holder \& Carlstrom, 1999; or with AMiBA
129: %\footnote{\url{http://www.asiaa.sinica.edu.tw/amiba/}}
130: ).
131: Once the power
132: spectrum is extracted from observations, comparison can be made with theory, to constrain
133: cosmological parameters and relevant cluster scale physics.
134: The SZE power spectrum as a cosmological probe has been well studied (Refregier \etal, 1999; Komatsu \& Kitayama,
135: 1999), although its use as a probe of ICM has seldom been looked at.
136:
137: Keeping such surveys in mind, in this {\it Letter},
138: we look at the SZE power spectrum as a probe of the ICM. We show it to be
139: a very sensitive probe of the evolution (or constancy) of \gmf. Measurements of the primary
140: anisotropy would give us `precise' values of cosmological parameters (like $h,
141: \Omega_m, \Omega_\Lambda, \Omega_b$). Hence, for our calculations, we assume that we know the
142: values of cosmological parameters and do not worry about their effect on the SZE power spectrum.
143: Any feature of the SZE power spectrum is attributed to specific cluster physics (like gas content).
144: We note that this method of probing \gmf is not biased from any selection effect that can occur while
145: doing pointed SZE observations of X-ray selected clusters of galaxies, and hence is more desirable.
146:
147: Current observations of primary CMB anisotropies suggest a flat universe with a cosmological constant
148: (Padmanabhan and Sethi, 2000). For
149: our calculations, we take a flat universe with $\Omega_m=0.35, \Omega_b=0.05$ and $h=0.65$ as our fiducial model.
150:
151:
152: The paper is structured as follows. In the next section we discuss the distribution of clusters and model the cluster
153: parameters. In \S 3, we compute the Poisson and clustering power spectrum from SZE and, finally, we discuss our
154: results and conclude in \S 4..
155:
156: \section{The Sunyaev-Zel'dovich effect power spectrum}
157:
158: \subsection{Distributing the galaxy clusters}
159: We set up an ensemble of galaxy clusters with masses between
160: $10^{13} \le M \le 10^{16}$
161: M$_{\odot}$, using the abundance of collapsed objects as predicted
162: by a modified version of the Press-Schechter (PS) mass function
163: (Press \& Schechter 1974) given by Sheth and Tormen (1999). The lower mass cutoff signifies the mass for which one
164: expects a well developed ICM.
165: The gas mass is supposed to sit in the halo potential and are distributed in the same manner.
166: Note, the results are not sensitive to either lower or upper cutoff.
167: We probe up to redshifts of 5. Most of the power, however, comes from objects distributed at $z \lsim 1$..
168:
169: We use the transfer function of Bardeen \etal (1986),
170: with the shape parameter given by Sugiyama (1995) and the Harrison -
171: Zel'dovich primordial spectrum to calculate the matter power spectrum
172: $P_m$(k). The resulting $COBE-FIRAS$ normalised (Bunn \& White, 1997) mass variance ($\sigma_8$) is 0.9 for our
173: fiducial model.
174:
175:
176:
177: \subsection{Modelling the cluster gas}
178: We assume the ICM to follow a $\beta$-profile with
179: $\beta=2/3$ for simplicity. The other physical parameters of the clusters are determined using the virial
180: theorem and spherical collapse model. We closely follow Colafrancesco \& Vittorio (1994) in our modelling.
181: We have for the gas density,
182: $n_e(r) = n_{e,0}{\left(1 + {{r^2}\over{r_c^2}} \right)}^{-3\beta/2}.$
183: We take the gas to be extended up to $R_v=pr_c$ with $p=10$.
184: The central gas density, $n_{e, 0}$ is given by
185: $n_{e,0} = f_g {{2\rho_0}\over{m_p(1+X)}}$
186: where $X=0.76$ is the average proton mass fraction and $\rho_0$ is the central gas mass density.
