Simulations Of Galaxy Cluster Formation Research Articles

Exploiting Machine Learning and Disequilibrium in Galaxy Clusters to Obtain a Mass Profile

We use 3D k-means clustering to characterize galaxy substructure in the A2146 cluster of galaxies (z = 0.2343). This method objectively characterizes the cluster’s substructure using projected position and velocity data for 67 galaxies within a 2.305 Mpc circular region centered on the cluster's optical center. The optimal number of substructures is found to be four. Four distinct substructures with rms velocity typical of galaxy groups or low-mass subclusters, when compared to cosmological simulations of galaxy cluster formation, suggest that A2146 is in the early stages of formation. We utilize this disequilibrium, which is so prevalent in galaxy clusters at all redshifts, to construct a radial mass distribution. Substructures are bound but not virialized. This method is in contrast to previous kinematical analyses, which have assumed virialization, and ignored the ubiquitous clumping of galaxies. The best-fitting radial mass profile is much less centrally concentrated than the well-known Navarro–Frenk–White profile, indicating that the dark-matter-dominated mass distribution is flatter pre-equilibrium, becoming more centrally peaked in equilibrium through the merging of the substructure.

Nonlinear dynamo in the intracluster medium

Hot plasma in galaxy clusters, the intracluster medium is observed to be magnetized with magnetic fields of around a μG and the correlation scales of tens of kiloparsecs, the largest scales of the magnetic field so far observed in the Universe. Can this magnetic field be used as a test of the primordial magnetic field in the early Universe? In this paper, we argue that if the cluster field was created by the nonlinear dynamo, the process would be insensitive to the value of the initial field. Our model combines state of the art hydrodynamic simulations of galaxy cluster formation in a fully cosmological context with nonlinear dynamo theory. Initial field is not a parameter in this model, yet it predicts magnetic scale and strength compatible with observations.

Constraints on the Optical Depth of Galaxy Groups and Clusters

Abstract Future data from galaxy redshift surveys, combined with high-resolutions maps of the cosmic microwave background, will enable measurements of the pairwise kinematic Sunyaev–Zel’dovich (kSZ) signal with unprecedented statistical significance. This signal probes the matter-velocity correlation function, scaled by the average optical depth (τ) of the galaxy groups and clusters in the sample, and is thus of fundamental importance for cosmology. However, in order to translate pairwise kSZ measurements into cosmological constraints, external constraints on τ are necessary. In this work, we present a new model for the intracluster medium, which takes into account star formation, feedback, non-thermal pressure, and gas cooling. Our semi-analytic model is computationally efficient and can reproduce results of recent hydrodynamical simulations of galaxy cluster formation. We calibrate the free parameters in the model using recent X-ray measurements of gas density profiles of clusters, and gas masses of groups and clusters. Our observationally calibrated model predicts the average (i.e., the integrated τ within a disk of size R 500) to better than 6% modeling uncertainty (at 95% confidence level). If the remaining uncertainties associated with other astrophysical uncertainties and X-ray selection effects can be better understood, our model for the optical depth should break the degeneracy between optical depth and cluster velocity in the analysis of future pairwise kSZ measurements and improve cosmological constraints with the combination of upcoming galaxy and CMB surveys, including the nature of dark energy, modified gravity, and neutrino mass.

Increasing Black Hole Feedback-induced Quenching with Anisotropic Thermal Conduction

Feedback from central supermassive blackholes is often invoked to explain the low star formation rates in massive galaxies at the centers of galaxy clusters. However, the detailed physics of the coupling of the injected feedback energy with the intracluster medium is still unclear. Using high-resolution magnetohydrodynamic cosmological simulations of galaxy cluster formation, we investigate the role of anisotropic thermal conduction in shaping the thermodynamic structure of clusters, and, in particular, in modifying the impact of black hole feedback. Stratified anisotropically conducting plasmas are formally always unstable, and thus more prone to mixing, an expectation borne out by our results. The increased mixing efficiently isotropizes the injected feedback energy which in turn significantly improves the coupling between the feedback energy and the intracluster medium. This facilitates an earlier disruption of the cool core, reduces the star formation rate by more than an order of magnitude, and results in earlier quenching despite an overall lower amount of feedback energy injected into the cluster core. With conduction, the metallicity gradients and dispersions are lowered, aligning them better with observational constraints. These results highlight the important role of thermal conduction in establishing and maintaining quiescence of massive galaxies.

