Abnormal values of TKE dissipation near surface

Question

Hi,

In a hydrodynamic simulation with 60-second timestep, the simulated TKE dissipation rate (output as 'DISSIPATION') showed frequent zero values near surface. Then I reduced the simulation timestep to 30 seconds, which helped to eliminate zero 'DISSIPATION' near surface but led to insufficient vertical mixing and unstable thermal structure that should not be there as shown by measured data. I tried to enhance vertical mixing by reducing coefficients MIXING_COEFF_CCT and MIXING_COEFF_CT, the zero 'DISSIPATION' issue showed up again.

Could anyone suggest a solution to this issue?

Thanks in advance.

Wencai Zhou

Earth Research Institute, University of California Santa Barbara.

Answer 1

Hi Wencai,

The issue here is that the dissipation output doesn't really have a 1:1 relationship with what would be measured in the field.  

The AEM3D dissipation output is just the reduction of TKE at the end of the mixing timestep (ie after mixing has occurred).  As you've discovered this is heavily dependent on timestep because if the timestep is large enough all TKE will be consumed by mixing.

It also doesn't account for TKE loss terms implicitly inherent in the parameterisations. For example, the shear mixing term has the coefficient Cs which is really an efficiency for the conversion of bulk KE to TKE with the inefficiency essentially being a dissipation.

The dissipation output is really meant more as a diagnostic output for the individual components in the TKE parametrisation.

Chris