187: To account for the fact that there is a final cutoff in the gas distribution we introduce a Gaussian filter at
188: the cluster edge $R_v$ given by $n_e(r)\rightarrow n_e(r) e^{-r^2/\xi R_v^2}$, where $\xi = 4/\pi$ is the fudge
189: factor.
190:
191: We parametrize the gas mass fraction as
192: \be
193: f_g = f_{g0} {\left(1+z\right)}^{-s} {\left({{M}\o{10^{15}h^{-1}M_\odot}}\right)}^{k},
194: \ee
195: where the normalization is taken to be $f_{g0}=0.15$, is based on local rich clusters. We look at
196: combinations of both mass and redshift dependence for a range of evolutionary
197: models. In particular, we look at a case of strong evolution given by $k=0.5, s=1$ (Colafrancesco \&
198: Vittorio, 1994); $k=0.1, s=0.5$ (as suggested by Ettori and Fabian, 1999); $k=0.1, s=0.1$ (weak
199: evolution); $k=0$ (no mass dependence) and $s=0$ (no redshift dependence).
200:
201: For the core radius $r_c$ and the temperature, we use
202: \bea
203: r_c\left(\Omega_0, M, z\right) &=& {{1.69h^{-1}Mpc}\o{p}} {{1}\o{1+z}} \times \\ \nonumber
204: && \left[ \left( {{M}\o{10^{15}h^{-1}M_\odot}} \right) {{178}\o{\Omega_0 \Delta_c}} \right]^{1/3} ,
205: \eea
206:
207:
208:
209: \be
210: k_B T_e = 7.76 \beta^{-1} {\left({{M}\o{10^{15}h^{-1}M_\odot}}\right)}^{2/3} (1+z) KeV.
211: \ee
212: Here, $\Delta_c(z)$ is the cluster overdensity relative to the background.
213:
214:
215: Putting everything in, we have the temperature distortion to be
216: $\frac{ {\mit\Delta}T(\theta)}{T_{\rm cmb}} = g(x)y(\theta)$, with
217:
218: \be
219: y(\theta) =
220: \left(\frac{\sigma_T n_{e0}r_{\rm c}k_{\rm B} T_e}{m_ec^2}\right)
221: \times \frac{\pi e^{1/\xi p^2}}{\sqrt{1+\left(\theta/\theta_{\rm c}\right)^2}}
222: {\rm Erfc}\left(
223: \sqrt{\frac{1+(\theta/\theta_{\rm c})^2}
224: {\xi p^2}}
225: \right).
226: \ee
227: The angular core radius $\theta_c = r_c/ d_{ang}$.
228: The spectral form of the thermal SZ effect is given by
229: \be
230: g(x) = {{x^4 e^{x}}\over{{(e^{x} - 1)}^2}}
231: \left[ x coth(x/2) - 4 \right] ,
232: \ee
233: where $x=h\nu/k_BT_{cmb}$.
234: This specific spectral dependence of the thermal SZ effect can be used to
235: separate it out from other CMB anisotropies (Cooray \etal , 2000).
236:
237:
238: \section{Computing the power spectrum}
239:
240:
241: The fluctuations of the CMB temperature produced by SZE can be quantified by their spherical harmonic
242: coefficients $a_{lm}$, which can be defined as
243: $\Delta T({\bf n}) = T_0^{-1} \sum_{lm} a_{lm} Y_{lm}({\bf n})$. The
244: angular power spectrum of SZE is then given by
245: $C_l=<{|a_{lm}|}^2>$, the brackets denoting an ensemble average.
246: The power spectrum for the
247: Poisson distribution of objects, can then be written as
248: (Cole \& Kaiser 1988, Peebles 1980)
249: \be
250: C_l^{Poisson} = \int_0^{z_{max}} dz {{dV(z)}\over{dz}} \int_{M_{min}}^{M_{max}}
251: dM {{dn(M,z_{in})}\over{dM}} {|y_l(M,z)|}^2 ,
252: \ee
253: where $V(z)$ is the comoving volume and $dn/dM$
254: is the number density of objects.
