Transcript Infrared Instrumentation & Observing Techniques

Optical Astronomy: Towards the HST,
VLT and Keck Era
Introduction & Overview
Chris O’Dea
Acknowledgements: Marc Postman, Jeff Valenti, &
Bernard Rauscher
Aims for this lecture

Historical overview
A

brief history of optical astronomy
trends in aperture and detector size
CCD Detection
 Observing Issues

 Effect
of the Atmosphere
 Effect of the Space Environment
Aims for this lecture II.

Optical Science

Pretty Pictures
– HST
– VLT

The synergy between optical and radio (real astrophysics)
– The radio loud/quiet quasar transition
– Time scales for fueling and activity in radio galaxies

Current `big’ issues in optical astronomy
Atmospheric Transmission (300-1100 nm)
History

Pre-history: mismatch between solar and lunar cycles
required astronomical observations to calibrate calendars
and predict times for natural and agricultural events
Newgrange, Ireland 3500 BC
 Stonehenge, England 3000 BC


First millenium BC – Greeks search for
Systematics of planetary motion
 Geometric model for planetary motion



Ptolemy’s Almagest (AD 145) presented robust geometric
model of planetary motion
12th century Islam- Need for more accurate measurements
of positions led to first “observatories” – dedicated
structures housing large, fixed instruments.
History


1575 Tycho Brahe’s Uraniborg – prototype of modern
observatory
1609 Galileo uses telescope for astronomy
Features on the moon
 Sattelites of Jupiter
 Stars remained unresolved


Development of reflecting telescopes (enables larger
collecting areas)


Gregory 1663, Newton 1668, Cassegrain 1672
Spectroscopy
1817 Fraunhofer combines narrow slit, prism and telescope to
make first spectrograph and discovers spectrum of the sun
 1859 Kirchoff shows that the solar spectrum reveals the
chemical composition

History

Photography
daguerreotype of sun – Focault & Fizeau
 1870’s - Improvements led to photography of faint
stars and nebulae
 1872 – Draper obtained photographic spectrum of
Vega
 1845

1875-1900 Combination of Photography and
Spectroscopy led to a shift of astronomy from
positional measurements to astrophysics
History
1970’s 4-m class telescopes become common
 1980’s CCDs are developed
 1990 HST launched
 1990’s 10-m class telescopes become available

Newgrange Megalithic Passage Tomb
Built ~3500 BC in County
Meath, Ireland
On winter solstice sun shines
down roof box and illuminates
central 62-ft passage.
Passage is
illuminated for 17
min after dawn
Dec 19-23
Tycho Brahe’s Uraniborg
Built 1576-1580
Prototype of “modern”
observatory
First “Big Science” –
required 1% of Danish
national budget!
Dedicated to precision
positional measurements
(one arcmin) – made
possible advances by
Copernicus and Kepler
Telescopes in Time
1858: Lassell 48”
First “Large” Reflector
1859: Clark 18.5”
1609
1672
Galileo 1.75” Newton 1.5”
1897 Yerkes 40”
Largest Refractor
1917 Hooker 100”
1948 Hale 200”
Hubble & Humason 1931, ApJ, 74, 43
Edwin Hubble
H~560 km/sec/Mpc
Aperture vs Time
450
Keck
400
Primary Aperture (inches)
350
300
250
200
150
100
50
Galileo
Newton
0
1500
1550
1600
1650
1700
1750
Year
1800
1850
1900
1950
2000
The Biggest Telescopes Today
Size Distribution of the 46 largest optical telescopes
Telescope Aperture (meters)
10
8
6
4
2
0
HST
CCD Camera Development for Ground Applications:
1.E+10
Luppino, 1998
# Pixels
1.E+09
DMT38k2
WFHRI36k2
OMEGA16k2
1.E+08
SDSS10kx12k
CFH8kx12k
UH8K2
Macho
8k2
EROS8k2
NOAO4k2
1.E+07
UH4k2
MOCAM4k2
BTC4k2
2k2
1.E+06
1990
1992
1994
1996
NOAO8k2
DEIMOS8k2
QUEST8k2
MDM8k2
MAGNUM8k2
CTIO8k2
ESO8k2
1998
Year
CFH_MEGA18k2
MMT_MEGA18k2
UW12kx16k
18k x 18k
8k x 8k
4k x 4k
8kx8k
4kx4k
2kx2k
2000
2k x 2k
2002
2004
2006
CCD Camera Development for Space Applications:
1.E+10
1.E+09
# Pixels
SNAP 250x2k2
GEST 60 3kx6k
1.E+08
18k x 18k
GAIA136x2k2
Fame 24 2kx4k
8k x 8k
Kepler 21x2k2
4k x 4k
1.E+07
ACS 4kx4k
WF3 4kx4k
2k x 2k
WFPC1
4x0.8k2
1.E+06
1990
WFPC2
4x0.8k2
1995
STIS 1kx1k
2000
Year
2005
2010
Astronomy at the end
of the 20th Century

Questions about the universe have become progressively more
sophisticated




From “Are there other galaxies? (ca. 1920)” to “What is the origin of
structure in the universe?”
From “How many planets in our solar system? (Pluto discovered 1930)”
to “How many extra-solar planetary systems lie within 100 light years of
the sun?” … and are any inhabited?
The basics of cosmology (age & density of universe), detailed maps of
the nearby galaxy dist’n, a basic theory of stellar evolution, and a
census of the stars in the solar neighborhood exist (or will exist within
5 years).
Astronomers today rely heavily on joint observations from ground &
space and data spanning large regions of the electromagnetic spectrum.
CCD Detection
MOS Capacitor:
Silicon Dioxide
Metal Electrode
+
Depletion Region
Silicon Substrate
•
•
•
•
•
CCDs are arrays of Metal Oxide Semiconductor (MOS) capacitors
separated by channel stops (implanted potential barriers).
Application of positive voltage repels majority carriers (holes) from
region underneath oxide layer, forming a potential well for electrons.
A photon produces an electron-hole pair: the hole is swept out of
depletion region and electron is attracted to the positive electrode.
Photoexcited charge collects in “depletion region” at PN junction.
Collected charge is shifted to amplifier (CCD) or sensed in situ (IR).
Structure of a 3-Phase CCD
Consider a 3-phase
CCD.
•Columns are separated
by non-conducting
channel stops.
•Rows are defined by
electrostatic potential.
•Charge is physically
moved within the
detector during readout.
CCD Vertical Structure



In the vertical direction, one sees a PN junction and control
electrodes.
Depletion regions form under both the metal gate and at
the PN junction.
Charge is collected where these depletion regions overlap.
Charge moves in a CCD



By changing electrode voltages, charge can be moved to
the output amplifier.
This process is called charge transfer.
In an IR array, this does not happen. Charge is sensed in
place.
CCD Readout Amplifier:
CCD
Readout
Amplifier
Packet of Q electrons is transferred through the output
gate onto a storage capacitor, producing a voltage V=Q/C.
The Atmosphere
Atmospheric absorption versus airmass

The amount of absorbed radiation depends upon
the number of absorbers along the line of sight
AM=1
Atmosphere
 mag / 2.5
I   I 0, 10
AM=2
, m ag  AM ,
where  is atm. extinction coefficient.
Atmospheric absorption versus altitude

Particle number densities (n) for most absorbers fall off
rapidly with increasing altitude.
I  I0, e
 
,where  is opticaldepth,
    ndx   e


 x/ x0
dx
x0,H20 ~ 2 km, x0,CO2 ~ 7 km, x0,O3 ~ 1530 km
So, 95% of atmospheric water vapor is below the altitude
of Mauna Kea.
Atmospheric Turbulence

A diffraction-limited point spread function (PSF) has a
full-width at half-maximum (FWHM) of:
 FWHM  1.2

{m}
D{m}

D{m}
{"}
In reality, atmospheric turbulence smears the image:
 FWHM  0.25

{radians}  0.25
{mm}
{mm}
r0{m}
{"}, where r0  6 / 5 .
At Mauna Kea, r0=0.2 m at 0.5 mm.
“Isoplanatic patch” is area on sky over which phase is
relatively constant.
Atmospheric Turbulence
1.4O seeing
0.5O seeing
no seeing!
Lick 3-m
Keck I 10-m
HST/NICMOS 2.4-m
Figer 1995
PhD Thesis
Serabyn, Shupe, & Figer
Nature 1998, 394, 448
Figer et al. 1999
ApJ. 525, 750
Adaptive Optics: “Eye Glasses” for
Ground-based Telescopes
Laser Guide Star
Wave
Front
Sensor
Adjust
Mirror
Shape
Adaptive Optics: “Eye Glasses” for
Ground-based Telescopes
Where does NGST win?
NGST should perform better than current
10m class ground-based telescopes.
 In the mid-IR range (wavelengths  3 m),
NGST will produce better quality (higher S/N)
images and spectra than a 50m AO corrected
ground-based telescope.
 For surveying large fields of view – AO only
works over a small field of view.
 Sky is much darker in space in NGST’s
wavelength range – better faint object
detection.

