Pith. sign in

REVIEW 3 major objections 3 minor 5 cited by

An Improved Distance to NGC 4258 and its Implications for the Hubble Constant

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read An improved geometric distance to NGC 4258 pins the Hubble constant to $H_0 = 73.5 \pm 1.4$ km s$^{-1}$ Mpc$^{-1}$ when combined with other anchors.

desk verdict A sharpened geometric anchor for the distance ladder: refitting the same NGC 4258 maser dataset with error floors as free parameters halves the distance uncertainty and keeps the Hubble tension at ~4.2σ, though the gain leans on a Gaussian-noise assumption that deserves scrutiny. read the letter →

arxiv 1908.05625 v2 pith:U7TOT6DJ submitted 2019-08-15 astro-ph.GA astro-ph.CO

M. J. Reid , D. W. Pesce , A. G. Riess This is my paper · ORCID
classification astro-ph.GAastro-ph.CO
keywords distancescaleHubbleconstantmegamaserNGC4258Cepheidladdertipoftheredgiantbranchgeometricverylongbaselineinterferometry
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper reanalyzes VLBI observations of water masers orbiting the black hole in NGC 4258, treating the data error floors as free parameters in an MCMC fit instead of fixed conservative values. It obtains a geometric distance of $7.576 \pm 0.082$ (stat.) $\pm 0.076$ (sys.) Mpc, reducing both statistical and systematic uncertainty relative to earlier work. Used to calibrate the Cepheid distance ladder, this distance yields a local Hubble constant of $72.0 \pm 1.9$ km s$^{-1}$ Mpc$^{-1}$ on its own and $73.5 \pm 1.4$ km s$^{-1}$ Mpc$^{-1}$ when combined with Milky Way parallaxes and LMC eclipsing binaries. A TRGB calibration on the same HST photometric system gives $M_{F814W} = -4.01 \pm 0.04$ mag and $H_0 = 71.1 \pm 1.9$ km s$^{-1}$ Mpc$^{-1}$, consistent with the Cepheid route.

What carries the argument

The load-bearing mechanism is the Keplerian modeling of water maser spots in the sub-parsec accretion disk of NGC 4258, where VLBI gives positions, Doppler velocities, and line-of-sight accelerations, and the ratio of angular to linear acceleration gives a purely geometric distance. The innovation is an MCMC fit in which the error floors are free parameters with flat priors, with the full Gaussian $\frac{1}{\sqrt{2\pi}}\frac{1}{\sigma}e^{-\Delta^2/2\sigma^2}$ likelihood normalization retained so the data can set their own weights; this removes the previous dependence on conservatively assumed error floors and lets their posterior distributions absorb part of the systematic budget. Minor changes include the $(1+z)$ velocity convention and defining warp parameters at the mean maser radius, and the results are cross-checked with an independent Hamiltonian MCMC code.

What would settle it

Re-fit the same 18-epoch VLBI data with the error floors fixed to independently measured astrometric and spectral calibration uncertainties; if the distance shifts by more than about $0.1$ Mpc, or if the epoch-to-epoch scatter of individual maser spots exceeds the inferred floors (e.g., $\sigma_x \approx 0.0016$ mas) by a factor of two, the error-floor model has missed real noise and the stated uncertainty is too small.

Watch

Extended reading notes

Core claim

The central claim is that the angular-diameter distance to NGC 4258 is $7.576 \pm 0.082$ (stat.) $\pm 0.076$ (sys.) Mpc, with the uncertainty reduced by nearly a factor of two compared to the previous best estimate of $7.596 \pm 0.170$ Mpc. The improvement comes from letting the five error floors (for eastward and northward positions, high-velocity and systemic velocities, and accelerations) be adjusted by the MCMC fit rather than fixed a priori. With this distance as the sole geometric calibrator of Cepheids, the paper derives $H_0 = 72.0 \pm 1.9$ km s$^{-1}$ Mpc$^{-1}$; combining all three geometric anchors gives $H_0 = 73.5 \pm 1.4$ km s$^{-1}$ Mpc$^{-1}$. A new TRGB absolute magnitude of $-4.01 \pm 0.04$ mag in F814W follows from the same distance and yields $H_0 = 71.1 \pm 1.9$ km s$^{-1}$ Mpc$^{-1}$, and the paper notes that using the same SN Ia intercept removes the residual Cepheid–TRGB difference.

Load-bearing premise

The five error floors are treated as free parameters with flat priors in the MCMC fit, so that the previously assumed systematic uncertainties are fully captured by their posterior distributions and can be removed from the systematic budget.

Editorial extensions

If this is right

  • The geometric distance to NGC 4258 now anchors the Cepheid ladder with roughly a 1.5% total uncertainty, down from about 2.6% in earlier joint analyses.
  • With all three geometric anchors (NGC 4258, Milky Way parallaxes, LMC eclipsing binaries), the best local value is $H_0 = 73.5 \pm 1.4$ km s$^{-1}$ Mpc$^{-1}$, which remains $4.2\sigma$ above the Planck + $\Lambda$CDM prediction.
  • The new TRGB absolute magnitude $M_{F814W} = -4.01 \pm 0.04$ mag is measured on the same HST photometric system and through similarly low extinction as SN Ia host halos, reducing systematic errors relative to LMC-based TRGB calibrations.
  • The small remaining offset between the Cepheid and TRGB routes ($H_0 = 72.0$ vs $71.1$) is traced to different SN Ia samples; using the same Hubble-diagram intercept brings the two routes into agreement.

Reading between the lines

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

  • If the error-floor marginalization is correct, the same fitting strategy could be applied to other megamaser galaxies with sufficient VLBI data, giving independent geometric anchors that do not depend on NGC 4258.
  • A direct test of the main assumption would be to compare the inferred position error floors (e.g., $\sigma_x \approx 0.0016$ mas) with the epoch-to-epoch scatter of the same maser spots; systematic underestimation would show up as scatter larger than the posterior floor.
  • Because the TRGB calibration is now on the native HST system, future TRGB surveys can avoid the blending and extinction corrections that have limited ground-based LMC calibrations, potentially pushing the TRGB $H_0$ uncertainty below $\pm 1.9$ km s$^{-1}$ Mpc$^{-1}$.
  • The paper's numbers imply that the Hubble tension is not driven by the NGC 4258 anchor: even the lowest value from this anchor alone ($72.0$ km s$^{-1}$ Mpc$^{-1}$) is still $2.4\sigma$ above Planck, so the discrepancy must come from elsewhere in the ladder or from new physics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The paper reanalyzes the VLBI maser data for NGC 4258, using a Markov-chain Monte Carlo approach in which the previously assumed error floors are treated as free parameters. The authors report a distance of D = 7.576 ± 0.082 (stat.) ± 0.076 (sys.) Mpc, about a factor of two improvement in statistical precision over Humphreys et al. (2013). They use this distance to recalibrate the Cepheid and TRGB distance ladders, obtaining H0 = 72.0 ± 1.9 km/s/Mpc from Cepheids alone, H0 = 71.1 ± 1.9 km/s/Mpc from the TRGB, and a combined Cepheid-anchor value of H0 = 73.5 ± 1.4 km/s/Mpc. The paper also derives a new TRGB absolute magnitude of M_F814W = -4.01 ± 0.04 mag.

