[Paper Review] The FLAMINGO Project: Galaxy clusters in comparison to X-ray observations

Joey Braspenning, Joop Schaye+11 more|arXiv (Cornell University)|Dec 13, 2023

Galaxies: Formation, Evolution, Phenomena4 citations

TL;DR

This paper evaluates the FLAMINGO cosmological hydrodynamical simulations by comparing predicted galaxy cluster thermodynamic profiles—temperature, density, pressure, entropy, and metallicity—to X-ray observations. It finds excellent agreement for most properties, except for overestimated core metallicities, and demonstrates that feedback physics and weighting schemes significantly affect simulated cluster properties, especially in cool-core fractions and core structure.

ABSTRACT

Galaxy clusters are important probes for both cosmology and galaxy formation physics. We test the cosmological, hydrodynamical FLAMINGO simulations by comparing to observations of the gaseous properties of clusters measured from X-ray observations. FLAMINGO contains unprecedented numbers of massive galaxy groups ($>10^6$) and clusters ($>10^5$) and includes variations in both cosmology and galaxy formation physics. We predict the evolution of cluster scaling relations as well as radial profiles of the temperature, density, pressure, entropy, and metallicity for different masses and redshifts. We show that the differences between volume-, and X-ray-weighting of particles in the simulations, and between cool-core non cool-core samples, are similar in size as the differences between simulations for which the stellar and AGN feedback has been calibrated to produce significantly different gas fractions. Compared to thermally-driven AGN feedback, kinetic jet feedback calibrated to produce the same gas fraction at $R_{ m 500c}$ yields a hotter core with higher entropies and lower densities, which translates into a smaller fraction of cool-core clusters. Stronger feedback, calibrated to produce lower gas fractions and hence lower gas densities, results in higher temperatures, entropies, and metallicities, but lower pressures. The scaling relations and thermodynamic profiles show almost no evolution with respect to self-similar expectations, except for the metallicity decreasing with redshift. We find that the temperature, density, pressure, and entropy profiles of clusters in the fiducial FLAMINGO simulation are in excellent agreement with observations, while the metallicities in the core are too high.

Motivation & Objective

• To test the FLAMINGO cosmological hydrodynamical simulations against observed X-ray properties of galaxy clusters.
• To assess how variations in feedback physics (thermal vs. kinetic jet AGN feedback) and particle weighting schemes (volume, mass, X-ray) affect simulated cluster profiles.
• To evaluate the evolution of cluster scaling relations and the cool-core fraction with redshift, comparing to observational expectations.
• To determine whether simulated clusters evolve self-similarly and how mass dependence influences normalized thermodynamic profiles.
• To identify discrepancies between simulations and observations, particularly in intracluster medium (ICM) metallicity and core structure.

Proposed method

• Simulating galaxy clusters using the FLAMINGO code, which includes both cosmological and galaxy formation physics variations across multiple realizations.
• Computing radial profiles of temperature, density, pressure, entropy, and metallicity from simulated particle data at different redshifts and masses.
• Applying three different particle weighting schemes—volume, mass, and X-ray luminosity—when constructing profiles to assess their impact on derived quantities.
• Defining cool-core clusters via central cooling time, enabling comparison between cool-core and non-cool-core sub-samples in simulations and observations.
• Normalizing profiles to virial quantities (e.g., $R_{ m 500c}$) to test for self-similar evolution across redshift and mass.
• Calibrating AGN feedback models (thermally-driven vs. kinetic jet) to match observed gas fractions at $R_{ m 500c}$, then comparing resulting thermodynamic profiles.

Experimental results

Research questions

• RQ1 How well do the FLAMINGO simulations reproduce observed X-ray scaling relations and radial thermodynamic profiles of galaxy clusters?
• RQ2 What is the impact of different particle weighting schemes (volume, mass, X-ray) on the derived thermodynamic profiles in simulations?
• RQ3 How does the choice of AGN feedback model (thermal vs. kinetic jet) affect cluster core properties such as temperature, entropy, and density?
• RQ4 To what extent do simulated cluster profiles evolve with redshift, and how does this compare to self-similar expectations?
• RQ5 Why is the observed cool-core fraction in simulations higher than in observations, and how does this depend on the definition and redshift evolution?

Key findings

• The fiducial FLAMINGO simulation shows excellent agreement with X-ray observations for temperature, density, pressure, and entropy profiles across different masses and redshifts.
• The iron abundance in the cluster core is overestimated in the simulation by approximately 0.2 dex compared to observations, indicating a key discrepancy in metallicity evolution.
• Kinetic jet feedback, calibrated to match the same gas fraction as thermally-driven feedback, produces hotter cores with higher entropy and lower density, reducing the cool-core fraction by ~20%.
• Volume- and mass-weighting of particles produce differences of up to 0.2 dex in density and entropy profiles, especially at small and large radii, while X-ray weighting yields the most observationally consistent results.
• Cool-core clusters in FLAMINGO have lower core temperatures and entropies, but higher densities, pressures, and metallicities—consistent with observational trends.
• After correcting for self-similar evolution, the cool-core fraction remains nearly constant with redshift, suggesting that the observed redshift evolution is primarily driven by selection effects rather than physical evolution.