Skip to main content
Sandbox Physics

AGN Reverberation Mapping

How does a light echo measure gas around a black hole? Gas re-emits a change in the central light after that light reaches it. The extra travel time reveals the gas distribution; the line width adds its motion. Connecting those measurements to a black-hole mass also requires the gas geometry and dynamics.

Drag the broad-line disk to change its tilt. Follow one gas distribution from its delayed glow to a velocity–delay map and an echoed light curve.

Interactive modelAGN Reverberation Mapping
Response-weighted delay⟨τ⟩=20 d\langle\tau\rangle=20\,\mathrm d
Line dispersionσv=1820 km s−1\sigma_v=1820\,\mathrm{km\,s^{-1}}
Correlation peak lagτpeak=15.5 d\tau_{\rm peak}=15.5\,\mathrm d
Virial productVP=12.9×106M⊙\mathrm{VP}=12.9\times10^6 M_\odot
Approaching gasReceding gasBright glow: current echoArrow: toward the observer
Model arrival timet=32 dt=32\,\mathrm d

The source and all curves are simulated. The same cloud weights produce the map and the line light curve; a correlation peak need not equal the average delay.

Velocity reveals which gas answers first
vlos [km s−1]v_{\rm los}\ [\mathrm{km\,s^{-1}}]τ [d]\tau\ [\mathrm d]

Brighter cells carry more response; blue gas approaches and red gas recedes. The gold line marks elapsed delay since the first continuum peak.

A sharp flash returns as a broader echo
Central continuumLine echo
t [d]t\ [\mathrm d]F/⟨Fbase⟩F/\langle F_{\rm base}\rangle
Response distribution, correlation and mass
How the light is spread in delay
τ [d]\tau\ [\mathrm d]∑vΨv,τ\sum_v\Psi_{v,\tau}
Cross-correlation of these two curves
τ [d]\tau\ [\mathrm d]r(τ)r(\tau)
VP=c⟨τ⟩σv2G,M=f VP\mathrm{VP}=\frac{c\langle\tau\rangle\sigma_v^2}{G},\qquad M=f\,\mathrm{VP}

The virial product combines the response-weighted delay with the simulated line dispersion. A geometric and dynamical factor is still needed for a mass. A face-on circular disk has no projected line width in this idealization.

The extra path turns a flash into a ruler

BackgroundThe central source brightens before the surrounding gas responds. Light travels from the source to the gas, and then from the gas to us. Subtract the direct route and the remaining travel time measures a geometric path difference.

Why it mattersEven when a telescope cannot resolve the broad-line region, a monitoring campaign can measure when its line emission changes. The delay and the Doppler width then supply different pieces of information about the gas.

Focus question
Do all parts of a ring at one radius respond at the same arrival time?
One-sentence intuition
Only a face-on ideal ring has one delay. Tilting the ring makes its near side answer before its far side, even though both lie at the same distance from the source.

Core mathematical model

Extra light-travel distance

τ=r−zc,n=ez\tau=\frac{r-z}{c},\qquad \boldsymbol n=\boldsymbol e_z

The coordinate toward the distant observer shortens the final path. The common source-to-observer path is subtracted. These model arrival times assume zero cosmological redshift.

The delay range of a thin ring

Rc(1−sin⁡i)≤τ≤Rc(1+sin⁡i)\frac{R}{c}(1-\sin i)\leq\tau\leq\frac{R}{c}(1+\sin i)

Inclination is measured from the disk normal. The mean delay stays at the radius divided by light speed for an axisymmetric ring, while the range widens as the disk tilts.

One distribution creates an entire light curve

L(t)=∫0∞Ψ(τ)C(t−τ) dτ∫0∞Ψ(τ) dτ=1\begin{gathered}L(t)=\int_0^\infty\Psi(\tau)C(t-\tau)\,\mathrm d\tau\\[0.5em]\int_0^\infty\Psi(\tau)\,\mathrm d\tau=1\end{gathered}

Each gas sample echoes an earlier part of the same continuum. Adding its weighted contribution broadens the flash. The line responds linearly and instantaneously in this model.

A line width and a delay give a virial product

VP=c⟨τ⟩σv2G,M=f VP\mathrm{VP}=\frac{c\langle\tau\rangle\sigma_v^2}{G},\qquad M=f\,\mathrm{VP}

The velocity dispersion is measured along the sightline. Converting this product to mass needs a factor that depends on geometry and dynamics. The correlation peak is separately computed from the two light curves.

Common difficulties

The source-to-gas distance is only part of the route

Typical misconceptionEvery gas cloud at the same radius has the same observed delay.

Better mental modelThe onward path to us also matters. Near-side and far-side photons travel different total distances.

A correlation peak is not a complete geometric map

Typical misconceptionThe correlation peak gives one exact gas radius and a unique black-hole mass.

Better mental modelThe peak depends on the response distribution and on the driving continuum. Projected velocity also depends on inclination and any radial motion.

Run the experiment

  1. 01

    Watch the first echo

    Leave the default disk playing. Compare the bright arc with the moving cursor on the two light curves.

    What to observe: Different parts of the same distribution respond at different times. The total line curve is broader than the driving flash.
  2. 02

    Tilt the disk

    Drag the gold handle, then compare the face-on ring with the strongly tilted preset.

    What to observe: A face-on ring has one delay and zero line-of-sight circular velocity. Tilting opens both the delay spread and the Doppler width.
  3. 03

    Add inward motion

    Choose the inward-motion preset and inspect the velocity–delay map. Compare it with circular motion at the same orientation.

    What to observe: The near side and far side acquire different radial-velocity signs. Which side of the line responds early now changes in a recognizable way.
  4. 04

    Save the response

    Pause and save the response. The download contains the same weighted gas samples, transfer bins and light curves shown here.

    What to observe: The accompanying method guide separately uses an actual monitoring record, where sampling and measurement errors enter the lag estimate.

Related experiments and articles

What this page links to directly

See every article series on the content map →

Measure a black hole with light echoes · Reading path

Measure a black hole with light echoes

Articles on this path that use this experiment, followed by the other experiments they suggest.

  1. ArticleHow a black hole can be measured with an echo
  2. ArticleFind the extra distance travelled by an echo
  3. ArticleTurn an orbiting ring into a velocity–delay map
  4. ArticleMeasure a lag before turning it into a mass
  5. ArticleRecompute an observed echo in NGC 5548
  6. ArticleRead a velocity–delay map as an inference from changing spectra
Something broken, confusing, or missing on this page?Tell us
What kind of feedback?

Sent with this page’s address, its Lab ID, the site language and your device type. Please leave out personal details. Feedback is kept for up to 12 months.Privacy policy