Changing-look Active Galactic Nuclei from SDSS, LAMOST and DESI Survey
Guohai Chen, Wenxin Yang, Xuhong Ye, Zhiqiang Chen, Zhiyuan Pei, Hubing Xiao, Junhui Fan · 2026
Read Changing-look Active Galactic Nuclei from SDSS, LAMOST and DESI Survey by Guohai Chen, Wenxin Yang, Xuhong Ye, Zhiqiang Chen, Zhiyuan Pei, Hubing Xiao, Junhui Fan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-24429.txt.
Introduction
Chen and six co-authors report 45 changing-look nuclei, 40 new. Turn-offs dominate, 43 against 2, perhaps because type 2 nuclei lack broad lines or hide them behind dust. Using DESI as a third epoch, the team identifies 12 repeaters. Of 14 new nuclei with three-epoch coverage, 7 are repeaters, and 5 of 16 previously reported nuclei also repeat. Repeaters show high-low-high accretion evolution, tying repeated changes to accretion swings. Transition times run about 10 years for the first change and 4 for the second.
Catalog record
| CREATOR | Guohai Chen, Wenxin Yang, Xuhong Ye, Zhiqiang Chen, Zhiyuan Pei, Hubing Xiao, Junhui Fan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 49757 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.24429
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Changing-look Active Galactic Nuclei from SDSS, LAMOST and DESI Survey Guohai Chen, Wenxin Yang, Xuhong Ye, Zhiqiang Chen, Zhiyuan Pei, Hubing Xiao, Junhui Fan Abstract Although more than 1000 optical changing-look active galactic nuclei (CLAGNs) have been reported to date, their physical origin remains unclear, and repeating CLAGNs (RCLAGNs) are still rare. Expanding the CLAGN sample, especially RCLAGNs, is therefore important for constraining the underlying mechanism. We systematically search for CLAGNs by cross-matching spectroscopic observations from the Sloan Digital Sky Survey (SDSS) and the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST), and further use spectra from the Dark Energy Spectroscopic Instrument (DESI) to investigate repeating CL behavior. We identify 45 CLAGNs, including 40 newly reported sources. The sample is dominated by turn-off events, with 43 turn-off and 2 turn-on sources, possibly because Type 2 AGNs either lack a detectable broad-line region or have their broad emission lines obscured by circumnuclear dust. Using DESI as a third spectroscopic epoch, we identify 12 RCLAGNs. This high detection rate of repeated CL behavior suggests that CL transitions may arise from recurrent physical processes, such as accretion-rate fluctuations or disk instabilities. In the log M BH \log M_{\rm BH} – log ( L bol / L Edd ) \log(L_{\rm bol}/L_{\rm Edd}) plane, RCLAGNs further show a clear high–low–high accretion-state evolution, supporting a close link between CL behavior and recurrent changes in accretion power. Finally, the rest-frame upper limits on the transition timescales are ∼ \sim 10 yr for the first transition and ∼ \sim 4 yr for the second, reflecting different survey time baselines rather than intrinsic differences in physical transition timescales. I Introduction Active galactic nuclei (AGNs) are powered by supermassive black holes (SMBHs) at their centres ( 45 ) . During accretion of surrounding gas and dust, an accretion disk forms, in which a portion of the gravitational potential energy of the infalling material is converted into electromagnetic radiation. At larger distances, the accretion disk is surrounded by the broad-line region (BLR), followed by the narrow-line region (NLR) on still larger scales, while a torus-like dusty structure may enshroud the system on intermediate scales ( 79 ) . AGNs are observationally divided into two main types based on optical spectroscopic features. Type 1 AGNs display both broad emission lines (BELs; line widths of 1000–20000 km s -1 ) and narrow emission lines (NELs; 300–1000 km s -1 ), while Type 2 AGNs show only NELs. According to the unified model, this difference arises from orientation: in Type 2 AGNs, the BELs are thought to be obscured from our line of sight by a dusty torus ( 6 ; 67 ) . However, it is now established that a population of “true” Type 2 AGNs exists, which appear to lack a BLR rather than being obscured Type 1 AGNs ( 78 ; 81 ; 28 ; 73 ; 17 ) . In recent years, spectroscopic observations have revealed a class of optical changing-look AGNs (CLAGNs), in which