Star formation in spiral vs
elliptical galaxies
Marie Martig
(MPIA)
With :
Frederic Bournaud, Avishai
Dekel, Romain Teyssier
Alison Crocker, Eric Emsellem, Tim
Davis, Martin Bureau, PierreAlain Duc + ATLAS3D
Gas and2 star formation in Bureau
early-type
et al.
galaxies
Molecular gas content of early-type galaxies
13
Young et al. 2011
re 6. Molecular masses and MK . Small crosses are ATLAS3D sample galaxies that are not detected in CO emission (3σ upper limits), and large circles
etections. The histograms in the right-hand panel show both the properties of the entire ATLAS3D sample (dotted line; left-side scale) and those of the
ies detected in CO (solid line; right-side scale). The relative scaling for the two histograms is the global detection rate. Binomial uncertainties are indicated
e histogram of detections.
Figure 1. Mosaic showing CO distributions (solid blue lines) overlaid
Bureau et al 2011, Alatalo et al. in prep
ATLAS3D:
on
volume(colours
limited
260
andsample
solid blackof
lines)
for local
selectedETGs
sample galaxies. A range of morph
22% detection rate in CO !
stellar velocity width σe . Figure 8 shows that the CO detection rate
for galaxies with σe ! 100 km s−1 is relatively high, 0.38 ± 0.05,
2. Molecular gas properties
−1
Gas and star formation in early-type
galaxies
Molecular gas content of early-type galaxies
13
Crocker et al. 2010
Young et al. 2011
re 6. Molecular masses and MK . Small crosses are ATLAS3D sample galaxies that are not detected in CO emission (3σ upper limits), and large circles
etections. The histograms in the right-hand panel show both the properties of the entire ATLAS3D sample (dotted line; left-side scale) and those of the
ies detected in CO (solid line; right-side scale). The relative scaling for the two histograms is the global detection rate. Binomial uncertainties are indicated
e histogram of detections.
ATLAS3D: volume limited sample of 260 local ETGs
stellar velocity width σ . Figure 8 shows that the CO detection rate
22% detection rate in CO !
e
for galaxies with σe ! 100 km s−1 is relatively high, 0.38 ± 0.05,
−1
Gas disk stability against local collapse
staragainst
formation
disk and
stability
local collapse and SF
Gas disk stability against local collapse and SF
Stability
results
fromfrom
a competition
between:
¨  Stability
results
a competition
between:
Stability results from a competition between:
� self-gravity
¤  self-gravity
� self-gravity
� velocity dispersion which inhibits the collapse
� velocity
¤  velocity
dispersion
dispersionwhich
whichinhibits
inhibitsthe
thecollapse
collapse
� differential rotation that shears gas clouds
� differential rotation that shears gas clouds
¤  differential
rotation that shears gas clouds
ToomreToomre
parameter
for a for
thina thin
rotating
gasgas
disk:
parameter
rotating
disk:
¨  Toomre parameter for a thin rotating gas disk:
κσg κσg
Qg =Qg =
πG ΣπG
g Σg
Stability
criterion:
¨  Stability
criterion:
QQg1g>>11
Stability
criterion:
Qg >
Foradisk
agas
gasembedded
diskembedded
embedded
disk
in aa stellar
stellar
disk:
aneffective
effective Toomre
For¨ a For
gas
in a in
stellar
disk:disk:
an an
effective
Toomre parameter
(stars contribute
to instability)
parameter
(stars
contribute
to instability)
Toomre
parameter
(stars
contribute
to instability)
1
1
1
1
1= 1+
Qg
Qs
=Q +
Gas disk stability in elliptical galaxies
¨ 
Gas disk is stabilized when stars are in a spheroid
instead of a disk:
¤  steeper
potential well
¤  reduced disk self-gravity
ETGs should have lower
star formation efficiencies:
morphological quenching
(Martig et al. 2009)
Evolution from z=2 to z=0
Morphological quenching in a cosmo
simulation
Elliptical galaxy with a
massive gas disk but
inefficient SF and red colors
(Martig et al. 2009)
MQ phase
Lower SF efficiency in elliptical galaxies
A comparison with high resolution AMR
simulations
AMR code RAMSES, 5 pc maximal resolution, star formation,
kinetic SN feedback (Martig, Crocker et al. 2013)
¨  Same gas disk embedded in spiral or elliptical galaxy
8
Mgas=7.5
M.¨ Martig
et al. x 10 Msun, fgas=1.3%
¨ 
6
!""#$%#&'"()*+',-./01
SFR = 0.1 Msun/yr
,$#2'"()*+',-./01
SFR = 2.5 Msun/yr
!""#$%#&'"()*+',-3/41
,$#2'"()*+',-3/41
A comparison with high resolution AMR
simulations
¨ 
¨ 
¨ 
AMR code RAMSES, 5 pc maximal resolution, star formation,
kinetic SN feedback
Same gas disk embedded in spiral or elliptical galaxy
Mgas=2.5 x 109Msun, fgas=4.5%
!""#$%#&'"()*+',-3/41
SFR = 4.8 Msun/yr
,$#2'"()*+',-3/41
SFR = 11.3 Msun/yr
The resulting Kennicutt relation
Molecular gas and star formation 21/12/07
in early-type
524
24/0
C
21/1
3489
23/04/07
C
3489
23/0
23/10/08
D
Table 1. Observing parameters
23/1
24/10/08
D
24/1
Galaxy
Date 4278Config.24/12/05
Ant.
