CHEX-MATE: Are we getting cluster thermodynamics right?

This study utilizes end-to-end simulations of CHEX-MATE-like galaxy clusters to demonstrate that while current X-ray analysis techniques can robustly recover gas density and mass profiles, they face significant challenges and potential biases in accurately measuring temperature profiles due to azimuthal variations and multi-temperature gas along the line of sight.

The Gist

Imagine the Universe is a giant, invisible web made of dark matter, and at the intersections of this web sit massive "cities" of galaxies called galaxy clusters. These aren't just collections of stars; they are mostly filled with super-hot gas (plasma) that glows in X-rays.

Scientists want to use these cosmic cities to understand the history of the Universe, but to do that, they need to know exactly how much "stuff" (mass) is in them and how hot the gas is. This is like trying to weigh a cloud by looking at its shadow.

This paper is essentially a quality control test for the tools scientists use to measure these clouds. The authors asked: "Are our rulers and thermometers accurate, or are we making mistakes?"

Here is the breakdown of their investigation using simple analogies:

1. The Setup: Building a "Fake" Universe

To test their tools, the scientists didn't just look at real clusters (because they don't know the "true" weight of a real cluster). Instead, they built three different virtual universes inside supercomputers.

• Think of these as three different video games (The300, Magneticum, and MACSIS) that simulate how gas and galaxies behave.
• They picked "twin" clusters from these games that look exactly like the real ones being studied by the CHEX-MATE project (a major real-world telescope survey).

2. The Simulation: Taking a "Fake" Photo

Next, they took these virtual clusters and ran them through a digital camera simulator.

• They didn't just look at the raw data; they simulated exactly what the XMM-Newton telescope would see.
• They added "noise," "blur" (like a camera lens that isn't perfect), and background stars, just like a real photo.
• This created a set of "fake" X-ray images that looked indistinguishable from real observations.

3. The Test: Measuring the Fake Photos

Now, the scientists took these fake photos and ran them through the standard analysis software that real astronomers use every day. They tried to measure two main things:

1. Gas Density: How crowded the gas is.
2. Temperature: How hot the gas is.

4. The Results: What Went Right and What Went Wrong

The Good News: The "Crowd" Count is Accurate

• Analogy: Imagine trying to count how many people are in a stadium by looking at the density of the crowd from above.
• Result: The scientists found that their tools were excellent at measuring gas density. They could reconstruct the "crowd" density with less than 2% error. Even the total weight of the gas (gas mass) was calculated with incredible precision (better than 1% error).
• Why it works: Density is like a shadow; it's hard to mess up because the gas glows brightly where it's dense.

The Bad News: The "Thermometer" is Tricky

• Analogy: Imagine trying to measure the temperature of a pot of soup that has hot chunks of meat and cold chunks of ice floating in it, all mixed together. If you stick a thermometer in, it might give you a "mixed" reading that doesn't represent the whole pot.
• Result: Measuring temperature was much harder. The tools often got it wrong, especially in the center of the clusters.
• The Culprit: The gas isn't uniform. It has "clumps" of different temperatures. When the telescope looks at the cluster, it sees a blend of hot and cold gas. The standard math assumes the gas is a smooth, single temperature, but it's actually a messy mix. This causes the "thermometer" to read lower than the true average temperature.

5. The Big Consequence: We Might Be Underestimating the Weight

This is the most important part of the paper.

• In physics, if you know the temperature and pressure of a gas, you can calculate how heavy the whole cluster is (using the "Hydrostatic Equilibrium" rule).
• The Problem: Because the "thermometer" reads too low (due to the messy mix of hot and cold gas), the calculation says the cluster is lighter than it actually is.
• The Analogy: It's like weighing a suitcase by guessing the weight of the clothes inside based on a temperature reading. If you think the clothes are colder (lighter) than they really are, you'll guess the suitcase is lighter than it actually is.

6. The Conclusion & Future

The paper concludes that while we are very good at counting the gas, we need to be very careful when interpreting temperature.

