Baryonic feedback in the small-scale matter power spectrum for cosmic shear: the S₈ systematic budget
With KiDS-Legacy, HSC Y3 and DES Y6 between 0 and 2.6σ from the CMB, the tension question has shrunk, but percent-level P(k) feedback calibration is a harder modelling task than the old S₈ entry (23) captured; still a visible route. Leverage is raised to 42, above the old S₈ entry (26.5): every Stage-IV cosmic-shear result (Euclid, LSST, Roman), including dark-energy constraints, depends on this calibration. Application value 10 because the emulators are pipeline infrastructure for those missions.
The problem
Pin down how much exploding stars and black-hole winds smooth out matter, to see if the 'clumpiness tension' is real.
- 1Early glow predicts how clumpy matter is
- 2Lensing finds matter a bit smoother
- 3Winds from stars and black holes smooth it
- 4Calibrate smoothing with gas observations
- 5See how much of the gap remains
The tension. S8 = σ₈ √(Ω_m/0.3) is a combination that weak lensing measures well; σ₈ is the amplitude of matter fluctuations on scales of about 8 Mpc/h. Cosmic shear surveys (KiDS-Legacy, HSC Y3, DES Y6) find S8 lower than the CMB prediction in ΛCDM by 0 to 2.6σ depending on survey and analysis.
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
- Baryonic feedback in the small-scale matter power spectrum for cosmic shear: the S₈ systematic budgetThis 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
- Baryonic feedback in the small-scale matter power spectrum for cosmic shear: the S₈ systematic budgetThe problem
State of the art
Amon & Efstathiou (arXiv:2206.11794) argued that a strong, scale-dependent suppression of the nonlinear matter power spectrum would remove the S₈ tension, and Preston, Amon & Efstathiou (arXiv:2305.09827) found that DES Y3 cosmic shear prefers more suppression than most hydrodynamic simulations produce. The FLAMINGO simulations (arXiv:2306.04024) include variants with stronger feedback calibrated to cluster gas fractions. Bigwood et al. (arXiv:2404.06098) combined weak lensing with kSZ measurements and found that data favour strong feedback, which can reduce the S₈ discrepancy, though the interpretation depends on the baryonification model. The survey numbers have converged upward. KiDS-Legacy (Wright et al., arXiv:2503.19441; A&A 703, A158, 2025) found S₈ = 0.815 (+0.016/−0.021), 0.73σ from Planck, attributing the rise mainly to improved redshift calibration, and it marginalises over baryon feedback. HSC Y3 with DESI clustering-based redshift calibration moves from 0.769 to 0.804 (+0.019/−0.018) (arXiv:2511.18134). DES Y6 cosmic shear gives 0.798 (+0.014/−0.015) with the NLA alignment model, about 2σ below the CMB (arXiv:2602.10065, February 2026), and DES Y6 3×2pt gives 0.789 ± 0.012, 2.6σ in S₈ alone (arXiv:2601.14559, January 2026). So the residual is 0–2.6σ depending on the survey, comparable to the spread from alignment and redshift choices, and the share attributable to feedback rather than to calibration is not settled. Euclid DR1 cosmology is not expected before mid-2027 and Rubin/LSST has no cosmic-shear results as of October 2026.
Realistic applications
A calibrated feedback model is required for every Stage-IV weak-lensing cosmology analysis (Euclid, LSST, Roman); an error of a few percent in P(k) at k = 5 h/Mpc is comparable to the target statistical errors. The emulators also constrain AGN feedback physics in galaxy formation.
What counts as solved
Resolved when a feedback model calibrated only on non-lensing gas observables predicts the nonlinear P(k) to within about 1–2% and the Stage-IV surveys (Euclid DR1, LSST Y1), with feedback marginalised using that model, give S₈ consistent with the CMB with this model, or when they give a residual discrepancy above 3σ after marginalising over it.
How it gets verified
ObservationSettled by new measurements or observations.
Settled by Stage-IV survey data (Euclid DR1, LSST Y1) together with gas observables: a feedback model calibrated only on non-lensing observables must predict the nonlinear P(k) to about 1 to 2%, and the resulting S8 must either match the CMB or retain a residual discrepancy above 3 sigma after marginalisation.
Claims under verification
+ add a claimNo claimed results on the Verification board for this problem yet.
Where the work is
Mostly simulations and emulator calibration, plus joint analysis with SZ and X-ray data.
Bottlenecks
ranked by votes- #1ComputeCalibrating feedback across simulations
Hydrodynamic simulations disagree on the P(k) suppression by several percent at k ≈ 5 h/Mpc, with sub-grid physics unconstrained by first principles.
Community grades
Sign in to add your own grade.
Connected problems
Progress
- Feb 10, 2026DES Y6 cosmic shearsource
S₈ = 0.798 (+0.014/−0.015), about 2σ below the CMB.
- Mar 25, 2025KiDS-Legacy cosmic shearsource
Final KiDS analysis finds S₈ consistent with Planck.
- Apr 9, 2024Lensing plus kSZ constraintssource
Joint analysis favours strong feedback.
- Jun 23, 2022Nonlinear suppression proposedsource
Amon & Efstathiou argue enhanced feedback can resolve the S₈ tension.