DOI: 10.1126/science.1209850, 952 (2011);334 Science

, et al.Todd M. TrippOutflowThe Hidden Mass and Large Spatial Extent of a Post-Starburst Galaxy

This copy is for your personal, non-commercial use only.

clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

here.following the guidelines

can be obtained byPermission to republish or repurpose articles or portions of articles

): November 27, 2011 www.sciencemag.org (this infomation is current as of

The following resources related to this article are available online at

http://www.sciencemag.org/content/334/6058/952.full.htmlversion of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/content/suppl/2011/11/16/334.6058.952.DC1.html can be found at: Supporting Online Material

http://www.sciencemag.org/content/334/6058/952.full.html#relatedfound at:

can berelated to this article A list of selected additional articles on the Science Web sites

http://www.sciencemag.org/content/334/6058/952.full.html#ref-list-1, 1 of which can be accessed free:cites 46 articlesThis article

http://www.sciencemag.org/content/334/6058/952.full.html#related-urls1 articles hosted by HighWire Press; see:cited by This article has been

http://www.sciencemag.org/cgi/collection/astronomyAstronomy

subject collections:This article appears in the following

registered trademark of AAAS. is aScience2011 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on

Nov

embe

r 27

, 201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

10. D. Keres, N. Katz, D. H. Weinberg, R. Davé, Mon. Not. R.Astron. Soc. 363, 2 (2005).

11. A. Dekel, Y. Birnboim, Mon. Not. R. Astron. Soc. 368,2 (2006).

12. G. Kauffmann et al., Mon. Not. R. Astron. Soc. 341,33 (2003).

13. T. M. Tripp, B. D. Savage, E. B. Jenkins, Astrophys. J.534, L1 (2000).

14. C. W. Danforth, J. M. Shull, Astrophys. J. 679, 194(2008).

15. C. Thom, H.-W. Chen, Astrophys. J. 683, 22 (2008).16. J. N. Bregman, Annu. Rev. Astron. Astrophys. 45,

221 (2007).17. J. T. Stocke et al., Astrophys. J. 641, 217 (2006).18. H.-W. Chen, J. S. Mulchaey, Astrophys. J. 701, 1219

(2009).19. J. X. Prochaska, B. Weiner, H.-W. Chen, J. S. Mulchaey,

K. L. Cooksey, http://arxiv.org/abs/1103.1891 (2011).20. K. R. Sembach et al., Astrophys. J. Suppl. Ser. 146,

165 (2003).21. See supporting material on Science Online.22. J. K. Werk et al., http://arxiv.org/abs/1108.3852

(2011).23. D. Schiminovich et al., Astrophys. J. Suppl. Ser. 173,

315 (2007).24. The typical log NOVI = 14.5 to 15.0 for star-forming

galaxies resembles the high end of the column-densitydistribution seen in blind surveys of intergalactic clouds

(14, 15, 38) and is higher than the mean value (14.0)measured by the Far Ultraviolet Spectroscopic Explorerthrough the halo of the Milky Way (20), which wouldbelong in our star-forming sample.

25. O VI emission is seen in elliptical galaxies (39), butthis gas is most likely associated with the ISM and notthe CGM.

26. M. Asplund, N. Grevesse, A. J. Sauval, P. Scott, Annu. Rev.Astron. Astrophys. 47, 481 (2009).

27. M. S. Peeples, F. Shankar, Mon. Not. R. Astron. Soc. 417,2962 (2011).

28. M. E. Putman, Astrophys. J. 645, 1164 (2006).29. N. Lehner, J. C. Howk, Science 334, 955 (2011);

10.1126/science.1209069.30. H.-W. Chen et al., Astrophys. J. 714, 1521 (2010).31. C. C. Steidel et al., Astrophys. J. 717, 289 (2010).32. D. Thomas, L. Greggio, R. Bender, Mon. Not. R.

Astron. Soc. 296, 119 (1998).33. T. M. Tripp et al., Science 334, 952 (2011).34. B. D. Oppenheimer et al., Mon. Not. R. Astron. Soc. 406,

2325 (2010).35. K. R. Stewart et al., Astrophys. J. 735, L1 (2011).36. M. Fumagalli et al., http://arxiv.org/abs/1103.2130

(2011).37. J. M. Gabor, R. Davé, K. Finlator, B. D. Oppenheimer,

Mon. Not. R. Astron. Soc. 407, 749 (2010).38. T. M. Tripp et al., Astrophys. J. Suppl. Ser. 177,

39 (2008).

39. J. N. Bregman, E. D. Miller, A. E. Athey, J. A. Irwin,Astrophys. J. 635, 1031 (2005).

Acknowledgments: We thank the anonymous reviewersfor constructive comments. This work is based onobservations made for program GO11598 with theNASA/ESA Hubble Space Telescope, obtained at theSpace Telescope Science Institute, operated by AURAunder NASA contract NAS 5-26555, and at theW. M. Keck Observatory, operated as a scientificpartnership of the California Institute of Technology, theUniversity of California, and NASA. The Observatory wasmade possible by the generous financial support of theW. M. Keck Foundation. The Hubble data are availablefrom the MAST archive at http://archive.stsci.edu.M.S.P. was supported by the Southern California Centerfor Galaxy Evolution, a multicampus research programfunded by the UC Office of Research.

Supporting Online Materialwww.sciencemag.org/cgi/content/full/334/6058/948/DC1SOM TextFigs. S1 to S5Tables S1 and S2References (40–62)

15 June 2011; accepted 27 September 201110.1126/science.1209840

The Hidden Mass and Large SpatialExtent of a Post-Starburst Galaxy OutflowTodd M. Tripp,1* Joseph D. Meiring,1 J. Xavier Prochaska,2 Christopher N. A. Willmer,3

J. Christopher Howk,4 Jessica K. Werk,2 Edward B. Jenkins,5 David V. Bowen,5 Nicolas Lehner,4

Kenneth R. Sembach,6 Christopher Thom,6 Jason Tumlinson6

Outflowing winds of multiphase plasma have been proposed to regulate the buildup of galaxies,but key aspects of these outflows have not been probed with observations. By using ultravioletabsorption spectroscopy, we show that “warm-hot” plasma at 105.5 kelvin contains 10 to 150 timesmore mass than the cold gas in a post-starburst galaxy wind. This wind extends to distances > 68kiloparsecs, and at least some portion of it will escape. Moreover, the kinematical correlation ofthe cold and warm-hot phases indicates that the warm-hot plasma is related to the interaction ofthe cold matter with a hotter (unseen) phase at >>106 kelvin. Such multiphase winds canremove substantial masses and alter the evolution of post-starburst galaxies.

Galaxies do not evolve in isolation. They in-teract with other galaxies and,more subtly,with the gas in their immediate environ-

ments. Mergers of comparable-mass, gas-richgalaxies trigger star-formation bursts by drivingmatter into galaxy centers, but theory predicts thatsuch starbursts are short-lived: The central gas israpidly driven away by escaping galactic windspowered by massive stars and supernova explo-sions or by a central supermassive black hole(1). Such feedback mechanisms could trans-form gas-rich spiral galaxies into post-starburst

galaxies (2) and eventually into elliptical-typegalaxies with little or no star formation (3).Mergers are not required to propel galaxy evo-lution, however. Even relatively secluded galaxiesaccrete matter from the intergalactic medium(IGM), form stars, and drive matter outflows intotheir halos or out of the galaxies entirely (4, 5).In either case, the competing processes of gasinflows and outflows are expected to regulategalaxy evolution.

