Seeing the Invisible: Observing the Dark Side of the Universe
Seeing the Distant Universe in
description
Transcript of Seeing the Distant Universe in
Seeing the Distant Universe inIntegral
Field Spectrosc
opyat high
redshift
3D3DAndrew Bunker, AAO & Oxford
After era probed by WMAP the Universe enters the so-called “dark ages” prior to formation of first stars
Hydrogen is then re-ionized by the newly-formed stars
When did this happen?
What did it?
DARK AGES
Redshift z
5
10
1100
2
0
B (0.45m) U (0.3m)
VV (0.6m) II (0.8m)
JJ (1.2m) HH (1.6m)
Near Infrared
Camera NICMOS
HUBBLE SPACE HUBBLE SPACE
TELESCOPETELESCOPE
z~1 HDF spiral
U (0.3m)B (0.45m)
II (0.8m) VV (0.6m)
JJ (1.2m) HH (1.6m)
"3D" SpectroscopyPreviously used a "long slit" in spectroscopy -
cut down background light, become more
sensitiveRelatively new technique
- integral field spectroscopy - arrange elements to survey a 2D area (rather than a 1D
line)The spectra gives a 3rd
dimension (wavelength, or velocity)
Integral Field Spectroscopy
Cambridge IR Panoramic Survey Spectrograph
What is CIRPASS?
Near-infrared integral field unit (spectra over a 2D area)
Built by the IoA with support of Sackler foundation & PPARC
Wavelengths 0.9-1.8m (z, J, H): doubles range of Gemini IFU science
490 spatial samples & variable image scales 0.05"-0.33" up to 5"x12" field
Large wavelength coverage (=2200Å) at R~4000: great sensitivity between OH sky lines
Limiting line flux on an 8m ~2x10-18 ergs/sec/cm^2 (53 hours)
Successfully demonstrated in August 2002 on Gemini-South telescope, community access 2003A
500 fibres IFU
Instrument cryostat
On dome floor
Sky "glow" in the near-IR
IFU Science
●Exquisitely sensitive to line emission redshifted between OH
●Star formation at z>1 (H, [OIII]5007Å, H, [OII]3727Å)
●Robust star formation rate measures down to
1M⊙/yr●Rotation curves, kinematics
●Masses, extinction, metallicity
●Nature of damped Lyman- systems at high-z
●Lensed galaxies/dark matter sub-clumping
●Ages of young star clusters
Gemini Integral Field Spectroscopy
– Program with Gemini Observatory to demonstrate the power of IFUs (5nights GMOS+8 nights CIRPASS)
Large interntational team (CIRPASS observations involve ~50 scientists) lead by Cambridge/Gemini/Durham
First demonstration of near-IR IFU science
Institute of Astronomy, Cambridge: Andy Bunker(AAO/Oxf), Joanna Smith (PhD student), Rachel Johnson (Oxf), Gerry Gilmore & Ian Parry, Rob Sharp, Andrew Dean etc CIRPASS team
Gemini: Matt Mountain, Kathy Roth, Marianne Takamiya, Inger Jørgensen, Jean-Rene Roy, Phil Puxley, Bryan Miller, etc. (Director's discretionary time)
Durham: Richard Bower, Roger Davies (Oxf), Simon Morris, Mark Swinbank etc. & GMOS team
Andrew Bunker, Gerry Gilmore
(IoA, Cambridge) & Roger Davies (Durham/Oxford
)
GMOS-IFU
GEMINI-NORTH
optical: Gemini Multi-Object Spectrograph
Hawaii June 02
GEMINI-SOUTH
Chile Aug '02,Mar/Jun 03
Q2237+03 - Einstein cross
Search for dark matter
substructure - Ben Metcalf,
Lexi Moustakas,
Bunker
z=1.7 QSO, z=0.04 lens
Substructure at 104M⊙<M<108M⊙ is 4%-7% of surface mass density - high compared to some CDM predictions
(but poss. variability/microlensing)
Q2237+03 - Einstein cross
Ben Metcalf, Lexi
Moustakas, Andy Bunker &
Ian Parry (2004,
accepted by ApJ, astro-ph/0309738)
Extended blue light over >5", aligned with radio
3C radio galaxy z=1.2 deep HST im.
studied by Spinrad & Dickinson
evidence of a cluster
size well-suited to GMOS/CIRPASS
study emission lines [OII] & [OIII]/H (kinematics)
A z=1.2 radio galaxy 3C324(Joanna Smith PhD)
[OIII] map in 3D of a z=1.2
galaxy (Smith, Bunker et al.)
