Flip status related to project 7

 

 

 

 

PV telecon march 2026

 

 

ACCEL² simulation

Corentin Ravoux - CNRS/LPCA Clermont-Ferrand

 

Flip structure evolution

  • flip.data: contains data for tests and examples

  • flip.data_vector: Modeling of different fields (vel, density, mesh…)

  • flip.power_spectra: Utilities for power spectra generation

  • flip.covariance: Main method, covariance field-level inference mainly targeting at growth rate

  • flip.comparison (mesh branch): Project for inference by field comparison (uses data_vector too)

  • flip.simulation (for later): might include forward modeling or SBI tests

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Non linear models in flip.covariance

  • Main flip modeling issues:

    • Unclear σu modeling + correlation with fσ8

    • Close SNe Ia breaks linearity assumptions (especially for LSST)

  • Start of project 7: Improvement of peculiar velocity modeling

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

"Simple" linear ZTF fit

P_{\mathrm{vv}} = (f\sigma_8)^2 \times \left(a^2 H^2 \frac{\mu^2}{k^2}\right) \times \left(P_{\mathrm{\theta\theta}}(k) D_u^2(k, \sigma_u)\right)

Non linear models: Dam et al. 2021

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

P_{\mathrm{\theta\theta}}(k)

  • 1-loop correction

  • Redshift-space velocity divergence power spectrum

  • Addition of multiple moments distortions  in
  • velocity power spectrum:
  • Still adding damping term:

Non linear models: Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Eulerian perturbation theory approach, fully developed model

Vlah et al. (2012, 2013); Okumura et al. (2014); Saito et al. (2014); Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Velocity-side only

  • Parameters in addition to growth rate (f):
    • Biases (b_1, b_2, b_s, b_3nl), some bias relation can reduce this
    • Velocity dispersion terms:
      • σv,1 in P12, P22, P13 vector part
      • σv,2 in P13 scalar part

Non linear models: Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Flip factorization

P_{\mathrm{ab}}(k, \mu_{\mathrm{a}}, \mu_{\mathrm{b}}) = \sum_{i} w_{\mathrm{ab},n} Z_{\mathrm{ab},n}(z_{\mathrm{a}}, z_{\mathrm{b}}) F_{\mathrm{ab},n} (k, \mu_{\mathrm{a}}, \mu_{\mathrm{b}}) \mathcal{P}_{\mathrm{ab},n}(k)

Parameters to fit

Terms to integrate analytically

Power spectra terms to integrate numerically

Redshift dependent terms

Non linear models: Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Flip factorization

P_{\mathrm{ab}}(k, \mu_{\mathrm{a}}, \mu_{\mathrm{b}}) = \sum_{i} w_{\mathrm{ab},n} Z_{\mathrm{ab},n}(z_{\mathrm{a}}, z_{\mathrm{b}}) F_{\mathrm{ab},n} (k, \mu_{\mathrm{a}}, \mu_{\mathrm{b}}) \mathcal{P}_{\mathrm{ab},n}(k)

Example:

Non linear models: Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Model fully developed in flip format for vv, vg and gg correlations

Non linear models: Howlett et al. 2019

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Large number of covariance terms to integrate: vv (56), vg (115) and gg (124)
  • A lot of variant implemented: biases links, different velocity dispersions between density & velocities
  • Ixx, Jxx, Kxx and Ksxx terms computed thanks to a SPT code (https://github.com/FeiQin-cosmologist/Galaxy_Power_Spectrum)

Non linear models: Howlett et al. 2019

Flip implementation

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Very similar to the linear model, with change in the input power spectra

What is left to do

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Test model on realistic simulations (start with ZTF)
  • Test the density model (DESI simulations ?)
  • Correct for numerical instabilities with density terms to integrate
  • Derive in wide-angle ? outch

Future step for the project 7

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Derive an EFTofLSS model
  • Chen et al. 2025: two-loop pairwise velocity power spectrum derivation
  • In contact with Stephen for the development of this model
  • Last step: model comparison with N-body sims (article ?)

Other unrelated improvements

Corentin Ravoux                                                                                                                                                              DESC PV telecon - march 2026

Text

  • Density & velocity gridding improvement 
  • Exploratory project: Galaxy vs GW density field rescaling, potential H_0 constraints similar to dark sirens (Project with IP2I team)
  • Flip to do list:
    • Improving and testing emulators
    • Improve documentation
    • Test non linear models
    • Derive likelihood with sample selection bias correction included
    • Creation of firecrown likelihood object to link flip with DESC pipeline (pyCCL already done).