Climate Model Resolutions

The numerical formulation of the equations included on the current climate models is a subject of research. The climate model resolution indicates information on this discretization process both in time and space. The two main discretization methods are the finite difference method to formulate the differential equations and the spectral method which allows analyitical expressions for the spaces derivatives. Here is specified several typical resolutions of both Spectral and Finite differences methods.

Spectral Model Grid

The spectral models uses a different methodology for space discretization based on spectral expansion (it can be considered a Galerkin approach)

More information can be found ECMWF and in the referece ¹

In general the main advantage of spectral models is that space derivatives can be calculated analytically instead of numerically. Also in general the pole problem is not present. However the main disadvantages are the difficulties to include directly typical space grid physical processed (althought it can be done with a transformation method). It is mainly designed for global models and not limited-area models.

Typical Model Spectral Resolutions

Finite Methods

As a comparative example here are some typical grids of the UKCA model (Hadley Centre version of the Unified Model at UK).

Climate Models used by UKCA

The UKCA model runs within the Hadley Centre version of the Unified Model. There are a number of different versions of this climate model at a number of resolutions which are used for UKCA development.

Up to table of contents