TNG-Cluster is a new addition to the IllustrisTNG suite of cosmological magnetohydrodynamical simulations of galaxy formation.
Its objective is to significantly increase the statistical sampling of the most massive and rare objects in the Universe: galaxy clusters with $M_{\rm halo} / M_\odot \gtrsim 10^{15.0}$.
Galaxy clusters are the ultimate outcome of the hierarchical assembly of structure. Their stellar bodies, supermassive black holes (SMBHs), dark matter halos, and gaseous atmospheres are all constructed from the collective merging of thousands, if not tens of thousands, of smaller constituents. Clusters are the most massive gravitationally bound structures in the Universe. They host the most massive galaxies, the most massive SMBHs, and the largest populations of satellite galaxies that exist. The dynamics and physical processes at play in the hot plasma of the intracluster medium (ICM) are complex, diverse, and observable. They are cosmologically rare, yet form a crucial observational probe of our cosmological model.
TNG-Cluster has three principal scientific goals:
By well-sampling these rare structures we have simulated a sizeable sample of massive clusters and galaxies. The excellent numerical resolution, combined with the rich features and physics of the TNG model, enables novel science.
In particular, we can study the interplay of galaxy interactions, SMBH feedback, turbulence, magnetic field amplification, non-thermal support, and shock formation and propagation in shaping the properties of the ICM of clusters. This is the gas that, having been heated to temperatures of 10 million Kelvin or more, fills massive dark matter halos and, in turn, affects the evolution of galaxies therein. For example, how do kinematic turbulence and SMBH-driven outflows influence the observable velocity structure of the ICM? How are the location, strength, and morphology of shock fronts regulated? What dominates the gas structure in the cluster core vs. its outskirts? How do the thermodynamical properties and kinematics of the ICM depend on the feedback from SMBHs? Do mergers between clusters change the picture?
TNG-Cluster enables us to develop mock observations of the ICM to compare against, and gain insights from, current and upcoming observations. These range from radio wavelengths from synchrotron emission, for connections to observations with e.g. LOFAR, MeerKAT, and in the future SKA, to the mm and X-ray bands. For the latter, two fundamental observables characterize galaxy clusters: the Sunyaev-Zelvodich effect (SZE), whereby photons from the cosmic microwave background interact with free electrons in the ICM plasma, and X-ray emission, primarily from thermal bremsstrahlung. Forward modeling will enable quantitative connections with existing and future X-ray observatories (Chandra, XMM-Newton, eROSITA, XRISM, and future ATHENA and LEM) and for the SZE from upcoming surveys (e.g. Atacama Cosmology Telescope (ACT), South Pole Telescope (SPT), Simons Observatory, CMB-S4).
In fact, X-ray detected clusters will play a key role in many precision cosmology experiments over the next decade. However, traditional cosmological pipelines suffer from an inadequate parameterization and marginalization over the key parameters of mass-observable relations. With TNG-Cluster, we can provide theoretically-motivated priors on the redshift evolution, shape, and intrinsic scatter of the relations between cluster mass and X-ray luminosity and temperature.
Finally, although only a few percent of all galaxies at the present day reside in the most massive and rarest dark-matter halos, these clusters contain a unique mixture of galaxy types. TNG-Cluster will provide an unmatched dataset for the theoretical study of brightest cluster galaxies (BCGs), satellite galaxies, as well as of proto-clusters at high redshifts. We can ask, for example, how do physical mechanisms such as mergers, star formation, gas accretion and SMBH feedback determine the mass and morphological evolution of BCGs? At what past epochs have they been disky and star forming? What causes galaxies shaped like flattened disks to transform into elliptical spheroids? What is the relationship between the BCG and cool-core clusters? How does the proto-cluster environment affect the evolution of galaxies that will become cluster galaxies?
The original IllustrisTNG project consisted of three volumes: TNG50, TNG100, and TNG300. These simulated regions of space 50, 100, and 300 comoving Mpc on a side, respectively. TNG-Cluster is instead a project that resimulates ~ 350 clusters drawn from a much larger, 1 Gpc3 volume, thirty-six times larger than TNG300.
We keep -- entirely unchanged -- the IllustrisTNG galaxy formation model, as well as the numerical resolution. This new sample of hundreds of highly resolved, massive galaxy clusters enables studies on the assembly of high-mass ellipticals and their supermassive black holes (SMBHs), brightest cluster galaxies (BCGs), satellite galaxy evolution and environmental processes, jellyfish galaxies, intracluster medium (ICM) properties, cooling and AGN feedback, mergers and relaxedness, magnetic field amplification, and the galaxy-halo connection at the high-mass end, with observables from the optical to radio synchrotron and the Sunyaev-Zeldovich (SZ) effect, to X-ray emission, as well as their cosmological applications.
