1: % iaus2esa.tex -- sample pages for Proceedings IAU Symposium document class
2: % (based on v1.0 cca2esam.tex)
3: % v1.04 released 17 May 2004 by TechBooks
4: %% small changes and additions made by KAvdH/IAU 4 June 2004
5: % Copyright (2004) International Astronomical Union
6:
7: \NeedsTeXFormat{LaTeX2e}
8:
9: \documentclass{iaus}
10: \usepackage{graphicx}
11:
12: \title[Are the most metal-poor galaxies young?] %% give here short title %%
13: {Are the most metal-poor galaxies young?}
14:
15: \author[Kunth \& \"Ostlin] %% give here short author list %%
16: {Daniel Kunth$^1$%
17: and G\"oran \"Ostlin$^2$}
18:
19: \affiliation{$^1$ Institut d'Astrophysique de Paris, 98bis Boulevard Arago, F75014 Paris, France \break email: kunth@iap.fr\\[\affilskip]
20: $^2$ Stockholm observatory, AlbaNova University Center, 10691 Stockholm, Sweden \break email: ostlin@astro.su.se}
21:
22: \pubyear{2006}
23: \volume{235} %% insert here IAU Symposium No.
24: \pagerange{1--4}
25: \date{20 oct 2006 and in revised form ??}
26: \setcounter{page}{119}
27: \jname{Galaxy Evolution across the Hubble Time}
28: \editors{F.Combes \& J. Palous, eds.}
29: \begin{document}
30:
31: \maketitle
32:
33: \begin{abstract}
34: We review the possibility that metallicity could provide a diagnostic for the age of a galaxy,
35: hence that the most metal-poor star forming galaxies in the local universe may be genuinely
36: young. Indeed, observational evidence for ``downsizing'' shows the average age of the stars in
37: a galaxy to decrease with decreasing mass and metallicity. However, we conclude both from
38: observational and theoretical viewpoints that metallicity is not an arrow of time. Consequently
39: the most metal--poor galaxies of our local universe are not necessarely young. Current
40: observations suggest that an old stellar population is present in all metal-poor galaxies,
41: although a couple of cases, e.g. IZw18, remain under debate. Further observations with more
42: sentitive equipement should settle this question in the coming years.
43: \keywords{galaxies: abundances, dwarf, evolution }
44: %% add here a maximum of 10 keywords, to be taken form the file <Keywords.txt>
45: \end{abstract}
46:
47: \firstsection % if your document starts with a section,
48: % remove some space above using this command.
49:
50: \section{What is a young galaxy?}
51:
52:
53: In view of the recent popularity of the downsizing scenarios, in which
54: mean stellar age and metallicity decrease with decreasing galactic mass
55: it might be tempting to conjecture that some metal-poor,
56: actively star forming dwarf galaxies may be genuinely young, now
57: forming their first generation of stars. On the other hand, some of
58: the most metal-poor galaxies in our vicinity
59: are found among the dSph satellites of the Milky Way, which have
60: old stellar populations. Hence, it is quite obvious that
61: metallicity cannot be used as a simple cosmic clock, and we can
62: straightforwardly answer the title question of this invited talk with a firm No.
63: However, the bare possibility that {\it some} local galaxies may
64: be young calls for a little more scrutiny in answering to this question.
65:
66: What criteria must be met before we decide that a clump
67: of matter is to be classified as a galaxy? Even if galaxies mostly
68: consists of dark matter, we classify them from their stellar population
69: and are inclined towards
70: using stars and the star formation history as the relevant
71: temporal measure for galaxy formation.
72: %It is now realised that
73: Galaxies do not live in isolation: they merge,
74: tear out gas and stars from each other, and accrete matter.
75: What is the proper age definition for a galaxy in
76: this context? We take here the working definition that
77: the age of a galaxy is equal to that of its oldest stars.
78: % (to the extent that
79: %these can be detected and measured).
