Transcript IFU observations of the high-z Universe

IFU observations of the high-z Universe
Constraints on feedback from deep field observations with
SAURON and VIMOS
Joris Gerssen
Overview
• Until a decade ago only extreme objects were
known in the distant universe
• Since then photometric redshift surveys and narrow
band surveys identified ( at z ~2 to ~4)
– Lyman Break Galaxies
– Ly-alpha galaxies
• Observational constraints on galaxy formation and
evolution
– e.g. morphology, star formation history, luminosty
functions, etc.
• Among the drivers behind this advancement are
– The 10m class telescopes and instruments
– Hubble Space Telescope
– Theoretical understanding of structure formation
• Integral Field Spectropscopy (IFS) is a recent
development with great potential to further galaxy
evolution studies
Integral Field Spectroscopy
Data cube: f(x, y, lambda)
VIMOS
- SINFONI
- MUSE
- SAURON
- PMAS
- …
-
Typical properties:
Field-of-View few (tens) of arcsec
Spectral resolution: R ~200 to ~2500
High-redshift science with IFUs
• (e.g. list of MUSE science drivers)
• Formation and evolution of galaxies:
– High-z Ly- emitters
– Feedback
– Luminosity functions (PPAK, VIRUS)
– Reionization
– ...
Feedback
• A longstanding problem in galaxy formation is to
understand how gas cools to form galaxies
• Discrepancy between observed baryon fraction
(~8%) and predicted fraction (> 50% )
• To solve this “cosmic cooling crisis” the cooling of
gas needs to be balanced by the injection of energy
(SNe/AGN)
Feedback
• Galactic outflows driven by AGN and/or SNe
– Resolve discrepancy between observed and predicted
baryon fraction
– Terminate star formation
– Enrich IGM
M82 (starburst)
NGC 6240 (ULIRG)
IFU Deep Field Observations
• Deep SAURON & VIMOS observations of blank sky
• But in practice centered on QSOs/high-z galaxies
– observe extended Ly- halo emission
– serendipitous detections
SAURON Deep Fields
• The SAURON IFU is optimized for the study of
internal kinematics in early type galaxies
•
•
•
•
DF observations of: SSA22a, SSA22b, HB89
Redshift range 2.9 - 3.3 (4900 - 5400 Angstrom)
Texp ~10 hours
FoV: 33 x 41 arcsec, R ~ 1500
SAURON observations: overview
SSA22a
SSA22b
HB89
1738+350
SSA22b (z = 3.09)
Wilman, Gerssen, Bower, Morris, Bacon, de Zeeuw & Davies
(Nature, 14 July 2005)
VolView rendering
Ly- distribution
1.0 arcsec = 7.6 kpc
Line profiles
• Emission lines ~ 1000 km/s wide
• Emission peaks shift by a few 100 km/s
• Absorption minima differ by at most a few tens of km/s
• Ly alpha is resonant scattered, naturally double peaked
• Yet, absorption by neutral gas is a more straighforward
explanation
Model cartoon
SSA22b results
• Assuming shock velocities of several 100 km/s
• Shell travels ~100 kpc in a few 108yr
• Shell can cool to ~104 K in this time
– Implied by the Voigt profile b parameter
– Required to be in photoionization equilibrium
•
•
•
•
Implied shell mass of 1011 M
Kinetic energy of the shell ~1058 erg
About 1060 erg available (IMF)
Superwind model provides a consistent, and
energetically feasible description
Comparison with SSA22a
• SSA22a
– Kinematical structure more irregular
– Luminous sub-mm source
• Suggests that a similar outflow may have just begun
• Probe a wider range of galaxies:
– SCUBA galaxy (observed last year)
– Radio galaxy (observed one last week)
– LBG
(a few hours last week)
SINFONI observations of SSA22b
Constrain the stellar properties
Link them to the superwind
Scheduled for P77 (B)
Foerster Schreiber et al.
Serendipitous emitters
• The correlation of Ly-alpha emitters with the
distribution of intergalactic gas provides another
route to observationally constrain feedback
• Based on Adelberger et al (2003) who find that the
mean transmission increases close to a QSO
– This result is derived from 3 Ly- sources only
Mean IGM transmission
z~3
z ~ 2.5
Adelberger et al. 2003
Adelberger et al. 2005
Advantage of IFUs
• IFUs cover a smaller FOV then narrow band imaging,
but
– IFUs are better matched to Ly-alpha line width
– Do not require spectroscopic follow-up
– Directly probe the volume around a central QSO
• Thus, IFUs should be more efficient than narrow band
surveys
IFU observations
• Search the data cube for emitters
• Use the QSO spectrum to measure the gas distribution
– Likely require the UVES spectra
• Available:
– One SAURON data cube
– 2 of 4 VIMOS IFU data cubes
SAURON example: HB89 +1738+350
VIMOS 'QSO2'
z = 3.92, Texp = 9 hours
LR mode
Search by eye for candidates
Need to identify/apply an automated procedure
Detection algorithms
• Matched kernel search
– Many
false detections
• IDL algorithm (van Breukelen & Jarvis 2005)
• FLEX: X-ray based technique (Braito et al. 2005)
• ELISE-3D: sextractor based (Foucaud 2005)
van Breukelen & Jarvis (MNRAS 2005)
• Similar data set:
– Radio galaxy at z = 2.9
– same instrumental set up
– similar exposure time
• Yet, they find more (14) and
brighter Ly- emitters
– Using an automated source
finder
In progress
• A direct comparison with the van Breukelen results
– Obtained their data from ESO archive
– And reduced and analyzed it with our procedures
• Preliminary results are in reasonably good agreement
– ‘Our’ data appears somwhat more noisy
– Find their emitters and their new type-II quasar (Jarvis et al
2005)
Preliminary results
• Number density of Ly alpha emitters agrees with model
predictions (fortuitous)
– The VIMOS fields contain 5 - 14 emitters
– Models (Deliou 2005) predict 9 in a similar volume
• IFUs are sensitive to at least a few 10E-18 erg/s/cm2
Summary
• IFUs provide a uniquely powerful way to study the
haloes around high redshift proto-galaxies
• Volumetric data are an efficient way to search for
Ly-alpha galaxies
– An alternative method to constrain feedback
• IFUs are a very valuable new tool to study the
formation and evolution of galaxies