Identifying tidal disruption events among radio transient galaxies

TL;DR

This paper examines optical and infrared properties for 24 radio transients discovered by VLASS to determine whether their radio variability arises from intrinsic AGN processes or from tidal disruption events (TDEs). Employing optical spectroscopy, emission-line diagnostics, and infrared monitoring (WISE/NEOWISE) alongside radio evolution, the study uses a mid-IR/radio diagram (W3 versus 1.4 GHz) to distinguish AGN-driven activity from potential TDEs. The analysis finds that most hosts are massive ellipticals with old stellar populations and low accretion rates, while a subset, especially in star-forming hosts, shows infrared variability and radio evolution consistent with TDEs; about eight sources are flagged as TDE candidates based on their W3/radio tracks. The work suggests that the W3/radio diagnostic is a promising tool for separating TDE-triggered radio transients from intrinsic AGN variability, though caveats related to selection effects, AGN contamination, and data sensitivity require further, deeper multiwavelength follow-up.

Abstract

We present the optical and infrared properties of a sample of 24 radio transient sources discovered in the Very Large Array Sky Survey (VLASS). Previous studies of their radio emission showed that these sources resemble young gigahertz-peaked spectrum (GPS) radio sources, but are less powerful and characterized by low-power jets. The bursts of radio activity in most cases are likely due to intrinsic changes in the accretion processes. However, for a few sources in this sample, we cannot rule out the possibility that their radio variability results from a tidal disruption event (TDE). In this work, we extend our analysis to the optical and infrared regimes, confirming that our sample of radio transients is not homogeneous in terms of their optical and infrared properties either. The host galaxies of most of these sources are massive ellipticals with emission dominated by active galactic nuclei (AGN). They host supermassive black holes (SMBHs) with masses typical of radio-loud AGNs ($\rm >10^7\,M_{\odot}$), but exhibit very low accretion activity. In contrast, the sources for which a TDE origin is suspected are either pure star-forming galaxies or show significant ongoing star formation, similar to radio-selected optically-detected TDEs. Additionally, two of them exhibit infrared flares characteristic of TDEs, while the remaining sources do not display significant variability outside the radio regime. Moreover, the evolution of their radio brightness in the W3/radio diagnostic diagram, which we employ in our analysis, also sets our TDE candidates apart from the rest of the sample and resembles the radio variability seen in optically discovered TDEs with radio emission. Finally, based on our findings, we hypothesize that the mid-IR/radio relation can serve as a tool to distinguish between radio transients caused by TDEs and those originating from intrinsic AGN variability.

Figures (10)

• Figure 1: Cutout stack g/r/i images $(1'\times1')$ from Pan-STARRS. The target source is located at the center of each image. The scale bar is size 10$"$.
• Figure 2: Emission line diagnostic diagrams: $\rm [O\,III]\lambda5007/H_{\beta}$ vs three different line ratios BaldwinKewley01Kewley2006Kauffmann and $W_{H_{\alpha}}$ vs $\rm [N\,II]\lambda6584/H_{\alpha}$CidFernandes2011. The lines demarcate boundaries between sources classified as star-forming (SF) galaxies, AGNs, low ionization nuclear emission line regions (LINERs), “Composite” sources, weak AGNs (wAGN), strong AGNs (sAGN), retired galaxies (RG) and passive galaxies. Classifications based on SDSS observations are indicated by blue squares, while red circles indicate the classifications based on Palomar and Keck observations. Numbers refer to source identifications in Table \ref{['table:basic']}.
• Figure 3: Light curves in the optical, infrared and radio (3 GHz) for selected objects. In case of optical and infrared curves the individual points represent monthly averaged values after $\rm 3\sigma$ clipping to reject outliers. For radio observations (NVSS, FIRST, VLA), the points correspond to single measurements. The VLA data come from the VLASS survey as well as dedicated observations, as reported in MKB2025.
• Figure 4: WISE color–color plot for sources presented in this work (red dots). Numbers refer to source identifications in Table \ref{['table:basic']}. Additional points (black dots, blue square and green triangle) represent optically discovered TDEs as described in the text. The colored density clouds represent the following objects: radio galaxies (green dots), star-forming galaxies (blue dots), and quasars (yellow dots). These data come from the ROGUE I Koziel2021 and ROGUE II (Kozieł-Wierzbowska, in prep.) catalogues.
• Figure 5: Mid-infrared (W3) vs. radio (1.4 GHz) flux density diagram for the sources presented in this work (red dots). Source numbers correspond to the identifications listed in Table \ref{['table:basic']}. Additional points—black dots, the blue square, and the green triangle—represent optically discovered TDEs. For clarity, error bars and upper limits are not shown in this figure. The solid and dashed lines indicate the evolution of radio flux density for individual sources, as described in detail in the text. To improve the readability of the plot, the main populations of objects are represented only by the contours of their density distributions: radio galaxies (green), star-forming galaxies (blue), and quasars (yellow). These reference data are taken from the ROGUE I Koziel2021 and ROGUE II (Kozieł-Wierzbowska, in prep.) catalogs. The main dashed line marks the relation $\rm S_{W3} = S_{1.4\,GHz}$, which serves to separate radio galaxies from star-forming galaxies based on the ROGUE I sample.