Observing in Space
HST Facts









Deployed 25 Apr 1990
Mass: 11600 kg
Length: 13.1 m
Primary diameter: 2.4 m
Secondary: 0.34 m
f/24 Ritchey-Chrétien
28 arcmin field-of-view
0.11 mm <  < 3 mm
0.043 arcsec FWHM at 5000 Å
HST Orbit:





Height = 590 km
Orbital period = 96.6 minutes
Precessional period = 56 days
Inclination = 28.5°
Continuous viewing zones (CVZ) at   ±61.5°
Space Environment:
Magnetic Flux Tubes:
CCD Radiation Damage:
ACS CCD
10 year
dose

Radiation damage limits the science lifetime of a CCD
 Ionization damage - flat band shifts
 Bulk damage
–
–

Displacement of Si atoms in lattice produces traps
Hot pixels created by electrons from silicon valence band
jump to trapping centers and generate high dark current
Annealing once a month to mitigate hot pixel accumulation.
 WFPC2 is warmed to +20o C
 STIS CCD is warmed -15o C
 80% of new hot pixels (>0.1 electron sec –1 pix –1 ) fixed
Losses Transferring Charge
SITe 1024  1024 CCD thinned backside
NGC 6752, 8  20s, ‘D’ amp at the top
Courtesy R. Gilliland (STScI)
Degradation of
Charge Transfer
Efficiency
Optical Science
 Pretty Pictures
 Astrophysics
Wide Field & Planetary Camera 2
Hubble Deep Field
There is a Synergy between High Resolution Optical and Radio
Observations
The Radio Loud/Quiet Transition
Overall SED is
similar for RL and RQ
quasars.
Why the difference in
radio power?
Sanders & Mirabel 1996, ARAA, 34, 749
Smooth Distribution in Radio Loudness
FIRST quasars. Solid line = all quasars, hatched region = newly
discovered quasars . Traditionally, radio loud objects have log R ~3-4.
Brinkmann etal 2000, A&A, 356, 445
Unimodal Distribution of Quasar Radio
Luminosity
5 GHz luminosity of
FIRST Bright Quasar
Survey II.
White etal. 2000, ApJS, 126, 133
Radio Luminosity – Optical Line
Correlation.
There is a strong correlation
between radio luminosity
and optical emission line
luminosity for both RL and
RQ objects. (see also Baum
& Heckman 1989)
Xu etal 1999, AJ, 118, 1169
Emission Lines are Powered by
Accretion Disk Luminosity.
There is a strong correlation
between X-ray luminosity
and optical emission line
luminosity for both RL and
RQ objects.
Xu etal 1999, AJ, 118, 1169
The AGN Paradigm
Annotated by M. Voit
What Causes the RL – RQ Transition?


Earlier data indicated a Bi-modal distribution of radio
loudness suggesting that the transition was very abrupt.
New data suggests a continuous distribution of radio
loudness. Thus, there is a more gradual transition.
Previously it was thought that there was a correlation with
host galaxy type – I.e., RQs are in Spirals and RLs in
Elliptical hosts. New data suggests that Ellipticals host
both RQ and RL quasars but only those with optically
luminous nuclei.
Quasar Host Galaxy Observations


Sample rest frame optical avoiding bright emission lines.
Match samples in optical luminosity at different z.
Kukula et al. 2001, MNRAS, 326 1533
Properties of the Host Galaxies

The surface
brightness profiles are
well fit by a r¼ law;
I.e. the host galaxies
are bulge dominated.
Dunlop etal 2001, astroph
Properties of the Host Galaxies
 The more luminous nuclei

live in galaxies which are
more bulge dominated.
Disk-dominated hosts
become increasingly rare
with increasing nuclear
power.
Relative contribution of the bulge to the total luminosity of the host galaxy.
RLQs are open, RQQs are filled circles, * are X-ray selected AGN from
Schade etal (2000). Dunlop etal 2001, astroph
BH Mass vs. Galaxy Bulge Mass
There is a relationship between BH mass and bulge
luminosity. And an even tighter relationship with the
bulge velocity dispersion. M(BH) ~ 10-3 M(Bulge).
Ferrarese & Merritt 2000, ApJ, 539, L9
Consistency Between Different Methods
 BH Mass vs bulge
magnitude relation is
similar for both active
and quiescent galaxies.
BH Mass vs bulge magnitude for quiescent galaxies, Seyferts and nearby
quasars. Size of symbol for AGN is proportional to the Hβ FWHM. Merritt
& Ferrarese 2001, astro-ph/0107134
BH Masses

BH Masses tend to be
high in these luminous
quasars.

Estimates of BH mass from
Hβ line widths and host
spheroid luminosity are in
rough agreement.
RLQs tend to have higher
BH mass than RQQs.
Assumes Mbh = 0.0025 Msph


Comparison between BH masses estimated from the host galaxy spheriod luminosity
and the Hβ line-width by McLure & Dunlop (2001). The shaded area marks BH
masses greater then 109 solar masses. RLQs are open, RQQs are filled circles.
Dunlop etal 2001, astroph
What Fraction of Eddington Luminosity?

RQQ and RLQs are
radiating at 1-10% of
their Eddington
luminosity.
Observed nuclear absolute magnitude vs that expected if the BH is emitting at the
Eddington luminosity. RLQs are open, RQQs are filled circles. Solid, dashed, and
dot-dashed are 100%, 10% and 1% of Eddington luminosity. Dunlop etal 2001,
astroph
The Paradigm Shift



Earlier data indicated a Bi-modal distribution of radio
loudness suggesting that the transition was very abrupt.
New data suggests a continuous distribution of radio
loudness. Thus, there is a more gradual transition.
Previously it was thought that there was a correlation with
host galaxy type – I.e., RQs are in Spirals and RLs in
Elliptical hosts. New data suggests that Ellipticals host
both RQ and RL quasars but only those with optically
luminous nuclei.
This is consistent with a correlation between optically
luminous nuclei and massive BHs and between BH mass
and host galaxy bulge mass.
Is it BH Spin ?

Possibilities include
 BH
Mass (but both RQs and RLs live in big bulges
and thus have high BH Mass)
 Mass accretion rate (but RQs and RLs have similar
optical luminosities)
 BH Spin
Time Scales for Gas Transport, Fueling,
and AGN Activity
Double-Doubles -- “Born-again”
Radio Sources


5-10% of > 1 Mpc radio
sources show doubledouble structure.
Working hypothesis: the
radio galaxy turned off
and then turned back on -creating a new double
propagating outwards
amidst the relic of the
previous activity.
Schoenmakers etal (2000)
Schematic of Supersonic Jet Model
Concept from Scheuer 1974, Blandford & Rees 1974. Illustration from Carvalho & O’Dea 2001.
Probing Time Scales of Activity
The double-doubles allow us to probe the timescale of
recurrent activity and the nature of the fuelling/triggering
of the activity.
Selection effects will limit the time scales which can be
detected in the double-doubles
If the source tuns off for < 106 yr the effects on the
larger source may not be noticable, and the younger
source may not be resolved from the core.
If the source turns off for > 108 yr, the larger source
will fade.
3C236 - 4 Mpc Radio Source
The largest
radio galaxy
known.
WSRT 92 cm image
(55”x96”) Mack etal.
1997) overlayed on
DSS image.
The Inner 2 Kpc Double
Inner 2 kpc double is well
aligned with outer 4 Mpc double
Global VLBI 1.66 GHz
image (Schilizzi etal 2001)
superposed on HST WFPC2
V band image
At z=0.1, and Ho=75,
1 arcsec = 1.7 kpc
O’Dea etal. 2001, AJ, 121,
1915
The Host Galaxy (in color)
Note
•
dust lane along major axis
•
tilted inner disk
•
blue knots along inner
edge of dust lane
3-color image.
STIS Near-UV MAMA
(F25SRF2 2300Å) 1440s
WFPC2 F555W (V) 600s
WFPC2 F702W (R) 560s
O’Dea etal. 2001, AJ, 121, 1915
STIS Near-UV Image
Note the 4 very blue regions in an arc along
the inner edge of the dust lane ~0.5”
(800 pc) from the nucleus, and
perpendicular to the radio source axis.
Regions are resolved with sizes ~0.3” (500
pc)
No strong emission lines in the F25SRF2
filter
Most likely to be due to relatively young
star formation
Bruzual-Charlot population synthesis
models are consistent with ages  5-10
Myr for knots 1,3 and ~100 Myr for
knots 2,4
STIS NUV image with global VLBI image
(Schilizzi etal 2001) superposed.
O’Dea etal. 2001, AJ, 121, 1915
Ages Estimated via Comparison with
Stellar Population Models
(Top) UV-V color as a function
of time for Bruzual-Charlot
models with both constant
star formation and an
instantaneous burst.
(Bottom) Evolution in colorcolor space of 3 models.
Plotted are the colors of the 4
knots, the nucleus, and the
older population in the host
galaxy.
Knots 1, 3 are consistent with 510 Myr, and knots 2, 4 with
~100 Myr.
O’Dea etal. 2001, AJ, 121, 1915
Star Formation Properties
Time Scales
Dynamical Ages:
Large radio source: t~7.8x108 (v/0.01c) yr
(comparable to the age of the oldest blue knots)
Small radio source: t~3.2x105 (v/0.01c) yr
(much younger than the youngest blue knots)
Dynamical time scale of the disk on the few
hundred pc scale t~107 yr
Alignments and the Bardeen-Petterson
effect
The small and large scale radio source are aligned to within about 10 deg.
 The radio sources are aligned to within a few degrees of perpendicular to the “inner"
(1 kpc) dust disk but are poorly aligned with the perpendicular to the larger dust lane.
 The Bardeen-Petterson effect will cause the black hole to swing its rotation axis into
alignment with the rotation axis of the disk of gas (on scales of hundreds to thousands of
Schwarzschild radii) which is feeding it; and conversely will keep the spin axis of the
inner disk aligned with the BH spin (e.g., Bardeen & Petterson 1975; Rees 1978)
The combination of the long term stability of the jet ejection axis and the alignment of
the jets with the inferred rotation axis of the inner kpc-scale dust disk suggests that the
orientation of the inner dust disk has also been stable over the lifetime of the radio source.
This also implies that the outer misaligned dust lane (which presumably feeds the disk)
settles into the same preferred plane as the disk.
The Scenario
The small and large radio sources are due to
two different events of mass infall.
Spectral aging estimates in the hot spotes of
the large source imply the radio source may have
turned off for ~ 107 yr in between the two
events.
The difference in the ages of the young and
old star formation regions also implies two
different triggers.
Implications
The two episodes of radio activity and the two
episodes of star formation are due to non-steady
transport of gas in the disk.
 If the young radio source and the young starburst
(knots 1,3) are related by the same mass transport
event, the gas must be transported from the
hundreds of pc scale to the sub-pc scale on the
dynamical time scale.

The Current Big Issues
Current `Big’ Issues for Optical Astronomy

Planet Formation & Evolution



When, where, & how frequently do planets form?
How important is dynamical evolution in planet
formation and consequent habitability?
Answers will require powerful (high S/N, high res.)
spectroscopic observations as 1 AU  0.002” at Orion
?
??
Current ‘Big’ Issues for Optical Astronomy

Star Formation & Evolution


Must have a more predictive and comprehensive theory
for star formation & evolution
Will require studies of stellar systems in hundreds of
other galaxies at (angular & spectral) resolutions
comparable with the work done in our own Galaxy
Current ‘Big’ Issues in Optical Astronomy

Galaxy Formation & Evolution

When do the first stars and galaxies form?
What processes trigger this formation and how do they
affect a galaxy’s evolution?

Develop a predictive theory of galaxy formation and evolution

HST Deep Field
Theory
Current ‘Big’ Issues for Optical Astronomy

Large-scale Structure




How are (proto) galaxies & clusters
distributed at when universe was only 25% of
it’s current age (z > 2)?
How do the distributions depend on the
galaxy’s mass, morphology, or star formation
rate in these early epochs?
How does structure evolve from the very
smooth pattern when universe was only a few
100,000 years old (z ~ 1000) to the highly
clumped and coherent pattern seen since last 6
billion years or so (z < 1)?
Answers will require large area telescope(s)
with large FOV and (moderate resolution)
spectrograph
The End