Outflows are evident in some nearby objects(6–9) and are ubiquitous in some types of gal-axies (10–15); their speeds can exceed the escapevelocity. Nevertheless, their broader impact ongalaxy evolution is poorly understood. First, theirfull spatial extent is unknown. Previous studies(6, 9, 16–22) have revealed flows with spatialextents ranging from a few parsecs up to ~20 kilo-parsecs (kpc). However, because of their lowdensities, outer regions of outflows may not havebeen detected with previously used techniques,and thus the flows could be much larger. Second,

the total column density andmass of the outflowsare poorly constrained. Previous outflow obser-vations were often limited to low-resolution spec-tra of only one or two ions (e.g., Na I or Mg II) orrelied on composite spectra that cannot yield precisecolumn densities.Without any constraints on hydro-gen (the vast bulk of the mass) or other elementsand ions, these studieswere forced tomake highlyuncertain assumptions to correct for ionization,elemental abundances, and depletion of speciesby dust. Lastly, galactic winds contain multiplephases with a broad range of physical conditions(6), and wind gas in the key temperature rangebetween 105 to 106 K (where radiative cooling ismaximized) is too cool to be observed in x-rays;detection of this so-called “warm-hot” phaserequires observations in the ultraviolet (UV).

To study the more extended gas around gal-axies, including regions affected by outflows, weused the Cosmic Origins Spectrograph (COS)on the Hubble Space Telescope (HST) to obtainhigh-resolution spectra of the quasi-stellar object(QSO) PG1206+459 (at redshift zQSO = 1.1625).By exploiting absorption lines imprinted on theQSO spectrum by foreground gaseous material,we can detect the low-density outer gaseous en-velopes of galaxies, regions inaccessible to othertechniques. We focus on far-ultraviolet (FUV) ab-sorption lines at rest wavelengths lrest < 912 Å.This FUV wavelength range is rich in diagnostictransitions (23), including the Ne VIII 770.409,780.324 Å doublet, a robust probe of warm-hotgas, as well as banks of adjacent ionization stages.The sight line to PG1206+459 pierces an absorp-tion system, at redshift zabs = 0.927, that providesinsights about galactic outflows. This absorberhas been studied before (24), but previous obser-vations did not cover Ne VIII and could not pro-vide accurate constraints on H I in the individualabsorption components.

1Department of Astronomy, University of Massachusetts, Am-herst, MA 01003, USA. 2University of California Observatories/Lick Observatory, University of California, Santa Cruz, CA 95064,USA. 3Steward Observatory, University of Arizona, Tucson, AZ85721, USA. 4Department of Physics, University of Notre Dame,Notre Dame, IN 46556, USA. 5Princeton University Obser-vatory, Princeton, NJ 08544, USA. 6Space Telescope ScienceInstitute, Baltimore, MD 21218, USA.

*To whom correspondence should be addressed. E-mail:tripp@astro.umass.edu

18 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org952

REPORTS

on

Nov

embe

r 27

, 201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

This absorber is illustrated in Figs. 1 to 3, in-cluding the COS data (25). The absorber is a“partial” Lyman limit (LL) system (i.e., the high-

er Lyman series lines are not saturated), whichenables accurate H I column density [N(H I)] mea-surement (Fig. 1). Awide variety of metals and

H I lines were detected in at least nine compo-nents (25) spanning a large velocity range from−317 to +1131 km s−1 (Figs. 1 and 2). TheNeVIIIdoublet was unambiguously detected (Fig. 2)with a totalN(Ne VIII) = 1014.9 cm−2 (25), whichis ~10 times higher than any previous N(Ne VIII)measurements in intervening absorbers (26, 27).The component at +1131 km s−1 exceeds vescapeof any individual galaxy, and the other compo-nents have very similar properties to the +1131km s−1 component (25), suggesting a commonorigin. Whether the other components have v >vescape depends on the (unknown) potential well,but allowing for projection effects and notingthat the gas is already far from the affiliated gal-axy (see below), several of the other componentscould also be escaping. Combined with detec-tion of Ne VIII, the detections of banks of adjacentions (N II, N III, N IV,NV;O III, O IV; S III, S IV,S V) place tight constraints on physical condi-tions of the gas. Notably, the velocity centroidsand profile shapes of lower and higher ioniza-tion stages are quite similar (Fig. 3).

