arxiv: 2511.22243 · v3 · submitted 2025-11-27 · 🌌 astro-ph.CO

Recognition: 1 theorem link

· Lean Theorem

Forecasting local Primordial Non-Gaussianities from UNIONS Lyman-Break Galaxies and Planck CMB lensing

Constantin Payerne , William d'Assignies , Christophe Y\`eche , Hendrik Hildebrandt , Dustin Lang , Thomas de Boer , S\'ebastien Fabbro

Authors on Pith no claims yet

Pith reviewed 2026-05-17 05:06 UTC · model grok-4.3

classification 🌌 astro-ph.CO

keywords local primordial non-GaussianityLyman-break galaxiesCMB lensingUNIONS surveycross-power spectrumforecastf_NL

0 comments

The pith

Cross-correlating Lyman-break galaxies from UNIONS with Planck CMB lensing constrains local primordial non-Gaussianity to σ(f_NL^loc)=34 for a photometric sample.

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

This paper forecasts how well the local primordial non-Gaussianity parameter can be measured by cross-correlating high-redshift Lyman-break galaxies selected from UNIONS ugriz imaging with maps of Planck CMB lensing. The cross-power spectrum approach is chosen because it cancels many imaging systematics that affect galaxy auto-spectra. A sympathetic reader cares because a tighter limit on f_NL^loc would give a sharper test of whether inflation was driven by a single field. For an idealized photometric sample at 1100 galaxies per square degree the forecast reaches 34, and spectroscopic follow-up with DESI-II improves it to 20. A denser sample drawn from early UNIONS data reaches the same 20 even in the photometric case, though redshift-distribution uncertainties degrade the result.

Core claim

From the angular cross-power spectrum between LBGs and Planck CMB lensing, the paper forecasts σ(f_NL^loc)=34 for an idealized photometric sample of r<24.3 LBGs at 1,100 deg^{-2} selected with a Random Forest from UNIONS-like imaging; the same analysis on a realistic denser sample at 1,400 deg^{-2} yields σ(f_NL^loc)=20, with both cases improving to 20 after DESI-II spectroscopic follow-up, while clustering-redshift uncertainties are shown to weaken the constraints substantially.

What carries the argument

The angular cross-power spectrum between the Lyman-break galaxy density field and the Planck CMB lensing convergence map, which isolates the scale-dependent bias signature of local PNG while canceling many imaging systematics.

If this is right

An idealized photometric LBG sample at surface density 1,100 deg^{-2} delivers σ(f_NL^loc)=34.
DESI-II spectroscopic follow-up improves the same forecast to σ(f_NL^loc)=20.
A denser realistic sample at 1,400 deg^{-2} from early UNIONS data reaches σ(f_NL^loc)=20 without follow-up.
Accounting for uncertainties in the clustering-redshift distribution significantly degrades the f_NL^loc precision.

Where Pith is reading between the lines

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

Applying the same cross-correlation method to future wider or deeper photometric surveys could push the precision below 20 even before spectroscopic follow-up.
Joint analyses that combine this probe with other large-scale structure tracers may help separate the PNG signal from galaxy bias uncertainties.
Dedicated efforts to improve redshift-distribution calibration using additional spectroscopic overlap would directly translate into tighter f_NL^loc bounds.

Load-bearing premise

The redshift distribution of the selected Lyman-break galaxies must be calibrated accurately enough that uncertainties in it do not dominate the error budget on f_NL^loc.

What would settle it

Performing the actual cross-power spectrum measurement on real UNIONS LBGs and Planck lensing data and obtaining a constraint on f_NL^loc whose uncertainty is substantially worse than the forecasted 34 due to redshift-distribution errors or unmodeled systematics.

Figures

Figures reproduced from arXiv: 2511.22243 by Christophe Y\`eche, Constantin Payerne, Dustin Lang, Hendrik Hildebrandt, S\'ebastien Fabbro, Thomas de Boer, William d'Assignies.

**Figure 1.** Figure 1: Left: Photometric redshift distributions of the photometric LBG samples UNIONSlike_RF in red (the convolution of the UNIONSlike_RFdistribution with the DESI spectroscopic efficiency is shown in dashed and dotted lines, they are not normalized to show the impact at z ∼ 2.5). The distribution of the samples UNIONSlike_RF_u180s, and LSSTY4like_RF_u2x180s are shown in blue and dashed cyan lines. Right: Photome… view at source ↗

**Figure 2.** Figure 2: Top panel shows the angular power spectra C gg ℓ , C κg ℓ , and C κκ ℓ (solid lines) for the baseline sample. The cosmological-only contribution is shown as dashed lines, while the noise contribution is shown as dotted lines. Bottom panel shows the signal-to-noise ratio, defined as the ratio of the cosmological-only term to the noise, for C gg ℓ and C κκ ℓ . We obtain σ(f loc NL ) = 7 from C gg ℓ alone. F… view at source ↗

**Figure 3.** Figure 3: Left: Posterior distribution of the parameters f loc NL and the large-scale linear bias (expressed in terms of relative error). Right: Forecasted error on f loc NL as a function of the lower multipole cut ℓmin. Solid lines show constraints using only C κg ℓ (ℓ > ℓmin), while dashed lines correspond to the combination of C κg ℓ and C gg ℓ (ℓ > ℓmin). Different colors indicate different LBG number densities:… view at source ↗

**Figure 4.** Figure 4: Left: Forecasted error on f loc NL as a function of the high-redshift bias prescription b¯ 2(z). Solid lines correspond to a full rescaling of the galaxy bias by b0 in Eq. (19). In contrast, dashed lines correspond to rescaling only the high-redshift population bias by b0 in Eq. (21). Right: Forecasted error on f loc NL for different choices of bΦ(b), beyond the prescription given in Eq. (9). results are s… view at source ↗

read the original abstract

Local Primordial non-Gaussianities (PNGs), characterized by $f_{\rm NL}^{\rm loc}$, provide a powerful window into the physics of inflation. Cross-correlating high-redshift tracer samples with the CMB lensing potential offers a particularly robust probe of PNGs, mitigating imaging systematics that typically affect large-scale measurements from tracer auto-spectra. In this context, UNIONS enables the selection of $u$-dropout high-redshift Lyman-Break Galaxies (LBGs). We perform a MCMC-based forecast to estimate the uncertainties on $f_{\rm NL}^{\rm loc}$ and on a galaxy bias parameter, which captures our uncertainty in the tracer bias. From the angular cross-power spectrum between LBGs and Planck CMB lensing, we forecast $\sigma(f_{\rm NL}^{\rm loc})=34$ for an idealized photometric sample of $r<24.3$ LBGs selected with a Random Forest classification algorithm from UNIONS-like $ugriz$ imaging, with a resulting surface density of $1,100$ deg$^{-2}$. This precision can be improved to $\sigma(f_{\rm NL}^{\rm loc})=20$ after spectroscopic follow-up with DESI, during its next phase starting in 2029, DESI-II. We test a more realistic $u$-dropout LBG selection using early UNIONS data, which yields a denser sample of $r<24.2$ objects at $1,400$ deg$^{-2}$. From this sample, covering a larger footprint and expected to have a higher large-scale galaxy bias, we forecast $\sigma(f_{\rm NL}^{\rm loc})=20$, with similar precision achievable after DESI spectroscopic follow-up. In addition, we perform preliminary validation of the redshift distribution using the clustering-redshift method with DESI DR1 data, confirming the calibration from deep, small-area photometric fields. However, accounting for uncertainties in the clustering-redshift distribution significantly degrades the $f_{\rm NL}^{\rm loc}$ constraining power.

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 quoted sigma(f_NL) forecasts of 20-34 assume fixed n(z) from clustering redshifts; the paper notes that marginalizing over those uncertainties degrades the constraints substantially.

read the letter

The headline result here is a set of MCMC forecasts for local f_NL from the angular cross-power spectrum of UNIONS-selected Lyman-break galaxies with Planck CMB lensing. They report sigma(f_NL^loc) = 34 for an idealized r<24.3 photometric sample at 1100 deg^{-2} and sigma = 20 for a denser realistic sample at 1400 deg^{-2}, with similar numbers after adding DESI-II spectroscopy. Those numbers treat the redshift distribution as fixed. The abstract states outright that accounting for uncertainties in the clustering-redshift distribution significantly degrades the constraining power, which is the realistic scenario given how the signal depends on the overlap integral with the lensing kernel. That is the central caveat to keep in mind when reading the numbers. What the paper does is apply a standard cross-correlation forecast to a new photometric dataset. They use a Random Forest selection on UNIONS-like ugriz imaging, produce surface densities and bias estimates for both idealized and early real data, and run a joint MCMC on f_NL and a single galaxy bias parameter. They also perform a preliminary validation of n(z) against DESI DR1 clustering redshifts. The cross-correlation choice itself is sensible for reducing imaging systematics on large scales. The soft spots are in the level of detail and quantification. The abstract supplies limited information on the exact covariance estimation, the modeling assumptions for the cross-spectrum, or how the degradation from n(z) errors was calculated. Without those steps shown more explicitly, it is hard to judge how much the final error budget inflates once redshift calibration is treated as uncertain. Jointly fitting bias is standard but means the f_NL errors are not independent of the bias modeling. This is a forecasting paper aimed at people working on PNG constraints with photometric LBG samples and CMB lensing cross-correlations. Readers planning UNIONS or DESI analyses would find the specific numbers and selection details useful. It is solid enough on its own terms to deserve a serious referee, even if the main forecasts need the redshift uncertainty quantification expanded in revision.

Referee Report

2 major / 1 minor

Summary. The manuscript performs an MCMC forecast for the uncertainty on local primordial non-Gaussianity f_NL^loc using the angular cross-power spectrum between Lyman-break galaxies (LBGs) selected from UNIONS-like ugriz imaging and Planck CMB lensing. It reports σ(f_NL^loc)=34 for an idealized photometric r<24.3 sample at 1,100 deg^{-2} and σ=20 after DESI-II spectroscopic follow-up; a more realistic UNIONS-like sample at 1,400 deg^{-2} also yields σ=20. The redshift distribution is validated with clustering redshifts from DESI DR1, but the text notes that uncertainties in this distribution significantly degrade the constraints.

Significance. If the forecasts remain competitive after proper marginalization over redshift-distribution uncertainties, the work would usefully demonstrate the value of high-redshift LBG samples from wide-field surveys for PNG constraints via CMB-lensing cross-correlations, which are robust to imaging systematics. The preliminary clustering-redshift validation with existing DESI data and the explicit consideration of spectroscopic follow-up are constructive elements that help anchor the forecast to near-term observations.

major comments (2)

[Abstract] Abstract: The headline forecasts σ(f_NL^loc)=34 (idealized sample) and =20 (realistic sample and DESI-II) are obtained while treating the redshift distribution n(z) as fixed. The manuscript states that “accounting for uncertainties in the clustering-redshift distribution significantly degrades the f_NL^loc constraining power,” yet no quantified degradation factor or marginalized result is supplied. Because the cross-spectrum amplitude depends directly on the overlap integral between n(z) and the CMB lensing kernel, this assumption is load-bearing for the central claim.
[Forecast methodology] Forecast methodology: The MCMC varies only f_NL^loc and a single galaxy-bias parameter; limited information is given on the precise modeling of the cross-power spectrum (including any scale cuts or bias evolution), the covariance estimation procedure, and the validation of the forecast against simulated realizations. These details are required to assess whether the reported uncertainties are robust.

minor comments (1)

The abstract would benefit from a clearer separation between the optimistic (fixed-n(z)) forecasts and the degraded constraints once redshift uncertainties are included, to avoid potential misreading of the expected precision.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed report. Their comments highlight important aspects of our forecast that merit clarification and expansion. We address each major comment below and indicate the revisions we will make to the manuscript.

read point-by-point responses

Referee: [Abstract] Abstract: The headline forecasts σ(f_NL^loc)=34 (idealized sample) and =20 (realistic sample and DESI-II) are obtained while treating the redshift distribution n(z) as fixed. The manuscript states that “accounting for uncertainties in the clustering-redshift distribution significantly degrades the f_NL^loc constraining power,” yet no quantified degradation factor or marginalized result is supplied. Because the cross-spectrum amplitude depends directly on the overlap integral between n(z) and the CMB lensing kernel, this assumption is load-bearing for the central claim.

Authors: We agree that a quantitative assessment of the impact from n(z) uncertainties would strengthen the presentation. The current manuscript notes the degradation qualitatively based on our clustering-redshift validation with DESI DR1, but does not report a specific marginalized constraint. In the revised version we will add an explicit calculation marginalizing over nuisance parameters for the mean and width of n(z), using the DESI DR1 clustering-redshift uncertainties as priors. The resulting degradation factor on σ(f_NL^loc) will be reported in the results section and referenced in the abstract. revision: yes
Referee: [Forecast methodology] Forecast methodology: The MCMC varies only f_NL^loc and a single galaxy-bias parameter; limited information is given on the precise modeling of the cross-power spectrum (including any scale cuts or bias evolution), the covariance estimation procedure, and the validation of the forecast against simulated realizations. These details are required to assess whether the reported uncertainties are robust.

Authors: We appreciate the request for additional methodological transparency. The cross-power spectrum is modeled under the Limber approximation with a single, scale-independent galaxy bias parameter, as stated in Section 3; scale cuts are imposed at multipoles corresponding to k_max ≈ 0.1 h Mpc^{-1} to remain in the linear regime. The covariance is computed analytically under the Gaussian approximation, including cosmic variance, shot noise, and Planck lensing noise (Section 4). Validation consists of MCMC recovery tests on mock data vectors generated from the fiducial model, which recover the input parameters without bias. We will expand the text in the revised manuscript to make these elements more explicit, including a short dedicated paragraph on the forecast pipeline and validation procedure. revision: yes

Circularity Check

0 steps flagged

No circularity: standard MCMC forecast with explicit caveats on n(z)

full rationale

The paper's derivation consists of an MCMC forecast on the LBG-Planck lensing cross-power spectrum, jointly sampling f_NL^loc and a single galaxy bias parameter to produce forecasted uncertainties such as σ(f_NL^loc)=34. This procedure does not reduce any claimed prediction to its inputs by construction, nor does it rely on self-definitional relations, fitted inputs renamed as predictions, or load-bearing self-citations. The abstract explicitly flags that marginalizing over clustering-redshift n(z) uncertainties degrades the constraints, so the reported numbers are presented with that limitation rather than smuggled in as independent results. No equations or ansatzes are shown to collapse the output to the input assumptions, making the forecast self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central forecast rests on standard cosmological modeling of angular power spectra plus a fitted galaxy bias parameter and a clustering-redshift calibrated redshift distribution whose uncertainties are shown to degrade constraints.

free parameters (1)

galaxy bias parameter
Captures uncertainty in the tracer bias and is jointly estimated with f_NL^loc in the MCMC forecast.

axioms (1)

domain assumption Standard assumptions for computing the angular cross-power spectrum between galaxies and CMB lensing convergence
Invoked throughout the forecast modeling of the observable.

pith-pipeline@v0.9.0 · 5706 in / 1469 out tokens · 73029 ms · 2026-05-17T05:06:48.982983+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We perform a Markov Chain Monte Carlo forecast to estimate the uncertainties on f_loc_NL and on a galaxy bias parameter b0... From the angular cross-power spectrum between LBGs and Planck CMB lensing, we forecast σ(f_NL^loc)=34...

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages · 1 internal anchor

[1]

Aihara, H. et al. 2022, PASJ, 74, 247 Alonso, D. et al. 2019, MNRAS , 484, 4127 Anbajagane, D., Chang, C., Lee, H., & Gatti, M. 2024, JCAP, 2024, 062 Barreira, A. 2020, JCAP, 2020, 031 Barreira, A. et al. 2020, JCAP, 2020, 013 Benítez, N. 2011, ASCL, ascl:1108.011 Bernstein, G. et al. 2025, arXiv e-prints, arXiv:2506.00758 Brown, M. L. et al. 2005, MNRAS ...

work page internal anchor Pith review Pith/arXiv arXiv 2022
[2]

decoupled

is given by bCXY ℓ = 1 2ℓ+1 ℓX m=−ℓ baX,∗ ℓmbaY ℓm 2 (C.4) On full-sky, the pseudo angular power spectrum averages to CXY ℓ = 2 π Z ∞ 0 dk k2 P(k)W X ℓ (k)W Y ℓ (k),(C.5) whereP(k) is the 3D matter power spectrum, andW X ℓ (k) and W Y ℓ (k) are the radial transfer functions of the fields, given by W X ℓ (k)= Z dz dχ dz qX(χ)j ℓ(kχ).(C.6) Following the MAS...

work page 2002