Pith. sign in

REVIEW 4 major objections 6 minor 3 cited by

Galaxy evolution in the post-merger regime. III -- The triggering of active galactic nuclei peaks immediately after coalescence

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Galaxy mergers trigger AGN most strongly in the first 160 million years after coalescence, with elevated rates persisting past 1.76 Gyr for mid-infrared and broad-line selected AGN.

desk verdict First time-resolved post-merger AGN census, plausible but the classifier systematics need to be opened up before I'd bet on the peak bin. read the letter →

arxiv 2412.02804 v2 pith:TF6Z2VD4 submitted 2024-12-03 astro-ph.GA

Sara L. Ellison , Leonardo Ferreira , Robert Bickley , Tess Grindlay , Samir Salim , Shoshannah Byrne-Mamahit , Shobita Satyapal , David R. Patton
show 1 more author
Jillian M. Scudder
This is my paper · ORCID
classification astro-ph.GA
keywords galaxymergersactivegalacticnucleipost-mergergalaxiestimeAGNtriggeringmachinevisionclassificationnuclearobscurationstarburst-AGNconnection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks when, along the sequence of a galaxy merger, a supermassive black hole is most likely to be actively accreting. The authors combine a sample of pre-coalescence galaxy pairs with post-merger remnants whose time since coalescence is estimated by a machine-vision pipeline, and compare AGN fractions in each stage with matched non-merger controls. They find that the excess of AGN peaks immediately after coalescence, in the bin 0–0.16 Gyr, regardless of whether AGN are selected by narrow emission lines, mid-infrared colours, or broad emission lines. The excess remains statistically significant out to the longest bin, 0.96–1.76 Gyr, for mid-infrared and broad-line selected AGN. This matters because it places the first direct observational timing constraint on when mergers feed black holes and connects that timing to nuclear dust clearing and starbursts.

What carries the argument

The load-bearing tool is MUMMI, a machine-vision ensemble of neural networks trained on simulated galaxies that both identifies post-merger remnants and predicts the time since coalescence in four bins: 0–0.16, 0.16–0.48, 0.48–0.96, and 0.96–1.76 Gyr. This converts the post-merger population into a timeline, which is what allows the paper to ask when AGN triggering peaks and how long it lasts. The three AGN diagnostics probe distinct physical regions—narrow lines from ionized gas on roughly kiloparsec scales, mid-infrared colours from the dusty torus, and broad lines from the innermost accretion region—and their different responses to merger stage carry the obscuration argument. The AGN excess, defined as the ratio of AGN fractions in mergers versus matched controls, is the common metric on which all trends are drawn.

What would settle it

Re-date the same post-merger galaxies with an independent method, such as fitting tidal features or stellar populations, and recompute the AGN excess in the same four time bins; if the peak moves out of the 0 to 0.16 Gyr bin, or the long-lived mid-infrared and broad-line excess disappears, the original bin assignments were creating the temporal signal.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the statistical enhancement of AGN in merging galaxies is largest right after the two galaxies coalesce, at $0 < T_{\rm PM} < 0.16$ Gyr, and then declines as the remnant ages. The result holds for all three AGN diagnostics, and the pattern differs by diagnostic in a way the authors interpret as evolution of nuclear obscuration: narrow-line and mid-IR selected AGN show an excess already in the pair phase, while broad-line AGN show a mild deficit before coalescence that becomes an excess afterward, implying the broad-line region is hidden by inflated dust covering fractions during the encounter and partially cleared after it. The excess is largest for the most luminous and bolometrically dominant AGN, mirroring simulations; merger-hosted narrow-line AGN are about 2.5 times more luminous than their secular counterparts. The paper also shows that the statistical peak of starbursts is contemporaneous with the AGN peak, with any lag between the two processes constrained to less than roughly 150 Myr.

Load-bearing premise

The timeline assumes the classifier puts each post-merger galaxy into the right time-since-coalescence bin, but it does so correctly only 70 to 80 percent of the time, so if the mistakes are tied to AGN properties the apparent peak right after coalescence could be an artifact.

