[MITgcm-support] Forcing theta within gchem

[Martin Losch]

Hi Roland,

the “generic” place to add something to the forcing of theta (i.e. the tendency terms of theta) is in “apply_forcing.F” (old code used external_forcing.F), or since your a dealing with latent heat, external_forcing_surf.F

gchem_forcing_sep is called at the end of the timestep, so you need to save the forcing terms for theta that you compute there for the next time step (which is a pain, because it would require saving them in the restart/pickup file) and then apply it in apply_forcing_t
If you need to use the gchem-package for thermal forcing, why don’t you do your calculations in gchem_calc_tendencies, which is called in the beginning of the time step (before thermodynamics, where temp_integrate and apply_forcing_t are called.

Martin


> On 20 Apr 2015, at 20:27, Roland Young <Roland.Young at physics.ox.ac.uk> wrote:
> 
> Hi,
> 
> I was wondering at what point in the code a tracer that forces the theta field should apply its forcing? I have various tracers which I want to force the theta field via latent heat exchanges. I calculate the relevant phase changes as a subroutine to gchem_forcing_sep, and at present I calculate the latent heat exchange there too. However if I try to apply the forcing to theta there (simply as theta = theta + dtheta, with appropriate looping), the total tracer mass rapidly increases (10% over a day). I’m pretty sure I’m updating theta in the wrong place, or doing it incorrectly. 
> 
> I could calculate the latent heat forcing in one timestep and apply it in the next in external_forcing_t - is that valid or are the tracers advected before that? Is there a way to update the prognostic fields within the gchem interface? I can’t find any examples that do it (the only one I can find updates CFC concentrations, but these don’t seem to force the model temperatures).
> 
> Thanks,
> 
> Roland
> 
> 
> 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support