This strong Ne VIII/LL absorber is affiliatedwith a galaxy near the QSO sight line (24, 25).This galaxy, which we refer to as 177_9, is thetype of galaxy expected to drive a galactic su-perwind (Fig. 4). Like post-starburst (11) andultraluminous infrared galaxies (28), galaxy 177_9is very luminous and blue (29); based on thecharacteristic magnitude (M*) of the z ~ 1 lumi-nosity function from the Deep EvolutionaryExploratory Probe 2 (DEEP2) (30), the galaxyluminosity L = 1.8 L*. The Multiple Mirror Tel-escope (MMT) spectrum in Fig. 4 is also similarto those of the post-starburst galaxies in (11),with higher Balmer series absorption lines, [O II]emission and [Ne V] emission indicative of anactive galactic nucleus (AGN) (25). Most impor-tantly, the galaxy has a large impact parameterfrom the QSO sight line, r = 68 kpc (31), whichimplies that the gaseous envelope of 177_9 hasa large spatial extent.

The component-to-component similarity ofthe absorption lines (Fig. 3) suggests a relatedorigin. To further investigate the nature of thisabsorber, we used photoionizationmodels (32) toderive ionization corrections and elemental abun-dances (25). These models indicate that the indi-vidual components have high abundances rangingfrom ~0.5 to 3 times those in the Sun (table S2).Such high abundances (or metallicities) favor anorigin in outflowing ejecta enriched by nucleo-synthesis products from stars; at the large impactparameter of 177_9, corotating outer-disk or halogas or tidal debris from a low-mass satellite gal-axy would be expected to have much lower me-tallicity. Tidal debris from a massive galaxy couldhave high metallicity, but we are currently awareof only one luminous galaxy near the sight line atthe absorber redshift (33); another luminous gal-axy interacting with 177_9 is not evident. Theabsorber could also be intragroup gas, but some-how it must have been metal-enriched, so sometype of galactic outflow is implicated in any case.

Fig. 2. Continuum-normalized absorption profiles (black lines) of various species detected in the LL /Mg IIabsorber shown in Fig. 1, plotted in velocity with respect to the galaxy 177_9 redshift (i.e., v = 0 km s−1 atz = 0.927). Labels below each absorption profile indicate the species and rest wavelength. We fitted ninecomponents to the COS and Space Telescope Imaging Spectrograph data (24). Component centroids areindicated by gray lines, and the Voigt-profile fits are overplotted with red lines (25). Yellow lines indicatecontaminating features from other redshifts or transitions. The two graphs at lower left compare apparentcolumn density profiles (39) of the N V and Ne VIII doublets.

Fig. 1. (Top) Small portion of the Keck HIRES spectrum of PG1206+459 (24). Tick marks at top indicatecomponents detected at various velocities in the Mg II 2803.53 Å transition. A velocity scale in the restframe of the affiliated galaxy 177_9 is inset at bottom. Gray indicates a feature not due to Mg II 2803.53 Å.(Bottom) Small portion of the ultraviolet spectrum of PG1206+459 recorded with the COS on HST thatshows H I Lyman series absorption lines (marked with ticks and labels) at the redshift of the Mg II complex inthe top graph, including H I Lyz through Lys (highest lines are marked but not labeled).

www.sciencemag.org SCIENCE VOL 334 18 NOVEMBER 2011 953

REPORTS

on

Nov

embe

r 27

, 201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

The photoionization models also constrainthe total hydrogen column (i.e., H I and H II),and, combined with spatial extent ≥ 68 kpc, thisallows mass estimates. By using fiducial thin-shell models (25), we find that the mass of cool,photoionized gas in individual componentsranges from 0.6 × 108 to 14 × 108 solar masses(M). However, photoionization fails (sometimesby orders of magnitude) to produce enough S V,N V, and Ne VIII; these species must arise in hotgas at temperature T > 105 K. By using equi-librium and nonequilibrium collision ionizationmodels (25), we find that the warm-hot gas con-tains much more mass than the cold gas, withindividual components harboring 10 × 108 to400 × 108 M in hot material. These are roughestimates with many uncertainties. For example,if the absorption arises in thin filaments analo-gous to those seen in starburst galaxies (6) orAGN bubbles (34), the cold-gas mass couldreduce to ~106 M per component. However, asin the thin-shell models, the warm-hot gas couldharbor 10 to 150 times more mass in suchfilaments (25). In either case (shells or filaments),given the similarity of the cold and warm-hotabsorption lines (Fig. 3), the Ne VIII–bearingplasma must be a transitional phase that links thecolder and hotter material and thus providesinsights on the outflow physics. The NeVIII/NVphase is not photoionized, so it must be generatedthrough interaction of the cold gas with a hotterambient medium analogous to x-ray–emittingregions seen in nearby galaxies. How this occursis an open question; the absorbers could bematerial cooling from the hot phase down to thecool gas, or the cool clouds could have a hotterand more-ionized surface that is evaporating.

Low-density plasma in the T = 105 to 106 Krange has been effectively hidden from most

Fig. 3. Comparison of apparent column density profiles (39) of the LL absorber affiliated withgalaxy 177_9. In each graph, the C II 687.05 Å profile (black histogram) is compared to anotherspecies (colored circles) as labeled at upper left; the comparison species profile is also scaled by thefactor in parentheses after the species label. Gray lines indicate regions contaminated by unrelatedabsorption. As in Fig. 2, v = 0 km s−1 at z = 0.927.

Fig. 4. Montage of observations of the galaxy atzgal = 0.927 that drives a large-scale outflow ofmetal-enriched plasma. (Top left) The galaxy, andthe background QSO that reveals the outflow viaabsorption spectroscopy, is shown in a multicolorimage obtained with the Large Binocular Telescope.This galaxy, which we refer to as 177_9, is the redobject 8.63 arc sec south of the bright QSO PG1206+459 (zQSO = 1.1625) at a position angle of 177° (Nthrough E) from the QSO. At the galaxy redshift, theangular separation from the QSO sight line corre-sponds to an impact parameter of 68 kpc. (Top right)The large red circle indicates the rest-frame U-Bcolor and absolute Bmagnitude of 177_9 comparedto all galaxies from the DEEP2 survey (gray scale)(30) and DEEP2 galaxies within T0.05 of z(177_9)(cyan points). The small purple circles show post-starburst galaxies from (11). (Bottom) An MMToptical spectrum of 177_9 (upper trace) with its1s uncertainty (lower trace). The strong featureat ≈ 7600 Å is partially due to telluric absorption.

18 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org954

REPORTS

on

Nov

embe

r 27

, 201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

outflow studies. In principle, theOVI 1032,1038Ådoublet can reveal such gas, but it is unclearwhether the O VI arises in photoionized 104 Kgas, hotter material at ~105.5 K, or both (35). TheNe VIII doublet avoids this ambiguity, and wehave found that this warm-hot matter is a sub-stantial component in the mass inventory of agalactic wind. Moreover, this wind has a largespatial extent, and the mass carried away by theoutflow will affect the evolution of the galaxy.Whereas earlier studies of poststarburst outflowsfocused on Mg II and could not precisely con-strain the metallicity, hydrogen column, andmass, these studies do indicate that post-starburstoutflows are common: 22/35 of the post-starburstsin (36) showed outflowingMg II absorption withmaximum (radial) velocities of 500 to 2400 km s−1,similar to the absorption near 177_9 (Fig. 1), and77 and 100% of the post-starburst and AGNgalaxies, respectively, in (37) drive outflows butwith lower maximum velocities, which may bedue to selection of wind-driving galaxies in alater evolutionary stage. With existing COS data,the effects of large-scale outflows on galaxy evo-lution can be studied with the techniques pre-sented here but with larger samples (38), withwhich it will be possible to statistically track howoutflows affect galaxies.

References and Notes1. P. F. Hopkins et al., Astrophys. J. Suppl. Ser. 163,

1 (2006).2. A. I. Zabludoff et al., Astrophys. J. 466, 104 (1996).3. G. F. Snyder, T. J. Cox, C. C. Hayward, L. Hernquist,

P. Jonsson, Astrophys. J. 741, 77 (2011).4. D. Kereš, N. Katz, R. Davé, M. Fardal, D. H. Weinberg,

Mon. Not. R. Astron. Soc. 396, 2332 (2009).5. B. D. Oppenheimer et al., Mon. Not. R. Astron. Soc. 406,

2325 (2010).6. S. Veilleux, G. Cecil, J. Bland-Hawthorn, Annu. Rev.

Astron. Astrophys. 43, 769 (2005).7. T. M. Heckman, M. D. Lehnert, D. K. Strickland, L. Armus,

Astrophys. J. Suppl. Ser. 129, 493 (2000).

8. D. S. Rupke, S. Veilleux, D. B. Sanders, Astrophys. J.Suppl. Ser. 160, 87 (2005).

9. C. L. Martin, Astrophys. J. 647, 222 (2006).10. M. Pettini et al., Astrophys. J. 554, 981 (2001).11. C. A. Tremonti, J. Moustakas, A. M. Diamond-Stanic,

Astrophys. J. 663, L77 (2007).12. C. C. Steidel et al., Astrophys. J. 717, 289 (2010).13. K. H. R. Rubin et al., Astrophys. J. 719, 1503 (2010).14. F. Hamann, G. Ferland, Annu. Rev. Astron. Astrophys. 37,

487 (1999).15. J. P. Grimes et al., Astrophys. J. Suppl. Ser. 181, 272 (2009).16. K. H. R. Rubin, J. X. Prochaska, D. C. Koo, A. C. Phillips,

B. J. Weiner, Astrophys. J. 712, 574 (2010).17. M. Moe, N. Arav, M. A. Bautista, K. T. Korista,

Astrophys. J. 706, 525 (2009).18. J. P. Dunn et al., Astrophys. J. 709, 611 (2010).19. D. Edmonds et al., Astrophys. J. 739, 7 (2011).20. F. Hamann et al., Mon. Not. R. Astron. Soc. 410, 1957

(2011).21. G. A. Kriss et al., Astron. Astrophys. 534, 41 (2011).22. D. M. Capellupo, F. Hamann, J. C. Shields, P. Rodríguez

Hidalgo, T. Barlow, Mon. Not. R. Astron. Soc. 413, 908(2011).

23. D. A. Verner, D. Tytler, P. D. Barthel, Astrophys. J. 430,186 (1994).

24. J. Ding, J. C. Charlton, C. W. Churchill, C. Palma,Astrophys. J. 590, 746 (2003).

25. See further information in supporting material on ScienceOnline.

26. B. D. Savage, N. Lehner, B. P. Wakker, K. R. Sembach,T. M. Tripp, Astrophys. J. 626, 776 (2005).

27. A. Narayanan et al., Astrophys. J. 730, 15 (2011).28. Y. Chen, J. D. Lowenthal, M. S. Yun, Astrophys. J. 712,

1385 (2010).29. The galaxy appears to be red in Fig. 4 because of its

redshift; in the rest frame of the galaxy, it has a veryultraviolet-blue (U-B) color.

30. C. N. A. Willmer et al., Astrophys. J. 647, 853 (2006).31. For distance calculations, we assume a cold dark matter

cosmology with Hubble constant H0 = 70 km s−1 Mpc−1

and dimensionless density parameters Ωm = 0.30, ΩL =0.70.

32. G. J. Ferland et al., Publ. Astron. Soc. Pac. 110, 761(1998).

33. As discussed in the supporting online material, the yellowgalaxy northwest of the QSO (Fig. 4) does not have aspectroscopic redshift but is likely to have z << 0.927.

34. A. C. Fabian et al., Nature 454, 968 (2008).35. T. M. Tripp et al., Astrophys. J. Suppl. Ser. 177, 39

(2008).

36. C. Tremonti, A. M. Diamond-Stanic, J. Moustakas, inGalaxy Evolution: Emerging Insights and FutureChallenges, S. Jogee, I. Marinova, L. Hao, G. Blanc, Eds,(Astronomical Society of the Pacific Conference Series,San Francisco, 2009), vol. 419, pp. 369–376.

37. A. L. Coil et al., http://arXiv.org/abs/1104.0681(2011).

38. J. Tumlinson et al., Science 334, 948 (2011).39. B. D. Savage, K. R. Sembach, Astrophys. J. 379, 245

(1991).Acknowledgments: This study has its basis in observations

made with the NASA/European Space Agency HubbleSpace Telescope (HST); the MMT, a joint facilityoperated by the Smithsonian Astrophysical Observatoryand the University of Arizona; and the Large BinocularTelescope, an international collaboration amonginstitutions in the United States, Italy, and Germany.Support for HST program number 11741 was providedby NASA through a grant from the Space TelescopeScience Institute, which is operated by the Associationof Universities for Research in Astronomy, Incorporated,under NASA contract NAS5-26555. Additional supportwas provided by NASA grant NNX08AJ44G. The DEEP2survey was supported by NSF grants AST 95-29098,00-711098, 05-07483, 08-08133, 00-71048,05-07428, and 08-07630. Funding for the Sloan DigitalSky Survey has been provided by the Alfred P. SloanFoundation, the Participating Institutions, NASA, NSF, theU.S. Department of Energy Office of Science, theJapanese Monbukagakusho, and the Max Planck Society.We thank C. Churchill for providing the archival Keckdata and the referees for review comments thatsignificantly improved this paper. We are also grateful tothe Hawaiian people for graciously allowing us to conductobservations from Mauna Kea, a revered place inthe culture of Hawaii. The HST data in this paperare available from the Multimission Archive at theSpace Telescope Science Institute (MAST) athttp://archive.stsci.edu.

Supporting Online Materialwww.sciencemag.org/cgi/content/full/334/6058/952/DC1Materials and MethodsSOM TextFigs. S1 to S5Tables S1 and S2References (40–54)

15 June 2011; accepted 26 October 201110.1126/science.1209850

A Reservoir of Ionized Gas in theGalactic Halo to Sustain StarFormation in the Milky WayNicolas Lehner* and J. Christopher Howk

Without a source of new gas, our Galaxy would exhaust its supply of gas through the formationof stars. Ionized gas clouds observed at high velocity may be a reservoir of such gas, but theirdistances are key for placing them in the galactic halo and unraveling their role. We have usedthe Hubble Space Telescope to blindly search for ionized high-velocity clouds (iHVCs) in theforeground of galactic stars. We show that iHVCs with 90 ≤ |vLSR| ≲ 170 kilometers per second(where vLSR is the velocity in the local standard of rest frame) are within one galactic radius of theSun and have enough mass to maintain star formation, whereas iHVCs with |vLSR| ≳ 170 kilometersper second are at larger distances. These may be the next wave of infalling material.

The time scale for gas consumption via starformation in spiral galaxies is far shorterthan a Hubble time (13.8 billion years),

requiring an ongoing replenishment of the gas-

eous fuel in the disks of galaxies for continuedstar formation. Analytical models and hydrody-namical simulations have emphasized the impor-tance of cold-stream accretion as a means for

metal-poor gas (metallicity that is less than 10%of that of the Sun, or Z ≲ 0.1 Z) to flow ontogalaxies along dense intergalactic filaments (1).However, galaxies may also exchange mass withthe local intergalactic medium (IGM) throughoutflows driven by galactic “feedback,” galacticwinds powered by massive stars and their deathand from massive black holes. Some of this ma-terial may return to the central galaxy as recycledinfalling matter—the galactic fountain mechanism(2, 3). The circumgalactic medium about a gal-axy is thus a complicated blend of outflowingmetal-rich and infalling metal-poor gas. The rela-tive importance of these processes is poorly con-strained observationally. Here, we demonstratethat ionized gas in the local galactic halo providesa major supply of matter for fueling ongoing starformation.

Department of Physics, University of Notre Dame, 225Nieuwland Science Hall, Notre Dame, IN 46556, USA.

*To whom correspondence should be addressed. E-mail:nlehner@nd.edu

www.sciencemag.org SCIENCE VOL 334 18 NOVEMBER 2011 955

REPORTS

on

Nov

embe

r 27

, 201

1w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from