255:
256: Since these fluctuations occur at
257: very small angular scales, we can use the small angle approximation of Legendre
258: transformation and write $y_l$ as the angular Fourier transform of $y(\theta)$
259: as $y_l = 2\pi \int y(\theta) J_0 [(l+1/2])\theta]\theta
260: d\theta $ (Peebles 1980, Molnar \& Birkinshaw 2000).
261:
262: In addition to Poisson power spectra, one would expect contribution to a
263: `correlation power spectrum' from the clustering of the galaxy clusters.
264: Following Komatsu and Kitayama (1999),
265: we estimate the clustering angular
266: power spectrum as
267: \bea
268: C_l^{Clustering} &=& \int_0^{z_{max}} dz {{dV(z)}\over{dz}} P_m \times \\ \nonumber
269: & &{\left[\int_{M_{min}}^{M_{max}} dM {{dn(M,z_{in})}\over{dM}}
270: b(M,z_{in}) y_l(M,z)\right]} ^2 ,
271: \eea
272: where $b(M,z)$ is the time dependent linear bias factor. The matter power
273: spectrum, $P_m(k,z)$, is related to the power spectrum of cluster correlation function
274: $P_c(k,M1,M2,z)$ through the bias, i.e
275: $P_c(k,M1,M2,z)=b(M1,z)b(M2,z)D^2(z)P_m(k,z=0)$ where we
276: adopt $b(M,z)$ given by
277: $b(M,z) = (1+0.5/\nu^4)^{0.06-0.02n}(1 + (\nu^2 -1)/\delta_c)$ (Jing 1999 for details).
278: This expression
279: for the bias factor matches accurately the results of
280: N-body simulations for a wide range in mass.
281: In the above equation $D(z)$ is the linear growth factor of density
282: fluctuation, $\delta_c=1.68$ and $\nu=\delta_c / \sigma(M)$.
283:
284:
285:
286:
287: \section{Results and Discussions}
288:
289:
290:
291: We study the power spectrum of SZE from clusters of galaxies, under the assumption of a `precise' and `a priori'
292: knowledge of
293: the cosmological parameters.
294: %(Note, that current observational data have already fully or partially detected the first three
295: %accoustic peaks between $l \approx 200 - 800$.)
296: We also assume that in the $l$-range of relevance, thermal
297: SZE from clusters of galaxies are the
298: dominant contributors to the temperature anisotropy. The other secondary anisotropies are either
299: smaller in strength or contribute at even higher $l's$ or have different spectral dependence (Aghanim
300: \etal, 2000; Majumdar \etal, 2000).
301:
302: We have plotted the Poisson SZE power spectrum in Fig1, left panel. Clearly,
303: the primary feature distinguishing a non-evolutionary constant \gmf model from an evolutionary one is the
304: position of the peak. The model with a constant \gmf peaks at a higher $l$-value and also has greater power.
305: The constant \gmf
306: model peaks at $l\sim4000$. This result is in agreement with that of Komatsu \& Kitayama (1999).
307: If one assumes that there is no evolution of \fg with redshift (i.e s=0), the peak is at
308: $l\sim 1100$, whereas in the case of no dependence on mass (k=0), the peak is at $l\sim 2500$. Based
309: on EMSS
310: data (David \etal, 1990), Colafrancesco \& Vittorio (1994) (and also Molnar \& Birkinshaw 2000) model
311: \gmf with
312: k=0.5 \& s=1. For this case, we see that the turnover is at a very low $l\sim 900$. Assuming a mild
313: evolution (k=0.1, s=0.1), we get the peak at $l\sim 2100$. We also show results for (k=0.5, s=0.5) and (k=0.1,
314: s=0.5). The last parametrization is based on the recent analysis of $ROSAT$ data by Ettori \& Fabian (1999).
