AstroWiki@UZH

GIZMO is a flexible, massively-parallel, multi-physics simulation code. The code lets you solve the fluid equations using a variety of different methods – whatever is best for the problem at hand.

It introduces new Lagrangian Godunov-type methods that allow you to solve the fluid equations with a moving particle distribution that is automatically adaptive in resolution and avoids the advection errors, angular momentum conservation errors, and excessive diffusion problems that limit the applicability of “adaptive mesh” (AMR) codes, while simultaneously avoiding the low-order errors inherent to simpler methods like smoothed-particle hydrodynamics (SPH).

Meanwhile, self-gravity is solved fast, with fully-adaptive gravitational softenings. And the code is massively parallel — it has been run on everything from a Mac laptop to >1 million CPUs on national supercomputers.

The code is originally descended from GADGET , and has retained many naming/use conventions as well as input/output cross-compatibility — if you have code or experience from GADGET, it should immediately be compatible with GIZMO.

* Hydrodynamics: with arbitrarily chosen solvers (new Lagrangian methods, or SPH, or fixed-grid Eulerian methods)

• Magnetic fields: ideal or non-ideal MHD, including Ohmic resistivity, ambipolar diffusion, and the Hall effect
• Cosmological integrations: large volume and cosmological “zoom in” simulations, with on-the-fly group finding
• Radiative heating/cooling & chemistry: including species-dependent cooling and pre-tabulated rates from 10-1e10 K, or hooks to external chemistry and cooling packages
• Galaxy/Star/Black hole formation and feedback: explicit modules for star and black-hole formation and feedback in galaxy-scale and cosmological simulations including detailed explicit and common sub-grid models
• Individual Star/Planet sink-particle formation: with explicit proto-stellar evolution and feedback, designed for individual star and planet formation simulations
• Non-standard cosmology (e.g. arbitrarily time-dependent dark energy, Hubble expansion, gravitational constants), and the development code includes in-testing modules for self-interacting and scalar-field dark matter;
• Self-gravity with adaptive gravitational resolution: fully-Lagrangian softening for gravitational forces, with optional sink particle formation and growth, and ability to trivially insert arbitrary external gravitational fields
• Anisotropic conduction & viscosity: fully-anisotropic kinetic MHD (e.g. Spitzer-Braginskii conduction/viscosity), or Navier-Stokes conduction/viscosity, and other non-ideal plasma physics effects
• Subgrid-scale turbulent “eddy diffusion” models, passive-scalar diffusion, and the ability to run arbitrary shearing-box or driven turbulent boxes (large-eddy simulations) with any specified driving spectrum
• Dusty fluids: explicit evolution of aerodynamic particles in the Epstein or Stokes limits, with particle-particle collisions, back-reaction of dust on gas, Lorentz forces on charged particles, and more
• Degenerate/stellar equations-of-state: appropriate Helmholtz equations of state and simple nuclear reaction networks
• Elastic/plastic/solid-body dynamics: arbitrary equations-of-state with pre-programmed modules for common substances (e.g. granite, basalt, ice, water, iron, olivine/dunite), including elastic stresses and plastic behavior
• Radiation-hydrodynamics: full and flexible, with a variety of modular solvers and the ability to trivially add or remove frequencies to the transported set (pick and choose which wavebands you want to evolve)
• Cosmic Ray transport and magneto-hydrodynamics. Modular choices of solvers for cosmic ray transport, and full coupling to gas including cooling and adiabatic, diffusion, Alfven-wave coupling, streaming, injection, and more.
• Other in-development features include detailed nuclear networks, relativistic hydrodynamics, and more

GIZMO-Public:

The publicly-available version of the code (hosted on a Bitbucket repository). It has full functionality for the “core” physics, including almost all physics listed above. This is intended for a wide range of science applications, as well as learning, testing, and code methods studies. A few modules are not yet in the public code, because they are in active development and not yet de-bugged or tested at the level required for use “out of the box” (but will be made public as soon as this stage is reached). In a few cases it is because they involve proprietary code developed by others with their own collaboration policies, which must be respected (see GIZMO User Guide for details).

GIZMO User’s Guide:

Everything you need to know to run and/or analyze simulations with GIZMO. Includes a suite of examples and test problems as well. If you have questions/bug reports/etc that go beyond the user guide, please join the GIZMO Google Group.

The GIZMO code/methods paper explains the hydrodynamic methods in detail and explores them in a large suite of test problems. Because there are many different physics modules in GIZMO, there are different methods papers for different code modules (for example, the GIZMO-MHD code/methods paper, which explains the implementation of ideal magneto-hydrodynamics). You should always read the appropriate methods paper[s] before using any given module, and cite them in papers using them! Links to the appropriate papers for each part of the code are in the User Guide. A broader overview of different numerical methods (suitable for an introduction before diving into the more detailed code papers) is available from one of my talks here.

If you wish to contribute to active development or new physics in GIZMO, you can branch off the public code, and eventually push to the development (not public) version of the code which is hosted on a separate Bitbucket repository here. For permissions you will need to contact me directly; if you do, please include an explanation of your plans and why you want to be involved in development.

• To download public version of GIZMO:

Download GIZMO