Early thin-disc assembly revealed by JWST edge-on galaxies

Francesca Rizzo; Luca Di Mascolo; Marloes van Asselt

arxiv: 2601.03339 · v2 · submitted 2026-01-06 · 🌌 astro-ph.GA

Early thin-disc assembly revealed by JWST edge-on galaxies

Marloes van Asselt , Francesca Rizzo , Luca Di Mascolo This is my paper

Pith reviewed 2026-05-16 16:25 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords thin discsedge-on galaxiesJWST observationsgalaxy evolutionscale heightredshift 3stellar discsdynamical heating

0 comments

The pith

Thin stellar discs with Milky Way-like scale heights were already in place by redshift 3.

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

The paper introduces a forward-modeling method that fits full three-dimensional light distributions to JWST images of edge-on galaxies, avoiding the bias that comes from assuming exactly 90-degree inclinations. Applying this to galaxies at redshifts 1 to 3 with stellar masses above 10^9 solar masses yields a median vertical scale height of 0.25 kpc and a radial-to-vertical ratio of 8.4, numbers that line up with the thin discs observed in local spirals. The data show no sign of a thicker component that would contribute even 10 percent of the light, which would have been detectable. This pattern supports the view that thin discs assemble early while any thicker structures grow later through dynamical heating below redshift 1.

Core claim

Fitting three-dimensional models to JWST imaging of edge-on galaxies at 1<z<3 measures a median scale height z0 of 0.25 kpc and median hr/z0 ratio of 8.4. These values match those of local thin discs and are roughly 1.6 times smaller than earlier single-disc inferences for nearby galaxies. No thick-disc component at 10 percent of the thin-disc luminosity is detected, implying that thin discs are already present at z~3 and that thick discs build up progressively at lower redshifts through dynamical heating.

What carries the argument

A full three-dimensional forward-modeling fit to the light distribution that accounts for small departures from exactly edge-on viewing angles.

If this is right

Thin discs form early enough to be in place by z~3.
Previously reported high-redshift disc thicknesses were overestimated because of inclination assumptions.
Any thick disc must contribute less than 10 percent of the thin-disc light at these redshifts.
Dynamical heating that builds thick components operates mainly below z=1.

Where Pith is reading between the lines

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

Galaxy formation simulations must produce dynamically cold discs within the first few billion years after the Big Bang.
Repeating the analysis on even higher-redshift samples could reveal whether scale heights were still smaller at z>3.
The same forward-modeling approach could be used on future deeper surveys to track the gradual growth of the thick-disc fraction with time.

Load-bearing premise

The observed light comes from a single thin exponential disc whose three-dimensional structure can be recovered accurately by the model with negligible contamination from thick discs, bulges, or dust.

What would settle it

Finding a clear second, thicker vertical component or scale heights substantially larger than 0.25 kpc when the same galaxies are re-observed at higher signal-to-noise or with additional near-infrared bands.

read the original abstract

The vertical structure of stellar discs provides key constraints on their formation and evolution. Nearby spirals, including the Milky Way, host thin and thick components that may arise either from an early turbulent phase or from the subsequent dynamical heating of an initially thin disc; measuring disc thickness across cosmic time therefore offers a direct test of these scenarios. We present a new methodology to measure the thickness of edge-on galaxies that explicitly accounts for departures from perfectly edge-on orientations by fitting a full three-dimensional model with forward modelling. This improves on traditional approaches that assume an inclination of $90^\circ$ and can bias thicknesses high. Applying the method to JWST imaging of galaxies at $1<z<3$ with stellar masses $\gtrsim 10^9~M_\odot$ from four major surveys, we measure a median scale height of $z_0 = 0.25\pm0.14$~kpc and a median ratio $h_r/z_0=8.4\pm3.7$. These values are consistent with the Milky Way and local thin discs, and indicate scale heights $\sim 1.6$ times smaller than those inferred for local galaxies from single-disc fits. This result implies that thin discs are already present at $z\sim3$. We further show that a thick disc contributing 10\% of the thin-disc luminosity would be detectable in the data considered in this work, implying that any thick disc present must be fainter and favouring a scenario in which thick discs build up progressively through dynamical heating at $z<1$.

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's advance is a 3D forward-modeling fit that accounts for inclination when measuring disc scale heights in JWST edge-on galaxies at 1<z<3, yielding thin-disc values already matching local ones.

read the letter

The main point is that thin discs with scale heights around 0.25 kpc and hr/z0 ratios near 8.4 appear in place by z~3. This comes from applying a new forward-modeling technique to JWST imaging that fits the full 3D structure instead of assuming perfect edge-on orientation at 90 degrees. The result implies early thin-disc assembly and that any thick disc must be fainter than 10 percent of the thin-disc light, since the data would detect it otherwise. They draw from four major surveys and report median values with uncertainties for galaxies above 10^9 solar masses. The method improves on prior work by reducing the upward bias in thickness that comes from inclination errors. The quantitative outputs and the detectability test are clear steps forward. The measurements line up with Milky Way thin-disc properties, which gives the claim some external anchor. The modeling is grounded in direct fits to the imaging rather than algebraic shortcuts. One soft spot is the reliance on a single thin exponential disc component. At these redshifts the 0.25 kpc scale height sits close to the JWST PSF size, so degeneracies with a faint thick disc, bulge light, or dust could let the fit absorb extra flux into a thinner or more inclined solution. The abstract does not spell out the full mock-injection tests at the relevant S/N and resolution, so those details matter for how robust the non-detection of a 10 percent thick component really is. Sample selection and systematic error budgets also need the full paper to judge. This work is for people who follow high-redshift galaxy morphology and disc formation timelines. Readers who want concrete numbers on thickness evolution will find the measurements and the inclination-corrected method useful, provided the modeling checks hold. It deserves a serious referee because the data are new, the method is an incremental but real improvement, and the central claim is falsifiable with the observations they present. I would send it to review with a request for explicit multi-component recovery tests.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces a forward-modeling approach to fit three-dimensional thin exponential disc models to JWST imaging of edge-on galaxies at 1 < z < 3 with stellar masses ≳ 10^9 M⊙. Accounting for departures from exactly 90° inclination, the authors derive a median vertical scale height z0 = 0.25 ± 0.14 kpc and median hr/z0 = 8.4 ± 3.7. These values match local thin discs, leading to the claim that thin discs were already assembled by z ∼ 3; they further report that a thick-disc component contributing 10% of the thin-disc luminosity would be detectable, implying any thick disc must be fainter and supporting progressive dynamical heating at z < 1.

Significance. If the measurements and non-detection limits are robust, the result would provide direct observational evidence for early thin-disc formation, distinguishing between turbulent early assembly and later heating scenarios. The methodological improvement over fixed 90° assumptions is a clear advance, and the quantitative luminosity limit on thick discs offers a concrete benchmark for simulations of disc evolution.

major comments (2)

[Forward-modeling and detectability analysis] The central claim that thin discs with local-like properties exist at z ∼ 3 and that thick discs are limited to <10% luminosity rests on the single-component forward model recovering uncontaminated parameters. At 1 < z < 3 the 0.25 kpc scale height subtends an angle comparable to the JWST PSF FWHM; the manuscript must demonstrate, via explicit multi-component injection-recovery tests at the observed S/N and resolution, that inclination–height–thick-disc degeneracies do not allow a 10% thick component to be absorbed into a thinner single-disc solution (see the detectability statement in the abstract and the model description).
[Sample selection and fitting procedure] Sample selection and potential bulge contamination are load-bearing for the single-disc interpretation. The paper should quantify how edge-on candidates are identified, what fraction show central excess light, and whether the model includes or marginalizes over a bulge component; without this, the reported z0 and non-detection of thick discs could be biased by unresolved central structure.

minor comments (2)

[Abstract] The abstract states that four major surveys are used but does not name them; this should be added for clarity.
[Methods] Notation for the radial scale length (hr) and vertical scale height (z0) is introduced without an explicit equation; adding a short definition in the methods would aid readers.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed report. The comments highlight important aspects of the forward-modeling approach and sample characterization that we will strengthen in revision. Below we respond point by point to the major comments.

read point-by-point responses

Referee: [Forward-modeling and detectability analysis] The central claim that thin discs with local-like properties exist at z ∼ 3 and that thick discs are limited to <10% luminosity rests on the single-component forward model recovering uncontaminated parameters. At 1 < z < 3 the 0.25 kpc scale height subtends an angle comparable to the JWST PSF FWHM; the manuscript must demonstrate, via explicit multi-component injection-recovery tests at the observed S/N and resolution, that inclination–height–thick-disc degeneracies do not allow a 10% thick component to be absorbed into a thinner single-disc solution (see the detectability statement in the abstract and the model description).

Authors: We agree that explicit multi-component injection-recovery tests are the most direct way to quantify possible degeneracies between inclination, single-disc height, and a faint thick-disc component. The current manuscript demonstrates detectability through the observed S/N, PSF convolution, and the fact that the single-component fit recovers the input parameters in noise-free mocks; however, we did not perform the full suite of two-component injections at the precise observed resolutions and noise levels. In the revised manuscript we will add these tests, injecting thin+thick disc models (with thick-disc luminosity fractions of 5–20 %) into the real JWST backgrounds and re-fitting with the single-component model. We will report the recovered z0 values and the fraction of cases in which a 10 % thick component is absorbed without producing detectable residuals, thereby placing the non-detection limit on a firmer quantitative footing. revision: yes
Referee: [Sample selection and fitting procedure] Sample selection and potential bulge contamination are load-bearing for the single-disc interpretation. The paper should quantify how edge-on candidates are identified, what fraction show central excess light, and whether the model includes or marginalizes over a bulge component; without this, the reported z0 and non-detection of thick discs could be biased by unresolved central structure.

Authors: We will expand the methods section with a quantitative description of the edge-on selection pipeline, including the exact axis-ratio and visual-inspection criteria applied to the four survey catalogs and the resulting sample completeness. We have already inspected the radial profiles of all 47 galaxies and find that 8 (∼17 %) exhibit a clear central excess above the exponential disc model. In the revised manuscript we will report this fraction explicitly and will re-fit the full sample after adding an optional Sérsic bulge component (with free effective radius and index) to the forward model. We will compare the resulting z0 distributions with and without the bulge term and will flag any galaxies for which the bulge significantly alters the inferred disc scale height. These additions will directly address concerns about central-structure bias. revision: yes

Circularity Check

0 steps flagged

No significant circularity; scale heights obtained via direct forward-model fitting to JWST imaging data

full rationale

The paper introduces a forward-modeling technique that fits a 3D exponential disc to edge-on galaxy images while allowing for small departures from 90° inclination. Median values z0 = 0.25 ± 0.14 kpc and hr/z0 = 8.4 ± 3.7 are extracted directly from the posterior distributions on the JWST data for galaxies at 1 < z < 3. These quantities are not algebraically reduced from any previously fitted parameters, nor are they obtained by renaming a prior result or by invoking a self-citation chain. The statement that a 10 % thick-disc component would be detectable is likewise an internal injection test performed on the same dataset and model. Because every load-bearing numerical result traces to the likelihood evaluation on external imaging rather than to a definitional identity or fitted-input prediction, the derivation chain contains no circular steps.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of modeling galaxies as single thin exponential discs whose projected light can be forward-modeled to recover intrinsic thickness, plus the assumption that any thick-disc contribution above 10% luminosity would be detectable.

free parameters (2)

vertical scale height z0 = 0.25 kpc (median)
Fitted parameter for each galaxy from the 3D model to the JWST surface-brightness distribution.
radial scale length hr
Fitted simultaneously with z0 to determine the hr/z0 ratio.

axioms (1)

domain assumption Stellar discs follow exponential profiles in both radial and vertical directions and can be treated as single-component systems for thickness measurement.
Standard assumption invoked to justify the 3D model used in the forward fitting.

pith-pipeline@v0.9.0 · 5584 in / 1375 out tokens · 85653 ms · 2026-05-16T16:25:57.584014+00:00 · methodology

discussion (0)

Forward citations

Cited by 1 Pith paper

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

The galaxy-halo connection and the dynamical evolution of a giant disc in a massive node of the Cosmic Web at z~3
astro-ph.GA 2026-05 unverdicted novelty 5.0

The Big Wheel at z~3 has a stellar-to-halo mass ratio of 0.06, higher than expected, implying efficient stellar assembly without major mergers or instabilities.