Hessian-based photometric substructure as an evolutionary tracer of OB cluster candidates in M31

Chao-Wei Tsai; Jingwen Wu; Yuan Liang

arxiv: 2604.23118 · v1 · submitted 2026-04-25 · 🌌 astro-ph.GA

Hessian-based photometric substructure as an evolutionary tracer of OB cluster candidates in M31

Yuan Liang , Chao-Wei Tsai , Jingwen Wu This is my paper

Pith reviewed 2026-05-08 07:37 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords OB clustersM31Hessian matrixphotometric substructurecluster evolutionHST photometryAndromedaage tracers

0 comments

The pith

Hessian-based photometric substructure traces the age of OB clusters in M31

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

The paper introduces a dimensionless metric called the trace coefficient of variation, or CV_tr, calculated from the Hessian matrix of Hubble Space Telescope images in four bands to quantify the internal photometric substructure of OB cluster candidates in M31. By matching 247 such candidates to clusters with independent age estimates from color-magnitude diagrams, the authors identify statistically significant anti-correlations between CV_tr and age specifically in the ultraviolet and blue bands. These trends indicate that the light distribution within clusters becomes progressively smoother as they evolve from approximately 10 to 300 million years old. Robustness is confirmed through bootstrap resampling and forward modeling of synthetic clusters that reproduce the monotonic relation. If correct, this means that second-order photometric properties can provide evolutionary information even for partially resolved clusters where full CMD analysis may not be feasible.

Core claim

Using Hubble Space Telescope images from the PHAT and PHAST surveys, the authors build a catalogue of 747 OB cluster candidates and introduce CV_tr derived from the Hessian matrix across four bands to measure internal photometric substructure. Cross-matching with 247 clusters that have CMD-based ages reveals statistically significant anti-correlations between CV_tr and age in UV and blue bands, with bootstrap resampling confirming the trends and synthetic cluster forward modeling recovering a monotonic CV_tr-age relation under simplified assumptions. The work concludes that second-order photometric structure contains measurable evolutionary information in the CMD-calibrated regime of roughly

What carries the argument

The trace coefficient of variation (CV_tr), a dimensionless metric from the Hessian matrix of the surface brightness distribution in four HST bands, which quantifies the degree of internal photometric substructure in partially resolved clusters.

Load-bearing premise

That the CV_tr metric primarily traces intrinsic evolutionary smoothing of the stellar light distribution rather than being dominated by observational effects such as varying resolution, dust extinction, or projection in the partially resolved clusters.

What would settle it

A larger sample of M31 clusters with precise CMD ages in which the anti-correlation between CV_tr and age disappears after correcting for dust, distance, and resolution variations across the sample.

Figures

Figures reproduced from arXiv: 2604.23118 by Chao-Wei Tsai, Jingwen Wu, Yuan Liang.

**Figure 3.** Figure 3: Spatial variation of the characteristic sizes of OBC candidates in the F275W band. (a) Median half-light radius (𝑅eff; blue circles, dashed line) and full-light radius (𝑅full; orange triangles, dot-dashed line) as a function of position angle, measured counterclockwise from the projected northern major axis of M31. The vertical dashed line at 𝜃 = 90◦ separates the northern and southern regions. (b) The sam… view at source ↗

**Figure 1.** Figure 1: Example of Hessian-based curvature analysis applied to a star cluster in two photometric bands. Panels (a) and (b) show the surface brightness map and corresponding curvature (Hessian trace) map in the F275W band, respectively, while panels (c) and (d) show the same quantities in the F814W band. In the curvature maps, blue regions indicate locally convex intensity distributions, while red regions indicate… view at source ↗

**Figure 2.** Figure 2: Spatial distribution of candidate OBCs in M31, overlaid on the dust emission map from Draine et al. (2013). Cyan dots represent the OBC candidates from Liang et al. (2025), and red dots are newly identified candidates from this work. The dashed box outlines the combined PHAT/PHAST survey footprint. The boundary between the PHAT (north) and PHAST (south) survey regions is marked by the central dashed line… view at source ↗

**Figure 4.** Figure 4: Evolution of the Hessian-based structure index 𝐶𝑉tr for OB cluster candidates in M31. (a): Median 𝐶𝑉tr as a function of CMD age for 247 clusters. The horizontal axis shows stellar age on a logarithmic scale, spanning ∼ 10 Myr to 300 Myr, with a bin width of 0.2 dex. Coloured symbols and line styles denote the four photometric bands: F275W (purple solid circles), F336W (blue dashed triangles), F475W (green … view at source ↗

**Figure 5.** Figure 5: Evolution of the Hessian-based structure index 𝐶𝑉tr derived from synthetic cluster simulations. (a): Median 𝐶𝑉tr as a function of input cluster age for 38 independent mock realizations. Age is shown on a logarithmic scale, spanning 106.5 yr to 109.2 yr with a bin width of 0.5 dex. Symbols represent the median values across realizations, and the vertical bars indicate the 16–84 percentile range. The annotat… view at source ↗

read the original abstract

Using \textit{Hubble Space Telescope} images from the PHAT and PHAST surveys, we construct an updated catalogue of 747 OB cluster (OBC) candidates. We introduce a dimensionless structural metric, the trace coefficient of variation ($CV_{\rm tr}$), derived from the Hessian matrix in four \textit{HST} bands, to quantify the internal photometric substructure of partially resolved OBC candidates. Cross-matching with the subset of M31 clusters that have independent colour--magnitude diagram (CMD) age estimates yields 247 objects in common. We find statistically significant anti-correlations between $CV_{\rm tr}$ and age in the UV and blue bands, suggesting a progressive smoothing of the light distribution as clusters evolve. Bootstrap resampling confirms the robustness of these trends. Forward modelling of synthetic clusters analysed with the same pipeline recovers a monotonic $CV_{\rm tr}$--age relation under simplified but physically motivated assumptions. These results show that second-order photometric structure contains measurable evolutionary information within the CMD-calibrated regime ($\sim10$--300~Myr).

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

The paper introduces a new Hessian-derived CV_tr metric that anti-correlates with CMD ages for M31 clusters, but the forward modeling leaves open whether observational effects like dust and crowding drive the trend instead of evolution.

read the letter

The core contribution here is the CV_tr metric, a dimensionless quantity pulled from the trace of the Hessian matrix across four HST bands on partially resolved OB cluster candidates. They cross-match 247 objects against existing CMD ages and report statistically significant anti-correlations in the UV and blue, with bootstrap support. Synthetic clusters run through the same pipeline recover a monotonic trend under their assumptions. That is the new piece: a photometric substructure measure tied directly to age in the 10-300 Myr range without needing full CMD fitting for every target. The catalogue update to 747 candidates from PHAT and PHAST data is also straightforward and useful for anyone working in M31. The forward modeling attempt is the right move to test whether the metric captures progressive light smoothing. It shows the idea is at least plausible on paper. The soft spot is exactly where the stress test flags it. The modeling relies on simplified assumptions that do not appear to include spatially varying extinction, position-dependent PSF variations, or differential crowding across the sample. M31 has plenty of dust and the clusters sit in different local environments, so any unaccounted gradient in those factors could produce an apparent smoothing trend without evolutionary content. The abstract does not spell out how those confounders were controlled or tested, which leaves the central claim vulnerable. This is for people who study young clusters in nearby galaxies and need faster age indicators than full CMDs. A reader focused on photometric structure or survey-scale methods would find the metric worth examining, even if the interpretation requires more checks. It deserves peer review because the data volume and the new metric are solid enough to warrant referee time, though the authors should expect pointed questions on whether the observed relation survives realistic observational modeling.

Referee Report

1 major / 3 minor

Summary. The manuscript constructs an updated catalogue of 747 OB cluster candidates in M31 from HST PHAT and PHAST imaging. It defines a new dimensionless metric, the trace coefficient of variation (CV_tr), computed from the Hessian matrix of the photometric distribution in four HST bands to quantify internal substructure. Cross-matching yields 247 objects with independent CMD ages; statistically significant anti-correlations between CV_tr and age are reported in the UV and blue bands. Bootstrap resampling and forward modelling of synthetic clusters under simplified assumptions recover a monotonic CV_tr-age trend, leading to the conclusion that second-order photometric structure encodes measurable evolutionary information within the ~10-300 Myr CMD-calibrated regime.

