[MITgcm-support] New user with problems with ptracers

[Martin Losch]

Hi Andrea,

your version of the code should work, but an upgrade wouldn't hurt 
either ...
I have two guesses of what could be going on:
1. There are two routines, ptracers_forcing.F and 
ptracers_forcing_surf.F, where the passive tracers get their forcing 
(the surface forcing is set in the latter and that is added to the 
overall forcing in the former). For testing purposes the forcing for 
all tracers is identical to that of salinity (surfaceTendencyS in 
ptracers_forcing_surf.F), so that all passive tracer are identical to 
salinity, except that they are passive. If you want to do something 
sensible with ptracers, you should, in ptracers_forcing_surf.F set 
surfaceTendencyPtr=0 or to whatever makes sense in your configuration, 
for example, you could specify a source of tracer (polution by a 
river?) here. For deep sources you need to modify ptracers_forcing.F. 
Also you need to specify an initial conditions file for your passive 
tracer, otherwise it's all zeros (but you probably have done that).
2. by default the turbulent mixing (diffusion) coefficients of all 
passive tracers equal those of salinity. There is little reason to 
assume different turbulent mixing coefficients for different passive 
tracers (even if they have different molecular diffusion coefficients). 
At the moment it's even quite difficult, or even impossible, to make 
the VERTICAL diffusivity (ptracers_diffkr) for ptracers different from 
that of salinity, especially if you are using a mixed layer model 
(KPP). I would not try that. The HORIZONTAl diffusivity parameters 
(ptracers_diffkh/diffk4) should have an effect.
I agree that the availability of all these runtime parameters is a 
little confusing.

I am quite confident that the first point (removing salinity forcing in 
ptracers_forcing_surf) will fix your problem.
Martin

On Sep 21, 2004, at 5:20 PM, cimatori at libero.it wrote:

> Hi, I'm a new MITGCM user from Italy. I'm working for my third year 
> thesis at the OGS in Trieste (Italy), with Stefano Querin and Cosimo 
> Solidoro.
> We want to set up a model for Trieste Gulf using ptracers package too, 
> we tried using checkpoint 52.e_pre, with minor changes but we get 
> strange results. All the tracers behave just like salt (or nearly like 
> salt), even if we changed diffKh by 10^3. We also get always the same 
> results (and in the same order), changing parameters.
> Do you think that trying with a newer version could work?
> Thanks.
> Andrea Cimatoribus
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
>
Martin Losch // mailto:mlosch at awi-bremerhaven.de
Alfred-Wegener-Institut für Polar- und Meeresforschung
Postfach 120161, 27515 Bremerhaven, Germany
Tel./Fax: ++49(471)4831-1872/1797
http://www.awi-bremerhaven.de/People/show?mlosch