80:
81: This subject has been reviewed by \cite{ko2000} and \cite{ko2001}, and
82: we reiterate one conclusion from this work: finding a genuinely young
83: galaxy would be such an interesting discovery that the burden of proof
84: must lie on the youth hypothesis. Only when it has been shown beyond
85: reasonable doubt, the limit being set by current technology, that a galaxy
86: contains no old stars should we term a galaxy young.
87:
88: \section{Which are the most metal-poor galaxies?}
89:
90: %Some low mass star forming galaxies, dwarf irregulars and blue compact
91: %galaxies.... \cite{ko2000}.
92: Among local actively star forming galaxies, most are dwarfs: dwarf irregulars
93: (dIrr) and Blue Compact Dwarfs (BCD). They remain, however, a minority among
94: the general population of dwarf galaxies, \cite{ko2000}.
95: At optical wavelengths, their spectra are dominated by young stars and
96: ionised gas, closely resembling those of giant HII regions in nearby spiral
97: and irregular galaxies. Analysis of their spectra show that most of them
98: are metal deficient. Towards the lower end of the metallicity distribution
99: (O/H $\sim 1/50 Z_\odot$) we find galaxies like IZw18 and SBS0335-052.
100: Due to their extreme properties it has been conjectured by \cite{ss1970}
101: that these chemically unevolved galaxies could be young systems still in
102: the proceess of forming. However, dwarf spheroidal galaxies (dSph) can be
103: even more metal-poor, if judged by [Fe/H] for the stellar population, but we
104: should keep in mind that this is not directly comparable to nebular [O/H] -
105: \cite{gr2003}. Yet, dSphs are undisputably old.
106: %, but we measure it in a very different
107: %way: sometimes by spectroscopy of individual stars, sometimes from
108: %the colour of the red giant branch.
109: %The old stars that some of us believe
110: %are present in IZw18, likely have lower [Fe/H] than the [O/H] value
111: %for the ISM, and Grebel et al (2003) argue that dSph tend to have
112: %higher [Fe/H] than dIrr at the same luminosity and to an extent that
113: %cannot be explained by simple luminosity evolution.
114:
115: Since metal-poor galaxies exist and are common at low redshift, one could
116: of course expect many more at high redshifts.
117: The first generation of
118: star forming galaxies must, by definition, have been both young and metal-poor.
119: %(but look at SCUBA galaxies!).
120: However, selection effects tend to discriminate against the smallest, likely
121: most metal-poor, galaxies in the distant universe, and the
122: rather few estimates for Lyman Break Galaxies
123: point to rather 'normal' abundances (say 1/3 to $ Z_\odot$) - \cite{pe2006}.
124: %In the low density gas between galaxies, in the Lya forest, has very
125: %metal-poor gas been detected at high z, but this is probably not a probe
126: %of the metallicity of any galaxy (NOT CLEAR!!).
127:
128: We note that many parameters control the observed metallicity in a given galaxy.
129: They are incorporated into chemical evolutionnary models and
130: include : stellar evolution and nucleosynthesis, inflows and outflows
131: and the problem of the mixing and dispersion time scale of freshly released
132: heavy elements. \cite{ko2000} caution the use of simple
133: classical closed box model although clearly, gas mass fraction seems
134: to play a major role
135: \cite{le1979}.
136: %The lack of even more metal-poor local star forming galaxies and the apparent disconnect
137: %in metallicity between star-forming galaxies and DLAs
138: %at the same cosmic epoch has led some questions concerning thofe a possible offset between
139: %abundances measured in HII and HI regions.
140: Moreover metallicity measurements may be relevant to only one
141: particular component of a galaxy. A suggestion of \cite{ku1986}
142: has been made that HII gas could enrich itself with metals expelled by CC-SNe of time
143: scales shorter than
144: the life time of a starburst. Recent FUSE observations of some dwarf galaxies
145: show a possible disconnect in metallicity between HI and HII regions - \cite{ca2005}, \cite{le2006} -
146: although possible
147: saturation effects on the line of sight may alter such a comparison - \cite{le2006}.
