Optical Spectroscopy

1
Optical Spectroscopy

• Introduction Overview
• Ian Browne Chris ODea

Acknowledgements Jerry Kriss Jeff Valenti
2
Aims for this lecture

• What is Spectroscopy?
• Spectrographs
• Information in a Spectrum
• Emission Lines
• Absorption Lines
• Astrophysical Results from Spectroscopy

3
What is Spectroscopy?
A picture may be worth a thousand words, but a
spectrum is worth a thousand pictures. Blair
Savage

• Spectroscopy is the study of radiation that has
  been dispersed into its component wavelengths.
• First astronomical spectrumthe Sun (Newton 1666
  Wollaston 1802 Fraunhofer 1814, 1817).

4
Spectroscopic Discoveries in Astronomy

• Chemical Abundances
• Discovery of Helium (in solar spectra, Lockyer
  Janssen 1868)
• Stellar evolution nucleosynthesis (Fowler,
  Burbidge2 1955)
• Big-bang nucleosynthesis (Peebles 1966)
• Measuring D/H is the primary mission of the Far
  Ultraviolet Spectroscopic Explorer (FUSE)
• Radial velocities/redshifts
• Galactic structure and rotation (Oort 1927)
• Expansion of the universe (Hubble 1929)
• Dark matter in clusters of galaxies (Zwicky 1937)
• Discovery of quasars (Schmidt 1963)
• Planets around nearby stars (Mayor,Queloz, Marcy,
  Butler 1995)
• g-ray bursters are at high redshift (Metzger et
  al. 1997)
• High-z supernovae/accelerating universe (Riess et
  al. 2000)

5
Spectroscopic Discoveries in Astronomy

• Line Widths
• Stellar surface gravities (white dwarfs)
• Stellar rotation (Schlesinger 1909)
• Velocity dispersions in ellipticals and bulges
• Ellipticals are not rotationally supported
  (Illingworth 1977 Schechter Gunn 1979)
• Black holes in galactic nuclei (e.g., Kormendy
  Richstone 1995)

6
What are those Squiggly Lines?

• Spectroscopic observations rarely receive press
  attention since the results arent as photogenic
  or as easily understood as astronomical images
• 2 of 32 HST press releases during 2000 were
  based on spectroscopic observations.
• Neither shows a spectrum!
• Some exceptions
• Black hole in M87, FOS WFPC2 (Ford, Harms, et
  al. 1994)
• He II in the IGM, HUT (Davidsen, Kriss, Zheng
  1996)
• Black hole in M84, STIS (Bower et al. 1997)
• SN1987A, STIS (Sonneborn et al. 1997)

7
Kinds of Spectrographs

• 1-dimensional (1D)
• Dispersed light is obtained from a single spatial
  point, or aperture
• Advantages
• Only requires a 1-D detector
• Simple optical design since entering light
  confined to the optical axis
• Examples FOS, GHRS, HUT, COS
• 2-dimensional (2D)
• Light entering through a long slit is dispersed
  at each point
• Advantage
• Spatial multiplexing increases efficiency by gt10x
• Disadvantages
• Requires a 2-D detector
• Greater optical complexity to handle off-axis
  rays
• Examples STIS, nearly all ground-based
  telescopes

8
Kinds of Spectrographs

• 3-dimensional (3D)Integral Field Spectrographs
• An entire area of the sky is imaged, and light
  from each pixel is separately dispersed into a
  spectrum. From this one can construct data
  cubes giving intensity as a function of (x, y,
  l).
• For compact objects, multiplexing the additional
  spatial element provides another order of
  magnitude increase in efficiency.
• (The tradeoff is the size of the field covered.)
• Examples
• Lenslet arrays TIGER, OASIS (CFHT)
• Fiber arrays DensePack (KPNO, retired), INTEGRAL
  (WHT)
• Image slicers popular for IR applications, MPEs
  3D
• Fabry-Perot interferometers Rutgers (CTIO),
  TAURUS-2 (AAT)