Editorial extensions

If this is right

  • The peak in AGN excess at $0 < T_{\rm PM} < 0.16$ Gyr appears with every AGN diagnostic, so the finding does not depend on a single selection method.
  • Mid-infrared and broad-line AGN remain more common than in controls out to the longest time bin, meaning merger-enhanced nuclear activity outlasts the star-formation enhancement, which fades by about 1 Gyr.
  • The excess is stronger for more luminous and bolometrically dominant AGN, and merger-hosted narrow-line AGN are about 2.5 times more luminous than secular ones, implying mergers preferentially produce powerful accretion episodes.
  • The pre-merger deficit and post-merger excess of broad-line AGN implies the nuclear dust covering fraction rises during the encounter and falls after coalescence, consistent with AGN feedback clearing some of the obscuring material.
  • Starburst and AGN excesses peak in the same 0.16 Gyr interval, so any statistical delay between the two triggers is less than about 150 Myr.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: because the first time bin is only 0.16 Gyr wide, the true peak of accretion could be even sharper and closer to coalescence than the bin-averaged excess shows; simulations or samples with finer time resolution in the first few hundred megayears would reveal it.
  • Editorial inference: the long-lived mid-infrared excess may be partly a byproduct of merger-driven quenching rather than continued black hole growth; splitting post-mergers by specific star formation rate would test whether less star-forming remnants are more likely to show a mid-infrared AGN signature.
  • Editorial inference: the covering-fraction picture predicts that X-ray column densities toward AGN in close pairs should be systematically higher than toward AGN in post-mergers; a dedicated X-ray census of these systems would directly test the blowout interpretation.
  • Editorial inference: if this timing is universal, merger-AGN surveys at higher redshift that find no excess may be washing out the signal by averaging over a wide range of merger stages; applying time-bin classification to those samples would sharpen the comparison.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper combines a sample of close galaxy pairs and post-merger galaxies identified by the MUMMI machine-vision pipeline in UNIONS, with SDSS DR7 spectroscopy and WISE photometry, to measure how the excess frequency of AGN (relative to a mass- and redshift-matched control sample) evolves across the merger sequence, from wide-separation pairs through coalescence to T_PM = 1.76 Gyr. AGN are selected with three independent diagnostics: narrow emission lines (NLAGN), broad emission lines (BLAGN), and mid-IR W1-W2 colors. The authors report that the AGN excess peaks in the first post-coalescence bin (0 < T_PM < 0.16 Gyr), that the excess is long-lived for mid-IR and BLAGN diagnostics, that more luminous AGN show larger excesses, that the fractional contribution of unobscured (broad-line) AGN rises after coalescence, and that the starburst excess peaks contemporaneously with the AGN excess.

Significance. If the T_PM assignments are reliable, this is the first observational measurement of the post-coalescence time evolution of AGN triggering, a result that directly tests simulation predictions and bears on feedback timescales. The study has several genuine strengths: the AGN diagnostics are independent of the machine-learning labels (so there is no circularity in the measured excess), the control matching is carefully constructed and demonstrated, the sample is large (8,141 post-mergers with quality cuts), and the paper makes clear, falsifiable predictions, such as the peak excess at 0-0.16 Gyr and the persistence of mid-IR/BLAGN excesses to 1.76 Gyr. The agreement with Illustris-TNG and disagreement with EAGLE, if correct, would be an important constraint on subgrid models of black hole fueling. However, the central claim depends entirely on the accuracy of MUMMI's time-post-merger bins, which is currently only summarized as a global 70-80% figure from an unpublished companion paper, so the observational result is not yet as secure as the abstract implies.

major comments (4)
  1. [2.3] The central claim that the AGN excess peaks at 0 < T_PM < 0.16 Gyr rests entirely on MUMMI's T_PM bin assignments, but the only accuracy information provided is a global '70-80 per cent' statement from an unpublished companion paper. No per-bin confusion matrix, purity/completeness values, or dependence of classification accuracy on AGN properties is given. The training images are TNG mock r-band images that do not include AGN point sources, so a bright unresolved nucleus in a real galaxy could plausibly make the remnant appear more concentrated and shift it into the shortest T_PM bin, creating the observed peak even if the true temporal trend is flat. This concern is compounded by the highly unbalanced bin populations (785, 740, 778, 5838 in the four T_PM bins), which suggest a strong classifier prior. The authors should provide the confusion matrix and per-bin accuracy from Ferreira et al. (in prep), and test for AGN-correlated biases by injecting point sources into the TNG mock images or by comparing T_PM assignments with an independent morphological or stellar-population age indicator.
  2. [3.1, Tables 1-4] The quoted uncertainties on the AGN excess are binomial errors on the fractions, and therefore propagate neither the 70-80% T_PM classification uncertainty nor the scatter in the control matching (which is iterative and reuses control galaxies across samples). The headline comparison (excess 3.1 +/- 0.2 vs 2.0 +/- 0.2 in the first two bins) is thus presented as more significant than the data actually support. The authors should propagate the T_PM confusion matrix (e.g., by Monte Carlo resampling of bin assignments) and the control-matching scatter (e.g., by bootstrap resampling of the control pools) into all excess values and significance claims.
  3. [2.4 and 4.1] The interpretation of the post-merger decline and the comparison with EAGLE presumes that the four T_PM bins have comparable purity and completeness. The fourth bin (0.96 < T_PM < 1.76 Gyr) contains 5,838 galaxies versus ~750 in each of the first three bins, and the bin lengths are deliberately unequal. If the longer bin has lower completeness or a different contamination rate (both plausible given the stated 70-80% average accuracy and the increasing difficulty of time prediction at late stages), the apparent decline in excess beyond ~1 Gyr - which is used to claim that the NLAGN excess ends by ~1 Gyr and that EAGLE's delayed ~300 Myr peak is ruled out - could be an artifact. Per-bin accuracy and completeness figures, plus a test of how the derived excesses change if the last bin is subdivided or if classification errors are simulated, are necessary to support the temporal-decay claims.
  4. [3.3 and Figure 6] The conclusion that there is a deficit of broad-line AGN in the pre-merger phase is based on excess values that the text itself describes as 'marginal,' and several bins appear consistent with unity within their error bars. Since this deficit is used to motivate the dust-covering-fraction narrative in the abstract and Section 5, the paper should either quote the relevant significances (or show the deficit is significant after including T_PM and matching uncertainties) or explicitly state that the pre-coalescence obscuration signal is only suggestive. As written, the strength of the claim in the abstract exceeds the statistical support in the figure.
minor comments (6)
  1. [Table 3] In the first row of Table 3, the post-merger AGN fraction is reported as '0.13 +/- 0.1'; the error should presumably be approximately 0.01, based on the binomial error for 104/785.
  2. [Table 4] In the first row of Table 4, the control BLAGN fraction is listed as '0.017 +/- 0.007'; for 435/25,905 the binomial error is approximately 0.0008, so the quoted error appears to be off by an order of magnitude.
  3. [2.5] The phrase 'Figure of Donley et al. 2012' should be 'Figure 1 of Donley et al. 2012' for clarity.
  4. [1] In the introduction, 'Omori et el. 2023' should be 'Omori et al. 2023'.
  5. [2.3] There is a typo in 'occured' in the sentence describing the first snapshot after coalescence; it should be 'occurred'.
  6. [5] The summary states that 'radiation pressure from the AGN redistributes this dust' as if it were a direct measurement; since only the BLAGN fraction is measured, consider softening this to 'is consistent with redistribution of nuclear dust by AGN feedback.'

Circularity Check

1 steps flagged · score 2.0 of 10

No significant circularity in the AGN-excess measurement; the only load-bearing self-citation is the unpublished T_PM bin accuracy underpinning the time axis.

  1. self citation load bearing [Section 2.3, post-merger sample paragraph (T_PM bin accuracy)]
    "After imposing a threshold for prediction accuracy, post-merger galaxies are correctly assigned into these T_PM bins 70–80 per cent of the time (Ferreira et al., in prep). This threshold is based on the quality of the probability distributions produced by mummi, and removes any uncertain or spurious classifications mitigating any performance degradation between simulation and observational domains (Ferreira et al. in prep)."

    The headline result, that the AGN excess peaks at 0 < T_PM < 0.16 Gyr, is located on a time axis that is entirely provided by MUMMI. The only evidence cited for the correctness of these T_PM bin assignments in observed UNIONS galaxies is an unpublished companion paper (Ferreira et al., in prep) by overlapping authors. No per-bin purity or completeness is given, and no test is presented for whether AGN light itself biases the morphological classifications. The temporal peak therefore inherits its meaning from a self-cited, unavailable validation.

full rationale

The core measurement is the fraction of AGN in post-merger galaxies versus matched controls, binned by T_PM. AGN are identified from three independent external diagnostics: MPA/JHU narrow-line ratios, Liu et al. (2019) broad-line fits, and unWISE W1-W2 colours. None of these enter the MUMMI training or the T_PM prediction; MUMMI is trained on TNG mock r-band images and predicts morphology/time only. The claimed excess at 0<T_PM<0.16 Gyr is therefore an empirical ratio of independently counted AGN fractions, not an output of the classifier or a fitted parameter. The luminosity, obscuration, and starburst analyses are likewise direct counts or catalogue comparisons. The paper does rely heavily on self-citations, and the T_PM axis is validated only by an in-prep companion paper from the same team, which is a legitimate concern for accuracy and reproducibility. However, that is a limitation of the time-axis calibration, not circularity: the result would not be forced by construction if the bin assignments were wrong. The AGN excess could have peaked in a different bin. I therefore assign a low score reflecting the self-referential validation of the temporal axis while recognizing that the central AGN claim is independent of, and not equivalent to, the paper's inputs.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities or fitted constants. The measurement depends on the ML tool MUMMI (self-cited, with the time-bin version unpublished), the adopted AGN selection thresholds, and the matching assumptions listed above.

free parameters (3)
  • T_PM bin edges
    The four post-merger time bins (0-0.16, 0.16-0.48, 0.48-0.96, 0.96-1.76 Gyr) are hand-chosen to maximize MUMMI classification performance. The claim of a peak in the first bin depends on this discretization, though the trend is monotonic across bins.
  • MUMMI T_PM quality threshold
    A threshold on the ML probability distributions selects 'confident' time predictions and reduces the post-merger sample from 11,309 to 8,141 galaxies; the threshold value is only cited to Ferreira et al. in prep, and it determines the sample used for the central claim.
  • W1-W2 AGN selection cut = 0.5 mag
    Mid-IR AGN are defined by W1-W2 > 0.5 (Vega), a literature choice; the paper shows the excess increases for a redder cut of 0.7, so the temporal peak is not an artifact of this threshold.
assumptions (5)
  • domain assumption MUMMI T_PM predictions are accurate: post-mergers are correctly assigned to time bins 70-80% of the time after the quality threshold.
    Section 2.3 invokes this accuracy from Ferreira et al. (in prep). The observed excess pattern as a function of T_PM is only meaningful if misclassifications do not correlate with AGN incidence.
  • domain assumption TNG mock images are a valid training ground for applying MUMMI to UNIONS observations.
    Section 2.3 trains on IllustrisTNG mock r-band images and applies directly to UNIONS r-band images; a quality threshold mitigates domain shift but does not eliminate it.
  • domain assumption Projected pair separation is a monotonic proxy for time to coalescence.
    Used to place the pair stage on the same axis as the post-merger stage in Figures 4-6, following Patton et al. (2024).
  • domain assumption AGN diagnostics identify the same physical population in mergers and controls.
    Kewley et al. (2001) line-ratio cut, W1-W2 > 0.5, and the Liu et al. (2019) broad-line catalog are applied uniformly; any AGN selection bias that varies with merger stage would shift the excess.
  • domain assumption Matching on stellar mass and redshift removes confounders for the AGN fraction.
    Section 2.4 matches until KS p<0.99 on these two axes. Unmeasured variables such as environment, gas fraction, or nuclear morphology are assumed not to bias the excess.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Galaxy evolution in the post-merger regime. III -- The triggering of active galactic nuclei peaks immediately after coalescence." pith.science (2026). https://pith.science/paper/TF6Z2VD4

@misc{pith2026241202804,
  author       = {Pith},
  title        = {Pith review of: Galaxy evolution in the post-merger regime. III -- The triggering of active galactic nuclei peaks immediately after coalescence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TF6Z2VD4}},
  note         = {Machine review of arXiv:2412.02804}
}
read the original abstract

Galaxy mergers have been shown to trigger AGN in the nearby universe, but the timescale over which this process happens remains unconstrained. The Multi-Model Merger Identifier (MUMMI) machine vision pipeline has been demonstrated to provide reliable predictions of time post-merger (T_PM) for galaxies selected from the Ultraviolet Near Infrared and Optical Northern Survey (UNIONS) up to T_PM=1.76 Gyr after coalescence. By combining the post-mergers identified in UNIONS with pre-coalescence galaxy pairs, we can study the triggering of AGN throughout the merger sequence. AGN are identified using a range of complementary metrics: mid-IR colours, narrow emission lines and broad emission lines, which can be combined to provide insight into the demographics of dust and luminosity of the AGN population. Our main results are: 1) Regardless of the metric used, we find that the peak AGN excess (compared with a matched control sample) occurs immediately after coalescence, at 0 < T_PM < 0.16 Gyr. 2) The excess of AGN is observed until long after coalescence; both the mid-IR selected AGN and broad line AGN are more common than in the control sample even in the longest time bin of our sample (0.96 < T_PM < 1.76 Gyr). 3) The AGN excess is larger for more luminous and bolometrically dominant AGN, and we find that AGN in post-mergers are generally more luminous than secularly triggered events. 4) A deficit of broad line AGN in the pre-merger phase, that evolves into an excess in post-mergers is consistent with evolution of the covering fraction of nuclear obscuring material. Before coalescence, tidally triggered inflows increase the covering fraction of nuclear dust; in the post-merger regime feedback from the AGN clears (at least some of) this material. 5) The statistical peak in the triggering of starbursts occurs contemporaneously with AGN, within 0.16 Gyr of coalescence.

Figures

Figures reproduced from arXiv: 2412.02804 by the authors.

Figure 1
Figure 1. — The stellar mass distributions of the post-merger sam￾ple in the four TPM time bins (blue) and matched control sample (red dashed). interaction-induced effects such as starbursts. Using the r-band images there ensures both a sensitive selection and one driven purely by morphology. Although DR5 covers a total of just under 5000 deg2 , we only apply mummi to the overlap with the SDSS DR7, thus providing complementar… view at source ↗
Figure 2
Figure 2. — The redshift distributions of the post-merger sample in the four TPM time bins (blue) and matched control sample (red dashed). al. (2013). There are 508,143 galaxies in the SDSS DR7 that fulfill these criteria and hence constitute the control pool for the pairs. For the post-mergers, the control pool consists of galaxies with no more than 2/20 of the neural net￾works assigning a merger classification (p > 0.5), yi… view at source ↗
Figure 3
Figure 3. presents the overall incidence of AGN in our post-merger sample as a Venn diagram. A separate Venn diagram is presented for each of the TPM bins. Although there is overlap between AGN samples selected using dif￾ferent diagnostics (although recall that the NLAGN and BLAGN classes are mutually exclusive by design), some sources are uniquely identified using just one metric. In order to attempt the most complete census… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: — The excess of AGN host galaxies identified by any of the diagnostics used in this paper (NLAGN, BLAGN and mid-IR colours) in the pairs (left side of the diagram) and post-mergers (right side of the diagram) compared with the control sample. The horizontal dashed line…
Figure 5
Figure 5. Figure 5: — The excess of AGN host galaxies identified by each individual diagnostic in the pairs (left side of the diagram) and post-mergers (right side of the diagram) compared with the control sample. From left to right the panels show the excess of NLAGN, mid-IR selected AGN…
Figure 6
Figure 6. Figure 6: — The excess of unobscured emission line AGN as a func￾tion of merger stage. The unobscured fraction of both merger and control samples is computed as NBLAGN /(NBLAGN +NNLAGN ), with the excess representing the ratio of fractions in the mergers relative to the controls…
Figure 7
Figure 7. Figure 7: — The excess of NLAGN host galaxies as a function of the [OIII]λ5007 line luminosity. Each panel counts NLAGN within a fixed range of L([OIII]). From left to right the panels show the excess of low (log L([OIII]) < 40 erg/s), medium (40 < log L([OIII]) < 41 erg/s) and …
Figure 8
Figure 8. Figure 8: — The enhancement in NLAGN luminosity (as measured by the luminosity of the [OIII]λ5007 line) of mergers compared with stellar mass and redshift matched control samples of non-merger NLAGN. On average, the AGN luminosity is enhanced by a factor of 2.5 immediately after…
Figure 9
Figure 9. Figure 9: — The excess of mid-IR AGN host galaxies as a function of W1 − W2 colour, which correlates with AGN luminosity. The left and right panels show AGN excess statistics for 0.5 < W1 − W2 < 0.7 and W1 − W2 > 0.7 respectively. The y-axis extent is fixed for both panels in or…
Figure 10
Figure 10. Figure 10: — The enhancement of W1−W2 colour of mid-IR AGN in mergers compared with stellar mass and redshift matched control samples of non-merger mid-IR selected AGN. The W1−W2 colour of mid-IR AGN in mergers is significantly redder than non-merger mid-IR AGN in the close pair…
Figure 11
Figure 11. Figure 11: shows that only NLAGN with log L([OIII]) ⪆ 41 erg/s have W1 − W2 > 0.5, and even then the selection is highly incomplete with many high luminos￾ity NLAGN falling below this mid-IR colour threshold. The combination of the figures presented in this Section therefore all…
Figure 12
Figure 12. Figure 12: — The excess of both starbursts (star symbols) and AGN (circles) identified by any of the diagnostics used in this paper (NLAGN, BLAGN and mid-IR colours, reproduced from [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. The Delay Time Distribution of Quasi-Periodic Eruptions

    astro-ph.GA 2026-08 conditional novelty 6.0 of 10

    QPE host galaxies are more likely to have recently formed a large burst of stars (burst mass fraction above 1%) than TDE host galaxies or mass- and redshift-matched controls.

  2. Cosmic Pairs: A DESI Census of Dual and Offset AGN as Precursors to Massive Black Hole Binaries

    astro-ph.GA 2026-07 conditional novelty 6.0 of 10

    DESI DR1 spectroscopy yields >7,000 candidate dual-AGN pairs, expanding the known kpc-scale census by ~4x and linking host-galaxy demographics to LISA-detectable massive black hole mergers.

  3. Performance of morphological classifiers for galaxy mergers compared to current machine learning methods

    astro-ph.GA 2026-07 conditional novelty 5.0 of 10

    Updated G-M20 and G-C morphological cuts achieve ~70% merger precision comparable to ML, with better high-z robustness, but only select pre-mergers.

Reference graph

Works this paper leans on

90 extracted references · 71 canonical work pages · cited by 3 Pith papers

  1. [1]

    S., Lambas, D

    Alonso, M. S., Lambas, D. G., Tissera, P. B., Coldwell, G., 2007, MNRAS, 375, 1017

    2007

  2. [2]

    J., et al., 2013, ApJ, 772, 26

    Assef, R. J., et al., 2013, ApJ, 772, 26

    2013

  3. [3]

    S., Comerford, J

    Barrows, R. S., Comerford, J. M., Stern, D., Assef, R. J., 2023, ApJ, 951, 92

    2023

  4. [4]

    Martin, M., 2014, MNRAS, 438, 1839

    2014

  5. [5]

    Martin, M., Cabrera-Lavers, A., 2017, MNRAS, 466, 3887

    2017

  6. [6]

    W., et al., 2021, MNRAS, 504, 372

    Bickley, R. W., et al., 2021, MNRAS, 504, 372

    2021

  7. [7]

    Gwyn, S., Hudson, M., 2022, MNRAS, 514, 3294

    2022

  8. [8]

    W., Ellison, S

    Bickley, R. W., Ellison, S. L., Patton, D. R., Wilkinson, S., 2023, MNRAS, 519, 6149

    2023

Show all 90 references
  1. [9]

    J., 2016, MNRAS, 462, 2246

    Blank, M., & Duschl, W. J., 2016, MNRAS, 462, 2246

    2016

  2. [10]

    F., Satyapal, S., Ellison, S

    Blecha, L., Snyder, G. F., Satyapal, S., Ellison, S. L., 2018, MNRAS, 478, 3056

    2018

  3. [11]

    Bottrell, C., et al., 2019, MNRAS, 490, 5390

    2019

  4. [12]

    R., 2023, MNRAS, 519, 4966

    Patton, D. R., 2023, MNRAS, 519, 4966

    2023

  5. [13]

    Ferreira, L., Hani, M., Quai, S., Wilkinson, S.,2024, MNRAS, 528, 5864

    2024

  6. [14]

    R., Dotti, M., Volonteri, M., Mayer, L., Bellovary, J

    Capelo, P. R., Dotti, M., Volonteri, M., Mayer, L., Bellovary, J. M., Shen, S., 2017, MNRAS, 469, 4437

    2017

  7. [15]

    2011, ApJ, 726, 57

    Cisternas, M., et al. 2011, ApJ, 726, 57

    2011

  8. [16]

    M., et al, 2024, ApJ, 963, 53

    Comerford, J. M., et al, 2024, ApJ, 963, 53

    2024

  9. [17]

    W., et al., 2010, MNRAS, 401, 1552

    Darg, D. W., et al., 2010, MNRAS, 401, 1552

    2010

  10. [18]

    L., et al., 2012, ApJ, 748, 142

    Donley, J. L., et al., 2012, ApJ, 748, 142

    2012

  11. [19]

    L., Harrison, C

    Dougherty, S. L., Harrison, C. M., Kocevski, D. D., Rosario, D. J., 2024, MNRAS, 527, 3146

    2024

  12. [20]

    G., Alonso, S., Coldwell, G

    Duplancic, F., Lambas, D. G., Alonso, S., Coldwell, G. V., 2021, MNRAS, 504, 4389

    2021

  13. [21]

    L., Patton, D

    Ellison, S. L., Patton, D. R., Simard, L., McConnachie, A. W., 2008 AJ, 135, 1877

    2008

  14. [22]

    L., Patton, D

    Ellison, S. L., Patton, D. R., Mendel, J. T., Scudder, J. M., 2011, MNRAS, 418, 2043

    2011

  15. [23]

    L., Mendel, J

    Ellison, S. L., Mendel, J. T., Patton, D. R., Scudder, J. M., 2013, MNRAS, 453, 3627

    2013

  16. [24]

    L., Patton, D

    Ellison, S. L., Patton, D. R., Hickox, R. C., 2015, MNRAS, 451, L35

    2015

  17. [25]

    W., Gwyn, S., Cuillandre, J.-C., 2019, MNRAS, 487, 2491

    McConnachie, A. W., Gwyn, S., Cuillandre, J.-C., 2019, MNRAS, 487, 2491

    2019

  18. [26]

    L., Ferreira, L., Wild, V., Wilkinson, S., Rowlands, K., Patton, D

    Ellison, S. L., Ferreira, L., Wild, V., Wilkinson, S., Rowlands, K., Patton, D. R., 2024, OJAp, 7, 121

    2024

  19. [27]

    Griffiths, A., Whitney, A., 2020, ApJ, 895, 115

    2020

  20. [28]

    Ferreira, L., et al., 2024, MNRAS, 533, 2547

    2024

  21. [29]

    Ferreira, L., et al., 2025, MNRAS, in press

    2025

  22. [30]

    M., et al., 2009, ApJ, 691, 705

    Gabor, J. M., et al., 2009, ApJ, 691, 705

    2009

  23. [31]

    Gao, F., et al., 2020, A&A, 637, 94

    2020

  24. [32]

    D., et al., 2018, PASJ, 70, 37

    Goulding, A. D., et al., 2018, PASJ, 70, 37

    2018

  25. [33]

    L., 2013, ApJ, 774, 145

    Zakamska, N. L., 2013, ApJ, 774, 145

    2013

  26. [34]

    H., Gosain, H., Ellison, S

    Hani, M. H., Gosain, H., Ellison, S. L., Patton, D. R., Torrey, P., 2020, MNRAS, 493, 3716

    2020

  27. [35]

    Hopkins, P., 2012, MNRAS, 420, L8

    2012

  28. [36]

    Hou, M., Li, Z., Liu, X., 2020, ApJ, 900, 79

    2020

  29. [37]

    Hou, M., et al., 2023, ApJ, 943, 50

    2023

  30. [38]

    C, & Alexander, D

    Hickox, R. C, & Alexander, D. M., 2018, ARA&A, 56, 625

    2018

  31. [39]

    J., Dopita, M

    Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., Trevena, J., 2001, ApJ, 556, 121

    2001

  32. [40]

    J., Nicholls, D

    Kewley, L. J., Nicholls, D. C., Sutherland, R., 2019, ARA&A, 57, 511

    2019

  33. [41]

    L., Mendel, J

    Ellison, S. L., Mendel, J. T., Patton, D. R., 2014, ApJ, 795, 62

    2014

  34. [42]

    2012, ApJ, 744, 148

    Kocevski, D., et al. 2012, ApJ, 744, 148

    2012

  35. [43]

    D., et al., 2015, ApJ, 814, 104

    Kocevski, D. D., et al., 2015, ApJ, 814, 104

    2015

  36. [44]

    2021, arXiv e-prints, arXiv:2102.05182

    Koppula, S., et al. 2021, arXiv e-prints, arXiv:2102.05182

    2021 arXiv

  37. [45]

    Koss, M., Mushotzky, R., Veilleux, S., Winter, L., 2010, ApJ, 716, L125

    2010

  38. [46]

    Vasudevan, R., Trippe, M., 2012, ApJ, 746, L22

    2012

  39. [47]

    Koss, M., et al., 2018, Nature, 563, 214 La Marca, A., et al., 2024, A&A, in press

    2018

  40. [48]

    C., Barcons, X., Carrera, F

    Lamastra, A., Bianchi, S., Matt, G., Perola, G. C., Barcons, X., Carrera, F. J., 2009, A&A, 504, 73

    2009

  41. [49]

    L., et al., 2021, ApJ, 919, 129

    Lambrides, E. L., et al., 2021, ApJ, 919, 129

    2021

  42. [50]

    W., Schlegel, D

    Lang, D., Hogg, D. W., Schlegel, D. J., 2016, AJ, 151, 36

    2016

  43. [51]

    P., 2008, MNRAS, 385, 1915

    Jing, Y. P., 2008, MNRAS, 385, 1915

    2008

  44. [52]

    Li, J., et al., 2020, ApJ, 903, 49

    2020

  45. [53]

    Li, C., et al., 2023, ApJ, 944, 168

    2023

  46. [54]

    A., 2012, ApJ, 745, 94

    Liu, X., Shen, Y., Strauss, M. A., 2012, ApJ, 745, 94

    2012

  47. [55]

    Liu, H.-Y., Liu, W.-J., Dong, X.-B., Zhou, H., Wang, T., Lu, H., Yuan, W., 2019, ApJS, 243, 21

    2019

  48. [56]

    Marian, V., et al., 2020, ApJ, 904, 79

    2020

  49. [57]

    M., Rosario, D

    McAlpine, S., Harrison, C. M., Rosario, D. J., Alexander, D. M., Ellison, S. L., Johansson, P. H., Patton, D. R., 2020, MNRAS, 494, 5713

    2020

  50. [58]

    Nelson, D., et al., 2019, ComAC, 6, 2

    2019

  51. [59]

    C., et al., 2023, A&A, 679, 142

    Omori, K. C., et al., 2023, A&A, 679, 142

    2023

  52. [60]

    T., Moreno, J., Torrey, P., 2016, MNRAS, 461, 2589

    Simard, L., Mendel, J. T., Moreno, J., Torrey, P., 2016, MNRAS, 461, 2589

    2016

  53. [61]

