Cepheid, TRGB and JAGB calibration of SN Ia with JWST: H₀ distance-ladder systematics at 1%
Both camps have JWST data and the disagreement is localised to named choices (SN sample, TRGB/JAGB calibration, zero points), so a path exists without new theory; difficulty sits 10 below the old Hubble tension (50), which also contained the theory half. Leverage is raised to 50, equal to the old Hubble tension's leverage: whether the ladder settles near 73 or near 68 decides whether the theory half (sound-horizon and late-time models) is needed at all. Application value 8 is unchanged: the output is the distance scale.
The problem
Measure how fast the universe expands today to 1% using a chain of distance yardsticks, and find why teams disagree.
- 1Geometry gives distances to nearby stars
- 2Pulsating and red-giant stars calibrate brightness
- 3Exploding stars reach far galaxies
- 4Teams get answers a few per cent apart
- 5Trace the gap to specific errors
The measurement. The Hubble constant H₀ is the ratio of the recession speed of a galaxy to its distance. The local distance ladder builds distances in steps. Geometric anchors (masers in the galaxy NGC 4258, parallaxes of stars in the Milky Way, and eclipsing binaries in the Large Magellanic Cloud) give absolute distances. Then standard candles are calibrated on them: Cepheid variables, whose pulsation period relates to luminosity (the period–luminosity relation), the tip of the red giant branch (TRGB), a sharp edge in the brightness of old red giants, and J-region asymptotic giant branch (JAGB) carbon stars. These calibrate Type Ia supernovae in nearby galaxies, which are then used out to distances where the expansion dominates.
Built from
earlier problems → this oneThe problems this one is built from: progress or tools it needs (builds on), special cases that come first (generalizes), and weaker results on the way (implies). On the right, what builds on it.
- Start with
- Cepheid, TRGB and JAGB calibration of SN Ia with JWST: H₀ distance-ladder systematics at 1%This problem
- Then
Path to understanding
foundations → problemThe research entries this problem rests on, in the order to read them. Each links to its explainer.
- Start with
- Step 2
- Step 3
- ΛCDM, the standard model of cosmologyTheory · host
- Then
- Cepheid, TRGB and JAGB calibration of SN Ia with JWST: H₀ distance-ladder systematics at 1%The problem
State of the art
SH0ES (Riess et al. 2022, arXiv:2112.04510) gave H₀ = 73.04 ± 1.04 km/s/Mpc. JWST photometry of more than 1,000 Cepheids rejected unrecognised HST crowding as the cause at 8.2σ (arXiv:2401.04773), and a September 2025 SH0ES update with JWST Cepheids in 19 hosts gives 73.49 ± 0.93, or 73.18 ± 0.88 when 35 TRGB calibrations are added (Riess et al., arXiv:2509.01667). The CCHP finds lower values with other calibrators: 70.39 ± 1.22 (stat) ± 1.33 (sys) ± 0.70 (SN) from HST+JWST TRGB, 68.81 ± 1.79 ± 1.32 from JWST-only TRGB and 67.80 ± 2.17 ± 1.64 from JWST-only JAGB (arXiv:2408.06153; ApJ 985, 203, 2025); a blinded March 2025 JWST TRGB update with ten hosts and eleven calibrator SNe gives 68.4–69.6 depending on which of four SN photometry sets is used, and shows that the Pantheon+ calibration alone shifts H₀ by +2 km/s/Mpc when the sample is extended to 24 SNe (arXiv:2503.11769). The H0DN community 'Local Distance Network' (arXiv:2510.23823; A&A 708, A166, 2026) combines Cepheids, TRGB, Miras, masers, SBF, Tully–Fisher and other rungs and finds a baseline H₀ = 73.50 ± 0.81 km/s/Mpc, 7.1σ from Planck+SPT+ACT ΛCDM; it reports that removing either Cepheids or TRGB changes little, but it is a covariance-weighted combination of existing calibrator data (largely the same data SH0ES and CCHP use), not a new independent measurement, and the CCHP-favoured lower values have not been reconciled with it. The early-universe values are 66.7–68.4 (SPT-3G alone 66.66 ± 0.60 and SPT+ACT+Planck 67.19 ± 0.38, arXiv:2506.20707; P-ACT with DESI DR2 68.43 ± 0.27, arXiv:2503.14452). The June 2026 decade review (Cai & Wang, arXiv:2606.20434) states the tension remains unresolved. No claim of a resolution was found as of October 2026.
Realistic applications
A 1%-controlled ladder is the reference against which every CMB-based or BAO-based H₀ inference in ΛCDM and its extensions is tested, and it fixes the cosmic-distance scale used by supernova, lensing and standard-siren cosmology. The photometric methods (crowded-field JWST photometry, TRGB edge calibration) transfer to other distance-indicator work. There is no consumer application.
What counts as solved
Resolved when a blinded, end-to-end ladder using at least two independent calibrator rungs (for example Cepheid and TRGB or JAGB) on a common SN Ia host sample returns H₀ values that agree within a quoted ≤1% total uncertainty, with the systematic budget published term by term and reproduced by an independent group. Either outcome counts: agreement near 73 or near 68.
How it gets verified
ObservationSettled by new measurements or observations.
Settled by new JWST observations of Cepheids, TRGB and JAGB in SN Ia host galaxies. A blinded end-to-end ladder with at least two independent calibrator rungs must agree within a quoted total uncertainty of 1% or better, with the systematic budget published term by term and reproduced by an independent group.
Claims under verification
+ add a claimNo claimed results on the Verification board for this problem yet.
Where the work is
The work is dominated by new observations and re-reductions: more JWST calibrator hosts and blinded re-analysis. Engineering is detector calibration, and compute is photometric pipelines and hierarchical fits.
Bottlenecks
ranked by votes- #1EmpiricalCalibrator sample size and rung overlap
Only 10–25 SN Ia hosts have JWST TRGB, JAGB or Cepheid data in common, and different selection of those hosts shifts H₀ by about 1–2 km/s/Mpc. The dependence of the answer on the SN sample is as large as the statistical error.
Community grades
Sign in to add your own grade.
Connected problems
Progress
- Oct 27, 2025H0DN Local Distance Networksource
Community fit gives 73.50 ± 0.81 km/s/Mpc; published in A&A in 2026.
- Sep 1, 2025SH0ES JWST Cepheids in 19 hostssource
H₀ = 73.49 ± 0.93, or 73.18 ± 0.88 with 35 TRGB calibrations added.
- Mar 14, 2025CCHP JWST TRGB updatesource
H₀ of 68.4–69.6 depending on the supernova sample, from ten SN hosts.
- Jan 9, 2024JWST rejects crowding biassource
JWST photometry of more than 1,000 Cepheids rejects unrecognised HST crowding as the cause at 8.2σ.