148:
149:
150: \section{Metallicity as an arrow of time}
151: The metal content of galaxies being an important diagnostic of galactic evolution,
152: the question is whether metallicity can be regarded as an arrow
153: of time? A useful connection between evolutionary state and metallicity is the
154: luminosity--metallicity relationship.
155: Such a relation can naturally arise if smaller galaxies
156: have larger gas mass fractions than larger galaxies, i.e. simply because low mass
157: galaxies have been less efficient in forming stars and therefore are
158: less evolved.
159: Another more likely possibility - see \cite{tr2004} - is that more massive galaxies
160: can retain metals
161: more easily than less massive galaxies because of their deeper gravitational wells,
162: making them less susceptible to metal loss in galactic winds - \cite{ga2002}.
163: %Indeed the metallicity-luminosity relation of galaxies inferred from the SDSS supports
164: %galactic winds, rather than gas mass fraction, as the main regulator of the
165: %observed metallicity - \cite{tr2004}.
166:
167: % On the other hand some arguments have been given that ascribe the correlation
168: % to the effects of selective losses of heavy elements from galaxies in supernova-driven
169: % outflows. This relationship is expected to evolve with cosmic epoch,
170: % due in part to ``feedback'' from supernova driven winds or infall of new, unprocessed gas.
171: % The metallicity-luminosity relation of galaxies inferred from the SDSS
172: % is providing evidence for galactic winds - \cite{tr2004}. Clearly, this relation
173: % cannot be fit by a simple closed box models while the reduction in the effective yields
174: % (most likely due to galactic winds) can explain the observations (note that \cite{ga2002}
175: % has shown already that metal loss due to supernova-driven
176: % winds is a process at work in Irr and some spiral galaxies).
177:
178: Numerical simulations by \cite{ce1999} predicting the evolution of the
179: metal content of the Universe
180: % by incorporating star formation and its
181: %feedback on the IGM. At a given gas density (corresponding e.g. to a rich
182: %cluster, a disc galaxy, dwarf etc...) these models predict evolution
183: %with redshift, but most importantly,
184: show that metallicity is a stronger
185: function of density than age; moreover with a considerable scatter.
186: At low redshift, one would expect a few percent of the gas-rich dwarfs to
187: have metallicity on the order of IZw18, without invoking their youth.
188:
189:
190: %-Refer to the evolution in the cosmic SFR (e.g. Chary and Elbaz)
191:
192:
193: In the past, the scatter in the N/O versus O/H
194: diagram had been considered to be larger than the observational uncertainties.
195: Time delays between
196: the production of oxygen due to massive stars and that of nitrogen
197: were a likely part of the explanation although this point of view was
198: challenged by \cite{iz1999}.
199: Indeed their observations not only suggested a small intrinsic
200: dispersion of log N/O ($\pm 0.02$ dex) at low metallicities but a similar behaviour
201: was found for C/O and some other ratios. They concluded that
202: galaxies with such low abundances are genuinely young (less than 40 Myr old),
203: now making their first generation of stars. Moreover they claimed that
204: all galaxies with 7.6 $\le$ 12+log(O/H) $\le$ 8.2 have ages from 100 to 500 Myr.
205: %Thus, the question raised by \cite{ss1970} more than
206: %30 years ago would after all have a positive answer.
207: However, independent data such as the CMDs suggest that these galaxies do in fact contain
208: old stars and there are definitely many BCDs with 12+ $\log({\rm O/H}) < 8.2$
209: which have been demonstrated to be much older than 500 Myr
210: (see next section). Nevertheless, the scatter is
211: surprisingly small considering the
212: short time scale for the production of oxygen (as compared to iron
213: production) because different stellar masses are involved.
214: %If real,
215: % it would imply that Fe could have been produced by explosive nucleosynthesis of
216: %core collapse SNe in the early stage of chemically unevolved galaxies.
