当前位置：文档库 › Mg II Absorbing Galaxies Morphologies and Kinematics

Mg II Absorbing Galaxies Morphologies and Kinematics

a r X i v :a s t

r o

Workshop on Extraplanar Gas ASP Conference Series,Vol.000,2004ed.Robert Braun Mg II Absorbing Galaxies:Morphologies and Kinematics Chris Churchill 1,Chuck Steidel 2,and Glenn Kacprzak 11Department of Astronomy,New Mexico State University,MSC 4500,P.O.Box 30001,Las Cruces,NM,880032Department of Astronomy,Caltech,MS 105-24,Pasadena,CA 91125Abstract.In this contribution,we review our current knowledge of the properties of galaxies,and their extended halos,selected by Mg ii absorption in the spectra of background quasars.We then describe recent e?orts to quantify the morphologies and orientations of galaxies and explore how these relate to the gas kinematics.In a sample of 26galaxies,we ?nd no clear connection between the orientation of the quasar line of sight through the galaxy and the velocity spread of the gas.However,it appears that the quantity of gas “stirred up”in the halo may be correlated to asymmetry in the galaxy morphology.Since the galaxies have fairly normal morphologies,this connection may suggest that galaxies with extended halos experienced an interaction or merging event a few dynamical times prior to the epoch of observation.1.Quasar Absorption Lines:Further and Farther From the study of the Galaxy and local galaxies,the disk/halo interface is now known to be a dynamic region,probably built in response to star forming regions and the infall of “high velocity clouds”(HVCs).Using sensitive 21–cm emission line imaging of local galaxies,several contributers of this meeting have reported himneys,fountains,superbubbles,and H i holes (see contributions throughout this volume).Extending somewhat further into the halo are the so-called “H i beards”(Sancisi et al.2001)and “forbidden gas”complexes (Fraternali et al.2002).Beards are cold neutral gas structures several kpc from the disks that

depart from galaxy rotation by as much as ～50–100km s ?1;they are suggestive of a “lagging halo”.Forbidden gas structures have kinematics that run counter to the galaxy.The di?use interstellar gas (DIGs)in local galaxies is seen out to ～10kpc above the galactic disks (e.g.,Swaters et al.1997;Rand 2000).DIG gas often exhibits decreasing rotational velocities with height above the disk (i.e.,“halo lag”).Simple fountain ?ow models are consistent with this behavior (Collins et al.2002).

In order to make greater sense of these observations and ultimately of the role of gas in galaxy evolution,it would be desirable to build an observational bridge connecting slightly earlier cosmic epochs to the present.It would also be desirable to probe further out into the galactic halos in order to ?rmly establish the nature of gas far from galaxy disks.The technique of quasar absorption lines is uniquely suited to the task;it provides a natural window on the gaseous conditions in galaxies over a broad range of redshifts (cosmic epochs)and allows for the study of both the inner and outer regions of galaxies and their halos.

Churchill,Steidel,and Kacprzak 2.Mg II Absorption:Case for the Ideal Probe

The strong resonant Mg ii λλ2796,2803doublet in absorption is quite arguably the best tracer of metal-enriched gas associated with galaxies .Mg ii absorption is common and easily spotted in quasar spectra.Furthermore,magnesium,an α–process element,is ejected by supernovae into interstellar,intergroup,and intergalactic space from the epoch of the very ?rst generation of stars at the highest redshifts.Since Mg ii absorption is known to arise in gas spanning ?ve decades of neutral

hydrogen column density,from log N (H i )=15.5to 20.5atoms cm ?2(e.g.,Bergeron &Stasi′n ska 1986;Churchill et al.1999b),it samples a large range of astrophysical environments near star forming and post–star forming regions (i.e.,galaxies).

3.Absorption–Selected Galaxies

Following the initial study by Bergeron &Boiss′e (1991),Steidel et al.(1994,hereafter SDP),obtained images of roughly 50quasar ?elds and identi?ed the “absorption–selected galaxies”associated with strong Mg ii systems.Their study ?rmly established a connection between the galaxy luminosities and colors and the extent of their gaseous “Mg ii absorbing envelopes”for 0.3≤z ≤1.0

-26-24-22

-22-20-18Figure 1.Global Galaxy Properties —(left,center)The B–band and K–band luminosity functions of Mg ii absorption selected galaxies.The dashed lines are the luminosity functions of z ～0?eld galaxies overlayed on the data.There is a paucity of faint blue galaxies relative to the present day distribution of ?eld galaxies.—(right)The rest–frame B ?K colors of Mg ii absorption selected galaxies.The average color is consistent with that of an Sb spiral.

