Lecture 7

• Continuum Emission in AGN

• UV-Optical Continuum

• Infrared Continuum

• High Energy Continuum

• Radio Continuum - Jets and superluminal motion

Goal: The foundation of all astrophysical observations is the photon. All morphological and spectral information about astrophysical sources is derived from the emitted radiation. We learned about the power of line emission (spectroscopy) Continuum radiation is a natural consequence of the principle that accelerating charges radiate.

Can have : thermal or nonthermal emission

Spectral Energy Distribution

AGN show emission lines in all astrophysically relevant wavelength regimes

Power Law Continuum• Emission observed from 108 Hz to 1027Hz:

α=energy index now know to differ in different

bands

ανAνL )(

Actual SED is a function of the AGN Class

From last class:AGN Taxonomy

• Seyfert galaxies 1 and 2

• Quasars (QSOs and QSRs)

• Radio Galaxies

• LINERs

• Blazars

• Related phenomena

• Definition: radio-loud if

is larger than 10 (Kellermann et al. 1989)

• RL AGN have prominent radio features 10% of AGN population • RL: BLRGs, NLRGs, QSRs, Blazars RQ: Seyferts, most QSOs

• Deep radio surveys show intermediate sources

opticalradio LLR /

The Continuum

A phenomenological approach:

• Power law continuum

• Thermal features

• Spectral Energy Distributions of

Radio-loud and Radio-quiet AGN

Observing the SEDs of AGN

Types of Continuum Spectra

• Blazars: non-thermal emission from radio to gamma-rays (2 components)

• Seyferts, QSOs, BLRGs: IR and UV bumps (thermal) radio, X-rays (non-thermal)

Spectral Energy Distributions (SEDs): plots of power per decade versus frequency (log-log)

Spectral Energy Distributions

IR bump

Big Blue BumpEUV gap

Sanders et al. 1989

The radio and IR bands

• Radio emission is two orders of magnitude or more larger in radio-loud than in radio-quiet

• Radio and IR are disconnected, implying different origins

The IR and Blue bumps

• LIR contains up to 1/3 of Lbol

LBBB contains a significant fraction of Lbol

• IR bump due to dust reradiation, BBB due to blackbody from an accretion disk

• The 3000 A bump in 4000-1800 A:• Balmer Continuum• Blended Balmer lines• Forest of FeII lines

The highest energies• Typically α=0.7-0.9 in 2-10 keV • Radio-loud AGN (BLRGs, QSRs) have flatter X-ray

continua than radio-quiet• Soft X-ray excess is also observed, often smoothly

connected to UV bump• The only AGN emitting at gamma-rays

( MeV) are blazars

Blazars’ SEDs

Red blazars: 3C279 Blue blazars: PKS 2155-398

Wehrle et al. 1999 Bertone et al. 2001

Blazar SEDs main features

• Two main components:• Radio to UV/X-rays • X-rays to gamma-rays

• Component 1 is polarized and variable Synchrotron emission from jet• Component 2: possibly inverse Compton scattering

A fundamental question

How much of the AGN radiation is primary and how much is secondary?

• Primary: due to particles powered directly by the central engine (e.g., synchrotron, accretion disk)

• Secondary: due to gas illuminated by primary and re-radiating

An important issue

Isotropy of emitted radiation

• Thermal radiation is usually isotropic• Non-thermal radiation can be highly

directed (“beamed”). In this case: • We can not obtain the true luminosity of

the AGN• We will not have a true picture of various

AGN emission processes

Interpreting the BBBFrom accretion disk theory (last class),

And the maximum emission frequency is at

i.e., in the EUV/soft X-ray emission region.

Hz 106.3 16max ν

BBB=thermal disk emission?!

1. UV-Optical Continuum

Model Spectrum of an Accretion Disk

Spectrum from an accretion disk• Optically thick, geometrically thin accretion disk

radiates locally as a blackbody due to sheer viscosity

• Total integrated spectrum goes like ~ν2 at low frequencies, decays exponentially at high frequencies

• For intermediate frequencies spectrum goes as ~ ν1/3

• T=T(R) and T is max in the inner regions in correspondence of UV emission

• After removing the small blue bump, the observed continuum goes as ν-0.3

• Removing the extrapolation of the IR power law gives ν-1/3 - but is the IR really described by a power law??

• More complex models predict Polarization and Lyman edge – neither convincingly observed

Observations of optical-to-UV continuum

Disk interpretation is controversial!

Alternative interpretation

• Optical-UV could be due to Free-free (bremsstrahlung) emission from many small clouds Barvainis 1993

• Slope consistent with observed (α~0.3), low polarization and weak Lyman edge predicted

• Requires high T~106 K

Is an accretion disk really there?