217: Now, more recent data using SDSS show a much larger increase of the scatter
218: at low (O/H) - \cite{iz2006} while other interpretations for the N/O behaviour are also possible.
219: A similar pattern is seen for N/O observations of H{\sc ii} regions in spirals - \cite{ze1998},
220: and old low surface brightness galaxies - \cite{ro1995}; \cite{be1999}, have N/O comparable
221: to those of the most metal-poor BCGs.
222:
223: \section{Ages of metal-poor galaxies}
224:
225: Colour-magnitude diagrams (CMDs) allows for the most direct determination
226: of the age of a stellar population. It is important to point out that in all
227: nearby galaxies where CMDs have been obtained to the depth required for a clear
228: detection of the red giant branch, has this also been found, showing
229: beyond doubt that these systems are not young
230: %({\bf reference, Schulte-Ladbeck?} NOT FOUND A RELEVANT REF).
231:
232: Following \cite{iz1999} a galaxy such as IZw18, with its record low
233: abundance (12+log(O/H)=7.18), should be genuinely young.
234: There are nevertheless independent evidence suggesting that IZw18 does
235: indeed host an old underlying population (see M. Tosi, this volume).
236: %IZw18 has been imaged in the optical with the HST/WFPC2 by two different
237: %groups - Hunter and Thronson 1995; \cite{du1996}, both finding a resolved
238: %population of young massive stars only. However, a re-analysis of these
239: %datasets combined (\cite{al1999}), and independent HST/NICMOS
240: %data (\cite{ostlin2000}), give support for an age in excess of 0.5 Gyr.
241: A careful analysis of archival HST/WFPC2 data (\cite{al1999}), as well
242: as independent HST/NICMOS observations (\cite{ostlin2000}, \cite{om2005}),
243: both give support for an age in excess of 0.5 Gyr.
244: % Using new, very sensitive, HST/ACS observations \cite{iz2004} failed
245: % to trace an old RGB in IZw18
246: %partly because the authors place this
247: %galaxy at a larger distance than customary, but also because their analysis
248: %was not performed deep enough.
249: %Indeed a re-analysis of
250: %the same HST/ACS data by \cite{mo2005} of I Zw 18 suggests that
251: %an older RGB population may be present, hence an age between 1--2 Gyr.
252: % Using new, very sensitive, HST/ACS observations \cite{iz2004} failed
253: % to trace an old RGB in IZw18
254: Using new, very sensitive, HST/ACS observations \cite{iz2004} failed
255: to trace an old RGB in IZw18, but a re-analysis of
256: the same data by \cite{mo2005} of IZw18 do suggest the presence of
257: an older RGB with an age of several Gyrs.
258:
259:
260: In cases where galaxies are too distant for a CMD to be obtained,
261: other ways have to be sought to estimate their age.
262: One such method involves using star clusters -
263: usually good representations of single stellar populations
264: whose ages can be determined with less degeneracies than
265: mixed stellar populations. Limitations are rather set by statistical effects
266: (in dwarf galaxies suitable massive
267: star clusters are rare). One example of a metal-poor galaxy whose
268: age can be tied to $\sim 10$ Gyr is the luminous compact galaxy
269: ESO\,338--04, a.k.a. Tol\,1924-416 (\"Ostlin et al. 1998).
270:
271: As most active star formation in metal-poor galaxies tends to
272: be centrally concentrated, a natural place to look for old stars
273: are the outskirts, or ``haloes'' of e.g. BCDs by comparing
274: optical and infrared colours to spectral synthesis models
275: (e.g. Bergvall \& \"Ostlin 2002). The further out
276: one looks, the smaller the contamination from young stars, but also
277: lesser is the light to analyse. A major problem is that of
278: wide spread ionised gas.
279: %, most likely ionised by
280: %photons originating from the central star forming regions.
281: In addition, the
282: ionised gas may have a different scale length from the stellar
283: population (Papaderos et al. 2002), further complicating the analysis.
