FLAMINGO: Baryonic effects on the weak lensing scattering transform

TL;DR

This study evaluates how baryonic feedback alters the weak-lensing scattering transform (ST) coefficients using the FLAMINGO suite of hydrodynamical and dark-matter-only simulations. The authors define a baryonic transfer function, $\mathcal{T}=\overline{\mathcal{S}^{\mathrm{HYDRO}}}/\overline{\mathcal{S}^{\mathrm{DMO}}}$, to quantify the imprint of feedback on ST coefficients, and demonstrate that $\mathcal{T}$ is largely insensitive to cosmology but highly sensitive to baryonic physics, with scale-dependent suppression strongest at small/intermediate scales ($j\approx 1-3$) and up to several percent for extreme feedback. Shape noise and smoothing significantly diminish the detectable baryonic signal, reducing $\mathcal{T}$-induced suppression to around $\sim 1\%$ or less, highlighting the need for high-resolution data and noise-mitigation strategies in Stage IV surveys. The results support using the transfer function as a correction factor in likelihood-free inference, emulation, or Bayesian model averaging to mitigate baryonic uncertainties while preserving cosmological information, and point toward extending the formalism to tomographic analyses.

Abstract

The scattering transform is a wavelet-based statistic capable of capturing non-Gaussian features in weak lensing (WL) convergence maps and has been proven to tighten cosmological parameter constraints by accessing information beyond two-point functions. However, its application in cosmological inference requires a clear understanding of its sensitivity to astrophysical systematics, the most significant of which are baryonic effects. These processes substantially modify the matter distribution on small to intermediate scales ($k\gtrsim 0.1\,h\,\mathrm{Mpc}^{-1}$), leaving scale-dependent imprints on the WL convergence field. We systematically examine the impact of baryonic feedback on scattering coefficients using full-sky WL convergence maps with Stage IV survey characteristics, generated from the FLAMINGO simulation suite. These simulations include a broad range of feedback models, calibrated to match the observed cluster gas fraction and galaxy stellar mass function, including systematically shifted variations, and incorporating either thermal or jet-mode AGN feedback. We characterise baryonic effects using a baryonic transfer function defined as the ratio of hydrodynamical to dark-matter-only scattering coefficients. While the coefficients themselves are sensitive to both cosmology and feedback, the transfer function remains largely insensitive to cosmology and shows a strong response to feedback, with suppression reaching up to $10\%$ on scales of $k\gtrsim 0.1\,h\,\mathrm{Mpc}^{-1}$. We also demonstrate that shape noise significantly diminishes the sensitivity of the scattering coefficients to baryonic effects, reducing the suppression from $\sim 2 - 10 \;\%$ to $\sim 1\;\%$, even with 1.5 arcmin Gaussian smoothing. This highlights the need for noise mitigation strategies and high-resolution data in future WL surveys.

Figures (11)

• Figure 1: Example of the full-sphere coverage by $5\times5$$\mathrm{deg}^2$ patches, with their centres distributed using a Fibonacci lattice. The left figure shows the sphere viewed from the equatorial direction ($x>0$, corresponding to $\varphi = 0$), while the right figure shows the view from the pole ($z>0$, $\theta = 0$). The patches are nearly non-overlapping and sample the sphere approximately uniformy.
• Figure 2: Convergence map patches for the DMO (left) and hydrodynamical (center) variations of L1_m9. Each patch covers $3.6 \times 3.6$$\text{deg}^2$ and includes structures up to $z = 3$ with a Euclid-like source redshift distribution. The right panel shows their difference, highlighting the non-Gaussian features created by baryonic effects. No noise or smoothing is applied. Maps are generated via backward ray-tracing, as described in Section \ref{['subsec:maps']}.
• Figure 3: Same as Figure \ref{['fig:diff_structures']}, but showing the moduli of the convolutions with Morlet wavelets at $(j = 4, l = 2)$, with azimuthal resolution $L = 8$. The mean value of the modulus of each convolved field is shown in the bottom-right corner of each panel. The wavelets effectively capture structural differences caused by baryonic effects, as is evident from the comparison with the right panel of Figure \ref{['fig:diff_structures']}: the regions of largest absolute values of $\Delta\kappa$ coincide, while areas with no difference remain near zero. Note that the third panel uses a different colour scale, which makes the smaller differences visible.
• Figure 4: Scattering coefficients $\mathcal{S}$ as a function of dyadic scale across all FLAMINGO cosmologies, computed from noiseless, unsmoothed convergence maps and averaged over angular orientations and patches. The top panel shows absolute values of the first-order ($\mathcal{S}_1$) and second-order ($\mathcal{S}_2$) scattering coefficients. For illustrative purposes, $\mathcal{S}_2$ were multiplied by a factor of $7$. The $\mathcal{S}_1$ coefficients correspond to the amplitudes of $\kappa$ fluctuations and resemble the convergence power spectrum, while $\mathcal{S}_2$ describe cross-scale clustering. The lower panel displays ratios relative to the fiducial D3A model (L1_m9). The green-shaded region corresponds to the cosmic variance in L1_m9. Cosmological variations affect all coefficients uniformly, leaving an approximately scale-independent imprint.
• Figure 5: Scattering coefficients $\mathcal{S}$ as a function of dyadic scale for the D3A cosmology across all FLAMINGO baryonic feedback models, computed from noiseless, unsmoothed convergence maps and averaged over angular orientations and patches. The top panel shows absolute values of the first-order ($\mathcal{S}_1$) and second-order ($\mathcal{S}_2$) coefficients. For illustrative purposes, $\mathcal{S}_2$ were multiplied by a factor of $7$. The $\mathcal{S}_1$ coefficients correspond to the amplitudes of $\kappa$ fluctuations and resemble the convergence power spectrum, while $\mathcal{S}_2$ describe cross-scale clustering. The lower panel displays ratios relative to the fiducial feedback model (L1_m9). The green-shaded region corresponds to the cosmic variance in the fiducial model. Baryonic feedback introduces scale-dependent suppression that strongly affects small scales.