Blazars and Relativistic Beaming

A blazar is the same object as a radio galaxy seen from a privileged seat. In a jetted active galactic nucleus a spinning supermassive black hole launches a magnetized plasma jet at bulk Lorentz factor $\Gamma\sim10$ – $50$ ; when that jet points within a few degrees of our line of sight, special relativity does something spectacular to the light. Aberration sweeps radiation the jet emits nearly isotropically in its own frame into a forward cone of half-angle $\sim1/\Gamma$ , and an observer inside that cone sees a source blueshifted, brightened by orders of magnitude, and flickering on impossibly short timescales. This one geometric accident — jet aimed at the eye — turns an ordinary radio galaxy into the most luminous and violently variable persistent object in the sky. Everything blazars do follows from a single number, the Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ , which governs how frequency, solid angle, intensity, and time transform between the jet frame and ours. Boost the frequency and a radio-band source shifts up to GeV and TeV gamma rays; boost the intensity by $\delta^{3+\alpha}$ and a sub-Eddington jet appears a thousandfold super-Eddington; compress the arrival times by $\delta$ and a light-week-sized region seems to double its output in minutes. The same kinematics makes jet knots appear to fly across the sky faster than light — a projection effect, not a violation of relativity. This article builds all of that from the relativistic Doppler transformation and shows how observed superluminal speeds, brightness temperatures, and variability are inverted to measure $\Gamma$ , $\delta$ , and the viewing angle $\theta$ directly.

Key takeaways

A blazar is a radio-loud active galactic nucleus whose relativistic jet points within a few degrees of the line of sight, so its synchrotron and inverse-Compton emission is strongly Doppler-boosted.
One kinematic quantity, the Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ , sets the boost of observed frequency ( $\nu_{\rm obs}=\delta\nu_{\rm em}$ ), timescale ( $\Delta t_{\rm obs}=\Delta t_{\rm em}/\delta$ ), and flux.
Flux boosts as $\delta^{3+\alpha}$ for a discrete moving blob and $\delta^{2+\alpha}$ for a steady jet; the bolometric apparent luminosity boosts as $\delta^{4}$ .
Beaming produces apparent superluminal motion, with sky speed $\beta_{\rm app}=\beta\sin\theta/(1-\beta\cos\theta)$ peaking at $\beta_{\rm app}^{\max}=\Gamma\beta$ near $\theta\approx1/\Gamma$ .
Blazars split into flat-spectrum radio quasars (FSRQs, broad lines, luminous disk) and BL Lacertae objects (BL Lacs, featureless continuum) — the beamed end of the AGN unified model — with double-humped radio-to-TeV spectral energy distributions.
Key idea: every extreme property of a blazar — huge apparent luminosity, minute-scale variability, superluminal knots — is intrinsically ordinary jet physics multiplied by powers of the Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ , because the jet is aimed almost straight at us.

The core insight

Every extreme property of a blazar — huge apparent luminosity, minute-scale variability, superluminal knots — is intrinsically ordinary jet physics multiplied by powers of the Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ , because the relativistic jet is aimed almost straight at us.

Where this lives today

Beaming is now a working instrument. The MOJAVE and Boston-University VLBA programmes monitor hundreds of blazar jets at milliarcsecond resolution, measuring $\beta_{\rm app}$ for individual knots; combined with Doppler factors from variability and synchrotron-self-Compton modelling, these fix $\Gamma\sim5$ – $40$ and $\theta$ of a few degrees, and anchor the misaligned parent radio galaxies. In the gamma-ray sky the Fermi-LAT catalogue is dominated by blazars, and imaging Cherenkov telescopes (H.E.S.S., MAGIC, VERITAS, and the coming Cherenkov Telescope Array) push the inverse-Compton hump into the TeV band, where spectral cutoffs measure the extragalactic background light and pair cascades constrain intergalactic magnetic fields. Minute-scale TeV flares from objects like PKS 2155−304 sharpen the causal-size bound of equation (6), demanding either very large $\delta$ or fine-scale jet structure and keeping the emission-region problem open. And multimessenger astronomy has made blazars candidate cosmic accelerators: the 2017 coincidence of IceCube-170922A with a flaring TXS 0506+056 implicated relativistic proton acceleration in a beamed jet — plausible precisely because both the neutrino output and the gamma-ray signal are enhanced along our line of sight.

The physical picture

Picture a searchlight that is not rotating but firing straight ahead, and whose beam narrows the faster its carriage runs. A source that radiates the same in every direction in its own rest frame appears, once it moves at $\beta\to1$ , to concentrate almost all of that light into a thin cone of half-angle $1/\Gamma$ about its direction of motion — relativistic aberration tips the rest-frame-isotropic photons forward in the frame where the source streams past. If your eye lies inside that cone you see a dazzling, blueshifted, rapidly flickering object; step a few degrees outside and the very same jet looks faint, red, and sluggish. The first figure fixes this geometry and attaches its numbers.

Figure

A relativistic jet launched near a spinning supermassive black hole moves at β ≈ 0.995 (Γ = 10) almost along the line of sight; special-relativistic aberration sweeps its radiation into a forward cone of half-angle ∼ 1/Γ ≈ 5.7°, and an observer inside that cone sees the emission Doppler-boosted by δ = Γ(1+β) ≈ 2Γ = 20 at θ = 0

Time compression is the inseparable twin of frequency boost. A blob approaching you at $\beta\to1$ almost keeps pace with the light it emits, so two pulses launched $\Delta t_{\rm em}$ apart in the jet frame pile up into $\Delta t_{\rm em}/\delta$ on your clock — which is why a blazar can double its gamma-ray flux in minutes though the intrinsic light-crossing time is longer by the full factor $\delta\sim10$ – $30$ . The same near-chase of light produces apparent superluminal motion: a knot moving toward us covers real ground between two flashes, and because the second flash has less far to travel, the sky-projected motion can appear to exceed $c$ . Nothing outruns light — the observer's clock is fooled by a shrinking light-travel time. The second figure plots how fast the sky motion can look for a jet with $\Gamma=10$ .

Figure

Apparent superluminal speed β_app(θ) for a jet with Γ = 10 (β ≈ 0.995): projection plus light-travel compression make knots appear to move at up to β_app = Γβ ≈ 9.95 times c, reached near θ ≈ 5.7° = 1/Γ; everywhere above the dashed line the sky motion looks faster than light

Two further clues confirm the picture. High optical and radio polarization — tens of percent, sometimes swinging during flares — is the fingerprint of synchrotron radiation in an ordered magnetic field. And the spectral energy distribution is unmistakably double-humped: a low-frequency synchrotron bump from relativistic electrons in the jet field, and a high-frequency bump, usually inverse-Compton scattering of soft photons up to GeV–TeV energies. Both humps ride the same Doppler boost, which is why blazars dominate the extragalactic gamma-ray sky.

The core equations

Beaming is pure kinematics, descending from the relativistic Doppler transformation of a photon's frequency combined with aberration. The one quantity that packages the transformation is the Doppler factor; every observable below is a power of it.

\delta = \frac{1}{\Gamma\,(1-\beta\cos\theta)} \qquad (1)

where: $\delta$ — Doppler factor; $\Gamma=(1-\beta^2)^{-1/2}$ — bulk Lorentz factor; $\beta$ — bulk speed in units of $c$ ; $\theta$ — angle between jet velocity and line of sight. Read it as the frequency multiplier: a photon emitted at $\nu_{\rm em}$ arrives at $\nu_{\rm obs}=\delta\,\nu_{\rm em}$ . For $\Gamma=10$ , $\theta=0$ : $\delta=\Gamma(1+\beta)\approx2\Gamma=20$ . Frequency and time transform oppositely under the same factor — crests come faster, and so do the events that mark time.

\nu_{\rm obs}=\delta\,\nu_{\rm em}, \qquad \Delta t_{\rm obs}=\frac{\Delta t_{\rm em}}{\delta} \qquad (2)

where: $\nu_{\rm em},\nu_{\rm obs}$ — jet-frame and observed frequencies; $\Delta t_{\rm em},\Delta t_{\rm obs}$ — emitted and observed timescales. A day-long intrinsic timescale, boosted by $\delta=20$ , appears as about an hour — the origin of blazar intraday variability. The flux boost is stronger than the frequency boost because it also gathers the beamed solid angle and shifts the spectrum. The exponent differs by one between a moving blob and a steady jet, and getting it wrong costs an order of magnitude.