284:
285: With good quality data it is even
286: possible to date the old population in the centra of star
287: forming galaxies, e.g. as has been performed with some SDSS spectra
288: of star forming dwarfs (Corbin et al. 2006).
289:
290: \section{Outlook}
291:
292: Our discussion lead us to conclude that no compelling observational evidence
293: for any young local galaxy has been provided so far. Moreover metallicity is
294: not an arrow of time, but a product of the past star formation history, combined
295: with the mass-loss and accretion history. Only few local galaxies
296: remain under debate but this is likely because
297: they remain beyond sensitivity thresholds!
298: With JWST we can probe a little further, but not much, e.g. getting a CMD
299: for SBS\,0335-052 will still be a tough task.
300: Do we at all expect young galaxies in our Local Universe?
301: %We expect galaxies to go at their own pace
302: %(this is what downsizing
303: %suggests) and density rather than metallicity
304: % seems to be the dominant factor.
305: Could some HI clouds that survived reionisation start to collapse now?
306: Then they need to have been unable to form any stars early on but still
307: be masssive enough not to boil off completely during
308: reionisation. It appears unlikely that both requirements can be fulfilled.
309:
310: \begin{acknowledgments}
311: We wrote this paper for money.
312: \end{acknowledgments}
313:
314: \begin{thebibliography}{}
315:
316:
317: \bibitem[Aloisi et al.(1999)]{al1999}
318: {Aloisi, A., Tosi, M., \& Greggio, L.}, 1999,
319: \textit{AJ}, 118, 302
320:
321: \bibitem[Bergvall \& \"Ostlin (2002)]{be2002}
322: {Bergvall, N., \"Ostlin, G.}, 1999,
323: \textit{A\&A}, 390, 891
324:
325: \bibitem[Bergvall et al. (1999)]{be1999}
326: {Bergvall, N., R"onnback, J., Masegosa, J., \"Ostlin, G.}, 1999,
327: \textit{A\&A}, 341, 697
328:
329:
330: \bibitem[Cannon et al. (2005)]{ca2005}
331: {Cannon, J.~M., Skillman, E.~D., Sembach, K.~R.,
332: \& Bomans, D.~J.}, 2005,
333: \textit{ApJ}, 618, 247
334:
335: \bibitem[Cen \& Ostriker (1999)]{ce1999}
336: {Cen R., Ostriker J.P.}, 1999,
337: \textit{ApJ}, 519, L109
338:
339: \bibitem[Corbin et al. (2006)]{corbin}
340: {Corbin M., Vacca W., Cid Fernandes R., et al.}, 2006,
341: \textit{ApJ}, in press.
342:
343: \bibitem[Dufour et al.(1996)]{du1996}
344: {Dufour, R.~J., Garnett, D.~R., Skillman, E.~D., \& Shields, G.~A.}, 1996,
345: \textit{Science with the Hubble Space Telescope} - II, 348
346:
347:
348: \bibitem[Garnett (2002)]{ga2002}
349: {Garnett, D.~R.}, 2002,
350: \textit{ApJ}, 581, 1019
351:
352: \bibitem[Grebel et al. (2003)]{gr2003}
353: {Grebel, E.~K., Gallagher, J.~S., III, \& Harbeck, D.}, 2003,
354: \textit{AJ}, 125, 1926
355:
356: \bibitem[Hunter \& Thronson(1995)]{hu1995}
