astro-ph0206292/ms.tex
1: \documentclass[letter,english]{article}
2: \usepackage{emulateapj,onecolfloat,epsfig}
3: 
4: %\topmargin 1.0cm
5: 
6: \begin{document}
7: \newcommand {\ds}{\displaystyle}
8: \def\sun{\hbox{$\odot$}}
9: 
10: \twocolumn
11: [
12: \title{Weak Lensing as a Calibrator of the Cluster Mass-Temperature Relation}
13: \author{Dragan Huterer$^1$ and Martin White$^2$}
14: \affil{$^1$Department of Physics, Case Western Reserve University, 
15: 	Cleveland, OH~~44106}
16: \affil{$^2$Departments of Astronomy and Physics, University of California, 
17: 	 Berkeley, CA~~94720}	
18: 
19: \begin{abstract}
20: The abundance of clusters at the present epoch and weak gravitational
21: lensing shear both constrain roughly the same combination of the power
22: spectrum normalization $\sigma_8$ and matter energy density
23: $\Omega_M$. The cluster constraint further depends on the
24: normalization of the mass-temperature relation. Therefore, combining
25: the weak lensing and cluster abundance data can be used to accurately
26: calibrate the mass-temperature relation.  We discuss this
27: approach and illustrate it using data from recent surveys.
28: \end{abstract}
29: \keywords{cosmology: theory -- large-scale structure of universe} ]
30: 
31: \section{Introduction}
32: 
33: The number density of galaxy clusters as a function of their mass, the mass
34: function, and its evolution can provide a powerful probe of models of
35: large-scale structure.
36: Historically the most important constraint coming from the present day
37: abundance of rich clusters has been the normalization of the {\it linear
38: theory\/} power spectrum of mass density perturbations (e.g.\
39: \cite{Evr89,FWED,BonMye91,HA91,Lil,OukBla,BahCen,WEF,VL96,VL98,Henry}).
40: The normalization is typically quoted in terms of $\sigma_8$, the rms density
41: contrast on scales $8\,h^{-1}\,$Mpc, with the abundance constraint forcing
42: models to a thin region in the $\Omega_M$-$\sigma_8$ plane.
43: 
44: Since the mass, suitably defined, of a cluster is not directly observable, one
45: typically measures the abundance of clusters as a function of some other
46: parameter which is used as a proxy for mass.
47: Several options exist, but much attention has been focused recently on the
48: X-ray temperature.
49: Cosmological N-body simulations and observations suggest that X-ray
50: temperature and mass are strongly correlated with little scatter
51: (\cite{EMN,BryNor,ENF,HorMS,NevMF}).
52: How well simulations agree with observational results is far from clear,
53: and several issues need to be resolved.  On the simulation side there are
54: the usual issues of numerical resolution and difficulties with including all
55: of the relevant physics.  On the observational side instrumental effects can
56: be important (especially for the older generation of X-ray facilities) in
57: addition to the worrying lack of a method for estimating ``the mass''.
58: In this respect it is worth noting that there are numerous differing
59: definitions of which ``M'' and ``T'' are to be related in the M--T relation
60: (\cite{White_mass})!
61: 
62: With current samples the {\it dominant\/} uncertainty in the normalization
63: in fact comes from the normalization of the M--T relation
64: (\cite{ECF96,VL96,DV99,Henry,PSW,Seljak}).
65: Or phrased another way, the cluster abundance is a sensitive probe of the
66: normalization of the M--T relation.
67: 
68: The abundance of clusters is, of course, not the only way to constrain the
69: cosmological parameters.
70: In this regard it is interesting to note that weak gravitational lensing 
71: provides a constraint on a very similar combination of $\Omega_M$ and
72: $\sigma_8$.
73: Therefore, the two constraints can be combined to check for consistency of
74: our cosmological model, to provide a normalization for the M--T relation,
75: to probe systematics in either method and/or to measure other parameters
76: not as yet included in the standard treatments.
77: 
78: While the cluster constraint comes primarily from scales of about
79: $R=10\,h^{-1}\,$Mpc, current weak lensing surveys constrain somewhat
80: smaller scales.  These surveys probe scales between roughly 1 and 10
81: arcmin, which for source galaxies located at $z\simeq 1$ in a
82: $\Lambda$CDM cosmology corresponds to $0.7\,h^{-1}\,{\rm
83: Mpc}<R<7\,h^{-1}\,{\rm Mpc}$.  Therefore, weak lensing probes
84: slightly smaller scales than clusters.  As lensing surveys push to larger
85: scales the overlap will become even better.
86: 
87: In this paper we argue that a natural application of combining the
88: cluster abundance and weak lensing constraints is to calibrate the
89: M--T relation for galaxy clusters (see also \cite{HuKrav}).  In
90: Sec.~\ref{sec:MT} we define the M--T relation and derive how cluster
91: abundance constraints depend on $\Omega_M$ and $\sigma_8$.  In
92: Sec.~\ref{sec:WL_clus} we illustrate how combining the two constraints
93: can fix the normalization of the M--T relation using two recently
94: obtained data sets.  Finally, in Sec.~\ref{sec:concl} we discuss this
95: approach further.
96: 
97: \section{The Mass-temperature relation}\label{sec:MT}
98: 
99: Throughout we shall be interested in the abundance of massive clusters at
100: low redshifts, so we parameterize the M--T relation as
101: \begin{equation}
102:   {M(T,z)\over M_{15}}
103:   = \left( {T\over T_*} \right)^{3/2}
104:     \left(\Delta_c E^2 \right)^{-1/2} 
105:     \left[ 1 -  2  {\Omega_{\Lambda}(z)\over\Delta_c  }\right]^{-3/2}
106: \label{eq:MT}
107: \end{equation}
108: where $M_{15}= 10^{15}\,h^{-1}\,{\rm M}_\odot$,
109: $\Delta_c$ is the mean overdensity inside the virial radius in units of the
110: critical density, which we compute using the spherical top-hat collapse
111: model, and $E^2=\Omega_M (1+z)^3 + \Omega_{\Lambda} + \Omega_{\rm k} (1+z)^2$.
112: $T_*$ is the normalization coefficient that we seek to constrain; it roughly
113: corresponds to the temperature of a
114: $M=7.5\times 10^{13}\,h^{-1}\,{\rm M}_\odot$ cluster.
115: If measured in keV, the value of $T_*$ is precisely equivalent to $\beta$
116: from Pierpaoli, Scott \& White (2002) and is $1.34f_T$ of 
117: Bryan \& Norman (1998).
118: 
119: Let us explore the sensitivity of cluster abundance on $\Omega_M$ and
120: $\sigma_8$.  The Press-Schechter formula gives the number of collapsed
121: objects $dn$ per mass interval $d\ln M$ (\cite{PS}); we define
122: ${\cal N}(M,z)=dn/d\ln M$.
123: Further defining $\nu\equiv \delta_c/\sigma(M, z)$, where $\sigma(M, z)$
124: is the rms density  fluctuation on mass-scale $M$ evaluated at redshift $z$
125: using linear theory and $\delta_c\approx 1.686$ is the linear threshold
126: overdensity for collapse, we have
127: \begin{equation}
128:   {\cal N}(M,z)= \sqrt{\frac{2}{\pi}} \frac{\rho_M}{M} 
129:   \frac{d\ln\sigma(M, z)}{d\ln\nu}\,
130:   \nu\,\exp\left( -\nu^2/2\right)
131: \label{eq:dndm}
132: \end{equation}
133: where $\rho_M$ is the present-day matter density. Assuming 
134: we are dealing with the current cluster abundance, $z\simeq 0$.
135: Following Pen (\cite{Pen}), for the mass scales of interest we can
136: approximate $\sigma(M)\propto M^{-\alpha}$ where $\alpha\simeq 0.27$ for
137: the currently popular $\Lambda$CDM cosmology.
138: 
139: Let us examine the dependence of ${\cal N}$ on $\Omega_M$, $\sigma_8$ and
140: $M$.  Ignoring the term $d\ln\sigma/d\ln\nu$ (which slowly varies) one
141: obtains
142: \begin{equation}
143:   {\delta{\cal N}\over {\cal N}} = \frac{\delta\Omega_M}{\Omega_M}
144:   (1-\alpha+\nu^2\alpha)+ \frac{\delta\sigma_8}{\sigma_8}(\nu^2-1)
145:   -\frac{\delta M}{M}(1-\alpha+\nu^2\alpha)
146: \end{equation}
147: Setting the left-hand side to zero and using the fact that $\delta
148: M/M=-3/2 \:\delta T_*/T_*$, for our fiducial cosmology and massive
149: clusters ($M\sim 10^{15}\,h^{-1}M_\odot$, or $\nu\simeq 2$) we
150: have\footnote{Note that the dependence of $M$ (or $T_*$) on $\sigma_8$ is
151: stronger for more massive clusters; a more detailed analysis gives
152: $T_*\propto \sigma^{-5/3}$ for the most massive clusters (\cite{Evr02}).}
153: \begin{equation}
154:   T_* \propto (\sigma_8\Omega_M^{0.6})^{-1.1}.
155: \end{equation}
156: 
157: Therefore, measurements of the cluster abundance at the present epoch
158: constrain a degenerate combination of $T_*$ and $\sigma_8\,\Omega_M^{0.6}$.
159: One of them cannot be determined without knowing the other.
160: Thankfully, weak lensing happens to measure roughly this combination of
161: $\Omega_M$ and $\sigma_8$ accurately, and the orthogonal combination much
162: less accurately (e.g.~\cite{Ber}).
163: Consequently, weak lensing in conjunction with cluster abundance can be
164: used to constrain $T_*$ quite strongly.
165: 
166: \section{Weak lensing plus clusters: an example}\label{sec:WL_clus}
167: 
168: As a more concrete example of these ideas, let us examine what value of
169: $T_*$ is required to bring current cluster and weak lensing results into
170: agreement.  This analysis will necessarily be illustrative, but is already
171: quite enlightening.
172: 
173: \subsection{The cluster data}
174: 
175: We compute $\sigma_8$ using a Monte-Carlo method following the steps
176: outlined in Pierpaoli, Scott \& White (2002).  Since some of the
177: details have changed we sketch the procedure here.
178: 
179: We use the HiFluGCS cluster sample of Reiprich \& B{\" o}hringer
180: (1999), restricted to clusters with $0.03<z<0.10$.  For simplicity we
181: do not include `additional' clusters of lower flux/temperature which
182: could scatter into the sample.  The cosmic microwave background (CMB)
183: frame redshifts from Struble \& Rood (\cite{StrRoo}) were used when
184: available and so were the two-component temperatures published in
185: Ikebe et al.~(2002).  For each $\Omega_{\rm m}$ we sample from a
186: distribution of cosmological parameters including $h$, $n$ and $T_*$
187: (the normalization of the M--T relation).  For each such realization
188: we generate 50 mass functions, where the temperature is chosen from a
189: Gaussian with the mean and variance appropriate to the observational
190: value and errors, and a scatter of 15\% in mass at fixed $T$ is
191: assumed for the M--T relation.  Using the mean values of the M--T
192: relation and the L--T relation from Ikebe et al.~(2002)
193: \begin{equation}
194:   L_X = 1.38\times 10^{35}\ \left( {kT\over 1{\rm keV}} \right)^{2.5}
195:   \quad h^{-2}{\rm W}
196: \end{equation}
197: we compute the volume to which clusters of mass $M$ could be seen above
198: the flux limit
199: $f_{\rm lim}=1.99\times 10^{-14}\,{\rm erg}\,s^{-1}\,{\rm cm}^{-2}$
200: of the survey.
201: For each realization of the mass function we compute the best fitting
202: $\sigma_8$ by maximizing the Poisson likelihood of obtaining that set of
203: masses from the theory with all parameters except $\sigma_8$ fixed.
204: The mass function can be computed using either the Press-Schechter (1974),
205: Sheth-Tormen (\cite{SheTor}) or Jenkins et al.~(\cite{JFWCCEY}) formulae.
206: We have used the Sheth-Tormen prescription throughout, with the mass variance
207: $\sigma^2(M)$ computed using the transfer function fits of Eisenstein
208: \& Hu~(\cite{EisHu}) and masses converted from $M_{180\Omega}$ to
209: $M_{\Delta_c}$ assuming an NFW profile (\cite{NFW}) with $c=5$.
210: The best fitting $\sigma_8$ is corrected from $\bar{z}$ to $z=0$.
211: The mean of the 50, $z=0$ normalizations is then taken as the fit for
212: that set of cosmological parameters (since the error from Poisson sampling
213: is completely sub-dominant to the error in the M--T normalization we do
214: not keep track of it here).  When quoting a best fit for a given triplet
215: of ($\Omega_{\rm m}$, $\sigma_8$, $T_*$), we marginalize (average) over the
216: other cosmological parameters $h$ and $n$.
217: 
218: \subsection{The weak lensing data}
219: 
220: As an example of weak lensing measurements, we use shear measurements
221: obtained using Keck and William Herschel telescopes (\cite{Bacon}).
222: These joint measurements used two independent telescopes covering 0.6
223: and 1 square degrees respectively, and enabled careful assessment of
224: instrument-specific systematics.
225: The authors compute the shear correlation function, and compare with the
226: theoretical prediction.
227: Assuming the shape parameter $\Gamma=0.21$, the results are well fit by
228: \begin{equation}
229:   \sigma_8 \left ({\Omega_M\over 0.3}\right )^{0.68}=0.97\pm 0.13
230: \end{equation}
231: which captures the total 68\% CL error: statistical, redshift uncertainty
232: and uncertainty in the ellipticity-shear conversion factor.
233: These results are consistent with other recent measurements of cosmic shear
234: (\cite{vW02,Refregier,Hoekstra}).
235: 
236: \subsection{Calibrating the M-T relation}
237: 
238: Fig.~\ref{fig:WL_clus} shows the constraints in the $\Omega_M$-$\sigma_8$
239: plane.  The cluster constraint has been marginalized over $h$ and $n$ as
240: explained above, and plotted for three different values of $T_*$.
241: We have checked that the allowed ranges for $h$ and $n$ are wide enough so
242: that essentially all of the likelihood is contained within those ranges.
243: The weak lensing constraints assume the shape parameter $\Gamma=0.21$.
244: Note that the constraint regions from the two methods are indeed parallel,
245: with very similar degeneracy directions.
246: This enables an accurate determination of the normalization $T_*$.
247: 
248: In the example above, we see that a relatively low $T_*$ is preferred
249: ($T_*\lesssim 1.7\,$keV) in order for cluster results to agree with
250: the weak lensing results.  While systematics in both methods could
251: still be important, it is interesting to note that this result is in line
252: with most earlier estimates (\cite{EMN,ENF,BryNor,YJS}), while it disagrees
253: with values adopted more recently (e.g.~\cite{Seljak}).
254: 
255: \begin{figure}[t]
256: \epsfig{file=f1.eps, height=3in, width=3.5in}
257: \caption{68\% CL uncertainty contours in the $\Omega_M$--$\sigma_8$ plane, 
258: for a weak lensing survey (\cite{Bacon}) and 95\% CL uncertainties
259: for a cluster survey (\cite{ReiBoh}).  The cluster results are shown
260: for three different values of the mass-temperature normalization
261: parameter $T_*$, and marginalized over $n$ and $h$.  The degeneracy
262: regions for the two methods are very similar, which in principle
263: enables an accurate determination of $T_*$.}
264: \label{fig:WL_clus}
265: \end{figure}
266: 
267: The fact that cluster abundance and weak lensing probe different
268: scales opens a possibility that one might be able to secure the
269: agreement between the two methods by varying the shape of the power
270: spectrum or the spectral index $n$ rather than the M--T normalization.
271: Unfortunately the constraints we have combined above have individually
272: been marginalized over $h$ and $n$.
273: Ideally, one would combine the cluster and weak lensing likelihood functions
274: and then marginalize over the relevant parameters to get the probability
275: distribution of $T_*$:
276: \begin{eqnarray}
277:   P(T_*) &=& \int \mathcal{L}_{\rm clus}(T_*, \Omega_M, \sigma_8, n, h)
278:   \nonumber \\ 
279:   &\times& \mathcal{L}_{\rm WL}(\Omega_M, \sigma_8, n, h)\,
280: 	d\Omega_M\, d\sigma_8\,dn\,dh.
281: \end{eqnarray}
282: Then the results would be manifestly independent of the power spectrum
283: parameters.  We do not have the ability to perform such an analysis here.
284: 
285: Note, however, that the scales probed by lensing and clusters are
286: quite close, separated an order of magnitude at most.  For example, it
287: would require a spectral tilt of $n\sim 1.2$ to make the recently
288: obtained ``low'' normalization from cluster abundance ($\sigma_8\sim
289: 0.6$) agree with the ``high'' normalization from weak lensing
290: ($\sigma_8\sim 0.9$), and such a high value of $n$ is already
291: disfavored by recent CMB experiments (\cite{max,boom,DASI,CBI}).