Semi-raw frame
Sky (xy)
(xz)
(yz)
HST B-band (rest-UV)
GMOS-IFU [OII]3727
CIRPASS [OIII]5007
HST R-band
3C324 alignment effect, with Joanna Smith (PhD student)
GMOS IFU Spectroscopy Gemini-N 3C324 z=1.21 radio galaxy -
"reduced" 2D (still has sky & cosmics, but extracted fibres)
8000Å 8300Å
[OII]3727Å @z=1.2
3C324 3-D data
cube
[OII]3727 structure has two velocity components at +/-400km/s
Wavelength/
velocity
HST B-band (rest-UV)
GMOS-IFU [OII]3727
CIRPASS [OIII]5007
HST R-band
3C324 - Smith, Bunker, et al. : alignment effect
Galaxy kinematics redshift 1!
H map of a CFRS disk galaxy
with CIRPASS (Smith, Bunker et al., submitted)
[OII]3727Å doublet, ~300km/s velocity shift
Wavelength/
velocity
z=1 arc 3D data cube
z=1 arc de-lensed
Mark Swinbank, Joanna Smith, Richard Bower,
Andrew Bunker et al
[OII]3727Å velocity map
HST/WFPC (B,R,I) F450W, F606W, F814W
sky (lensed) de-lensed
Galaxy Kinematics at High Redshift:
Why do we care?
- For disk galaxies, velocity at flat part of rotation curve correlates with the stellar mass of the galaxy (I- or K-band) - the Tully Fisher relation-How does this scaling relation evolve with time?- In "classical" model, dark halo forms first, and disk forms later: M/L decreases with time.-So circular velocity at a fixed stellar mass less in the past- BUT in hierarchical assembly, make galaxies through mergers, so stellar mass vs. circular velocity follows same relation over a wide range of redshifts- Can test this through rotation curves of z~1 galaxies- Use rest-optical lines redshifted into near-infrared- IFUs ideal - no uncertainty of slit axis vs. galaxy axis
Emission lines ⇒ Star formation rates,
metallicity, dust extinction, kinematics
Damped Ly- QSO Absorption Systems
Bunker, Warren et al.
Star formation in damped Ly- systems(Joanna Smith PhD)
CIRPASS refereed Publications "Spectroscopic Gravitational Lensing and Limits on the Dark Matter Substructure in Q2237+0305" R.B. Metcalf, L.A. Moustakas, A.J. Bunker & I.R. Parry ApJ (astro-ph/0309738)
"Extragalactic integral field spectroscopy on Gemini" A. Bunker, J. Smith, I. Parry, R. Sharp, A. Dean, G. Gilmore, R. Bower, A.M. Swinbank, R. Davies, R.B. Metcalf & R. de Grijs (astro-ph/0401002)
"CIRPASS near-IR integral field spectroscopy of massive star clusters in the starburst galaxy NGC1140" R. de Grijs, L.J. Smith, A. Bunker, R. Sharp, J. Gallagher, P. Anders, A. Lancon, R. O'Connell & I. Parry; MNRAS (astro-ph/0404422)
"The Tully-Fisher Relation at z~1 from CIRPASS near-IR IFU H-alpha spectroscopy" J. Smith, A. Bunker, N. Vogt et al. MNRAS 2004
Seeing fluorescence from neutral hydrogen
5"
200Å
20"
zem=4.487
Spatially Extended Ly- Emission
z=4.5 QSO illuminating its protogalaxy
Extended Ly- , narrow (FWHM~1000km/s)
Central QSO (solid line)
broad Ly-Extended narrow Ly- (dashed
line),no continuum
Recombination line probably powered by reprocessed QSO UV flux rather than by local star formation.
The HI cloud of the host galaxy is
~>35kpc/h70 (=0.3)
SPH simulations, distribution of neutral gas at z~3 (from Katz et al. and Rauch, Haehnelt & Steinmetz).
Left box is 22Mpc comoving, 15arcmin; right zoomed x10
Wavelength/Å
The catch: very faint low surface brightness
The deepest spectrum in the Universe?
Rauch, Haehnelt, Bunker, Becker et al. (2007)
Win with IFUs rather than long-slit: MUSE?
DAZLE - Dark Ages 'z' Lyman-alpha Explorer (IoA - Richard McMahon, Ian Parry; AAO - Joss
Bland-Hawthorne
"Lyman break
technique" - sharp
drop in flux at
below Ly-.
Steidel et al. have
>1000 z~3 objects,
"drop" in U-band.
Pushing to higher
redshift- Finding
Lyman break galaxies
at z~6 : using i-drops.
The Star Formation
History of the Univese
Bunker, Stanway, z=5.8
Ellis, McMahon
& McCarthy (2003)
Keck/DEIMOS
spectral follow-up
& confirmation
I-drops in the Chandra Deep
Field South with HST/ACS
Elizabeth Stanway, Andrew
Bunker, Richard McMahon
2003 (MNRAS)
Galaxies at z~6 are small - barely resolved by HST. E-
ELT diffraction limit ~0.01” (~50-100pc). See individual HII regions?
What is JWST?● 6.55 m deployable primary
● Diffraction-limited at 2 µm
● Wavelength range 0.6-28 µm
● Passively cooled to <50 K
● Zodiacal-limited below 10 µm
● Sun-Earth L2 orbit
● 4 instruments
– 0.6-5 µm wide field camera (NIRCam)
– 1-5 µm multiobject spectrometer (NIRSpec)
– 5-28 µm camera/spectrometer (MIRI)
– 0.8-5 µm guider camera (FGS/TF)
● 5 year lifetime, 10 year goal
● 2014 launch
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
NASA/ESA/CSA - JWST● NIRSpec
– ESA near-IR MOS to 5um, 3’x3’
● NIRCAM - 3’x3’ imager <5um
● FGS (Canada) - has tunable
1% narrow-band NIR filters
in
● MIRI - mid-infrared
Europe/US
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
(closely similar to HST model…)
NIRSpec IST
Absorption lines at z>5 - a single v. bright Lyman
break z=5.5 galaxy, Dow-Hygelund et al (2005),
AB=23-24, VLT spectrum (22 hours), R~3000; S/N=3-10 at
R=1000,2700 in 1000sec NIRSpec
E-ELT
For I-drops (z~6) would only get ~1 per NIRSpec field bright enough for S/N~3-10 in continuum in 1000sec for abs line
studies
Does AO Help you?
-If Ly-alpha is compact, AO will boost point-source sensitivity-- Unclear if this will be the case - extended Ly-alpha haloes known, and expected through resonant scattering (see the far edge of the ionized bubble)
--For morphological analysis, unclear that high-tech ELT AO is better than a poorer but better-quantified PSF (e.g. from space)--If you can’t quantify where 10-20% of the light goes from a centrally-condensed core, that’s the difference between a disk and bulge morphology when fitting Sersic index--Even worse when looking for QSO host galaxies…
Conclusions-- 3D IFU spectroscopy at high redshift is (finally) realising its potential, but still small sample sizes--Important as a probe of galaxy kinematics, and spatially-resolved maps of stellar populations, metallicity-- Trace the evolution of the assembly of stellar mass--Explore the nature of gravitational lenses (dark matter)-- Explore the nature of the galaxies responsible for QSO absorption lines--In future might see fluorescence of the HI gas--Compact galaxies at high-z: need AO on ELTs to get real IFU benefit
GMOS-IFU (Swinbank et al. 2003)
Tully-Fisher at redshift 1!
Swinbank, Smith, Bower, Bunker et al.
HEALTH
WARNING!
CFRS22.1313
CIRPASS H
Lensed arc z=1
GMOS [OII]