315: It is evident that the difference, in the $l$-value of the peak of the constant
316: \gmf scenario from an evolutionary one, can range between $l \sim 1500 - 3200$..
317: The position of the peak thus is a
318: strong distcriminatory signature of any evolution of \gmf.
319:
320: It is easy to understand the shift in the peak of the SZE power spectrum. Let us consider the case s=0, i.e.
321: \gmf depends only on total mass. From Eqn.(1), this means an enhanced reduction of \gmf of smaller mass clusters
322: relative to the
323: larger masses and so a reduction of power at larger $l's$, (since smaller masses contribute at larger
324: $l$). Hence, the peak shifts to a lower $l$. For the case k=0, (i.e only redshift dependence),
325: we now have structures at high $z$ contributing less to the power (than without a redshift dependence).
326: Since from PS formalism, less massive structures
327: are more abundant at high $z$, this negative dependence of \gmf on redshift cuts off their contribution more
328: than the more massive clusters. Hence, once again there is less power at high $l$ and the peak
329: shifts to lower $l$-value. The parametrization of Eqn. (1) affects the larger masses less, as evident from
330: almost equal power seen at $l \lsim 600$, for all models.
331: The net effect is a reduction of power at smaller angular scales, and
332: hence a shift in the position of the peak to a smaller multipole value.
333:
334:
335: We note that, these results are irrespective of the
336: arguments given (see Rines \etal, 1999) to explain any possible evolution of \gmf. In their case,
337: they {\it assume} \gmf to be constant and relies on the cosmology to change the angular diameter distance, so
338: that there is an `apparent' change in \gmf. In such a case, if there is `actually' even a slight evolution
339: of \gmf, then one can still account for it with a non-evolutionary model, by simply changing the cosmological
340: parameters. Our method {\it does not assume a priori} any constancy (or evolution) of \gmf and tries to look for it.
341:
342: In Fig1, right panel, we show the SZE clustering power spectrum. For all models, it falls of at a smaller $l$
343: w.r.t Poisson power spectrum. Since for clustering, the peak depends on the average inter-cluster separation, which
344: is fixed once the cosmology is fixed, there is no appreciable spread of the peaks in $l$-space.
345: The only difference is in their relative
346: power w.r.t each other which depends on the total gas mass available to distort the CMB.
347: Addition of the clustering power spectrum to the Poisson case results in slight shift of the peaks to lower
348: $l$'s.
349:
350: It maybe possible to measure the power spectrum of SZE with the ongoing and future high angular resolution
351: CMB observations. In principle, observations with SUZIE, OVRO, BIMA and ATCA can probe the range in $l$ from $\approx
352: 1000 - 7000$ and a frequency range of $\approx 2 - 350$ GHz. The SZE power spectrum would also be measured with
353: increased precision by the proposed ALMA
354: %\footnote{\url{http://www.alma.nrao.edu}}
355: and AMiBA (which is geared for blank sky surveys).
356:
357: Finally, let us comment on the validity and robustness of our results.
358: In Fig2, we show results for an open universe ($\Omega_0=0.35, h=0.65$). It is clearly seen that the
359: difference in the peak position of constant \gmf and evolutionary models remain far apart (Infact, for
360: same parameters of $k=0.5, s=1$, the difference increases to $\approx 4500$ from that of
361: $\approx 3000$ in a flat universe).
362: It is seen that the turnover of the SZE power spectrum is insensitive to the mass cutoff, since
363: main contribution to the anisotropy comes from clusters with
364: $10^{14}M_\odot <M<10^{15}M_\odot$.
365: In Fig2, we also indicate the effect of having a more compact gas distribution with $p=7$.
366: We see that shift in the peaks are negligible (though the height is reduced a little) and an
367: uncertainty as to how far the gas extends is not major.
368: The use of a single $\beta$ to model the full gas distribution introduces little error, though
369: a $\beta$-model fits the inner cluster regions better.