    J., Wang, L., Trayford, J

    Pearson, W. J., Wang, L., Trayford, J. W., Petrillo, C. E., van der Tak, F. F. S., 2019, A&A, 626A, 49

    2019

  54. [62]

    J., et al., 2022, A&A, 661, 52

    Pearson, W. J., et al., 2022, A&A, 661, 52

    2022

  55. [63]

    Margalef-Bentabol, B., 2024, A&A, 687, 45

    2024

  56. [64]

    Pierce, J. C. S., et al., 2022, MNRAS, 510, 1163

    2022

  57. [65]

    Pierce, J. C. S., et al., 2023, MNRAS, 522, 1736 Ramos Almeida, C., Tadhunter, C. N., Inskip, K. J.,

    2023

  58. [66]

    Morganti, R., Holt, J.; Dicken, D., 2011, MNRAS, 410, 1550 Ramos Almeida, C., & Ricci, C., 2017, NatAs, 1, 679

    2011

  59. [67]

    Ricci, C., et al., 2017, MNRAS, 468, 1273

    2017

  60. [68]

    Ricci, C., et al., 2021, MNRAS, 506, 5935

    2021

  61. [69]

    N., Argudo-Fernandez, M., 2013, MNRAS, 430, 638

    Sabater, J., Best, P. N., Argudo-Fernandez, M., 2013, MNRAS, 430, 638

    2013

  62. [70]

    Salim, S., et al., 2016, ApJS, 227, 2

    2016

  63. [71]

    Z.., 1988, ApJ, 325, 74

    Matthews, K., Neugebauer, G., Scoville, N. Z.., 1988, ApJ, 325, 74

    1988

  64. [72]

    R., Mendel, J

    Patton, D. R., Mendel, J. T., 2014, MNRAS, 441, 1297

    2014

  65. [73]

    Satyapal, S., et al., 2017, ApJ, 848, 126

    2017

  66. [74]

    M., Treister, E., Kaviraj, S., Kushkuley, B., 2009, ApJ, 692, L19

    Schawinski, K., Virani, S., Simmons, B., Urry, C. M., Treister, E., Kaviraj, S., Kushkuley, B., 2009, ApJ, 692, L19

    2009

  67. [75]

    Scott, C., & Kaviraj, S., 2014, MNRAS, 437, 2137

    2014

  68. [76]

    T., 2012, MNRAS, 426, 549

    Mendel, J. T., 2012, MNRAS, 426, 549

    2012

  69. [77]

    Shah, E., et al., 2020, ApJ, 904, 107 15

    2020

  70. [78]

    D., Martis, N., Iono, D., Espada, D., Skelton, R., 2021, ApJ, 909, 124

    Silva, A., Marchesini, D., Silverman, J. D., Martis, N., Iono, D., Espada, D., Skelton, R., 2021, ApJ, 909, 124

    2021

  71. [79]

    D., et al., 2011, ApJ, 743, 2

    Silverman, J. D., et al., 2011, ApJ, 743, 2

    2011

  72. [80]

    Springel, V., et al., 2005, Nature, 435, 629

    2005

  73. [81]

    K., et al., 2018, MNRAS, 481, 341

    Steinborn, L. K., et al., 2018, MNRAS, 481, 341

    2018

  74. [82]

    M., Barrows, R

    Stemo, A., Comerford, J. M., Barrows, R. S., Stern, D., Assef, R. J., Griffith, R. L., Schechter, A., 2021, ApJ, 923, 36

    2021

  75. [83]

    Stern, D., et al., 2012, ApJ, 753, 30

    2012

  76. [84]

    M., Simmons, B

    Treister, E., Schawinski, K., Urry, C. M., Simmons, B. D., 2012, ApJ, 758, L39

    2012

  77. [85]

    Villforth, C., et al., 2017, MNRAS, 466, 812

    2017

  78. [86]

    Villforth, C., 2023, OJAp, 6, 34

    2023

  79. [87]

    Walmsley, M., et al., 2022, MNRAS, 509, 3966

    2022

  80. [88]

    L., 2017, MNRAS, 464, 3882

    Cooper, A., McConnell, A., Nielsen, J. L., 2017, MNRAS, 464, 3882

    2017

  81. [89]

    Wild, V., Heckman, T., Charlot, S., 2010, MNRAS, 405, 933

    2010

  82. [90]

    F., Geller, M

    Woods, D. F., Geller, M. J., 2007, AJ, 134, 527 This paper was built using the Open Journal of As- trophysics LATEX template. The OJA is a journal which provides fast and easy peer review for new papers in the astro-phsectionofthearXiv, makingthereviewingpro- cess simpler for ...

    2007

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.