• The "Real" Issue: For years, scientists thought the clusters were "lighter" than expected because the gas was moving around wildly (turbulence). But new data from a mission called XRISM shows the gas is actually quite calm.
• The New Insight: This paper suggests the "missing weight" isn't because the gas is moving; it's because our math is too simple to handle the complex, multi-temperature soup inside the clusters.

In a nutshell:
We have great rulers to measure the size of the galaxy clusters, but our thermometers are a bit fuzzy because the gas is a complex mix of hot and cold. If we don't fix how we read the temperature, we will keep underestimating the mass of the Universe's largest structures. The authors are now working on better ways to "un-mix" that soup to get the true weight.

Technical Summary

Here is a detailed technical summary of the paper "CHEX-MATE: Are we getting cluster thermodynamics right?" by Seppi et al. (2026).

1. Problem Statement

Galaxy clusters are critical laboratories for cosmology and astrophysics, encoding information about dark matter, dark energy, and baryonic physics. However, the scientific exploitation of cluster surveys (such as the CHEX-MATE project) relies heavily on the accurate measurement of thermodynamic properties (density, temperature, pressure, entropy) from X-ray observations.

A persistent challenge in X-ray cluster cosmology is the hydrostatic mass bias, where masses derived assuming hydrostatic equilibrium are systematically lower than the true mass. While non-thermal pressure support (turbulence, bulk motions) is often cited as the cause, recent high-resolution spectroscopic measurements (e.g., from XRISM) suggest gas velocity dispersions are lower than expected, implying non-thermal pressure alone may not explain the full bias. This raises the question: Do current X-ray analysis techniques accurately recover intrinsic thermodynamic profiles, or are systematic biases introduced by multi-phase gas structures, projection effects, and analysis methodologies?

2. Methodology

The authors developed a rigorous end-to-end simulation pipeline to test the reliability of standard X-ray analysis tools against "ground truth" data from hydrodynamical simulations.

• Simulation Suites: The study utilized three state-of-the-art hydrodynamical simulations with different physical implementations and cosmological parameters:
  • The300: 324 re-simulated clusters using GADGET-X (SPH with improved mixing).
  • Magneticum: A suite using P-GADGET3, known for preserving cool substructures due to limited mixing in its SPH formulation.
  • MACSIS: 390 re-simulated massive clusters using BAHAMAS physics (OWLS subgrid models).
• Mock Observation Generation:
  • Gas particles were projected along the line of sight to create emissivity maps.
  • A newly developed XMM-Newton simulator (xmm_simulator) was used to generate realistic mock EPIC event files. This included telescope response (PSF, effective area), instrumental background, sky background (CXB, Galactic halo, Local Hot Bubble), and AGN contamination.
  • Exposure times were set to 25 ks, representative of CHEX-MATE observations.
• Analysis Pipeline:
  • Mock data were processed using standard X-ray tools: XMM-SAS, pyproffit, and hydromass.
  • Surface Brightness: Extracted from Voronoi-tessellated images (median profiles) and standard binned images (mean profiles).
  • Spectral Fitting: Performed using the X-COP pipeline (XSPEC with C-stat) to derive radial temperature profiles.
  • Deprojection: Three modeling approaches were tested:
    1. NFW: Parametric mass model.
    2. NP (Non-Parametric): Linear combination of log-normal functions for temperature.
    3. FM (Forward Model): Generalized NFW for pressure.
• Comparison Metrics: Recovered profiles were compared to intrinsic simulation values. Crucially, the authors compared X-ray spectral fitting results against two intrinsic temperature definitions:
  • Mass-Weighted (MW): $T_{MW} = \sum m_i T_i / \sum m_i$
  • Spectroscopic-Like (SL): $T_{SL} = \sum n_i^2 T_i^{-3/4} T_i / \sum n_i^2 T_i^{-3/4}$ (designed to mimic X-ray emissivity weighting).

