SIMBA (simulation)

SIMBA (simulation)
Name	SIMBA
Initial release	2019
Programming language	C, C++
License	proprietary
Platform	Unix-like

SIMBA (simulation) is a cosmological hydrodynamic simulation suite designed to model galaxy formation and evolution across cosmic time. It couples gravity, gas dynamics, radiative cooling and heating, star formation, chemical enrichment, and active galactic nucleus feedback within a cosmological volume to study the coevolution of galaxies, black holes, and large-scale structure. The project builds on prior simulation programs and engages with observational programs to interpret surveys of galaxies, quasars, and the intergalactic medium.

Overview

SIMBA integrates N-body methods for dark matter with mesh-free finite-mass hydrodynamics to follow baryons in volumes that span from tens to hundreds of comoving megaparsecs. It targets comparisons with surveys such as the Sloan Digital Sky Survey, COSMOS (survey), CANDELS, ALMA, and Hubble Space Telescope deep fields by predicting properties measurable by instruments like JWST, Chandra X-ray Observatory, and Spitzer Space Telescope. The simulation is tuned to reproduce empirical relations established by studies involving the Tully–Fisher relation, Faber–Jackson relation, and the stellar mass–halo mass relation while also engaging with constraints from the Lyman-alpha forest, Planck (spacecraft) cosmological parameters, and the Galaxy And Mass Assembly project.

Physical and Numerical Methods

SIMBA employs a gravity solver based on a TreePM scheme adapted from prior codes used in simulations such as GADGET and GIZMO. Hydrodynamics uses a meshless finite-mass method that inherits numerical techniques developed in GIZMO to reduce numerical diffusion and improve conservation compared with classical particle approaches like those in SPH (smoothed particle hydrodynamics). Time integration leverages adaptive hierarchical timestepping similar to methods in AREPO and the Enzo (software) community, while parallelization follows strategies implemented on high-performance computing systems such as NERSC and national supercomputing centers like XSEDE and PRACE allocations. Radiative cooling tables incorporate atomic and molecular processes calibrated against calculations from groups associated with CLOUDY and CHIANTI (database), while heating includes a metagalactic ultraviolet background consistent with models from the Haardt & Madau family.

Subgrid Physics and Feedback Models

SIMBA implements subgrid modules for star formation, stellar feedback, metal enrichment, and black hole physics informed by empirical and theoretical work from groups behind Kennicutt–Schmidt law studies, Kroupa initial mass function analyses, and supernova yield calculations associated with research from the Woosley & Weaver models. Stellar feedback includes kinetic winds and mass-loading scalings calibrated to match observations compiled by teams working on SDSS, GAMA, and DEEP2. Black hole seeding and growth utilize torque-limited accretion prescriptions influenced by analytic models from the Shakura–Sunyaev framework and comparisons to Bondi-like accretion explored in literature including work on Active Galactic Nuclei. AGN feedback in SIMBA comprises both radiative-mode winds and jet-mode kinetic feedback patterned after phenomenology studied in Seyfert galaxies, radio galaxies, and observations from the Very Large Array; these implementations aim to reproduce quenching trends identified in surveys like COS-Halos and cluster studies such as those with XMM-Newton.

Validation and Comparison with Observations

SIMBA’s outputs are compared against multiwavelength datasets, including stellar mass functions from SDSS and ZFOURGE, star formation rate densities from compilations tied to GOODS, gas fractions probed by ALFALFA and PHIBSS, and black hole–host scaling relations established by studies using HST imaging and Keck Observatory spectroscopy. The simulation is validated by reproducing galaxy color bimodality seen in COMBO-17 and size–mass relations discussed in the CANDELS collaboration. Its predictions for circumgalactic medium absorption are tested against quasar sightline surveys like COS-Halos and the Keck Baryonic Structure Survey, and comparisons to cosmic X-ray background measurements leverage data from Chandra and XMM-Newton.

Major Results and Applications

Key results from SIMBA include self-consistent emergence of galaxy quenching correlated with black hole growth, realistic cold gas fractions across stellar mass and redshift, and predictions for molecular gas scaling relations that inform interpretations from ALMA and NOEMA. The simulation has been used to study the buildup of the mass–metallicity relation relevant to work by the MaNGA and SAMI teams, the evolution of the baryon cycle connected to COS-Halos findings, and feedback-driven outflows comparable to observations in MUSE and integral field spectrograph studies. SIMBA’s cosmological volumes enable applications to clustering and halo occupation modeling that interface with analyses from the Baryon Oscillation Spectroscopic Survey and large-scale structure probes used by eBOSS.

Development History and Code Architecture

Development of SIMBA followed an evolution of codes and modules contributed by researchers with backgrounds in projects like GADGET, GIZMO, and earlier cosmological efforts associated with the Illustris and EAGLE collaborations. The code architecture emphasizes modular subgrid physics, scalability on MPI-based clusters, and on-the-fly analysis tools to produce synthetic observations comparable to data from HST, JWST, and ground-based facilities such as Subaru Telescope. Ongoing development involves engagement with community efforts in reproducible simulation science and interoperability with analysis frameworks used by teams at institutions such as Kavli Institute for Cosmology, Max Planck Institute for Astrophysics, and national laboratories.

Category:Cosmological simulations