357: {Hunter, D.~A., \& Thronson, H.~A., Jr.}, 1995,
358: \textit{ApJ}, 452, 238
359:
360: \bibitem[Izotov \& Thuan (1999)]{iz1999}
361: {Izotov, Y.~I., \& Thuan, T.~X.}, 1999,
362: \textit{ApJ}, 511, 639
363:
364: \bibitem[Izotov \& Thuan(2004)]{iz2004}
365: {Izotov, Y.~I., \& Thuan, T.~X.}, 2004,
366: \textit{ApJ}, 616, 768
367:
368: \bibitem[Izotov et al.(2006)]{iz2006}
369: {Izotov, Y.~I., Stasi{\'n}ska, G., Meynet, G., Guseva, N.~G., \& Thuan, T.~X.}, 2006,
370: \textit{ApJ}, 448, 955
371:
372:
373: \bibitem[Kunth \& \"Ostlin (2000)]{ko2000}
374: {Kunth D., \"Ostlin G.}, 2000,
375: \textit{A\&ARv}, 10, 1
376:
377: \bibitem[Kunth \& \"Ostlin (2001)]{ko2001}
378: {Kunth D., \"Ostlin G.}, 2001,
379: \textit{ApSSS }, 277, 281
380:
381: \bibitem[Kunth \& Sargent (1986)]{ku1986}
382: {Kunth, D., \& Sargent, W.~L.~W.}, 1986,
383: \textit{ApJ}, 300, 496
384:
385: \bibitem[Lebouteiller et al.(2006)]{le2006}
386: {Lebouteiller, V., Kunth, D., Lequeux, J., Aloisi, A., Desert,
387: J.~-., Hebrard, G., Lecavelier des Etangs, A., \& Vidal-Madjar, A.}, 2006,
388: \textit{Astrophysics e-prints}, astro-ph/0608445
389:
390: \bibitem[Lequeux et al. (1981)]{le1981}
391: {Lequeux, J., Maucherat-Joubert, M., Deharveng, J.~M., \& Kunth, D.}, 1981,
392: \textit{A\&A}, 103, 305
393:
394: \bibitem[Lequeux et al. (1979)]{le1979}
395: {Lequeux, J., Peimbert, M.; Rayo, J. F.; Serrano, A.; Torres-Peimbert} 1979,
396: \textit{A\&A}, 80, 155
397:
398: \bibitem[Momany et al.(2005)]{mo2005}
399: {Momany, Y., et al.}, 2005,
400: \textit{ApJ}, 439, 111
401:
402: \bibitem[\"Ostlin et al. (1998)]{os1998}
403: {\"Ostlin, G., Bergvall, N., \& Roennback, J.}, 1998,
404: \textit{A\&A}, 335, 85
405:
406: \bibitem[\"Ostlin (2000)]{ostlin2000}
407: { \"Ostlin G.}, 2000,
408: \textit{ApJ}, 535, L99
409:
410: \bibitem[\"Ostlin \& Mouhcine (2005)]{om2005}
411: {\"Ostlin G., Mouhcine, M.}, 2005,
412: \textit{A\&A}, 433, 979
413:
414: \bibitem[Papaderos et al. (2002)]{papa}
415: {Papaderos P., Izotov Y., Thuan T., et al.}, 2002,
416: \textit{A\&A}, 393, 461
417:
418: \bibitem[Pettini(2006)]{pe2006}
419: {Pettini, M.}, 2006,
420: \textit{ASP Conf.~Ser.}, 353, 363
421:
422: \bibitem[R\"onnback \& Bergvall (1995)]{ro1995}
423: {R\"onnback, J., \& Bergvall, N.}, 1995,
424: \textit{A\&A}, 302, 353
425:
426: \bibitem[Sargent \& Searle (1970)]{ss1970}
427: {Sargent, W.~L.~W., \& Searle, L.}, 1970,
428: \textit {ApJ}, 162, L155
429:
430: \bibitem[Tremonti et al. (2004)]{tr2004}
431: {Tremonti, C.~A., et al.}, 2004,
432: \textit{ApJ}, 613, 898
433:
434: \bibitem[van Zee et al.(1998)]{ze1998}
435: {van Zee, L., Salzer, J.~J., \& Haynes, M.~P.}, 1998,
436: \textit {ApJ}, 497, L1
437:
438:
439: \end{thebibliography}
440:
441:
442:
443: \end{document}
444: