EUCLID spectroscopy : overview and synergies with WFXT

1
EUCLID spectroscopy overview and synergies
with WFXT

Gianni Zamorani
INAF - OABo

the EUCLID-NIS Team


WFXT Workshop, Bologna, 25 -
26 November 2009
2
EUCLID in a few words
EUCLID is an ESA Dark Energy Mission, currently
under study in the context of the Cosmic Vision
programme for a possible launch, if selected,
around 2017/2018.
It consists of two parts a spectroscopic part
in the near infrared (E-NIS) and an imaging part
in the optical and near infrared (EIC)
The main scientific goal is to measure with
exquisite accuracy the cosmological parameters,
including the Dark Energy parameter (w0 1-2)
and its (possible) evolution with redshift (wa
10). In order to reduce possible systematic
effects, this will be done using a set of
different cosmological probes
3

• Weak gravitational lensing
• Type Ia Supernovae
• Baryonic Acoustic Oscillations
• Redshift-space distorsions
• Integrated Sachs-Wolfe effect
• Clusters of galaxies
• Age redshift relation

Massive spectroscopic survey
4
Top Level Requirements

• Number of spectroscopic redshifts gt 5 x 107
• Redshift accuracy sz/(1z) 0.001
• Redshift range 0.5 lt z lt 2
• Sky coverage 20,000 deg2
• Deep Survey 2 magnitudes deeper, 40 deg2
• Mission duration 5 years

5
Slitless spectroscopy (baseline) Star-forming
galaxies Ha emission at 0.5ltzlt2
Baseline
6
DMD slit spectroscopy (option) All types of
galaxies at 0 ltzlt 2.5 selected in the H-band
(?obs 1.6 µm) (HABlt 22)
All galaxy types
7
EUCLID-NIS
Euclid-NIS is an extremely efficient redshift
machine. It will produce 70-160 million redshifts
in the redshift range at 0.5 lt z lt 1.2
(slitless/DMD) over half of the sky

Such a number of spectroscopic redshifts, by
itself, will have an enormous legacy potential,
which will be greatly enhanced by the products of
the imaging part of EUCLID (morphology, mass,
colors etc.) and by any other all-sky missions at
different wavelengths (e.g.X-rays - WFXT)
This combination of products will be a unique
resource to the astronomical community and impact
all areas of astronomy.
8
Additional Science (slitless)

• 70x106 galaxies AGNs star formation,
  co-evolution of distribution functions,
  environment
• Clusters of galaxies (mostly at z lt 1)
• Clustering and halo statistics
• The largest unbiased survey for high-z QSOs
• Most luminous objects at z gt 7 (Deep Survey)
• Our Galaxy (ultracool dwarfs, IMF), GAIA
• SNe (Deep Survey)
• Synergy with VIS/NIP, multi-? surveys (WFXT),
  JWST, ALMA..

galaxy spectra
High legacy value !
High-z QSOs
9
Comparison with some of the top existing
spectroscopic surveys
z range area
spectra Papers increase factor
(sq.deg.) in of redshifts SDSS
lt 0.3 10,000 1x106 gt 2000
70 - 160 DEEP2 lt 1.4
3.5 50,000 50 1400 -
3200 VVDS 0.2 - 5.0 8.0
50,000 50 1400 - 3200 zCOSMOS
0.2 - 3.0 1.7 30,000 25
2300 - 5000 Vipers 0.5 - 1.2
24 100,000 --- 700 -
1600
Increase by a factor more than 1,000 with respect
to the top present spectroscopic surveys at z 1,
and a factor gt 70 with respect to the local SDSS.
This, by itself, assures a very high legacy value
10
Other important features of the E-NIS Survey

• Well defined selection criteria
• either emission line flux limit (Slitless) or
  magnitude limit (DMD)
• This will make easy the scientific analysis
• Main differences between the baseline (slitless)
  and the DMD option
• Number of redshifts a factor 2.5
• Targets Slitless emission line galaxies
• DMD all types of
  galaxies, including passive galaxies
• Spectral features
• Slitless only emission
  lines
• DMD also absorption lines

