TL;DR
This paper investigates the relationship between the velocity dispersion of galaxies within clusters and the temperature of the hot gas in those clusters. They find a correlation that deviates from the simple prediction of the virial theorem, suggesting that processes like protogalactic winds and dynamical friction play a significant role in shaping this relationship.
What Problem Does This Paper Address?
The virial theorem provides a theoretical prediction for how the velocity dispersion of galaxies ($\sigma_r$) in a cluster should relate to the temperature ($T$) of the intracluster medium (ICM). Specifically, it predicts a relationship of $\sigma_r \propto T^{0.5}$, assuming the cluster is in a state of virial equilibrium, with isotropic galaxy orbits, and gravity as the sole energy source. However, previous studies have shown a large scatter in this observed correlation, and some have found a steeper slope than predicted. This paper aims to re-examine this correlation using a carefully selected sample of galaxy clusters, focusing on those with minimal substructure and dominant central galaxies, to better understand the underlying physics and the reasons for the deviation from the virial theorem prediction.
Methods
The authors employ a combination of observational data and statistical analysis. They use a limited sample of 22 galaxy clusters, specifically chosen from Bird (1994)'s "cD database," which are known to have dominant central galaxies and have undergone substructure correction. This selection aims to minimize the impact of complex dynamical states that can introduce scatter. They then perform linear regression analysis on the velocity dispersion and temperature data. To assess the statistical significance of the observed deviation from the virial theorem's $T^{0.5}$ prediction, they utilize a Monte Carlo computer routine to simulate data and estimate the probability of obtaining their observed slope by chance. They also explore various regression techniques, including the bisector method, which is considered more appropriate for independent variables.
Datasets
The primary dataset consists of 22 galaxy clusters selected from Bird (1994). This sample is characterized by having dominant central galaxies and being corrected for substructure. The X-ray temperature data for these clusters were supplemented with new, more accurate measurements from ASCA and Ginga, in addition to previously published data from EXOSAT and Einstein. Velocity dispersion data were also taken from the literature, with substructure corrections applied.
Key Insights and Ideas
- Substructure Matters: The presence of substructure in galaxy clusters significantly affects the scatter in the velocity dispersion-temperature correlation. Correcting for substructure helps to reveal a clearer underlying relationship.
- Beyond Virial Equilibrium: The observed correlation is consistently steeper than the $T^{0.5}$ predicted by the virial theorem, even after accounting for substructure. This indicates that other physical processes are at play.
- Protogalactic Winds and Dynamical Friction: The paper strongly suggests that protogalactic winds and dynamical friction are key mechanisms that can explain the observed steeper slope. These processes can inject energy into the ICM and influence the relative motions of galaxies, leading to a temperature dependence that deviates from the simple virial prediction.
- Regression Technique Sensitivity: The choice of linear regression method (e.g., ordinary least squares vs. bisector) can significantly impact the derived slope of the correlation, highlighting the importance of using appropriate statistical tools for astrophysical data.
Key Results
- The velocity dispersion-temperature correlation for their limited cluster sample is found to be $\sigma_r \propto T^{0.61 \pm 0.13}$.
- Monte Carlo simulations indicate that the probability of observing a correlation as steep as or steeper than their result, assuming only virial equilibrium, is only 4%, suggesting the deviation is statistically significant.
- The protogalactic winds model, when combined with dynamical friction, reproduces the observed relationship between velocity dispersion and temperature remarkably well.
Limitations and Future Work
- Sample Size: While the authors carefully selected their sample, it is still a "limited cluster sample." Larger datasets with higher-quality X-ray spectra and more clusters are needed to further refine the understanding of this fundamental correlation.
- Model Assumptions: The agreement between the simulated ensemble and real clusters is acknowledged as potentially fortuitous, with the wind model in simulations having a higher luminosity than expected for real galaxies. The dynamical accuracy of modeling galaxies as collisionless particles is also noted as an area of uncertainty.
- Future Work: The authors suggest that future work should focus on obtaining larger datasets, improving X-ray spectral resolution, and further investigating the role of protogalactic winds and velocity bias. They also plan to present non-parametric determinations of galaxy and gas density profiles to test cluster-by-cluster variations in these profiles.
Summary updated