Illustris

Illustris
NASA / WMAP Science Team · Public domain · source
Name	Illustris
Genre	Astrophysical simulation
Developer	Illustris Collaboration
Initial release	2014
Latest release	2018
Platform	High-performance computing

Illustris is a large cosmological hydrodynamical simulation project that modeled the formation and evolution of galaxies, dark matter, and baryonic structures in a representative volume of the Universe. The project combined gravitational dynamics, gas hydrodynamics, radiative cooling, star formation, and feedback processes to produce synthetic universes for comparison with observational surveys and theoretical models. Illustris produced datasets used by researchers associated with observatories, institutes, and surveys to study galaxy morphology, large-scale structure, and baryon cycles.

Overview

The project was led by a collaboration including researchers from institutions such as the Harvard–Smithsonian Center for Astrophysics, Max Planck Institute for Astrophysics, MIT, Princeton University, University of California, Berkeley, Yale University, University of Cambridge, Columbia University, Stanford University, and Kavli Institute for Cosmology. Illustris used supercomputers at centers like the National Energy Research Scientific Computing Center, Lawrence Livermore National Laboratory, Harvard Odyssey, Leibniz-Rechenzentrum, and Jülich Supercomputing Centre. The project interfaced with datasets and surveys including Sloan Digital Sky Survey, Hubble Space Telescope, Chandra X-ray Observatory, Spitzer Space Telescope, and GALEX for validation. Collaborators published results in journals such as Nature, Monthly Notices of the Royal Astronomical Society, Astrophysical Journal, and Astronomy & Astrophysics.

Simulation Design and Methods

Illustris employed a moving-mesh hydrodynamics code developed by teams at Heidelberg University and Princeton, building on methodologies related to codes like AREPO and concepts from Smoothed Particle Hydrodynamics research groups at Durham University and Max Planck Society. The simulation used cosmological parameters consistent with measurements from missions and experiments such as Planck (spacecraft), Wilkinson Microwave Anisotropy Probe, and constraints tied to the Lambda-CDM model as interpreted in frameworks used by WMAP. Initial conditions were generated using perturbation theory techniques similar to those applied in analyses by teams at University of Chicago and Carnegie Institution for Science, with transfer functions informed by software developed at CERN and Los Alamos National Laboratory. Physics modules implemented radiative cooling and heating following prescriptions from studies at KIPAC, star formation subgrid models inspired by work at Caltech and University of Illinois Urbana-Champaign, stellar evolution yields calibrated against Geneva Observatory and Padova (stellar models), and feedback routines referencing active galactic nucleus models from groups at Max Planck Institute for Astronomy and University of Tokyo.

Large halo and galaxy catalogs were identified using halo finders originally developed at University of Washington, Flatiron Institute, and University of Groningen, while merger trees utilized algorithms previously applied in projects like Millennium Simulation and EAGLE (project). Data analysis pipelines integrated tools and libraries similar to those from Astropy, NumPy, SciPy, and visualization approaches used by Matplotlib and groups at Jet Propulsion Laboratory.

Results and Scientific Findings

Illustris produced mock observables enabling comparisons to results from the COSMOS survey, CANDELS, SDSS, and GAMA survey. Key findings addressed galaxy stellar mass functions, rotation curves, morphological transformations, and baryon fractions in halos compared against inferences from Chandra, XMM-Newton, and lensing measurements from teams connected to CFHTLenS and DES (Dark Energy Survey). The simulation reproduced phenomena investigated by researchers at Max Planck Institute for Astrophysics and Observatoire de Paris such as bimodal color distributions, star formation rate density evolution previously mapped by GALEX and Herschel Space Observatory, and mass-metallicity relations similar to work at NOAO and STScI.

Illustris elucidated processes like gas accretion along filaments identified in studies at Princeton, angular momentum acquisition compared to theories from Tully–Fisher research groups, the role of mergers analyzed in the context of results from Hubble Deep Field teams, and AGN quenching mechanisms with reference to models developed at University of Oxford and University of Bologna. The outputs informed comparisons with theoretical frameworks tested at Perimeter Institute and Institute for Advanced Study.

Applications and Impact

Datasets from Illustris were used by researchers at Cambridge University Observatory, Institut d'Astrophysique de Paris, ETH Zurich, Barcelona Supercomputing Center, University of Toronto, University of Edinburgh, Monash University, and Australian National University for projects ranging from galaxy morphology classification to interpreting observations from instruments like Very Large Telescope, Atacama Large Millimeter Array, and Keck Observatory. The public release fostered cross-collaboration with groups at Flatiron Institute, Harvard-Smithsonian Center for Astrophysics, and Princeton and supported teaching applications in courses at University College London and University of Michigan. Illustris influenced successor simulations and initiatives at Max Planck Institute, Lawrence Berkeley National Laboratory, and Space Telescope Science Institute.

Limitations and Criticism

Critiques arose from researchers at institutions including Cambridge University, Leiden University, University of Colorado Boulder, and Rutgers University who noted tensions between Illustris predictions and observations such as overproduced stellar masses in group-scale halos and discrepancies in circumgalactic medium properties measured by COS (HST) teams. Methodological limitations were discussed in relation to subgrid physics choices compared with approaches used in EAGLE (project), Horizon-AGN, and semi-analytic models from Kentucky University-affiliated groups. Computational constraints tied to resources at NERSC and trade-offs similar to those in Millennium Simulation led to debates about resolution, box size, and the fidelity of feedback prescriptions developed at MPIA and Rutgers.

Related Projects and Successors

Illustris is part of a lineage that includes related projects and successors such as EAGLE (project), Horizon-AGN, Millennium Simulation, Bolshoi (simulation), Aquarius Project, FIRE (project), IllustrisTNG, SIMBA (simulation), and initiatives at Max Planck Institute for Astrophysics and Flatiron Institute. These projects continued development of hydrodynamic codes like AREPO, improved subgrid physics informed by groups at ETH Zurich and University of California, Santa Cruz, and exploited facilities including XSEDE and PRACE for larger-volume, higher-resolution runs.

Category:Cosmological simulations