Transcript MOS Scientific Applications

MOS Scientific Applications
Michael Balogh
University of Durham
MOS Scientific Applications
(mostly galaxy clusters)
Michael Balogh
University of Durham
Outline
1. Galaxy Clusters: scientific motivation
2. Canadian Network for Observational Cosmology
(CNOC)
3. More clusters and groups with MOS
4. Ultra-plex spectroscopy: Ha spectroscopy of 4
clusters at z~0.4
5. Future considerations
Why Clusters?
Why clusters?
1. Very rare objects, in
the tail of the mass
distribution. So very
sensitive to cosmology
2. Largest structures just
virialising today, so
we can study the
process of structure
formation
3. Extreme environments
can affect galaxy
properties
150 Mpc/h
University of Durham
dalla Vechia, Jenkins & Frenk
Institute for Computational Cosmology
A Press-Schechter plot showing the growth of the mass
structure of the universe
LCDM cosmology
Rapid growth of structure
Groups
Clusters
Renormalised relative
to 1011 M☼
Why Does Star Formation Stop?
Steidel et al. 1999
Cluster environments inhibit
star formation (Balogh et al.
1997; 1998)
Can the growth in the
abundance of clusters
explain the global decline
of star formation?
Or is it related to internal
galaxy properties?
Why MOS?
Clusters are ideal targets for MOS because:
1. high density of galaxies
2. close in velocity space, so can use narrow
wavelength range to increase multiplex
3. Learn about cluster dynamics and galaxy
properties from the same set of spectra
CNOC: Cluster masses
CNOC: Goals
Sample: 15 X-ray luminous clusters from EMSS,
0.2<z<0.55
Goal to obtain 50-200 member redshifts per cluster,
for a total of ~1500 (r~22)
Observations over large fields (~0.5 degree) to
sample virialised region
Carlberg, Yee, Ellingson 1996 ApJS 102, 269
CNOC: Survey Strategy
MOS on CFHT
1. 4 band-limiting filters to sample rest-frame
~3500Å – 4300 Å, at 4 redshift slices
2. Obtain ~30 spectra per 9′ field of view; 2-3
masks per field, 1-5 fields per cluster
3. Real-time operations:
•
Imaging, mask design, mask cutting, and
spectroscopy all done at the telescope
Carlberg, Yee, Ellingson 1996 ApJS 102, 269
CNOC: Survey Strategy
Carlberg, Yee, Ellingson 1996 ApJS 102, 269
CNOC: Results
1. Dynamical measurement of Wm
2. Mass profiles of clusters
3. Cluster galaxy properties
Dynamical measurement of Wm
1. From velocity and spatial distribution, determine
cluster mass M and virial radius, R
2. Calculate mass-to-light ratio M/L
Carlberg et al. 1996 ApJ 462, 32
Dynamical measurement of Wm
1. Assume average galaxy M/L is the same in
clusters and in the field
2. Use the field sample from same survey to
measure (M/L)crit = rcrit/j, where j is the
luminosity density of the Universe
3. This calculation yields Wm~0.3; the most
convincing evidence for low Wm at the time.
Carlberg et al. 1996 ApJ 462, 32
CNOC: Average mass profiles
Dynamically determined
average mass profile of the
most massive clusters
In good agreement with
predictions from
simulations (Navarro,
Frenk & White 1996)
Carlberg et al. 1997 ApJ 478, 462
Carlberg et al. 1997 ApJ 485, L13
CNOC: Galaxy populations
Measurements of [OII]
emission line for galaxies
in clusters and the
surrounding field at z~0.3
[OII] closely related to star
formation rate (SFR)
Showed that average SFR
within the virialised regions
of clusters is much lower
than in lower density
regions
Balogh et al. 1997, ApJ 488, L75
CNOC: Galaxy populations
Morph-density relation
Field
Showed presence of
strong radial gradient in
SFR. Always lower than
the field
Gradient much steeper
than expected from
morphology-density
relation
Observed relation
Balogh et al. 1998, ApJ 504, L75
CNOC: Galaxy populations
Use numerical model of
infall to estimate timescale
for disruption of SFR
Radial gradients in CNOC
clusters suggest t ~2 Gyr
Balogh, Navarro & Morris 2000
CNOC: Remaining Questions
1. Are X-ray luminous clusters unusual?
2. Dust-obscured starburts? Is [OII] a good
enough SFR indicator? Are data complete
enough to rule out a small fraction of intense,
cluster-induced starbursts?
