Pith. Sign in

REVIEW

Tilting at the Turnover: Modeling the Faint-End of the UV Luminosity Function Behind Abell s1063 with JWST

PaperPDF from arXiv

data files

ReviewThe objections and verdict, as JSON Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Caio Moreira Goolsby, Christopher J. Conselice, Duncan Austin, Tom Harvey, Jose Diego, Nathan Adams, Julien Marabotto, Jordan D'Silva
show 1 more author
Qiong Li

Astrophysics of Galaxies · Submitted 1 Jul 2026 · 178 references

Pith's verdict: The paper supplies new lower limits on faint-galaxy contributions to reionization at z=6 via a gradual quadratic turnover model on lensed JWST data, but the numbers rest on the Abell S1063 magnification map holding up at high factors.

T0 · AI review

How this review was made: An AI referee (grok-4.3) read the paper on July 2, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

  1. T0 AI review An AI read the paper and answered a fixed list of questions.
  2. T1 Audit Automatic checks that the references and files the paper relies on exist and match.
  3. T2 Compute The paper's code or calculations re-run with the same results.
  4. T3 Formal statement The claims written out in a formal math language a computer can read.
  5. T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 2 major objections · 2 minor · Ask about this paper · Read the full review →
Challenge this review Comment (0) Read on arXiv ↗
Cite ▾
arXiv ID2607.01129 (version 1)
Pith IDUW33EQXU, Pith's permanent record of this paper and its review
CitationCaio Moreira Goolsby, Christopher J. Conselice, Duncan Austin, Jordan D'Silva, Jose Diego, Julien Marabotto, Nathan Adams, Qiong Li, Tom Harvey. "Tilting at the Turnover: Modeling the Faint-End of the UV Luminosity Function Behind Abell s1063 with JWST." arXiv:2607.01129 (2026). https://pith.science/paper/UW33EQXU
Cite this reviewBibTeX for Pith's review of this paper
Claim paper

Why Pith says this

Highlight any text to comment on it.

Referee report

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

Galaxies fainter than M_UV=-17 supply more than half the UV luminosity density and at least 64% of ionizing photons at z=6.

The reading

This paper uses ultra-deep JWST and HST imaging behind the lensing cluster Abell S1063 to reach faint high-redshift galaxies and build the UV luminosity function at z approximately 6 to 11. The authors replace any sharp turnover cutoff with a gradual quadratic suppression that still permits some contribution from galaxies below the turnover magnitude. They detect no turnover signature down to M_UV=-13.5 at z=6 and report lower limits on the total UV output, star-formation rate density, and ionizing photon production. The results show that the faintest star-forming galaxies dominate the ultraviolet light budget and photon supply during the epoch of reionization.

What carries the argument

The gradual quadratic suppression model for the UV luminosity function turnover, fitted to counts of strongly lensed high-redshift candidates, which allows a reduced but non-zero contribution from galaxies fainter than the turnover magnitude.

Extended reading notes

Core claim

By constructing a photometric catalogue of lensed high-redshift candidates and modeling the UVLF turnover as a gradual quadratic suppression rather than a hard cutoff, the analysis shows that galaxies fainter than the conventional M_UV=-17 limit contribute more than half of the UV luminosity density and at least ∼64% of the ionizing photons produced by star-forming galaxies at z=6. Lower limits are derived as ρ_UV ≥22×10^25 erg s^-1 Hz^-1 Mpc^-3, SFRD ≥25×10^-3 M_⊙ yr^-1 Mpc^-3, and log10(n_ion/s^-1 Mpc^-3) ≥51.02. The model permits a suppressed but non-zero population beyond the turnover, so sources fainter than M_t still add to both ρ_UV and n_ion.

Load-bearing premise

The strong-lensing magnification model for Abell S1063 together with the photometric selection and completeness corrections accurately recover the intrinsic luminosities and redshifts of the high-z candidates down to M_UV≈-13.5 without large systematic biases.

Editorial extensions

If this is right

  • Galaxies fainter than M_UV=-17 contribute more than half the UV luminosity density at z=6.
  • These faint galaxies produce at least 64% of the ionizing photons from star-forming galaxies at z=6.
  • The UV luminosity density at z=6 has a lower limit of 22×10^25 erg s^-1 Hz^-1 Mpc^-3.
  • The star formation rate density at z=6 has a lower limit of 25×10^-3 M_⊙ yr^-1 Mpc^-3.
  • Reionization models must account for the shape of any turnover to capture the contribution from sources fainter than M_t.

Reading between the lines

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

  • If the lensing model holds, deeper imaging could directly detect the population below M_UV=-13.5 and test whether the quadratic suppression continues or flattens.
  • The large share of ionizing photons from faint galaxies implies that reionization calculations limited to brighter sources will under-predict the total photon budget.
  • Repeating the same lensing analysis on additional clusters would test whether the faint-end behavior is universal or varies with environment.
  • The absence of a detected turnover suggests the luminosity function may keep rising slowly at the faintest luminosities probed so far.
Original abstract

We leverage the strong gravitational field of Abell S1063 to identify faint, highly magnified galaxies using ultra-deep James Webb Space Telescope (JWST)/NIRCam imaging from the GLIMPSE survey and ancillary Hubble Space Telescope (HST)/ACS imaging from the Hubble Frontier Fields program. We construct a photometric catalogue of lensed high-redshift candidates and use these sources to constrain the faint end of the rest-frame UV luminosity function (UVLF) over $z\simeq6$--11. Rather than treating the UVLF turnover ($M_{\rm t}$) as a hard cutoff, we model it as a gradual quadratic suppression and explicitly account for the potential continued contribution of galaxies beyond the turnover. In a shallow-turnover scenario, up to one-third of the UV luminosity density can arise from sources fainter than $M_{\rm t}$. While we find no direct evidence for a turnover down to $M_{\rm UV}=-13.5$ at $z=6$, our analysis can only confidently exclude weak, medium, and strong turnover models down to $M_{\rm t}=-15.9$, $-15.1$, and $-14.8$, respectively. Across these models, we infer lower limits of the UV luminosity, star formation density, and the ionization rate as: $\rho_{\rm UV}\geq22\times10^{25}\,{\rm erg\,s^{-1}\,Hz^{-1}\,Mpc^{-3}}$, ${\rm SFRD}\geq25\times10^{-3}\,M_\odot\,{\rm yr^{-1}\,Mpc^{-3}}$, and $\log_{10}(\dot{n}_{\rm ion}/{\rm s^{-1}\,Mpc^{-3}})\geq51.02$. We find that galaxies fainter than the conventional $M_{\rm UV}=-17$ limit contribute more than half of the UV luminosity density and at least $\sim64\%$ of the ionizing photons produced by star-forming galaxies at $z=6$. Because our turnover model permits a suppressed, but non-zero, galaxy population beyond $M_{\rm t}$, sources fainter than the turnover remain contributors to both $\rho_{\rm UV}$ and $\dot{n}_{\rm ion}$, emphasizing the need to consider the turnover and its shape during reionization.

Figures
Figure 1. RGB Composite of the processed, calibrated, and background subtracted F200W, F356W, and F444W cluster module GLIMPSE NIRCam observations of the Abell S1063 Cluster. The mosaic images are combined non-linearly using a hyperbolic stretch to highlight the details of the cluster galaxies, ICL and background lensed arcs. systems, suggest that faint galaxies may contribute a larger share of the ionizing photon budget than previously inferred (Endsley et al. 2023; Donnan et al. 2024; Adams et al. 2024). These results place renewed emphasis on the broader problem of balancing the ionizing…
Figure 2. RGB Composite of the processed, calibrated, and background sub￾tracted F200W, F356W, and F444W non-cluster module GLIMPSE NIRCam data products. The mosaic images are combined non-linearly using a hyper￾bolic stretch. due to the extreme depth and sensitivity of the GLIMPSE survey, the background subtraction was done on a more aggressive 32x32 pixel grid rather than the more common 64x64 background calibration. This was done with a custom two-dimensional background model￾ing procedure (implemented via photutils; Bradley et al. 2022). The results of this are visible in
Figure 3. RGB composite of processed, calibrated, and background￾subtracted Hubble Fronteir Fields F814W, F606W, and F435W imaging of AS1063. Yellow region represents the critical curve (magnification greater than 100) at redshift 8 in the image plane. wavelet decomposition of the science image to attempt to extract dif￾fuse, large-scale structure. However, due to the nature of Abell S1063, the vast majority of the structure identified and removed was diffuse light from the brightest central galaxy of the cluster, rather than the intra-cluster light that dominates discussions of cluster…
Figure 4. Best-fit SED for galaxy ID:8777, one of a multiply-imaged lensed galaxy in AS1063. The top panel shows the best high-redshift and low-redshift EAZY SED fit, along side the probability distribution for each solution on the top right. Below, the apertures for our photometery (green circle) are overlaid on the cutout for the galaxy in each filter. On the bottom is the BAGPIPES SED fit for galaxy 8777, alongside its BAGPIPES star formation history (top left), de-lensed UV-Magnitude PDF (top right), and stellar-mass PDF (bottom right). This image of the galaxy is highly magnified, with…
Figure 5. Redshift–𝑀UV distribution of the selected galaxy sample behind Abell S1063. For each galaxy, the observed UV absolute magnitude is shown together with the corresponding intrinsic, lensing-corrected value, with faint connecting lines linking the two measurements. Points are coloured by the gravitational magnification factor, 𝜇, shown on a logarithmic scale. The lens￾ing correction shifts galaxies to intrinsically fainter 𝑀UV values, illustrating how strong magnification enables the recovery of galaxies that would other￾wise lie below the blank-field detection limit. We adopt the…
Figure 6. Photometric vs. spectroscopic redshifts for all grade = 3 objects in our matching of spectroscopic redshifts with our catalog. Sources which were not selected by our methods to be reliable galaxies are shown in gray; selected sources are highlighted. The solid line indicates 𝑧phot = 𝑧spec and the dashed lines show ±0.5 in |Δ𝑧 |. For the selected, grade = 3 subset with robust spectroscopic matches, we obtain median[Δ𝑧/(1 + 𝑧spec ) ] = 0.005, 𝜎NMAD = 0.0147, std = 0.024, and RMSE = 0.023, with zero catastrophic outliers under either criterion. 4 RESULTS 4.1 Highly-Magnified Galaxy Cand
Figure 7. Ultraviolet luminosity function for galaxies at redshifts z = 6, where we probe the deepest into the faintest systems down to 𝑀𝑈𝑉 ∼ −13. Our data points are red triangles. The solid and dashed grey lines represent Schechter and Double Power Law (DPL) fits to the UVLF at redshift 𝑧 ∼ 6 respectively. The solid blue line and the dashed purple line represent the same Schechter and DPL, but with the faint end slope fitted to our data. The outline of a red triangle is one of our data points, but brighter than the knee of the Schechter and DPL function, and therefore excluded when the…
Figure 8. Ultraviolet Luminosity function for various redshift bins (top left) Redshift z=7; (top right) Redshift z=8; (bottom left) Redshift z=9; (bottom right) Redshift z=11. Our data points are red triangles. The solid and dashed grey lines represent Schechter and Double Power Law (DPL) fits to the UVLF at redshift z=6 respectively. The solid blue line and the dashed purple line represent the same Schechter and DPL, but with the faint end slope fitted to our data. Comparisons to previous work on these measurements are also shown. These are parameterized by the faint-end slope 𝛼,…
Figure 9. Evolution of the faint end slope across the Epoch of Reionization. The best-fitted faint end slopes for each of our redshift bins are displayed in red for the Schechter fit and in blue for the DPL fit. Comparable literature faint-end slope values are also visibile, with the points being grey for non-lensing surveys and colour for surveys that leveraged gravitational lensing. Results from previous work using the GLIMPSE data are green
Figure 10. This figure illustrates our attempt to constrain the turnover by directly comparing a turnover to a no-turnover model. In the top figure, you can see how by assuming a strong turnover, and by having the turnover strength (𝛿) and the turnover magnitude as free parameters, we observe the Δ BIC significantly favoring the no-turnover model, despite the very low Δ𝜒 2 . Applying this generally-used methodology to every redshift bin, we can ostensibly put a lower limit on the turnover magnitude at various redshifts, as seen in the lower figure. However, for a more robust treatment of the…
Figure 11. A comparison of the best-fit Double Power Law (DPL) fit of the Redshift 6 UVLF with the UVLF DPL with various turnover models with 𝛿 = 0.1, 0.2, 0.4, and 0.8 and 𝑀t = −14, −15, and -17. This illustrates how little variation there is in the faint end between the turnover and no turnover models, especially when you assume a weak suppression mechanism
Figure 12. A heat map of the Δ𝜒 2 , of how much better the no-turnover model fits the observed redshift 𝑧 = 6 UVLF when compared to a turnover model with a given 𝛿 and 𝑀t . Stronger (higher 𝛿) and higher mass/ lower magnitude (lower 𝑀t) turnovers are strongly excluded by the comparison, while it is difficult to robustly exclude shallow turnovers or turnovers at higher magnitudes as the faint end of the UVLF is not constrained enough yet. Our fixed-grid turnover analysis, in which the DPL UVLF is modi￾fied by a quadratic faint-end suppression, more robustly tests a range of turnover magnitudes…
Figure 13. Redshift evolution of the minimum allowed non-ionizing UV luminosity density, 𝜌UV, inferred from fixed-strength faint-end turnover models. Coloured diamond markers show the lower-limit values from this work, obtained by placing the turnover at the limiting magnitude still allowed by the data at the 90 per cent confidence level. The plotted values thus represent conservative minimum estimates of 𝜌UV. The three turnover prescriptions correspond to a quadratic suppression applied to the DPL faint end, with weak, medium, and strong turnover strengths given by 𝛿 = 0.2, 0.5, and 0.8,…
Figure 14. Constraints on the faint-end UV galaxy population from this work. Left: comparison of turnover magnitude constraints with previous studies, showing that, aside from Livermore et al. (2017), this work places among the tightest constraints on the UVLF turnover to date. Right: comparison of the inferred 𝜌UV lower limits from this work with literature values from Oesch et al. (2013); Bouwens et al. (2015); Finkelstein et al. (2015); Harikane et al. (2023); Donnan et al. (2023); Atek et al. (2024); Adams et al. (2024). where we are able to place the strongest constraints on the…
Figure 15. Lower limits on the ionizing photon emissivity, 𝑛¤ion, inferred from the turnover-constrained UV luminosity densities using a luminosity-dependent Duncan-style 𝜉ion (𝑀UV ) relation and a Chisholm-style escape-fraction prescription, 𝑓esc (𝛽UV ). Literature measurements and model constraints are shown for comparison. These estimates should be interpreted cautiously, since the calculation extrapolates the 𝜉ion and 𝑓esc relations to extremely faint galaxies, beyond the magnitude range over which these empirical relations were originally calibrated. of the UV luminosity density in our…
Figure 16. Fraction of the ionizing photon emissivity, 𝑛¤ion, contributed by galaxies fainter than 𝑀UV = −17 at redshift 6 as a function of turnover magnitude and quadratic turnover strength. The calculation combines the DPL UVLF with the luminosity-dependent 𝜉ion and 𝛽UV relations from Austin et al. (2025), together with the 𝑓esc (𝛽UV ) prescription from Chisholm et al. (2022). Across the allowed parameter space, at least ∼ 64% of the ionizing photons are produced by galaxies fainter than 𝑀UV = −17, highlighting the dominant role of very faint galaxies in sustaining the ionizing photon…
Figure 17. Estimated f-ractional contribution to the total UV luminosity density, 𝜌UV, as a function of the adopted turnover strength, 𝛿, for the 𝑧 = 6 double power-law luminosity function. The calculation uses the DPL parameters 𝑀∗ UV = −21.20, log10 (𝜙 ∗ /Mpc−3 mag−1 ) = −3.72, faint-end slope 𝛼 = −2.071, and bright-end slope 𝛽 = −5.1. The two panels show different assumed turnover magnitudes, 𝑀turn = −17 and −15. The stacked regions separate the relative contribution from bright galaxies, 𝑀UV < −21, typical galaxies, −21 < 𝑀UV < −17, faint galaxies, −17 < 𝑀UV < −11, and ultra-faint…

Figures reproduced from arXiv:2607.01129 by the authors.

References

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 12 sentences without a confident label

Uses the method or tool of 2 papers
Uses data or values from 2 papers
Compares with 1 paper
  • These lower limits on $ṅ_ion$ and $ρ_UV$ are elevated when compared to previous observational estimates as seen in previous works, notably [ref] , as seen in Figure [ref] and Figure [ref], particularly at $z=6$ where we are able to place the strongest constraints on the population of faint galaxies.

    2512.10839 · The Ionizing Photon Budget · 1.00

Cites as support 4 papers
  • While our adopted $ξ_ion$ and $f_esc$ relations must be extrapolated into a magnitude regime where they are not yet well calibrated, recent spectroscopic studies have found that our assumptions may even be conservative in the faint-regime, leading to an under-estimation of the importance of the faint galaxy population [ref] .

    A GLIMPSE into the UV Continuum Slopes of the Faintest Galaxies in the Epoch of Reionization · Conclusion · 0.99

  • While our adopted $ξ_ion$ and $f_esc$ relations must be extrapolated into a magnitude regime where they are not yet well calibrated, recent spectroscopic studies have found that our assumptions may even be conservative in the faint-regime, leading to an under-estimation of the importance of the faint galaxy population [ref] .

    2601.20045 · Conclusion · 0.99

  • This picture is supported by recent observational and simulation-based work suggesting that low-mass galaxies dominate the escaped ionizing budget at earlier times, even if more massive systems become increasingly important toward lower redshift [ref] .

    Bridging the Gap between Cosmic Dawn and Reionization favors Faint Galaxies-dominated Models · Introduction · 0.98

Cites as background 3 papers
  • A more complete discussion of this survey can be found in [ref] .

    2601.20045 · Spectroscopic Comparison · 0.90

  • A full discussion of this survey can be seen in [ref] .

    2511.07542 · JWST GLIMPSE Survey · 0.85

  • The GLIMPSE survey, as one of the most unique and deepest surveys in JWSTs history, has already been the subject of much analysis and research, especially by those looking at the population of faint, highly magnified galaxies [ref] .

    2512.11790 · Comparison to Previous GLIMPSE Works · 0.99

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.