Dynamical Studies of the Globular Cluster Systems around the Giant Elliptical Galaxies NGC4636 and NGC1399 [Elektronische Ressource] / Ylva Schuberth. Mathematisch–Naturwissenschaftliche Fakultät

Ylva Schuberth

Sujets

• Astronomie

Informations

• Publié par: rheinische_friedrich-wilhelms-universitat_bonn
• Publié le: 01 janvier 2011
• Nombre de lectures: 11
• Langue: English
• Poids de l'ouvrage: 6 Mo

Extrait

DynamicalStudiesoftheGlobularClusterSystems
aroundtheGiantEllipticalGalaxies
NGC4636andNGC1399
Dissertation
zur
Erlangung des Doktorgrades (Dr. rer. nat.)
der
Mathematisch–Naturwissenschaftlichen Fakultät
der
Rheinischen Friedrich–Wilhelms–Universität Bonn
vorgelegt von
Ylva Schuberth
aus
Bonn
Bonn, 2010AngefertigtmitGenehmigungderMathematisch-Naturwissenschaftlichen
Fakultät der Rheinischen Friedrich-Wilhelms-Universität Bonn
1. Referent: Prof. Dr. T. Richtler
2. Referent: Prof. Dr. P. Kroupa
Tag der Promotion: 15. November 2010
Erscheinungsjahr: 2011Unicuique proprium dat Natura munus:
ego numquam potui scribere ieiunus,
me ieiunum vincere posset puer unus.
sitim et ieiunium odi tamquam funus.
Archipoeta
Meinen ElternAbstract
Darkmatterstudiesinellipticalgalaxieswerelonghamperedbythelackof
suitable dynamical tracers. Being largely devoid of neutral gas and show-
ing no on–going star formation, the observational methods which are used
tounveilthepresenceofdarkmatterinspiralgalaxiescannotbeemployed
in the case of ellipticals. Moreover, the steepness of their surface bright-
ness profiles limits spectroscopic studies of the integrated stellar light in
elliptical galaxies to their central parts. Even the most recent studies just
marginally probe the regions where dark matter is expected to become
dominant.
The advent of 8m–class telescopes equipped with multi–object spectro-
graphs has made it possible to use globular clusters (GCs, i.e. roughly
spherical,denselypackedgravitationallyboundofstars)asdynam-
icalprobesconstrainingtheirhostgalaxy’sgravitationalpotential. However,
when using discrete tracers, a large number of radial velocities is required
to constrain the velocity dispersion profile which is the quantity linked to
the total enclosed mass via the Jeans equation. Being surrounded by ex-
tremely populous and very extended globular cluster systems, the giant
ellipticals in nearby galaxy clusters are the prime targets for this method.
Massivestarclustersareformedwhenevertheoverallstarformationrate
ishigh. Therefore,GCsystemscanberegardedasfossilrecordsofthechem-
ical and dynamical conditions at the time the host galaxy was formed. A
feature shared by all giant ellipticals is the bimodality of the colour distribu-
tion of their GC systems. The presence of two colour peaks results from a
metallicity difference between two old subpopulations, a metal–rich pop-
ulation with red photometric colours and a blue, metal–poor population.
Thesesubsystemsdifferwithregardtotheirspatialdistributionsandkine-
matic properties and hence have to be treated separately in the dynamical
analysis.
This work presents the two largest samples of globular cluster veloci-
ties obtained for giant elliptical galaxies to date: The galaxies studied are
NGC4636 located in the very outskirts of the Virgo cluster of galaxies and
NGC1399, the central galaxy of the Fornax cluster.
NGC4636 has a very rich GC system and is known for its unusually
bright X–ray halo which earned it the reputation of being extremely dark
matter dominated. Using 460 velocities of GCs out to a projected galacto-
icentric distance of 60kpc, we confirm that the blue GCs have a declining
line–of–sight velocity dispersion profile. The corresponding Jeans models
require significantly less dark matter than suggested by the X–ray stud-
ies, unless the latter incorporate a very strong (and probably unrealistic)
metal–abundance gradient.
The extremely populous globular cluster system of NGC1399 has an ex-
tentofatleast250kpc,whichiscomparabletothecoreradiusoftheFornax
cluster itself. Hence, the question arises whether there exists a population
of intra cluster globular clusters (ICGCs), i.e. GCs which are not bound to
anyindividualgalaxybut,rather,movefreelythroughthepotentialwellof
the Fornax cluster as a whole. Using a catalogue of candidate ICGCs from
the literature (150 velocities of GCs with projected distances of up to 230
kpc from NGC1399), and combining these data with photometry obtained
byourgroup,Ishowthatthevastmajorityofthemetal–poorGCsfoundin
between the galaxies of the Fornax cluster have velocities that are compat-
ible with their being members of the very extended NGC1399 GC system.
The line–of–sight velocity dispersion profile obtained for the GCs with the
most accurate velocity measurements declines with galactocentric distance
