MPAS-A Gap Flow Experiment: Exploring the Coriolis Effect

This set of experiments explores how planetary rotation, via the Coriolis effect, influences atmospheric flow through a topographic gap in an idealized setup using the MPAS-Atmosphere (MPAS-A) model.

All cases are based on the standard MPAS-A gap flow configuration, with varying rotational and initial condition parameters to isolate and assess Coriolis-induced dynamics.

1. With and without Coriolis effect

1. a) control : No coriolis

2. b) sensitivity: Yes coriolis

2. Doubling Initial Zonal wind Uo

a) control : No coriolis; Uo =10 m/s

b) sensitivity: No coriolis; Uo =20 m/s

MPAS-A Gap Flow Experiment: Horizontal Diffusion Sensitivity Tests

This set of experiments investigates the impact of different horizontal diffusion strategies on idealized gap flow simulation results using the MPAS-A model with a 120-km small-Earth grid (mpasa120). The experiments vary only in the choice and configuration of horizontal eddy viscosity via the mpas_horiz_mixing parameter and associated diffusion coefficients.

1. Control: Smagorinsky-type Flow-dependent Mixing  ( mpas_horiz_mixing = '2d_smagorinsky' (default))

• • Horizontal diffusion coefficient is diagnosed based on local flow deformation (strain and shear).

  • Smagorinsky-type mixing activates primarily where small-scale turbulent features exist, allowing scale-selective damping.

  • Suitable for real-data and high-resolution simulations due to minimal artificial damping in quiescent regions.

2.  Sensitivity: Constant value for the domain  - mpas_horiz_mixing = '2d_fixed'; mpas_h_mom_eddy_visc2 = 0.0D0

• • Represents a zero-diffusion baseline to isolate natural numerical diffusion or test model stability in absence of explicit horizontal damping.

  • Useful as a reference to assess how much diffusion is introduced inherently by the numerical scheme alone.

3. Fixed Coefficient Diffusion (Constant 2nd-order Mixing) - mpas_horiz_mixing = '2d_fixed'; mpas_h_mom_eddy_visc2 = 200.0D0 (m²/s)

• • Applies a uniform horizontal viscosity across the entire domain.

  • This approach ensures consistent damping but may overly suppress physical features, especially at large scales.

  • Primarily used in idealized tests where predictable diffusion is desirable.