Cosmology

Hubble Tension

Disagreement between local and cosmological measurements of the universe's expansion rate

The Hubble tension is the persistent disagreement between two methods of measuring the Hubble constant H₀: (1) Local distance ladder (Cepheids → Type Ia SN) gives H₀ ≈ 73 km/s/Mpc. (2) Early-universe CMB measurements (Planck) give H₀ ≈ 67.4 km/s/Mpc. Discrepancy: ~5σ — too large to be statistical fluke. Implies either (a) systematic errors in one method, or (b) new physics — modification of standard cosmological model. Active area of research; potentially most important problem in cosmology today.

  • Local H₀73.0 ± 1.0 km/s/Mpc (Cepheid + SN)
  • CMB H₀67.4 ± 0.5 km/s/Mpc (Planck)
  • Discrepancy~5σ — beyond statistical fluke
  • ImplicationsSystematic errors OR new physics
  • New physics optionsEarly dark energy, sterile neutrinos, evolving DE
  • Future testsJWST, LSST, Roman, more SN

Interactive visualization

Press play, or step through manually. The visualization is yours to drive — try it before reading on.

Open visualization fullscreen ↗

Watch the 60-second explainer

A condensed visual walkthrough — narrated, captioned, under a minute.

Why Hubble tension matters

  • Cosmological models. Tests ΛCDM.
  • New physics search. Possible signature.
  • Distance ladder. Cross-validation across methods.
  • Universe expansion. Direct measurement.
  • Future surveys. Major target for upcoming missions.
  • Theoretical models. Discriminate among proposals.
  • Cosmology future. Resolution determines next steps.

Common misconceptions

  • One measurement is right. Both rigorous; difference unclear.
  • Tension is small. 5σ is significant.
  • It's clearly new physics. Could be systematic; uncertain.
  • Tension solved. Active research; not resolved.
  • Hubble constant is well-defined. Value depends on cosmological epoch.
  • Tension affects all of cosmology. Specific to expansion rate; rest of cosmology robust.

Frequently asked questions

How is H₀ measured locally?

Distance ladder. (1) Geometric parallax for nearby stars. (2) Cepheid variable stars — period-luminosity relation calibrated by parallaxes. (3) Type Ia supernovae — calibrated using Cepheids in same galaxy. (4) Hubble's law — v = H₀ d for distant SN. Adam Riess's SH0ES collaboration: H₀ = 73.0 ± 1.0.

How is H₀ measured from CMB?

Planck mission (2013). Power spectrum of CMB anisotropies fit with cosmological model (ΛCDM). H₀ derived. Result: H₀ = 67.4 ± 0.5 km/s/Mpc. Highly precise but indirect — depends on cosmological model assumptions. Independent of local distance ladder.

Why is the difference so significant?

Both measurements are systematic-error-controlled and independently confirmed. Both have multiple cross-checks. ~5σ discrepancy. If real: indicates new physics. If not: indicates one or both methods have unrecognized systematic errors.

What new physics could explain?

Possibilities. (1) Early dark energy — additional energy in early universe, makes H₀ from CMB lower. (2) Sterile neutrinos — affect early universe expansion. (3) Modified gravity — affects distance scales. (4) Evolving dark energy — w(z) varies. None definitively confirmed; multiple proposals.

Could one measurement be wrong?

Yes — extensively studied. Local: dust extinction, Cepheid metallicity, SN systematic errors. CMB: assumes ΛCDM model, neutrino mass, recombination physics. Independent measurements (TRGB stars, masers, BAO) help — generally support tension exists.

What's TRGB?

Tip of Red Giant Branch — alternative distance indicator. Tip luminosity is constant (calibratable). Wendy Freedman's CCHP collaboration: H₀ = 70 km/s/Mpc, intermediate. Doesn't fully resolve tension. Difference may be statistical — not as decisive.

Is this an existential cosmology problem?

Major problem; "crisis" is overstated. ΛCDM fits remarkably well overall. Could be: systematic in one method, modest cosmological extension, or revolutionary new physics. JWST observations of Cepheids (2023+) tightening local H₀; CMB-S4 will tighten CMB H₀. Coming years: clarification expected.