The table above shows details of the TNG-Cluster (right-most column) simulation in comparison to its smaller volume siblings. The values are: volume, box side-length, number of initial gas cells, dark matter particles, and tracers; the mean baryonic cell mass and dark matter particle mass; the gravitational softening length minimum for the gas, and the softening for the collisionless components at $z=0$; the number of groups (13.0 to 14.0), lower mass clusters (14.0-14.5), intermediate-mass (14.5-15.0), and high mass ($>$15.0) clusters. $\dagger$ = effective full-volume equivalent. We increase the statistics for halos $\gtrsim 10^{14.5} M_\odot$ by an order of magnitude, from ~ 30 objects to ~ 350. Likewise, we increase the sampling of the most massive and rare $\gtrsim 10^{15} M_\odot$ halos by a factor of thirty, from just three halos in TNG300 to ninety in TNG-Cluster. The combined $M_{\rm halo} > 10^{14} M_\odot$ cluster sample increases from 280 to 636 halos.
In total we simulate 352 clusters. The resulting mass function, stacking on top of the statistics already available from TNG300, is shown above. The TNG-Cluster halos are chosen randomly in 0.1 dex mass bins such that we (i) include all halos above $10^{15} M_\odot$, and (ii) compensate the drop-off of statistics in TNG300 from $10^{14.3}$ to $10^{15.0}$ to produce a flat distribution in halo mass over this range. The resulting sample of simulated clusters overlaps with current observational large-survey programs targeting clusters in the local as well as high redshift Universe, such as PSZ1-cosmo, ACT-DR5, the ROSAT MCXC catalog, SPT-ESC, the XXL 365 catalog, and the eROSITA all-sky catalogs. We also simulate numerous analogs of well known local clusters such as Coma, Perseus and Virgo, as well as rare examples such as the Bullet Cluster and El Gordo.
The TNG-Cluster simulation occupies a unique combination of large volume and high resolution. The figure above places it into context. We show TNG-Cluster (large red diamond) in comparison to other cosmological volumes (circles) and zoom simulation projects (diamonds) at z ~ 0. The x-axis shows the number of simulated massive clusters, with halo masses above $10^{15} M_\odot$. The y-axis shows the resolution, given in terms of the mass of the baryonic mass element. The challenge in simulating massive clusters with high resolution is in moving towards the upper right corner of this diagram.
Institute for Theoretical Astrophysics, Heidelberg University
Max Planck Institute for Astronomy (MPIA), Heidelberg
First results presentation team:
ITA, Heidelberg University
Yonsei University, Seoul
ITA, Heidelberg University
MPIA, Heidelberg
NASA Goddard & University of Maryland
The motion and velocity structure of the ionized gas which makes up the gaseous atmospheres of clusters is a window into the complex plasma physics at play. Turbulence in the ICM is an important factor that may affect the evolution of galaxy clusters and their constituent galaxies. The contribution of turbulent motions to the total pressure in the ICM can be significant, affecting observational inferences of halo mass. Turbulence and non-gravitational motions in general can be driven by many mechanisms, including accretion and merging, AGN feedback, hydrodynamical processes. In Ayromlou et al. (2023) we study the kinematics of the ICM in TNG-Cluster, assessed in terms of line-of-sight velocities, velocity dispersions (shown in the image above), and the velocity structure function. We show that cluster kinematics are potentially connected to halo properties such as formation time as well as galaxy properties such as supermassive black hole energetics. With Truong et al. (2023) we make specific predictions for the observable velocity dispersion of Perseus-like cluster cores for the upcoming XRISM mission.
One of the most unique features of observed galaxy clusters is the existence of extended radio-emission, a sign for the presence of magnetic fields and high-energy electrons in these systems. As IllustrisTNG models the presence and amplification of primordial magnetic fields during the collapse of structure in the early universe, we are able to self-consistently study the magnetic properties of the gas in different environments. In collapsed objects there is an efficient amplification of the magnetic field to about 5 orders of magnitude above the value expected from adiabatic compression alone. The topology of the magnetic field in these regions is consequently strongly correlated with the topology of the gas flows -- in clusters, especially due to mergers. These events produce so-called "radio relics" which are large features of radio emission with distinctive morphologies and shapes. Lee et al. (2023) explores the occurence and properties of radio relics in TNG-Cluster mergers. The image above shows one example: a TNG-Cluster halo, with X-ray emission in the purple background emission, overlaid with radio emission (green contours) and dark matter density (white contours). The resulting library of radio relic features, up to several Mpc in extent, can be used to better understand their physical origins and interpret the results from upcoming radio surveys.
Cool-core (CC) and non-cool-core galaxy (NCC) clusters may represent two classes of galaxy clusters that differ in their central thermodynamical properties, such as temperature and density. While all galaxy clusters are surrounded by the hot intracluster medium, cool-core clusters have a dense, cool core at their center. The cooling time of the gas in the core is relatively short, such that the gas should be rapidly cooling and possibly forming stars. In contrast, non-cool-core clusters have a flatter temperature profile and lack a "cool core". Do mergers transform CC into NCC clusters? What is the role of AGN feedback? Lehle et al. (2023) examines the CC/NCC population in TNG-Cluster, assess whether or not CC/NCC clusters represent two distinct classes, and study the underlying physical differences in central cluster properties. The diverse simulated cluster population can be used to better understand the ultimate origin of CC versus NCC clusters.
AGN feedback can have a significant impact on the evolution of galaxy clusters. In the ICM of galaxy clusters, AGN feedback can redistribute gas and lower halo-scale gas fractions. In Nelson et al. (2023) we present an overview of ICM properties as captured by TNG-Cluster. AGN feedback can also produce cavities or "bubbles" in the hot gas, thought to be initially created by winds or jets from the central supermassive black hole, after which they can rise buoyantly through the ICM. In doing so they deposit energy, heat and disrupting the gas, and drive turbulence. Pressure waves, which are sound waves propagating through the ICM, can be observed through the Sunyaev-Zeldovich effect, or with X-ray measurements. In Pillepich et al. (in prep) we study the diverse morphologies of such pressure and bubble perturbations in the central regions of the ICM. X-ray surface brightness maps show related features, highlighted in their small-scale spatial gradients (images above, left and right). At the same time, massive merging satellite systems produce significant inhomogeneities in the ICM gas, including surface density and/or surface brightness fluctuations. In Rohr et al. (2023) we study how the CGM of cluster satellites can be retained and produce observable signatures.
D. Nelson
The formation of a massive TNG-Cluster halo, starting from 100 million years after the Big Bang and progressing to the present day. By redshift zero, this galaxy cluster has a total mass of 2x1015 times the mass of our sun, and represents one of the most massive objects in the Universe. The visualization first depicts dark matter density, from low-density cosmic voids (black) to the highest density central core of the proto-cluster at high redshift (white). It then shifts to gas metallicity, before transitioning to gas density towards redshift zero. (01:14) [4K version]
D. Nelson
The formation of a massive TNG-Cluster halo, starting from 100 million years after the Big Bang and progressing to the present day. By redshift zero, this galaxy cluster has a total mass of 2x1015 times the mass of our sun, and represents one of the most massive objects in the Universe. The visualization includes eight views (panels) of the same region of space, showing the simultaneous evolution of eight physical properties: dark matter density, gas density, magnetic field strength, stellar density, temperature, gas-phase metallicity, X-ray luminosity (0.5-2 keV), and synchrotron emission. (01:14) [4K version]
D. Nelson
Visualization of the formation of a massive TNG-Cluster halo that, by redshift zero, will be a galaxy cluster with a total mass of 2x1015 Msun, one of the most massive virialized structures in the Universe. The movie starts from 100 million years after the Big Bang and ends at the present day. It transitions from a projection of dark matter density, following the early collapse of filaments and halos in the cosmic web, before showing the magnetic field strength in the gas, which amplifies from a vanishing small primordial seed field due to turbulent, small-scale dynamo processes. (01:14) [4K version]
D. Nelson
Visualization of the formation of the same, massive TNG-Cluster halo that, by redshift zero, will be a galaxy cluster with a total mass of 2x1015 Msun, one of the most massive virialized structures in the Universe. The movie starts from 100 million years after the Big Bang and ends at the present day. It transitions from a projection of gas density, following the early collapse of filaments and halos in the cosmic web, before showing X-ray emission from hot gas in the soft (0.5-2 keV) band. (01:14) [4K version]
