GFDL_atmos_cubed_sphere
GFDL_atmos_cubed_sphere copied to clipboard
NOAA-GFDL
Implement Rayleigh damping on vertical velocity only through semi-implicit solver
Is your feature request related to a problem? Please describe. Damping of the horizontal velocity, as is done currently, is more consistent with a hydrostatic model. A nonhydrostatic model can simply damp the vertical velocity directly through the semi-implicit solver to prevent reflections from the upper boundary, but it requires a much shorter damping time scale. This is generally useful, but may also solve a spread problem near the model top in the ensemble system.
Describe the solution you'd like I want to implement the Klemp 2008 (MWR) sponge mechanism on vertical velocity through the vertical solver with a new namelist parameter tau_nh which is the damping time scale in seconds (~0.2 seconds). This provides a very effective, relatively thin, sponge layer.
Describe alternatives you've considered I have tried a damper on perturbation pressure through the solver (since that fits the boundary conditions), but that was not very effective. The results with the w-sponge have been robust.
Additional context Add any other context or screenshots about the feature request here.
Hi, Kevin. Thank you for working on this.
After some thought I remembered that since perturbation pressure is not a prognostic variable, damping it will only have a very indirect effect on vertical velocities, and will not remove the source of the strongest vertical velocities which are typically thermally-driven.
I can take a look once you have a PR ready.
Lucas
On Tue, Dec 6, 2022 at 9:32 AM KevinViner-NOAA @.***> wrote:
Is your feature request related to a problem? Please describe. Damping of the horizontal velocity, as is done currently, is more consistent with a hydrostatic model. A nonhydrostatic model can simply damp the vertical velocity directly through the semi-implicit solver to prevent reflections from the upper boundary, but it requires a much shorter damping time scale. This is generally useful, but may also solve a spread problem near the model top in the ensemble system.
Describe the solution you'd like I want to implement the Klemp 2008 (MWR) sponge mechanism on vertical velocity through the vertical solver with a new namelist parameter tau_nh which is the damping time scale in seconds (~0.2 seconds). This provides a very effective, relatively thin, sponge layer.
Describe alternatives you've considered I have tried a damper on perturbation pressure through the solver (since that fits the boundary conditions), but that was not very effective. The results with the w-sponge have been robust.
Additional context Add any other context or screenshots about the feature request here.
— Reply to this email directly, view it on GitHub https://github.com/NOAA-GFDL/GFDL_atmos_cubed_sphere/issues/229, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMUQRVDII547LFSPZNA7CHDWL5FAVANCNFSM6AAAAAASVSMI3I . You are receiving this because you are subscribed to this thread.Message ID: @.***>