broader scientific output from the DMD survey
(evolution and physical properties of ALL galaxy
types)
11
EUCLID dn/dz at 0.5 lt z lt 2.0
Expected counts of Ha emitters for different flux
limits and redshift bins (Geach et al. 2009) At
S 4 x 10-16 cgs we expect 7x106 galaxies in
?z bins 0.1 at 0.5 lt z lt 1.4 and 1 x106
galaxies in ?z bin 0.1 at 1.8 lt z lt 2
The ENIS Wide spectroscopic survey will allow us
to probe galaxy evolution in very narrow bins of
mass, type, star-formation and environment a
necessary step forward to pin-down the main
mechanisms of galaxy mass assembly. In addition,
the Deep Euclid survey will provide 46,000 Ha
emitters per deg2 to a flux limit of F(Ha) 5
x10-17 erg s-1 cm-2.
Baseline
12
Star formation history at 0.5 lt z lt 2.0
Reconstructed SFR from the wide (yellow close to
zero Poissonian errors) and deep (red excellent
constraint on the faint end slope) surveys
Limits in SF (in solar masses/year) z
Wide Deep 0.5 5
lt 1 1.0 20
2 2.0 120 10
accuracy of a few (modulo uncertainty on
extinction correction ) in the SFR Density in
each dz 0.1 bin with a single SFR indicator
over a wide redshift range
Excellent example of synergy between the Wide and
Deep Surveys
13
How will we handle extinction?
Usually, an average extinction of about 1 mag at
H? is ASSUMED
We will be able to do much better using, for
example, the existing correlation between
extinction and mass (mass from Euclid SED
fitting) In addition, for the higher fluxes, we
will detect also H? at z gt 1.05 ? this will allow
a direct spectral estimate of the extinction
This will help us in significantly reducing the
uncertainty in the extinction correction on a
galaxy-by-galaxy basis
Daddi et al., 2009
14
AGN
About a few of our spectra will be AGN ( 106
of them ! - an order of magnitude more than the
SDSS quasars) 1. These will cover an
impressively large redshift range with at least
two lines z range Ha
0.52 - 2.0 H? OIII 1.05 -
3.0 MgII 2.60 - 7.1 CIV
5.45 - 11.9 Lya 7.10 -
16.4 2. About equally splitted in type 1
(broad lines easily recognizable from the
spectra) and type 2 (narrow lines). Note that
because of the color-color selection, the type 2
AGN are severely under-represented in SDSS and
as of today (much) less than 100 type 2 AGN are
spectroscopically confirmed in the redshift range
1 lt z lt 2.
Two strong lines (CIV and Ly?) at 7.1 lt z lt11.9 !
15
An example TYPE 2 AGN
Most of our type 2 AGN will be in the same
redshift range as our star-forming galaxies and,
because of the adopted spectral resolution, we
will be able to recognize most of them on the
basis of the NII/Ha ratio.
Very few emission line galaxies with S(Ha) gt 8 x
10-16 cgs (NII/H? 0.3) have a measured
(NII/H? gt 0.6). This will allow the selection
of a clean and large sample of type 2 AGN 200 x
103 at 1ltzlt2.
A unique sample to study the f(AGN) as a function
of mass, SF, environment and co-evolution of AGN
and galaxies in a very interesting redshift range
A gain of a factor more than 103 with respect to
the number of currently known type 2 AGN in this
redshift range!
16
With WFXT
Two-ways synergy with WFXT WFXT --gt Deep X-ray
data for a large sample of optically selected
sample of type 2 AGN at z gt1 EUCLID --gt redshift
for X-ray detected type 2 AGN Who will benefit
more from this synergy? Probably us (almost all
our optically detected AGN will have an X-ray
detection), but redshift (even if only for a
not-negligible fraction ( 20 ? )of the sources)
will be very useful to WFXT. Similar arguments
apply to all other classes of AGN, including the
very high redshift AGN ( 6 lt z lt 12) (how many of
them?) where, in addition, also the EIC data
(dropouts) will play a very important role, for a
three-ways synergy
17
How many clusters will Euclid-NIS detect? and how
many galaxies in them? Andrea Biviano (INAF/Oss.
Astr. Trieste) Stefano Borgani (Univ. of
Trieste) Barbara Sartoris (Univ. of
Trieste) Need - cluster galaxies luminosity
functions (LF) - relation between mass and
richness - the evolution of the two above -
theoretical mass function of clusters
Not an easy job (see predictions of SZ clusters
for Planck)
18
Limiting M200 of detectable richness n clusters
DMD H? n50 n20 n5
A factor of 3 deeper in mass with DMD up to z
1.3 and much more at higher redshift
19
Number of clusters with richness ??n above given
z
DMD H? n50 n20 n5
Total Number of groups/clusters N(z)
slitless DMD Gain(DMD) gt 5
220,000 800,000 4 gt20
5,000 50,000 10 gt50 100
3,000 30
20
The LSS How well will we recover it ?
This mainly depends on the density of the
detected galaxies
Simulations in the YB have predicted the number
of groups/clusters expected to be detected with
E-NIS. These predictions have been verified
using some recent real data at 0.6 lt z lt 1.0
In the redshift range 0.6 - 1.0, our expected
density (dn/dz) is only 30 smaller than that in
the 10K zCOSMOS sample, from which a well defined
group catalog has been derived (Knobel et al.
2009) Then the EUCLID-NIS dn/dz will stay
constant up to at least z 1.35
21
The LSS what will we see?
zCOSMOS 10 k data in redshift slices, with groups
(red circles) overimposed
This is approximately what we expect to see
(filamentary web groups) over 20,000 sq.deg.
In z COSMOS there are 24 groups with Ngt5
members and 0.55ltzlt0.95. Reducing this number by
a factor of two (because of the worse EUCLID
redshift accuracy), it implies 140,000 Euclid
groups with Nmemb gt 5, in excellent agreement
with the theoretically expected number
22
The LSS which questions can we answer ?
Number of groups/clusters N(z)
slitless DMD Gain(DMD) gt 5
220,000 800,000 4 gt20
5,000 50,000 10 gt50 100
3,000 30
DMD Slitless
Ngt5
With more than 106 galaxies in groups (or gt105 in
clusters with Nmembgt20) photo-z for passive
galaxies we will study in detail a) mass
function of galaxies in groups vs.field and its
evolution with redshift b) SF as a function of z
and environment up to at least z 1.4 in
groups c) evolution of the morphology-density
relation, an essential step to understanding
structure formation d) physical processes which
produce differential evolution in fields and
clusters e) .
Ngt20
Ngt50
With DMD, this science (field vs. environment)
would be more powerful by about one order of
magnitude number of clusters spectra of
passive galaxies
23
Synergy with an X-ray missions eROSITA and
WFXT Euclid-NIS (Slitless or DMD) ltzgt
estimates (based on 5 galaxies) for all eROSITA
clusters (105 gt3 1014 M) DMD needed for deeper
WFXT survey ?improve cosmological
constraints Euclid-NIS (DMD) internal structure
and detailed kinematics for ?200 massive
clusters (based on 100 galaxies) ?several Bullet
Clusters, constrain DM properties (synergy with
Weak Lensing from EIC WFXT)
24
The mass calibration problems and
improvements Masses based on Emission-line
galaxies may be biased
...but things get better in more
distant clusters, in the redshift range where
EUCLID will find clusters
Low-z
25
The mass calibration problems and
improvements Masses based on Emission-line
galaxies may be biased
High-z
Low-z
26
The mass calibration
problems and improvements Bias and scatter in
mass determinations can profit significantly
from cross-calibration Euclid-NIS masses from
kinematics vs. Euclid-VIS Weak Lensing masses vs.
Euclid-NIP mass-estimates from richness/luminosity
vs. mass-estimates from X-ray and SZ
27
Conclusion
Both EUCLID and WFXT are very exciting and
powerful missions, and the output of each of them
will be greatly enhanced by the synergy between
the two missions.
28
Selecting galaxies into the epoch of reionization
Two basic techniques 1. Lyman-break
selection (LBGs) 2. Lyman alpha emitters via
narrow-band or spectroscopy (LAEs)
McLure, Dunlop, Cirasuolo et al. 2009, arXiv
0909.2437
Short wavelength light (?lt1215 Å) is absorbed by
neutral hydrogen
28
To find the first galaxies at zgt7, near-IR
observations are crucial
29
Selecting galaxies into the epoch of reionization
Two basic techniques 1. Lyman-break
selection (LBGs) 2. Lyman alpha emitters via
narrow-band or spectroscopy (LAEs)
Narrow-band filters sample dark regions between
sky lines Clean selection method if combined with
deep multi-wavelength imaging data Currently only
one object with spectroscopic confirmation at z
7
z4.7
z5.7
z6.6
30
Evolution of Lyman-alpha emitters
The bright end very poorly constrained Euclid
NIS down a flux limit of 5 x 10-17 erg s-1
cm-2 will detect (in 1 square degree) 10-100
galaxies at zgt6 1 - 20 galaxies at zgt7 0-5
galaxies at zgt8
Kashikawa 2006 Ouchi 2008 Ota 2008
Even higher redshift by targeting the critical
gravitational lensing lines in clusters (e.g.
Bradley 2008)
Similar numbers of quasars (Ly? and CIV) are
expected at the same redshifts
31
Evolution of Lyman-alpha emitters
The bright end very poorly constrained -gt need
for large FoV Euclid NIS down a flux limit of 5
x 10-17 erg s-1 cm-2 will detect (in 1 square
degree) 10-100 galaxies at zgt6 1 - 20
galaxies at zgt7 0-5 galaxies at zgt8
Kashikawa 2006 Ouchi 2008 Ota 2008
Even higher redshift by targeting the critical
gravitational lensing lines in clusters (e.g.
Bradley 2008)
With DMDs we can reach the knee of the LF
(pre-selection with optical/near-IR colours),
better constraining the shape of the LF
32
Giant Lyman-a nebulae
Extended Lyman-alpha nebulae are rare sources and
mostly found at z2-3, but recently Ouchi et al.
(2009) discovered one at z6.5 The nature of
these nebulae is still unknown 1.
Proto-galaxies in their very early stages
(ltlt107yr) with large outflows contributing to the
metal enrichment and cosmic re-ionization. 2.
Cooling clouds accreting onto a massive dark
halo. Based on the results from Ouchi et al.
(2009) Euclid-NIS will detect 50 spatially
extended Lyman-alpha nebulae at zgt7 in the Deep
Survey
Ouchi et al. 2009
33
Galactic Science

• Ultra-cool dwarfs in the solar neighborhood (Teff
  lt 2000 K)
• Characterization of atmospheres and complete mass
  function of free-floating very low-mass stars,
  brown dwarfs, and giant planets.
• Includes spectral types L (1300-2200 K), T
  (1300-700 K), and Y (lt 700 K).
• Slitless EUCLID allows to explore the presence of
  free-floating giant planets as old as the Sun up
  to r 8 pc.
• DMD slit EUCLID increases the explored volume by
  factors of 25 (Y-type) and 125 (L- and T-types).
• Ultra-cool dwarfs in star-forming regions
• Is there a minimum mass cut-off in the initial
  mass function (IMF)? Where is it?
• Slitless EUCLID allows to determine the complete
  brown dwarf IMF of most nearby star-forming
  regions.
• Slit EUCLID allows to study the universality of
  the IMF including the planetary-mass regime, and
  its dependence on various environments.
• Dependence of ultra-cool atmospheres on age
  (gravity) and metallicity.
• Binaries containing compact objects (black
  holes, neutron stars).
• Cool white dwarfs to explore the age of the
  Galactic disk and halo.
• Trans-Neptunian objects in the Kuiper Belt
  (pristine chemical composition of solar nebula).
• EUCLID will complement GAIA.

34
Galactic Science An example
Extremely low-temperature brown dwarfs a bridge
to terrestrial planet characterization
KI
KI
CH4
H2O
H2O
Between the coolest known dwarfs (T-type) and
Jupiter there is an interval of temperatures
typical of planet atmospheres (including
Earth-like temperature of 300 K) where EUCLID can
make relevant contribution to the identification
and characterization of the smallest population
beyond the Solar System. EUCLID wavelength
interval covers important features due to water
vapor, methane, oxygen, and ammonia.
L5, 1600 K
???
T5, 1100K
Jupiter, 150 K
Observed spectra from Rayner et al. (2009, ApJS)
degraded to the resolution of Euclid.
35
Conclusion
In addition to the main cosmological products,
E-NIS will produce exciting science, with a large
legacy value, in all areas of astronomy, from
galactic studies to the evolution of galaxies,
AGN, and LSS and the high redshift Universe,
close to the re-ionization epoch. This is true
for the current baseline configuration
(slitless). However, there is no doubt that the
E-NIS legacy value would be even (significantly)
larger if the DMD option becomes the baseline.

36
Additional Science the extra gain with DMD
spectroscopy
Unfeasible Limited to most luminous
objects Biased towards some class(es) of
objects Feasible
37
Why in the near-infrared ?

• Instantaneous coverage of 0.5 lt z lt 2 with Ha
  emission (1 2 µm)
• Much less affected by dust extinction than
  optical
• Rest-frame optical spectra ? high legacy value

Euclid-NIS is an extremely efficient redshift
machine. It will produce 70-160 million redshifts
in the redshift range at 0.5 lt z lt 1.2
(slitless/DMD) over half of the sky
Such a number of spectroscopic redshifts will, by
itself, have an enormous legacy potential, which
will be greatly enhanced by the products of the
imaging part of EUCLID (morphology, mass, colors
etc.).
38
Some of the main scientific goals of the slitless
survey
Galaxies The evolution of the multivariate
distribution functions (e.g., luminosity, stellar
mass, morphology, environment) of star-forming
galaxies The cosmic evolution of the star
formation density and activity at 0.5ltzlt2, for
different Hubble types Star forming galaxies
in the LSS Evolution of the merger rate of
different galaxy populations up to z lt 2 The
identification of the rarest and most massive
early-type galaxies at zgt1.5, to constrain the
evolution of the exponential ends of their
luminosity and mass functions a key test for
galaxy formation models AGN AGN galaxy
co-evolution Large samples of Type 1 and type 2
AGN a unique sample of type 2 AGN at z gt 1 and
fraction of AGN as a function of properties of
the host galaxy (mass, morphology, environment
etc) The High redshift Universe blind
spectroscopic searches for Lya emission and
spatially extended Lya nebulae (Lya blobs)
high redshift quasars Galactic Science
ultra-cool dwarfs (L, T, Y) and many others