S_{\nu,\rm obs}=\delta^{\,3+\alpha}\,S_{\nu,\rm em}\ \ (\text{moving blob}), \qquad S_{\nu,\rm obs}=\delta^{\,2+\alpha}\,S_{\nu,\rm em}\ \ (\text{continuous jet}) \qquad (3)

where: $S_\nu$ — monochromatic flux density; $\alpha$ — spectral index, $S_\nu\propto\nu^{-\alpha}$ . Integrated over frequency the blob's bolometric apparent luminosity scales even more steeply, $L_{\rm iso}=\delta^{4}L_{\rm em}$ — the extra power comes from boosting the band itself. The most arresting consequence is a knot that appears to cross the sky faster than light, its apparent transverse speed set by true motion and light-travel compression together.

\beta_{\rm app}=\frac{\beta\sin\theta}{1-\beta\cos\theta} \qquad (4)

where: $\beta_{\rm app}$ — sky-projected knot speed in units of $c$ . This is not a Lorentz-transformed velocity but a ratio of projected displacement to observed elapsed time, and it can exceed unity without any physical superluminal motion. Maximizing (4) over viewing angle picks out a special geometry and a clean cap on the apparent speed.

\beta_{\rm app}^{\max}=\Gamma\beta \quad\text{at}\quad \cos\theta=\beta,\ \ \theta\approx1/\Gamma \qquad (5)

where: $\beta_{\rm app}^{\max}$ — largest observable apparent speed. The angle $\cos\theta=\beta$ is also where $\delta=\Gamma$ , so the most strongly superluminal sources are also strongly beamed — which is why flux-limited samples are full of them. Because the timescale is compressed by $\delta$ , the light-crossing bound on an emitting region's size is relaxed by the same factor.

R\lesssim\frac{c\,\delta\,\Delta t_{\rm obs}}{1+z} \qquad (6)

where: $R$ — radius of the emitting region; $\Delta t_{\rm obs}$ — observed variability timescale; $z$ — redshift. Without the $\delta$ , minute-scale gamma-ray flares would demand regions smaller than the black hole's gravitational radius; beaming rescues causality. Finally the beaming geometry itself: aberration collimates the rest-frame-isotropic emission into a forward cone whose narrowness measures $\Gamma$ .

\theta_{\rm cone}\sim\frac{1}{\Gamma}, \qquad \Omega_{\rm beam}\sim\frac{\pi}{\Gamma^2} \qquad (7)

where: $\theta_{\rm cone}$ — half-angle of the forward cone; $\Omega_{\rm beam}$ — solid angle into which the emission is concentrated. Only a fraction $\sim1/(4\Gamma^2)$ of random sightlines lands inside a jet's cone, so for every blazar we see there are $\sim2\Gamma^2$ jets pointing elsewhere — the misaligned radio-galaxy parents.

Derivation AWhy a jet knot can appear to outrun light

The goal is to show that a knot moving strictly slower than light, at speed $\beta c$ toward the observer, can nonetheless appear to sweep across the sky at many times $c$ , and to derive both the apparent-speed formula and its maximum $\Gamma\beta$ . The whole effect is a light-travel-time illusion: because the knot chases its own photons, the second of two flashes has a shorter path to us, so the observed time between them is compressed relative to the true motion. We track two emission events, project onto the sky, correct the arrival times, and maximize over the viewing angle.

Step 1. Let the knot be ejected at speed $\beta c$ at angle $\theta$ to the line of sight, and take two emission events separated by a coordinate interval $\Delta t$ in the host-galaxy frame; the knot travels $\beta c\,\Delta t$ between them. Why: fixing two well-defined events turns a continuous motion into a measurable proper motion on the sky.

Step 2. Only the component perpendicular to the line of sight is seen as motion across the sky, so the projected displacement is $\Delta r_\perp=\beta c\,\Delta t\,\sin\theta$ . Why: an astronomer measures angular position, which responds to transverse motion only; the radial part is invisible.

Step 3. Between the two flashes the knot has approached us by $\beta c\,\Delta t\cos\theta$ , so the second photon travels a shorter path and arrives sooner; the observed interval shrinks to $\Delta t_{\rm obs}=\Delta t\,(1-\beta\cos\theta)$ . Why: this is the crux — the denominator $1-\beta\cos\theta$ that later defines $\delta$ enters purely from light-travel bookkeeping, not from any Lorentz transformation.

Step 4. The apparent sky speed is projected distance over observed time, $\beta_{\rm app}=\Delta r_\perp/(c\,\Delta t_{\rm obs})=\beta\sin\theta/(1-\beta\cos\theta)$ . Why: this can exceed one because the denominator becomes arbitrarily small as $\beta\to1$ and $\theta\to0$ , with no physical speed exceeding $c$ .

Step 5. Setting $\mathrm{d}\beta_{\rm app}/\mathrm{d}\theta=0$ gives $\cos\theta=\beta$ , i.e. $\sin\theta=1/\Gamma$ ; substituting back, $\beta_{\rm app}^{\max}=\Gamma\beta$ , using $1-\beta^2=1/\Gamma^2$ . Why: this names the optimal viewing angle $\theta\approx1/\Gamma$ and turns a measured $\beta_{\rm app}$ into a firm lower bound $\Gamma\ge\sqrt{1+\beta_{\rm app}^2}$ . For $\Gamma=10$ ( $\beta\approx0.995$ ), $\beta_{\rm app}^{\max}=\Gamma\beta\approx9.95$ : a knot can look like it is crossing the sky at nearly ten times the speed of light. Verified against Rybicki & Lightman ch. 4 and Blandford, Meier & Readhead ARA&A 2019.

Go deeper · Derivation B · 1 minThe Doppler boost of flux from an invariant

Derivation BThe Doppler boost of flux from an invariant

The goal is to derive the $\delta^{3+\alpha}$ scaling of observed flux for a discrete moving emitter, and the reduced $\delta^{2+\alpha}$ for a steady jet, from a single relativistic fact: the phase-space photon density $I_\nu/\nu^3$ is a Lorentz invariant. This one statement fixes how specific intensity transforms; the power-law spectrum supplies one extra factor, and the blob-versus-filled-beam geometry supplies the one-power offset. We transform the intensity, re-express it at fixed observed frequency, and insert the spectral shape.

Step 1. The combination $I_\nu/\nu^3$ is invariant because photon number is conserved and the phase-space volume $\mathrm{d}\nu\,\mathrm{d}\Omega\,\mathrm{d}A\,\mathrm{d}t$ transforms to leave it unchanged; hence $I_{\nu,\rm obs}/\nu_{\rm obs}^3=I_{\nu,\rm em}/\nu_{\rm em}^3$ . Why: starting from an invariant means we never transform intensity, solid angle, and time separately — they are already bundled correctly.

Step 2. Since $\nu_{\rm obs}=\delta\,\nu_{\rm em}$ , the intensity at the transformed frequency is boosted by three powers, $I_{\nu,\rm obs}=\delta^{3}\,I_{\nu,\rm em}$ . Why: the three factors of $\delta$ are the frequency boost, the solid-angle compression, and the arrival-time compression, delivered at once by the invariant.

Step 3. We want the intensity at fixed observed frequency $\nu$ , so we evaluate the emitted spectrum at $\nu_{\rm em}=\nu/\delta$ ; for a power law $I_\nu\propto\nu^{-\alpha}$ this gives $I_{\nu,\rm em}(\nu/\delta)=\delta^{\alpha}I_{\nu,\rm em}(\nu)$ . Why: comparing like with like — the same observed band — introduces the spectral index into the exponent.

Step 4. Combining, $I_{\nu,\rm obs}(\nu)=\delta^{3+\alpha}I_{\nu,\rm em}(\nu)$ , and for an unresolved source the flux tracks the specific intensity, so $S_\nu\propto\delta^{3+\alpha}$ . Why: this steep boost ( $\alpha\approx0.7$ gives $\delta^{3.7}$ ) is what makes flux-limited samples so dramatically dominated by beamed sources.