Recei
C
4278
24/1
(NGC)
01/11/07
D
01/1
524 star
24/08/07
D
5
new:
Molecular gas and
formation
in
early-type
ga
4477
09/01/06
C3
4477
09/0
21/12/07
C
6
new: 3
30/04/06
D
30/0
3489
23/04/07
C
6
new:
3
7457
29/06/07
D
7457
29/0
Table 1. Observing
23/10/08parameters
D
6
new:
3
13/07/07
D
13/0
24/10/08
D
new:
3
Galaxy
Date
Config.
Ant.6 Receiver
4278
24/12/05
C
6
old
(NGC)
01/11/07
D 5666
new:
1
NGC
35
Mp
NGC
566
524
24/08/07
D
5 6 (at
new:
3mm
4477
09/01/06Donzelli
5/6 new:
old
&3mm
Da
21/12/07
C C & Davoust
6Donzelli
(2003
30/04/06
D
6
old
3489
23/04/07 much
C gas and
6much
new:
3mm
gasstruc
and
spiral
7457
29/06/07
D
4/5
new:
3h
23/10/08 also Dhas too6also
new:
3mm
has
too
high
a
bulge
13/07/07
5 12
new:
3
Figure
1.PA
DSS
R-band
images
of6populations
the
sample
24/10/08
DD
new:
3mm
f
H2 mass
H i mass
f3.6
Axis
Ratio
populations
for
a
normal
!
!
with
gas. Each Cimage is 67 × 7 . old
4278 molecular
24/12/05
Resolved observations of molecular gas
and star formation in ETGs
Martig et al.
sample of observed ETGs
Name
Distance
(Mpc)
maps : BIMA (Young
et al.
SpitzerImagin
Multi
Spitzer
Multiband
01/11/07
D
6
new: 1mm
NGC
5666(Crocker
(at
35brightness
Mpc)et
isface
a profile
difficult
ga
brightnes
2008)
PdBI
face
4477and 09/01/06
C
5/6al.
oldrea
NGC 524
23.3
7.8
< 6.4
0.293
36.5
1.06
Donzelli30/04/06
& Davoust
(2003)
conclude
Youngthat
etold
al.NG
2
D et al.62009), hintin
Young
2009,2011)
NGC 2768
21.8
7.8
7.8
0.258
180and spiral
1.20D
Vaucouleurs
p
much gas
structure
to be
a true
7457
new:
3mm
profile.
We
th
of
Es and29/06/07
S0s and Vaucouleurs
cluster
and4/5
non-cluster
m
NGC 3032
21.4
8.7
7.8HI mass:
0.293
92.5
2.06
but
rema
also
has
tooE/S0
highgalaxies
a bulge-to-total
ratio
and
13/07/07
5ple,
new:
3mm
(Serra
et
al.
Figure 1. DSS R-band images
of the
12Westerbork
sample
tionally
trying
to uniformly
populate
elliptici
ple, D
but remain
aware
of
Figure
R-band
images
of the
12 image
sample
E/S0
NGC
4150 1. DSS
13.4
7.7 gas.
6.0
0.264
146 for a 1.44
populations
normal Scd ticular,
galaxy. we
Annor
!×
! .galaxies
with molecular
Each
is
7
7
space (de Zeeuw etticular,
al. 2002).
we note that it d
!