3. Key Contributions

• End-to-End Validation: Unlike previous studies that often analyzed raw simulation particles, this work processes full mock X-ray observations through a realistic instrument simulator and standard analysis pipelines, bridging the gap between theory and observation.
• Quantification of Temperature Bias: The study isolates the impact of multi-temperature gas structures and azimuthal variations on temperature measurements, distinguishing them from pure projection effects.
• Voronoi Binning Efficacy: Demonstrated that Voronoi-tessellated surface brightness profiles significantly reduce biases in gas density reconstruction caused by cool gas clumps compared to standard mean profiles.
• Empirical Correction Model: Developed a statistical model linking the ratio of $T_{SL}/T_{MW}$ to the width (standard deviation) and median of local temperature fluctuations, providing a tool to correct for multi-phase biases.

4. Key Results

A. Gas Density and Mass

• Robust Recovery: Gas density profiles are robustly recovered across all simulations and methods.
• Accuracy: Using azimuthal median profiles from Voronoi images, density is recovered within ~2% (at most) of the true value.
• Mass: Gas mass is reconstructed with better than 1% accuracy.
• Bias Source: Standard mean profiles tend to overestimate density and mass by ~5% due to the $n_e^2$ dependence of X-ray emissivity, which biases the measurement toward dense, cool clumps. Voronoi binning mitigates this.
• Miscentering: Small discrepancies in the innermost cores ($<0.1 R_{500}$) are largely attributed to miscentering (offset between X-ray peak and dark matter potential minimum), which resolves when well-centered systems are selected.

B. Temperature Profiles

• SL vs. MW: The spectroscopic-like ($T_{SL}$) temperature is generally in excellent agreement with X-ray spectral fitting results ($T_X$). However, $T_{SL}$ is systematically lower than the mass-weighted temperature ($T_{MW}$) at all radii.
• Multi-Temperature Impact: The discrepancy between $T_{SL}$ and $T_{MW}$ is driven by unresolved multi-temperature structures (clumps, mixing, shocks) rather than simple projection effects.
  • In simulations with strong cool cores (e.g., MACSIS), $T_{SL}$ underestimates the core temperature by up to 15% compared to $T_{MW}$.
  • The bias correlates with the standard deviation of temperature fluctuations within a radial bin. A broader temperature distribution leads to a larger suppression of the spectroscopic temperature.
• Reconstruction Methods: The Non-Parametric (NP) model provides the best recovery of $T_{SL}$. The NFW and Forward Model (FM) approaches show larger scatter, particularly in simulations with complex thermodynamics (Magneticum).

C. Derived Thermodynamics (Pressure and Entropy)

• Systematic Underestimation: Thermal pressure and entropy profiles are systematically underestimated relative to true values.
  • The300: Pressure underestimated by ~5-10%.
  • Magneticum & MACSIS: Pressure underestimated by ~15-20%.
• Cause: Since density is recovered accurately, the pressure bias is driven primarily by the underestimation of temperature due to multi-phase gas structures. This leads to a suppression of the inferred thermal pressure.

5. Significance and Conclusions

• Hydrostatic Mass Bias: The study suggests that a significant portion of the hydrostatic mass bias may stem from spectroscopic temperature biases in multi-phase gas, rather than solely from non-thermal pressure support. If $T$ is underestimated due to cool clumps, the derived hydrostatic mass ($M \propto T \times \nabla T$) will also be biased low, even in systems in hydrostatic equilibrium.
• Implications for XRISM/Athena: The findings highlight the necessity of high-resolution spectroscopy (e.g., XRISM, Athena) to resolve multi-temperature structures. Line ratios (e.g., Fe XXV/XXVI) can provide direct diagnostics of thermal complexity that continuum fitting misses.
• Methodological Recommendations:
  • Use Voronoi-tessellated median profiles for density and mass reconstruction to avoid clump bias.
  • Interpret temperature profiles with caution, acknowledging that $T_{SL}$ is a weighted average that may not represent the gravitational potential depth ($T_{MW}$) in systems with high thermal inhomogeneity.
  • Future mass calibration must account for the bias introduced by multi-phase gas, not just non-thermal pressure.

In summary, while X-ray analysis techniques are highly reliable for gas density and mass, temperature measurements are subject to significant biases driven by the complex, multi-phase nature of the Intra-Cluster Medium (ICM). Correcting for these biases is essential for accurate cosmological constraints and understanding the true thermodynamic state of galaxy clusters.