Step 5. A steady continuous jet is not a moving point but a static distribution filling the beam in the observer frame; integrating along it absorbs one factor of $\delta$ , leaving $S_\nu\propto\delta^{2+\alpha}$ . Why: using the blob exponent for a smooth jet biases inferred Doppler factors by a full power of $\delta$ . *Verified against Urry & Padovani (1995) eqs. 1–4; the same invariant gives the counter-jet-to-jet flux ratio $[(1+\beta\cos\theta)/(1-\beta\cos\theta)]^{3+\alpha}$ .*

Go deeper · Derivation C · 1 minHow beaming fakes a super-Eddington luminosity

Derivation CHow beaming fakes a super-Eddington luminosity

The goal is to explain why blazars routinely display apparent isotropic luminosities that exceed, sometimes by factors of hundreds, the Eddington limit of their inferred black-hole masses, without any real violation of that limit. The resolution is that the Eddington argument assumes isotropic emission, whereas a beamed jet radiates anisotropically toward us, so the apparent isotropic-equivalent luminosity over-counts by the bolometric boost $\delta^{4}$ . We write the boost, put in numbers, and compare with the Eddington luminosity.

Step 1. Integrating the blob boost $S_\nu\propto\delta^{3+\alpha}$ over frequency adds one power from shifting the whole band, giving a bolometric apparent luminosity $L_{\rm iso}=\delta^{4}L_{\rm em}$ , where $L_{\rm em}$ is what the jet would radiate if isotropic. Why: $L_{\rm iso}$ is what an observer computes from $4\pi d_L^2\times$ (observed flux), naively assuming the source shines equally in all directions — which a jet does not.

Step 2. A modest Doppler factor produces an enormous boost: $\delta=20$ gives $\delta^4=1.6\times10^{5}$ , five orders of magnitude. Why: entering as the fourth power, even $\delta=10$ versus $\delta=20$ is a factor of sixteen in apparent luminosity.

Step 3. Compare with the Eddington luminosity $L_{\rm Edd}=1.26\times10^{38}\,(M/M_\odot)\ \mathrm{erg\,s^{-1}}$ ; for $M=10^{9}M_\odot$ this is $L_{\rm Edd}\approx1.3\times10^{47}\,\mathrm{erg\,s^{-1}}$ . Why: this is the isotropic luminosity at which radiation pressure balances gravity — the ceiling the apparent luminosity seems to breach.

Step 4. An intrinsic $L_{\rm em}\sim10^{44}\,\mathrm{erg\,s^{-1}}$ , comfortably sub-Eddington, appears as $L_{\rm iso}\sim1.6\times10^{49}\,\mathrm{erg\,s^{-1}}$ — matching the brightest gamma-ray flares and $\sim100\times$ Eddington. Why: no accretion physics is stressed; the excess is entirely an anisotropy artifact — the decisive argument that the emission really is collimated toward us. *Verified: Eddington coefficient from Carroll & Ostlie; the $\delta^{4}$ bolometric scaling follows from integrating the $\delta^{3+\alpha}$ boost over frequency.*

Limiting cases

The Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ controls every beaming effect; the four regimes below map its behaviour from head-on to transverse.

Case	Result	Meaning
Head-on ( $\theta=0$ )	$\delta=\Gamma(1+\beta)\approx2\Gamma$	maximum blueshift and boost; $\Gamma=10$ gives $\delta\approx20$
Beaming angle ( $\cos\theta=\beta$ )	$\delta=\Gamma$	at $\theta\approx1/\Gamma$ the Doppler factor equals the Lorentz factor, and $\beta_{\rm app}$ is maximal at $\Gamma\beta$
Transverse ( $\theta=90^\circ$ )	$\delta=1/\Gamma<1$	emission de-boosted, redshifted, dimmed — why the receding counter-jet is nearly invisible
Non-relativistic ( $\beta\ll1$ )	$\delta\to1$	beaming vanishes; blazar phenomenology disappears, leaving an ordinary two-sided radio source

Common trapConfusing the Lorentz factor Γ with the Doppler factor δ, and mixing up the flux exponents

The bulk Lorentz factor $\Gamma$ and the Doppler factor $\delta$ are not the same number; they coincide only at the single angle $\cos\theta=\beta$ . At $\theta=0$ the Doppler factor is nearly twice as large, $\delta=\Gamma(1+\beta)\approx2\Gamma$ , and it falls rapidly as $\theta$ grows, dropping below $\delta=1$ once $\theta\gtrsim\sqrt{2/\Gamma}$ — so quoting a "Doppler factor" without its viewing angle is meaningless. Equally treacherous is the flux exponent: a discrete moving blob boosts as $\delta^{3+\alpha}$ but a steady continuous jet only as $\delta^{2+\alpha}$ , and the bolometric apparent luminosity as $\delta^{4}$ — using the wrong one throws inferred jet powers and Doppler factors off by a full power of $\delta$ .