Indirect evidence:

Fitting of SEDs Double-peaked line profiles

Direct evidence: Water maser in NGC 4258

Optical emission lines

Eracleous and Halpern 1984

Water Masers in NGC 4258

Within the innermost 0.7 ly, Doppler-shifted molecular clouds:

• Obey Kepler’s Law• Massive object at

center

2. The IR emission

• In most radio-quiet AGN, there is evidence that the IR emission is thermal and due to heated dust

• However, in some radio-loud AGN and blazars the IR emission is non-thermal and due to synchrotron emission from a jet

Evidence for IR thermal emission

• Obscuration :

Many IR-bright AGN are obscured (UV and optical radiation is strongly attenuated)

IR excess is due to re-radiation by dust

Radial dependence of dust temperature

From the balance between emission and absorption:

With R in pc, Leff in erg/s, T in Kelvin

5/12

6 )(10R

LT eff

Hotter dust lies closer to the AGN

Evidence for IR thermal emission

• IR continuum variability :

IR continuum shows same variations as UV/optical but with significant delay

variations arise as dust emissivity changes in response to changes of UV/optical that heats it

Emerging picture

• The 2μ-1mm region is dominated by thermal emission from dust (except in blazars and some other radio-loud AGN)

• Different regions of the IR come from different distances because of the radial dependence of temperature

The 1μ minimum

• General feature of AGN

• Consistent with the above picture: hottest dust has T~2000 K (sublimation temperature) and is at 0.1 pc

• This temperature limit gives a natural explanation for constancy of the 1μ minimum flux

3. Radio properties of AGN

I) Basic features of radio morphologyII) Observed phenomena• Superluminal motion• Beaming

Radio features

Core

Lobes

JetHotspot

Speed of Jets

What is the speed of radio jets in AGN? Since this is non-thermal plasma where no spectral lines are seen, the Doppler-shift cannot be used to derive a jet velocity for the nucleus!

Radio Telescopes: VLA, VLBI• The Very Large Array has angular resolution

• At z=0.5 this is ~2 kpc• For the Very Long Baseline Interferometry, R~1m.a.s.• At z=0.5 this is ~2 pc

"1106

kmcm

Dλ

R

The power of resolution

Energy is transported

by jets from the cores

to the outer regions

Superluminal Motion

• VLBI observations of the inner jet of 3C273 shows ejected blobs moving at v~3-4c

• This is called superluminal motion

How is this possible??

Historgram of observed v/c in 33 jets

Explanation of apparent superluminal motionExplain apparent superluminal motion as an optical illusion caused by the finite speed of light. Consider a knot in the jet moving almost directly towards us at high speed:

The blobs are moving towards us at anangle measured from theline of sight.

Photon emitted along theline of sight at time t=0, travelsa distance d to us, taking a time t1 to arrive: t1 = d/c

A second photon is emitted at a time te later, whenThe blob is a distance d – vte cos away from us. The second photon

arrives at t2 = te + (d - vte cos)/c

The observed difference in the time of arrival from photon 1 & 2 is:tobs = t2 - t1 = te (1 – vcos/c) < te

The apparent transverse velocityis vT = vte sin / t = v sin / (1 – v cos /c)

As v approaches c, vT canappear > than c! Superluminal motion, typically 5-10c!

Let = 1/(1- v2/c2)1/2, this is the Lorentz factor. Then:vT v (the maximum observed velocity) which occurswhen cos = v/c. We will only observe superluminal motion whenthe jets are pointed within an angle of 1/ towards the line of sight,but this light will be beamed and brightened.

Relativistic motion of plasma

• Relativistic bulk motion in radio sources has important consequences on the following observed quantities:

1. Frequency

2. Length and time

3. Intensity

4. Direction light is emitted

Relativistic Doppler Effect

Assume an emitting source moving at a speed v c at an angle with respect to the observer.

Time-dilation tells us that t in the observers rest frame for a periodic signal with frequency ’ in the co-moving (primed) frame is:

However, since the emitting source is moving almost as fast as the

emitted photon, the source will be catching up on the photon, and

travel a distance s = v tcos . The time difference in the arrival

time of the two photons will therefore be reduced by s/c, i.e

and the observed frequency is

This is the relativistic Doppler effect which defines the Doppler factor

One can show (i.e. Rybicki & Lightman, chap. 4.9) that the ratio of the flux density S and the frequency cubed is invariant under Lorentz transormation:

Since the observed frequency is =D’, = we find that also the observed flux has to be (S’= flux density in co-moving frame)

Even for relatively modest relativistic velocities of v=0.97c, for example, the flux in the forward direction can be boosted by a factor 1000, while it is reduced by a factor 1000 in the backward direction!