Significance. If the central claim is robust, the work supplies a practical photometric tracer of cluster evolution that does not require full CMD fitting, potentially applicable to larger samples or other galaxies. The cross-match to independent ages, bootstrap validation, and forward modelling constitute concrete strengths; the metric appears free of ad-hoc parameters and is tested against a falsifiable synthetic prediction.

major comments (1)

[Abstract] Abstract (forward modelling paragraph): the recovery of a monotonic CV_tr-age relation is demonstrated only under 'simplified but physically motivated assumptions.' These assumptions are not shown to incorporate spatially varying dust extinction or position-dependent crowding, both of which are known to vary across the M31 disk and across the 247-object sample. Because CV_tr is computed directly on raw photometry, any unmodelled observational gradient could induce the reported anti-correlation without evolutionary content. This assumption is load-bearing for the claim that CV_tr primarily traces progressive smoothing.

minor comments (3)

The manuscript would benefit from explicit tabulation of the 747 candidates and the 247 cross-matched subset, including per-object CV_tr values, CMD ages, uncertainties, and local environment parameters.
Provide the precise definition of CV_tr (including how the Hessian trace is normalised) and the exact pipeline steps for its computation on the HST images.
Expand the bootstrap and forward-modelling sections to include the full set of observational confounders tested (or not tested) and the quantitative impact on the recovered slope.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive review and for recognizing the strengths of the cross-matched sample, bootstrap tests, and forward modelling. We address the single major comment below and will incorporate clarifications in the revised manuscript.

read point-by-point responses

Referee: [Abstract] Abstract (forward modelling paragraph): the recovery of a monotonic CV_tr-age relation is demonstrated only under 'simplified but physically motivated assumptions.' These assumptions are not shown to incorporate spatially varying dust extinction or position-dependent crowding, both of which are known to vary across the M31 disk and across the 247-object sample. Because CV_tr is computed directly on raw photometry, any unmodelled observational gradient could induce the reported anti-correlation without evolutionary content. This assumption is load-bearing for the claim that CV_tr primarily traces progressive smoothing.

Authors: We agree that the forward modelling (Section 4.3) employs simplified assumptions that omit spatially varying dust extinction and position-dependent crowding. The text describes these as 'simplified but physically motivated' to isolate the effect of dynamical and stellar evolution on the light distribution. The primary evidence for the CV_tr–age anti-correlation is the direct measurement on the observed 247-object sample (Section 3.2), which is independent of the modelling. The synthetic test is presented only as a proof-of-concept that monotonic smoothing can produce the observed trend direction under basic conditions. We acknowledge that unmodelled gradients could contribute to the signal and will revise the abstract to qualify the forward-modelling result more explicitly. In the main text we will add a dedicated limitations paragraph discussing the possible impact of dust and crowding on raw-photometry metrics and note that more complete simulations including these effects lie beyond the present scope. These changes constitute a partial revision that directly responds to the concern while preserving the distinction between data-driven and modelling results. revision: partial

Circularity Check

0 steps flagged

No significant circularity; empirical correlation with independent ages

full rationale

The paper computes the dimensionless CV_tr metric directly from the Hessian matrix of HST photometry in four bands for 747 OBC candidates. It then cross-matches 247 objects to an external CMD-based age catalogue and reports anti-correlations. Forward modelling of synthetic clusters is performed with the same pipeline under explicitly stated simplified assumptions to show that a monotonic CV_tr-age trend can be recovered. No equation defines CV_tr in terms of age or fits it to the target data; the central claim rests on the observed statistical relation to an independent age estimator rather than on any self-referential reduction or load-bearing self-citation. The derivation chain is therefore self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; full paper details on assumptions unavailable.

axioms (1)

domain assumption Independent CMD age estimates provide a reliable ground truth for calibration.
Cross-matching with 247 objects relies on this to establish the CV_tr-age relation.

pith-pipeline@v0.9.0 · 5487 in / 1162 out tokens · 40276 ms · 2026-05-08T07:37:41.590678+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

1 extracted references · 1 canonical work pages

[1]

doi:10.5281/zenodo.10967176

Adamo A., et al., 2017, The Astrophysical Journal, 841, 131 Adamo A., et al., 2020, Space Science Reviews, 216, 69 Allen L., Megeath S. T., Gutermuth R., Myers P. C., Wolk S., Adams F. C., Muzerolle J., 2007, Protostars and planets V, 361 Barmby P., et al., 2009, The Astronomical Journal, 138, 1667 Beerman L. C., et al., 2012, The Astrophysical Journal, 7...

work page doi:10.5281/zenodo.10967176 2017

[1] [1]

doi:10.5281/zenodo.10967176

Adamo A., et al., 2017, The Astrophysical Journal, 841, 131 Adamo A., et al., 2020, Space Science Reviews, 216, 69 Allen L., Megeath S. T., Gutermuth R., Myers P. C., Wolk S., Adams F. C., Muzerolle J., 2007, Protostars and planets V, 361 Barmby P., et al., 2009, The Astronomical Journal, 138, 1667 Beerman L. C., et al., 2012, The Astrophysical Journal, 7...

work page doi:10.5281/zenodo.10967176 2017