9
The OASIS Integral Field Spectrograph at the CFHT
10
Sample Data from an Integral Field Spectrograph
11
Atmospheric Transmission (300-1100 nm)
12
Definition of Spectral Resolution
Intrinsic Thorium Profile
Resolution
Resolving Power
Observed Profile
FWHM
13
Morphological Features in Spectra
Continuum Fit
Continuum
Emission Lines
Absorption Lines
14
Information in a Spectrum

• A spectroscopic observation provides the
  following information
• Spatial location (point, one, or two dimensions)
• Spatial resolution is instrument dependent
• Intensity (flux) as a function of wavelength
• Spectral resolution is instrument dependent
• Polarization as a function of wavelength
• The FOS could do spectropolarimetry, but STIS
  cannot
• Spectroscopic observations provide a direct view
  of atomic and molecular processes via their
  radiative transitions, thus enabling us to probe
  physical conditions in astronomical sources.

15
Quantitative Measurements of Emission Lines

• Flux, Centroid, Full-width at Half Maximum (FWHM)
• 0th, 1st, and 2nd moments of a spectral feature
• Fluxes ? physical conditions (density,
  temperature)
• ionization state
• abundances
• Centroids ? Kinematics (velocities)
• Outflow? Inflow? Rotation? ? Black Hole Mass
• FWHM ? Dynamics, temperature
• Making physical inferences
• Use individual lines as plasma diagnostics
  (Osterbrock 1989)
• Compare to models
• Collisional (or, coronal) equilibrium models
• Photoionization (CLOUDY, XSTAR)
• Shock models (MAPPINGS)

16
Optical Temperature Diagnostic
From Osterbrock (1989)
17
Optical Density Diagnostics
From Osterbrock (1989)
18
UV Density Diagnostic
From Osterbrock (1989)
19
Residual Intensity
Residual Intensity is the Flux Spectrum Divided
by Continuum Fit
Line Depth
Line Width
Equivalent Width
20
Quantitative Measurements of Absorption Lines

• Equivalent Width (EW), Centroid, FWHM
• Again, these are related to the 0th, 1st, and 2nd
  moments
• EW ? (f(l) fc(l)) / fc(l) d l Flux/ fc(lo)
• EW ? Column density ? physical conditions
• ionization state
• abundances
• Centroids ? Kinematics. Stellar lines ? Black
  Hole Mass
• FWHM ? Dynamics. Thermal motion? Turbulence?
• Opacity and Line Profiles
• Absorption cross section is s f (pe2/mc),
• where f is the oscillator strength.
• Opacity t (n) N s f(n) N f (pe2/mc) f(n)
• Lorentzian profile f(n) Fo (g/4p2) / ((n -
  no)2 (g/4p)2)
• Doppler profile f(n) Fo exp(-(n - no)2 c2
  /b2 no2)(c/(bnovp))
• Voigt profile Convolve the Lorentzian and
  Doppler profiles

21
Absorption Line Profiles
Lorentzian
22
Curves of Growth

• Curve of growth for the line equivalent width is
• Wn ? (1 - e-tn) dn

Square-root portion Wl /l ?(Nfl)
Flat portion Wl /l ? ln(Nfl)
Linear portion Wl /l Nfl
23
Spectral Features due to Hydrogen
24
Ultra-deep Echelle Spectra of the Orion Nebula
Baldwin etal 2000, ApJS, 129, 229
Region of the Balmer limit. Hydrogen lines up to
n28 are detected.
Emission lines of OII multiplet line 1 and very
week NIII and NII lines
25
Measuring the Mass of Black Holes in Galaxies

• Use stellar motions (rotation and velocity
  dispersion) to constrain models of stellar orbit
  distributions in the potential of a galaxy plus a
  central supermassive black hole (e.g., van der
  Marel et al. 1997).
• When gas disks are present, rotational velocities
  can be measured using line emission from the gas.
  Model as Keplerian rotation in the potential of
  the galaxy plus a central supermassive black hole
  (e.g., Harms et al. 1994).