As shown in the left and center panels of Figure 1,the luminosities of Mg ii selected galaxies exhibit little evolution with redshift.However,there is di?eren-tial evolution in the luminosity function (see Lilly et al.1996;Guillemin &Bergeron 1997)in that faint blue galaxies are apparently under abundant;they are not selected in a survey using Mg ii gas cross section,seen as a turn down at the faint–end of the B–band luminosity.This is a quandary in that low mass halos of star forming galaxies are most expected to have large gas cross sections (as will be discussed below,this may be a selection bias).

Steidel (1995)showed that the sizes of Mg ii “halos”scale weakly with galaxy luminosity,following,R (L K )?38h ?1(L K /L ?K )

0.15kpc.This relation-

Mg II Absorbing Galaxies

-80-60-40-20020406080-80

-60

-40

-20

6080

Figure 2.Galaxy Gas Cross Section —(left)Impact parameter,D ,versus galaxy luminosity (Steidel 1995).The extent of the absorbing gas has a weak dependence upon galaxy luminosity,R ∝L 0.15,as shown by the dashed line.Open circles are absorbing galaxies and solid triangles are non–absorbing galaxies.—(right)The inferred statistical cross sections of Mg ii absorbers assuming they are associated with normal bright galaxies.

ship,illustrated in the left panel of Figure 2,was determined by plotting the impact parameter as a function of host galaxy luminosity and maximizing the number of absorbing galaxies (open circles)below the R (L K )relation and the number of non–absorbing galaxies (solid triangles)above the R (L K )relation.From the clean demarcation (as marked by the dashed line)it was inferred that the halos are roughly spherical and have unity covering factors and sharp bound-aries for W (Mg ii )>0.3?A (the dotted line is the canonical Holmberg slope 0.4,normalized at D =38h ?1kpc).

As illustrated in the right panel of Figure 2,the gas cross section for W (Mg ii )>0.3?A is ～40kpc and is statistically consistent the extent of gas op-tically thick at the Lyman limit of neutral hydrogen (i.e.,LLS).However,for the so–called weak Mg ii absorbers,those with W (Mg ii )<0.3?A (Churchill et al.1999b),little is known about what galaxy types are selected.If it is assumed that normal bright galaxies give rise to weak Mg ii absorption,then,using the normalization appropriate for the luminosity functions presented in Figure 1,their gas cross section is ～70kpc and is consistent with the region occupied by strong C iv absorbing gas.

4.Need for Complete Census

For all that has been accomplished to date,it is probable the scenario of a spherical distribution of gas with unity covering factor may be oversimpli?ed,or a result of survey methodology.Indeed,the SDP survey remains relatively imcomplete in that 30%of the absorbing galaxies in the ?elds do not

to be at the absorption

redshift).

Though 25“control”?elds,in which none of the galaxies gave rise to strong Mg ii absorption,were systematically studied (i.e.,all galaxies having L >0.05L ?within 10′′have measured photometry and spectroscopic redshifts),this was not the case for the 52?elds with absorbing galaxies.In these lat-

4Churchill,Steidel,and Kacprzak

ter?elds,the spectroscopic redshifts were not completed out to a?xed radius around the quasars;once a galaxy was identi?ed at the absorber redshift,no further redshifts were determined in the?eld.The galaxy spectroscopy was per-formed in an outward pattern from the quasar and galaxies that were targeted were based upon their magnitudes and colors being consistent with the absorber redshift.Close in galaxies of unexpected luminosity and color and further out galaxies could have been missed as potential absorbers(see Bowen et al.1995).

Thus,there is the very real possibility that some galaxies have been misiden-ti?ed as the absorber,or their may be more than one galaxy giving rise to the Mg ii absorption.Such incompleteness in the survey could have strong implica-tions on our inference of the covering factor,luminosity functions,and geometric model of the absorbing gas halos.Charlton&Churchill(1996)studied the ef-fects of incompleteness and possible misidenti?cations in the SDP survey and concluded that the realization of absorbing/non-absorbing galaxies shown in Figure2obtained by Steidel(1995)was not improbable for the survey methods even if the covering factor was as small as70%.

For only a single quasar?eld,that of3C336has a thorough census of absorbing and non–absorbing galaxies been completed(Steidel et al.1997).This rich?eld had four known strong Mg ii absorbers,but in the course of the study two additional weaker systems were found with W(Mg ii)<0.3?A!A deep WFPC–2/HST image of the?eld was obtained and every galaxy within50′′of the quasar in the?eld with L>0.05L?K was spectroscopically redshifted.