and is consistent with mass models derived from NGC1399 GCs within
80 kpc of the galaxy. Thus, no additional cluster–wide halo component is
required. However, we do identify one ‘vagrant’ GC whose radial velocity
suggests that it is not bound to any galaxy unless its orbit has a very large
apogalactic distance.
The data set used for the dynamical analysis of the NGC1399 GC sys-
tem presented in this work, comprises the velocities of about 700 GCs with
projected galactocentric radii between 6 and 100kpc. Our sample is fur-
ther augmented by including the above–mentioned ICGC velocities. The
most important results are: The metal–rich (red) GCs resemble the stellar
field population of NGC1399 in terms of radial distribution and velocity
dispersion. Themetal–poor(blue)GCshaveashallowerradialdistribution
and show a more erratic kinematic behaviour. Both subpopulations are
kinematically distinct and do not show a smooth transition. It is not possi-
ble to find a common dark matter halo which reproduces the properties of
both red and blue GCs. Some blue GCs within 100kpc of NGC1399 have
velocities that can only be explained by orbits with very large apogalactic
distances, thus indicating a contamination by GCs stripped from nearby
elliptical galaxies e.g. NGC1404 which is known to possess unusually few
GCs. The mass estimates obtained from the combined analysis of the red
GCs and the stellar velocity dispersion profile agree with the values from
iiX–ray studies in the inner 100kpc. At larger radii, however, we do not find
any evidence for a transition from a galaxy to a cluster halo, as suggested
by X–ray work.
Finally, we compare our GC–based NGC1399 mass profile to the dy-
namics of the Fornax cluster. We compile a catalogue of about 180 Fornax
cluster galaxies and present the velocity dispersion profiles obtained for
different morphological types of galaxies. The dynamical analysis (which
also makes use of recent distance measurements for a large fraction of the
early–type galaxies) suggests that the early-type giants form a subsystem
which is in dynamical equilibrium. The kinematics of these galaxies agree
with the extrapolation of the GC based mass estimate of NGC1399, i.e. no
separate ‘cluster halo’ dark matter component is needed. The late-type gi-
ants, on the other hand, tend to avoid the Fornax cluster core and their
velocity distribution indicates that these galaxies are an infalling popula-
tion.
iiiContents
Abstract i
TableofContents v
ListofFigures ix
ListofTables xiii
Acronyms&Abbreviations xv
1. Introduction 1
1.1. Dark matter on cosmological and galactic scales . . . . . . . . 1
1.2. Dark in disk galaxies . . . . . . . . . . . . . . . . . . . . 2
1.3. Dark matter in elliptical galaxies . . . . . . . . . . . . . . . . . 9
1.4. Giant elliptical galaxies and their globular cluster systems . . 10
1.5. Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2. Globularclustersasdynamicalprobes 23
2.1. Collisionless dynamics and the Jeans equation . . . . . . . . . 23
2.2. Solving the Jeans equation . . . . . . . . . . . . . . . . . . . . . 26
2.3. Effect of the parameters on the velocity dispersion σ (R) . . 27los
2.4. Observations required for a Jeans analysis . . . . . . . . . . . 28
3. ThegiantellipticalgalaxiesNGC1399andNGC4636 35
3.1. The environments of NGC1399 and NGC4636 . . . . . . . . . 35
3.2. The X–ray halos surrounding NGC1399 and NGC4636 . . . . 39
3.3. The globular cluster systems of NGC1399 and NGC4636. . . 40
4. ThedynamicsoftheNGC4636globularclustersystem 45
4.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.2. Observations and data reduction . . . . . . . . . . . . . . . . . 50
4.3. The combined data set . . . . . . . . . . . . . . . . . . . . . . . 54
4.4. Properties of the GC sample . . . . . . . . . . . . . . . . . . . . 56
4.5. Definition of the subsamples . . . . . . . . . . . . . . . . . . . 63
4.6. The line–of sight velocity distribution . . . . . . . . . . . . . . 66
4.7. Rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
v4.8. Globular cluster velocity dispersion profiles . . . . . . . . . . 71
4.9. NGC4636 stellar kinematics . . . . . . . . . . . . . . . . . . . . 73
4.10. Jeans models for NGC4636 . . . . . . . . . . . . . . . . . . . . 76
4.11. The mass profile of NGC4636 . . . . . . . . . . . . . . . . . . . 83
4.12. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.13. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5. ThedynamicsoftheNGC1399globularclustersystem 109
5.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.2. Observations and data reduction . . . . . . . . . . . . . . . . . 114
5.3. The velocity data base . . . . . . . . . . . . . . . . . . . . . . . 118
5.4. Propertie