Repeating flares, X-ray outbursts and delayed infrared emission: A comprehensive compilation of optical tidal disruption events

TL;DR

We present TDECat, a comprehensive catalogue of 134 confirmed TDEs with multi-wavelength photometry and publicly available spectra up to 2024, enabling population studies across wavelengths and spectral classes. Using Bayesian Blocks, we quantify optical flare timescales and find that durations, rise times, and decay times are well described by log-normal distributions, with a positive rise-decay correlation captured by log10(t_rise) ≈ 0.915 log10(t_decay) − 0.31. Spectral analysis shows a predominance of the TDE-H+He class, with three newly identified repeating-flare TDEs (AT 2024pvu, AT 2022exr, AT 2021uvz) and a notable IR–X-ray correlation across the sample, including all coronal-line TDEs exhibiting both. The catalogue also reveals IR delays and X-ray delays in several events, highlighting complex multi-wavelength emission mechanisms. Overall, TDECat provides a valuable framework for statistical population studies and will be instrumental for LSST-era TDE research.

Abstract

TDEs have been proposed as valuable laboratories for studying dormant black holes. However, progress in this field has been hampered by the limited number of observed events. In this work, we present TDECat, a comprehensive catalogue of 134 confirmed TDEs (131 optical TDEs and three jetted TDEs) discovered up to the end of 2024, accompanied by multi-wavelength photometry (X-ray, UV, optical, and IR) and publicly available spectra. We also study the statistical properties, spectral classifications, and multi-band variability of these events. Using a Bayesian Blocks algorithm, we determined the duration, rise time, decay time, and their ratio for 103 flares in our sample. We find that these timescales follow a log-normal distribution. Furthermore, our spectral analysis shows that most optical TDEs belong to the TDE-H+He class, followed by the TDE-H, TDE-He, and TDE-featureless classes, which is consistent with expectations from main-sequence star disruption. Using archival observations, we identified three new potentially repeating TDEs, namely, AT2024pvu, AT2022exr, and AT2021uvz, increasing the number of known repeating events. In both new and previously known cases, the secondary flares exhibit a similar shape to the primary. We also examined the infrared and X-ray emission from the TDEs in our catalogue, and find that 14 out of the 18 infrared events have associated X-ray emission, strongly suggesting a potential correlation. Finally, we find that for three sub-samples (repeating flares, infrared-emitting events, and X-ray events), the spectral classes are unlikely to be randomly distributed, suggesting a connection between spectral characteristics and multi-wavelength emission. TDEcat enables large-scale population studies across wavelengths and spectral classes, providing essential tools for navigating the data-rich era of upcoming surveys such as the Legacy Survey of Space and Time.

Figures (11)

• Figure 1: Example of the implementation of the Bayesian Blocks algorithm to the ZTF light curve of AT 2022fpx. The red solid line shows the optimal blocks and the dashed, dotted and solid black lines represent the rise, decay, and peak times, respectively.
• Figure 2: Distribution of the $t_{rise}/t_{decay}$ ratio for 103 TDE flares in our catalogue (black step histogram). The dashed red line indicates a Gaussian distribution with $\mu=-0.48$ and $\sigma=0.30$ overplotted on the data. The grey data points and boxes represent the median of the mock distributions for each bin and corresponding uncertainty respectively.
• Figure 3: Distributions of the durations, rise times, and decay times of the flares (top panels and bottom-left panel, respectively), as well as the scatter plot of $t_{rise}$ versus $t_{decay}$ (bottom-right panel). The red dashed lines indicate Gaussian distributions overplotted on the data. The grey data points and boxes represent the median of the generated distributions for each bin and corresponding uncertainty, respectively. In the bottom-right panel the solid red line indicates the best-fit ODR line, while the grey contour represents the 95% confidence interval.
• Figure 4: Distribution of redshift for the TDEs in our sample (black step histogram) as well as the optical TDEs (grey step-filled histogram). The dotted black line indicates a normal distribution overplotted on the full sample, while the dashed red line represents a normal distribution overplotted on the optical TDEs.
• Figure 5: Percentage of each TDE spectral class in the catalogue.