---
title: "Read the evidence for colliding tidal streams"
description: "Read pinned stream-precession and AT 2020mot papers, distinguish measured polarization from a collision model, and test host-light dilution."
canonical_url: "https://sandboxphysics.com/stories/tidal-disruption-fallback/reading-stream-collisions-and-polarization/"
language: "en"
page_type: "story"
alternate_languages:
  zh-CN: "https://sandboxphysics.com/zh/stories/tidal-disruption-fallback/reading-stream-collisions-and-polarization/"
last_modified: "2026-11-16"
---

L4 · Method guide

# Read the evidence for colliding tidal streams

Sandbox Physics · 2026-11-16

Open Liodakis and colleagues’ [arXiv:2208.14465v2](https://arxiv.org/pdf/2208.14465v2), Figures 1–2 and supplementary Table S1. You will distinguish the measured polarization of AT 2020mot from the proposed stream-collision explanation, then calculate how unpolarized host light dilutes a signal. This is a reading and analytic exercise.

## Read the three panels on one clock

Figure 1 places an optical light curve, polarization degree and polarization angle above the same Modified Julian Date axis. Follow a single observing date vertically before comparing peaks. The green symbols in the middle panel include a host-galaxy correction; the uncorrected symbols answer a different measurement question. Supplementary Table S1 supplies the observed polarization and normalized Stokes parameters. The error bars represent 68-percent intervals. The paper’s corrected peak is about twenty-five percent, with an uncertainty of four percentage points.

Polarization describes an imbalance between directions of the light’s electric-field oscillation. A polarimeter compares intensities through different analyser orientations. It provides more information than the total brightness alone, because the directions can change even when the summed light remains similar.

Define the total intensity $I$ and linear Stokes parameters $Q$ and $U$. They have the same intensity units. The polarization fraction and orientation are

$$
\begin{gathered}
p=\frac{\sqrt{Q^2+U^2}}{I},\\
\chi=\frac12\operatorname{atan2}(U,Q).
\end{gathered}
$$

The factor of one half reflects the fact that a linear-polarization direction repeats after half a turn. Use the two-argument arctangent to retain the correct quadrant. Multiplying $p$ by one hundred expresses it as a percentage.

Here is a useful independent exercise: add light with the same intensity and polarization fraction at two perpendicular orientations. The Stokes vectors cancel, so the combined light can have zero polarization even though each component is polarized. Low polarization by itself therefore permits several geometries. Also, taking the magnitude of noisy $Q$ and $U$ produces a positive value even when their true means vanish; a small measured fraction requires an uncertainty and bias treatment.

## Remove host light before comparing fractions

Suppose an unresolved source sits inside a galaxy. If the galaxy light is unpolarized, it adds to $I$ but contributes zero to $Q$ and $U$. Let the host intensity be $I_{\rm h}$ and source intensity $I_{\rm s}=I-I_{\rm h}$. The polarized intensity stays the same, so

$$
p_{\rm s}=p_{\rm obs}\frac{I}{I-I_{\rm h}}.
$$

This simple derivation also explains why the measurement aperture matters: it determines how much galaxy light enters the denominator. Colour and time matching matter when source and host measurements come from different filters or epochs. Read the supplementary polarimetry discussion beside the corrected points in Figure 1.

**Worked example.** Take an observed polarization of seven percent and an unpolarized host contributing seventy-two percent of the aperture’s total light. The source supplies the remaining twenty-eight percent. Dividing seven by $0.28$ gives an intrinsic twenty-five percent. These are chosen teaching inputs, not a remeasurement of one Table S1 row.

![For seven percent observed polarization, correcting increasing unpolarized host light raises intrinsic polarization; a teaching host fraction of seventy-two percent gives twenty-five percent.](https://sandboxphysics.com/astronomy/ac53/figures/host-dilution.en.svg)

Analytic dilution example with seven-percent observed polarization. The correction becomes sensitive to the assumed host intensity when the source supplies only a small fraction of the light. The marked seventy-two-percent host fraction is a teaching choice.

Differentiate the correction with respect to the host fraction $h=I_{\rm h}/I$. At fixed observed fraction, $dp_{\rm s}/dh=p_{\rm obs}/(1-h)^2$. For the example, an uncertainty of $0.03$ in host fraction alone contributes about $0.027$ to the inferred source fraction, or $2.7$ percentage points. The full uncertainty must also include the observed polarization and total flux, with their correlations where relevant.

## Read Figure 2 as a physical proposal

Figure 2 sketches the authors’ proposed sequence of shocks during disk formation. It is a qualitative model diagram, distinct from Figure 1’s observations. The authors use polarization together with radio and X-ray constraints to argue for tidal-stream shocks rather than a bright jet. Their interpretation leaves room for a weak disk below the observational limits. Those statements do not turn the polarization angle into a resolved photograph of a collision site.

Ask what calculation would connect the two figures. It needs gas trajectories that meet at the same place and time, shock heating, magnetic-field geometry and radiation capable of producing the measured Stokes parameters. A ballistic orbit alone supplies only part of that chain.

## Add relativistic precession to the orbital picture

For the geometric step, open [Dai, McKinney and Miller, arXiv:1507.04333v2](https://arxiv.org/pdf/1507.04333v2), Section 2, Equations 1 and 7, and Figure 1. A Kepler ellipse closes after each turn. Around a nonrotating black hole, the nearest point instead advances. In the weak-field approximation,

$$
\begin{gathered}
\Delta\omega\simeq\frac{6\pi r_{\rm g}}{a(1-e^2)},\\
r_{\rm g}=\frac{GM_\bullet}{c^2}.
\end{gathered}
$$

Here $a$ is semimajor axis, $e$ eccentricity and $r_{\rm g}$ the gravitational radius. Imagine two identical ellipses rotated by this angle and put $\ell=a(1-e^2)$. Equating their polar equations gives a far-side geometric crossing radius

$$
r_{\rm cross}=\frac{\ell}{1-e\cos(\Delta\omega/2)}.
$$

The expression reveals why a small change in precession can move a high-eccentricity crossing substantially. A geometric crossing still needs simultaneous gas occupation to make a collision. Stream thickness, cooling and spin-induced orbital-plane changes add further physics. The browser’s Newtonian impulse model omits these effects; the paper guide adds the geometric argument as a separate calculation.

## Check how representative the event is

An unusually polarized event need not typify every disruption. [Floris and colleagues, arXiv:2509.08895v1](https://arxiv.org/pdf/2509.08895v1), Section 4 and Figure 2, compare thirteen events and discuss why AT 2020mot remains unusual in polarization even when several other inferred properties resemble the sample. That accepted 2025 manuscript is an additional comparison, not a rerun of the 2023 analysis.

For the next paper, mark which panels contain observations, which contain corrections and which contain model predictions. Note the aperture and host subtraction, the origin of every time axis, the limits behind a non-detection and the gas and radiation assumptions needed to predict polarization. Then ask which one new observation could distinguish the competing explanations.

## Sources

[Liodakis and colleagues, arXiv:2208.14465v2](https://arxiv.org/pdf/2208.14465v2), version dated 27 November 2023; published in Science 380, 656–658. Reading scope: main Figures 1–2, observational and interpretation sections, supplementary Table S1 and host-light correction discussion. No new polarimetric reduction or full shock-model reproduction is claimed.

[Dai, McKinney and Miller, arXiv:1507.04333v2](https://arxiv.org/pdf/1507.04333v2), Section 2, Equations 1 and 7 and Figure 1, for the simplified precessing-ellipse construction.

[Floris and colleagues, arXiv:2509.08895v1](https://arxiv.org/pdf/2509.08895v1), Section 4 and Figure 2, for the thirteen-event polarization comparison. Sources and selected passages checked on 16 November 2026.

## Try the prediction

### [Find the point where ballistic motion ends](https://sandboxphysics.com/labs/tidal-disruption-fallback/)

1. Predict what happens when two gas streams occupy the same place at the same time.
2. Inspect the returning stream in the Lab and note which forces are included.
3. Explain which shock, magnetic-field and radiation calculations the paper adds to a ballistic trajectory.

## Keep reading

Follow stellar debris back to a black hole · Reading path

### Follow stellar debris back to a black hole

1. [Article **Why a shredded star can take weeks to brighten**](https://sandboxphysics.com/stories/tidal-disruption-fallback/why-a-shredded-star-brightens-late/)
2. [Article **Give the debris its orbital energies**](https://sandboxphysics.com/stories/tidal-disruption-fallback/tides-and-frozen-orbital-energy/)
3. [Article **Turn orbital energies into a return-rate curve**](https://sandboxphysics.com/stories/tidal-disruption-fallback/orbital-periods-and-fallback/)
4. [Article **Let returning gas become escaping light**](https://sandboxphysics.com/stories/tidal-disruption-fallback/diffusion-and-observed-light/)
5. [Article **Replay debris orbits beside a real X-ray light curve**](https://sandboxphysics.com/stories/tidal-disruption-fallback/replaying-debris-and-a-light-curve/)
6. [You are here **Read the evidence for colliding tidal streams**](https://sandboxphysics.com/stories/tidal-disruption-fallback/reading-stream-collisions-and-polarization/)