3. How far does the cluster’s influence extend?
4. Is star formation sensitive to local effects (i.e.
density) or global ones (i.e. clusters vs. groups)
1. Low Lx Clusters
Low Lx Clusters at z~0.25
Cl0841
z=0.24
s=390
Cl0849
z=0.23
s=750
Cl1701
z=0.24
s=590
Cl1702
z=0.22
s=370
Cl0818
z=0.27
s=630
Cl0819
z=0.23
s=340
Cl1309
z=0.29
s=640
Cl1444
z=0.29
s=500
Lx ~ 1043 - 1044 ergs/s, ~ 10 X less massive than CNOC
Low Lx Clusters at z~0.25
Multiobject spectroscopy with MOSCA (Calar
Alto) and LDSS2 (WHT)
No band-limiting filter, to allow measurement of
Ha in some cases
Star Formation in Low-Lx
Clusters
Spectroscopy for 172 cluster
members Mr< -19 (h=1)
Balogh et al.
1997
SFR from [OII] emission line
Identical to more massive
clusters
Balogh, et al. 2002, MNRAS 337, 256
2. Dust-obscured starbursts?
30
200
AC114 (z=0.31)
-100
0
Does star formation take
place in clusters at z>0 ?
-200
-300
DRA
100
Butcher-Oemler effect?
-200
-100
0
100
200
Couch et al. 2001, ApJ 549, 820
Nod & Shuffle: LDSS++ (AAT)
Band-limiting filter +
microslit = ~800
galaxies per 7’ field
Nod & Shuffle: LDSS++ (AAT)
Advantages:
1. Perfect sky subtraction. Allows
observation of Ha at z=0.31
(8600 Å)
2. Short slits = maximum multiplex
3. Trivial data reduction
Disadvantages:
1. Lose 2/3 of detector, unless
you use an oversized CCD
2. Need √2 more exposure time,
unless you nod along the slit
Ha in Rich Clusters at z~0.3
LDSS++ with nod
and shuffle sky
subtraction, on
AAT
(Field)
No evidence for
enhanced star
formation
Couch et al. 2001 ApJ 549, 820
Balogh et al. 2002 MNRAS, 335, 110
3. Cluster sphere of influence
Cluster sphere of influence
Fibre based wide field surveys:
1. 2dF galaxy redshift survey
•
Ha in 11000 galaxies within 20 Mpc of 17 clusters,
down to MB=-19 (Lewis et al. 2002, MNRAS 334, 673)
2. Sloan digital sky survey
•
Volume-limited sample of 8600 galaxies from the
EDR, MR<-20.5 (Gomez et al. 2003, ApJ 584, 210)
SFR-Environment Relation in the
2dFGRS
Lewis et al. 2002
MNRAS 334, 673
4. Galaxy Groups
The CNOC2 Field survey
1. Similar strategy to cluster survey, using MOS
on CFHT to study field galaxies out to z~0.6
•
Yee et al. (2000) ApJS 129, 475
2. Main goal to measure evolution of correlation
function and star formation rates
•
•
Carlberg et al. (2000) ApJ 542, 57
Lin et al. (1999) ApJ 518, 533
CNOC2 Groups
1. Identified a sample of groups from original
survey (Carlberg et al. 2001 ApJ 552, 427)
2. Properties of these groups can be directly
compared with low redshift counterparts from
2dFgrs and SDSS
3. Durham involvement: follow-up observations
with Magellan to gain higher completeness
confirming complete samples of group
members using LDSS-2
CNOC2 Groups at z~0.45
LDSS2 on Magellan
CNOC2 Groups at z~0.45
Combined with CNOC2
multicolour photometry and
spectroscopy, we can
determine group structure,
dynamics, stellar mass, and
star formation history
CNOC2 Groups at z=0.45
[OII]
[OII]
CNOC2 Groups at z~0.45
[OII]
[OII]
CNOC2 Groups at z~0.45
Preliminary results
based on only 12
CNOC2 groups
Have observed >30
groups to date
Balogh et al.
1997
The Future: Clusters at z>1
Groups at z > 1
1. Deep multicolour (VRi′z′JKs) images of Lynx
and Q1335+28 (z=1.2).
2. Proposals to observe high redshift radio galaxies
and radio-loud quasars: known to reside in dense
environments
• IRIS2 narrow band Ha and [OIII] at z=2.3
• GMOS/FORS2 narrow band filter + grism Ha and
[OII] spectroscopy at z=1.4, 1.47, 2.3
Lynx clusters: z=1.2
Subaru VRi’z’
INGRID JKs
Y (arcmin)
Identified 7 groups
around the clusters
from photometric
redshifts.
GMOS spectroscopy
pending
X (arcmin)
Nakata et al. (2002)
Overdensities around HizRG
z=1.44
z=1.59
Best et al. 2003
Conclusions
1. Clusters and groups have a large impact on
galaxy star formation rates at the present day
2. Need to understand how cluster populations
evolve to disentangle internal and external
effects
3. MOS at high redshift essential. Nod-and-shuffle
required to work at red wavelengths, but need
full field of view.