Significance. If the error-floor treatment is valid, the paper delivers a more precise geometric anchor for the local distance scale and provides a TRGB calibration on the HST photometric system that reduces systematic errors relative to the LMC-based calibration. The analysis has several genuine strengths: two independent MCMC implementations give consistent results; mock datasets generated with different levels of Gaussian random noise recover the injected noise levels; the reduced chi-squared improves from 1.4 to 1.2; and the new distance is consistent with previous estimates. The paper also makes explicit, testable predictions for H0 that are directly relevant to the current Hubble tension. However, the central gain in precision rests on an assumption about the noise model that is not fully validated, and a few internal inconsistencies need attention.

major comments (3)
  1. [Section 2, Table 2] The text states that the position error floors previously adopted by Humphreys et al. (2013) were (sigma_x, sigma_y) = (±0.010, ±0.020) mas, but Table 2 lists the same assumed values in brackets as [0.0200] and [0.0300] mas. This is an internal inconsistency in a comparison that is central to the paper's claim of reduced uncertainty, and it must be corrected.
  2. [Section 2, paragraph beginning 'The position error floors...'] The reduction in the statistical distance uncertainty from ±0.170 to ±0.082 Mpc relies on the fitted Gaussian error floors (Table 2) being accurate descriptions of the measurement noise. The validation on mock datasets only tests recovery of injected Gaussian random noise levels; it does not test robustness to epoch-to-epoch correlations, non-Gaussian tails, or maser-structure-induced outliers. The earlier conservative floors were motivated by potential interferometric delay errors, which would produce correlated position errors across maser spots. If such correlations are present, the fitted floors (sigma_x ≈ 0.0016 mas, sigma_y ≈ 0.0041 mas) could be biased low, and the quoted statistical uncertainty would be underestimated. The authors should either demonstrate robustness with a correlated-noise model or retain a corresponding systematic term.
  3. [Section 2, paragraph beginning 'Further gains in distance accuracy...'] The paper removes the systematic contributions listed in Table 4 of Humphreys et al. (2013) because 'their uncertainties are now incorporated into the marginalized distance estimate.' This is only valid if the five fitted error floors fully represent each of those systematics. Since the error floors are single per-coordinate constants, they cannot capture epoch-dependent or spatially varying systematic errors. The justification for keeping only the spiral-structure term of ±0.076 Mpc therefore needs stronger support; otherwise the systematic uncertainty is understated.
minor comments (3)
  1. [Section 3, TRGB paragraph] The phrase 'greater depth (2.6 ks versus 8.8 ks in F814W)' appears to have the exposure times reversed; the GO 9810 observation with 8.8 ks is deeper than the GO 9477 observation with 2.6 ks.
  2. [Table 4, note d] The text says the average of the two TRGB measurements is adopted with the larger error, but note d describes a 'variance-weighted average' with ±0.022 mag; the text and table note should be reconciled.
  3. [References] The Roe reference is cited as 'arXiv:1906:09077'; the arXiv identifier format appears incorrect and should be checked.

Circularity Check

0 steps flagged · score 0.0 of 10

Geometric maser distance is derived independently of the distance-ladder data it calibrates; no circular step found.

full rationale

The central result, D = 7.576 +/- 0.082 (stat.) +/- 0.076 (sys.) Mpc, is obtained from VLBI mapping and spectral monitoring of H2O masers in the Keplerian accretion disk around the NGC 4258 black hole. The fitted data are maser positions, Doppler velocities, and accelerations; none of these inputs come from Cepheid, TRGB, or SN Ia observations. The distance is therefore not defined in terms of, and does not reduce to, the H0 calibration it subsequently anchors. The H0 estimates are applications of this distance to external data and previously published formalisms (Riess et al. 2016, 2019); these self-citations are to independent data sets and do not supply the maser distance itself. The adjustable error floors are nuisance parameters in the maser likelihood, fit to the same maser data; their effect on the uncertainty is a statistical modeling choice, not a case of fitting a target quantity and then predicting it. The consistency between the Cepheid and TRGB routes when a common SN Ia intercept is adopted is a comparison, not a derivation of H0 from itself. Concerns about the Gaussian error-floor model and possible correlated systematics are correctness risks, not circularity, because they do not reduce the stated distance to its inputs by construction.

Assumptions & free parameters 19 free parameters · 4 assumptions · 0 invented entities

The central distance estimate is produced by fitting a warped Keplerian disk model with 14 geometric and dynamical parameters plus five error floors to the VLBI maser data. These are the free parameters the paper pays for; no new physical entities are introduced. The main domain assumptions are that the masers trace the disk and that residual errors are Gaussian. The H0 results additionally assume the Cepheid and TRGB distance-ladder calibrations from prior literature.

free parameters (19)
  • Distance to NGC 4258 = 7.576 +/- 0.082 (stat.) +/- 0.076 (sys.) Mpc
    Primary fitted parameter of the maser disk model and the central result of the paper.
  • Black hole mass = 3.98 +/- 0.04 x 10^7 solar masses
    Fitted disk model parameter.
  • Galaxy systemic velocity = 473.3 +/- 0.4 km/s
    Fitted disk model parameter.
  • Dynamical center x offset = -0.152 +/- 0.003 mas
    Fitted disk model parameter.
  • Dynamical center y offset = 0.556 +/- 0.004 mas
    Fitted disk model parameter.
  • Disk inclination = 87.05 +/- 0.09 deg
    Fitted disk model parameter at r = 6.1 mas.
  • Inclination warp, 1st order = 2.59 +/- 0.07 deg/mas
    Fitted disk model parameter.
  • Inclination warp, 2nd order = 0.041 +/- 0.018 deg/mas^2
    Fitted disk model parameter.
  • Disk position angle = 88.43 +/- 0.04 deg
    Fitted disk model parameter.
  • Position angle warp, 1st order = 2.21 +/- 0.02 deg/mas
    Fitted disk model parameter.
  • Position angle warp, 2nd order = -0.13 +/- 0.01 deg/mas^2
    Fitted disk model parameter.
  • Orbital eccentricity = 0.007 +/- 0.001
    Fitted disk model parameter.
  • Periapsis angle = 318 +/- 13 deg
    Fitted disk model parameter.
  • Periapsis angle warp = 123 +/- 7 deg/mas
    Fitted disk model parameter.
  • Position error floor, eastward (sigma_x) = 0.0016 +/- 0.0005 mas
    Fitted as a free parameter; replaces the assumed 0.020 mas floor used by Humphreys et al. (2013).
  • Position error floor, northward (sigma_y) = 0.0041 +/- 0.0005 mas
    Fitted as a free parameter; replaces the assumed 0.030 mas floor used by Humphreys et al. (2013).
  • Systemic velocity error floor (sigma_v,sys) = 0.31 +/- 0.20 km/s
    Fitted as a free parameter; replaces the assumed 1.00 km/s floor.
  • High-velocity error floor (sigma_v,hv) = 2.25 +/- 0.31 km/s
    Fitted as a free parameter; replaces the assumed 1.00 km/s floor.
  • Acceleration error floor (sigma_a) = 0.46 +/- 0.04 km/s/yr
    Fitted as a free parameter; replaces the assumed 0.30 km/s/yr floor.
assumptions (4)
  • domain assumption Maser features in NGC 4258 trace a geometrically thin, warped, mildly eccentric Keplerian disk around a central point mass.
    The entire distance estimate follows from this disk model; the model form and parameterization are carried over from Humphreys et al. (2013) with modest modifications.
  • domain assumption Residual scatter in maser positions, velocities, and accelerations after model subtraction is Gaussian and can be represented by constant, per-component error floors.
    This is what makes the likelihood and the marginalization over error floors valid; mock tests are offered as support.
  • domain assumption The Cepheid period-luminosity relation and SN Ia standardization formalism of Riess et al. (2016) and Riess et al. (2019) are correct for the H0 estimates.
    The H0 results inherit these ladder calibrations; they are not rederived in this paper.
  • domain assumption The TRGB detections in NGC 4258 by Macri et al. (2006) and Jang and Lee (2017) are reliable on the HST system, with negligible halo extinction by convention.
    Used to derive the TRGB absolute calibration M_F814W = -4.01 +/- 0.04 mag.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Improved Distance to NGC 4258 and its Implications for the Hubble Constant." pith.science (2026). https://pith.science/paper/U7TOT6DJ

@misc{pith2026190805625,
  author       = {Pith},
  title        = {Pith review of: An Improved Distance to NGC 4258 and its Implications for the Hubble Constant},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/U7TOT6DJ}},
  note         = {Machine review of arXiv:1908.05625}
}
abstract

NGC 4258 is a critical galaxy for establishing the extragalactic distance scale and estimating the Hubble constant (Ho). Water masers in the nucleus of the galaxy orbit about its supermassive black hole, and very long baseline interferometric observations of their positions, velocities, and accelerations can be modeled to give a geometric estimate of the angular-diameter distance to the galaxy. We have improved the technique to obtain model parameter values, reducing both statistical and systematic uncertainties compared to previous analyses. We find the distance to NGC 4258 to be 7.576 +/- 0.082 (stat.) +/- 0.076 (sys.) Mpc. Using this as the sole source of calibration of the Cepheid-SN Ia distance ladder results in Ho = 72.0 +/- 1.9 km/s/Mpc, and in concert with geometric distances from Milky Way parallaxes and detached eclipsing binaries in the LMC we find Ho = 73.5 +/- 1.4 km/s/Mpc. The improved distance to NGC 4258 also provides a new calibration of the tip of the red giant branch of M_{F814W} = -4.01 +/- 0.04$ mag, with reduced systematic errors for the determination of Ho compared to the LMC-based calibration, because it is measured on the same Hubble Space Telescope photometric system and through similarly low extinction as SN Ia host halos. The result is Ho = 71.1 +/- 1.9 km/s/Mpc, in good agreement with the result from the Cepheid route, and there is no difference in Ho when using the same calibration from NGC 4258 and the same SN Ia Hubble diagram intercept to start and end both distance ladders.

Figures

Figures reproduced from arXiv: 1908.05625 by the authors.

Figure 1
Figure 1. — Marginalized probability densities for selected parameters: distance (D), black hole mass (Mbh), and error floors for the eastward (σx) and northward (σy) positions, the high (σv,hv) and systemic (σv,sys) velocities, and the accelerations (σa) [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

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

  1. JWST Detects a Dusty AGB-like Source Before the Type Ia-CSM Supernova 2026sqf

    astro-ph.SR 2026-08 conditional novelty 7.0 of 10

    Pre-explosion JWST and HST images reveal a dusty, luminous AGB-like point source at the site of SN 2026sqf, the first candidate progenitor system for a Type Ia-CSM supernova.

  2. Hubble Space Telescope Observations of Mira Variables in the Type Ia Supernova Host NGC 1559: An Alternative Candle to Measure the Hubble Constant

    astro-ph.CO 2019-08 conditional novelty 7.0 of 10

    The first Mira-variable-based calibration of a Type Ia supernova yields a distance to NGC 1559 and a Hubble constant of about 73 km/s/Mpc, consistent with Cepheid-based values.

  3. The potential broader complex linked to the key Cepheid SV Vul

    astro-ph.SR 2026-07 conditional novelty 5.0 of 10

    SV Vul and its host cluster Alicante 13 lie at 2.30 ± 0.13 kpc and may share a common origin with the nearby cluster Liu-Pang 1738.

  4. Constraints on Dark Energy Models Using Late Universe Probes

    astro-ph.CO 2025-06 conditional novelty 4.0 of 10

    Using late-universe probes only, the authors find all dark energy models remain within 1-2 sigma of Lambda CDM, which wins the Bayesian model comparison, and confirm that DESI's LRG1 and LRG2 points drive the dynamica...

  5. Cosmology With Low-Redshift Observations: No Signal For New Physics

    astro-ph.CO 2019-08 conditional novelty 4.0 of 10

    Combining SN, BAO, lensing, cosmic chronometer, and growth data, the authors find ΛCDM remains the best dark energy model, with H0 and S8 consistent with most early and late universe probes within 2σ.

Reference graph

Works this paper leans on

24 extracted references · 22 canonical work pages · cited by 5 Pith papers

  1. [1]

    L., Greenhill, L

    Argon, A. L., Greenhill, L. J., Reid, M. J., Moran, J. M. & Humphreys, E. M. L. 2007, , 659, 1040

    work page 2007

  2. [2]

    R., Parent, E., Phillips, M

    Burns, C. R., Parent, E., Phillips, M. M., , , 869, 56

    work page

  3. [3]

    L., Madore, B

    Freedman, W. L., Madore, B. F., Hatt, D. \ 2019, , 882, 34

    work page 2019

  4. [4]

    R., Moran, J

    Herrnstein, J. R., Moran, J. M., Greenhill, L. J., \ 1999, , 400, 539

    work page 1999

  5. [5]

    Hoffman, S. L. 2013, PhD Thesis, Texas A & M University, 164

    work page 2013

  6. [6]

    D., Riess, A

    Huang, C. D., Riess, A. G., Hoffmann, S. L. \ 2018, , 857, 67

    work page 2018

  7. [7]

    Humphreys, E. M. L., Reid, M. J., Greenhill, L. J., Moran, J. M. & Argon, A. L. 2007, , 672, 800

    work page 2007

  8. [8]

    Humphreys, E. M. L., Reid, M. J., Moran, J. M., Greenhill, L. J. & Argon, A. L. 2013, , 775, 13

    work page 2013

Show all 24 references
  1. [9]

    Jang, I. S. & Lee, M. G. 2017, , 836, 74

    2017

  2. [10]

    D., Scolnic, D., Riess, A

    Kenworthy, W. D., Scolnic, D., Riess, A. 2019, , 875, 145

    2019

  3. [11]

    M., Stanek, K

    Macri, L. M., Stanek, K. Z., Bersier, D., Greenhill, L. J. & Reid, M. J. 2006, , 652, 1133

    2006

  4. [12]

    F., Mager, V

    Madore, B. F., Mager, V. A. & Freedman, W. L. 2009, , 690, 389

    2009

  5. [13]

    A., Madore, B

    Mager, V. A., Madore, B. F. & Freedman, W. L. 2008, , 689, 721

    2008

  6. [14]

    Newman, J

    Maoz, E. Newman, J. A., Ferrarese, L. \ 1999, , 401, 351

    1999

  7. [15]

    & Scranton, R

    M\'enard, B., Kilbinger, M. & Scranton, R. 2010, , 406, 1815

    2010

  8. [16]

    & Richards, G

    M\'enard, B., Scranton, R., Fukugita, M. & Richards, G. 2010, , 405, 1025

    2010

  9. [17]

    1995, , 373, 127

    Miyoshi, M., Moran, J., Herrnstein, J. 1995, , 373, 127

    1995

  10. [18]

    \ 2019, , 567, 200

    Pietrzy\'nski, G., Graczyk, D., Gallenne, A. \ 2019, , 567, 200

    2019

  11. [19]

    Planck, 2018, Results VI Cosmological Parameters, arXiv:1807.06209

    2018 arXiv

  12. [20]

    G., Macri, L

    Riess, A. G., Macri, L. M., Hoffmann, S. L., \ 2016, , 826, 56

    2016

  13. [21]

    G., Casertano, S., Yuan, W., Macri, L

    Riess, A. G., Casertano, S., Yuan, W., Macri, L. M. & Scolnic, D. 2019, , 876, 85

    2019

  14. [22]

    2015, arXiv:1906:09077

    Roe, B. 2015, arXiv:1906:09077

    2015

  15. [23]

    Scolnic, D., Casertano, S., Riess, A., \ 2015, , 815, 117

    2015

  16. [24]

    G., Macri, L

    Yuan, W., Riess, A. G., Macri, L. M. \ 2019, arXiv1908.00993

    2019 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.