Transcript absorption lines

Ch.5
absorption lines
they allow to use quasars as cosmological probes to study the
Universe at large distances and large look-back times
QSO
zabs<zem
*
zem
absorption line system:
system of absorption lines at
the same zabs, presumably
associated with the same
absorber
absorption lines
study of great potential interest to investigate the gas distribution in the Universe
difficulties:
* weak and unresolved lines
need high spectral resolution, and high S/N ratio for weak
sources
need big telescopes
* study must be done as a function of redshift
need wide spectral range
every given QSO can include between 0 and hundreds of absorption lines in his
spectrum, depending on
(1) redshift zem
(2) observed spectral region
(3) limit EW (function of spectral resolution and S/N ratio)
most common lines: Lyalpha 1216, CIV 1548,1551, MgII
2795,2802
other common lines: CII 1335, Si IV1394,1403, MgI 2852
absorption lines
“metal” line system
“damped” Lyα
large column
density,
probably due to
the cross of a
galactic disk
absorption lines
Lyα systems
Ly limit system
Broad Absorption Lines and BAL QSOs
intrinsic to the quasar,
not intergalactic !
profiles PCygni-like,
absorptions shifted by ~ 30000
km/s
probably associated with
outflows from nuclear region
broad (104 km/s)
Gibson et al 2008
BAL QSOs are:
•X-ray-weak,
lower than for non-BAL
•usually radio quiet
BAL QSOs
~15-20% of radio-quiet AGNs
two classes of
models
model by Elvis (2000)
outflows are evolutionary
phenomenon, independent on
orientation
outflows are present in every
quasar, but cover only ~20% of solid
angle
statistic of absorption lines
number of absorbers crossed per unit path length
proper length
cross section of
the absorbers
density of the absorbers:
constant comoving
density
n(z)=n0(1+z)3
number of absorption lines per unit redshift
it is assumed that
comoving density and
cross section are
both constant
more generally, dependence on z is assumed
parametrically
for the non evolutionary case, no, σo const, we have:
q0=0
q0=1/2
implies evolution
for Λ≠0 change to:
risultati statistici
clear evidence
of evolution
possibly, lower
metallicity at
high z
effects near the QSO
number of absorbers can increase for zabs~zem because some absorbers
could be physically related to the QSO
viceversa in many cases (e.g. Lyα) number of absorption lines decreases
due to the higher ionization level (proximity effect or inverse effect)
to remove the effect, absorption systems within an
appropriate velocity interval from QSO are
excluded
in the QSO rest-frame, a cloud moving toward
the observer produces an absorption line at
in the observer
frame:
the corresponding
velocity can be
found:
typically, bias is removed excluding absorptions with zabs less
than the value corresponding to β~0.1
high redshift galaxies
they appear different from nearby galaxies, both for observational effects and for
intrinsic differences
main effects:
* redshift-dependence of surface brightness
* K-correction
* passive and active evolution
light of distant galaxies comes mainly from massive, young stars: observing at high
redshift, we see cosmic epochs of vigorous star formation
portion of the Hubble Deep Field
galaxies appear more irregular than
present day galaxies
we see them through the light emitted
in UV by the young stars
but in UV also nearby galaxies
appear less regular
cosmic distances and surface brightness
like for quasars, also for galaxies we must use luminosity distance
moreover, because galaxies have extended images, it is also important the
angular diameter distance
the different dipendence on redshift has important consequences
on surface brightness, and is due to the fact that in one case
photons are dispersed on detector’s area at z=0,
while in the other we observe photons emitted by an
area of the source at zem
surface brightness falls rapidly, making photometry difficult
the apparent sizes of corresponding isophotes shrink
fall of the isophotal diameter R25
exponential disk
spheroid R1/4
at low z there is a larger effect for the spheroid, at high z for the exponential
disk
luminosity distance
in units of c/H0
A(z)
0
z
angular diameter distance
in units of c/H0
Aa(z)
z
K-correction
(evolutionary
correction)
same spectrum
shifted in lambda
we cannot
measure the
spectrum of a
distant galaxy like
it is now, but we
can compute how
a galaxy identical
to one present
day galaxy would
appear if placed
at a redshift z
effect for z=0.5
elliptical, falls rapidly in the rest-frame
UV corresponding to observed B
starburst, small or no decrease,
because of young stars emitting in this
band
effect in the I band is lower
for both spectra
K-correction
K-correction for ellipticals, Sb spirals, blue
irregolar/starburst galaxies, in the bands BJ,
I, K
evolutionary correction
different spectrum at t0 and te
3 possibilities:
•burst of star formation, and then rapid death of massive stars and
progressive dimming of the other stars (passive evolution)
•further episodes of star formation
•addition of stars and/or gas in merging episodes
approximate expression in terms of the evolutionary
luminosity change dL/dt, for small z and Λ=0:
Δt being the look-back time
Hubble diagram
in the K band for some
samples of radiogalaxies
curves show the effect of two
models of passive evolution
with star formation burst at
z=20
passive and active evolution
passive evolution is called the change of galactic properties due to the aging of stellar
population born initially in the original star formation burst.
active evolution indicates instead the effect on galactic properties due to secondary events
of star formation, e.g. produced by merging
population synthesis: a galactic spectrum can be written simply as sum of the spectra of
constituent stars (ignoring complications such as internal absorption by dust or co-evolving
binary systems):
theoretical stellar spectra can be used, or even empirical stellar spectra, if they can be
observed for a grid of values of temperature, luminosity, and chemical composition
need to specify the initial mass function (IMF) with which stars are born. at high redshift IMF
can be much different than present IMF, probably peaked toward very massive stars
most common models use star formation with a single burst, or exponentially decreasing, or
constant. results show that much of the initial emission is in the UV. later, a strong
characteristic spectral feature is produced, called HK break or 4000Å break, a blend of
absorption lines near the HK CaII doublet. the amplitude of the break increases with age
and is little dependent on other factors
evolution of a galactic spectrum
color bimodality
luminosity, mass, color, morphology, stellar population of galaxies are strongly related. analysis of such
properties in the cosmic time started first with the study of the luminosity function but later included galaxy
counts as function of the various parameters
however, almost all these properties are unimodal, and galaxies tend to occupy a big cloud in the parameter
space, and it is often difficult to distinguish if a change in a particular cell of the parameter space is due to a
global number change or to a shift towards/from nearby cells
in this sea of unimodal functions, one function appears different for his bimodal character, the color function.
bimodality is evident, e.g. in the color-magnitude diagram (CMD), where two populations are clearly
distingushed, the BLUE CLOUD and the RED SEQUENCE
Baldry et al 2004
Hogg et al 2003
color bimodality
otherwise, this can be viewed with color distributions in bins of absolute magnitude, approximated by double
Gaussians
Baldry et al 2004
bimodality is present also for other parameters, morphology, metallicity, SFR, but color bimodality is much more
clear, and is observed up to z~1, and partially for z>1.
blue and red luminosity function
bimodal behavior is also clear from the
luminosity function, where a steepening is
observed in the low luminosity part of the LF of
blue galaxies for z > ~0.5, and instead a
substantial lack of evolution for red galaxies
(Lilly et al 1995)
these observations were interpreted with the
conclusion that red galaxies formed first, in
accordance with the so-called “monolithic
collapse” scenario (Eggen Lynden-Bell
Sandage 1962), and that blue galaxies are still
evolving
Lilly et al 1995
blue and red luminosity function
more recent studies based on 39,000
galaxies from surveys DEEP2 and
COMBO-17 (Faber et al 2007) have
provided evidence also for evolution of
the LF of red galaxies, with a decrease
of MB* and an increase of φ*
(parameters of the Schechter LF)
it is found also a substantial constancy
of the luminosity density for z<1
as stellar evolution models for red
galaxies predict an increase of the ratio
M/LB of 1-2 mag, constancy of jB implies
that stellar mass of red galaxies is at
least doubled from z=1
Faber et al 2007
color-stellar mass diagram
besides the color-magnitude diagram, bimodality is represented also
with the color-stellar mass diagram
e.g. Taylor et al 2009
color-stellar mass diagram
Bundy et al 2005
estimate of stellar masses uses
multiband photometry and redshift to
compare the observed SED with a grid
of synthetic SEDs depending on star
formation history, age, metallicity, dust
content.
for each grid model the computed
quantities are M*/LK, M*, chi2, and the
probability that the model represents
the data
probabilities are then summed on the
grid and probability histograms by
stellar mass are produced. so for each
galaxy a probability distribution of M* is
found, and the median value is adopted
as measure of M*
evolution in the color-stellar mass diagram
Faber et al 2007 assume that galaxies can transit from BLUE CLOUD to RED SEQUENCE when star
formation stops during a “major merger” (merging between galaxies with nearly equal masses). the stop of
star formation(quenching) is represented by nearly vertical lines. mergers are gas-reach (wet mergers)
because progenitor galaxies are blue galaxies with star formation. once on the red sequence, galaxies can
be subject to gas-poor mergers (dry mergers), described by the white arrows. three cases are proposed:
Track A represents an early
quenching of star formation,
when galaxy fragments are
still small. in this case, most
of the galaxy growth occurs
in “dry mergers”
Track B is the opposite
extreme, with a late star
formation quenching. in this
case, galaxies collect most
of their mass in the blue
phase, and then are
subject to merging and
become red, without further
“dry merging”
Track C is intermediate, with
contributions by both mechanisms. this
scenario is in better agreement with the
properties of elliptical galaxies, both
distant and local
variants of the color-stellar
mass diagram
Dekel et al 2006
Cattaneo et al 2009
variants of the color-stellar
mass diagram
Cattaneo et al 2006
green valley
Hasinger 2008
Smolcic 2009