WEIGHING GALAXY CLUSTERS WITH GAS. I. ON THE METHODS OF COMPUTING HYDROSTATIC MASS BIAS

Mass estimates of galaxy clusters from X-ray and Sunyeav-Zel'dovich observations assume the intracluster gas is in hydrostatic equilibrium with their gravitational potential. However, since galaxy clusters are dynamically active objects whose dynamical states can deviate significantly from the equilibrium configuration, the departure from the hydrostatic equilibrium assumption is one of the largest sources of systematic uncertainties in cluster cosmology. In the literature there has been two methods for computing the hydrostatic mass bias based on the Euler and the modified Jeans equations, respectively, and there has been some confusion about the validity of these two methods. The word "Jeans" was a misnomer, which incorrectly implies that the gas is collisionless. To avoid further confusion, we instead refer these methods as "summation" and "averaging" methods respectively. In this work, we show that these two methods for computing the hydrostatic mass bias are equivalent by demonstrating that the equation used in the second method can be derived from taking spatial averages of the Euler equation. Specifically, we identify the correspondences of individual terms in these two methods mathematically and show that these correspondences are valid to within a few percent level using hydrodynamical simulations of galaxy cluster formation. In addition, we compute the mass bias associated with the acceleration of gas and show that its contribution is small in the virialized regions in the interior of galaxy cluster, but becomes non-negligible in the cluster outskirts of massive clusters where materials are being actively accreted. We discuss implications for understanding the origin of hydrostatic mass biases and for characterizing them using hydrodynamical simulations as well as observations.

PREDICTING MERGER-INDUCED GAS MOTIONS IN ΛCDM GALAXY CLUSTERS

In the hierarchical structure formation model, clusters of galaxies form through a sequence of mergers and continuous mass accretion, which generate significant random gas motions especially in their outskirts where material is actively accreting. Non-thermal pressure provided by the internal gas motions affects the thermodynamic structure of the X-ray emitting intracluster plasma and introduces biases in the physical interpretation of X-ray and Sunyaev-Zeldovich effect observations. However, we know very little about the nature of gas motions in galaxy clusters. The ASTRO-H X-ray mission, scheduled to launch in 2015, will have a calorimeter capable of measuring gas motions in galaxy clusters at the level of <100 km/s. In this work, we predict the level of merger-induced gas motions expected in the Lambda-CDM model using hydrodynamical simulations of galaxy cluster formation. We show that the gas velocity dispersion is larger in more massive clusters, but exhibits a large scatter. We show that systems with large gas motions are morphologically disturbed, while early forming, relaxed groups show a smaller level of gas motions. By analyzing mock ASTRO-H observations of simulated clusters, we show that such observations can accurately measure the gas velocity dispersion out to the outskirts of nearby relaxed galaxy clusters. ASTRO-H analysis of merging clusters, on the other hand, requires multi-component spectral fitting and enables unique studies of substructures in galaxy clusters by measuring both the peculiar velocities and the velocity dispersion of gas within individual sub-clusters.

MAGNETOHYDRODYNAMIC TURBULENCE AND COSMIC-RAY REACCELERATION IN GALAXY CLUSTERS

Cosmological MHD simulations of galaxy cluster formation show a significant amplification of seed magnetic fields. We developed a novel method to decompose cluster magnetized turbulence into modes and showed that the fraction of the fast mode is fairly large, around 1/4 in terms of energy. This is larger than that was estimated before, which implies that cluster turbulence interacts with cosmic rays rather efficiently. We propose a framework to deal with electron and proton reacceleration in galaxy clusters that includes feedback on turbulence. In particular, we establish a new upper limit on proton and electron fluxes based on turbulence intensity. These findings, along with detailed modeling of reacceleration, will help to reconcile the observed giant radio haloes and the unobserved diffuse gamma-ray emission from these clusters.

