Transcript Searching for HI emission from distant galaxies

HI in galaxies at
intermediate redshifts
Philip Lah (ANU)
Frank Briggs (ANU)
Matthew Colless (AAO)
Roberto De Propris (CTIO)
Michael Pracy (ANU)
Erwin de Blok (ANU)
Jayaram N Chengalur
NCRA/TIFR
Evolution of Wgas
Star formation Rate
Cold gas density
SDSS
DLAS
Very limited constraints on the gas content
(i.e. raw material
Ly-a line not observable from
for star formation) exist in the redshift range in which
the star
the earth
HIPASS HI emission survey
formation
shows
very
rapid
evolution
at z =rate
0 (Zwaan
et al.
2005
)
Sensitivity Issues – need for
SKA
The flux from an MHI* galaxy at z ~ 1 is
M HI
200km / s 6.2 103 Mpc 1  z
S  7 Jy (
)(
)(
)(
)
9
6 10 M sun
V
DL
2
The time required to make a 3s detection with
the GMRT is
  9000hr (
Tsys / 70 K
( N / 30)(G / 0.25K / Jy)( S / 7 Jy)
)2 (
200km / s
)
V
Clearly a much larger sensitivity is needed,
i.e. the SKA
What can one do now?
• The volume of space observed by the GMRT
telescope in a single observation ~ (FoV x
Bandwidth) could contain ~ 100 or more bright
galaxies
• One could try to detect the average HI emission
of all of these galaxies by stacking
• Stacking requires one to know the position and
redshift of all galaxies
Proof of concept A3128
• Naively if one co-adds the HI emission signal from N
galaxies, the SNR should improve by N½
– Redshift measurement errors lead to errors in aligning HI spectra
– Unknown HI mass of each galaxy leads to non optimal weights
while co-adding
• HI mass depends on morphological type, optical diameter.
– Unknown HI extent of each galaxy also leads to non optimal
SNR
• HI diameter correlates with optical diameter.
• Low redshift cluster A3128 observed as “proof of
concept”
Begum, Chengalur, Karachentsev et al. (2008)
A3128
• A 3128 is a z ~ 0.06, richness
class 3, Bautz-Morgan type I-II
cluster
Chengalur et al. 2001
(also Zwaan et al. 2001)
All late types
• Redshifts available for 193
galaxies, of which 148 lie inside
the ATCA cube
• Co-added emission detected from
cluster galaxies.
• Late type galaxies located outside
the X-ray contours have the
highest HI content
• MHI = 16.7 ± 2.6
• MHI = 8.6 ± 2
(late type, outer)
(all galaxies)
Late types outside
X-ray contours
Control Sample
The Subaru field
Fujita et al. (2003),
narrow band Ha
imaging at z ~0.24
Narrow band Ha selected galaxies
24’ × 30’
Fujita et al. 2003 did a narrow band
imaging survey for Ha emission at z=0.24
Total of 348 galaxies in the sample
Ha at
z = 0.24
The Giant Meterwave Radio
Telescope (GMRT)
• Aperture Synthesis Radio
Telescope (interferometer)
•30 Antennas each 45m in
diameter
•About 70 km N of Pune, 160
km E of Mumbai.
1 km x 1
km
• Hybrid configuration
• 14 dishes in central
compact array
• Remaining along 3 “Y”
Low and
high angular resolution
arms
GMRT images of CH3CHO
• Allows one to
emission from Sgr B2 (Chengalur
simultaneously
make
& Kanekar
2003) made
from a
low
and
high
resolution
single GMRT observation.
images
14
km
GMRT Observations
• 121 galaxies within the GMRT
data cube
• Total of ~ 40 hours of on
source time
• Most of these galaxies are fainter
than L* (i.e. low HI mass)
• Redshifts obtained using the 2dF
instrument on the AAT
• Optical imaging with the ANU 40”
telescope.
• Smoothing sized fixed using DHI Dopt relation from Broeils & Rhee
(1997)
Stacked HI Spectrum and WHI
121 redshifts - weighted average
MHI = (2.26 ± 0.90) ×109 M
GMRT
Measurement
Star Formation Rate at z = 0.24 shows same
correlations as for z=0 galaxies
SFR vs MHI
z = 0 relation from Doyle &
Drinkwater (2006)
SFR vs Radio
Continuum
z = 0 relation from Sullivan (2001)
Abell 370
a Galaxy Cluster at z = 0.37
Abell 370, a galaxy cluster at z = 0.37
large galaxy cluster of
order same size as
Coma
optical imaging ANU
40 inch telescope
spectroscopic followup with the AAT
GMRT ~34 hours on
cluster
AbellAbell
370370galaxy
cluster
galaxy cluster
Extent of X-ray
gas
324 galaxies
105 blue
(B-V  0.57)
R200  radius
at which cluster
200 times
denser than the
general field
219 red
(B-V > 0.57)
redshift histogram
324 useful
redshifts
GMRT
sideband
frequency
limits
HI all spectrum
324 redshifts (all available)
MHI = (6.6 ± 3.5) ×109 M
HI Mass in the inner regions of clusters
HI mass within 2.5 Mpc
of cluster centers
A 370 has substantially more HI mass than the comparable richess Coma cluster
HI mass to luminosity ratios
HI Mass to Light Ratios
HI mass to
optical B band
luminosity for
Abell 370
galaxies
Uppsala General
Catalog
Local Super
Cluster
(Roberts & Haynes 1994)
HI Mass vs Star Formation Rate in Abell 370
all 168
[OII]
emission
galaxies
Average
line from
Doyle &
Drinkwater
2006
Summary
•
Galaxies in A370 (z ~ 0.37, Tlookback ~ 4 Gyr)
have significantly more gas than those in the
similar size nearby Coma cluster
•
A370 shows similar trends as for nearby
clusters, e.g.
– decrease in HI mass for central galaxies
– Correlation of SFR with total HI content
– Calibration between O[II] derived SFR and radio
continuum derived SFR is the same as in the
local universe
•
At the observed SFR, A370 will evolve into a
gas poor cluster like Coma by z ~ 0
•
Co-adding is a powerful method to study the
HI content of star forming galaxies, galaxy
evolution in clusters, substructure in clusters
etc.
Thank you
HI mass
324 galaxies
219 galaxies
105 galaxies
94 galaxies
156 galaxies
168 galaxies
110 galaxies
214 galaxies
A3128: Inhomogeneous distribution
of gas rich galaxies
Gas Rich
Co-added spectra of the most gas
rich group
MHI ~ 26 x 109 Msun
Gas Poor