D. Nelson
This movie follows the collapse and formation of a massive TNG-Cluster halo through time, tracking it through its high-redshift proto-cluster phase until it reaches a final mass of 2x1015 Msun by redshift zero. The visualization begins showing projected gas density, and then transitions to gas temperature around z=2, where the complex kinematics of the ICM as driven by both mergers and AGN feedback are visible. At z=0 it revolves around the cluster, transitioning from gas density, to the distribution of stars -- where the central BCG, intracluster light, satellites, and nearby galaxies are all visible. It ends with a view of the energy dissipated in hydrodynamical shocks, revealing a network of filamentary and quasi-spherical accretion and feedback-driven shocks surrounding the cluster. (01:40) [4K version]
D. Nelson
Visualization of the gaseous, stellar, and dark matter distribution and physical properties for the third most massive halo in TNG-Cluster at z=0, with a total M200c = 1015.2 solar masses. The main panel shows soft-band (0.5-2 keV) X-ray surface brightness. The smaller panels show dark matter density, stellar density, the Sunyaev-Zeldovich y-parameter map, gas-phase metallicity, magnetic field strength, and neutral HI column density. In this final panel, the location and size of the 100 most massive subhalos are also indicated with circles. In all cases, the white circle marks R200c. [ref]
W. Lee
The most massive galaxy cluster merger in the TNG-Cluster simulation at redshift zero. The total halo mass is 1015.4 solar masses, and from top to bottom the panels show: dark matter column density, gas column density, stellar column density, density-weighted tempearture, mass-weighted magnetic field strength, shock dissipated energy, and the shock Mach number. Each image is 32 Mpc wide and 6 Mpc tall. [large] [ref]
D. Nelson
Gallery of the first 72 halos of TNG-Cluster at z=0, showing soft-band (0.5-2 keV) X-ray luminosity. White circles mark the halo virial radii in each panel. A variety of X-ray morphologies are evident, from relaxed, centrally peaked surface brightness distributions, to un-relaxed and binary merging clusters, triple and multi-component merging systems, and large scale sloshing and gas displacement effects. [large] [ref]
W. Lee
Gallery of multi-wavelength observables of 16 systems from TNG-Cluster. The background image shows the X-ray emission surface brightness, while the green contours show the radio synchrotron emission, and the white contours show the dark matter column density (i.e. weak lensing shear). [large] [ref]
W. Lee
Radio surface brightness projection map (top) and multi-wavelength map (bottom) of a massive merging cluster. The yellow circles mark R500c = 1.7 Mpc and R200c = 2.8 Mpc from the cluster center. The green and white contours on the X-ray map depict the (smoothed) radio surface brightness and the dark matter column density maps. Double radio relics are elongated tangentially with respect to the mass centre and their largest linear size of ~ 4 Mpc. The double radio relics form a non-uniform radio halo-like feature in the z-axis projection (right panel) that fills R500c. [large] [ref]
N. Truong
Gallery of X-ray surface brightness maps of 16 TNG-Cluster systems that are all cool-cores at z=0. The maps have been processed with the Gaussian gradient magnitude (GGM) filter to highlight fluctuations and variations. Ripples, cavities, bubbles, and shock fronts are visible, similar to observations of the Perseus cluster. [large] [ref]
D. Nelson
A large-scale view (roughly 25 Mpc across) of the structure of hydrodynamical shocks surrounding a massive TNG-Cluster halo at z=0. Color shocks the energy dissipation rate (white higher, blue lower). Filamentary as well as quasi-spherical shock surfaces are visible due to the interplay of gas inflows, accretion shocks, and feedback-driven shocks. [large] [ref]
The TNG-Cluster simulation was presented in a 'first results' series of papers:
The heart of galaxy clusters: demographics and physical properties of cool-core and non-cool-core halos in the TNG-Cluster simulation.
Katrin Lehle, Dylan Nelson, Annalisa Pillepich, Nhut Truong, Eric Rohr
A&A (2024) [ads] [arXiv:2311.06333] (Nov. 10, 2023)
The hot circumgalactic media of massive cluster satellites in the TNG-Cluster simulation: existence and detectability.
Eric Rohr, Annalisa Pillepich, Dylan Nelson, Mohammadreza Ayromlou, Elad Zinger
A&A (2024) [ads] [arXiv:2311.06337] (Nov. 10, 2023)
Radio relics in massive galaxy cluster mergers in the TNG-Cluster simulation.
W. Lee, A. Pillepich, J. ZuHone, D. Nelson, M. J. Jee, D. Nagai, K. Finner
A&A (2024) [ads] [arXiv:2311.06340] (Nov. 10, 2023)
An Atlas of Gas Motions in the TNG-Cluster Simulation: from Cluster Cores to the Outskirts.
Mohammadreza Ayromlou, Dylan Nelson, Annalisa Pillepich, Eric Rohr, Nhut Truong, Yuan Li, Aurora Simionescu, Katrin Lehle, Wonki Lee
A&A (2024) [ads] [arXiv:2311.06339] (Nov. 10, 2023)
X-ray inferred kinematics of the core ICM in Perseus-like clusters: insights from the TNG-Cluster simulation.
Nhut Truong, Annalisa Pillepich, Dylan Nelson, Irina Zhuravleva, Wonki Lee, Mohammadreza Ayromlou, Katrin Lehle
A&A (2024) [ads] [arXiv:2311.06334] (Nov. 10, 2023)
Introducing the TNG-Cluster Simulation: overview and physical properties of the gaseous intracluster medium.
Dylan Nelson, Annalisa Pillepich, Mohammadreza Ayromlou, Wonki Lee, Katrin Lehle, Eric Rohr, Nhut Truong
A&A (2024) [ads] [arXiv:2311.06338] (Nov. 10, 2023)
We list here all subsequent papers which have made direct use of TNG-Cluster:
Atacama Large Aperture Submillimeter Telescope (AtLAST) Science: Resolving the Hot and Ionized Universe through the Sunyaev-Zeldovich effect
Luca Di Mascolo, Yvette Perrott, Tony Mroczkowski, Stefano Andreon, Stefano Ettori, Aurora Simionescu, Srinivasan Raghunathan, Joshiwa van Marrewijk, Claudia Cicone, Minju Lee, Dylan Nelson, Laura Sommovigo, Mark Booth, Pamela Klaassen, Paola Andreani, Martin A. Cordiner, Doug Johnstone, Eelco van Kampen, Daizhong Liu, Thomas J. Maccarone, Thomas W. Morris, Amélie Saintonge, Matthew Smith, Alexander E. Thelen, Sven Wedemeyer
arXiv pre-print [ads] [arXiv:2403.00909] (March 1, 2024)
The cooler past of the intracluster medium in TNG-Cluster
Eric Rohr, Annalisa Pillepich, Dylan Nelson, Mohammadreza Ayromlou, Céline Péroux, Elad Zinger
MNRAS (2024) [ads] [arXiv:2410.19900] (Oct. 25, 2024)
X-ray cavities in TNG-Cluster: AGN phenomena in the full cosmological context
Marine Prunier, Julie Hlavacek-Larrondo, Annalisa Pillepich, Katrin Lehle, Dylan Nelson
arXiv pre-print [ads] [arXiv:2410.21366] (Oct. 28, 2024)
ERGO-ML: A continuous organization of the X-ray galaxy cluster population in TNG-Cluster with contrastive learning
Urmila Chadayammuri, Lukas Eisert, Annalisa Pillepich, Katrin Lehle, Mohammadreza Ayromlou, Dylan Nelson
arXiv pre-print [ads] [arXiv:2410.22416] (Oct. 29, 2024)
TNG-Cluster will be publicly released in 2024 on this website, as with all the other TNG simulations.
In the meantime, please get in touch if you have an interest in using the simulation for a project! We are always open to collaborations, new and old alike, and are usually happy to provide access to the simulation data in support of your science.
The TNG-Cluster Catalog is a dynamic, online-accessible table which contains numerous derived properties for all the TNG-Cluster halos. It is continually updated with specialized and contributed analysis of all kinds. An example of a selection of fields is shown above (Table B1 from Nelson et al. 2023). The full, unabridged table will be available for download (in HDF5, FITS, and CSV formats) shortly. Please get in touch in the meantime.
![[large]](/static/tng/media/TNG-Cluster_halo0_overview_4k.jpg){kind=link}
![[large]](/static/tng/media/TNG-Cluster_Xray_gallery_lg.jpg){kind=link}
![[large]](/static/tng/media/TNG-Cluster_radio_relics_xray_gallery_lg.png){kind=link}
![[large]](/static/tng/media/TNG-Cluster_radio_relics_xray_4k.png){kind=link}
![[large]](/static/tng/media/TNG-Cluster_radio_relics_gallery_4k.png){kind=link}
![[large]](/static/tng/media/TNG-Cluster_CCsXray_GGM_2k.png){kind=link}