arXiv subjects

A. Herle

Publications and source records attributed to A. Herle.

4 recordsLinked to original sources

arXiv2026

Assembly bias and the redshift evolution of intrinsic alignments for LRGs

The intrinsic alignment (IA) of galaxies is one of the main contaminants to the weak lensing shear signal. Efforts to model it often assume that the alignment strength depends only on halo mass. In this work, we use the 2.8 Gpc box-size run of the $\mathtt{FLAMINGO}$ suite to show that alignment amplitude, in addition to halo mass, depends on the formation redshift of the host haloes for an LRG-like sample. We show that the assembly histories of galaxies and haloes influence the alignment signal. After correcting for their mass evolution, we find that haloes that formed earlier have a higher alignment amplitude, as do their central galaxies. We also explore the redshift evolution of the alignment signal by fitting the amplitude with a power-law in mass at different snapshots of the simulation. We find that the amplitude of this power-law increases steadily with redshift, while the slope decreases with redshift until $z\sim1$ and then flattens. We provide an empirical mass-redshift intrinsic alignment model fit on the $\mathtt{FLAMINGO}$ simulation. Furthermore, by tracking central galaxies across snapshots, we show that the alignment signal changes with redshift beyond that associated with the change in mass, and that galaxies tracked from higher redshifts have a larger amplitude. Our results indicate that IA modeling in weak lensing surveys cannot have arbitrarily small prior ranges, and complicate the implementation of HOD-based alignment models for gravity-only simulations. They also provide simulation-based guidelines for a redshift evolution model of IA for use in observational studies.

astro-ph.CO↗

arXiv2026

Intrinsic alignments in the FLAMINGO simulations with two-point statistics

Intrinsic alignments are a major astrophysical contaminant for next generation large-sky surveys like Euclid and LSST. Large hydrodynamic simulations are crucial for informing the alignment modelling for these surveys. We measure position-position and position-shape correlations of a Luminous Red Galaxy sample from the FLAMINGO suite of hydrodynamical simulations, measuring the alignment signal for more than 4.9 million galaxies at redshift 0. We jointly model the clustering and alignment correlations to provide the tightest constraints on the alignment amplitude to date from a hydrodynamic simulation. We find that both the Non-Linear Alignment (NLA) and the more complex Tidal Alignment Tidal Torquing (TATT) models provide good fits to the data. We compare the measured $A_1$ amplitude to observational data and find good agreement. We measure the dependence of the NLA and TATT free parameters on halo mass. We also introduce a mass-dependent TATT model, TATT-M, by finding empirical relations between the halo mass and the TATT parameters. This allows us to fit TATT with only one parameter, $A_1$, with $A_2/A_1$ being a constant and $A_{1δ}/A_1$ being a function of halo mass. Using a Bayesian approach, we find that TATT-M is very strongly preferred by the data over NLA. Using the baryonic feedback variations of the FLAMINGO simulation suite, we test whether the TATT parameters are sensitive to feedback. Variations in AGN and supernova feedback do not significantly change the alignment amplitude beyond the change associated with the dependence of galaxy stellar mass on the strength of feedback. Our results inform the IA modelling for upcoming surveys by providing guidance on model choices, priors and sensitivities to feedback.

astro-ph.CO↗

arXiv2025

Unbiased estimates of the shapes of haloes using the positions of satellite galaxies

The shapes of dark matter haloes are sensitive to both cosmology and baryon physics, but are difficult to measure observationally. A promising way to constrain them is to use the positions of satellite galaxies as tracers of the underlying dark matter, but there are typically too few galaxies per halo for reliable shape estimates, resulting in biased shapes. We present a method to model sampling noise to correct for the shape bias. We compare our predicted median shape bias with that obtained from the FLAMINGO suite of simulations and find reasonable agreement. We check that our results are robust to resolution effects and baryonic feedback. We also explore the validity of our bias correction at various redshifts and we discuss how our method might be applied to observations in the future. We show that median projected halo axis ratios are on average biased low by 0.31 when they are traced by only 5 satellites. Using the satellite galaxies, the projected host halo axis ratio can be corrected with a residual bias of ~ 0.1, by accounting for sampling bias. Hence, about two-thirds of the projected axis ratio bias can be explained by sampling noise. This enables the statistical measurement of halo shapes at lower masses than previously possible. Our method will also allow improved estimates of halo shapes in cosmological simulations using fewer particles than currently required.

astro-ph.CO↗

arXiv2023

Selection functions of strong lens finding neural networks

Convolution Neural Networks trained for the task of lens finding with similar architecture and training data as is commonly found in the literature are biased classifiers. An understanding of the selection function of lens finding neural networks will be key to fully realising the potential of the large samples of strong gravitational lens systems that will be found in upcoming wide-field surveys. We use three training datasets, representative of those used to train galaxy-galaxy and galaxy-quasar lens finding neural networks. The networks preferentially select systems with larger Einstein radii and larger sources with more concentrated source-light distributions. Increasing the detection significance threshold to 12$σ$ from 8$σ$ results in 50 per cent of the selected strong lens systems having Einstein radii $θ_\mathrm{E}$ $\ge$ 1.04 arcsec from $θ_\mathrm{E}$ $\ge$ 0.879 arcsec, source radii $R_S$ $\ge$ 0.194 arcsec from $R_S$ $\ge$ 0.178 arcsec and source Sérsic indices $n_{\mathrm{Sc}}^{\mathrm{S}}$ $\ge$ 2.62 from $n_{\mathrm{Sc}}^{\mathrm{S}}$ $\ge$ 2.55. The model trained to find lensed quasars shows a stronger preference for higher lens ellipticities than those trained to find lensed galaxies. The selection function is independent of the slope of the power-law of the mass profiles, hence measurements of this quantity will be unaffected. The lens finder selection function reinforces that of the lensing cross-section, and thus we expect our findings to be a general result for all galaxy-galaxy and galaxy-quasar lens finding neural networks.

astro-ph.CO↗