1: \documentclass[useAMS,usenatbib]{mn2e}
2: \usepackage{epsfig}
3: \usepackage{amssymb}
4:
5: % correct for funky A4 spacing
6: \voffset -0.5in
7:
8: \newcommand{\msun}{\mbox{M$_{\odot}$}}
9: \newcommand{\lsun}{\mbox{L$_{\odot}$}}
10: \newcommand{\kms}{\mbox{km s$^{-1}$}}
11:
12: % ---------------
13: \newcommand{\mnras}{MNRAS}
14: \newcommand{\aap}{A\&A}
15: \newcommand{\aaps}{A\&AS}
16: \newcommand{\aj}{AJ}
17: \newcommand{\apj}{ApJ}
18: \newcommand{\apjl}{ApJL}
19: \newcommand{\apjs}{ApJS}
20: \newcommand{\apss}{Ap\&SS}
21: \newcommand{\araa}{ARA\&A}
22: \newcommand{\ass}{A\&A}
23: \newcommand{\nat}{Nature}
24: \newcommand{\pasj}{PASJ}
25: \newcommand{\prd}{PhRvD}
26: \newcommand{\physrep}{Physics Reports}
27: \newcommand{\skytel}{S\&T}
28: % ---------------
29:
30:
31:
32: %-----------------------------
33: % Title, Abstract & Whosthat
34: %-----------------------------
35: \title[The Collision Between The Milky Way And Andromeda]
36: {The Collision Between The Milky Way And Andromeda}
37:
38: \author[Cox \& Loeb]
39: {T. J. Cox\footnotemark[1] and Abraham Loeb\footnotemark[2] \\
40: Harvard-Smithsonian Center for Astrophysics,
41: 60 Garden Street, Cambridge, MA 02138, USA \\
42: }
43:
44:
45: \begin{document}
46:
47:
48: \maketitle
49:
50:
51: \begin{abstract}
52:
53: We use a N--body/hydrodynamic simulation to forecast the future encounter
54: between the Milky Way and the Andromeda galaxies, given current
55: observational constraints on their relative distance, relative velocity,
56: and masses. Allowing for a comparable amount of diffuse mass to fill the
57: volume of the Local Group, we find that the two galaxies are likely to
58: collide in a few billion years - within the Sun's lifetime. During the the
59: interaction, there is a chance that the Sun will be pulled away from its
60: present orbital radius and reside in an extended tidal tail. The
61: likelihood for this outcome increases as the merger progresses, and there
62: is a remote possibility that our Sun will be more tightly bound to
63: Andromeda than to the Milky Way before the final merger. Eventually, after
64: the merger has completed, the Sun is most likely to be scattered to the
65: outer halo and reside at much larger radii ($>30$~kpc). The density
66: profiles of the stars, gas and dark matter in the merger product resemble
67: those of elliptical galaxies. Our Local Group model therefore provides a
68: prototype progenitor of late--forming elliptical galaxies.
69:
70: \end{abstract}
71:
72:
73: \begin{keywords}
74: galaxy:evolution --- galaxies:evolution --- galaxies:formation ---
75: galaxies:interactions --- Local Group --- methods:N-body simulations.
76: \end{keywords}
77:
78:
79: \renewcommand{\thefootnote}{\fnsymbol{footnote}}
80: \footnotetext[1]{tcox@cfa.harvard.edu}
81: \footnotetext[2]{loeb@cfa.harvard.edu}
82:
83:
84: %---------------------
85: %---------------------
86: %---------------------
87: \section{Introduction}
88: \label{sec:intro}
89:
90:
91: It is well known that the Milky Way (MW) and Andromeda (M31) are the two
92: largest members of the Local Group of galaxies. Together with their
93: $\sim40$ smaller companions, the Milky Way and Andromeda comprise our
94: galactic neighborhood, and as such, represent the nearest laboratory,
95: and therefore the most powerful tool, to study the formation and
96: evolution of galactic structure.
97:
98: Like most extragalactic groups, the Local Group is very likely to be
99: decoupled from the cosmological expansion and is now a gravitationally
100: bound collection of galaxies. This notion is supported by the observed
101: relative motion between its two largest galaxies; namely, the Milky Way and
102: Andromeda are moving toward each other at $\sim120$~\kms \citep{BT}.
103: Unfortunately, this motion alone does not indicate whether the Local Group
104: is bound or not. The unknown magnitude of Andromeda's transverse velocity
105: adds uncertainty into the present day orbital parameters and therefore the
106: past and future evolution of the Local Group.
107:
108: Barring the uncertain transverse velocity of Andromeda, a considerable
109: amount of information can be inferred about the Local Group provided a
110: plausible set of assumptions. Nearly 50 years ago \citet{KW59} pioneered
111: the ``timing argument,'' in which the Milky Way and Andromeda are assumed
112: to form within close proximity to each other, during the dense early stages
113: of the Universe, before they were pulled apart by the general cosmological
114: expansion. They have subsequently reversed their path and are approaching
115: one another owing to their mutual gravitational attraction. According to
116: the timing argument, the Milky Way and Andromeda have now traced out nearly
117: a full period of their orbital motion which is governed by Kepler's laws.
118: By assuming that the system has no angular momentum, and given the current
119: separation, velocity of approach, and the age of the Universe, the timing
120: argument yields estimates for the mass of the Local Group
121: ($>3\times10^{12}$\msun), the semi--major axis of the orbit
122: ($<580$~kpc), and the time of the next close passage ($>4$~Gyr)
123: \citep[see Sec. 10.2 of ][]{BT}.
124:
125: While the seminal results of \citet{KW59} were an early indication of the
126: large mass--to--light ratio in the Local Group and therefore the presence
127: of dark matter, they also began a nearly five decade long quest to understand
128: the past, present, and future of our Local Group. In particular, a number of
129: studies have extended the original timing argument by allowing for
130: various angular momenta, by including more realistic or time--dependent
131: mass distributions, by adding the effects of mass at scales beyond that
132: of the Local Group, or testing its validity using numerical simulations
133: \citep[see, e.g.,][]{Peeb89,FT91,Val93,P94,P01,SF05,Loeb05,LW07,vdM07}.
134:
135: One of the most intriguing developments stemming from the various studies
136: of the Local Group is an estimate of the transverse velocity of Andromeda.
137: By employing the action principle to the motions of galaxies within and
138: near ($<20$~Mpc) the Local Group, \citet{P01} concluded that the transverse
139: velocity of Andromeda is less than $200$~\kms. Using the well measured
140: transverse velocity of M33 \citep{Bru05} and numerical simulations that tracked
141: the potential tidal disruption during M33's past encounters with Andromeda,
142: \citet{Loeb05} found an even smaller estimate, $\sim 100$~\kms, for the
143: transverse velocity. While future astrometric observations using
144: SIM\footnote{http://planetquest.jpl.nasa.gov/SIM/} and
145: GAIA\footnote{http://www.sci.esa.int/gaia/}
146: will be able to accurately measure the proper motion of Andromeda, the
147: low values favored by these papers suggests that the Local Group is indeed
148: a gravitationally bound system.
149:
150: Provided that the Local Group is gravitationally bound, and that the Milky
151: Way and Andromeda are heading towards each other, one must admit the
152: possibility that they will eventually interact and merge. This outcome
153: appears inevitable given the massive halos of dark matter that likely
154: surround the Milky Way and Andromeda. Numerical experiments have robustly
155: concluded that dark matter halos can exert significant dynamical friction,
156: and are sponges that soak up energy and angular momentum leading to a rapid
157: merger \citep{B88}.
158:
159: Even though the eventual merger between the Milky Way and Andromeda is
160: common lore in Astronomy, the merger process has not been addressed by a
161: comprehensive numerical study. The one exception is a paper by
162: \citet{Dub96} that presented a viable model for the Local Group and
163: numerically simulated the eventual merger between the Milky Way and
164: Andromeda. However, \citet{Dub96} utilized this Local Group model and its
165: numerical evolution to study the production of tidal tails during
166: such an encounter and the possibility to use the structure of this tidal
167: material to probe the dark matter potential. While the study by
168: \citet{Dub96} provided the first enticing picture of the future encounter
169: between the Milky Way and Andromeda, \citep[for a more recent and higher
170: resolution version of this simulation, see][]{Dub06}, it was neither
171: designed to detail the merger dynamics including intergalactic material,
172: nor outline the possible outcomes for the dynamics of our Sun, nor quantify
173: properties of the merger remnant. In addition, the last decade has
174: produced a number of improved models for the structure of the Milky Way and
175: Andromeda as well as the properties of the intragroup medium.
176:
177: In this paper we quantitatively predict when the interaction and merger of
178: the Milky Way and Andromeda will likely occur and forecast the probable
179: dynamics of the Sun during this event. We achieve this goal by
180: constructing a model for the Local Group in \S\ref{sec:model} that
181: satisfies all observational constraints. We then evolve this model using a
182: self-consistent N-body/hydrodynamic simulation, as described in
183: \S\ref{sec:meths}. The generic properties of the merger, including the
184: merger timescale, the possible evolution of our Solar System, and
185: properties of the merger remnant, are outlined in \S\ref{sec:results}.
186: Finally, we conclude in \S\ref{sec:conc}.
187:
188:
189:
190:
191:
192: %-------------------------
193: %-------------------------
194: %-------------------------
195: \section{A Model of the Local Group}
196: \label{sec:model}
197:
198:
199: The distribution of mass within our Local Group of galaxies has
200: been a long--standing question in astrophysics. It is clear that
201: much of the matter is associated with the two largest galaxies
202: in the Local Group: the Milky Way and Andromeda. Moreover, these
203: two spiral galaxies are likely to be embedded in an ambient medium
204: of dark matter and gas.
205:
206: \subsection{The Milky Way and Andromeda}
207: \label{ssec:mwand}
208:
209: There are a number of different models for both the Milky Way and Andromeda
210: galaxies \citep[see, e.g.,][and reference therein] {KZS02,WD05,SBB07}. These
211: studies generally enlist a myriad of observational data to infer the
212: distribution of baryons, while the dark matter, which dominates the
213: gravitational potential, is set to match distributions extracted from
214: cosmological N-body simulations \citep[e.g.,][]{NFW96}. Together, these
215: models specify the total mass distribution out to the virial radius
216: ($\sim200-300$~kpc).
217:
218: In our model of the Local Group we start by adopting the models for the
219: Milky Way and Andromeda favored by \citet{KZS02}. Within these models, the
220: baryons are contained entirely within the rotationally supported
221: exponential disk and central bulge. These components are then surrounded
222: by a massive dark--matter halo, which has nearly 20 times the mass as the
223: baryons, as specified by the mass fractions, $m_b$ and $m_d$, defined as
224: the bulge and disk mass, respectively, divided by the total mass. The
225: exponential disk, of radial disk scale radius $R_d$, also contains a set
226: fraction $f$ of its mass in collisional gas that can cool and form stars.
227: Both the bulge and dark halo components are assumed to follow the
228: \citet{H90} profile. The bulge scale radius $a$ is fixed to be 20\% of the
229: radial disk scale radius $R_d$. The dark--matter profile is defined by its
230: concentration $c$, spin parameter $\lambda$, and total virial mass
231: $M_{200}$ and virial circular velocity $V_{200}$ (at the radius $r_{200}$
232: where the average interior density is 200 times the critical cosmic density
233: today, $rho_{\rm crit}= 10^{-29}~{\rm g~cm^{-3}}$), which are all listed in
234: Table~\ref{tab:galps}. The numerical construction of these models employs
235: methods commonly used to construct equilibrium disk galaxies \citep[see,
236: e.g.,][]{H93,SW99,Sp00,Cox06,SdMH05}.
237:
238:
239:
240: \begin{table}
241: \begin{center}
242: \caption{
243: Properties of the Milky Way (MW) and Andromeda (M31) models
244: used in the work (see text for definitions).
245: }
246: \begin{tabular}{lcc}
247: \hline
248: Property & MW & Andromeda \\
249: \hline
250: $V_{200}$ (\kms) & 145 & 170 \\
251: $M_{200}$ ($10^{12}$\msun) & 1.0 & 1.6 \\
252: $c$ & 12 & 12 \\
253: $\lambda$ & 0.031 & 0.036 \\
254: $m_d$ & 0.041 & 0.044 \\
255: $R_d$ & 2.2 & 3.6 \\
256: $f$ & 0.3 & 0.3 \\
257: $m_b$ & 0.008 & 0.012 \\
258: $a$ & 0.4 & 0.7 \\
259: N$_{dm}$ & 475,500 & 755,200 \\
260: N$_{disk}$ & 14,350 & 24,640 \\
261: N$_{gas}$ & 6,150 & 10,560 \\
262: N$_{bulge}$ & 4,000 & 9,600 \\
263: \hline
264: \end{tabular}
265: \label{tab:galps}
266: \end{center}
267: \end{table}
268:
269:
270:
271: \subsection{The Orbit}
272: \label{ssec:mwandorb}
273:
274: Given the adopted parameters of the two largest galaxies in the Local
275: Group, we must now define their orbital parameters and any ambient medium
276: in which the system will be embedded. There are a few empirical
277: constraints that must be considered. First, at the current epoch, the
278: separation between the Milky Way and Andromeda is 780~kpc
279: \citep{McC05,Rib05}. Second, the Milky Way and Andromeda are approaching
280: each other at a radial speed of 120~\kms, assuming a local circular speed
281: of 220~\kms \citep[see Sec. 10.2 of][]{BT}. These observational facts
282: tightly constrain any dynamical model of the Local Group since their
283: fractional error bars are estimated to be less than $\sim5$\%.
284:
285: Less well constrained is the current estimate for the transverse velocity
286: of Andromeda. As mentioned in \S\ref{sec:intro}, speeds of $<200$~\kms are
287: favored by recent models \citep{P01,Loeb05,vdM07}, but depend upon
288: assumptions regarding the distribution of mass within the Local Group and
289: its initial state. We will therefore gauge the success of our Local Group
290: model by its ability to reproduce the above three observations, but it
291: should be kept in mind that the first two observations have significantly
292: less leeway than the third.
293:
294: Constraints on the spin orientation of both the Milky Way and Andromeda
295: with respect to the orbital plane of the merger originate from the current
296: position and orientation of Andromeda in the night sky. While these
297: details are necessary for a complete model of the Local Group, they do not
298: significantly influence the timing of the merger between the Milky Way and
299: Andromeda. The spin orientation will, however, affect the disk morphology
300: during the merger, a detail that will be addressed when we attempt to track
301: the possible dynamical fate of our Sun in \S\ref{ssec:earth}.
302:
303:
304: \subsection{The Local Intragroup Medium}
305: \label{ssec:lgmed}
306:
307: One plausible starting model for the Local Group is to follow the logic
308: originally employed by the timing argument \citep{KW59}, i.e., that the
309: mass of the Local Group is entirely contained within the Milky Way and
310: Andromeda and their motion is a simple two body problem governed by
311: Kepler's equations. In practice, however, most of the recent
312: implementations of the timing argument \citep[see, e.g.,][] {BT,FT91,LW07}
313: generally yield masses for the Local Group ($>3\times10^{12}$~\msun) that
314: exceed the total masses in our Milky Way and Andromeda models
315: ($2.6\times10^{12}$~\msun). This discrepancy suggests that the Milky Way
316: and Andromeda do not contain the entire quantity of mass in the Local Group
317: and are instead the most massive concentrations of mass within a larger
318: all--encompassing medium. This point of view is very natural in a
319: cosmological context where galaxies are not isolated islands, but rather
320: mountain peaks within a vast continent of land.
321:
322:
323: \begin{figure*}
324: \begin{center}
325: \resizebox{14.0cm}{!}{\includegraphics{f1.eps}}\\
326: \caption{
327: Sketch of the initial configuration of our Local Group model, which
328: consists of the Milky Way and Andromeda embedded in a diffuse,
329: constant--density intragroup medium of equivalent mass.
330: See the text in \S\ref{sec:model} and Table 1 for more details.
331: \label{fig:lgmodel}}
332: \end{center}
333: \end{figure*}
334:
335:
336: For these reasons our Local Group model supplements the Milky Way and
337: Andromeda galaxy models with a diffuse and extended intragroup medium, as
338: schematically depicted in Figure~\ref{fig:lgmodel}. Put within a
339: cosmological framework this initial configuration may represent a point in
340: time when the entire Local Group has decoupled from the general Hubble flow
341: and is in the process of collapsing to become a virialized structure.
342:
343: For simplicity, we assume that the Local Group medium is initially a
344: constant density cube of 1.5 Mpc on a side composed of both dark matter and
345: gas. The total diffuse mass within this cube is set equal to the total
346: mass of the two galaxy halos today, $2.6\times 10^{12}$~\msun, yielding a
347: net mass interior to the MW/M31 orbit that is consistent with the timing
348: argument. Our choice is motivated by other cosmological simulations
349: \citep{Gao04a} which indicate that a substantial fraction of the dark
350: matter is likely to be diffuse and resides in between virialized halos
351: (within the unvirialized Local Group) at the initial time of our simulation
352: $\sim5$~Gyr ago.
353: \footnote{The initial scale of the Local group in our simulation is an
354: order of magnitude larger than the virial radius $r_{200}$ of the Milky Way
355: or Andromeda galaxies. If one were to extend the \citet{NFW97} density
356: profile of the envelope around each galaxy (with an asymptotic radial
357: dependence of $r^{-3}$) out to our initial Local Group scale, then one
358: would roughly double the mass found within the virial radius of each
359: halo. The added mass in that case would be similar to the amount we indeed
360: assume for the intragroup medium.}
361:
362: We also postulate that the Local Group
363: volume contains close to the cosmic mean value of baryons, namely 16\%
364: \citep{Sp03}. Since the Milky Way and Andromeda galaxy models only include
365: baryons in the galactic disk and bulge components, they are far short of
366: this value. We therefore set the Local Group medium to be 20\% primordial
367: gas, by mass, so that the entire region approaches the cosmic mean value.
368: The gaseous component, like the dark matter, is initialized to have a
369: constant density, and its temperature is fixed to be $3\times10^5$~K,
370: consistent with estimates for the warm--hot intergalactic medium at
371: comparable overdensities from cosmological simulations \citep[see, e.g.,
372: Fig.~6 in][]{Dav01} and with observations of apparent RAM pressure
373: stripping in the Local Group dwarf galaxy Pegasus \citep{McC07}.
374:
375: We note that the initial gas temperature is far below the virialized
376: temperature of the Local Group, however this assumption is consistent with
377: the post--turn around, and pre--virialized initial conditions that we
378: adopt. The constant density is also not in line with the assumed growth of
379: the initial stellar disks from a virialized gaseous halo \citet[see,
380: e.g.,][ and references therein]{MMW98}. Within a halo dynamical time, some
381: of the initial intragroup gas is expected be accreted by each galaxy, and
382: most of the remaining intragroup gas will end up being shock--heated during
383: the collapse and relaxation of the Local Group. The dynamics of the merger
384: between the Milky Way and Andromeda is dominated by the distribution of
385: dark matter and stellar disks, and so the basic results of our simulation,
386: involving for example the timing of the merger, are not directly affected
387: by the initial gas temperature and density distribution.
388:
389: Since the total mass of the diffuse medium is equivalent to the two
390: galaxies, the two components are represented with an equivalent number of
391: particles; 1.3 million (1.04 million dark matter, and 260,000 gas). With
392: these assumptions, the overdensity of our initial Local Group model is
393: $\delta_{\rm LG} = \rho_{\rm LG}/\rho_{\rm crit} \approx 10$.
394:
395:
396: \subsection{Other Considerations}
397: \label{ssec:other}
398:
399: Unfortunately, once the Local Group contains this extended mass
400: distribution the relative motion of the Milky Way and Andromeda no
401: longer becomes a trivial application of Kepler's laws. The diffuse
402: mass will steadily extract orbital energy and angular momentum owing
403: to dynamical friction, and the deep potential wells of the galaxies
404: will slowly accrete diffuse matter. Both of these effects act to
405: gradually extract energy from the binary system, hardening its
406: orbit, and accelerating the merging process.
407:
408: Owing to the complicated nature of the orbit, and its deviation from
409: simple two--body motion, we are left with a fair degree of ambiguity
410: regarding the initial state of our Local Group model. Because the
411: configuration envisaged by Figure~\ref{fig:lgmodel} resembles the cosmological
412: collapse of an overdense region of the Universe, a natural starting
413: point is when this fluctuation decoupled from the general cosmological
414: expansion, turned around, and began to collapse under its own self gravity.
415: Like the original timing argument, we then presume that this scenario also
416: occurred for the Milky Way and Andromeda as well, i.e., we initialize their
417: radial velocity to be zero as if they have just turned around and are now
418: about to begin their gravitational collapse.
419:
420: While some insight about the value of the turn around radius can be gained
421: from an application of the timing argument, in practice we find that the
422: position and velocity of the Milky Way and Andromeda quickly deviated from
423: simple two--body motion owing to the diffuse intragroup medium. In
424: general, the velocities quickly grew larger than the original orbit
425: specified and the direction became highly radial. We therefore adopt a
426: trial and error approach where we build an initial model, run it until the
427: Milky Way and Andromeda are separated by 780~kpc (as this is their current
428: separation), and inspect the relative velocity between the two galaxies to
429: assess the validity of the model.
430:
431: Employing this procedure results in a number of models that are all honest
432: representations of the Local Group at the present time. In practice, we
433: find a trade off between the initial separation and the initial orbital
434: energy such that models which start with a larger separation require a
435: smaller initial eccentricity. In all cases, the apparent orbit of the two
436: galaxies becomes (or was to begin with) nearly parabolic at the present
437: time, consistent with recent estimates of the Milky Way--Andromeda orbit
438: \citep[see, e.g.,][] {vdM07}. While the rest of this paper will primarily
439: present the results of one particular model, we will also show that all
440: models yield similar estimates for the eventual merger between the Milky
441: Way and Andromeda. We will argue in \S\ref{ssec:time} that this
442: convergence results naturally from our assumed intragroup medium.
443:
444: The model we choose to focus upon begins with an initial separation of 1.3
445: Mpc, and initializes the Milky Way and Andromeda on an eccentric orbit
446: $\epsilon=0.494$, with a distance at perigalacticon of 450~kpc. With this
447: orbit the initial angular velocity is 65~\kms, which could likely originate
448: from tidal torques \citep{GT78,RLB89}. This particular model begins with
449: the largest separation of all our models and therefore may be the best
450: representation of the evolution of the Local Group since its decoupling from
451: the universal expansion. Since this model tracks the Local Group the farthest
452: into the past, the intragroup medium also has a significant amount of time to
453: react to the two galaxies and therefore is likely to be the most insensitive
454: to its initial configuration.
455:
456:
457:
458:
459: %-------------------------
460: %-------------------------
461: %-------------------------
462: \section{Numerical Methods}
463: \label{sec:meths}
464:
465: To simulate the evolution of our Local Group and in particular the
466: interaction between the Milky Way and Andromeda use the publically
467: available N-body/hydrodynamic code {\small GADGET2} \citep{SpGad2}. This
468: version of the code employs the ``conservative--entropy'' formulation of
469: Smoothed Particle Hydrodynamics \citep[SPH,][]{SHEnt} that conserves both
470: energy and entropy \citep[unlike earlier versions of SPH; see
471: e.g.,][]{H93sph}, while improving shock-capturing. We assume that the gas
472: is of primordial composition, and include the effects of radiative cooling.
473:
474: Star formation and its associated feedback are both included in a
475: manner very similar to that described in \citet{Cox06}. As is
476: commonly assumed, stars are stochastically formed at a rate
477: determined by the SPH gas density \citep[see, e.g.][]{Kz92,SH03,SdMH05,
478: Cox06} with an efficiency set to match the observed correlation
479: between star formation and gas density \citep{Kenn98}.
480:
481: Feedback from stellar winds and supernovae is treated in a very simplistic
482: manner, namely the SPH particles that have sufficient density to form
483: stars are fixed to have an effective temperature of $10^5$ K. This
484: methodology is similar in principle to most of the currently favored models
485: for feedback \citep[see, e.g.,][]{Sp00,SH03, Stin06,Cox06}, and is easy to
486: implement. Since the focus of this work is the large--scale evolution of
487: the Local Group and the generic dynamics of the collision between the Milky
488: Way and Andromeda, the detailed treatment of the inter--stellar medium does
489: not influence our primary conclusions.
490:
491: Numerical resolution is a significant consideration for any computational
492: problem. For our purposes here, we require sufficient resolution to
493: reliably follow the interaction and merger of the Milky Way and Andromeda,
494: while maintaining the ability to perform a number of simulations with the
495: available computational resources. These considerations motivated the
496: particle number choices outlined in \S\ref{sec:model}. Given the large
497: number of components in these simulations (stellar disks, dark halos, and
498: intragroup medium) and the desire to reduce two--body effects, we required
499: all particles to have an identical mass of $2\times 10^{7}$\msun and we
500: employed a universal gravitational softening length of 150~pc. To test the
501: sensitivity of the results to these parameter choices, we also ran a higher
502: resolution version of one model with 30 times the baryonic disk mass
503: resolution (and therefore number of particles) and 2 times the resolution
504: of the dark matter and intragroup medium. In this case we also decreased
505: the gravitational softening length of the baryons by a factor of 3, and
506: increased that of the dark matter by a factor of 3. While this test
507: yielded much better resolution of the stellar disks and in particular the
508: tidal material, the general merger dynamics were identical to the low
509: resolution version.
510:
511:
512:
513:
514:
515: %------------------------------
516: %------------------------------
517: %------------------------------
518: \section{The Collision Between The Milky Way and Andromeda}
519: \label{sec:results}
520:
521: \subsection{Generic Features of the Merger}
522: \label{ssec:generic}
523:
524: In Figures~\ref{fig:starimages}~through~\ref{fig:crelvel} we present the
525: basic properties of the dynamical evolution of our Local Group, from 5~Gyr
526: in the past and until 10~Gyr into the future, beyond the merger time
527: between the Milky Way and Andromeda. Most of the features present in these
528: figures are generic to binary galaxy interactions, and have been described
529: in great detail by prior studies \citep[see, e.g.,][]{TT72,BH91,
530: BH92rev,MH96,Cox06}. However, we will review some of the details that are
531: particularly relevant to the Local Group, and subsequently highlight the
532: unique status of our own Sun which will be a participant in this galaxy
533: interaction. The future evolution of structures beyond the local group was
534: simulated elsewhere \citep{NL03,NL04,Bus03,Bus05}.
535:
536:
537:
538: \begin{figure*}
539: \begin{center}
540: \resizebox{16.0cm}{!}{\includegraphics{f2.eps}}\\
541: \caption{ Time sequence of the projected stellar density, shown in
542: red-scale, for the merger of the Milky Way (MW) and Andromeda (M31).
543: Andromeda is the larger of the two galaxies and begins the simulation in
544: the upper-right. The Milky Way begins on the edge of the image in the
545: lower-left. Each panel is 1.5~Mpc square, and the simulation time, in Gyr
546: relative to today, appear on the top-left label of each panel. The
547: trajectories of the Milky Way and Andromeda are depicted by the blue solid
548: lines. }
549: \label{fig:starimages}
550: \end{center}
551: \end{figure*}
552:
553:
554:
555: \begin{figure*}
556: \begin{center}
557: \resizebox{16.0cm}{!}{\includegraphics{f3.eps}}\\
558: \caption{ Same as Figure~\ref{fig:starimages} only the projected gas
559: density is shown. }
560: \label{fig:gasimages}
561: \end{center}
562: \end{figure*}
563:
564:
565:
566: \begin{figure*}
567: \begin{center}
568: \resizebox{16.0cm}{!}{\includegraphics{f4.eps}}\\
569: \caption{ Time sequence of the projected surface mass density of stars,
570: during the final merger between Andromeda and the Milky Way. Panels have
571: varying scales as specified by the label, in kpc, on the lower--right of
572: each panel. The top--left panel is identical to the ``Today'' panel in
573: Figure~\ref{fig:starimages}. The simulation time, with respect to today,
574: is shown in the top--left of each panel.}
575: \label{fig:zoomimage}
576: \end{center}
577: \end{figure*}
578:
579:
580:
581: To begin, Figures~\ref{fig:starimages}~and~\ref{fig:gasimages} present
582: the entire evolution of the Local Group from the point of view of a
583: distant observer. These images begin at the start of our simulation,
584: when the Milky Way and Andromeda are separated by 1.3~Mpc, and include
585: the present state of the Local Group (labeled ``Today'')
586: and the eventual merger of the Milky Way and Andromeda. As a guide to
587: the eye, each panel includes the trajectory of both the Milky Way and
588: Andromeda.
589:
590: Shown in Figure~\ref{fig:starimages} is the evolution of the stellar
591: component, which in our simulation only has contributions from the Milky
592: Way and Andromeda as we ignore any structure smaller than the two largest
593: galaxies in the Local Group.
594:
595: Figure~\ref{fig:gasimages} presents the projected gas distribution during
596: the interaction, with panels shown at the same times as in
597: Figure~\ref{fig:starimages}. Here the color--scale has been stretched to
598: emphasize the abundant quantity of low--density gas that is spread
599: throughout the local group. The initial condition of our Local Group model
600: assumes a uniform distribution of warm gas, however the gas quickly
601: responds to the non--uniform potential. In particular gas is accreted and
602: shocked to form a hydrostatic halo of warm gas around the Milky Way and
603: Andromeda galaxies. The gas distribution is also clearly affected by the
604: interaction itself, as shocks develop once the galaxy halos begin to
605: interpenetrate at close separation.
606:
607: While a detailed analysis of the diffuse gaseous intragroup medium is
608: beyond the primary focus of our work, the basic properties of this
609: component appear to be consistent with observations. In particular, our
610: model predicts that, at the present state, the intragroup medium is
611: predominantly warm $10^5-10^6$~K, has a fairly low density, $10^{-4} -
612: 10^{-6}$~cm$^{-3}$, and fills the entire volume that we simulate. We note
613: this medium is also expected to extend to larger scales into what has been
614: termed the ``Warm-Hot Intergalactic Medium" \citep{CO99,HGM98,Dav01}.
615: While the presence of our intragroup medium is consistent with all current
616: data \citep{Oso02}. Owing to the difficulty in directly observing gas in
617: this state, most evidence for its existence comes from indirect means. For
618: example, $Chandra$ and $FUSE$ observations of $z\sim0$ Oxygen and Neon
619: absorption along numerous sight lines suggests the presence of a local,
620: volume--filling, diffuse, warm medium \citep{Nic02,Nic03,Sem03,Sav03},
621: although detailed analysis of the ionization states suggest that this is a
622: complex multi--phase medium that our simulation does not have the
623: resolution to model.
624:
625: While Figure~\ref{fig:starimages} presented the dynamical evolution of
626: the stellar mass on large scales, this vantage point makes it difficult
627: to distinguish the tell-tale signs of a galaxy merger. We therefore
628: zoom into the central regions of the Local Group and specifically show the
629: merger between the Milky Way and Andromeda in Figure~\ref{fig:zoomimage}.
630: From this viewpoint the classic signatures of a galaxy interaction, such
631: as tidal tails, plumes, and shells are clearly evident.
632:
633: The physical separation between the Milky Way and Andromeda is presented in
634: Figure~\ref{fig:csep}. The current state of the Local Group occurs
635: $\sim5$~Gyr after the start of our simulation. For the standard set of
636: cosmological parameters \citep{Teg06short}, this implies that we initiate the
637: simulation at a redshift of $z\approx 0.5$, around the time when the Sun
638: was born in the Milky Way disk\footnote{Note that extending the simulation
639: to significantly earlier times is not adequate since stellar ages imply
640: that the two galactic disks (and presumably their halos) have not been
641: fully assembled at $z\ga2$ \citep{Wyse07}.}. Figure~\ref{fig:csep} also
642: clearly shows the prediction that our models makes for the future collision
643: between the Milky Way and Andromeda. Their first close passage will occur
644: less than 2~Gyr from the present, and the centers are fully coalesced in
645: less than 5~Gyr. These time scales are comparable to the lifetime of our
646: Sun \citep{SBK93} and admit the possibility that an observer
647: in the Solar System will witness some (or all) of the galaxy collision.
648: We will return to this possibility in \S\ref{ssec:earth}.
649:
650:
651:
652: \begin{figure}
653: \begin{center}
654: \resizebox{8.0cm}{!}{\includegraphics{f5.eps}}\\
655: \caption{ The separation between the centers of Andromeda and the Milky Way
656: during the course of their merger. The current separation of $\sim780$~kpc
657: is shown with a horizontal dashed line and occurs at $T\approx4.7$~Gyr. }
658: \label{fig:csep}
659: \end{center}
660: \end{figure}
661:
662:
663:
664: \begin{figure}
665: \begin{center}
666: \resizebox{8.0cm}{!}{\includegraphics{f6.eps}}\\
667: \caption{ The relative velocity between the centers of Andromeda and the
668: Milky Way galaxies during the course of their merger. The present
669: state of the Local Group occurs at $T\approx4.7$~Gyr. }
670: \label{fig:crelvel}
671: \end{center}
672: \end{figure}
673:
674:
675:
676: Finally, we present the relative velocity between the Milky Way and
677: Andromeda in Figure~\ref{fig:crelvel}. As in Figure~\ref{fig:csep}
678: we clearly delineate the present state of the Local Group with a
679: hatched region whose size corresponds to the errors estimated with the
680: velocity measurements. We note that the velocities presented are
681: relative to each galaxies center of mass, and do not correct for any
682: motion relative to that.
683:
684: Now that the general features of our Local Group model have been
685: outlined, we next explore the validity of our model, the merger
686: dynamics, the star formation during the interaction, and the
687: properties of the merger remnant.
688:
689:
690:
691:
692: %----------------------
693: %----------------------
694: \subsection{The Current State of the Local Group}
695: \label{ssec:present}
696:
697:
698: Although our model for the Local Group presents its past, current, and
699: future evolution, the only way to test its validity is by comparing it to
700: the empirical data on its present-day state. As we focus primarily on the
701: evolution of the two largest galaxies in the Local Group, the Milky Way and
702: Andromeda, it is only the relative separation and motion of these two
703: galaxies that can be compared to data. As mentioned in both
704: \S\ref{sec:intro} and \S\ref{sec:results}, the separation and
705: line--of--sight velocity of the Milky Way and Andromeda are currently
706: measured to be 780~kpc \citep[][and references therein]{McC05, Rib05}, and
707: -120~\kms \citep{BT}, respectively. The proper motion of Andromeda
708: perpendicular to our line of sight is less well constrained, but current
709: estimates suggest that it is $<200$~\kms \citep{P01,Loeb05}.
710:
711: Figures~\ref{fig:csep} and \ref{fig:crelvel} present the direct comparisons
712: between the observational constraints and our model, and clearly
713: demonstrate that it is viable. In particular, at a time of $4.7$~Gyr after
714: the start of the simulation (close to the present cosmic time) the
715: separation between the Milky Way and Andromeda is 780~kpc, while the
716: relative radial velocity is 135~\kms and the tangential velocity is
717: 132~\kms. While the tangential velocity is well within the limits
718: currently favored, the line--of--sight velocity is slightly larger than
719: (but within 2.0~$\sigma$ of) the observations. Even though a number of our
720: model assumptions can be manipulated to reduce these values, e.g., the
721: initial separation and eccentricity of the fiducial orbit, we note that our
722: model is a good fit to the velocity when the separation between the Milky
723: Way and Andromeda is larger than 780~kpc. Given the observational
724: uncertainties in these values we feel that there are likely to be a large
725: number of models that can simultaneously fit all the data within its
726: 2$\sigma$ error bars.
727:
728:
729: %----------------------
730: %----------------------
731: \subsubsection{An Ensemble of Models}
732: \label{ssec:ensemble}
733:
734:
735: As mentioned in \S\ref{ssec:other}, the model described up to this point is
736: one of twenty Local Group models that we have simulated. Since the primary
737: conclusion of this paper, involving the timescale for the eventual merger
738: between the Milky Way and Andromeda, may be influenced by any number of our
739: model assumptions, we explicitly show the separation between the Milky Way
740: and Andromeda and therefore the time of the merger, for all of our models
741: in Figure~\ref{fig:csep_ensemble}.
742:
743:
744: \begin{figure}
745: \begin{center}
746: \resizebox{8.0cm}{!}{\includegraphics{f7.eps}}\\
747: \caption{ The separation between the centers of Andromeda and the Milky Way
748: during the course of their merger for a large ensemble of Local Group
749: models. The current separation of $\sim780$~kpc is shown with a horizontal
750: dashed line and all models are normalized to this particular time. }
751: \label{fig:csep_ensemble}
752: \end{center}
753: \end{figure}
754:
755:
756: Figure~\ref{fig:csep_ensemble} demonstrates several interesting features.
757: Nearly all of the models provide a similar outcome for time when the two
758: galaxies make their first passage (ensemble average and standard deviation
759: are $T=2.8\pm0.5$~Gyr). The same holds true for the final merger time
760: ($T=5.4\pm0.4$~Gyr). We also note that these average values are slightly
761: larger than one provided by the model described up to this point. This
762: highlights a general trend, namely the mergers that start with a larger
763: separation and have to traverse a longer path through the intragroup
764: medium, usually (but not always) have a quicker merger dynamics.
765: Regardless of the relatively small differences between merger times in
766: these models, they are all completely coalesced by 6.2~Gyr from today. In
767: the following section we will argue that this result is a direct byproduct
768: of our inclusion of an intragroup medium.
769:
770:
771: %----------------------
772: %----------------------
773: \subsection{Merger Dynamics}
774: \label{ssec:mergerdyn}
775:
776: \subsubsection{Timescale}
777: \label{ssec:time}
778:
779:
780: One of the most intriguing characteristics of our Local Group model is the
781: relatively quick timescale for the interaction and merger between the
782: Milky Way and Andromeda. As stated in the previous section, their first
783: close passage will occur in less than 2~Gyr, and the final coalescence will
784: occur in less than 5~Gyr. While the following section will specifically
785: address the possible dynamics of the Sun during the merger, we show here
786: that the cause for the short merger timescale is the intragroup medium.
787:
788:
789:
790: \begin{figure}
791: \begin{center}
792: \resizebox{8.0cm}{!}{\includegraphics{f8.eps}}\\
793: \caption{ The separation between Andromeda and the Milky Way during the
794: course of their merger. Our fiducial model in red is compared to a case
795: with no intragroup medium (dark matter$+$gas) in blue. Dynamical friction
796: on the diffuse medium shortens the MW/M31 coalescence time by a factor of a
797: few. The present time is 5 Gyr after the start of the simulation. }
798: \label{fig:csepboth}
799: \end{center}
800: \end{figure}
801:
802:
803:
804: Our model for the Local Group assumes that the Milky Way and Andromeda are
805: embedded in a diffuse medium of dark matter and gas. The medium itself
806: (minus the evolved galaxies) is unvirialized and of low overdensity
807: ($\delta\sim5$) corresponding to a region of the Universe that has
808: decoupled from the Hubble flow and started its evolution toward a collapsed
809: virialized state. This medium exerts dynamical friction on the Milky Way
810: and Andromeda and speeds up the merger dynamics by soaking orbital energy
811: and angular momentum. This is shown explicitly in
812: Figure~\ref{fig:csepboth}, which compares the Milky Way -- Andromeda
813: separation in our fiducial model to an identical model without a diffuse
814: medium. Without the intragroup medium, the merger timescale is nearly
815: three times longer than with it.
816:
817: The effect of the intragroup medium should scale similarly to the standard
818: (Chandrasekhar) formula for dynamical friction \citep[Eq. 7-18]{BT}, in
819: which the deceleration of a massive object is proportional to the
820: background matter density, $d{\rm v}/dt \propto \rho$. Therefore, the rate
821: at which angular momentum is extracted from the orbit depends on the
822: assumed intragroup medium density, a quantity which is poorly constrained
823: observationally. Once the dark matter halos begin to interpenetrate, i.e.,
824: when the Milky Way -- Andromeda separation is $\sim100$~kpc, the merger
825: completes relatively quickly because the dark matter halos dominate over
826: the background density.
827:
828: This last point is particularly relevant to simulations of binary galaxy
829: mergers, which typically omit any background overdensity. While this
830: omission does not significantly alter the dynamical friction estimates once
831: the halos overlap and therefore dominate the background density, it may
832: still change the distribution of orbits in high density environments where
833: the galaxies traverse through significant overdensities prior to their
834: interaction event. In this sense, the merger timescales extracted from
835: binary merger simulations may be considered an upper limit that is most
836: applicable to galaxies in the field.
837:
838:
839:
840:
841:
842: %---------------------------------------------
843: %---------------------------------------------
844: \subsubsection{The Fate of our Solar System}
845: \label{ssec:earth}
846:
847:
848:
849: \begin{figure*}
850: \begin{center}
851: \resizebox{4.5cm}{!}{\includegraphics{f9a.eps}}
852: \resizebox{4.5cm}{!}{\includegraphics{f9b.eps}}
853: \resizebox{4.5cm}{!}{\includegraphics{f9c.eps}}\\
854: \resizebox{4.5cm}{!}{\includegraphics{f9d.eps}}
855: \resizebox{4.5cm}{!}{\includegraphics{f9e.eps}}
856: \resizebox{4.5cm}{!}{\includegraphics{f9f.eps}}\\
857: \caption{ The possible location of the Sun during various stages of the
858: merger between the Milky Way and Andromeda. The top panel shows all
859: stellar particles in our simulation that have a present-day galactocentric
860: radius of 8$\pm0.1$~kpc with a red cross and tracks their position into
861: the future. The bottom panel presents a histogram of the radial distance
862: from the center of the Milky Way. }
863: \label{fig:whe}
864: \end{center}
865: \end{figure*}
866:
867:
868: An interesting consequence of the short timescale for the merger between
869: the Milky Way and Andromeda is the possibility that a human (or decedent
870: thereof) observer will witness the interaction and merger. At the heart of
871: this issue is the comparison between the lifetime of the Sun and the
872: interaction timescale.
873:
874: Current evolutionary models \citep[see, e.g.,][]{SBK93} predict that the
875: Sun will steadily increase its size and luminosity for the next 7~Gyr as it
876: slowly consumes all available hydrogen and evolves towards a red giant
877: phase. While this places a strong upper limit to the extent of life on
878: Earth, it is likely that much smaller changes ($<50$\%) in the Sun's
879: luminosity will significantly alter the Earth's atmosphere and thus its
880: habitability within the next 1.1-3.5~Gyr \citep{Kas88}. \citet{Kory01}
881: suggested that the onset of these effects could be delayed by increasing
882: the orbital radius of the Earth through a sequence of interactions with
883: bodies in the outer Solar System, and we can not rule out the possible
884: colonization of habitable planets in nearby stars, especially long-lived
885: M-dwarfs \citep{Udry07} whose lifetime may exceed $10^{12}$ years
886: \citep{ABL05}. In short, it is conceivable that life may exists for as
887: little as 1.1~Gyr into the future or, if interstellar travel is possible,
888: much longer.
889:
890: Regardless of the prospects for life in the future, we can attempt to
891: predict what any potential observer at the solar Galactic circle might see
892: by tracking candidate Suns in our simulation. In particular, we flag all
893: stellar particles with a galactocentric orbital radius of 8$\pm0.5$~kpc,
894: corresponding to the observed value for the Sun \citep{Eis03}, and
895: subsequently follow the location of these particles forward in time. This
896: procedure typically yields $\sim700$ particles as ``candidate Suns''
897: (although it should be kept in mind that owing to our limited resolution,
898: each of the simulated particles is many orders of magnitude more massive
899: than the Sun). The results are presented in Figure~\ref{fig:whe} and
900: demonstrate some of the possible outcomes for the future location of our
901: Sun. Given the uncertainties in our model parameters and the fact that the
902: orbital period of the Sun around the Milky Way is much shorter than the
903: merger timescale, it is not possible to forecast reliably the actual phase
904: of the Galactic orbit of the Sun at the time of closest approach to
905: Andromeda. Therefore we regard all the stellar particles at the
906: galactocentric radius of the Sun as equally probable of representing the
907: Sun. We will first outline some of the general features of our fiducial
908: model, before showing similar distributions from a subset of models to
909: assess the reliability of these results.
910:
911: Figure~\ref{fig:whe} outlines the wide variety of potential locations for
912: our candidate Suns during the future evolution of the Local Group. For
913: example, the top--right panel in Figure~\ref{fig:whe} demonstrates the
914: location of the candidate Suns after the first passage of Andromeda. At
915: this point, the observer will most likely still be in the (now disturbed)
916: disk of the Milky Way, but there exists a 12\% chance that the Sun will be
917: tidally ejected and take part in the tidal tail material that is
918: $>20$~kpc away from the Milky Way center.
919:
920: The probability that the candidate Sun will be located farther than
921: $20$~kpc from the Milky Way center steadily increases as the interaction
922: progresses. At the second passage, the percentage of Suns that are farther
923: than $20$~kpc is 30\%. This number increases to 48\% after the second
924: passage. and is 68\% in the merger remnant. In fact, there is a 54\%
925: chance that the Sun will be at radii larger than 30~kpc in the merger
926: remnant. However, we caution that this probability is derived at one point
927: in time. Individual stars will generally spend much of their orbital time
928: at large radii, even if their orbit is eccentric and they come much closer
929: to the galactic center at other times.
930:
931: One unexpected possibility for the location of the future Sun is
932: demonstrated in the lower--middle panel of Figure~\ref{fig:whe}, namely the
933: Sun may actually become bound to Andromeda instead of the Milky Way before
934: the two galaxies coalesce. Such a situation occurs when material becomes
935: loosely bound after the first passage and is later captured by the
936: gravitational potential of Andromeda during its second passage. While this
937: outcome is unexpected and certainly exciting, only 2.7\% of the candidate
938: Suns became bound to Andromeda and so this outcome is relatively unlikely.
939:
940:
941: \begin{figure*}
942: \begin{center}
943: \resizebox{5.3cm}{!}{\includegraphics{f10a.eps}}
944: \resizebox{5.3cm}{!}{\includegraphics{f10b.eps}}
945: \resizebox{5.3cm}{!}{\includegraphics{f10c.eps}}\\
946: \caption{ The possible location of the Sun during the first passage (left
947: panel), second passage (middle panel), and final merger (right panel) for
948: several different models of the merger between the Milky Way and Andromeda.
949: The histograms are difficult to distinguish, which highlights the generic
950: probability for the Sun to reside at various galactocentric radii. The
951: distribution of radii from the fiducial model, which is presented in
952: Figure~\ref{fig:whe}, is denoted by a hatched histogram. }
953: \label{fig:whe_ensemble}
954: \end{center}
955: \end{figure*}
956:
957:
958: In Figure~\ref{fig:whe_ensemble} we attempt to assess whether the
959: percentages just quoted for our fiducial model are characteristic of all
960: our models and are therefore robust. Each panel in
961: Figure~\ref{fig:whe_ensemble} shows the distribution of galactocentric
962: radii for candidate Suns during three periods of the interaction in several
963: representative Local Group models, including the fiducial one. The
964: similarity between the models highlights the generic features of the
965: fiducial model, although there is a large spread in the percentages quoted
966: in the previous paragraphs. The common distributions might have been
967: expected given the similar orbital evolution demonstrated in
968: Figure~\ref{fig:csep_ensemble} and the tendency for mergers to preserve the
969: hierarchy of the initial binding energies of the collisionless particles in
970: the merger remnant.
971:
972: We note that several of the interactions generate unique tidal features
973: which result in subtle differences in the distributions shown in
974: Figure~\ref{fig:whe_ensemble}. These differences are an outcome of
975: different disk-spin orientations that were examined in the different
976: models. One model in particular yields a remnant where some of the
977: candidate Suns reside in a shell--like structure at $\sim50-60$~kpc (see
978: right--most panel in Figure~\ref{fig:whe_ensemble}). Only half of the
979: models include the possibility for the Sun to be ``stolen'' by Andromeda,
980: however one model yielded a probability that was nearly three times larger
981: than in the fiducial case.
982:
983:
984:
985: %-------------------------
986: %-------------------------
987: \subsection{Star Formation}
988: \label{ssec:sf}
989:
990:
991:
992: \begin{figure}
993: \begin{center}
994: \resizebox{8.0cm}{!}{\includegraphics{f11.eps}}\\
995: \caption{
996: The cumulative star-formation rate during the merger
997: of the Milky Way and Andromeda compared to the star
998: formation for models of the Milky Way and Andromeda
999: evolved in isolation.
1000: }
1001: \label{fig:sfr}
1002: \end{center}
1003: \end{figure}
1004:
1005:
1006:
1007: There is mounting evidence that galaxy interactions are the predominant
1008: mechanism for producing large bursts of star formation, such as in the
1009: ultra--luminous infrared galaxies \citep[ULIRGs,][]{SM96,BH92rev}. During
1010: the interactions, gravitational torques extract angular momentum from the
1011: gas and funnel it to the center of the merger galaxy where it participates
1012: in a centrally--concentrated starburst. Because the future of the Local
1013: Group entails a major galaxy merger, it is natural to examine whether the
1014: merger of the Milky Way with Andromeda will become a ULIRG.
1015:
1016: This question is addressed explicitly in Figure~\ref{fig:sfr}, which shows
1017: the star--formation rate during the entire evolution of the Local Group,
1018: including the merger between the Milky Way and Andromeda. Throughout the
1019: entire evolution of the Local Group, the star--formation rate steadily
1020: decreases with time, and hence we conclude that the merger Local Group will
1021: not become a ULIRG in the future. In fact, the final coalescence yields
1022: star formation that is barely enhanced above the that which would occur if
1023: the Milky Way and Andromeda has not participated in the merger at all.
1024:
1025: The weak starburst event triggered by the merger between the Milky Way
1026: and Andromeda is a direct result of their present low gas content.
1027: Moreover, a large fraction ($>75$\%) of this gas will be consumed by
1028: quiescent star formation by the time the merger actually occurs. In short,
1029: both disks will be extremely gas--poor during the final coalescence and
1030: there will be no fuel for the starburst.
1031:
1032: While we have not explicitly tracked the black holes at the center of the
1033: Milky Way and Andromeda, it is interesting to speculate whether the merger
1034: will produce a luminous quasar, which many models argue is intricately
1035: linked to galaxy mergers and starbursts \citep{Hop06big}. Even though
1036: Figure~\ref{fig:sfr} demonstrates that there is not enough gas to fuel a
1037: powerful starburst, this gas content is clearly sufficient to ignite a
1038: luminous quasar if $\sim1$\% of it is accreted by the black hole. While
1039: the current work cannot address this possibility in detail, our model
1040: provides a framework to study the formation of quasars in the future.
1041:
1042:
1043: %------------------------------
1044: %------------------------------
1045: \subsection{The Merger Remnant}
1046: \label{ssec:mrem}
1047:
1048:
1049: \begin{figure*}
1050: \begin{center}
1051: \resizebox{6.0cm}{!}{\includegraphics{f12a.eps}}
1052: \resizebox{6.0cm}{!}{\includegraphics{f12b.eps}}\\
1053: \resizebox{6.0cm}{!}{\includegraphics{f12c.eps}}
1054: \resizebox{6.0cm}{!}{\includegraphics{f12d.eps}}\\
1055: \resizebox{6.0cm}{!}{\includegraphics{f12e.eps}}
1056: \resizebox{6.0cm}{!}{\includegraphics{f12f.eps}}\\
1057: \caption{ Properties of the Milky Way and Andromeda merger remnant, which
1058: we abbreviate as {\it Milkomeda}. Panel (a) shows the spherically averaged
1059: distribution of dark matter, stars, and gas. The mass density is expressed
1060: in units of the present--day critical density, $\rho_{\rm crit}=
1061: 10^{-29}~{\rm g~cm^{-3}}$. Panel (b) shows the projected distribution of
1062: stellar mass, including the contribution from stars formed during the
1063: course of the simulation which are termed ``New Stars,'' plotted against
1064: $R^{1/4}$, where $R$ is the projected radius. Panel (c) presents the
1065: kinematics of {\it Milkomeda}. For 195 projections, the ellipticity of the
1066: half--mass isophote $\varepsilon$, the rotation velocity ($V$) along the major
1067: axis $V_{\rm maj}$ and the central velocity dispersion $\sigma$, are plotted
1068: as a colored histogram, with darker shades representing a higher concentration
1069: of projected properties. The solid line represents what is expected from an
1070: oblate isotropic rotation and overplotted is data extracted from the literature.
1071: Panel (d) shows the velocity anisotropy, defined as the ratio $V/\sigma$
1072: normalized by the value expected for an oblate isotropic rotator, versus the
1073: isophotal shape parameter $a_4/a$, where positive values are disky isophotes and
1074: negative values are boxy. Overplotted in black are current data (Ralf Bender,
1075: private communication). Panel (e) shows the X-ray surface brightness profile,
1076: and panel (f) shows {\it Milkomeda} on the near--IR fundamental plane along with
1077: the relation observed by \citet{Pah98}. }
1078: \label{fig:remnant}
1079: \end{center}
1080: \end{figure*}
1081:
1082:
1083:
1084: Galaxy mergers have become an area of intense study owing to their proposed
1085: role in shaping galaxy morphology. In particular, the ``merger
1086: hypothesis'' \citep{T77} posits that the interaction and merger of two
1087: spiral galaxies leaves behind a remnant that is morphologically and
1088: kinematically similar to an elliptical galaxy. Since the future evolution
1089: of the Local Group contains such an event, it is natural to examine
1090: whether the local group will eventually consist of a single elliptical
1091: galaxy, and if so, how do the properties of this galaxy differ from
1092: present--day ellipticals which were formed at earlier epochs.
1093:
1094: Note that the morphology of the Milky Way and Andromeda merger remnant,
1095: which we abbreviate hereafter as {\it Milkomeda}, was presented in
1096: Figure~\ref{fig:zoomimage}. This figure confirms the notion that galaxy
1097: mergers leave behind remnants that are spheroidal in shape, contain stars
1098: with a wide range of orbits and a large velocity dispersion, and possess very
1099: modest net rotation.
1100:
1101: In Figure~\ref{fig:remnant}, we present a series of plots that further
1102: support the assertion that {\it Milkomeda} resembles an elliptical galaxy
1103: and also serve to better quantify its properties. In panel (a), we present
1104: the spherically averaged mass profile, decomposed by the three components;
1105: dark matter, stellar and gaseous mass. At radii greater than $\sim2-3$~kpc,
1106: the dark matter dominates the mass density, and at radii greater than
1107: $\sim20$~kpc, the profile is well--fit by the NFW \citep{NFW96} or Hernquist
1108: \citep{H90} profile.
1109:
1110: The projected mass distribution shown in panel (b) of
1111: Figure~\ref{fig:remnant} presents the first direct evidence that
1112: {\it Milkomeda} resembles an elliptical galaxy. Specifically, the stellar
1113: surface density is close to a pure R$^{1/4}$ distribution, with the exception
1114: of the inner $\sim400$~pc ($\sim2.5\times$ the gravitational softening
1115: length), where the surface density flattens to a nearly constant
1116: density core.
1117:
1118: We have also quantified the kinematics and isophotal shape of the {\it
1119: Milkomeda} galaxy following the methods of \citet{Cox06rot}. Panel (c) in
1120: Figure~\ref{fig:remnant} presents the anisotropy diagram, a measure of the
1121: half--mass isophote ellipticity versus the maximum rotation along its major
1122: axis divided by the central velocity dispersion. The shaded region
1123: represents the distribution of values for {\it Milkomeda} if viewed
1124: from 195 directions that uniformly sample the unit sphere \citep[with
1125: angles selected using HEALPIX,][]{Gor05}. Also plotted in this figure is
1126: the relation expected from an oblate isotropic rotator as a solid line.
1127: Owing to the high concentration of projections that closely track the solid
1128: line, this analysis demonstrates that {\it Milkomeda} is nearly an oblate
1129: isotropic rotator. In addition, the
1130: deviations from a perfect ellipse are also quantified and are presented in
1131: panel (d) in Figure~\ref{fig:remnant}. Depending on the viewing direction,
1132: {\it Milkomeda} may appear to be either disky or boxy, with a slightly larger
1133: fraction of views producing disky isophotes.
1134:
1135: We have also analyzed the properties of the hot gas in and around {\it
1136: Milkomeda}. The evolution of this component was presented in
1137: Figure~\ref{fig:gasimages} and clearly demonstrates the formation of an
1138: extended gaseous halo, primarily accreted from the large reservoir of the
1139: intragroup medium. Although the gas temperature was originally
1140: $3\times10^5$~K, it has been shock-heated to the virial temperature of
1141: $\sim3\times10^6$~K in {\it Milkomeda} and has a gradient to cooler
1142: temperatures at large radii. This hot gas leads to an X-ray surface
1143: brightness profile shown in panel (e) of Figure~\ref{fig:remnant} and
1144: a total X-ray luminosity of
1145: $\sim10^{41}$~ergs~s$^{-1}$, which is consistent with present--day
1146: elliptical galaxies of equivalent B-band luminosity
1147: ($\sim3\times10^{10}$~\lsun).
1148:
1149: In general, {\it Milkomeda} resembles the remnants of gas--rich major
1150: mergers, which in--turn resemble the general population of low-- and
1151: moderate--luminosity elliptical galaxies \citep{NJB06,Cox06rot}. However,
1152: there are some systematic differences that likely arise because of the much
1153: smaller gas content of the Milky Way and Andromeda when they merge. In
1154: particular, the inner regions of {\it Milkomeda} have a much lower stellar
1155: density than present--day ellipticals, which are often observed to have
1156: steep spikes of newly formed stars \citep{RJ04,Kor07}. These central
1157: excesses arise from high gas concentrations that fuel nuclear starbursts
1158: during the merger event \citep[see,e.g.,][]{MH94dsc,Sp00,Cox06}. This
1159: process does not occur during the formation of {\it Milkomeda} (see
1160: \S\ref{ssec:sf}), for which the stars formed during the merger simulation
1161: have an identical profile to the entire stellar population (see panel (b)
1162: in Figure~\ref{fig:remnant}).
1163:
1164: The diffuse nature of {\it Milkomeda} is also evident if it is projected
1165: onto the near-IR fundamental plane as shown in panel (f) of
1166: Figure~\ref{fig:remnant}, where it lies above the observed relation. The
1167: half--mass radius is 4.9~kpc, which is larger than the mean relation found in
1168: the Sloan Digital Sky Survey \citep{Shen03,Des07} for galaxies of
1169: equivalent stellar mass ($1.3 \times 10^{12}$\msun) or r--band absolute
1170: magnitude (-21.2). These comparisons gives credence to the claim that
1171: present--day ellipticals can not have formed from the merger of present day
1172: spirals \citep{NO07}.
1173:
1174:
1175:
1176:
1177:
1178: %------------------------------
1179: %------------------------------
1180: \section{Conclusions}
1181: \label{sec:conc}
1182:
1183: In this paper we have used an N-body/hydrodynamic simulation to track the
1184: evolution of the Local Group, focusing primarily upon the two most massive
1185: galaxies: the Milky Way and Andromeda. In contrast to most prior work,
1186: which typically employed models for the Local Group to infer its total mass
1187: or the proper motion of its constituents, we simulated the large-scale
1188: dynamics of all matter in the Local Group including dynamical friction on
1189: the intragroup medium, leading to the eventual merger of the Milky Way and
1190: Andromeda.
1191:
1192: Owing to the diffuse intragroup medium that was assumed to pervade the
1193: Local Group with a total mass comparable to that the galaxies, we have
1194: found that the interaction and merger between the Milky Way and Andromeda
1195: will occur in less than 5~Gyr, a timescale comparable to the lifetime of
1196: the Sun \citep{SBK93}. This Local Group model therefore admits
1197: the possibility that future astronomers in the Solar System will witness
1198: parts of, or the entire interaction and merger of the Milky Way and
1199: Andromeda.
1200:
1201: With this in mind we have calculated the probable location of our Solar
1202: System during specific points of the future interaction and find several
1203: interesting outcomes. First, there is a chance that the Sun will be
1204: ejected along with other tidal material into a long tidal tail following
1205: the first passage of Andromeda. Second, as a result of the disruptive
1206: effects of each close tidal passage, there is an increasing chance that the
1207: Sun will inhabit extended tidal features as the interaction proceeds.
1208: Moreover, there is a small chance that the Sun will be more tightly bound
1209: to Andromeda at some point during the merger. In such a case, Andromeda
1210: will capture the Sun and future astronomers in the solar system might see
1211: the Milky Way as an external galaxy in the night sky\footnote{Aside from
1212: the changed night sky during the interaction between the Milky Way and
1213: Andromeda, future observers might witness enhanced comet showers due to the
1214: increased flux of stars passing by the solar system and perturbing the Oort
1215: cloud. Additional observables would include a slightly enhanced star
1216: formation in the two galaxies and the production of hot gas by the shocks
1217: surrounding the interaction region.}.
1218:
1219: While this paper highlights the possible outcomes of the future interaction
1220: between the Milky Way and Andromeda, we emphasize that our model is likely
1221: to be one of many plausible models within an ensemble of possibilities that
1222: span the uncertain value of the transverse velocity of Andromeda and the
1223: density (and other properties) of the intragroup medium. We have performed
1224: 20 additional runs in order to test the sensitivity of our results to
1225: various assumptions of our model -- mainly involving the initial orbit of
1226: the Milky Way and Andromeda. These runs yield similar estimates for the
1227: merger timescale as well as for the possible locations of the Sun in the
1228: future, provided that the intragroup medium is indeed similar to our
1229: fiducial case. While this gives us some confidence that our results are
1230: robust, an even larger suite of models, that spans a much wider set of
1231: model assumptions, will provide better statistics on these results. In
1232: addition, employing higher resolution simulations with increased
1233: complexity, may shed light on the nature of the intragroup medium, the soft
1234: X-ray background \citep{Oso02}, galactic substructure \citep{Will05}, the
1235: origin of the Magellanic Clouds and Stream \citep{Besla07}, and the future
1236: evolution of globular clusters \citep{For00}.
1237:
1238: Finally, we note that the simulated views from the distribution of
1239: locations for the candidate Suns in the merger remnant (see
1240: Fig.~\ref{fig:whe}), which we have termed {\it Milkomeda}, represent
1241: the {\it only} views available for a future
1242: local astronomer. Extragalactic astronomy will come to an end within 100
1243: billion years if the cosmological constant will not evolve with time. Owing
1244: to the accelerated expansion caused by a steady cosmological constant, all
1245: galaxies not bound to the Local Group will eventually recede away from the
1246: Local Group and exit our event horizon \citep{Loeb02}. At that point, the
1247: merger product of the Milky Way and Andromeda (with its bound satellites)
1248: will constitute the entire visible Universe \citep{NL03}.
1249:
1250:
1251:
1252:
1253: \section*{Acknowledgments}
1254:
1255: The simulations were performed at the Center for Parallel Astrophysical
1256: Computing at the Institute for Theory and Computation at the
1257: Harvard-Smithsonian Center for Astrophysics. This research was supported
1258: in part by a grant from The Foundational Questions Institute. We
1259: acknowledge helpful discussions with G. Besla, J. Dubinski, S. Dutta,
1260: L. Hernquist, P. Hopkins, and B. Robertson. We thank Ralf Bender for
1261: kindly providing data used in Figure~\ref{fig:remnant}.
1262:
1263:
1264:
1265: \bibliographystyle{mn2e}
1266: \bibliography{lg}
1267:
1268:
1269: \end{document}
1270:
1271:
1272:
1273:
1274: