arxiv: 2605.08537 · v1 · submitted 2026-05-08 · 🌌 astro-ph.GA

Recognition: no theorem link

Specific Star Formation Rate Enhancement across the Galaxy Merger Sequence: Insights from Citizen Science Classifications

Jacob Lee , Alexandra Le Reste , Claudia Scarlata , Kameswara Mantha Bharadwaj

Authors on Pith no claims yet

Pith reviewed 2026-05-12 00:48 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords galaxy mergersspecific star formation ratecitizen sciencemerger sequenceSDSSstar formation enhancementZooniverse classifications

0 comments

The pith

Citizen science shows specific star formation rates rise weakly with galaxy merger stage.

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

The paper examines specific star formation rates in galaxies at different visual merger stages using volunteer classifications from a Zooniverse project. It identifies a weak but statistically significant positive correlation in a sample of over 3600 merger systems drawn from SDSS data at low redshift. The result indicates that star formation activity tends to increase as mergers advance from early interactions to final coalescence, while also showing that citizen science can serve as a practical method for tracing these evolutionary stages.

Core claim

We find a weak but statistically significant positive correlation between log(sSFR) and visual merger stage (r = 0.161, p = 7.23 × 10^{-23}), with a best-fit relation log(sSFR)=(0.148±0.015) S_Merg-(1.865±0.038). The large RMS scatter of 0.661 dex reflects the wide range of physical timescales spanned by visual stages, and the trend corroborates prior evidence that star formation is enhanced as mergers progress.

What carries the argument

Visual merger stage S_Merg assigned by citizen scientists to pre-selected SDSS galaxy systems, used as the independent variable in a linear fit to log(sSFR).

If this is right

Specific star formation rate increases by roughly 0.15 dex per step along the visual merger sequence.
Enhancement occurs gradually across the full merger timeline rather than in one abrupt event.
Citizen science classifications provide a scalable complement to automated or pair-based merger studies.
The observed scatter implies that each visual stage corresponds to a broad window of dynamical evolution.

Where Pith is reading between the lines

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

If the relation holds, models of merger-driven star formation must produce gradual rather than instantaneous bursts over hundreds of Myr.
The slope could be used to assign approximate merger ages to larger photometric samples where spectroscopic confirmation is unavailable.
Extending the same citizen-science approach to higher-redshift surveys would test whether the enhancement amplitude grows with cosmic time.

Load-bearing premise

Visual merger stages assigned by citizen scientists accurately map onto physical merger timescales without systematic distortion from selection effects or classification biases.

What would settle it

No positive correlation appears when the same galaxies are reclassified using automated morphological metrics or when the relation is tested in hydrodynamical simulations with matching mass and redshift cuts.

Figures

Figures reproduced from arXiv: 2605.08537 by Alexandra Le Reste, Claudia Scarlata, Jacob Lee, Kameswara Mantha Bharadwaj.

**Figure 1.** Figure 1: Log(sSFR) versus SMerg for Cosmic Disco galaxies with Pmerg ≥ 0.5. Merger stages trace galaxies from pre-interaction (SMerg = 1) to post-coalescence (SMerg = 4). We find a weak but statistically significant correlation, with linear fit shown in red and 1σ uncertainty shown as the shaded region. Median log(sSFR) for Cosmic Disco non-mergers (Pmerg < 0.5; blue dashed) and SDSS control galaxies (green dashed)… view at source ↗

read the original abstract

We present an analysis of specific star formation rates (sSFR) across the galaxy merger sequence using visual classifications from the Zooniverse citizen science project "Cosmic Disco: Characterizing Galaxy Collisions". Our sample comprises 4884 galaxy systems pre-selected as merger candidates from SDSS DR17 ($0.01 < z < 0.05$, $M_* > 10^{8.5}M_\odot$) using Zoobot, of which 3690 were classified as mergers spanning pre-interaction through post-coalescence stages by citizen scientist volunteers. We find a weak but statistically significant positive correlation between $\log(\mathrm{sSFR})$ and visual merger stage ($r = 0.161$, $p = 7.23 \times 10^{-23}$), with a best-fit relation $\log\left(sSFR\right)=(0.148\pm0.015)\, S_{\rm Merg}-(1.865\pm0.038)$. The large RMS scatter (0.661 dex) reflects visual merger stages capturing wide merger timescales, and our results corroborate previous findings of increasing SFR enhancement with merger progression. This work shows that citizen science is a viable complement to automated and pair-based approaches to evaluate timescales for galaxies across the merger sequence.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Citizen science labels recover the usual weak sSFR rise along the merger sequence on this SDSS sample, but the correlation stays modest and the work adds mainly a new data point rather than new insight.

read the letter

The paper recovers a known weak trend: sSFR rises slightly with visual merger stage in this SDSS sample classified by Zooniverse volunteers. The correlation coefficient is 0.161 with a slope around 0.15, but scatter is high at 0.66 dex. The main result is the best-fit relation log(sSFR) = (0.148 ± 0.015) S_Merg - (1.865 ± 0.038) on 3690 classified mergers from a 4884-galaxy parent sample pre-selected by Zoobot. It lines up with earlier pair-based and simulation work on merger-driven star formation enhancement. What is new is the specific slope, intercept, and scatter values obtained from volunteer stages on this particular SDSS DR17 slice. The work shows that citizen science can scale to thousands of galaxies and still detect the same statistical pattern seen in smaller expert samples. That demonstration is useful on its own. The soft spots sit in the numbers themselves. The correlation is statistically significant but weak enough that it does not tightly constrain anything, and the 0.66 dex RMS scatter means the stages capture very broad timescales. The abstract gives no per-stage error bars, no completeness corrections, and no quantitative tests of whether the ordered visual stages map monotonically to physical time or whether Zoobot pre-selection introduces sSFR-dependent biases. If those checks exist in the full text they are not foregrounded, so the central claim rests partly on the assumption that volunteer labels track merger progression cleanly. This paper is for researchers already working on galaxy mergers or citizen-science applications who want one more empirical comparison point. A reader looking for strong new constraints or methodological advances will not find them here. It deserves a serious referee because the sample and measurement approach are described clearly enough for others to reproduce or extend, and a referee can ask for the missing validation details. Send it out for review rather than desk reject.

Referee Report

3 major / 2 minor

Summary. The manuscript analyzes specific star formation rates (sSFR) across the galaxy merger sequence in a sample of 4884 SDSS DR17 galaxies (0.01 < z < 0.05, M_* > 10^{8.5} M_⊙) pre-selected as merger candidates via Zoobot. Of these, 3690 systems receive visual merger-stage classifications (pre-interaction through post-coalescence) from citizen scientists in the Zooniverse project 'Cosmic Disco'. The central result is a weak but statistically significant positive correlation between log(sSFR) and visual merger stage (r = 0.161, p = 7.23 × 10^{-23}), with best-fit relation log(sSFR) = (0.148 ± 0.015) S_Merg − (1.865 ± 0.038) and RMS scatter 0.661 dex. The work concludes that citizen-science classifications provide a viable complement to automated and pair-based methods for assessing sSFR enhancement along the merger sequence.

Significance. If the visual stages reliably trace physical merger progression without dominant selection biases, the result supplies an independent empirical corroboration of prior findings that sSFR increases with merger advancement, now derived from a large volunteer-classified sample. The purely empirical approach (no circularity between stage assignment and sSFR measurement) and the demonstration that citizen science can scale merger-sequence studies are positive contributions. However, the modest correlation coefficient and large scatter imply that any quantitative timescale inferences remain limited even if the trend holds.

major comments (3)

[Abstract / Results] Abstract and results section: the reported Pearson r = 0.161 and linear fit are presented without per-stage error bars, completeness corrections, or explicit propagation of sSFR photometric uncertainties into the correlation statistic or slope uncertainty. Given the acknowledged 0.661 dex RMS scatter, this omission prevents quantitative assessment of whether the trend remains significant after realistic error treatment and is therefore load-bearing for the central claim of statistically significant enhancement.
[Methods / Classification] Classification and sample-selection sections: no quantitative validation is provided for the mapping of citizen-science visual stages onto physical merger timescales (e.g., agreement metrics with expert labels, simulation-based stage recovery fractions, or monotonicity tests). This assumption is load-bearing because the headline correlation interpretation requires that the ordered stages correspond to increasing merger time rather than classification subjectivity.
[Sample Selection] Sample-selection section: potential sSFR-dependent biases introduced by the Zoobot pre-selection of the 4884-galaxy parent sample are not tested (e.g., via comparison of sSFR distributions against a non-merger control sample or retention-rate checks across stages). Such biases could artificially produce or inflate the reported positive correlation and must be quantified before the result can be interpreted as intrinsic sSFR enhancement.

minor comments (2)

[Abstract] The abstract states that results 'corroborate previous findings' but does not cite the specific prior works being referenced; adding these citations would improve traceability.
[Abstract] Notation for the merger stage variable (S_Merg) is introduced in the abstract without an explicit definition or reference to the classification scheme table; a brief parenthetical definition on first use would aid readability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed comments, which have identified important areas for clarification and strengthening of our analysis. We address each major comment below and indicate the revisions we plan to implement.

read point-by-point responses

Referee: [Abstract / Results] Abstract and results section: the reported Pearson r = 0.161 and linear fit are presented without per-stage error bars, completeness corrections, or explicit propagation of sSFR photometric uncertainties into the correlation statistic or slope uncertainty. Given the acknowledged 0.661 dex RMS scatter, this omission prevents quantitative assessment of whether the trend remains significant after realistic error treatment and is therefore load-bearing for the central claim of statistically significant enhancement.

Authors: We agree that per-stage error bars and explicit discussion of uncertainties would improve the manuscript. In the revision, we will add error bars (standard error of the mean) to the binned log(sSFR) values per visual stage in the results section and a new figure. Photometric uncertainties on sSFR are typically 0.1-0.2 dex and are subsumed in the 0.661 dex RMS scatter (which includes physical variation across broad timescales); the reported slope uncertainty already incorporates the observed scatter, and the p-value remains robust under bootstrap resampling. We will add a brief bootstrap test and note on this. For completeness corrections, our focus is the trend within the pre-selected sample rather than absolute merger rates, but we will clarify this limitation and discuss potential effects without claiming completeness. These updates will be made to the results and abstract. revision: yes
Referee: [Methods / Classification] Classification and sample-selection sections: no quantitative validation is provided for the mapping of citizen-science visual stages onto physical merger timescales (e.g., agreement metrics with expert labels, simulation-based stage recovery fractions, or monotonicity tests). This assumption is load-bearing because the headline correlation interpretation requires that the ordered stages correspond to increasing merger time rather than classification subjectivity.

Authors: We acknowledge that additional quantitative validation would strengthen the interpretation. The stages follow standard morphological definitions (pre-interaction pairs, bridges/tails, post-coalescence) established in the literature and shown by simulations to progress temporally. In the revised methods section, we will add a paragraph citing prior works on visual stage timescales, include available Zooniverse inter-volunteer agreement metrics, and perform a simple monotonicity check on the binned data. A full simulation-based recovery analysis (requiring new mock images and expert reclassification) is beyond the scope of this empirical study, but we will emphasize that the ordering is not arbitrary and aligns with established merger-sequence frameworks used in previous studies. revision: partial
Referee: [Sample Selection] Sample-selection section: potential sSFR-dependent biases introduced by the Zoobot pre-selection of the 4884-galaxy parent sample are not tested (e.g., via comparison of sSFR distributions against a non-merger control sample or retention-rate checks across stages). Such biases could artificially produce or inflate the reported positive correlation and must be quantified before the result can be interpreted as intrinsic sSFR enhancement.

Authors: We agree this test is necessary to rule out selection effects. In the revised manuscript, we will add a control sample of SDSS galaxies matched in redshift and stellar mass but not flagged as mergers by Zoobot, and compare their sSFR distributions to our sample. We will also compute the fraction of galaxies assigned to each stage as a function of sSFR bins to check for retention-rate biases. These results will be presented in the methods or results section with a new table or panel, and discussed to confirm that the observed trend is not driven by the pre-selection. If biases are minimal, this supports interpreting the correlation as intrinsic enhancement. revision: yes

Circularity Check

0 steps flagged

No significant circularity: empirical correlation from independent data

full rationale

The paper reports a direct statistical correlation (r = 0.161) and linear fit between log(sSFR) from photometry and visual merger stages assigned by citizen scientists on a pre-selected sample. No equation or derivation reduces the reported slope, intercept, or significance to a fitted parameter by construction. Stages and sSFR are measured independently; the result does not invoke self-citations for uniqueness, ansatz smuggling, or renaming. The central claim remains a straightforward empirical measurement without tautological reduction.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claim rests on two fitted parameters from ordinary least-squares regression and on the domain assumption that visual stages track physical merger progression. No new entities are postulated.

free parameters (2)

slope = 0.148
Linear-regression coefficient relating log(sSFR) to merger stage; value 0.148 reported with uncertainty.
intercept = -1.865
Linear-regression intercept; value -1.865 reported with uncertainty.

axioms (1)

domain assumption Volunteer visual classifications reliably order galaxies along the physical merger sequence
Invoked when interpreting the positive correlation as evidence of merger-driven sSFR enhancement.

pith-pipeline@v0.9.0 · 5540 in / 1358 out tokens · 45979 ms · 2026-05-12T00:48:51.659050+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[1]

Journal of Open Source Software , year = 2023, volume =

Zoobot: Adaptable Deep Learning Models for Galaxy Morphology. Journal of Open Source Software , year = 2023, volume =. doi:10.21105/joss.05312 , adsurl =

work page doi:10.21105/joss.05312 2023
[2]

The orbital extent of enhanced star formation in interacting galaxies

Galaxy pairs in the Sloan Digital Sky Survey - VI. The orbital extent of enhanced star formation in interacting galaxies. , keywords =. doi:10.1093/mnrasl/slt058 , archivePrefix =. 1305.1595 , primaryClass =

work page doi:10.1093/mnrasl/slt058
[3]

, keywords =

SDSS-IV MaNGA: Spatial Evolution of Star Formation Triggered by Galaxy Interactions. , keywords =. doi:10.3847/1538-4357/ab311c , archivePrefix =. 1907.04491 , primaryClass =

work page doi:10.3847/1538-4357/ab311c 1907
[4]

Comprehensiveanalyticformulaeforstellarevolutionasafunctionofmassandmetallicity,

Hubble Space Telescope imaging of the CFRS and LDSS redshift surveys - IV. Influence of mergers in the evolution of faint field galaxies from z -0.5ex 1. , keywords =. doi:10.1046/j.1365-8711.2000.03083.x , archivePrefix =. astro-ph/9909211 , primaryClass =

work page doi:10.1046/j.1365-8711.2000.03083.x 2000
[5]

, keywords =

The Seventeenth Data Release of the Sloan Digital Sky Surveys: Complete Release of MaNGA, MaStar, and APOGEE-2 Data. , keywords =. doi:10.3847/1538-4365/ac4414 , archivePrefix =. 2112.02026 , primaryClass =

work page doi:10.3847/1538-4365/ac4414
[6]

, publisher =

Conselice , Christopher J. , publisher =. Galaxy Mergers and Interactions through Cosmic Time , booktitle =. doi:https://doi.org/10.1002/9781119818007.ch5 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1002/9781119818007.ch5 , year =

work page doi:10.1002/9781119818007.ch5
[7]

The observational properties of post-merger galaxies

Galaxy pairs in the Sloan Digital Sky Survey - VIII. The observational properties of post-merger galaxies. , keywords =. doi:10.1093/mnras/stt1562 , archivePrefix =. 1308.3707 , primaryClass =

work page doi:10.1093/mnras/stt1562
[8]

J., Lang, D., et al

Overview of the DESI Legacy Imaging Surveys. , keywords =. doi:10.3847/1538-3881/ab089d , archivePrefix =. 1804.08657 , primaryClass =

work page doi:10.3847/1538-3881/ab089d
[9]

2010, MNRAS, 401, 1670, doi: 10.1111/j.1365-2966.2009.15794.x

Galaxy Zoo: the fraction of merging galaxies in the SDSS and their morphologies. , keywords =. doi:10.1111/j.1365-2966.2009.15686.x , archivePrefix =. 0903.4937 , primaryClass =

work page doi:10.1111/j.1365-2966.2009.15686.x 2009
[10]

, year = 2000, volume = 533, pages =

The Dust Content and Opacity of Actively Star-forming Galaxies. , year = 2000, volume = 533, pages =

work page 2000
[11]

, year = 2003, volume = 346, pages =

The host galaxies of active galactic nuclei. , year = 2003, volume = 346, pages =

work page 2003
[12]

, year = 2001, volume = 322, pages =

On the variation of the initial mass function. , year = 2001, volume = 322, pages =

work page 2001
[13]

, year = 2011, volume = 737, pages = 67, doi =

Calibrating Extinction-free Star Formation Rate Diagnostics with 33 GHz Free-free Emission in NGC 6946. , year = 2011, volume = 737, pages = 67, doi =

work page 2011
[14]

Comparingthepropertiesoflocalglobularclustersystems:implicationsfortheformationoftheGalactic halo,

The physical properties of star-forming galaxies in the low-redshift Universe. , keywords =. doi:10.1111/j.1365-2966.2004.07881.x , archivePrefix =. astro-ph/0311060 , primaryClass =

work page doi:10.1111/j.1365-2966.2004.07881.x 2004