26
Ford et al. (1994) Harms et al.
(1994)
27
Model for Disk Velocities in M87
Courtesy L. Dressel
28
STIS Observations of the LINER NGC 3998
29
STIS Long-Slit Spectrum of NGC 3998
L. Dressel/STScI
30
Fitting the HaN II and S II Emission Lines
Flux
Wavelength (Å)
Courtesy L. Dressel
31
Rotation Curve of NGC 3998
Courtesy L. Dressel
32
The Mass of the Black Hole in NGC 3998
2.0x108 Msun
1.5x108 Msun
Courtesy L. Dressel
33
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
34
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 Hß FWHM.
Merritt Ferrarese 2001, astro-ph/0107134
35
The Structure of AGN
Seyfert 1
Narrow Line Region
Torus
Central Engine Accretion DiskBlack Hole
Seyfert 2
Broad Line Region
36
The AGN Paradigm

• Annotated by M. Voit

37
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
38
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
39
The Alignment Effect in CSS Sources
CSS radio galaxies show extended emission line
gas which is aligned with the radio source axis
(De Vries etal 1998, Axon etal 2000)
40
The Alignment Effect in CSS Sources
The emission line gas is more strongly aligned in
the CSS radio galaxies than in high redshift
radio galaxies.
Histogram of difference in radio and optical
position angle. De Vries etal 1999, ApJ, 526, 27
41
The Alignment Effect in CSS Sources
HST STIS long slit spectroscopy of CSS Sources
ODea etal in preparation.
42
HST STIS Long Slit Spectroscopy of CSS Sources
Distance along slit
Wavelength (Velocity)
ODea etal in preparation.
43
HST STIS Long Slit Spectroscopy of CSS Sources

• There are systematic offsets in velocity on the
  two sides of the radio source
• There are complex line profiles with possibly
  multiple components
• Velocity shifts are 300-500 km/s
• Association of velocity shifts with radio lobes
  suggests that the CSS radio lobes are
  accelerating gas to these velocities.

ODea etal in preparation.
44
Intrinsic absorption in the FUSE spectrum of a
Seyfert 1 galaxy (Kriss et al. 2000).
Relative Flux
45
Physical Properties of the Absorbers in Mrk 509
46
(No Transcript)
47
PG 1634 706 (z1.335)
Ly?
STIS E140M
Ly?
48
He II in the Intergalactic Medium
Optical depth to H I in a Standard Cold Dark
Matter Model at z 2.336
Optical depth to He II From Croft et al. (1997)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
Ford et al. (1994) Harms et al.
(1994)
53
Supermassive Black Hole
8
M 3?10 Msun
STIS Slit
Nucleus
400 km/s
16 pc
400 km/s
54
(No Transcript)
55
References

• Allen 2000, Allens Astrophysical Quantities, ed.
  A. Cox, (Springer New York)
• Bower et al. 1997, ApJ, 492, L111
• Croft et al. 1997, ApJ, 488, 532
• Davidsen, Kriss, Zheng 1996, Nature, 380, 47
• Ford et al. 1994, ApJ, 435, L27
• Fowler, Burbidge, Burbidge 1955, ApJ, 122, 271
• Fraunhofer 1817, Denkschriften der Münchner
  Akademie der Wissenschaften, 5, 193
• Harms et al. 1994, ApJ, 435, L35
• Hubble 1929, Proceedings of the National Academy
  of Sciences, 15, 171
• Hubble Humason 1931, ApJ, 74, 43
• Illingworth 1977, ApJ, 218, L43
• Kormendy Richstone 1995, ARAA, 33, 581
• Kriss et al. 2000, ApJ, 538, L17
• Lockyer Janssen 1868. See http//ww.hao.ucar.edu
  /public/education/sp/images/lockyer.html
• Mayor Queloz 1995, Nature, 378, 355
• Marcy Butler 1995, ApJ, 464, L147
• Metzger et al. 1997, Nature, 387, 878
• Osterbrock 1989, Astrophysics of Gaseous Nebulae
  and Active Galactic Nuclei, (University Science
  Books Mill Valley)
• Oort 1927, Bull. Astron. Inst. Netherlands, 3,
  275

56
The End