Idealized test cases for primitive equation based dynamical cores have become a standard assessment tool for the strengths and weaknesses of current dynamics packages in GCMs. But although tests like the Held-Suarez test or the breaking polar vortex pattern show important model characteristics, additional test cases will be needed for future model intercomparisons.

The talk will propose a new standard test sequence for dynamical cores with pressure-based vertical coordinates. The test suite comprises two parts. First, the model will be initialized with steady state, balanced initial conditions that are an analytical solution to the primitive equations. This initial flow consists of a zonally symmetric basic state with two jets in midlatitudes and a realistic temperature profile. The careful design guarantees static, inertial and symmetric stability properties, but is unstable with respect to baroclinic or barotropic instability mechanisms. Model integrations over thirty days reveal how well the model can keep its initial state. Second, a well-resolved small amplitude Gaussian hill perturbation is superimposed on the initial state. This triggers baroclinic wave activities that lead to explosive cyclogenesis.

The test sequence has been applied to NCAR's newest dynamical cores (CAM2 model framework), to NASA's finite volume model (developed at GSFC) and the icosahedral model GME (developed at DWD). In addition, model results based on cubed meshes and reduced grids are available. The intercomparison reveals interesting characteristics that have already led to model improvements. The diagnostics include error norm statistics, grid point data comparisons and wave number analyses.

In the future the test will serve as a standard test case for adaptive grid simulations that are capable of tracking localized phenomena at high resolutions. A brief outline of this project will be given.