370: This is because the major contribution to the anisotropy comes from around the
371: core region, and increasing $\beta$ slightly decreases the overall distortion, without touching the peak.
372: Also, a modified $M-T$ relation (more suitable for $\Lambda$CDM) does not change the conclusions of this paper
373: (although amplitude of distortion slightly changes).
374: For a more detailed analysis, however, one should take better observationally supported gas density and temperature profiles
375: (see Yoshikawa \& Suto, 1999). These points will be discussed in greater detail in a future publication.
376:
377: In conclusion, we have computed the angular power spectrum of SZE from clusters of galaxies.
378: We have shown the position of the peak of the power
379: spectrum to bear a {\it strong discriminatory signature} of different \gmf models. One of the goals of arc minute scale
380: observations of the CMB
381: anisotropy is to measure the SZE power spectrum from blank sky surveys. Such observational results can be used to constrain
382: \gmf models. This also has the added advantage of being devoid of uncertainties that can creep in through `selection biases' in
383: estimating the \gmf using pointed studies of X-ray selected galaxy cluster. Our method, thus, provides a powerful
384: probe of evolution (or constancy) of gas mass fraction and can potentially resolve the decade long debate.
385:
386:
387: \acknowledgments
388: The author wishes to thank Joe Mohr and Biman Nath for critical comments, Sunita Nair and Dipankar Bhattacharya for
389: valuable suggestions and Pijush Bhattacharjee for encouragements. He also wishes to thank the Raman Reseach Institute,
390: Bangalore. After the completion of this work, we learnt about a complimentary work
391: by Carlstrom \& Holder (2001) which looks at changes in gas profiles due to preheating and its effect on SZE.
392:
393:
394: % ================================================================================
395:
396:
397: \beb
398:
399: \bi Aghanim, N., Balland, C., \& Silk, J., 2000, A\&A, 357, 1
400: \bi Allen, S. W., \& Fabian, A., 1998, MNRAS, 297, L57
401: %\bi Atrio-Barandela, F., \& M\"ucket, J. P. 1999,ApJ, 515, 465
402: \bi Bardeen, J. M., Bond, J. R.,Kaisaer, N. \& Szalay, A. S., 1986
403: ApJ, 304. 15
404: \bi Bialek, J. J., Evrard., A. E., \& Mohr, J, 2000, preprint, (astro-ph/0010584)
405: \bi Birkinshaw, M. 1999;Phys. Rep., 310, 97
406: \bi Bunn, E. \& White, M. 1997, ApJ, 480, 6
407: \bi Carlstrom, J. E., \& Holder, G., 2001, private communication
408: \bi Colafrancesco, S., \& Vittorio, N., 1994, ApJ, 422, 443
409: \bi Cole, S. \& Kaiser, N., 1988, MNRAS, 233, 637
410: \bi Cooray, A., Hu, W., \& Tegmark, M., 2000, preprint, (astro-ph/0002238)
411: \bi David, L. P., Arnaud, K. A., Forman, W., \& Jones, C., 1990. ApJ, 356, 32
412: \bi David, L. P., Jones, C., \& Forman, W., 1995, ApJ, 445, 578
413: \bi Ettori, S., \& Fabian, A. C., 1999, MNRAS, 30. 834
414: \bi Grego, L., \etal, 2000, preprint, (astro-ph/0012067)
415: \bi Holder, G. P., \& Carlstrom, J. E., 1999, preprint, (astro-ph/9904220)
416: %\bi Holzapfel,W.L, Carlstrom, J.E., Grego, L., Holder, G, Joy, M., \& Reese, 2000, ApJ, 539, 57