!
with
molecular16.1
gas. Each image
selection crite
Spitzer 89
Multiband1.47
Imaging Photometer
(M
NGC
4459
8.2 is 7 × 7<. 6.62012)
0.264
NGC 5666 (at
35
Mpc) is
a difficult
galax
Combining
a selection
detection
campaign
with
criteria
for
the
follows
face brightness
profile
reasonably
wellupon
(base
NGC 4477
16.5
7.4
< 7.0SFR 0.261
227
1.20
Donzelli
&
Davoust
(2003)
conclude
that
NGC
maps:
non-stellar
8μm
de
Radioastronomie
Millimétrique
(IRAM)
3s
follows
upon
the
SAURON
ration).
Young
et
al.
2009),
hinting
at
the
dominanc
NGC 4526
16.4
8.8
< 7.9
0.268
108
2.28
much gas et
and
structure
be a the
truelite
ea
(Combes
al. spiral
2007)
and datatofrom
ration).
WeNGC
there5
(Shapiro
et
al.
2010)
Vaucouleurs
profile.
We
thus
include
S0s and cluster
and non-cluster
members,
addiNGC 4550
15.5of Es and6.9
< 6.5emission
0.268
179
3.55
also has too
highmain
a bulge-to-total
and t
galaxies
of the
sample areratio
detected
log(M! )
CO
log(M!
)
have
a1/4t
of Es and S0s and
cluster
and tonon-cluster
members,
addiand CO
map
ple, ellipticity-magnitude
but
remain awareWe
of therefore
its uncertain
mor
tionally
trying
uniformly populate
populations measurements.
for a normal
Scd
galaxy.
An
single-dish
Follow-ups
to robt
Figure 1. DSS R-band images of the 12 sample E/S0 galaxies
and CO
maps
(10
from
Unfortunately
tionally trying to
uniformly
populate
ellipticity-magnitude
ticular,
we
note
that
it
did
not
meet
the4tv
space
(de
Zeeuw
et
al.
2002).
!
!
20
−2
−1
−1
Spitzer
Multiband
Imaging
Photometer
(MIPS
with
molecular
gas.
Each
image
is
7
×
7
.
metric
molecular
gas
data
have
mapped
from Paper I. H2 masses from Crocker et al. (2011) and use XCO = 3 × 10 cm Unfortunately
(K km s ) the. H
i mas
CO
dist
lar ring
outsid
spacewith
(de Zeeuw
etfrom
al.
2002).
Combining
a detection
campaign
with
the
Institut
selection
criteria
for
the complete
Atlas3D
+12
spirals
data
THINGS
(Walter
et
al.
2008),
BIMA-SONG
face
brightness
profile
reasonably
well
(based
the
Berkeley-Illinois-Maryland
Array
(BIMA
or Crocker et al. (2011).Combining
f3.6 is the de
factorMillimétrique
applied
the
IRAC
3.6µm
image
whensoit(Cappellari
subtra
lar
ring survey
outside
ofis the
reg
sometimes
amultiplicative
detection
campaign
with to
the
Institut
Radioastronomie
(IRAM)
30m
follows
upon
thetelescope
SAURON
Young
et
al.
2009),
hinting
at
the
dominance
o
2008)
and
7
galaxies
with
the
IRAM
Plateau
n order to
remove et
the
stellar
emission
at Millimétrique
8µm
et
al. data
2010).
Position
angles
are
from Cappellari
(Helfer
2003)
and
SINGS
(Kennicutt
et
al.
2003)
surveys
the
of
thb
so
sometimes
we rest
will et
notal
(Combes
et (Shapiro
al. 2007)
and
from
the
literature,
13/48
deal.
Radioastronomie
(IRAM)
30m
telescope
ration).
Vaucouleurs profile.
thus include
NGC2009
566
terferometer
(PdBI; We
Crocker
et al. 2008,
axis ratios
are derived
from and
best
inclinations
from
Davis
etWe
al. therefore
(2011a,
Paper
unmapped
ga
the
rest
of the
sample.
A
galaxies
of data
the from
main the
sample
are detected
in CO
from
(Combes
et al. 2007)
literature,
13/48
have
a V).
total
of 12
galaxies
Observed LTGs and ETGs on the
relation
XXII.Kennicutt
Low-efficiency
star formation in ETGs 9
SF efficiency
1.8 x lower
(Martig, Crocker et al. 2013)
Conclusion
¨ 
Morphological quenching: gas disks are stabilized
against star formation when embedded in a stellar
spheroid instead of a disk, whatever:
¤  the
mechanism forming the spheroid
¤  the origin of the gas (left over after mergers or
accreted)
SF efficiency 2-5 times lower in ETGs
¨  There is an upper limit to the amount of gas that
can be stabilized
¨