A Damped Lyαabsorber(DLA),still remains unidenti?ed.Interestingly, there is a galaxy at the DLA redshift;if it is the absorbing galaxy it lies well outside the R(L K)relation(L=0.6L?K and D=65h?1kpc)and has an ex-traordinarly large N(H i)for this impact parameter.

Furthermore,at least one non–absorbing galaxy was identi?ed that is well below the R(L K)relation so that it would be predicted to give rise to strong Mg ii.An after–the–fact search for Mg ii absorbtion in the HIRES/Keck spec-trum was inconclusive;this galaxy may host very weak Mg ii absorber.For each of two additional cases,there are two galaxies having redshifts coincident with the Mg ii absorption redshift.

Over the last decade,several quasar?elds have been reexamined by us in a patch work e?ort and a disturbing number of misidenti?cations and multiple galaxies at the absorption redshift have indeed been found!For example,in one case,a galaxy at70h?1kpc was found to be the absorber,not the assumed galaxy at30h?1kpc.This case again demonstrates that some of the absorbing galaxies lie well above the R(L K)relation(see Figure2).

In another case,the spectroscopic redshift of a galaxy thought to be associ-ated with a strong Mg ii absorber revealed it to be at a redshift with no known absorption.When reexamined with higher senstivity quasar spectra,the galaxy was found to align with a weak Mg ii absorber at a di?erent redshift;there is no obvious galaxy candidate in the?eld for the strong absorber!In the SDP survey,the galaxy associated with weak Mg ii,if properly identi?ed,would have been a non–absorber below the R(L K)relation.

In yet another?eld,a double galaxy was found at the absorber redshift; the assumed galaxy was not the absorber.Their are at least two other?elds in which three or more galaxies are clustered at the absorption redshift.

Mg II Absorbing Galaxies5 Clearly,our recent work and the complete analysis of the3C336?eld has provided examples of departures from the conclusions drawn by SDP.What is also very important to realize is that the HIRES/Keck quasar spectroscopy that has uncovered weak Mg ii absorption has shed new levels of sensitivity and complications on the whole Mg ii halo gas–galaxy connection.Some galaxies show weak absorption at smaller galactocentric distances and some show it at larger galactocentric distances.Armed with a factor of10increase in sensitivity to Mg ii absorption using HIRES/Keck,we now have the potential leverage to probe the patchiness of the halos and measure the covering factor at much more sensitive column density thresholds.All this,of course,hinges on the paramount objective of obtaining an unbiased,accurate,and complete census of the galaxies.

5.Galaxy Morphologies and Gas Kinematics

There has been a fair amount of activity in examining the kinematics of z≤1 Mg ii

absorbers and comparing them with the gross properties of the absorbing galaxies(Lanzetta&Bowen1992;Churchill et al.1996,2000b;Churchill&Vogt 2001).However,knowledge of the galaxy has been limited to the photometric magnitudes,B?K colors,and the observed impact parameters.

|||||

||||

|||

-200-1000100200

|||

-200-1000100200

|||

|||||||||

-200-1000100200

|||||||

-200-1000100200

|||||

-200-1000100200

||||||||

Figure3.Morphology,Impact Parameter,and Gas Kinematics—

The galaxy morphology(from WFPC-2/HST images)and the Mg iiλ2796

line–of–sight kinematics(from HIRES/Keck quasar spectra)shown with in-

creasing impact parameter order for10of our26galaxies.The vertical bars

above the Mg iiλ2796pro?les(in velocity)mark the number of subcompo-

nents and their velocities based upon Voigt pro?le?tting.

The sample size has remained relatively small,about15absorption sys-tem/absorbing galaxy pairs.These studies have not provided any clear system-

6Churchill,Steidel,and Kacprzak

atic trends to elucidate the nature of absorbing gas in relation to the galaxies. The“scatter”in the properties is simply too high for statisitcally signi?cant trends to be seen in the small sample.If systematic kinematics do exists,it is highly probable that additional“parameters”are at play that once invoked can illucidate the systematics.Thus,it is also of supreme importance to explore Mg ii kinematic connections with the galaxy morphologies,the line of sight paths of the quasar light,and the kinematics of the galaxies themselves.