The transformation from a spherical to an elliptical polar diagram shows that angles are also transformed by relativistic effects. The so-called relativistic aberration (see Rybicki & Lightman, chap. 4.1) is given by:

In the rest frame of the source, half of the radiation will be emitted from –/2 to /2, hence setting ’ = /2 will give

thus for >> 1 half of the radiation will be emitted in a cone with half-opening angle

Jet-sidedness

Since we expect jets to be two-sided, we always have two angles

under which the emission is seen by an observer: and + . We

can now calculate the flux ratio R between jet and counter-jet under

the assumption of intrinsically symmetric jets:

Even for mildly relativistic jets one side will always be significantly brighter than the other

•Most of the strong, compact radio cores seem to come from sources where the angle to the line of sight is small, these jets are always one-sided.

•Even most of the large scale jets appear to be one- sided, even though 2 extended lobes are seen indicating that

really two jets are present.

Nearby FRI radio galaxy and LINER galaxy M87 - no counter-

Jet observed

Summary: evidence for relativistic motion in AGN

• Superluminal motion• One-sided jets (pc and kpc scales)

Caveats• None of the above evidence proves that relativistic motion

exists• Alternative explanation exist for each observed property

(e.g., one-side jets)• But relativistic motion=beaming is the only and the

simplest explanation for all of them at once

Physics of AGNThe Emission-Line Regions (BLR, NRL)

An AGN produces a lot of ionizing radiation, most likely from theaccretion disk.

This emission is intercepted by gas and dust in the host galaxy.Correspondingly an AGN spectrum shows reprocessed radiation fromthis gas and dust. The respective features are:

• Broad-Line Region (BLR)• Narrow-Line Region (NLR)• IR-bump from a molecular (dusty) “torus” (which we talked about last class)

Reprocessed Radiation

• FWHM several thousand km/sec up to 30000 km/sec FWZI (zero

intensity)• derived gas temperatures are several 104 K• Doppler broadening through bulk motion of gas in gravitational field

• with velocities as high as 0.1c, the distance from the Black Hole

can be as close as 100 Rs

• Comparison of continuum and BLR fluxes indicate that only 10% of

the continuum radiation is absorbed by BLR clouds• The volume filling factor is very low - a few millionth of the central

region is occupied by BLR 'clouds'• The necessary mass in the BLR to produce the observed luminosity

is only a few solar masses

• Broad-lines are very smooth - they are either made up of a huge

number of small clouds or represent a coherent structure

BLR:Properties

Broad, permitted emission lines (e.g., H) in the optical spectrum:

Lines from highly ionized gas (He II 1640, C IV 1549) respondfaster than lines from lower ionization levels (e.g. Balmer lines)

ionization structure in BLR

more highly excited lines are further in

Reverberation Mapping

For Keplerian rotation, the FWHM of the lines should correspond to the typical velocity dispersion at the radius where the line is produced.

More highly ionized lines, which are closer in, should have largerFWHM and shorter time-lags.

Size of BLR

Reverberation Mapping

Luminous AGN are classified as:

• Seyfert galaxies (Type I and II)• Quasars, QSOs• BL Lacs• Radio galaxies (in `Broad line’ and `Narrow line’ variants)• LINERs

All powered by accretion onto supermassive black holes.But why so many classes - are these all physically distinctobjects?

AGN Unification Schemes

An AGN consists of the following basic ingredients:• Black Hole (power source)• Accretion Disk (UV/x-rays)• Jet (radio)• -Core (compact, flat-spectrum, radio-to-gamma emission)• -Jet• -Lobes & Hotspots (extended, steep spectrum)

• Broad-Line Region (BLR)• Narrow-Line Region (NLR)• molecular (dusty) “torus" (feeding and obscuration)• host galaxy (feeding)

AGN: Basic Ingredients

Seyferts 1 and 2: Unification Scheme

In the broad-line region (BLR)

The Keplerian orbital speeds of the clouds around the central massive body will be large => lines are Doppler broadened.

Density is high => no forbidden lines are emitted

In the narrow-line region (NLR)

The Keplerian orbital speeds of the clouds will be much smaller => lines are narrow

Density is low => forbidden lines are emitted

So, if the above Seyfert were viewed from direction (1), you would see:

Broad permitted lines

Narrow Forbidden Lines

Bright continuum from the central engine

i.e. a Seyfert 1

If, on the other hand, it were viewed from direction (2), you would see:

No broad permitted lines (obscured by dust torus)

Narrow Forbidden Lines

No bright continuum from the obscured central engine

except in the infrared and X-ray region, which gets through the dust

i.e. a Seyfert 2

Seyferts 1 and 2: Unification Scheme

Seyferts 1 and 2: Unification Scheme: Evidence for Torus