417: \bi Hu, W., Fukugita, M., Zaldarriaga, M., \& Tegmark, M., 2000, preprint, (astro-ph/006436)
418: \bi Jing, Y. P., 1999, ApJ, 515, L45
419: \bi Komatsu, E. \& Kitayama, T. 1999, ApJ, 526, L1
420: %\bi Limber, D., 1954, ApJ, 380, 1
421: %\bi Majumdar, S. \& Subrahmanyan, R., 2000, MNRAS, 312, 724
422: \bi Majumdar, S., Nath, B. B., \& Chiba, M., 2000, preprint, (astro-ph/0012016)
423: \bi Mohr, J., Mathiesen, B., \& Evrard, A., 1999, ApJ, 517, 627
424: \bi Molnar, S. M. \& Birkinshaw, M., 2000, ApJ, 537, 542
425: \bi Padmanabhan, T., \& Sethi, S., 2000, preprint, (astro-ph/0010309)
426: \bi Peebles, P. J. E. 1980;The Large Scale Structure of the
427: Universe;Princeton;Princeton Univ. Press;
428: \bi Ponman, T. J., Bourner, P. D. J., Ebeling, H., \& Bohringer, H., 1996, MNRAS, 283, 690
429: \bi Press, W. H., \& Schechter, P. 1974, ApJ, 187, 425
430: \bi Refregier, A., Komatsu, E., Spergel, D. N., \& Pen, U, 1999, preprint, (astro-ph/9912180)
431: \bi Rines, K., Forman, W., Pen, U., Jones, C., \& Burg, R., 1999, ApJ, 517, 70
432: \bi Sadat, R. \& Blanchard, A., 2001, preprint (astro-ph/0102010)
433: \bi Sasaki, S., 1996, PASJ, 48, L119
434: \bi Schindler, S., 1999, A\&S, 349, 435
435: \bi Sheth, R. K., \& Tormen, G., 1999, MNRAS, 308, 119
436: %\bi Subrahmanyan, R., Kesteven, M. J., Ekers, R. D., Sinclair, M., \& Silk, J. 1998, MNRAS, 298, 1189
437: \bi Sugiyama, N., 1995, ApJS, 100, 281
438: %\bi Suto, Y., Sasaki, S., \& Makino, N., 1998, ApJ, 509, 544
439: \bi Tsuru T.G., \etal, 1997, preprint, (astro-ph/9711353)
440: \bi White, D. A., \& Fabian, A. C., 1995, MNRAS, 273, 72
441: \bi Yoshikawa, K., \& Suto, Y., 1999, ApJ ,513, 549
442: \bi Zel'dovich, Ya. B., \& Sunyaev, R. A. 1969, Ap\&SS 4, 301;
443:
444: \eeb
445:
446:
447: % ==============================================================================================
448:
449: \clearpage
450: \begin{figure}
451: \plottwo{fig1a.eps}{fig1b.eps}
452: \caption{
453: The Poisson (left panel) and clustering (right panel) power spectra due to SZE from galaxy
454: clusters for different \gmf models. For both the panels, the thick solid line correponds to constant \gmf model, the
455: thick dashed line has no evolution with redshift and the thick dash-dotted line has no evolution with total mass.
456: The thin lines are for the cases: a) k=0.5, s=1; b) k=0.5, s=0.5; c) k=0.1, s=0.5 and d) k=0.1, s=0.1.
457: \label{fig1}
458: }
459: \end{figure}
460:
461: \clearpage
462: \begin{figure}
463: \plotone{fig2.eps}
464: \caption{
465: The Poisson SZE power spectra are plotted for different cosmologies and with different extension of the
466: gas mass. The solid lines are for a $\Lambda CDM$, with $\Omega_m=0.35, \Omega_\Lambda=0.65, h=0.65$, and the thin
467: lines are for $OCDM$ with $\Omega_m=0.35,h=0.65$. The $OCDM$ lines have been multiplied by a factor of 10 in the
468: plot. The solid and dashed lines are for gas mass extending upto
469: $10 r_c$, whereas the dash-dotted and the dotted lines are for extension upto $7 r_c$.
470: \label{fig2}
471: }
472: \end{figure}
473:
474:
475: \end{document}
476: %
477: ==================================================================================
478: