1: \documentclass[12pt,preprint]{aastex}
2:
3: \def\ra#1#2#3{#1$^{\rm h}$#2$^{\rm m}$#3$^{\rm s}$}
4: \def\dec#1#2#3{$#1^\circ#2'#3''$}
5: \def\swift{{\it Swift}}
6: \def\hete{{\it HETE-2}}
7: \def\nod{\nodata}
8:
9: \def\prince{1}
10: \def\ociw{2}
11: \def\hubble{3}
12:
13: \shorttitle{Galaxy Clusters Associated with Short GRBs II}
14: \shortauthors{Shin and Berger}
15:
16: \begin{document}
17:
18: \title{Galaxy Clusters Associated with Short GRBs. II. Predictions
19: for the Rate of Short GRBs in Field and Cluster Early-Type Galaxies}
20:
21: \author{
22: M.-S.~Shin\altaffilmark{\prince}
23: and E.~Berger\altaffilmark{\ociw,}\altaffilmark{\prince,}\altaffilmark{\hubble}
24: }
25:
26: \altaffiltext{\prince}{Princeton University Observatory, Peyton Hall,
27: Ivy Lane, Princeton, NJ 08544}
28:
29: \altaffiltext{\ociw}{Observatories of the Carnegie Institution of
30: Washington, 813 Santa Barbara Street, Pasadena, CA 91101}
31:
32: \altaffiltext{\hubble}{Hubble Fellow}
33:
34: \begin{abstract}
35: We determine the relative rates of short GRBs in cluster and field
36: early-type galaxies as a function of the age probability distribution
37: of their progenitors, $P(\tau)\propto \tau^n$. This analysis takes
38: advantage of the difference in the growth of stellar mass in clusters
39: and in the field, which arises from the combined effects of the galaxy
40: stellar mass function, the early-type fraction, and the dependence of
41: star formation history on mass and environment. This approach
42: complements the use of the early- to late-type host galaxy ratio, with
43: the added benefit that the star formation histories of early-type
44: galaxies are simpler than those of late-type galaxies, and any
45: systematic differences between progenitors in early- and late-type
46: galaxies are removed. We find that the ratio varies from $R_{\rm
47: cluster}/R_{\rm field}\sim 0.5$ for $n=-2$ to $\sim 3$ for $n=2$.
48: Current observations indicate a ratio of about $2$, corresponding to
49: $n\sim 0-1$. This is similar to the value inferred from the ratio of
50: short GRBs in early- and late-type hosts, but it differs from the
51: value of $n\approx -1$ for NS binaries in the Milky Way. We stress
52: that this general approach can be easily modified with improved
53: knowledge of the effects of environment and mass on the build-up of
54: stellar mass, as well as the effect of globular clusters on the short
55: GRB rate. It can also be used to assess the age distribution of Type
56: Ia supernova progenitors.
57: \end{abstract}
58:
59:
60: \keywords{gamma-rays:bursts --- galaxies:clusters --- galaxies:formation}
61:
62: \section{Introduction}
63: \label{sec:intro}
64:
65: Gamma-ray bursts (GRBs) are divided into two broad classes of
66: short/hard and long/soft bursts \citep{kmf+93}, which appear to have
67: different progenitor populations. Observations of Type Ic supernovae
68: in association with long GRBs provide a direct confirmation that they
69: arise from the death of massive stars \citep{hsm+03,smg+03}. Short
70: GRBs, on the other hand, have long been suspected to arise from the
71: merger of neutron star and/or black hole binaries (NS-NS, NS-BH; e.g.,
72: \citealt{elp+89,npp92,rr02,ajm05}). During the last year, follow-up
73: observations of \swift\ and \hete\ short GRBs have provided initial
74: confirmation to this idea, based in particular on the localization of
75: some short GRBs to elliptical galaxies \citep{bpc+05,gso+05,bpp+06},
76: their lack of association with bright supernovae
77: \citep{ffp+05,hwf+05,sbk+06}, and their lower energy release and wider
78: beaming angles compared to long GRBs \citep{bgc+06,sbk+06}.
79:
80: Despite this observational progress we are still missing a clear
81: understanding of the progenitor population, due to the lack of direct
82: observations (circumburst chemical abundances, gravitational waves, or
83: a sub-relativistic, radioactive component: \citealt{lp98,kul05}).
84: Thus, statistical studies of the burst properties can be highly
85: effective in understanding their progenitor population(s) (e.g.,
86: \citealt{gno+05,ngf05,gp06}). Recently, \citet{zr06} investigated the
87: ratio of short GRBs in early- and late-type galaxies as a constraint
88: on the age distribution of the progenitors. Their analysis combines
89: the global star formation rate in each galaxy type with a local galaxy
90: stellar mass function, assuming the formation process of short GRB
91: progenitors does not depend on any other physical parameters. Since
92: each galaxy type has experienced a globally different star formation
93: history, the ratio of bursts in each type is predicted to vary
94: as a function of the progenitor age distribution.
95:
96: A complementary way to constrain the progenitor age distribution is to
97: use the rates of short GRBs in clusters and the field
98: (\citealt{bsm+06}; hereafter Paper I). This approach takes advantage
99: of the following differences between cluster and field environment.
100: First, the galaxy stellar mass
101: function of clusters is more heavily dominated by massive galaxies
102: than in the field \citep{cfn+05,bbb+06}. Since the star formation
103: history is mainly determined by galaxy mass, the overall growth of
104: stellar mass is in turn affected by the large-scale environment.
105: Second, the fraction of early-type galaxies is larger in clusters than
106: in the field \citep{dre80,wgj93,bbb+06}. Finally, there appears to be
107: a systematic offset in the star formation histories of cluster and
108: field early-type galaxies of $\sim 1-3$ Gyr (e.g.,
109: \citealt{brc+98,ksc+02,tmb+05}). These effects lead to an overall
110: difference in cluster and field star formation histories, which
111: combined with the progenitor age distribution, is expected to affect
112: the relative fraction of short GRBs in each environment.
113:
114: In this paper we quantify these effects and show how the ratio of
115: short GRBs in cluster and field early-type galaxies can be used to
116: understand the age distribution of the progenitors. This is part of
117: our on-going systematic study of galaxy clusters hosting short GRBs,
118: using multi-slit optical spectroscopy and X-ray observations (Paper
119: I). We find that the current observations, albeit with a small number
120: of events, favor $P(\tau)\propto \tau^n$ with $n\sim 0-1$. Finally,
121: we provide a comparison of the systematic effects in this approach and
122: the approach of \citet{zr06}.
123:
124:
125:
126: \section{The Rate of Short Bursts in Clusters and the Field}
127: \label{sec:theory}
128:
129: The approach used by \citet{zr06} can be generalized to formulate the
130: short GRB rate per unit volume in cluster and field early-type
131: galaxies at $z\sim 0$, using the star formation history function,
132: ${\rm SFH}(\tau)$ instead of a star formation rate function:
133: \begin{equation}
134: R_i=C\int^{t(z=\infty)}_{0} {\rm SFH}_i(\tau)\, P(\tau)\, d\tau.
135: \label{eqn:rate}
136: \end{equation}
137: Here $\tau$ is both the look-back time and the time delay of a short
138: GRB progenitor, and $i$ designates a cluster or field environment.
139: $P(\tau)$ represents the time delay probability distribution of the
140: progenitors with a normalization constant $C$. In the context of
141: NS-NS or NS-BH mergers we adopt the standard power-law form, $P(\tau)
142: \propto\tau^n$. We note that NS-NS binaries in the Milky Way appear
143: to follow $P(\tau)\propto \tau^{-1}$ \citep{clm+04}. It is thus the
144: convolution of $P(\tau)$ with ${\rm SFH(\tau)}$ that determines the
145: relative rate of short GRBs in cluster and field early-type galaxies.
146:
147: Since the star formation history of early-type galaxies is determined by
148: both the mass and environment of a galaxy, the total star formation
149: history of each environment can be described in the following manner:
150: \begin{equation}
151: {\rm SFH}_i(\tau)=\int^{M_u}_{M_l} \phi_i(M)\, {\rm SFH}_{{\rm
152: gal},i}(\tau,M)\, dM,
153: \label{eqn:sfh}
154: \end{equation}
155: where $\phi_i(M)$ is the galaxy number density function, ${\rm
156: SFH}_{{\rm gal},i}(\tau,M)$ is the star formation history function of
157: a single galaxy that has a stellar mass, $M$, and $M_u$ and $M_l$ are
158: the appropriate upper and lower mass integration limits (see
159: \S\ref{sec:apply}).
160:
161: From SDSS observations, it has been determined that the galaxy number
162: density is well described by a double Schechter function
163: \citep{bbb+06}:
164: \begin{equation}
165: \phi(M)dM = e^{-M/M^*}[\phi^*_1 (M/M^*)^{\alpha_1}+\phi^*_2
166: (M/M^*)^{\alpha_2}]\frac{dM}{M^*},
167: \label{eqn:phi}
168: \end{equation}
169: where the parameters ${\rm log}\, M^*$, $\phi^*_1$, $\alpha_1$, $\phi^*_2$ and
170: $\alpha_2$ for a cluster environment (${\rm log}\,\Sigma\sim 1.3$) are
171: 11.06, 0.74, $-1.09$, 0.07, and $-1.5$, and for a field environment
172: (${\rm log}\,\Sigma\sim -0.9$) they are 10.44, 2.7, $-0.2$, 0.8, and
173: $-1.5$ \citep{bbb+06}. Here, $\Sigma$ is the projected density of
174: neighboring galaxies, and the specific values are determined from a
175: sample of $\sim 1.5\times 10^5$ galaxies in the redshift range
176: $0.01-0.085$ in the SDSS Data Release Four \citep{bbb+06}.
177:
178: We are here only interested in the rates of short GRBs in early-type
179: galaxies, whose star formation history is better understood than those
180: of late-type galaxies, and in order to avoid any systematic
181: differences between progenitors in late- and early-type host galaxies.
182: We therefore need to modify $\phi(M)$ by the fraction of early-type
183: galaxies in each environment \citep{bbb+06}:
184: \begin{equation}
185: f_{r,i}(\Sigma,M)=1-{\rm exp}\{-[(\Sigma/b_1)^{b_2}+(M/b_3)^{b_4}]\},
186: \label{eqn:fri}
187: \end{equation}
188: where the values of the parameters $b_1$, $b_2$, $b_3$, and $b_4$ are
189: $10^{0.91}$ Mpc$^{-2}$ , 0.69, $10^{10.72}$ M$_\odot$, and $0.59$.
190: Thus, $\phi_{i}(M)=\phi(M)f_{r,i}(\Sigma,M)$. In a cluster
191: environment, with a higher $\Sigma$ and systematically larger masses,
192: the early-type fraction is larger than in the field. A plot of
193: $f_r\phi(M)$ for cluster and field environments is shown in
194: Figure~\ref{fig:panels}a. Clearly, the most massive galaxies reside
195: preferentially in clusters, while the bulk of the mass in galaxies
196: with $M\lesssim 10^{10.5}$ is in the field.
197:
198: Finally, we turn to the star formation history function, ${\rm
199: SFH_{gal}}(\tau,M)$. Early-type galaxies in different environments
200: show different star formation histories for a given galaxy mass
201: \citep{ksc+02,tmb+05,dsw+06,schawinski06}. Moreover, the star formation history is
202: also determined by the galaxy mass. Following \citet{tmb+05}, we use
203: a Gaussian form for the star formation history:
204: \begin{equation}
205: {\rm SFH_{gal}}(\tau,M)=\frac{M}{\sqrt{2\pi}\Delta t}{\rm exp}
206: [-\frac{(\tau-t_{\rm peak})^2}{2(\Delta t)^2}],
207: \label{eqn:sfhgal}
208: \end{equation}
209: where the peak of the star formation history function, $t_{\rm peak}$,
210: is determined by both the mass of a galaxy and its environment
211: (Equations 2 and 3 of \citealt{tmb+05}), and $\Delta t$ is determined
212: by the mass of a galaxy. Here, we assume that the low-density
213: environment of \citet{tmb+05} corresponds to the field, while clusters
214: correspond to the high-density
215: environment\footnotemark\footnotetext{Following \citet{tmb+05} we
216: adopt the following cosmological parameters: $\Omega_m=0.3$,
217: $\Omega_\Lambda=0.7$, and $H_0=75$ km s$^{-1}$ Mpc$^{-1}$.}. The
218: overall trend is that less massive galaxies form their stars later
219: (i.e., smaller $t_{\rm peak}$) and over a wider timescale (i.e.,
220: larger $\Delta t$). In addition, $t_{\rm peak}$ in clusters is
221: systematically larger than in the field. These effects are shown in
222: Figure~\ref{fig:panels}b, where we plot the star formation histories
223: of galaxies with $10^8$, $10^9$, $10^{10}$, and $10^{11}$ M$_\odot$ in
224: both environments.
225:
226: The final ingredient is the overall normalization of
227: Equation~\ref{eqn:phi}. In its current form this equation was
228: normalized to a total mass of $10^{10}$ M$_\odot$ for each environment
229: by \citet{bbb+06}. Since we are interested in the overall rate per
230: unit volume, we therefore need to know what fraction of the stellar
231: mass is in clusters versus the field at $z\sim 0$. From the study of
232: \citet{bbb+06}, as well as \citet{fhp98} and \citet{ebc+05}, it
233: appears that about $20\%$ of the total stellar mass is included in
234: cluster environments\footnotemark\footnotetext{This corresponds to the
235: assumption that cluster environments can be described by ${\rm log}\,
236: \Sigma\gtrsim 0.5$; see Appendix A of \citet{bbb+06} for details.},
237: which we adopt here.
238:
239:
240: \section{Constraints on the Age Distribution of Short GRB Progenitors}
241: \label{sec:apply}
242:
243: To illustrate the combined effect of the trends discussed in the
244: previous section we begin by making the simplified assumption that
245: there is a single typical galaxy mass, $M_{\rm typ}$, for each
246: environment. This typical mass in turn determines the typical star
247: formation history of each environment. In this scenario,
248: Equation~\ref{eqn:sfh} can be simplified as:
249: \begin{equation}
250: {\rm SFH}_i(\tau)=\phi_i(M_{\rm typ})\times{\rm SFH}_{{\rm
251: gal},i}(\tau,M_{\rm typ}),
252: \label{eqn:sfhsimple}
253: \end{equation}
254: while $M_{\rm typ}$ is determined by \citep{bbb+06}:
255: \begin{equation}
256: {\rm log}\,(M_{\rm typ})=10.73+0.15{\rm log}\,\Sigma.
257: \end{equation}
258: Therefore, $M_{\rm typ}$ is $\sim 10^{10.6}$ M$_\odot$ and $\sim
259: 10^{10.9}$ M$_\odot$ for field and cluster environments, respectively.
260:
261: We are now in a position to use Equations~\ref{eqn:rate},
262: \ref{eqn:phi}, \ref{eqn:fri}, \ref{eqn:sfhgal}, and
263: \ref{eqn:sfhsimple} to predict the relative rate of short GRBs in
264: cluster and field early-type galaxies. The result as a function of
265: the power law index $n$ is shown in Figure~\ref{fig:simple}. We note
266: that the absolute scale in this plot is irrelevant since we did not
267: integrate over the full mass function, but it provides insight into
268: the overall trend. Namely, as $n$ increases, i.e., as the
269: distribution is more heavily weighted to older progenitors, the
270: fraction of short GRBs in clusters increases. This can be understood
271: as the combined effect of a systematically earlier star formation
272: episode and a higher typical mass in clusters.
273:
274: A quantitative determination of the relative rates requires a full
275: integration of Equation~\ref{eqn:sfh}. The results of this
276: integration are shown in Figure~\ref{fig:full}. Overall, the
277: dependence of the ratio of short GRBs in clusters and the field on $n$
278: is similar to the one found in the simple case. However, the
279: integration over the full mass function of each environment brings out
280: additional trends. For a low mass cutoff smaller than $10^{10}$
281: M$_\odot$, there is sharp downturn in the ratio for $n\lesssim -1$.
282: This can be understood from the plots of ${\rm SFH}_{\rm
283: gal}(\tau)\times P(\tau)$ shown in Figures~\ref{fig:panels}c and
284: \ref{fig:panels}d. In particular, for $n=-2$ the short GRB rate in
285: the field is larger than in clusters since the field has a higher
286: abundance of low mass ($\lesssim 10^{10}$ M$_\odot$) early-type
287: galaxies (Figure~\ref{fig:panels}a), which have young or intermediate
288: stellar populations. For such low values of $n$ the ratio is
289: sensitive primarily to recent star formation. We note that the same
290: effect is seen in the analysis of \citet{zr06} for the ratio of early-
291: to late-type hosts.
292:
293: In this context ($n\lesssim -1$) the lower mass cutoff in
294: Equation~\ref{eqn:sfh} plays an important role. In
295: Figure~\ref{fig:full} we show the effect of setting $M_l=10^8$,
296: $10^9$, and $10^{10}$ M$_\odot$. As noted above, for $M_l=10^{10}$
297: M$_\odot$ we do not see an obvious downturn because galaxies above
298: this limit do not exhibit any obvious recent star formation activity.
299: For the lower values of $M_l$, we find that the largest downturn is
300: for $M_l=10^{9}$ M$_\odot$. The reason for this is evident in
301: Figure~\ref{fig:panels}a, which shows that the largest difference
302: between the field and cluster mass functions is at $M\gtrsim 10^9$
303: M$_\odot$. At lower masses the mass functions converge, leading to a
304: ratio that is $\sim 1$.
305:
306: For $n\gtrsim -1$, on the other hand, the rate is dominated by the
307: oldest, and hence most massive galaxies (Figure~\ref{fig:panels}d).
308: In this case the ratio does not depend on $M_l$, and the predominance
309: of massive galaxies in clusters, along with the systematically earlier
310: star formation episodes, results in an increased fraction of short
311: GRBs in clusters for larger $n$. In fact, for $n=2$, we find that
312: there are three times as many short GRBs in cluster early-type
313: galaxies as there are in field early-type galaxies.
314:
315:
316: \section{Discussion}
317: \label{sec:disc}
318:
319: We now turn to a comparison of our model with observation of short
320: GRBs. To date two short bursts have been localized to clusters at
321: $z\sim 0.2$: GRB\,050509b \citep{gso+05,bpp+06} and GRB\,050911
322: (Paper1). We note that in the latter case the large error
323: circle prevents an association with a specific cluster galaxy, but the
324: large early-type fraction of $80\%$ \citep{bsm+06} suggests that the
325: burst was likely hosted by an early-type galaxy. On the other hand,
326: only one short GRB has been localized to a field early-type galaxy,
327: GRB\,050724 \citep{bpc+05}. We do not consider GRB\,050813, which was
328: hosted by a cluster at a much higher redshift ($z\sim 1.8$;
329: \citealt{ber06}), and GRB\,060502b, which may be hosted by an
330: early-type galaxy \citep{bpc+06}, but whose large-scale environment
331: has not been fully explored yet\footnotemark\footnotetext{The limit on
332: the X-ray luminosity at the redshift of the putative host galaxy,
333: $L_X\lesssim 6\times 10^{42}$ erg s$^{-1}$, is lower by a factor of
334: eight than that of the cluster hosting GRB\,050509b, but it is
335: somewhat higher than that of the cluster hosting GRB\,050911
336: (Paper I).}. Thus, the current ratio of short GRBs in cluster
337: versus field early-type galaxies is about $2\!:\!1$, with a large
338: uncertainty due to the small number of events. From
339: Figure~\ref{fig:full}, we find that this ratio corresponds to $n\sim
340: 0-1$. This value is lower than $n\gtrsim 3/2$ claimed by \citet{zr06},
341: but is in rough agreement with $-1 \lesssim n \lesssim 0$ found by \citet{bfp+06} based
342: on their revised redshift distribution with $1/4-2/3$ of all short GRBs at
343: $z\gtrsim 1$.
344:
345:
346: Clearly, in both methods of estimating the age of short GRB
347: progenitors the uncertainty in the inferred value of $n$ is currently
348: dominated by the small number of bursts with a known redshift, host
349: galaxy, and large-scale environment type. Since this uncertainty will
350: eventually diminish with a larger sample of events, it is interesting
351: to consider systematic uncertainties in both theoretical approaches.
352: Our analysis suffers from the somewhat poor definition of cluster and
353: field environments. We have used an overall cluster mass fraction of
354: $20\%$, as indicated by several researchers, but this number may range
355: from $10$ to $30\%$. Second, we have used the simplified bimodal star
356: formation history model of \citet{tmb+05}, but these authors do not
357: use the same quantitative definition of galaxy environment that was
358: used for the mass functions by \citet{bbb+06}. Since we have used
359: representative mass functions, and then scaled the results by the
360: overall mass fraction in clusters and the field, this effect should
361: not be significant. Third, the uncertainty in the definition of field
362: galaxy environment leads to an overall uncertainty of about $20\%$ in
363: our calculated ratio. Finally, since for $n\lesssim -1$ the ratio
364: depends on $M_l$, it is essential to understand the appropriate low
365: mass limit for early-type galaxies. With the current inferred value
366: of $n\sim 0-1$, however, this may not be a relevant issue for short
367: GRBs.
368:
369: Similarly, the analysis of the short GRB rate in early- and late-type
370: galaxies performed by \citet{zr06} also suffers from systematic
371: effects. First, the ratio is affected by the uncertain star formation
372: history at high redshift when early-type galaxies formed most of their
373: stars. Second, their estimation of star formation rate does not
374: consider environmental effects that appear to be important in current
375: observations from SDSS and 2dF, and which we have accounted for here.
376: Finally, as noted by \citet{zr06}, it is possible that late-type
377: galaxies have an altogether different age distribution of short GRB
378: progenitors than early-type galaxies. This problem is overcome by our
379: method since it considers only early-type host galaxies. We note that
380: if both derivations are in fact correct, then the estimated values of
381: $n$ can be used to assess any systematic differences of short GRB
382: progenitors in early- and late-type galaxies, as suspected to exist
383: for type Ia supernovae \citep{slp+06}.
384:
385: Future applications of our approach will include the effect of
386: globular clusters, which are thought to provide an efficient
387: environment for the production of NS-NS binaries, and may account for
388: a substantial fraction of all short GRB progenitors \citep{gpm06,hopman+06}. We
389: expect that since the specific frequency of globular clusters
390: increases significantly with galaxy mass \citep{har91}, an association
391: with globular clusters will increase the fraction of short GRBs in
392: galaxy clusters compared to the trend shown in Figure~\ref{fig:full}.
393: Similarly, our approach can be extended to higher redshift to
394: investigate the evolution in the fraction of short GRBs in clusters.
395: We expect that the lack of strong evolution in the last several Gyr
396: likely makes our analysis applicable out to $z\sim 1$. However, if
397: some short GRBs are in fact associated with clusters at $z\sim 2$,
398: this presents an opportunity to assess any systematic changes in the
399: value of $n$ with redshift.
400:
401: We end with the following conclusion. If our current estimate of
402: $n\gtrsim 0$ continues to be supported by future observations, then
403: this implies that the majority of short GRBs in early-type galaxies
404: will occur in clusters. This therefore suggests that short GRBs can
405: provide an efficient tool for finding forming galaxy clusters at high
406: redshift, as already appears to be the case for GRB\,050813
407: \citep{ber06}. Continued near-IR imaging and optical spectroscopic
408: observations of short GRB fields may therefore provide an efficient
409: method for finding the highest redshift clusters.
410:
411:
412:
413: \acknowledgements
414: We thank I.~K.~Baldry, Y.-T.~Lin, M.~Brown,
415: A.~Dressler, and J.~Mulchaey for helpful discussions.
416: M.-S.~S.~acknowledges support from the Observatories of the Carnegie
417: Institution of Washington.
418: E.B.~is supported by NASA through Hubble Fellowship grant
419: HST-01171.01 awarded by STSCI, which is operated by AURA, Inc., for
420: NASA under contract NAS5-26555.
421:
422: \begin{thebibliography}{}
423:
424: \bibitem[\protect\citeauthoryear{{Aloy}, {Janka}, \& {M{\"u}ller}}{{Aloy}
425: et~al.}{2005}]{ajm05}
426: {Aloy}, M.~A., {Janka}, H.-T., \& {M{\"u}ller}, E. 2005, \aap, 436, 273
427:
428: \bibitem[\protect\citeauthoryear{{Baldry} et~al.}{{Baldry}
429: et~al.}{2006}]{bbb+06}
430: {Baldry}, I., {Balogh}, M., {Bower}, R., {Glazebrook}, K., {Nichol}, R.,
431: {Bamford}, S., \& {Budavari}, T. 2006, ApJ, in press
432:
433: \bibitem[\protect\citeauthoryear{{Berger} et~al.}{{Berger}
434: et~al.}{2005}]{bpc+05}
435: {Berger}, E., et~al. 2005, \nat, 438, 988
436:
437: \bibitem[\protect\citeauthoryear{{Berger}}{{Berger}}{2006a}]{ber06}
438: {Berger}, E. 2006a, in AIP Conf. Proc. 838: Gamma-Ray Bursts in the Swift Era,
439: ed. S.~S. {Holt}, N.~{Gehrels}, \& J.~A. {Nousek}, 33
440:
441: \bibitem[\protect\citeauthoryear{{Berger} et~al.}{{Berger}
442: et~al.}{2006b}]{bsm+06}
443: {Berger}, E., {Shin}, M.~S., {Mulchaey}, J.~S., \& {Jeltema}, T.~E. 2006b,
444: preprint (astro-ph/0608498)
445:
446: \bibitem[\protect\citeauthoryear{{Berger} et~al.}{{Berger}
447: et~al.}{2006c}]{bfp+06}
448: {Berger}, E., et~al. 2006c, preprint (astro-ph/0611128)
449:
450: \bibitem[\protect\citeauthoryear{{Bernardi} et~al.}{{Bernardi}
451: et~al.}{1998}]{brc+98}
452: {Bernardi}, M., {Renzini}, A., {da Costa}, L.~N., {Wegner}, G., {Alonso},
453: M.~V., {Pellegrini}, P.~S., {Rit{\'e}}, C., \& {Willmer}, C.~N.~A. 1998,
454: \apjl, 508, L143
455:
456: \bibitem[\protect\citeauthoryear{{Bloom} et~al.}{{Bloom}
457: et~al.}{2006b}]{bpc+06}
458: {Bloom}, J.~S., et~al. 2006b, ApJ, in press
459:
460: \bibitem[\protect\citeauthoryear{{Bloom} et~al.}{{Bloom}
461: et~al.}{2006a}]{bpp+06}
462: {Bloom}, J.~S., et~al. 2006a, \apj, 638, 354
463:
464: \bibitem[\protect\citeauthoryear{{Burrows} et~al.}{{Burrows}
465: et~al.}{2006}]{bgc+06}
466: {Burrows}, D.~N., et~al. 2006, ApJ, in press
467:
468: \bibitem[\protect\citeauthoryear{{Champion} et~al.}{{Champion}
469: et~al.}{2004}]{clm+04}
470: {Champion}, D.~J., {Lorimer}, D.~R., {McLaughlin}, M.~A., {Cordes}, J.~M.,
471: {Arzoumanian}, Z., {Weisberg}, J.~M., \& {Taylor}, J.~H. 2004, \mnras, 350,
472: L61
473:
474: \bibitem[\protect\citeauthoryear{{Croton} et~al.}{{Croton}
475: et~al.}{2005}]{cfn+05}
476: {Croton}, D.~J., et~al. 2005, \mnras, 356, 1155
477:
478: \bibitem[\protect\citeauthoryear{{De Lucia} et~al.}{{De Lucia}
479: et~al.}{2006}]{dsw+06}
480: {De Lucia}, G., {Springel}, V., {White}, S.~D.~M., {Croton}, D., \&
481: {Kauffmann}, G. 2006, \mnras, 366, 499
482:
483: \bibitem[\protect\citeauthoryear{{Dressler}}{{Dressler}}{1980}]{dre80}
484: {Dressler}, A. 1980, \apj, 236, 351
485:
486: \bibitem[\protect\citeauthoryear{{Eichler} et~al.}{{Eichler}
487: et~al.}{1989}]{elp+89}
488: {Eichler}, D., {Livio}, M., {Piran}, T., \& {Schramm}, D.~N. 1989, \nat, 340,
489: 126
490:
491: \bibitem[\protect\citeauthoryear{{Eke} et~al.}{{Eke} et~al.}{2005}]{ebc+05}
492: {Eke}, V.~R., {Baugh}, C.~M., {Cole}, S., {Frenk}, C.~S., {King}, H.~M., \&
493: {Peacock}, J.~A. 2005, \mnras, 362, 1233
494:
495: \bibitem[\protect\citeauthoryear{{Fox} et~al.}{{Fox} et~al.}{2005}]{ffp+05}
496: {Fox}, D.~B., et~al. 2005, \nat, 437, 845
497:
498: \bibitem[\protect\citeauthoryear{{Fukugita}, {Hogan}, \& {Peebles}}{{Fukugita}
499: et~al.}{1998}]{fhp98}
500: {Fukugita}, M., {Hogan}, C.~J., \& {Peebles}, P.~J.~E. 1998, \apj, 503, 518
501:
502: \bibitem[\protect\citeauthoryear{{Gal-Yam} et~al.}{{Gal-Yam}
503: et~al.}{2005}]{gno+05}
504: {Gal-Yam}, A., et~al. 2005, preprint (astro-ph/0509891)
505:
506: \bibitem[\protect\citeauthoryear{{Gehrels} et~al.}{{Gehrels}
507: et~al.}{2005}]{gso+05}
508: {Gehrels}, N., et~al. 2005, \nat, 437, 851
509:
510: \bibitem[\protect\citeauthoryear{{Grindlay}, {Portegies Zwart}, \&
511: {McMillan}}{{Grindlay} et~al.}{2006}]{gpm06}
512: {Grindlay}, J., {Portegies Zwart}, S., \& {McMillan}, S. 2006, Nature Physics,
513: 2, 116
514:
515: \bibitem[\protect\citeauthoryear{{Guetta} \& {Piran}}{{Guetta} \&
516: {Piran}}{2006}]{gp06}
517: {Guetta}, D., \& {Piran}, T. 2006, \aap, 453, 823
518:
519: \bibitem[\protect\citeauthoryear{{Harris}}{{Harris}}{1991}]{har91}
520: {Harris}, W.~E. 1991, \araa, 29, 543
521:
522: \bibitem[\protect\citeauthoryear{{Hjorth} et~al.}{{Hjorth}
523: et~al.}{2003}]{hsm+03}
524: {Hjorth}, J., et~al. 2003, \nat, 423, 847
525:
526: \bibitem[\protect\citeauthoryear{{Hjorth} et~al.}{{Hjorth}
527: et~al.}{2005}]{hwf+05}
528: {Hjorth}, J., et~al. 2005, \nat, 437, 859
529:
530: \bibitem[\protect\citeauthoryear{{Hopman} et~al.}{{Hopman}
531: et~al.}{2006}]{hopman+06} {Hopman}, C., et~al. 2006, \apjl, 643, L91
532:
533: \bibitem[\protect\citeauthoryear{{Kouveliotou} et~al.}{{Kouveliotou}
534: et~al.}{1993}]{kmf+93} {Kouveliotou}, C., {Meegan}, C.~A., {Fishman}, G.~J., {Bhat}, N.~P., {Briggs},
535: M.~S., {Koshut}, T.~M., {Paciesas}, W.~S., \& {Pendleton}, G.~N. 1993,
536: \apjl, 413, L101
537:
538: \bibitem[\protect\citeauthoryear{{Kulkarni}}{{Kulkarni}}{2005}]{kul05}
539: {Kulkarni}, S.~R. 2005, preprint (astro-ph/0510256)
540:
541: \bibitem[\protect\citeauthoryear{{Kuntschner} et~al.}{{Kuntschner}
542: et~al.}{2002}]{ksc+02}
543: {Kuntschner}, H., {Smith}, R.~J., {Colless}, M., {Davies}, R.~L., {Kaldare},
544: R., \& {Vazdekis}, A. 2002, \mnras, 337, 172
545:
546: \bibitem[\protect\citeauthoryear{{Li} \& {Paczy{\'n}ski}}{{Li} \&
547: {Paczy{\'n}ski}}{1998}]{lp98}
548: {Li}, L.-X., \& {Paczy{\'n}ski}, B. 1998, \apjl, 507, L59
549:
550: \bibitem[\protect\citeauthoryear{{Nakar}, {Gal-Yam}, \& {Fox}}{{Nakar}
551: et~al.}{2005}]{ngf05}
552: {Nakar}, E., {Gal-Yam}, A., \& {Fox}, D.~B. 2005, preprint (astro-ph/0511254)
553:
554: \bibitem[\protect\citeauthoryear{{Narayan}, {Paczynski}, \& {Piran}}{{Narayan}
555: et~al.}{1992}]{npp92}
556: {Narayan}, R., {Paczynski}, B., \& {Piran}, T. 1992, \apjl, 395, L83
557:
558: \bibitem[\protect\citeauthoryear{{Rosswog} \& {Ramirez-Ruiz}}{{Rosswog} \&
559: {Ramirez-Ruiz}}{2002}]{rr02}
560: {Rosswog}, S., \& {Ramirez-Ruiz}, E. 2002, \mnras, 336, L7
561:
562: \bibitem[\protect\citeauthoryear{{Schawinski} et~al.}{{Schawinski}
563: et~al.}{2006}]{schawinski06}
564: {Schawinski}, K., et~al. 2006, preprint (astro-ph/0601036)
565:
566: \bibitem[\protect\citeauthoryear{{Soderberg} et~al.}{{Soderberg}
567: et~al.}{2006}]{sbk+06}
568: {Soderberg}, A.~M., et~al. 2006, preprint (astro-ph/0601455)
569:
570: \bibitem[\protect\citeauthoryear{{Stanek} et~al.}{{Stanek}
571: et~al.}{2003}]{smg+03}
572: {Stanek}, K.~Z., et~al. 2003, \apjl, 591, L17
573:
574: \bibitem[\protect\citeauthoryear{{Sullivan} et~al.}{{Sullivan}
575: et~al.}{2006}]{slp+06}
576: {Sullivan}, M., et~al. 2006, preprint (astro-ph/0605455)
577:
578: \bibitem[\protect\citeauthoryear{{Thomas} et~al.}{{Thomas}
579: et~al.}{2005}]{tmb+05}
580: {Thomas}, D., {Maraston}, C., {Bender}, R., \& {Mendes de Oliveira}, C. 2005,
581: \apj, 621, 673
582:
583: \bibitem[\protect\citeauthoryear{{Whitmore}, {Gilmore}, \& {Jones}}{{Whitmore}
584: et~al.}{1993}]{wgj93}
585: {Whitmore}, B.~C., {Gilmore}, D.~M., \& {Jones}, C. 1993, \apj, 407, 489
586:
587: \bibitem[\protect\citeauthoryear{{Zheng} \& {Ramirez-Ruiz}}{{Zheng} \&
588: {Ramirez-Ruiz}}{2006}]{zr06}
589: {Zheng}, Z., \& {Ramirez-Ruiz}, E. 2006, preprint (astro-ph/0601622)
590:
591: \end{thebibliography}
592:
593: \begin{figure}
594: \plotone{f1.eps}
595: \caption{A summary of the model ingredients that determine the
596: relative rate of short GRBs in cluster and field early-type galaxies.
597: Panel (a) shows the mass function in each environment (solid line:
598: cluster; dashed line: field), including the different early-type
599: fractions (Equation~\ref{eqn:fri}), and an overall cluster mass
600: fraction of $20\%$ (\S\ref{sec:theory}). Panel (b) shows the star
601: formation history as a function of environment and galaxy mass (red:
602: $10^{11}$ M$_\odot$; purple: $10^{10}$ M$_\odot$; green: $10^9$
603: M$_\odot$; blue: $10^8$ M$_\odot$). Bottom panels show the product of
604: the star formation history with the short GRB progenitor age
605: distribution function for a power law index $n=-2$ (c), and $n=2$ (d).
606: Clearly, a distribution weighted to short merger timescales heavily
607: favors lower mass host galaxies (and hence the field), while a
608: distribution weighted to long merger timescales favors massive host
609: galaxies (and hence clusters).
610: \label{fig:panels}}
611: \end{figure}
612:
613: \begin{figure}
614: \plotone{f2.eps}
615: \caption{Ratio of the short GRB rate in cluster and field early-type
616: galaxies as a function of the age distribution power law index, $n$,
617: assuming that each environment is described by a typical galaxy mass
618: (\S\ref{sec:apply}). The scale on the ordinate is arbitrary since we
619: do not consider the full mass function, but the overall trend is
620: representative. Since the typical galaxy mass is higher in clusters
621: than in the field, the typical star formation epoch is earlier in
622: clusters. We therefore expect more short GRBs in cluster early-type
623: galaxies when $n$ is high.
624: \label{fig:simple}}
625: \end{figure}
626:
627: \begin{figure}
628: \plotone{f3.eps}
629: \caption{Ratio of the short GRB rate in cluster and field early-type
630: galaxies as a function of the age distribution power law index, $n$,
631: considering the full mass function in each environment. Different
632: colors represent the effect of different mass integration limits:
633: $10^8<M<10^{12}$ M$_\odot$ (black), $10^9<M<10^{12}$ M$_\odot$ (blue),
634: and $10^{10}<M<10^{12}$ M$_\odot$ (red). The current observed ratio,
635: based on only three events, is about 2, suggesting that $n\sim 0-1$.
636: \label{fig:full}}
637: \end{figure}
638:
639: \end{document}
640: