General Circulation Model

A general circulation model (GCM) is a computer model that simulates the motion of the atmosphere, the ocean, or both, by solving the equations of fluid flow and energy on a three-dimensional grid around the planet. GCMs are the main tools used to project future climate.

The model divides the atmosphere into grid boxes, typically tens to a hundred kilometers wide with dozens of vertical layers, and steps forward in time using the equations of motion, a form of the Navier–Stokes equations, together with the laws of energy and mass conservation. Processes smaller than the grid, such as clouds and convection, are parameterized from simpler rules. The result reproduces features such as the Hadley cell and the jet streams.

Norman Phillips ran the first successful general circulation experiment in 1956, and Syukuro Manabe and Richard Wetherald made early calculations of the effect of carbon dioxide in the 1960s. Today's coupled models link atmosphere, ocean, land and sea ice, and the largest Earth system models add the carbon cycle and vegetation. Manabe and Klaus Hasselmann shared the 2021 Nobel Prize in Physics for work that laid the basis of such models.

Groups around the world run their models under shared protocols, in the Coupled Model Intercomparison Project, so that results can be compared and combined in the assessments of the IPCC. They estimate climate sensitivity and the response to radiative forcing. Coarse grids limit regional detail, which is addressed by statistical downscaling, and cloud processes are a main source of uncertainty.

• Earth System Model — the broader class that adds biogeochemistry
• Navier-Stokes Equations — the equations of flow behind the models
• Climate Sensitivity — a key quantity the models estimate