In Figure3we show the galaxy morphologies and Mg iiλ2796absorption pro?le kinematics as a function of the quasar–galaxy impact parameter for10 galaxies.The galaxy images have been obtained with WFPC-2/HST and the Mg ii pro?les with HIRES/Keck.The images are2′′×2′′;the galaxy impact parameters are given in the upper portion of the panels and the redshifts in the lower portion.The absorbing gas is shown in the galaxy rest–frame velocity from?220to220km s?1(the zero points are arbitrary).The Mg ii absorp-tion exhibits a wide range of kinematic complexity,with velocity spreads of ～150km s?1,at various impact parameters.Ticks above the pro?les give the numbers and velocities of subcomponents,based uponχ2Voigt pro?le?tting (Churchill&Vogt2001;Churchill et al.2003).

We have used GIM2D(Simard et al.2002)to model the galaxy morpholo-gies for26absorbing galaxies(see Kacpzrak et al.2004).In Figure4,we show, from top left to top right,the HST image,the two–component GIM2D model, and the model residuals.,i.e.,data?model.North is oreinted up and east to the left.Note the striking asymmetric bar structure and the“tail”extending to the southeast.These structures are not immediately apparent in the data prior to modeling.We have found that similar levels of asymmetric structures are present in many of the galaxies in our sample.That is,the galaxy morphologies appear fairly normal prior to examination of the model residuals(Steidel1998; Kacpzrak et al.2004).

For each galaxy in our sample,we have quanti?ed the residual asymmetries using the parameter R A(Schade et al.1995)for3σresiduals out to twice the half light radius,R1/2.When R A>0.05,the galaxy is considered to have signi?cant morphological assymetry(rotated180?about the galaxy axis).In the lower panel of Figure4,we plot R A versus N cl,the number of“clouds”(Voigt pro?le components)comprising the MgII ii absorption pro?le.Filled data points have W(Mg ii)≤1.0?A and open data points have W(Mg ii)>1.0?A. Note that18of the26galaxies are catagorized as having signi?cant asymmetric morphologies.

We limit further discussion to those systems with W(Mg ii)≤1.0?A.These are“classical”systems,whereas those with larger W(Mg ii)typically are“dou-ble”systems;those with W(Mg ii)≥1.5?A are typically DLAs with severe pro?le saturation(Churchill et al.2000b).There is a2.8σcorrelation between R A for the galaxy and N cl for the absorbing gas.The dotted line through the data is not a?t,only a guide for the eye.

We also tested for correlations between gas kinematics and galaxy orien-tation.Orientation was parameterized by the angle between the galaxy major axis and the quasar sightline,and by the inclination of the galaxy.Only three of our26galaxies are ellipticals.We also examined orientation normalized to

Mg II Absorbing Galaxies

01020

304050

0.05

.1.15.2

Ncl G a l a x y A s y m m e t r y (3σ, 2R 1/2)Figure 4.Morpholgy and Kinematic Spread —(top,left)The WFPC-2/HST image of an absorbing galaxy quasar.—(top,center)The GIM2D model of the galaxy.—(top,right)The model residual.Note the asymmetric morphology.—(bottom)The morphology asymmetry versus the number of clouds in Mg ii absorption.The data suggest a correlation between morpho-logical “disturbance”and qas quantity and/or gas kinematics.

the galaxy luminosity and the quasar–galaxy impact parameter.No signi?cant correlations were found.

6.First Hints of a Galaxy–Halo Gas Kinematic Connection

At least in our moderate sized sample,there is no clear indication that the ge-ometric cross section and kinematics of the gas is closely related to orientation of the line of sight through the galaxy “halo”.However,the correlation between galaxy morphological asymmetry and the number of clouds intercepted by the quasar sightline is our ?rst hint of a connection between the galaxy proporties and the gaseous conditions in extended gas.It suggests that gravitational distur-bance,possibly due to a previous merging or accretion event,may have induced a stirring of gas surrounding the galaxy.Such disturbances are often seen in local galaxies from 21–cm studies (see Hibbard et al.2001,The Rogue Catalog).The fact that,for our sample,the luminous morphologies of the galaxies are fairly normal would imply that the events are not neccessarily violent,or that they occured a few dynamical times prior to the epoch of observation.

