6. Boundary conditions and time intergration methods

6.1. Boundary Conditions

NUMA reads in boundary conditions via the file numa3d.in through the variables x_boundary, y_boundary, and z_boundary. NUMA allows for the following types of boundary conditions:

  • 0 = do nothing boundary conditions

  • 2 = no-slip boundary condition

  • 4 = free-slip boundary condition

  1. Do Nothing Boundary Condition

    Setting z_boundary=(0,0) tells h-NUMO to ignore these boundaries.

  2. No-slip boundary condition

    This boundary condition cancels all direction velocity at a boundary (\(\mathbf{u} = 0\)).

  3. No-slip or No-flux Boundary Condition

    This boundary condition cancels the normal velocity at a boundary (\(\mathbf{n}\cdot\mathbf{u} = 0\)).

6.2. Time-Integrators

h-NUMO is equipped with a suite of time-integrators which include:

  • Explicit 5-stage 3rd Order Runge-Kutta Methods (used only for the barotropic equations),

  • Two-level or predictor-corrector method for both barotropic and baroclinic equations.