Measuring the expansion of the universe.
CANDLE turns the Hubble Tension from an unresolved mystery into a definitive measurement — either confirming the standard model or revealing the first cracks in it.
Imagery · ESO / S. Brunier
Key Project
CANDLE.
Cepheid And Nearby Distance Ladder Experiment — a focused observational campaign to measure the local Hubble constant to sub-percent precision.
CANDLE is the primary observational program of the CELESTE mission. Its goal is unambiguous: measure H₀ to better than one percent accuracy, resolving the 5σ tension between local and CMB-based estimates of the cosmic expansion rate.
The core strategy is straightforward but technically demanding. CELESTE provides a 12 mas laser guide star; the Keck telescope delivers a 10-meter aperture; ORKID collects the photons. The result is diffraction-limited Cepheid photometry in galaxies out to 100 Mpc — the same galaxies where Type Ia supernovae are calibrated, but now with the crowding bias removed.
CANDLE will observe approximately 2,000 Cepheid variables across 12 target galaxies, spanning the period–luminosity relation across three decades of pulsation period. The statistical power of this sample, combined with the removal of systematic crowding bias, is projected to reduce the Cepheid-anchored H₀ uncertainty from 2.3% (SH0ES 2022) to below 1%.
The Brief
i.
Science Objectives
Measure H₀ to < 1% accuracy using a new, crowding-free Cepheid distance ladder. Provide the definitive local measurement against which CMB-based predictions are tested. Determine whether the Hubble Tension represents new physics or unaccounted systematics.
ii.
Observables & Measurements
Period–luminosity relations for ~2,000 Cepheid variables in 12 galaxies out to 100 Mpc. Multi-epoch R and I-band light curves with cadence matched to expected pulsation periods (5 – 100 days).
iii.
Key Functional Requirements
Diffraction-limited PSF at R and I bands (0.6 – 0.9 μm). Photometric stability of 0.5% across the multi-year campaign. Access to the same host galaxies as the SH0ES and CCHP supernova samples.
iv.
CELESTE Uniqueness
CANDLE is the only proposed program that removes the crowding systematic by resolving individual Cepheids in crowded spiral arms — rather than modelling or correcting for it statistically. This is only possible at CELESTE angular resolution.