Unified Z-grid Icosahedral Model (UZIM)

Institution: Colorado State University, Fort Collins, CO

Model Metadata

• Metadata Entry Form
• View Metadata

Description:

Nonhydrostatic vorticity-divergence dynamical core for global cloud resolving models. Model equations are discretized on the icosahedral Z-grid using 3th-order Adams-Bashforth scheme in time and finite-defferences in space. The dynamical core predicts the vorticity (vertical component) and divergence directly on the Z-grid defined at cell centers.

Physics is based on SAM (Marat Khairoutdinov's cloud-resolving model).

Key words:

Members of this modeling group during DCMIP-2012:

UZIM model equations.

UZIM model is based on the unified system described by (Arakava, Konor, 2009).  The unified system is a nonhydrostatic system that is designed for global cloud resolving models which filters vertically propagating sound waves while allowing elasticity due to thermal expansion. 

To filter acoustic waves the density is devided into two parts:

where

The system obtained is such a way is

Information on the computational grid:

The horizontal grid is based on an icosahedral hexagon-pentagon discretization. This grid have the property that all neighbors of a given cell lie acroll cell walls. There are no "vertex neighbors". We choose the Z-grid because:

• It does a good job with geostrophic adjustment;
• It has no computational modes.

The grid generation procedure contains of two steps. At first a "raw" grid is generated. "Raw" grid obtained using triangulation of the icosahedron which inscribed inside a sphere. Otherwise each face of the icosahedron is subdevided into four triangular faces by bisecting the edges of the triangles. The vertuces of the new triangles are projected onto the surface of the sphere.

The "raw" grid is optimized using a variational method. The goal of the optimization is to minimize the distance between the mid-point of the cell wall and the point where the grid segment intersects the cell wall, by displacing the cell centers relative to those of the raw grid (Augenbaum, Peskin, 1985).

Grid resolution and total number of the nodes is presented in the table below.

Grids 5,6 and 7 are used for test calculations as the model with low, medium and hight resolution.

Test cases.

All of the tests presented below were obtained using hydrostatic version of the UZIM model without orography.

1. Pure advection

  1.1 3D deformational flow

  1.2 Hadley-like meridional circulation

2. Impact of orography on a non-rotating planet

3. (Non)-Hydrostatic Gravity Wave

  3.1 Perturbed potential temperature

4. Baroclinic instability

  4.1 Dry baroclinic instability on a small planet with dynamic tracers

  4.2 Moist variant with large-scale condensation

  4.3 Moist variant driven by "Simple Physics"

5. Idealized tropical cyclone

  5.1 Coupling to "Simple Physics"

  5.2 Coupling to Model's Full Physics Aqua-Planet mode

 Bibliography:

• Augenbaum, Jeffrey M.; Peskin, Charles S. On the construction of the Voronoi meth on the sphere // Journal of Computational Physics, Vol. 59, p.177