Worked examples

1 · Reading Γ and θ off a blazar's proper motion and boost (fully worked)

VLBI monitoring shows a jet knot with apparent speed $\beta_{\rm app}=10$ , and the variability brightness temperature implies $\delta=16$ . Find the bulk Lorentz factor $\Gamma$ and viewing angle $\theta$ , and verify self-consistency.

Step 1. Two observables, two unknowns. Solving equations (1) and (4) simultaneously for $\beta$ and $\theta$ gives the standard inversion

\Gamma=\frac{\beta_{\rm app}^2+\delta^2+1}{2\delta}, \qquad \tan\theta=\frac{2\beta_{\rm app}}{\beta_{\rm app}^2+\delta^2-1}.

Step 2. With $\beta_{\rm app}^2=100$ , $\delta^2=256$ : $\Gamma=(100+256+1)/(2\times16)=357/32=\mathbf{11.2}$ , and the true bulk speed $\beta=\sqrt{1-1/\Gamma^2}=\mathbf{0.996}$ .

Step 3. The angle: $\tan\theta=2\times10/(100+256-1)=20/355=0.0563$ , so $\theta=\mathbf{3.2^\circ}$ . The jet points within a few degrees of us — the defining blazar geometry.

Step 4. Consistency check. Reinserting $\Gamma=11.2$ , $\beta=0.996$ , $\theta=3.2^\circ$ into equations (1) and (4) recovers $\delta=16.0$ and $\beta_{\rm app}=10.0$ exactly. Γ ≈ 11.2, θ ≈ 3.2°, β ≈ 0.996 — a fast jet aimed almost straight at us.

2 · The size of a flaring region and its "impossible" luminosity

A blazar at $z=0.34$ doubles its gamma-ray flux in $\Delta t_{\rm obs}=15$ min with $\delta=20$ , reaching apparent isotropic luminosity $L_{\rm iso}=10^{48}\,\mathrm{erg\,s^{-1}}$ ; its black hole is $M=3\times10^{8}M_\odot$ . How big can the emitting region be, and is it really super-Eddington?

Step 1. Causal size from variability, equation (6): with $c=3\times10^{10}\,\mathrm{cm\,s^{-1}}$ , $\delta=20$ , $\Delta t_{\rm obs}=900\,\mathrm{s}$ , $1+z=1.34$ , $R\lesssim(3\times10^{10})(20)(900)/1.34=\mathbf{4.0\times10^{14}\ cm}\approx\mathbf{27\ AU}$ . Without the $\delta$ the inferred size would be twenty times smaller and physically untenable.

Step 2. The Eddington limit: $L_{\rm Edd}=1.26\times10^{38}\times3\times10^{8}=\mathbf{3.8\times10^{46}\ erg\,s^{-1}}$ , so $L_{\rm iso}/L_{\rm Edd}=\mathbf{26}$ — apparently twenty-six times super-Eddington.

Step 3. De-boost to the intrinsic luminosity, $L_{\rm em}=L_{\rm iso}/\delta^4$ . With $\delta^4=20^4=1.6\times10^5$ : $L_{\rm em}=\mathbf{6.3\times10^{42}\ erg\,s^{-1}}$ , so $L_{\rm em}/L_{\rm Edd}=1.7\times10^{-4}$ — under a ten-thousandth of Eddington. The flare looks 26× super-Eddington but is intrinsically deeply sub-Eddington — beaming, not a broken accretion limit, explains it.

Physics of the portrait

Relativistic beaming is less a discovery by one person than a chain of realizations that a factor of $\delta$ sits behind the strangest facts in extragalactic astronomy — predicted before it was seen, confirmed the moment interferometers grew large enough to watch jets move.

Origins · 1 minrelativistic beaming and how this was discovered

relativistic beaming

the kinematic Doppler boost δ = 1/(Γ(1−β cosθ)) of a jet aimed near the line of sight · special-relativistic astrophysics of AGN jets

The idea that a source could appear to move faster than light is older than any blazar. The modern account belongs to Martin Rees, who in 1966 pointed out that a synchrotron source expanding relativistically toward us would show apparent superluminal separations — a projection-plus-light-travel illusion, not a violation of relativity. When very-long-baseline interferometry resolved the quasars 3C 273 and 3C 279 in 1971 and caught their radio knots separating at several times $c$ , Rees's prediction was vindicated and beaming became a cornerstone. The same bulk motion that faked superluminal speeds had to boost the flux and compress the timescales; Blandford and Königl built the boosting into a quantitative model of compact radio jets at the end of the 1970s, explaining why one-sided jets, sky-high brightness temperatures, and apparent super-Eddington luminosities were the rule. Unification followed: Urry and Padovani in 1995 assembled the evidence that BL Lacs and flat-spectrum radio quasars are pole-on views of the two radio-galaxy classes, an orientation sequence governed by $\theta$ and $\delta$ . What began as a way to save relativity from a superluminal embarrassment is now the tool that measures jet speeds and — since the 2017 neutrino from TXS 0506+056 — even the sites of cosmic-ray acceleration.

History

1929

— Hoffmeister catalogues BL Lacertae as an irregular variable star

1966

— Rees predicts apparent superluminal motion from relativistic bulk flow toward the observer

1971

— VLBI reveals radio knots in 3C 279 and 3C 273 separating at several times c

1995

— Urry and Padovani frame radio-loud AGN as a single orientation sequence

2017

— an IceCube neutrino coincides with a gamma-ray flare of the blazar TXS 0506+056

Going further

Texts. Carroll & Ostlie, An Introduction to Modern Astrophysics, 2nd ed., ch. 28 (Active Galaxies) · Rybicki & Lightman, Radiative Processes in Astrophysics, ch. 4 (relativistic beaming and Doppler transformations) · Ryden, Introduction to Cosmology, 2nd ed. (luminosity distance for AGN) · Krolik, Active Galactic Nuclei · Beckmann & Shrader, Active Galactic Nuclei.

Provenance. The Doppler factor, aberration relation, and $I_\nu/\nu^3$ invariance follow Rybicki & Lightman ch. 4; a verifier should reproduce $\beta_{\rm app}=\beta\sin\theta/(1-\beta\cos\theta)$ and its maximum $\Gamma\beta$ by differentiation (yielding $\cos\theta=\beta$ ), and the $\delta^{3+\alpha}$ flux boost from Lorentz invariance. The inversion $\Gamma=(\beta_{\rm app}^2+\delta^2+1)/2\delta$ , $\tan\theta=2\beta_{\rm app}/(\beta_{\rm app}^2+\delta^2-1)$ solves equations (1) and (4) simultaneously and is checked numerically in worked example 1. The Eddington coefficient $1.26\times10^{38}(M/M_\odot)\,\mathrm{erg\,s^{-1}}$ is from Carroll & Ostlie; the $\delta^4$ bolometric boost follows by integrating $\delta^{3+\alpha}$ over frequency. FSRQ/BL Lac unification and typical $\Gamma\sim5$ – $40$ , $\theta$ of a few degrees follow Urry & Padovani, PASP 107, 803 (1995); the superluminal prediction is Rees, Nature 211, 468 (1966); the neutrino association is IceCube Collaboration et al., Science 361 (2018).

Builds on: Active Galactic Nuclei: Unified Model · Relativistic Jets and Launching Mechanisms · Supermassive Black Holes · Black Hole Spin and the Ergosphere. Leads to: Quasars and Quasar Evolution · AGN Feedback · Eddington Luminosity and Accretion Limits.

Ask the tutor: Ask me to walk you through inverting a real blazar's proper motion and brightness temperature into its jet speed and viewing angle, or to show why the counter-jet vanishes.

What is the difference between the Lorentz factor Γ and the Doppler factor δ?

The Lorentz factor $\Gamma=(1-\beta^2)^{-1/2}$ describes how fast the jet plasma actually moves — a property of the flow alone. The Doppler factor $\delta=[\Gamma(1-\beta\cos\theta)]^{-1}$ describes how that motion looks to you, so it depends on both the speed and the viewing angle. They are equal only at $\cos\theta=\beta$ . Straight down the jet ( $\theta=0$ ), $\delta\approx2\Gamma$ , nearly double; from the side ( $\theta=90^\circ$ ), $\delta=1/\Gamma<1$ and the source is dimmed. All the observable boosts — flux, frequency, timescale — are powers of $\delta$ , not of $\Gamma$ , so quoting one without the angle is meaningless.

How can a jet knot appear to move faster than light without breaking relativity?

Nothing physical exceeds $c$ ; the illusion is a light-travel-time effect. A knot moving almost toward us covers real ground between two flashes, but because it has come closer, the second flash arrives sooner than a naive clock expects. The observed interval is compressed by $1-\beta\cos\theta$ while the sky-projected distance is $\beta c\,\Delta t\sin\theta$ ; their ratio $\beta_{\rm app}=\beta\sin\theta/(1-\beta\cos\theta)$ can exceed one — reaching $\Gamma\beta$ near $\theta\approx1/\Gamma$ — even though the true speed is strictly below $c$ . The apparent speed is not a velocity in any frame but a bookkeeping ratio of projected displacement to compressed arrival time.

Why can blazars vary in minutes when the black hole is light-hours across?

Because observed timescales are Doppler-compressed by $\Delta t_{\rm obs}=\Delta t_{\rm em}/\delta$ . A region that physically takes an hour to respond appears to vary in a few minutes when $\delta\sim20$ . Turned around as $R\lesssim c\,\delta\,\Delta t_{\rm obs}/(1+z)$ , a fifteen-minute flare implies a region tens of AU across rather than the sub-gravitational-radius size the un-boosted estimate would demand. Without beaming, minute-scale TeV flares would be causally impossible.

What distinguishes a BL Lac object from a flat-spectrum radio quasar?

Both are blazars — jets aimed at us — but they differ in accretion state and emission lines. Flat-spectrum radio quasars (FSRQs) have a luminous disk and broad emission lines, and their high-energy hump is usually inverse-Compton scattering of external photons (disk, broad-line region, torus). BL Lacertae objects (BL Lacs) show weak or absent lines and a nearly featureless jet continuum, with the high-energy hump often from synchrotron-self-Compton. In the unified picture they are the beamed counterparts of the two radio-galaxy classes (FR II and FR I), along a sequence in which more luminous, external-photon-dominated sources peak at lower frequencies.

If we cannot measure the viewing angle directly, how do we get δ?

Several independent routes, all exploiting a quantity that beaming distorts. Variability brightness temperatures far above the $\sim10^{12}\,\mathrm{K}$ inverse-Compton ceiling imply $T_{B,\rm obs}=\delta\,T_{B,\rm em}$ , so the excess measures $\delta$ . Synchrotron-self-Compton modelling compares predicted and observed X-ray flux for a source of known size, and the mismatch scales with a high power of $\delta$ . The requirement that GeV–TeV gamma rays escape without being absorbed on the source's own radiation sets a lower bound on $\delta$ . Combined with the VLBI proper motion $\beta_{\rm app}$ , any one $\delta$ estimate yields both $\Gamma$ and $\theta$ through the inversion in worked example 1.

If beaming is so selective, how many jets are we missing?

Many. The forward cone subtends a solid angle $\Omega_{\rm beam}\sim\pi/\Gamma^2$ , so a randomly oriented jet points within it with probability only $\sim1/(4\Gamma^2)$ — for $\Gamma=10$ , under one in four hundred. For every blazar we catch beamed toward us there are of order $2\Gamma^2\sim200$ identical jets pointing elsewhere, seen instead as ordinary radio galaxies with de-boosted, often two-sided jets. This is the quantitative backbone of the unified model: the parent population's space density, inferred from the beamed sample, must match the observed radio-galaxy counts — and it does, which is among the strongest evidence that beaming and unification are correct.