arxiv: 2601.16229 · v2 · submitted 2026-01-20 · 🌌 astro-ph.CO · gr-qc

Recognition: no theorem link

Calibration-independent consistency test of BAO and SNIa data: update

Bikash R. Dinda , Roy Maartens , Chris Clarkson

Authors on Pith no claims yet

Pith reviewed 2026-05-16 12:31 UTC · model grok-4.3

classification 🌌 astro-ph.CO gr-qc

keywords BAOSNIaconsistency testAlcock-Paczynski parameterGaussian processesDESIDES-Dovekiemodel-independent

0 comments

The pith

Uncalibrated BAO measurements from DESI DR2 agree with three SNIa catalogs within about 1 sigma via a shared distortion parameter.

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

The paper updates an earlier consistency test between baryon acoustic oscillation data and supernova Ia distance measurements by applying it to the latest releases. It reconstructs the Alcock-Paczynski parameter separately for each uncalibrated data set using Gaussian processes, then checks whether the implied expansion histories match without assuming any dark energy model or absolute scale such as the sound horizon or supernova peak magnitude. The test finds that DESI DR2 BAO data line up with the Union3, Pantheon+, and DES-Dovekie supernova samples at the level of roughly one standard deviation. A reader would care because persistent mismatch at this level would signal hidden systematics that could bias combined cosmological inferences, while agreement supports treating the data sets as mutually reliable for joint analyses.

Core claim

The authors show that the uncalibrated BAO data from DESI DR2 and the supernova Ia data from Union3, Pantheon+, and DES-Dovekie are consistent with one another at approximately the 1 sigma level when compared through their separately reconstructed Alcock-Paczynski parameters. This agreement is obtained independently of any assumed cosmological model, modified gravity theory, sound-horizon scale, or supernova absolute magnitude.

What carries the argument

Gaussian process reconstruction of the Alcock-Paczynski parameter from each data set, used to compare the shape of the expansion history without calibration constants.

If this is right

The four data sets can be combined for joint cosmological parameter estimation without introducing calibration-dependent biases.
The resolution of earlier tension after the DES-Dovekie update indicates that the prior discrepancy was tied to specific processing details in the earlier DES-Y5 release.
The same reconstruction method can be reapplied to future data releases to monitor consistency before they enter combined analyses.
No evidence appears for calibration-independent systematics that would require new physics to explain.
The test remains valid even if the underlying cosmology includes evolving dark energy or modified gravity.

Where Pith is reading between the lines

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

The approach could be extended to cross-check against other distance probes such as strong lensing time delays or gamma-ray bursts for broader model-independent validation.
If upcoming high-redshift BAO or supernova measurements from Euclid or the Vera Rubin Observatory produce tension in this test, the discrepancy would point to measurement issues independent of calibration scales.
Because the method isolates shape information, it could help isolate whether any future tensions arise from low-redshift calibration or from high-redshift behavior.

Load-bearing premise

The Gaussian process reconstruction of the Alcock-Paczynski parameter from each data set is unbiased and accurate enough to reveal any real inconsistencies.

What would settle it

A calculation showing that the reconstructed Alcock-Paczynski parameter from DESI BAO and one of the supernova sets differs by more than 2 sigma over a well-sampled redshift range would falsify the consistency claim.

read the original abstract

In a recent paper, arXiv:2509.19899, we presented a new method to test the consistency between uncalibrated BAO and SNIa data through a common parameter, the Alcock-Paczynski variable. Using Gaussian Processes, we can determine if various datasets are consistent, independently of dark energy or modified gravity models, and of the sound horizon and SNIa peak magnitude. We found that the DES-Y5 SNIa data showed non-negligible tension with other datasets. However, the recent update DES-Dovekie removes this tension. We find that all uncalibrated data from DESI DR2 BAO and three SNIa datasets, Union3, Pantheon+, and DES-Dovekie, are consistent with each other within $\sim 1\sigma$.

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

This update resolves the earlier DES-Y5 tension with the new DES-Dovekie data, leaving DESI DR2 BAO consistent with Union3, Pantheon+, and DES-Dovekie at ~1σ in the AP parameter test.

read the letter

The main thing to know is that swapping in the DES-Dovekie SNIa sample removes the tension reported in their prior work, so the uncalibrated BAO and SNIa sets now agree within about 1 sigma on the Alcock-Paczynski parameter. They reconstruct that parameter separately from each dataset using Gaussian processes and check overlap without assuming a dark energy model or fixing the sound horizon or SNIa absolute magnitude. That keeps the test genuinely calibration-independent and focused on whether the datasets are telling the same story about the expansion history. The approach is straightforward and uses public data releases, which is useful for anyone who wants a quick cross-check before combining probes in joint analyses. The update itself is incremental rather than a new framework, but it directly addresses the previous result with the latest sample. The soft spot is that the abstract gives no detail on kernel choice, hyperparameter optimization, or cross-validation for the GP step. If the full paper shows those checks and confirms the uncertainties are well-calibrated across datasets, the consistency claim holds up cleanly; without them the result rests on the standard RBF kernel behaving the same way on BAO and SNIa points. This is the sort of targeted consistency test that matters for people building model-independent expansion histories. It is solid enough on its own terms to go to a serious referee, even if the payoff is mainly confirmatory rather than revolutionary.

Referee Report

2 major / 1 minor

Summary. The manuscript updates a prior analysis (arXiv:2509.19899) by applying Gaussian Process reconstructions of the Alcock-Paczynski parameter to perform a calibration-independent consistency test between uncalibrated DESI DR2 BAO data and three SNIa datasets (Union3, Pantheon+, DES-Dovekie). It concludes that these datasets agree within ~1σ, resolving the tension previously reported with DES-Y5.

Significance. If the GP reconstructions prove robust, the work supplies a model-independent cross-check on BAO and SNIa datasets that avoids assumptions about dark energy, modified gravity, sound horizon, or SNIa absolute magnitude. This is useful for isolating systematics in public data releases and for guiding future analyses that combine these probes.

major comments (2)

[§3] §3 (Gaussian Process implementation): the central claim that the reconstructions are unbiased and yield well-calibrated uncertainties for the ~1σ consistency test is load-bearing, yet the text provides no robustness checks against kernel choice (e.g., RBF versus Matérn) or hyperparameter optimization procedure; without these, apparent agreement could be an artifact of the GP prior rather than data concordance.
[§4] §4 (results and posterior comparison): the quantitative measure of consistency (overlap integrals, tension metrics, or effective σ values) between the four independent AP reconstructions is not reported in sufficient detail to allow independent verification of the ~1σ statement.

minor comments (1)

[Abstract] The abstract and introduction should explicitly reference the kernel and covariance function used in the GP, even if details appear later in the text.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and constructive comments. We address each major comment below and have revised the manuscript to incorporate the suggested improvements.

read point-by-point responses

Referee: [§3] §3 (Gaussian Process implementation): the central claim that the reconstructions are unbiased and yield well-calibrated uncertainties for the ~1σ consistency test is load-bearing, yet the text provides no robustness checks against kernel choice (e.g., RBF versus Matérn) or hyperparameter optimization procedure; without these, apparent agreement could be an artifact of the GP prior rather than data concordance.

Authors: We thank the referee for highlighting this important aspect of the analysis. The original manuscript used the RBF kernel with hyperparameters optimized via marginal likelihood, which is a standard choice for smooth cosmological functions. However, we acknowledge that explicit robustness checks against alternative kernels and optimization procedures were not presented. In the revised version, we have added a dedicated subsection in §3 that includes reconstructions using the Matérn 3/2 and 5/2 kernels, as well as tests with fixed versus optimized length scales. The resulting AP parameter reconstructions and their uncertainties are consistent across these choices, and the inter-dataset agreement remains within ~1σ. A new figure and accompanying text document these tests, demonstrating that the reported consistency is not driven by the specific GP prior. revision: yes
Referee: [§4] §4 (results and posterior comparison): the quantitative measure of consistency (overlap integrals, tension metrics, or effective σ values) between the four independent AP reconstructions is not reported in sufficient detail to allow independent verification of the ~1σ statement.

Authors: We agree that additional quantitative detail on the consistency metric would improve verifiability. The revised manuscript now expands §4 to include explicit overlap integrals between the GP-reconstructed posterior distributions for the AP parameter, together with the corresponding effective tension values expressed in σ units. These quantities are summarized in a new Table 2, which reports all pairwise consistencies as lying below 1.2σ and thereby supports the ~1σ statement. The text also describes the precise computational procedure used to obtain the overlap integrals, enabling independent verification by readers. revision: yes

Circularity Check

0 steps flagged

No significant circularity: independent GP reconstructions compared directly

full rationale

The paper applies Gaussian Process reconstruction independently to each dataset (DESI DR2 BAO, Union3, Pantheon+, DES-Dovekie) to obtain the Alcock-Paczynski parameter as a function of redshift, then checks for overlap within ~1σ. This is a direct empirical comparison with no shared cosmological model fit, no parameter tuned on a subset then called a prediction, and no self-definitional reduction. The citation to arXiv:2509.19899 describes the method but does not bear the load of the consistency result itself, which remains falsifiable against the input data. No uniqueness theorems, ansatze, or renamings are involved. The derivation chain is self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The approach assumes Gaussian Processes can faithfully reconstruct the AP parameter from BAO and SNIa observables without model assumptions and that the AP variable serves as a sufficient model-independent consistency metric.

axioms (1)

domain assumption Gaussian Processes provide unbiased reconstruction of the Alcock-Paczynski parameter from uncalibrated BAO and SNIa data
Invoked to enable model-independent comparison

pith-pipeline@v0.9.0 · 5433 in / 1177 out tokens · 42376 ms · 2026-05-16T12:31:37.623684+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.

Model-independent consistency tests of DESI DR2 BAO and SN Ia
astro-ph.CO 2026-04 unverdicted novelty 4.0

DESI DR2 BAO and Pantheon+/Union3 SN Ia datasets are mutually consistent at 1-2 sigma using crossing statistics, supporting a reconstruction suggestive of evolving dark energy at low redshift.