the BELs in the optical spectra can appear (turn-on) or disappear (turn-off) on timescales ranging from months to years ( 66 ) . Since their discovery, this phenomenon has attracted significant attention in astronomy. A robust identification of CLAGNs requires both the emergence or disappearance of BELs and significant photometric variability. To date, more than 1000 optical CLAGNs have been identified, primarily through two main approaches ( 64 ; 40 ; 89 ; 36 ; 37 ; 25 ; 15 ) . One approach relies on repeat spectroscopic observations from multi-epoch spectral surveys, such as the Sloan Digital Sky Survey ( 62 , SDSS;) , the Large Sky Area Multi-Object Fiber Spectroscopic Telescope ( 18 , LAMOST;) , the Dark Energy Spectroscopic Instrument ( 21 , DESI;) , and the 6dF Galaxy Survey ( 41 ) . Sources exhibiting changes in BELs are identified, and photometric variability is further examined to rule out BEL variations caused by instrumental effects or other observational uncertainties. Another approach begins with the selection of candidates based on large-amplitude photometric variability, which are subsequently confirmed as CLAGNs through follow-up spectroscopic observations that capture such BEL transitions ( 91 ; 90 ; 53 ; 74 ; 97 ; 98 , e.g.) . In both approaches, visual inspection is often applied as a final verification step. Studies of CLAGNs point to three main mechanisms for their CL behavior: intrinsic variations in accretion rate, dust obscuration effects, and tidal disruption events (TDEs). Each mechanism has observational limitations and cannot explain all cases. Among them, accretion-rate variations are most widely supported, with accretion disk state transitions driving the observed spectral changes ( 59 ; 69 ; 76 ; 29 ; 66 ; 88 ) . In contrast, the obscuration scenario attributes BEL variations to dusty material moving into and out of the line of sight ( 38 ) . For most optical CLAGNs, the transition timescales are much shorter than the expected dynamical timescales of obscuring dust ( 46 ; 70 ; 75 ) , and this scenario cannot explain the significant variability observed in their light curves ( 54 ; 68 ; 70 ) . Nevertheless, 94 reported a rapid CLAGN transition on a rest-frame timescale of \lesssim 2 months, whose spectral evolution favored obscuration, although the short timescale and mid-infrared variability made this interpretation physically challenging. The TDE scenario ( 57 ; 60 ) can explain only a limited number of cases ( 57 ) and has difficulty accounting for repeating CLAGNs (RCLAGNs). It remains unclear whether all CLAGNs undergo repeated CL transitions. Despite the growing number of identified CLAGNs, recurrent CL events are still rare, because their confirmation requires multi-epoch spectroscopic evidence for multiple state changes, which is often obtained through long-term monitoring over years to decades. In this context, the nearby Seyfert galaxy Mrk 1018 provides one of the earliest and most striking examples of repeating CL behavior. Its H \beta emission increased by more than a factor of six within a few years, before fading again and disappearing several decades later ( 16 ; 56 ) . Long-term spectroscopic monitoring further revealed that Mrk 1018 experienced a complete sequence of spectral-type transitions, evolving through Seyfert types 1.0, 1.2, 1.5, 1.8, and 2.0 over a span of approximately 45 years ( 49 ) . A similar evolutionary pattern has been observed in Mrk 590. The H \beta line first emerged in 1989, vanished by 2003 ( 19 ) , and subsequently reappeared in 2017 together with strong continuum variability ( 65 ; 55 ) . More recently, 61 reported that Mrk 590 is once again transitioning toward a type 1.2 state, closely resembling its spectral classification in the 1990s. To date, the number of confirmed RCLAGNs remains small, with fewer than about 50 sources reported in the literature ( 83 ; 84 ; 52 ; 24 ; 25 ) . Therefore, identifying more RCLAGNs through long-term photometric monitoring and multi-epoch spectroscopic observations is essential for constraining the physical origin and characteristic timescales of the CL phenomenon, and constitutes one of the main motivations of this work. Low-redshift ( z<0.9 ) CLAGNs have been systematically identified using SDSS in combination with DESI ( 36 ; 37 ) , as