Pith. sign in

REVIEW 4 major objections 5 minor 119 references

A random forest trained on simulated clusters can identify backsplash galaxies—galaxies that have already passed through a cluster—from observable properties, with up to ~70% purity and completeness.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 00:39 UTC pith:NOC4KBAF

load-bearing objection A genuinely useful simulation-based classifier for backsplash galaxies, with a Virgo application that rests on an untested mass extrapolation. the 4 major comments →

arxiv 2607.26136 v1 pith:NOC4KBAF submitted 2026-07-28 astro-ph.GA

Identifying backsplash galaxies using machine learning

Roan Haggar , Elizaveta Sazonova , Cameron R. Morgan , Alexander Knebe , Rhys Jordan , Weiguang Cui , Frazer R. Pearce , James E. Taylor This is my paper
classification astro-ph.GA
keywords backsplash galaxiesgalaxy clustersrandom forest classifiermachine learning classificationgalaxy infallram pressure strippingVirgo Clusterphase space
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper claims that a machine-learning classifier can separate two populations that look nearly identical in ordinary observations: backsplash galaxies, which have previously crossed a cluster's centre and are now back in its outskirts, and galaxies falling into the cluster for the first time. Trained on a suite of hydrodynamical simulations of massive clusters, the random forest reaches up to ~70% purity and completeness for backsplash galaxies and over 80% for first-time infallers, using only quantities that can be measured in real surveys. The payoff is a way to remove a known contaminant from studies of galaxy pre-processing: without orbital information, quenched galaxies in cluster outskirts could have been changed by their own prior passage through the cluster rather than by groups and filaments. As a proof of concept, the model classifies all ten Virgo galaxies with asymmetric HI tails as first-time infallers, supporting the idea that cold gas is stripped during a galaxy's first approach to a cluster.

Core claim

The paper's central claim is that an ensemble of decision trees can infer a galaxy's orbital history from observable projected quantities alone. On simulated test clusters, the full model reaches 84% accuracy in the annulus between one and two cluster radii, with 75% purity and completeness for backsplash galaxies and 88% for first-time infallers; the threshold can be tuned to trade purity against completeness. In the simplest version, using only distance, velocity, and the cluster's magnitude gap, accuracy is 74%. Applied to ten Virgo galaxies with asymmetric HI distributions, the model labels all ten as first-time infallers, which the paper interprets as evidence that ram pressure removes

What carries the argument

The carrying mechanism is a user-tunable random forest—an ensemble of decision trees. Its 14 input features are all observationally measurable: projected distance from the cluster centre, line-of-sight velocity, stellar and halo mass, velocity direction in the plane of the sky, distance to the fourth-nearest neighbour, morphologies, and a cluster magnitude gap. Because the forest can be retrained on any subset of these, users with sparse survey data get the best model their features allow. The crucial property is that it learns a blurred, cluster-dependent boundary in projected phase space, rather than the hard 'backsplash region' cuts used in previous work.

Load-bearing premise

The load-bearing assumption is that the simulated galaxy clusters used for training are representative of real clusters such as Virgo, even though Virgo's halo mass lies below the lowest-mass cluster in the training sample, and all performance metrics are computed on simulated test clusters rather than on any observed ground truth (Section 4.2).

What would settle it

A concrete test would be to apply the published classifier to a sample of nearby galaxies whose orbital status can be checked independently—for example, via proper-motion orbits in the Local Group—and compare the labels. The paper itself notes that one Virgo galaxy (VCC 2066) flips from first infall to backsplash when the nearest-neighbour feature is excluded, so a larger sample of such flips would settle whether the model is really tracking orbital history or just local environment.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • If the model works on real clusters, spectroscopic surveys can produce per-galaxy lists of backsplash and infalling galaxies in cluster outskirts instead of statistical counts or hard phase-space cuts.
  • Users can tune the classification threshold to favour high purity or high completeness; the paper shows overall accuracy stays above 65% even at extreme thresholds.
  • The Virgo application implies that galaxies with asymmetric HI tails are preferentially first-time infallers, giving observers a direct way to select recently stripped galaxies in nearby clusters.
  • Because the model accepts any subset of its 14 features, existing surveys with only positions, velocities, and a magnitude gap can use it immediately, at 74% accuracy rather than 84%.
  • A public web app retrains the forest on-the-fly for the user's chosen feature set, letting any observer generate backsplash likelihoods for their own catalogue.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A testable extension beyond the paper's scope: retrain the classifier on simulations that span Virgo's halo mass to see whether the Virgo classifications remain stable; if they do not, the model is extrapolating below its training range.
  • The paper deliberately excludes gas content from the features; combining the classifier with gas-based selection could look for rare backsplash galaxies that kept their gas, testing stripping timescales in a way the paper does not.
  • Because the blurred boundary depends partly on the cluster's magnitude gap, a reader could probe how sensitive the classifications are to the assumed R200 and velocity dispersion of a real cluster; the paper tests only one set of Virgo assumptions.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper presents a random forest classifier trained on the Gizmo-Simba run of The Three Hundred cluster simulations to label galaxies in the projected annulus [R200, 2R200] as either backsplash or first-time infalling, using up to 14 observable or semi-observable features. It reports purity/completeness for five feature subsets, compares with a literature phase-space cut (Rhee et al. 2017), and shows that the full model reaches P=C=75% for backsplash and P=C=88% for infalling galaxies at 84% accuracy. The authors then apply a five-feature version to 10 Virgo galaxies with asymmetric H i tails, concluding that all 10 are likely first-time infallers, supporting the idea that cold gas is stripped on first passage. A public web app is provided. The main claims are the simulation-internal classifier performance and the Virgo proof-of-concept.

Significance. If the reported P/C values are robust, the classifier is a practical step beyond hard phase-space cuts: it provides per-galaxy probabilities, can be tuned for purity or completeness, and is delivered as a web app. The feature-importance analysis and explicit comparison to R17 are useful, and the Virgo application, while small, is the kind of proof-of-concept the community needs. The result's significance is currently capped by the fact that all performance metrics are internal to The300 and the observational application rests on an untested mass extrapolation and an unvalidated feature combination; those issues are fixable but block the strong 'demonstrated' claim.

major comments (4)
  1. [§4.2, §2.1] The Virgo mass extrapolation is the weakest link in the observational application. The text in §4.2 acknowledges that R200=0.97 Mpc places Virgo below the mass range of The300 (M200 in 5e14–2.6e15 h^-1 M_sun; R200 in 1.3–2.2 h^-1 Mpc), i.e., roughly a factor of two below the minimum training radius. The only defense given is that the features are re-scaled by R200/σ and that the impact 'should be minimal'. That is not a test. Moreover, §3.1 reports a Spearman test between accuracy and cluster mass with ρ_s=0.23, p=0.07; p≈0.07 is at best marginal evidence of mass independence, and the tested mass range does not include Virgo. I request a deliberate test—e.g., training on the full suite and showing performance as a function of M200 for the lowest-mass clusters, or a rescaled/mass-limited mock—or, failing that, re-framing the Virgo result as an explicit extrapolation. As written, the abstr
  2. [§4.2, Table 1] The exact feature combination used for Virgo is not evaluated. The paper itself notes that 'the combination of parameters we use in Section 4.2 does not exactly match any of the combinations listed above'. The Virgo model uses d_proj, v_LOS, M12, θ_v, and d_proj,nn4. Table 1 reports P/C for 'Simplest + velocity angle' (without d_proj,nn4) and 'Simplest + neighbours' (without θ_v), but not the actual combination deployed. The VCC 2066 discussion shows that removing d_proj,nn4 changes the classification, so the P/C of the exact model that produced all 10 Virgo labels is unknown. Please add this feature set to Table 1—with accuracy, P/C, threshold, and bootstrap uncertainties—so the reader can see how the deployed model performs on the simulated test set.
  3. [§3 (intro), §2.1] All reported validation is internal to The300: the classifier is trained and tested on the same simulation suite, and the class labels come from the same merger trees that define the classes. The paper states that verification against real data is not easy and that all verification is carried out on the simulated test set. Since several input features (e.g., m_h, θ_v, morphological terms) are affected by the simulation's baryonic and subgrid modeling, the 74–84% accuracies and, importantly, the calibrated threshold p(Backsplash) may not transfer to observations. I ask for at least one cross-simulation validation with independently defined orbits (e.g., TNG300 or Magneticum) or, if that is not feasible in this paper, a substantial softening of the claims and a clear statement that the P/C are simulation-internal.
  4. [§4.2] The Virgo sample selection is not fully quantified. The sample is 10 galaxies chosen by visual inspection of H i tails, after excluding cases where two or more of the three authors could not identify a clear tail direction. No inter-rater scatter is reported, no selection function is given for the 14 galaxies with H i maps out of the 62 above the stellar mass cut, and the completeness of the VCC/EVCC parent sample is not discussed. With N=10, the statement that the sample is 'strongly biased towards galaxies approaching the cluster for the first time' is fragile; VCC 2066 lies close to the threshold and changes classification when d_proj,nn4 is removed. Please show the distribution of backsplash probabilities with bootstrap confidence intervals, and quantify the tail-angle measurement uncertainty so the reader can see how many of the 10 classifications are secure.
minor comments (5)
  1. [Fig. 9 caption] The caption says 'arrows pointing up show galaxies moving perpendicular to the cluster, and arrows pointing up show galaxies receding from the cluster centre'; the second occurrence should presumably be a different direction. Please correct.
  2. [Table 2] The column header 'cos(θv),' contains a stray comma; also, consider giving units in the header (R200 and σ).
  3. [Fig. 8] The KDE in the top row extends below the stated stellar mass limit 10^9.5 h^-1 M_sun. The text explains this is a bandwidth artifact, but a vertical dashed line marking the limit would avoid confusion.
  4. [Section 2.3.1] The threshold is chosen to match the simulated backsplash fraction on the training set. This is not circular for the Virgo predictions, but if The300's backsplash fraction is biased, the operational point of the classifier will be biased. A brief caveat to this effect would help.
  5. [Data availability / web app] The web app is a valuable deliverable, but the training catalogue and classifier code are not archived or versioned. For reproducibility, please provide a permanent DOI/code repository or at least a versioned release of the app's backend alongside the paper.

Circularity Check

0 steps flagged

No significant circularity: the classifier is trained on simulated orbits and tested on held-out clusters; the Virgo result is an out-of-sample application with no fitting.

full rationale

The derivation is self-contained. Backsplash/infalling labels are defined from 3D orbits traced through The300 merger trees (Section 2.2), independent of the observable features used by the random forest. Performance metrics (Section 3, Table 1) are computed on a held-out set of 65 clusters never used in training. The threshold is calibrated to the training-set backsplash fraction (Section 2.3.1), but this is standard model selection and does not enter the test metrics or the Virgo prediction. The Virgo application (Section 4.2) feeds five observed features into the pre-trained classifier and uses no Virgo measurements to fit or update the model; none of the 10 HI-tail galaxies' classifications are used as training data. The paper even avoids a genuinely circular use of morphology (Section 3.5) by not analyzing morphology as an outcome when morphology is included as a feature. The acknowledged self-citations (Haggar et al. 2020, 2023) motivate feature choices and radial-scope choices and are corroborated by external citations (Borrow et al. 2023; Boselli et al. 2014, 2023); they are not the sole support for any prediction. The main limitation is extrapolation of the simulation-trained model to Virgo at R200 below the training mass range (Section 4.2), and the lack of observational ground truth for backsplash membership; these are generalizability/validity concerns, not circularity.

Axiom & Free-Parameter Ledger

2 free parameters · 4 axioms · 0 invented entities

The paper introduces no new physical entities or free physical constants. Its central claim rests on the choice of R200 as the backsplash boundary, the two-class split of the outer annulus, and the realism of The300 simulations; the ML threshold and hyperparameters are model choices rather than physical parameters.

free parameters (2)
  • Classification threshold p(Backsplash) = 0.58–0.61 (default)
    Tuned cluster-by-cluster so the predicted backsplash fraction matches the simulated truth; directly controls the purity/completeness trade-off (Section 2.3.1).
  • Random forest hyperparameters = 50 trees, depth 30, 15,000 per class
    Chosen by accuracy-plateau tests and threshold granularity; not fit to external data but affect results (Section 2.3).
axioms (4)
  • domain assumption R200 defines the cluster boundary for backsplash classification
    Galaxies that previously crossed R200 are 'backsplash'; R200 is a standard but arbitrary choice (Section 2.2).
  • domain assumption Galaxies in the projected annulus [R200, 2R200] are either backsplash or first-time infallers
    Cluster members inside the annulus are ignored; the two-class simplification underlies the classifier (Section 2.2).
  • domain assumption The300 Gizmo-Simba simulations faithfully reproduce the dynamical observables of real clusters
    The model's training data is entirely simulated; cross-simulation validation is planned, not performed (Sections 2.1, 5).
  • standard math Projecting each cluster along three orthogonal lines of sight provides independent samples
    Used to triple the dataset; assumes projection directions are equivalent (Section 2.2).

pith-pipeline@v1.3.0-alltime-deepseek · 29095 in / 9539 out tokens · 88421 ms · 2026-08-01T00:39:13.657009+00:00 · methodology

0 comments
read the original abstract

The galaxy population in the outskirts of a cluster contains members that have been pre-processed in groups and filaments, as well as backsplash galaxies -- those that have recently passed through the cluster's center. However, disentangling these two pathways is challenging observationally. In this work, we present a machine-learning-powered model, trained on simulations of galaxy clusters from The Three Hundred suite of simulations, which can identify individual backsplash galaxies in astronomical observations. This model can build samples of backsplash galaxies with a purity and completeness of up to ~70%, and galaxies on their first infall with a purity and completeness of over 80%. It can be tuned to optimise either of these two metrics, and can be used with any combination of a set of observable quantities. We have also applied this model to galaxies with asymmetric HI distributions in the Virgo Cluster, and have demonstrated that these galaxies are all likely approaching the cluster for the first time. This supports the idea that cold gas is removed from these galaxies soon after entering a cluster, and demonstrates how this classifier can provide a better understanding of which properties of galaxies are caused by a previous passage through a cluster. We have made this model publicly available in the form of a web app, with a link in the Conclusions of this paper.

Figures

Figures reproduced from arXiv: 2607.26136 by Alexander Knebe, Cameron R. Morgan, Elizaveta Sazonova, Frazer R. Pearce, James E. Taylor, Rhys Jordan, Roan Haggar, Weiguang Cui.

Figure 1
Figure 1. Figure 1: Plot showing region of cluster-centric phase space stated to contain backsplash galaxies, from a selection of previous works in the literature (Muriel & Coenda 2014; Rhee et al. 2017; Ferreras et al. 2023; Martínez et al. 2023). In each case, the ‘backsplash region’ is the region below these lines. In some of these cases (Rhee et al. 2017; Ferreras et al. 2023), the backsplash region extends to radii outsi… view at source ↗
Figure 2
Figure 2. Figure 2: Relative importance of all 14 measurable quantities in classifying backsplash galaxies, plus a random number generator as a baseline measure￾ment. This is used to identify metrics that provide no classifying power, which are shown as unfilled circles. 3.1 Complete set of galaxy properties The first iteration of this model we tested is the complete random forest classifier. In this case, all 14 of the param… view at source ↗
Figure 4
Figure 4. Figure 4: Backsplash fraction as a function of projected distance from the cluster centre, 𝑑proj, in units of the cluster radius, 𝑅200, and line-of-sight velocity relative to the cluster centre, 𝑣LOS, in units of the line-of-sight cluster velocity dispersion, 𝜎. Left panel shows the true fraction of backsplash galaxies in our training set. Centre panel shows the fraction from our test set, predicted using the random… view at source ↗
Figure 5
Figure 5. Figure 5: Purity and completeness of backsplash (left) and infalling (centre) samples, as a function of backsplash galaxy threshold value. Right panel shows the overall accuracy of the model as a function of the threshold value. Vertical line represents the fiducial threshold value, p(Backsplash) = 0.61, as described in Section 2.3.1. Shaded regions are the uncertainties derived by bootstrapping from a binomial dist… view at source ↗
Figure 6
Figure 6. Figure 6: Equivalent figures to [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Example cluster from The300 simulations (ID: cluster_0141). Left panel shows the mass distribution of gas in this cluster, right panel shows the distribution of stars. The large white circles represent distances of 𝑅200 and 2𝑅200 from the cluster centre. The right panel shows backsplash galaxies and infalling galaxies identified by our model, along with their status as true/false backsplash/infalling galax… view at source ↗
Figure 8
Figure 8. Figure 8: Distributions of stellar masses (top), and halo-to-stellar mass ratios (bottom), for galaxies in the outskirts of our example cluster, shown in [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Distribution of 10 galaxies with H i tails in the outskirts of the Virgo cluster, in cluster-centric position-velocity space. The size of each point corresponds to 𝑑proj,nn4, such that larger points are more isolated from nearby galaxies. The arrows represent 𝜃v; arrows pointing left represent galaxies approaching the cluster centre, arrows pointing up show galaxies moving perpendicular to the cluster, and… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

119 extracted references · 2 canonical work pages

  1. [1]

    , keywords =

    Probing Galaxy Evolution in Massive Clusters Using ACT and DES: Splashback as a Cosmic Clock. , keywords =. doi:10.3847/1538-4357/ac0bbc , archivePrefix =. 2008.11663 , primaryClass =

    Pith/arXiv arXiv 2008

  2. [2]

    arXiv e-prints , keywords =

    Cluster properties as a function of dynamical state in the DESI Legacy x UNIONS surveys. arXiv e-prints , keywords =. doi:10.48550/arXiv.2512.14636 , archivePrefix =. 2512.14636 , primaryClass =

    doi:10.48550/arxiv.2512.14636

  3. [3]

    , keywords =

    The effect of local and large-scale environments on nuclear activity and star formation. , keywords =. doi:10.1051/0004-6361/201628232 , archivePrefix =. 1605.05642 , primaryClass =

    Pith/arXiv arXiv

  4. [4]

    , keywords =

    The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package. , keywords =. doi:10.3847/1538-4357/ac7c74 , archivePrefix =. 2206.14220 , primaryClass =

    Pith/arXiv arXiv

  5. [5]

    , keywords =

    Why does the environmental influence on group and cluster galaxies extend beyond the virial radius?. , keywords =. doi:10.1093/mnras/stt109 , archivePrefix =. 1210.8407 , primaryClass =

    Pith/arXiv arXiv

  6. [6]

    , keywords =

    Galaxy bimodality versus stellar mass and environment. , keywords =. doi:10.1111/j.1365-2966.2006.11081.x , archivePrefix =. astro-ph/0607648 , primaryClass =

    arXiv 2006

  7. [7]

    , keywords =

    Differential Galaxy Evolution in Cluster and Field Galaxies at z -0.5ex 0.3. , keywords =. doi:10.1086/308056 , archivePrefix =. astro-ph/9906470 , primaryClass =

    Pith/arXiv arXiv

  8. [8]

    , keywords =

    The Origin of Star Formation Gradients in Rich Galaxy Clusters. , keywords =. doi:10.1086/309323 , archivePrefix =. astro-ph/0004078 , primaryClass =

    Pith/arXiv arXiv

  9. [10]

    GASP. II. A MUSE View of Extreme Ram-Pressure Stripping along the Line of Sight: Kinematics of the Jellyfish Galaxy JO201. , keywords =. doi:10.3847/1538-4357/aa7875 , archivePrefix =. 1704.05087 , primaryClass =

    Pith/arXiv arXiv

  10. [11]

    , keywords =

    Accretion of galaxy groups into galaxy clusters. , keywords =. doi:10.1093/mnras/staa2636 , archivePrefix =. 2005.05344 , primaryClass =

    Pith/arXiv arXiv 2005

  11. [12]

    , keywords =

    The Orbits of Isolated Dwarfs in the Local Group from New 3D Kinematics: Constraints on First Infall, Backsplash, and Quenching Mechanisms. , keywords =. doi:10.3847/1538-4357/ae0733 , archivePrefix =. 2509.11299 , primaryClass =

    arXiv

  12. [13]

    , keywords =

    Convolutional neural network identification of galaxy post-mergers in UNIONS using IllustrisTNG. , keywords =. doi:10.1093/mnras/stab806 , archivePrefix =. 2103.09367 , primaryClass =

    Pith/arXiv arXiv

  13. [14]

    Studies of the Virgo cluster. II. A catalog of 2096 galaxies in the Virgo cluster area. , keywords =. doi:10.1086/113874 , adsurl =

    doi:10.1086/113874 2096

  14. [15]

    , keywords =

    There and back again: Understanding the critical properties of backsplash galaxies. , keywords =. doi:10.1093/mnras/stad045 , archivePrefix =. 2205.10376 , primaryClass =

    Pith/arXiv arXiv

  15. [16]

    The GALEX Ultraviolet Virgo Cluster Survey (GUViCS). IV. The role of the cluster environment on galaxy evolution. , keywords =. doi:10.1051/0004-6361/201424419 , archivePrefix =. 1407.4986 , primaryClass =

    Pith/arXiv arXiv

  16. [17]

    A Virgo Environmental Survey Tracing Ionised Gas Emission (VESTIGE). XV. The H luminosity function of the Virgo cluster. , keywords =. doi:10.1051/0004-6361/202346506 , archivePrefix =. 2305.15919 , primaryClass =

    Pith/arXiv arXiv

  17. [18]

    Machine Learning , keywords =

    Random Forests. Machine Learning , keywords =. doi:10.1023/A:1010933404324 , adsurl =

    doi:10.1023/a:1010933404324

  18. [19]

    The evolution of galaxies in clusters. V. A study of populations since Z 0.5. , keywords =. doi:10.1086/162519 , adsurl =

    doi:10.1086/162519

  19. [20]

    Benchmark for early-type galaxies scaling relations from 260 dynamical models: mass-to-light ratio, dark matter, Fundamental Plane and Mass Plane

    The ATLAS ^ 3D project - XV. Benchmark for early-type galaxies scaling relations from 260 dynamical models: mass-to-light ratio, dark matter, Fundamental Plane and Mass Plane. , keywords =. doi:10.1093/mnras/stt562 , archivePrefix =. 1208.3522 , primaryClass =

    Pith/arXiv arXiv

  20. [21]

    , keywords =

    A Physically Motivated Framework to Compare the Merger Timescales of Isolated Low- and High-mass Galaxy Pairs Across Cosmic Time. , keywords =. doi:10.3847/1538-4357/ad7bad , archivePrefix =. 2409.02233 , primaryClass =

    Pith/arXiv arXiv

  21. [22]

    , keywords =

    The GOGREEN Survey: Evidence of an Excess of Quiescent Disks in Clusters at 1.0. , keywords =. doi:10.3847/1538-4357/ac1117 , archivePrefix =. 2107.03403 , primaryClass =

    Pith/arXiv arXiv

  22. [23]

    VLA Imaging of Virgo Spirals in Atomic Gas (VIVA). I. The Atlas and the H I Properties. , keywords =. doi:10.1088/0004-6256/138/6/1741 , adsurl =

    doi:10.1088/0004-6256/138/6/1741

  23. [24]

    Dark matter halo properties of intermediate-z star-forming galaxies

    MUSE-DARK: I. Dark matter halo properties of intermediate-z star-forming galaxies. , keywords =. doi:10.1051/0004-6361/202557396 , adsurl =

    work page doi:10.1051/0004-6361/202557396

  24. [25]

    Synthesis of environmental and star-forming regulation mechanisms

    Semi-analytic galaxies - I. Synthesis of environmental and star-forming regulation mechanisms. , keywords =. doi:10.1093/mnras/sty1131 , archivePrefix =. 1801.03883 , primaryClass =

    Pith/arXiv arXiv

  25. [26]

    , keywords =

    The Three Hundred project: a large catalogue of theoretically modelled galaxy clusters for cosmological and astrophysical applications. , keywords =. doi:10.1093/mnras/sty2111 , archivePrefix =. 1809.04622 , primaryClass =

    Pith/arXiv arXiv

  26. [27]

    , keywords =

    THE THREE HUNDRED project: The GIZMO-SIMBA run. , keywords =. doi:10.1093/mnras/stac1402 , archivePrefix =. 2202.14038 , primaryClass =

    Pith/arXiv arXiv

  27. [28]

    , keywords =

    MUFASA: galaxy formation simulations with meshless hydrodynamics. , keywords =. doi:10.1093/mnras/stw1862 , archivePrefix =. 1604.01418 , primaryClass =

    Pith/arXiv arXiv

  28. [29]

    , keywords =

    SIMBA: Cosmological simulations with black hole growth and feedback. , keywords =. doi:10.1093/mnras/stz937 , archivePrefix =. 1901.10203 , primaryClass =

    Pith/arXiv arXiv 1901

  29. [30]

    , keywords =

    ROGER: Reconstructing orbits of galaxies in extreme regions using machine learning techniques. , keywords =. doi:10.1093/mnras/staa3339 , archivePrefix =. 2010.11959 , primaryClass =

    Pith/arXiv arXiv 2010

  30. [31]

    , keywords =

    Clusters' far-reaching influence on narrow-angle tail radio galaxies. , keywords =. doi:10.1093/mnrasl/slab075 , archivePrefix =. 2107.00449 , primaryClass =

    Pith/arXiv arXiv

  31. [32]

    , keywords =

    Improving galaxy morphologies for SDSS with Deep Learning. , keywords =. doi:10.1093/mnras/sty338 , archivePrefix =. 1711.05744 , primaryClass =

    Pith/arXiv arXiv

  32. [33]

    , keywords =

    Galaxy morphology in rich clusters: implications for the formation and evolution of galaxies. , keywords =. doi:10.1086/157753 , adsurl =

    doi:10.1086/157753

  33. [34]

    The IMACS Cluster Building Survey. II. Spectral Evolution of Galaxies in the Epoch of Cluster Assembly. , keywords =. doi:10.1088/0004-637X/770/1/62 , archivePrefix =. 1303.4272 , primaryClass =

    Pith/arXiv arXiv

  34. [35]

    , keywords =

    VICTORIA project: The LOFAR HBA Virgo Cluster Survey. , keywords =. doi:10.1051/0004-6361/202346458 , archivePrefix =. 2306.04513 , primaryClass =

    Pith/arXiv arXiv

  35. [36]

    Overview of the Euclid mission

    Euclid: I. Overview of the Euclid mission. , keywords =. doi:10.1051/0004-6361/202450810 , archivePrefix =. 2405.13491 , primaryClass =

    arXiv

  36. [37]

    The Next Generation Virgo Cluster Survey (NGVS). I. Introduction to the Survey. , keywords =. doi:10.1088/0067-0049/200/1/4 , adsurl =

    doi:10.1088/0067-0049/200/1/4

  37. [38]

    , keywords =

    Exploring the stellar populations of backsplash galaxies. , keywords =. doi:10.1093/mnras/stad001 , archivePrefix =. 2301.01776 , primaryClass =

    Pith/arXiv arXiv

  38. [39]

    , keywords =

    Dark matter haloes within clusters. , keywords =. doi:10.1046/j.1365-8711.1998.01918.x , archivePrefix =. astro-ph/9801192 , primaryClass =

    arXiv 1998

  39. [40]

    Virgo Filaments. V. Disrupting the Baryon Cycle in the NGC 5364 Galaxy Group. , keywords =. doi:10.3847/1538-4357/adc566 , archivePrefix =. 2505.09782 , primaryClass =

    Pith/arXiv arXiv

  40. [43]

    The role of group environment in quenching star formation

    The miniJPAS survey. The role of group environment in quenching star formation. , keywords =. doi:10.1051/0004-6361/202244030 , archivePrefix =. 2207.05770 , primaryClass =

    Pith/arXiv arXiv

  41. [44]

    , year = 1972, month = aug, volume =

    On the Infall of Matter Into Clusters of Galaxies and Some Effects on Their Evolution. , year = 1972, month = aug, volume =. doi:10.1086/151605 , adsurl =

    doi:10.1086/151605 1972

  42. [45]

    , keywords =

    The Three Hundred project: backsplash galaxies in simulations of clusters. , keywords =. doi:10.1093/mnras/staa273 , archivePrefix =. 2001.11518 , primaryClass =

    Pith/arXiv arXiv 2001

  43. [46]

    , keywords =

    The Three Hundred project: galaxy groups do not survive cluster infall. , keywords =. doi:10.1093/mnras/stac2809 , archivePrefix =. 2209.13604 , primaryClass =

    Pith/arXiv arXiv

  44. [47]

    , keywords =

    Constraining Cosmological Parameters Using the Splashback Radius of Galaxy Clusters. , keywords =. doi:10.3847/1538-4357/ad5cee , archivePrefix =. 2406.17849 , primaryClass =

    Pith/arXiv arXiv

  45. [48]

    , keywords =

    Reconsidering the dynamical states of galaxy clusters using PCA and UMAP. , keywords =. doi:10.1093/mnras/stae1566 , archivePrefix =. 2406.15555 , primaryClass =

    Pith/arXiv arXiv

  46. [49]

    , keywords =

    YZiCS: Preprocessing of Dark Halos in the Hydrodynamic Zoom-in Simulation of Clusters. , keywords =. doi:10.3847/1538-4357/aadfe2 , archivePrefix =. 1809.02763 , primaryClass =

    Pith/arXiv arXiv

  47. [50]

    The Open Journal of Astrophysics , keywords =

    Finding the boundary: Using galaxy membership to inform galaxy cluster extent through machine learning. The Open Journal of Astrophysics , keywords =. doi:10.33232/001c.159081 , archivePrefix =. 2511.07516 , primaryClass =

    arXiv

  48. [51]

    , keywords =

    A new class of accurate, mesh-free hydrodynamic simulation methods. , keywords =. doi:10.1093/mnras/stv195 , archivePrefix =. 1409.7395 , primaryClass =

    Pith/arXiv arXiv

  49. [52]

    , keywords =

    The Three Hundred Project: Connection between star formation quenching and dynamical evolution in and around simulated galaxy clusters. , keywords =. doi:10.1093/mnras/stac3209 , archivePrefix =. 2211.04485 , primaryClass =

    Pith/arXiv arXiv

  50. [53]

    Radial alignment of satellites towards cluster centres

    Intrinsic alignment in redMaPPer clusters - II. Radial alignment of satellites towards cluster centres. , keywords =. doi:10.1093/mnras/stx2995 , archivePrefix =. 1704.06273 , primaryClass =

    Pith/arXiv arXiv

  51. [54]

    J. D. Comput. Sci. Eng. , year=

  52. [55]

    GASP. IX. Jellyfish galaxies in phase-space: an orbital study of intense ram-pressure stripping in clusters. , keywords =. doi:10.1093/mnras/sty500 , archivePrefix =. 1802.07297 , primaryClass =

    Pith/arXiv arXiv

  53. [56]

    , keywords =

    Environmental Dependence of the Galaxy Merger Rate in a CDM Universe. , keywords =. doi:10.1088/0004-637X/754/1/26 , archivePrefix =. 1205.1588 , primaryClass =

    Pith/arXiv arXiv

  54. [57]

    , keywords =

    The wide-field, multiplexed, spectroscopic facility WEAVE: Survey design, overview, and simulated implementation. , keywords =. doi:10.1093/mnras/stad557 , archivePrefix =. 2212.03981 , primaryClass =

    Pith/arXiv arXiv

  55. [58]

    , keywords =

    Tidal features around simulated groups and cluster galaxies: enhancement and suppression of merger events through environment in LSST-like mock observations. , keywords =. doi:10.1093/mnras/staf1635 , archivePrefix =. 2509.20723 , primaryClass =

    arXiv

  56. [59]

    , keywords =

    The Extended Virgo Cluster Catalog. , keywords =. doi:10.1088/0067-0049/215/2/22 , archivePrefix =. 1409.3283 , primaryClass =

    Pith/arXiv arXiv

  57. [60]

    , keywords =

    MultiDark simulations: the story of dark matter halo concentrations and density profiles. , keywords =. doi:10.1093/mnras/stw248 , archivePrefix =. 1411.4001 , primaryClass =

    Pith/arXiv arXiv

  58. [61]

    , keywords =

    ASKAP reveals the radio tail structure of the Corkscrew Galaxy shaped by its passage through the Abell 3627 cluster. , keywords =. doi:10.1093/mnras/stae1838 , archivePrefix =. 2405.04374 , primaryClass =

    arXiv

  59. [62]

    , keywords =

    Haloes gone MAD: The Halo-Finder Comparison Project. , keywords =. doi:10.1111/j.1365-2966.2011.18858.x , archivePrefix =. 1104.0949 , primaryClass =

    arXiv 2011

  60. [63]

    , keywords =

    Structure finding in cosmological simulations: the state of affairs. , keywords =. doi:10.1093/mnras/stt1403 , archivePrefix =. 1304.0585 , primaryClass =

    Pith/arXiv arXiv

  61. [64]

    , keywords =

    The Three Hundred project: shapes and radial alignment of satellite, infalling, and backsplash galaxies. , keywords =. doi:10.1093/mnras/staa1407 , archivePrefix =. 2005.09896 , primaryClass =

    Pith/arXiv arXiv 2005

  62. [65]

    , keywords =

    AHF: Amiga's Halo Finder. , keywords =. doi:10.1088/0067-0049/182/2/608 , archivePrefix =. 0904.3662 , primaryClass =

    Pith/arXiv arXiv

  63. [66]

    , keywords =

    Cosmic filaments delay quenching inside clusters. , keywords =. doi:10.1093/mnras/stac300 , archivePrefix =. 2110.13419 , primaryClass =

    Pith/arXiv arXiv

  64. [67]

    , keywords =

    The clustering of X-ray AGN at 0.5 < z < 4.5: host galaxies dictate dark matter halo mass. , keywords =. doi:10.1093/mnras/staa815 , archivePrefix =. 2003.10461 , primaryClass =

    Pith/arXiv arXiv 2003

  65. [68]

    , keywords =

    A Comparison of Methods for Determining the Molecular Content of Model Galaxies. , keywords =. doi:10.1088/0004-637X/729/1/36 , archivePrefix =. 1011.4065 , primaryClass =

    Pith/arXiv arXiv

  66. [69]

    , keywords =

    Mapping and characterization of cosmic filaments in galaxy cluster outskirts: strategies and forecasts for observations from simulations. , keywords =. doi:10.1093/mnras/staa1083 , archivePrefix =. 2004.08408 , primaryClass =

    Pith/arXiv arXiv 2004

  67. [70]

    , keywords =

    An inventory of galaxies in cosmic filaments feeding galaxy clusters: galaxy groups, backsplash galaxies, and pristine galaxies. , keywords =. doi:10.1093/mnras/stab3419 , archivePrefix =. 2111.11467 , primaryClass =

    Pith/arXiv arXiv

  68. [71]

    , keywords =

    From voids to filaments: environmental transformations of galaxies in the SDSS. , keywords =. doi:10.1051/0004-6361/201730526 , archivePrefix =. 1703.04338 , primaryClass =

    Pith/arXiv arXiv

  69. [72]

    , keywords =

    The 2dF Galaxy Redshift Survey: the environmental dependence of galaxy star formation rates near clusters. , keywords =. doi:10.1046/j.1365-8711.2002.05558.x , archivePrefix =. astro-ph/0203336 , primaryClass =

    arXiv 2002

  70. [73]

    The origin of H I-deficiency in galaxies on the outskirts of the Virgo cluster. I. How far can galaxies bounce out of clusters?. , keywords =. doi:10.1051/0004-6361:20034155 , archivePrefix =. astro-ph/0310709 , primaryClass =

    Pith/arXiv arXiv

  71. [74]

    Sample selection

    Cosmology and astrophysics from relaxed galaxy clusters - I. Sample selection. , keywords =. doi:10.1093/mnras/stv219 , archivePrefix =. 1502.06020 , primaryClass =

    Pith/arXiv arXiv

  72. [75]

    , keywords =

    Machine learning to identify ICL and BCG in simulated galaxy clusters. , keywords =. doi:10.1093/mnras/stac1558 , archivePrefix =. 2203.03360 , primaryClass =

    Pith/arXiv arXiv

  73. [76]

    , keywords =

    Reconstructing orbits of galaxies in extreme regions (ROGER) III: Galaxy evolution patterns in projected phase space around massive X-ray clusters. , keywords =. doi:10.1093/mnras/stac3746 , archivePrefix =. 2212.09780 , primaryClass =

    Pith/arXiv arXiv

  74. [78]

    2010 , doi=

    Wes McKinney , journal=. 2010 , doi=

    2010

  75. [79]

    Clusters of Galaxies: Probes of Cosmological Structure and Galaxy Evolution , year = 2004, editor =

    Interactions and Mergers of Cluster Galaxies. Clusters of Galaxies: Probes of Cosmological Structure and Galaxy Evolution , year = 2004, editor =

    2004

  76. [80]

    , keywords =

    An X-Ray Method for Detecting Substructure in Galaxy Clusters: Application to Perseus, A2256, Centaurus, Coma, and Sersic 40/6. , keywords =. doi:10.1086/173019 , adsurl =

    doi:10.1086/173019

  77. [81]

    , keywords =

    Galaxy harassment and the evolution of clusters of galaxies. , keywords =. doi:10.1038/379613a0 , archivePrefix =. astro-ph/9510034 , primaryClass =

    Pith/arXiv arXiv

  78. [82]

    , keywords =

    The Splashback Radius as a Physical Halo Boundary and the Growth of Halo Mass. , keywords =. doi:10.1088/0004-637X/810/1/36 , archivePrefix =. 1504.05591 , primaryClass =

    Pith/arXiv arXiv

  79. [83]

    The ubiquity of truncated star-forming discs across the Virgo cluster environment

    A Virgo Environmental Survey Tracing Ionised Gas Emission (VESTIGE): XVI. The ubiquity of truncated star-forming discs across the Virgo cluster environment. , keywords =. doi:10.1051/0004-6361/202449225 , archivePrefix =. 2409.08339 , primaryClass =

    Pith/arXiv arXiv

  80. [84]

    , keywords =

    Deep Galaxy Stellar Mass Functions As a Function of Star Formation Rate in the Virgo Cluster Environment. , keywords =. doi:10.3847/1538-4357/addc6c , archivePrefix =. 2505.13605 , primaryClass =

    Pith/arXiv arXiv

Showing first 80 references.