292: 
293: \section{Conclusions}\label{sec:concl}
294: 
295: There has been a lot of discussion recently regarding the value of
296: cluster normalization $\sigma_8$.  While the ``old'' results favor
297: $\sigma_8\sim 1$ (\cite{VL98,PSW} and references therein), several new
298: cluster abundance analyses favor a significantly lower normalization
299: (\cite{ReiBoh,Bor01,Viana_Nichol,Seljak,Ikebe,SDSS}).  The lower
300: normalization is also favored by the combined analysis of 2dF Galaxy
301: Redshift Survey and CMB data (\cite{2dF}).  On the other hand, recent
302: weak lensing results (\cite{vW02,Bacon,Refregier,Hoekstra}) tend to
303: favor a higher value of $\sigma_8$. The cause is of this discrepancy
304: between various measurements has not been identified yet; one
305: candidate is larger than anticipated systematic errors in one or both
306: methods.  Another possibility is the bias in the relation between the
307: mass and the observable quantity --- temperature or luminosity ---
308: used to construct the abundance of clusters.
309: 
310: \vspace{0.2cm}
311: \begin{figure}[!ht]
312: \epsfig{file=f2.eps, height=3in, width=3.5in}
313: \caption{Estimates of the M--T normalization $T_*$ 
314: collected from the literature. The red (light) points are estimates
315: from simulations, while the black (dark) points are from the
316: observations. Points with no error bars had none quoted. The shaded
317: region is roughly our favored range of values of $T_*$.}
318: \label{fig:MT_lit}
319: \end{figure}
320: 
321: The cluster abundance constraint on $\sigma_8$ crucially depends on
322: the M--T normalization $T_*$.  Figure~\ref{fig:MT_lit} summarizes the
323: current status of our knowledge of $T_*$. It shows seven
324: determinations from N-body simulations and three from direct
325: observations, as compiled in Pierpaoli, Scott \& White (2002) and Muanwong et
326: al.\ (2002). The shaded region is roughly our favored range of values
327: of $T_*$. Points without error bars had none quoted, and the three
328: observed values of $T_*$ assumed the isothermal-$\beta$
329: model. The measurement due to Muanwong et al.\ corresponds to their
330: ``radiative'' and ``preheating'' cases that are cooling-flow
331: corrected, while the value due to Pierpaoli, Scott \& White is an
332: average over the simulations.  The large discrepancy between the
333: different measurements is apparent, and it also appears that the
334: observed values are systematically higher than the ones obtained from
335: simulations (see Muanwong et al.~2002 for further discussion).
336: 
337: We argue here that the cluster abundance -- weak lensing complementarity can
338: be used to cross-check the M--T relation.
339: By combining recent weak lensing constraints from Bacon et al.\ and the
340: HiFluGCS cluster sample of Reiprich \& B{\" o}hringer, we have demonstrated
341: the utility of this method.
342: While potential systematic errors in both data sets are still a concern,
343: the example we used prefers relatively low values of the M--T normalization
344: ($T_*\lesssim 1.7\,$keV).
345: We conclude that future weak lensing surveys (Vista, LSST, SNAP) combined
346: with new cluster data from Chandra and XMM-Newton observations will provide
347: a strong probe of the M-T relation.
348: 
349: \section*{Acknowledgments}
350: DH is supported by the DOE grant to CWRU.  MW is supported by NASA and
351: by the Sloan Foundation.
352: 
353: \begin{thebibliography}{99}
354: \bibitem[Bacon et al.\ 2002]{Bacon}
355: 	Bacon, D.J., Massey, R.J., Refregier, A.R., \& Ellis, R.S.
356: 	2002, submitted to MNRAS [astro-ph/0203134]
357: 
358: \bibitem[Bahcall \& Cen 1993]{BahCen}
359: 	Bahcall, N.\ \& Cen, R.\ 1993, ApJ, 407, L49
360: 
361: \bibitem[Bahcall et al.\ 2002]{SDSS}
362: 	Bahcall, N.\ et al.\ 2002, submitted to ApJ [astro-ph/0205490]
363: 
364: \bibitem[Balbi et al.\ 2000]{max}
365: 	Balbi, A.\ et al.\ 2000, ApJ, 545, L1
366: 
367: \bibitem[Bernardeau et al.\ 1997]{Ber}
368: 	Bernardeau, Y, van Waerbeke, L., \& Mellier, Y. 1997, A\&A 322, 1
369: 
370: \bibitem[Bond \& Myers 1991]{BonMye91}
371: 	Bond, J.R., \& Myers, S.T.\ 1991, 
372: 	Trends in Astroparticle Physics, eds.~D. Cline,
373: 	R. Peccei, World Scientific, Singapore, p.\,262
374: 
375: \bibitem[Borgani et al.\ 2001]{Bor01}
376: 	Borgani, S.\ et al.\ 2001, ApJ, 561, 13
377: 
378: \bibitem[Bryan \& Norman 1998]{BryNor}
379: 	Bryan, G.L.\ \& Norman, M.L.\ 1998, ApJ, 495, 80
380: 
381: \bibitem[Donahue \& Voit 1999]{DV99}
382: 	Donahue, M., \& Voit, G.M.\ 1999, ApJ, 523, L137
383: 
384: \bibitem[1999]{EisHu}
385: 	Eisenstein, D., \& Hu, W.\ 1999, ApJ, 511, 5
386: 
387: \bibitem[Eke, Cole \& Frenk 1996]{ECF96}
388: 	Eke, V., Cole, S., \& Frenk, C.S.\ 1996, MNRAS, 282, 263
389: 
390: \bibitem[Eke, Navarro \& Frenk 1998]{ENF}
391: 	Eke, V., Navarro, J.F., \& Frenk, C.S. 1998, ApJ, 503, 569
392: 
393: \bibitem[Evrard 1989]{Evr89}
394: 	Evrard, A.E.\ 1989, ApJ, 341, L71
395: 
396: \bibitem[Evrard, Metzler \& Navarro 1996]{EMN}
397: 	Evrard, A.E., Metzler, C., \& Navarro, J.F.\ 1996, ApJ, 469, 494
398: 
399: \bibitem[Evrard et al.\ 2002]{Evr02}
400: 	Evrard, A.E.\ et al.\ 2002, ApJ, 573, 7
401: 
402: \bibitem[Frenk et al.\ 1990]{FWED}
403: 	Frenk, C.S., White, S.D.M., Efstathiou, G., \& Davis M.\ 1990, 
404: 	ApJ, 351, 10
405: 
406: \bibitem[Henry 2000]{Henry}
407: 	Henry, J.P.\ 2000, ApJ, 534, 565
408: 
409: \bibitem[Henry \& Arnaud 1991]{HA91}
410: 	Henry, J.P., \& Arnaud, K.A.\ 1991, ApJ, 372, 410
411: 
412: \bibitem[Hoekstra, Yee \& Gladders 2002]{Hoekstra}
413: 	Hoekstra, H., Yee, H.K.C., \& Gladders, M.D.\ 2002,	
414: 	submitted to ApJ [astro-ph/0204295]
415: 
416: \bibitem[Horner, Mushotsky \& Scharf 1999]{HorMS}
417: 	Horner, D.J., Mushotsky, R.F., \& Scharf, C.A.\ 1999, ApJ, 520, 78
418: 
419: \bibitem[Hu \& Kravtsov 2002]{HuKrav}
420: 	Hu, W., \& Kravtsov, A. 2002, ApJ, submitted [astro-ph/0203169]
421: 
422: \bibitem[Ikebe et al.\ 2002]{Ikebe}
423: 	Ikebe, Y., Reiprich, T.H., B{\" o}hringer, H., Tanaka, Y.,
424: 	\& Kitayama, T. 2002, A\&A, 383, 773
425: 
426: \bibitem[2001]{JFWCCEY}
427: 	Jenkins, A., Frenk, C.S., White, S.D.M., Colberg, J.M., Cole, S., 
428: 	Evrard, A.E., \& Yoshida, N.\ 2001, MNRAS, 321, 371 
429: 
430: \bibitem[Lahav et al.\ 2001]{2dF}
431: 	Lahav, O., et al.\ 2001, MNRAS, 333, 961
432: 
433: \bibitem[Lilje 1992]{Lil}
434: 	Lilje, P.B.\ 1992, ApJ, 386, L33
435: 
436: \bibitem[1998]{Mar98}
437: 	Markevitch, M.\ 1998, ApJ, 504, 27
438: 
439: \bibitem[Muanwong et al.\ 2002]{Muanwong}
440: 	Muanwong, O., Thomas, P.A., Kay, S.T., \& Pearce, F.R.\ 2002,
441: 	MNRAS, in press [astro-ph/0205137]
442: 
443: \bibitem[Navarro, Frenk \& White 1997]{NFW}
444: 	Navarro J.F., Frenk C.S., White S.D.M., 1997, \apj, 490, 493
445: 
446: \bibitem[Netterfield et al.\ 2002]{boom}
447: 	Netterfield, C.B.\ et al.\ 2002, ApJ, 571, 604
448: 
449: \bibitem[Nevalainen, Markevitch \& Forman 2000]{NevMF}
450: 	Nevalainen, J., Markevitch, M., \& Forman, W. 2000, ApJ, 532, 694
451: 
452: \bibitem[Oukbir \& Blanchard 1992]{OukBla}
453: 	Oukbir, J., \& Blanchard, A.\ 1992, A\&A, 262, L21
454: 
455: \bibitem[1998]{Pen}
456: 	Pen, U.-L.\ 1998 ApJ, 498, 60
457: 
458: \bibitem[Pierpaoli, Scott \& White 2002]{PSW}
459: 	Pierpaoli, E., Scott, D. and White, M.\ 2002, MNRAS, 325, 77
460: 
461: \bibitem[Press \& Schechter 1974]{PS}
462: 	Press, W.H., \& Schechter, P.\ 1974, ApJ, 187, 452
463: 
464: \bibitem[Pryke et al.\ 2002]{DASI}
465: 	Pryke, C.\ et al.\ 2002, ApJ, 568, 46
466: 
467: \bibitem[Refregier, Rhodes \& Groth 2002]{Refregier}
468: 	Refregier, A., Rhodes, J. \& Groth, E.J.\ 2002,
469: 	ApJ, 572, L131
470: 
471: \bibitem[Reiprich \& B{\" o}hringer 1999]{ReiBoh}
472: 	Reiprich, T.H., \& B{\" o}hringer H.\ 2002, ApJ, 567, 716
473: 
474: \bibitem[Seljak 2002]{Seljak}
475: 	Seljak, U.\ 2001, submitted to MNRAS [astro-ph/0111362]
476: 
477: \bibitem[1999]{SheTor}
478: 	Sheth, R.K., \& Tormen, G.\ 1999, MNRAS, 308, 119
479: 
480: \bibitem[Sievers et al.\ 2002]{CBI}
481: 	Sievers, J.L.\ et al.\ 2002, submitted to ApJ [astro-ph/0205387]
482: 
483: \bibitem[1999]{StrRoo}
484: 	Struble, M.F., \& Rood, H.\ 1999, ApJS, 125, 35
485: 
486: \bibitem[van Waerbeke et al.\ 2002]{vW02}
487: 	van Waerbeke, L., Mellier, Y., Pell\'{o}, R., Pen, U.-L., 
488: 	McCracken, H.J., \& Jain, B.\ 2002, submitted to A\&A 	
489: 	[astro-ph/0202503]
490: 
491: \bibitem[Viana \& Liddle 1996]{VL96}
492: 	Viana, P.T.P., \& Liddle, A.\ 1996, MNRAS, 281, 323
493: 
494: \bibitem[Viana \& Liddle 1999]{VL98}
495: 	Viana, P.T.P., \& Liddle, A.\ 1999, MNRAS, 303, 535
496: 
497: \bibitem[Viana et al.\ 2002]{Viana_Nichol}
498: 	Viana, P.T.P., Nichol, R.C., \& Liddle, A.R.\ 2002, ApJ, 569, L75
499: 
500: \bibitem[White 2001]{White_mass}
501: 	White, M.\ 2001, A\&A 367, 27
502: 
503: \bibitem[White, Efstathiou \& Frenk 1993]{WEF}
504: 	White, S.D.M., Efstathiou, G., \& Frenk, C.S.\ 1993, MNRAS, 262, 1023
505: 
506: \bibitem[Xu et al.\ 2001]{Xu}
507: 	Xu, H., Jin, G., \& Wu, X.-P.\ 2001, ApJ, 553, 78
508: 
509: \bibitem[Yoshikawa, Jing \& Suto 2000]{YJS}
510: 	Yoshikawa, K., Jing, Y.P., \& Suto, Y.\ 2000, ApJ, 535, 593
511: 
512: \end{thebibliography}
513: 
514: \end{document}
515: