Euclid preparation. Testing analytic models of galaxy intrinsic alignments in the Euclid Flagship simulation

arXiv:2601.07784 · astro-ph.CO · Submitted 2026-01-12 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: Today's paper: "Euclid preparation. Testing analytic models of galaxy intrinsic alignments in the Euclid Flagship simulation".

Jocelyn: As a diligent researcher,

Vera: First, who's behind it and why it matters.

Title and authors: Vera: So, looking at the title "Euclid preparation. Testing analytic models of galaxy intrinsic alignments in the Euclid Flagship simulation," it really tells us this work is directly aimed at preparing for the actual science phase of the Euclid mission.

Jocelyn: That’s right, and it emphasizes that they aren't just doing abstract theory; they are testing these models within a simulation specifically designed to represent what we expect Euclid to observe.

Subrahmanyan: The authors, including R. Paviot and many others from the Euclid Collaboration, are bringing together expertise from both observational cosmology and simulation work to tackle this problem head-on.

Vera: It’s interesting how they're using these established models like NLA and TATT to see if they actually match what the simulation spits out, which helps us gauge our confidence in using them later.

Jocelyn: So, essentially, this paper is laying the groundwork by vetting these analytical frameworks against a high-fidelity model of a future survey's expected data.

Subrahmanyan: This validation step is necessary because if the models don't hold up in the simulation, then any constraints derived from real Euclid data might be flawed.

Vera: That makes sense, and it sets a really strong baseline for what we can expect when we look at the actual lensing maps.

Jocelyn: It’s about ensuring that when we talk about IA contamination in our final cosmological constraints, the underlying tools are sound.

The paper's summary: Vera: In terms of what they actually found in "Euclid preparation. Testing analytic models of galaxy intrinsic alignments in the Euclid Flagship simulation," they modeled how intrinsic alignments affect the weak lensing signal using photometric properties and dark matter halos together.

Jocelyn: They compared their simulation results against two major IA models, NLA and TATT, finding that both could accurately describe the alignment signal down to scales of six–seven h−one Mpc.

Subrahmanyan: The key result here is that they found the alignment amplitudes for red galaxies were comparable to what we see in actual observations, even for samples that weren't part of their initial calibration set.

Vera: But they also looked at blue galaxies and noted that their constraints are consistent with zero alignments in the lowest redshift bin, but a signal does appear at higher redshifts.

Jocelyn: They also found something about evolution; they showed that the commonly used redshift power-law for IA doesn't work above z = one point one, suggesting we need to look closer at how things change with time.

Subrahmanyan: The paper highlights that incorporating a luminosity dependence into the modeling gives them a much better fit, showing that galaxy linear alignment is tightly linked to luminosity, which is something we already knew from prior studies but this simulation confirms it.

Vera: So, in short, they've validated the tools and found that while simple power laws fail at higher redshifts, including luminosity dependence makes the modeling much more accurate for predicting what Euclid will measure.

The paper's improvements: Jocelyn: The paper suggests several improvements for how we should approach this, specifically pointing out that the simple redshift power-law model for IA simply isn't sufficient when looking at higher redshifts, like above z = one point one.

Vera: And they stress that incorporating luminosity dependence is a substantial improvement because it captures the way intrinsic luminosity evolves in magnitude-limited surveys, which is what Euclid will be observing.

Subrahmanyan: This suggests that for future work, we should focus on linking the IA signal more directly to halo mass evolution through a single power-law model, as this seems to be what drives that luminosity dependence.

Jocelyn: They also mention that they've seen how this luminosity evolution might depend on the morphological properties of red samples, which adds another layer of complexity we need to consider.

Vera: And from an observational data side, the paper suggests using these results to set conservative scale cuts for TATT models, like setting a minimum scale cut around R min = six–seven h-one Mpc during likelihood fits.

Subrahmanyan: That scale-dependent prior is smart because it prevents the AI from overfitting to small-scale noise where the nonlinear modeling assumptions are less certain, which is a good engineering constraint.

Jocelyn: So, the suggested improvements move us toward a system that doesn't just use one model but uses context—redshift, luminosity, and scale—to constrain IA better.

Conclusion: Vera: So to wrap up "Euclid preparation. Testing analytic models of galaxy intrinsic alignments in the Euclid Flagship simulation," they successfully validated NLA and TATT models against a realistic simulation setup to check their performance on the expected Euclid data.

Jocelyn: The main implication is that while we need more complex modeling than simple power laws, incorporating luminosity dependence is essential for accurately forecasting contamination across different redshift bins in our weak lensing analyses.

Subrahmanyan: This work solidifies the idea that linking IA to halo mass evolution via luminosity dependence provides a more physically motivated description of how alignments change over cosmic time compared to simpler models.

Vera: It sets a clear path forward for using these IA predictions as reliable nuisance terms when we run our cosmological inference pipelines for Euclid data.

Jocelyn: It’s encouraging because it gives us concrete parameters and scale priors to use in the actual analysis phase, making the forecasting process much more robust.

Subrahmanyan: This paper provides a solid foundation for understanding how galaxy properties influence the large-scale structure we see through weak lensing, which is fundamental to our entire cosmic picture.

R. Paviot, B. Joachimi, K. Hoffmann, S. Codis, I. Tutusaus, D. Navarro-Gironés, J. Blazek, F. Hervas-Peters, B. Altieri, S. Andreon, N. Auricchio, C. Baccigalupi, M., Baldi, S., Bardelli, S., A. Biviano, E., Branchini, E., Brescia, M., Camera, S., Cañas-Herrera, G, Capobianco, Carbone, Cardone, Carretero, Castander et al., 2025, Castignani, G, Cavuoti, S., Chambers, K. C., Cimatti, A., Colodro-Conde, C., Congedo, G, Conversi, L, Copin, Y., Courbin, F, Courtois, H. M., Da Silva, A, Degaudenzi, H, de la Torre, S de la, De Lucia, G de , Dole, H, Dubath, F, Duncan C. A. J., Dusini S., Escoffier S., Farina M., Farinelli R, Farrens S, Ferriol F, Finelli F, Fosalba P., Frailis M, Franceschi E, Galeotta S Galeotta K George B Gillis C Giocoli J Gracia-Carpio A Grazian F Grupp S V. H. Haugan H Hoekstra W Holmes F Hormuth A Hornstrup K Jahnke M Jhabvala E Keihänen S Kermiche A Kiessling M Kilbinger B Kubik M Kümmel M Kunz H Kurki-Suonio, Le Brun A. M. C., Ligori S, Lilje P. B., Lindholm V., Lloro I., Mainetti G Maino D Maino E Maiorano O Mansutti S Marcin O Marggraf M Martinelli N Martinet F Marulli R Massey J Maurogordato E Medinaceli S Mei Y Mellier M Meneghetti E Merlin G Meylan A Mora, Moresco M L Moscardini C Neissner S.-M. Niemi C Padilla S Paltani F Pasian K Pedersen V Pettorino S Pires G Polenta M Poncet L A Popa L Pozzetti F Raison R Rebolo A Renzi J Rhodes G Riccio E Romelli M Roncarelli R Saglia Z Sakr A G Sánchez D Sapone B Sartoris T Schrabback A Secroun E Sefusatti, Seidel G Serrano P Simon C Sirignano G Sirri A Spurio Mancini L Stanco J Steinwagner P Tallada-Crespí A N. Taylor I Tereno N Tessore S Toft R Toledo-Moreo F Torradeflot L Valenziano J Valiviita T Vassallo G Verdoes Kleijn A Veropalumbo, Wang Y., Weller J., Zacchei A, Zamorani G, Zerbi F M. Zerbi E Zucca E Bozzo C Burigana R Cabanac M Calabrese J A Escartin Vigo L Gabarra W G Hartley S Matthew M Maturi N Mauri R B Metcalf A Pezzotta M Pöntinen C Porciani I Risso V Scottez, Sereno M Tenti M Viel, Wiesmann S Y Akrami I T. Andika, Anselmi S, Archidiacono F Atrio-Barandela D Bertacca B Bethermin A Blanchard L Blot H Böhringer M Bonici S Borgani, Calabro A Camacho Quevedo F Caro C. S. Carvalho T Castro F Cogato, Conseil A R Cooray O Cucciati S Davini F De Paolis G Desprez A Díaz-Sánchez J J Diaz S Di Domizio P Dimauro M Y. Elkhashab A Enia Y Fang, Ferrari A Finoguenov K Ganga J García-Bellido T Gasparetto V Gautard R Gavazzi E Gaztanaga F Giacomini F Gianotti G Gozaliasl M Guidi C Gutierrez A Hall S Hemmati H Hildebrandt J Hjorth J. J. E. Kajava Y Kang V Kansal D Karagiannis K Kiiveri J Kim C C Kirkpatrick S Kruk, Le Graet L Legrand M Lembo F Lepori G Leroy G F Lesci T I Liaudat A Loureiro J Macias-Perez G Maggio M Magliocchetti F Mannucci R Maoli, Martins C. J. A. P., Maurin L, Miluzio P Monaco C Moretti G Morgante S Nadathur K Naidoo P Natoli A Navarro-Alsina S Nesseris L Pagano D Paoletti F Passalacqua K Paterson L Patrizii A Pisani D Potter S Quai M Radovich, Sacquegna S M Sahlén D B. Sanders E Sarpa A Schneider M Schultheis, Sciotti D Sellentin L C Smith J G Sorce K Tanidis C Tao G Testera R Teyssier S Tosi A Troja M Tucci C Valieri A Venhola D Vergani F Vernizzi G Verza, Vielzeuf P and N. A Walton, Castander et al., 2025

Euclid Collaboration

astro-ph.CO

Submitted: 2026-01-12

Updated: 2026-09-29

Journal ref: A&A A&A Volume 713, September 2026

DOI: 10.1051/0004-6361/202658933

Code: https://github.com/cosmodesi/pycorr

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 87/100

The gist: As a diligent researcher, I have meticulously reviewed and synthesized these excerpts from what appears to be a technical paper concerning Intrinsic Alignments (IA) within cosmological simulations

Key concepts

Galaxy Intrinsic Alignments (IA)
IA refers to how galaxies are intrinsically aligned with the surrounding dark matter structure. This alignment affects the weak lensing signal that Euclid will measure. The paper tests models to understand this contamination.
NLA and TATT Models
These are two major analytic models used to describe galaxy intrinsic alignments. The study compared simulation results against these models to see if they accurately predict the alignment signal in a realistic simulation.
Luminosity Dependence
The research found that incorporating how galaxy luminosity changes with redshift provides a much better fit for IA modeling than using simple redshift power laws. This dependence is linked to how intrinsic luminosity evolves in magnitude-limited surveys.

Terminology

Summary

As a diligent researcher, I have meticulously reviewed and synthesized these excerpts from what appears to be a technical paper concerning Intrinsic Alignments (IA) within cosmological simulations for weak lensing studies, specifically in preparation for Euclid observations. The information spans model implementation, calibration against observational data, comparison of IA models (NLA vs. TATT), and the impact of galaxy properties like luminosity and redshift on the IA signal.

Here is a detailed synthesis combining all provided summaries:

This research focuses on modeling Intrinsic Alignments (IA) within the Euclid Flagship simulation to rigorously investigate their potential impact on the weak lensing signal expected from the Euclid survey. The study employs a sophisticated IA implementation that accounts for both the photometric properties of galaxies and their underlying dark matter host halos.

The simulation parameters were meticulously calibrated using a combination of constraints derived from existing observational data (e.g., SDSS) and cosmological hydrodynamical simulations (specifically the Horizon-AGN simulation at z=1). The IA implementation itself is tested against two of the most widely utilized IA models: the Non-Linear Alignment (NLA) and the Tidal Alignment and Tidal Torquing (TATT) models.

The analysis measures the amplitude of this simulated IA signal as a function of galaxy magnitude and color across a broad redshift range (0.1 < z < 2.1), mirroring the characteristics of Euclid’s main galaxy sample. The simulation successfully validates both NLA and TATT models, demonstrating their capability to accurately describe the IA signal down to small physical scales, specifically r min = 6–7, h-1 Mpc.

The study provides detailed constraints across different galaxy populations:

  • Red Galaxies: The measured alignment amplitudes for red galaxies are found to be comparable to those reported in actual observations, even for samples that were not included in the initial calibration procedure.

  • Blue Galaxies: Constraints on blue galaxies show consistency with zero alignments in the lower redshift bin (0.1 < z < 0.3). However, a non-negligible signal is detected at higher redshifts, which remains consistent with existing observational upper limits.

  • Spiral Galaxies: The findings align with predictions from several hydrodynamical simulations, which also suggest alignment for spiral galaxies.

A key finding relates to the evolution of IA with redshift:

  1. General Redshift Evolution: The evolution of alignment with redshift is deemed realistic and comparable to that determined in observational studies.

  2. Power-Law Failure: A significant limitation identified is that the **commonly adopted redshift power-law for IA fails to reproduce the simulation alignments above z = 1.1 **.

  3. Luminosity Dependence: A substantially better agreement with the simulation results is achieved when a luminosity dependence is incorporated. This captures the intrinsic luminosity evolution inherent in magnitude-limited surveys, showing that galaxy linear alignment (A 1) is tightly correlated with luminosity, as previously established in prior studies.

  4. Model Equivalence: The comparison between the two primary models reveals that using the TATT model instead of the NLA model **does not significantly alter the redshift evolution of the linear alignment parameter A 1 **.

In summary, this Flagship IA simulation serves as a valuable tool for translating current IA constraints into reliable predictions regarding IA contamination in Euclid-like samples. The analysis successfully demonstrates that while simple power-law models are insufficient, incorporating luminosity dependence is crucial for accurately modeling the intrinsic evolution of galaxy alignments across the observed redshift range.

The provided excerpts also detail the broader framework used to analyze these IA signals within a full cosmological context:

The analysis incorporates a comprehensive model for the galaxy power spectrum, including contributions from linear and nonlinear matter power spectra, as implemented in packages like FASTPT. The model includes all contributions listed in Eq. (A.1).

  • Matter Spectra: The linear power spectrum is calculated using CAMB (Lewis et al. 2000), and the nonlinear matter power spectrum P delta delta(k) is estimated using the latest HMcode version (Mead et al. 2021).

  • Bias Parameters: Second-order (b s squared) and third-order (b 3nl) non-local biases are fixed based on a local Lagrangian bias prescription: b s squared = -4/7(b 1 - 1) and b 3nl = b 1 - 1 (Eq. A.2).

Improvements for AI systems

As a fastidious researcher, I have analyzed this manuscript, which focuses on testing analytical models of galaxy intrinsic alignments (IA) within Euclid's Flagship simulation to forecast its impact on weak lensing observations.

The primary improvements suggested for AI systems stem from integrating the validated IA modeling frameworks (NLA and TATT) and their derived luminosity/redshift evolution into downstream cosmological inference pipelines.

Here are the specific improvements and what they enable:

) Improved AI System Capabilities: Forecasting Cosmological Contamination in Weak Lensing Surveys

The proposed AI system, informed by this paper's results, will be a sophisticated pipeline capable of generating statistically robust forecasts for Euclid-like weak lensing surveys. It moves beyond simple simulation comparison to actively model and mitigate systematic uncertainties arising from galaxy intrinsic alignments (IA).

Here are the specific improvements:

  1. Improved AI System Capability Specific Functionality Enabled

2.:---:---

  1. Model-Informed IA Parameter Inference Engine The system will integrate the validated analytical IA models (NLA and TATT) directly into likelihood analyses (as described in Section 4). This allows the AI to perform parameter inference on observational data vectors, specifically extracting constraints on key IA parameters:

4.:---:---

  1. Luminosity-Dependent Bias Modeling Module The system will incorporate the derived luminosity dependence of the IA amplitude (Eq. 16/30) into its modeling framework. This enables it to predict how the measured IA signal will evolve as a function of galaxy magnitude and redshift, allowing for accurate forecasting of contamination in magnitude-limited surveys (e.g., Euclid's DR1).

6.:---:---

  1. Redshift Evolution Forecasting Module (z-NLA and z-TATT) The system will utilize the derived redshift evolution power laws (Eq. 15/19) for the linear alignment amplitude, as well as the evolution of nonlinear parameters like tidal torquing amplitude (A2). This capability allows it to forecast IA contamination across different redshift bins, providing a crucial input for tomographic 3x2pt analyses.

8.:---:---

  1. Scale-Dependent Constraint Prior Generator Based on the scale cuts derived from mock validation (Section 5.1), the AI will implement dynamic scale priors (e.g., setting conservative limits like Rmin = 6–7 h−1Mpc for TATT) during likelihood fitting. This prevents overfitting to unphysical small-scale noise and ensures constraints are robust against nonlinear modeling assumptions.

10.:---:---

  1. Model Comparison and Systematic Bias Detector The system will be designed to compare the performance of different IA models (NLA vs. TATT) across various redshift slices and luminosity bins, specifically flagging where the model deviates significantly from simulation results (as seen in Section 6.2). This allows researchers to identify systematic biases introduced by incorrect IA modeling before applying the forecasts to real data.

12.:---:---

  1. Predictive Contamination Mapping for 3x2pt Analyses The ultimate output will be a forecast that maps the expected IA contamination (as a nuisance term) onto cosmological parameter constraints derived from 3x2pt analyses, explicitly accounting for the interplay between galaxy bias and IA evolution, which is the core challenge identified in Section 7.

Abstract

We model intrinsic alignments (IA) in Euclid's Flagship simulation to investigate its impact on Euclid's weak lensing signal. Our IA implementation in the Flagship simulation takes into account photometric properties of galaxies as well as their dark matter host halos. We compare simulations against theory predictions, determining the parameters of two of the most widely used IA models: the Non Linear Alignment (NLA) and the Tidal Alignment and Tidal Torquing (TATT) models. We measure the amplitude of the simulated IA signal as a function of galaxy magnitude and colour in the redshift range 0.1<z<2.1. We find that both NLA and TATT can accurately describe the IA signal in the simulation down to scales of 6 - 7,h-1, Mpc. We measure alignment amplitudes for red galaxies comparable to those of the observations, with samples not used in the calibration procedure. For blue galaxies, our constraints are consistent with zero alignments in our first redshift bin 0.1 < z < 0.3, but we detect a non-negligible signal at higher redshift, which is, however, consistent with the upper limits set by observational constraints. Additionally, several hydrodynamical simulations predict alignment for spiral galaxies, in agreement with our findings. Finally, the evolution of alignment with redshift is realistic and comparable to that determined in the observations. However, we find that the commonly adopted redshift power-law for IA fails to reproduce the simulation alignments above z=1.1. A significantly better agreement is obtained when a luminosity dependence is included, capturing the intrinsic luminosity evolution with redshift in magnitude-limited surveys. We conclude that the Flagship IA simulation is a useful tool for translating current IA constraints into predictions for IA contamination of Euclid-like samples.

Sources

Related papers