🔭 astrophysics

UV and Optical Signatures of Late-time Disk Instabilities in Tidal Disruption Events

This paper models the brief UV and optical transients predicted to arise from late-time thermal instabilities in tidal disruption event accretion disks, estimates their detectability by upcoming surveys, and reports initial constraints from existing Zwicky Transient Facility data to test the hypothesis that such instabilities drive late-time radio emission.

Original authors: Daichi Tsuna, V. Ashley Villar, Anthony L. Piro, Samantha C. Wu

Published 2026-02-18

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Daichi Tsuna, V. Ashley Villar, Anthony L. Piro, Samantha C. Wu

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a cosmic drama playing out in the deep universe. A star wanders too close to a giant, invisible monster—a supermassive black hole. The monster's gravity is so strong it rips the star apart, like a piece of taffy being stretched until it snaps. This event is called a Tidal Disruption Event (TDE).

Usually, when this happens, the star's debris swirls around the black hole, forming a hot, glowing disk of gas. It shines brightly in ultraviolet (UV) light and then slowly fades away.

But here is the plot twist: Astronomers have noticed that about 40% of these events have a weird "sequel." Hundreds or even thousands of days after the star is destroyed, they suddenly start glowing in radio waves again. It's like a campfire that died out, only to suddenly burst into flames again weeks later. Nobody knew exactly why this happened.

The New Theory: The Black Hole's "Heartburn"

This paper proposes a new explanation. The authors suggest that the disk of gas swirling around the black hole isn't smooth and steady. Instead, it gets "sick" with thermal instabilities.

Think of the accretion disk like a pot of soup on a stove. Usually, it simmers gently. But sometimes, the heat gets too intense, and the soup suddenly boils over violently. In the black hole's case, the disk gets so hot and unstable that it suddenly dumps a massive amount of material onto the black hole all at once.

This "dump" happens so fast that the black hole can't eat it all. It gets overwhelmed (a state called super-Eddington accretion). Because it's eating too fast, it starts "spitting" out a powerful wind of gas, shooting it away at incredible speeds.

The Hidden Clue: A Flash of UV Light

The authors realized that if this "spitting" wind happens, it shouldn't just create radio waves. It should also create a very specific, short-lived flash of ultraviolet (UV) and visible light.

Here is the analogy:

Imagine the black hole is a giant, invisible engine.
The "instability" is a sudden surge of fuel.
The engine revs up, creating a massive exhaust wind.
As that wind expands and cools, it glows brightly for a few days, like a flare from a rocket booster.

The Prediction:
The paper predicts that if this theory is true, we should see a brief, bright flash of UV light (lasting only a few days) right when the disk gets unstable.

Brightness: It would be incredibly bright, outshining entire galaxies in UV light for a moment.
Duration: It would be very short, like a camera flash, lasting only 2 to 5 days.
Timing: It would happen years after the star was first destroyed.

The Hunt: Looking for the Flash

The team didn't just sit back and guess; they went looking for these flashes.

The Future Hunt: They calculated that powerful new telescopes (like ULTRASAT and the Vera C. Rubin Observatory) will be able to spot these flashes in galaxies up to 1 billion light-years away. If we see these UV flashes and the late radio waves at the same time, it will be the "smoking gun" proving the disk instability theory.
The Past Hunt: They also looked at old data from the Zwicky Transient Facility (ZTF), a telescope that has been scanning the sky for years. They checked 28 nearby TDEs to see if any of them had these mysterious flashes in the past.

The Result:
They didn't find any definite flashes yet. However, they didn't find nothing, either. They found that if these flashes do happen, they can't be happening more than once or twice a year for any single black hole. This puts a limit on how wild the black hole's "heartburn" can be.

Why Does This Matter?

This is a detective story about the universe.

If they find the flash: It proves that black hole disks are unstable and chaotic, and it explains why we see those late radio signals. It tells us how black holes "eat" and how they push gas back out into the universe.
If they don't find it: It means the theory is wrong, and we need to come up with a new explanation for the late radio signals.

In a nutshell: The authors are saying, "Black holes might have a bad temper, causing them to spit out gas in violent bursts. If we look closely with the right UV cameras, we should see a bright flash every time they do it. We haven't seen it yet, but we're getting better at looking, and soon we might catch them in the act."

1. Problem Statement

Tidal Disruption Events (TDEs) occur when a star is torn apart by a supermassive black hole (SMBH). While early-time emission is well-studied, a significant fraction of TDEs exhibit late-time radio brightening (hundreds to thousands of days after the optical peak). The origin of the delayed, mildly relativistic outflows powering this radio emission remains unclear.
Recent theoretical models suggest that TDE accretion disks may undergo late-time thermal instabilities, leading to brief episodes of super-Eddington accretion. These instabilities could launch the delayed outflows. However, a critical gap exists: what are the unique electromagnetic signatures of these instabilities in the UV and optical bands? Detecting these signatures is essential to confirm the disk instability scenario as the driver of late-time radio emission.

2. Methodology

The authors developed a semi-analytical model to predict the UV/optical emission resulting from accretion flares driven by disk instabilities.

Accretion Flare Model:
- Inspired by time-dependent one-zone modeling (Piro & Mockler 2025), the model assumes a flare ejects a mass $M_{fl}$ over a duration $t_{fl}$ .
- Parameters: The flare duration is fixed at $t_{fl} \approx 1\text{--}2$ days (viscous timescale). The ejected mass is empirically set to $M_{fl} = 10^{-2}\text{--}10^{-1} M_\odot$ to match late-time radio observations. The characteristic disk radius $r_d$ is varied, typically around $5r_T$ (tidal radius).
- Super-Eddington Outflow: The flare drives a quasi-spherical, optically thick wind with a mass-loss rate $\dot{M}_w \gg \dot{M}_{Edd}$ . The wind is launched from a radius $r_{in}$ (either the spherization radius or the disk radius).
Radiative Transfer & Light Curve Calculation:
- The wind is modeled as a series of expanding "shells."
- Energy Evolution: The authors solve the energy equation for internal energy ( $E_{int}$ ), accounting for adiabatic cooling ($PdV$ work) and radiative cooling regulated by photon diffusion ( $t_{diff}$ ).
- Luminosity: The observed luminosity $L_{obs}$ is derived by integrating the radiative output of all shells, accounting for adiabatic losses which reduce the observed luminosity below the input accretion power.
- Temperature Evolution: The authors calculate the "color radius" ( $r_c$ $r_{c}$ ), the outermost radius where thermalization occurs. They distinguish between two regimes:
  1. Trapping-dominated: When the trapping radius $r_{tr} > r_c$ , temperature is set by diffusion at $r_{tr}$ .
  2. Thermalization-dominated: When $r_c > r_{tr}$ , temperature is determined by the opacity (Kramers' law) at the color radius.
- Spectra: The model assumes greybody spectra to convert bolometric luminosity and temperature into AB magnitudes for Near-UV (250 nm) and Optical (600 nm) bands.
Observational Search:
- The authors conducted a search for these transients in 28 nearby TDEs ( $z < 0.1$ ) using data from the Zwicky Transient Facility (ZTF).
- They searched for short-duration flux excesses ( $\pm 3\text{--}50$ days) on top of the late-time plateau emission.
- Upper Limits: For three well-monitored TDEs (AT 2019qiz, AT 2019azh, AT 2021ehb), they performed artificial signal injections to derive 90% confidence upper limits on the flare rate.

3. Key Contributions

Prediction of UV/Optical Transients: The paper provides the first detailed prediction of the UV/optical light curves associated with late-time disk instabilities in TDEs.
Analytical Framework: It establishes a robust semi-analytical framework for calculating the emission from time-dependent, optically thick super-Eddington outflows, specifically handling the transition between advection-dominated and diffusion-dominated regimes.
Survey Feasibility Study: It quantifies the detectability of these events for current and future surveys (ULTRASAT, Rubin/LSST, Argus Array).
Empirical Constraints: It places the first observational upper limits on the rate of such flares in existing TDE samples.

4. Results

Theoretical Predictions

Light Curve Morphology: The model predicts a brief, UV-bright transient lasting 3–5 days. The light curve rises, settles into a short plateau (due to the top-hat assumption of the flare), and then drops sharply as the outflow shuts off and diffusion timescales evolve.
Luminosities:
- Near-UV (250 nm): Peak luminosities of $10^{42} \text{--} 10^{43} \text{ erg s}^{-1}$ , corresponding to magnitudes of $-17$ to $-19$ mag.
- Optical (600 nm): Peak luminosities of $10^{41} \text{--} 10^{42} \text{ erg s}^{-1}$ , corresponding to magnitudes of $-14$ to $-17$ mag.
Temperature: The emission is very hot, with temperatures reaching $\sim 10^5$ K early on, cooling to $\sim 10^4$ K.
Detectability:
- ULTRASAT: Capable of detecting these UV flares in TDEs out to redshift $z \approx 0.1$ due to its high sensitivity (22.5 mag) and 4-day cadence.
- Rubin/Argus: Optical surveys like LSST and the Argus Array could detect the optical counterparts, particularly for nearby events ( $z < 0.1$ ), though they are fainter than the UV emission.

Observational Search (ZTF Data)

Candidates: No strong candidates consistent with the disk instability model were found.
- Two high-significance excesses (AT 2020vdq, AT 2022dbl) were identified but ruled out as they exhibited slow declines and high luminosities inconsistent with the short-duration flare model.
- Two lower-significance candidates (AT 2021nwa, AT 2019azh) were likely noise or non-flare variations, as they lacked multi-band consistency.
Rate Constraints:
- For the three most constraining TDEs, the non-detection implies an upper limit on the flare rate of $\lesssim 1\text{--}2 \text{ yr}^{-1}$ per TDE.
- This constraint is marginal; it does not yet rule out the disk instability model (which predicts $\sim 1$ flare per event), but it suggests the events are rare or the model parameters (mass, radius) need refinement.

5. Significance

Testing Disk Instability Theory: This work provides a concrete, testable prediction (brief UV/optical flares) for the disk instability hypothesis. A detection would simultaneously explain the late-time radio emission and the disk physics.
Multi-messenger Synergy: The paper emphasizes that combined UV/optical and radio monitoring is crucial. Radio data can constrain the outflow launch time and radius, while UV/optical data can confirm the thermal instability mechanism.
Future Prospects: The results highlight the critical role of upcoming high-cadence UV surveys (ULTRASAT) and deep optical surveys (Rubin/LSST, Argus) in the next decade. These facilities will be able to probe the parameter space of $M_{fl}$ and $r_d$ effectively, potentially confirming or ruling out the thermal instability scenario as the driver of late-time TDE activity.
Model Refinement: The authors note that future improvements should include 1D/2D disk modeling, delayed circularization effects, and detailed radiative transfer to better predict optical fluxes and flare timing.