A divergence-cleaning scheme for cosmological SPMHD simulations

In magnetohydrodynamics (MHD), the magnetic field is evolved by the induction equation and coupled to the gas dynamics by the Lorentz force. We perform numerical smoothed particle magnetohydrodynamics (Spmhd) simulations and study the influence of a numerical magnetic divergence. For instabilities arising from divergence B related errors, we find the hyperbolic/parabolic cleaning scheme suggested by Dedner et al. 2002 to give good results and prevent numerical artifacts from growing. Additionally, we demonstrate that certain current Spmhd implementations of magnetic field regularizations give rise to unphysical instabilities in long-time simulations. We also find this effect when employing Euler potentials (divergenceless by definition), which are not able to follow the winding-up process of magnetic field lines properly. Furthermore, we present cosmological simulations of galaxy cluster formation at extremely high resolution including the evolution of magnetic fields. We show synthetic Faraday rotation maps and derive structure functions to compare them with observations. Comparing all the simulations with and without divergence cleaning, we are able to confirm the results of previous simulations performed with the standard implementation of MHD in Spmhd at normal resolution. However, at extremely high resolution, a cleaning scheme is needed to prevent the growth of numerical errors at small scales.

COSMOLOGICAL MAGNETOHYDRODYNAMIC SIMULATIONS OF CLUSTER FORMATION WITH ANISOTROPIC THERMAL CONDUCTION

(abridged) The ICM has been suggested to be buoyantly unstable in the presence of magnetic field and anisotropic thermal conduction. We perform first cosmological simulations of galaxy cluster formation that simultaneously include magnetic fields, radiative cooling and anisotropic thermal conduction. In isolated and idealized cluster models, the magnetothermal instability (MTI) tends to reorient the magnetic fields radially. Using cosmological simulations of the Santa Barbara cluster we detect radial bias in the velocity and magnetic fields. Such radial bias is consistent with either the inhomogeneous radial gas flows due to substructures or residual MTI-driven field rearangements that are expected even in the presence of turbulence. Although disentangling the two scenarios is challenging, we do not detect excess bias in the runs that include anisotropic thermal conduction. The anisotropy effect is potentially detectable via radio polarization measurements with LOFAR and SKA and future X-ray spectroscopic studies with the IXO. We demonstrate that radiative cooling boosts the amplification of the magnetic field by about two orders of magnitude beyond what is expected in the non-radiative cases. At z=0 the field is amplified by a factor of about 10^6 compared to the uniform magnetic field evolved due to the universal expansion alone. Interestingly, the runs that include both radiative cooling and anisotropic thermal conduction exhibit stronger magnetic field amplification than purely radiative runs at the off-center locations. In these runs, shallow temperature gradients away from the cluster center make the ICM neutrally buoyant. The ICM is more easily mixed in these regions and the winding up of the frozen-in magnetic field is more efficient resulting in stronger magnetic field amplification.

GAS CLUMPING IN THE OUTSKIRTS OF ΛCDM CLUSTERS

Recent Suzaku X-ray observations revealed that the observed entropy profile of the intracluster medium (ICM) deviates significantly from the prediction of hydrodynamical simulations of galaxy clusters. In this work, we show that gas clumping introduces significant biases in X-ray measurements of the ICM profiles in the outskirts of galaxy clusters. Using hydrodynamical simulations of galaxy cluster formation in a concordance Lambda-CDM model, we demonstrate that gas clumping leads to an overestimate of the observed gas density and causes flattening of the entropy profile. Our results suggest that gas clumping must be taken into account when interpreting X-ray measurements of cluster outskirts.

Simulations of cosmic-ray feedback by active galactic nuclei in galaxy clusters

Feedback processes by active galactic nuclei (AGN) appear to be a key for understanding the nature of the very X-ray luminous cool cores found in many clusters of galaxies. We investigate a numerical model for AGN feedback where for the first time a relativistic particle population in AGN-inflated bubbles is followed within a full cosmological context. In our high-resolution simulations of galaxy cluster formation, we assume that black hole accretion is accompanied by energy feedback that occurs in two different modes, depending on the accretion rate itself. At high accretion rates, a small fraction of the radiated energy is coupled thermally to the gas surrounding the quasar, while in a low-accretion state, mechanically efficient feedback in the form of hot, buoyant bubbles that are inflated by radio activity is considered. Unlike previous work, we inject a non-thermal particle population of relativistic protons into the AGN bubbles, instead of adopting a purely thermal heating. We then follow the subsequent evolution of the cosmic-ray (CR) plasma inside the bubbles, considering both its hydrodynamical interactions and dissipation processes relevant to the CR population. This permits us to analyse the impact of CR bubbles on the surrounding intracluster medium, and in particular, how this contrasts with the purely thermal case. Due to the different buoyancy of relativistic plasma and the comparatively long CR dissipation time-scale, we find substantial changes in the evolution of clusters as a result of CR feedback. In particular, the non-thermal population can provide significant pressure support in central cluster regions at low thermal temperatures, providing a natural explanation for the decreasing temperature profiles found in cool core clusters. At the same time, the morphologies of the bubbles and of the induced X-ray cavities show a striking similarity to observational findings. AGN feedback with CRs also proves efficient in regulating cluster cooling flows so that the total baryon fraction in stars becomes limited to realistic values of the order of ∼10 per cent, more than a factor of 3 reduction compared with cosmological simulations that only consider radiative cooling and supernova feedback. We find that the partial CR support of the intracluster gas also affects the expected signal of the thermal Sunyaev–Zel'dovich effect, with typical modifications of the integrated Compton-y parameter within the virial radius of the order of ∼10 per cent.

Physical viscosity in smoothed particle hydrodynamics simulations of galaxy clusters

Most hydrodynamical simulations of galaxy cluster formation carried out to date have tried to model the cosmic gas as an ideal, inviscid fluid, where only a small amount of (unwanted) numerical viscosity is present, arising from practical limitations of the numerical method employed, and with a strength that depends on numerical resolution. However, the physical viscosity of the gas in hot galaxy clusters may in fact not be negligible, suggesting that a self-consistent treatment that accounts for the internal gas friction would be more appropriate. To allow such simulations using the smoothed particle hydrodynamics (SPH) method, we derive a novel SPH formulation of the Navier–Stokes and general heat transfer equations and implement them in the gadget-2 code. We include both shear and bulk viscosity stress tensors, as well as saturation criteria that limit viscous stress transport where appropriate. Our scheme integrates consistently into the entropy and energy conserving formulation of SPH employed by the code. Using a number of simple hydrodynamical test problems, e.g. the flow of a viscous fluid through a pipe, we demonstrate the validity of our implementation. Adopting Braginskii parametrization for the shear viscosity of hot gaseous plasmas, we then study the influence of viscosity on the interplay between AGN-inflated bubbles and the surrounding intracluster medium (ICM). We find that certain bubble properties like morphology, maximum clustercentric radius reached or survival time depend quite sensitively on the assumed level of viscosity. Interestingly, the sound waves launched into the ICM by the bubble injection are damped by physical viscosity, establishing a non-local heating process. However, we find that the associated heating is rather weak due to the overall small energy content of the sound waves. Finally, we carry out cosmological simulations of galaxy cluster formation with a viscous ICM. We find that the presence of physical viscosity induces new modes of entropy generation, including a significant production of entropy in filamentary regions perpendicular to the direction of the clusters encounter. Viscosity also modifies the dynamics of mergers and the motion of substructures through the cluster atmosphere. Substructures are generally more efficiently stripped of their gas, leading to prominent long gaseous tails behind infalling massive haloes.