8Churchill,Steidel,and Kacprzak

Ultimatley,we need to obtain the kinematics of the luminous components of the galaxies.Steidel et al.(2002)recently conducted a detailed study comparing the galaxy kinematics to the gas kinematics for?ve Mg ii absorbers.We caution that these galaxies are all edge–on spirals,and not representative of the overall population of Mg ii absorption–selected galaxies.For four of the?ve,the ab-sorbing gas kinematics is rotating in the same sense as the galaxy;however,the total spread of gas velocities is inconsistent with simple disk rotation.In three of the?ve cases,the gas kinematics can be explained if the halo rotation“lags”the galaxy disk rotation,as is seen with DIGs and“beards”in local galaxies (Swaters et al.1997;Rand2000),but we are probing much higher in the halo. As a caveat,we o?er a counter example:Ellison et al.(2003)showed that the Mg ii gas kinematics in an edge–on spiral galaxy toward MC1331+170does not fully follow the disk rotation.Some of the gas is akin to the aforementioned “forbidden”gas,moving opposite to the rotation(Fraternali et al.2002). References

Bergeron,J.&Boiss′e,P.1991,A&A,243,344

Bergeron,J.,&Stasi?nska,G.1986,A&A,169,1

Bowen,D.V.,Blades,J.C.,&Pettini,M.1995,ApJ,448,662

Charlton,J.C.,&Churchill,C.W.1996,ApJ,465,631

Churchill,C.W.,Mellon,R.R.,Charlton,J.C.,Jannuzi,B.T.,Kirhakos,S.,Steidel,

C.C.,&Schneider,

D.P.2000b,ApJ,543,577

Churchill,C.W.,Rigby,J.R.,Charlton,J.C.,&Vogt,S.S.1999,ApJS,120,51 Churchill,C.W.,Steidel,C.S.,&Vogt,S.S.1996,ApJ,471164

Churchill,C.W.,&Vogt,S.S.2001,AJ,122,679

Churchill,C.W.,Vogt,S.S.,&Charlton,J.C.2003,AJ,125,98

Collins,J.A.,Benjamin,R.A.,&Rand,R.J.2002,ApJ,578,98

Ellison,S.L.,Mall′e n-Ornelas,Gabriela,S.M.2003,ApJ,589,709

Fraternali,F.,van Moorsel,G.,Sancisi,R.,&Oosterloo,T.2002,AJ,123,312 Guillemin,P.&Bergeron,J.1997,A&A,328,399

Hibbard,J.E.,Rupen,M.,van Gorkom,J.H.2001,“Gas and Galaxy Evolution,”ASP Conference Proceedings,Vol.240(San Francisco:ASP)

Lanzetta,K.M.,&Bowen,D.V.1992,ApJ,391,4

Lilly,S.J.,Le F`e vre,O.,Hammer,F.,&Crampton,D.1996,ApJ,460,L1 Kacprzak,G.G.,Churchill,C.W.,Steidel,C.C.,&Shapely,A.2004,in The e–Proceedings,“The Environments of Galaxies”(held in Crete),ed.D.Forbes, https://www.wendangku.net/doc/bf16231682.html,.au/conferences/crete2004/proc.html

Rand,R.2000,ApJ,537,13

Sancisi,R.,Fraternali,F.,Oosterloo,T.,&van Moorsel,G.2001“The HI Halos of Spiral Galaxies,”in Gas and Galaxy Evolution,ASP Conference Proceedings,Vol.240.

eds.J.E.Hibbard,M.R.Rupen,and J.H.van Gorkom,(San Francisco:ASP), p241

Schade,D.,Lilly,S.J.,Crampton,D,Hammer,F.,Le Fevr′e,O.,&Tresse,L.1995, ApJ,451,1

Simard,L.,Willmar,C.N.A.,Vogt,N.,et al.2002,ApJS,142,1

Steidel,C.C.1995,in QSO Absorption Lines,Proc.ESO Workshop,ed.G.Meylan (Springer:Heidelberg),p139

Steidel,C.C.1998,“Gaseous Halos and High Redshift Galaxies,”in Galactic Halos:

A UC Santa Cruz Workshop,ASP Conf.Series,V136,ed.D.Zaritsky(San

Francisco:PASP),167

Steidel,C.C.,Dickinson,M.,Meyer,D.M.,Adelberger,K.L.,&Sembach,K.R.1997, ApJ,480,568

Mg II Absorbing Galaxies9 Steidel,C.C.,Dickinson,M.,&Persson,E.1994,ApJ,437,L75

Steidel,C.C.,Kollmeier,J.A.,Shapley,A.E.,Churchill,C.W.,Dickinson,M.& Pettini,M.2002,ApJ,570,526

Swaters,R.,Sancisi,R.,&van der Hulst,J.M.1997,ApJ,491,140