🔭 astrophysics

Rate of Repeating Tidal Disruption Events with 5--19 years interval

By analyzing CRTS data for a sample of 16 ZTF BTS TDEs, this study identifies two repeating TDE candidates with 13–17 year intervals, suggesting that repeating partial TDEs may constitute 25–60% of the observed sample and implying that current optical TDE rates are likely overestimated.

Original authors: Yujun Yao, Luming Sun, Tao Wu, Ning Jiang, Shiyan Zhong, Xinwen Shu

Published 2026-04-01

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Yujun Yao, Luming Sun, Tao Wu, Ning Jiang, Shiyan Zhong, Xinwen Shu

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 the center of a galaxy as a massive, invisible vacuum cleaner—a Supermassive Black Hole (SMBH)—sitting in a cosmic neighborhood filled with stars. Usually, these stars dance safely around the vacuum cleaner. But sometimes, a star gets too close, gets stretched like taffy by the black hole's gravity, and gets ripped apart. This violent event is called a Tidal Disruption Event (TDE). It's like a cosmic fireworks display that flares up brightly in visible light and X-rays before fading away.

For a long time, astronomers thought these fireworks were one-time events. Once a star is eaten, it's gone. But recently, scientists started noticing something weird: in some galaxies, the fireworks seemed to go off, fade, and then go off again years or even decades later. These are called Repeating TDEs (rTDEs).

This paper is a detective story about how often these "repeat offenders" happen and what they actually are.

The Detective Work: Looking Back in Time

The authors, led by Yujun Yao and Luming Sun, decided to investigate a specific group of 16 recent TDEs found by the Zwicky Transient Facility (ZTF), a powerful telescope that scans the sky nightly. They wanted to know: Did any of these galaxies have a "previous life" as a TDE host?

To answer this, they acted like time travelers. They used data from the Catalina Real-time Sky Survey (CRTS), which has been taking photos of the sky since 2005. They looked at the host galaxies of their 16 ZTF TDEs to see if there were any bright flashes 5 to 19 years before the ZTF event.

Think of it like checking a security camera tape from 15 years ago to see if a specific house had a fire before the one you just saw.

The Findings: Two "Repeat Offenders"

Out of the 16 galaxies they checked, they found two that had a bright flash in the past:

AT 2019azh: Had a flash 13.2 years before the main event.
AT 2024pvu: Had a flash 17.1 years before the main event.

But wait, could these be something else?
The biggest worry for astronomers is that these flashes might just be Supernovae (exploding stars), which are common and can look similar to TDEs. It's like mistaking a firework for a lightning strike.

To solve this, the team played detective with the "temperature" of the light:

The Temperature Test: They used data from the GALEX space telescope (which sees ultraviolet light) for the AT 2024pvu event. They found the flash was incredibly hot—about 19,500 degrees Kelvin.
The Analogy: A supernova is like a campfire; it's bright but relatively "cool." A TDE is like a blowtorch; it's scorching hot. The temperature they measured was too hot for a supernova, strongly suggesting it was indeed a TDE.
The Math: They also calculated the odds of two supernovae accidentally happening in these specific spots at these specific times. The odds were roughly 0.3% (or 1 in 300). They concluded it's almost certain these were TDEs.

The Big Surprise: The Rate is Off the Charts

Here is where the plot thickens. If you assume TDEs happen randomly and rarely (which is the standard theory), finding two "repeat" events in a sample of 16 is like flipping a coin and getting "heads" 10 times in a row. It shouldn't happen by chance.

The authors calculated that the rate of TDEs in these specific galaxies was 100 to 1,000 times higher than the average for the rest of the universe.

This leads to two possible explanations for the "Repeat Offenders":

Theory A: The "Double Trouble" (Independent Events)

Maybe these galaxies are just incredibly unlucky (or lucky, depending on how you look at it). Perhaps they have a secret second black hole nearby (a binary black hole system) that is shaking up the neighborhood, throwing stars into the main black hole's mouth much more frequently than usual.

The Metaphor: Imagine a neighborhood where a second, hidden tornado is constantly pushing cars into a giant drain. The drain (black hole) is eating cars (stars) way faster than normal.

Theory B: The "Partial Bite" (Repeating Partial TDEs)

This is the authors' favorite theory. Instead of the star being eaten whole, imagine a star on a very long, stretched-out oval orbit. Every time it swings close to the black hole, the black hole takes a bite out of it but doesn't swallow it whole. The star survives, gets pushed back out, and comes back for another bite years later.

The Metaphor: Think of a dog (the star) running around a fence post (the black hole). Every time it runs close, the fence post nips a piece of its fur. The dog runs away, grows the fur back, and comes back for another nip. The dog isn't dead; it's just getting "nibbled" repeatedly.

Why Does This Matter?

If Theory B (the "Partial Bite") is correct, it changes everything we know about the universe's statistics.

Currently, astronomers count every flash of light as a "new" TDE. But if many of these are actually the same star being nibbled multiple times, we have been overcounting.

The Result: The actual rate of stars being destroyed might be 40% to 80% lower than we thought. We thought we were seeing 100 different stars die, but we might have only seen 20 stars getting nibbled 5 times each.

The Conclusion

The paper concludes that Repeating TDEs are likely much more common than we thought, possibly making up 25% to 60% of all the TDEs we see.

They prefer the "Partial Bite" model because it fits the data well, but they admit we need more observations to be sure. They predict that if we keep watching these galaxies, we will see the flashes happen again in the future (like a clockwork mechanism), which would prove it's the same star returning for more.

In short: The universe is full of "repeat customers" at the black hole buffet, and we might have been double-counting the menu items. This discovery forces us to rethink how often stars actually get eaten by black holes.

1. Problem Statement

Tidal Disruption Events (TDEs) are transient flares caused by stars being torn apart by supermassive black holes (SMBHs). Recent statistical studies of TDE rates rely on large samples from surveys like the Zwicky Transient Facility (ZTF). However, these statistics may be contaminated by Repeating TDEs (rTDEs), where a single star undergoes multiple partial disruptions or independent TDEs occur in the same galaxy.

The Gap: While short-interval rTDEs (periods < 5 years, e.g., ASASSN-14ko) are known, there has been no systematic statistical study of rTDEs with longer intervals (5–19 years).
The Challenge: Distinguishing long-interval rTDEs from independent TDEs or contaminating supernovae (SNe) is difficult due to the sparsity of historical data and the need to rule out SN contamination.
The Goal: To search for rTDEs with intervals > 5 years in a well-defined ZTF sample, verify their nature (ruling out SNe), and estimate the fraction and rate of rTDEs to correct the global TDE rate statistics.

2. Methodology

The authors employed a multi-step approach combining archival data mining, photometric modeling, and statistical rate estimation.

Sample Selection:
- Started with the ZTF Bright Transient Survey (BTS) TDE catalog.
- Applied a redshift cut ( $z < 0.05$ ) to ensure the shallower Catalina Real-time Sky Survey (CRTS) could detect potential past flares.
- Removed 2 objects (one in a merging galaxy, one off-nuclear), resulting in a final sample of 16 TDEs.
Search for Past Flares (CRTS):
- Utilized CRTS data (2005–2013) to search for flares 5–19 years prior to the ZTF detection.
- Applied a detection threshold of consecutive > 5 $\sigma$ flux excesses over the background.
- Identified two candidates: AT 2019azh and AT 2024pvu.
Nature Verification (TDE vs. SN vs. AGN):
- Light Curve Fitting: Modeled the CRTS flares using a Gaussian rise and power-law decline (GP model).
- Multi-band Analysis (AT 2024pvu):
  - Used GALEX UV observations taken near the CRTS peak.
  - Performed Spectral Energy Distribution (SED) fitting using CIGALE on archival SDSS, 2MASS, and WISE data to model the host galaxy's quiescent emission.
  - Compared GALEX fluxes against the host model and Swift UVOT follow-up observations (taken after the flare faded) to isolate the transient's UV contribution.
  - Derived a blackbody temperature ( $T_{BB}$ ) for the flare.
- SN Contamination Estimation:
  - Calculated the expected number of SNe (Type Ia and Core-Collapse) detectable by CRTS in the 16 galaxies.
  - Estimated the Effective Monitoring Duration (EMD) for SNe based on CRTS depth and cadence, assuming typical SN light curves.
  - Used Poisson statistics to calculate the probability of the observed flares being SNe.
- AGN Check: Verified the absence of AGN signatures (broad lines, X-ray emission, MIR colors) in the host galaxies.
Rate and Fraction Estimation:
- Calculated the EMD for TDEs (assuming various rise/decline timescales).
- Used a Bayesian method to infer the TDE Luminosity Function (LF) and rate normalization ( $\dot{N}_0$ ) for the sample.
- Compared the derived rate against the average ZTF TDE rate.
- Estimated the fraction of rTDEs by accounting for missed events (intervals < 20 years) and combining with known short-interval rTDEs in the sample.

3. Key Results

Discovery of Candidates: Two rTDE candidates were found with intervals of 13.2 years (AT 2019azh) and 17.1 years (AT 2024pvu).
Nature Confirmation:
- AT 2024pvu.I: The GALEX data revealed a blackbody temperature of $\sim$ 19,500 K, consistent with TDEs but significantly higher than typical SNe ( $<10^4$ K). The peak luminosity ( $M_V \approx -19.7$ ) also exceeds typical SNe.
- SN Probability: The expected number of SNe detectable in the sample was calculated to be $\lesssim 0.08$ . The probability that both flares are SNe is only 0.3%.
- Conclusion: It is highly probable that both candidates are genuine rTDEs, not SNe or AGN flares.
TDE Rate Anomaly:
- The detection of 2 rTDEs in 16 galaxies over a 5–19 year window implies a TDE rate 2–3 orders of magnitude higher than the global average ( $3.2 \times 10^{-5}$ galaxy $^{-1}$ yr $^{-1}$ ) for these specific host galaxies.
rTDE Fraction:
- Including two additional short-interval rTDEs (AT 2020vdq, AT 2022dbl) found in the same sample, at least 4/16 (25%) are rTDEs.
- Accounting for missed events (due to CRTS depth/cadence), the authors estimate that 25%–60% of optical TDEs with intervals < 20 years are rTDEs.
- The inferred rate of rTDEs is $2–6 \times 10^{-6}$ galaxy $^{-1}$ yr $^{-1}$ .

4. Interpretation and Origin

The authors discuss two primary scenarios for the origin of these rTDEs:

Independent TDEs: The host galaxies (post-starburst/green valley) harbor close SMBH binaries that boost the TDE rate via the Lidov-Kozai mechanism. However, this requires an extremely high rate ( $\sim 10^{-2}$ galaxy $^{-1}$ yr $^{-1}$ ) and is disfavored for AT 2024pvu due to its high SMBH mass ( $\sim 10^8 M_\odot$ ), where stars would likely be swallowed whole before disruption.
Repeating Partial TDEs (rpTDEs) - Preferred Model:
- A single star in a highly eccentric orbit ( $1-e \sim 10^{-3}$ ) undergoes partial disruption at each periapsis passage.
- The orbital periods (13–17 years) imply semi-major axes of $\sim$ 780–2850 AU.
- This scenario suggests the observed optical TDE rate is overestimated because multiple flares from one star are counted as separate events.
- Correction Factor: If each rpTDE produces $\sim$ 3–10 flares, the true optical TDE rate should be reduced by 40%–80%.
- Formation Channel: Likely formed via the Hills mechanism (tidal dissociation of a close binary) or scattering in a dense stellar disk left by a faded AGN.

5. Significance

Statistical Correction: This study provides the first quantitative estimate of long-interval rTDEs, suggesting that current TDE rates may be significantly overestimated (by up to a factor of 2–3) if rTDEs are not accounted for.
Methodological Advance: Demonstrates the utility of combining modern high-cadence surveys (ZTF) with historical data (CRTS) and UV follow-up (GALEX/Swift) to characterize transient populations over decades.
Astrophysical Insight: Supports the rpTDE model as a dominant channel for rTDEs, offering a new probe into stellar dynamics near SMBHs and the population of close binaries in galactic nuclei.
Future Outlook: Predicts that future observations of these specific sources (e.g., IC 3599, AT 2019azh) will reveal periodicity, confirming the rpTDE nature and refining the TDE rate statistics.