[MITgcm-devel] pkg/frazil to pkg/seaice hook

[Gael Forget]

Ian,

I am transferring this thread to devel list, in case somebody is interested.

I would insist we start with what we discussed, rather than anything more 
complicated (for which a justification may emerge at some point, or not).

Recalling what I believe we agreed on a week ago : right now, pkg/frazil acts on THETA(1), 
and then pkg/seaice SEAICE_frazilFrac term will make ice in case pkg/frazil results in 
THETA(1) < freezing point. That may be all we justifiably need by the way. To by-pass
that supercooling step, it should suffice to add the "heat flux" out of pkg/frazil
to either a_QbyATM_cover or a_QbyATM_open (depending on your "vision" of how 
this term should affect the time stepping of AREA) right before they get converted 
to ice meters units. Either way you really just need to add one line of code in 
seaice_growth.F (plus the run time switch, with a suitable name -- see my previous email).

Cheers,
Gael




On Mar 9, 2012, at 4:08 PM, Ian Fenty wrote:

> What I mean is that we have two sources of "frazil"  - that from Dimitris' package and that from T < T_f.  Right now we are lumping the handling of T < T_f as part of r_QbyOCN but I envision a separation of a_QbyOCN into a frazil term: a_QbyOCN_frazil and a turbulent melting term: a_QbyOCN_turb.  
> 
> Adding Dimitris' term within growth would involve adding his contribution to the term a_QbyOCN_frazil.  We need a name for the run-time switch to include or exclude this interior frazil term.  I suggest we use SEAICE_useInteriorFrazil.  
> 
> Handling the expansion of ice cover due to the frazil should be done in the same way as we handle open water growth.  Hence,  d_HEFFbyOCN_frazil will be added to d_HEFFbyATM_open to determine the positive ice cover expansion rate.
> 
> -Ian
> 
> On Mar 9, 2012, at 12:48 PM, Gael Forget wrote:
> 
>> I dont get what you mean by "we will soon be treating all T < T_f as frazil",
>> but in any case useFrazil is not available (that is the one for pkg/frazil).
>> Gael
>> 
>> 
>> 
>> On Mar 9, 2012, at 3:38 PM, Ian Fenty wrote:
>> 
>>> Gael,
>>> How about SEAICE_useFrazilPackage or useInteriorFrazil instead of useFrazil because we will soon be treating all T < T_f as "frazil" regardless of whether we use Dimitris' new frazil package?
>>> -Ian
>>> 
>>> 
>>> On Mar 9, 2012, at 12:30 PM, Gael Forget wrote:
>>> 
>>>> Hi Hong,
>>>>> #undef SEAICE_VARIABLE_SALINITY
>>>>> How to keep SEAICEadvSalt on?
>>>> you dont need to advect a constant salinity.
>>>> Cheers,
>>>> Gael
>>>> 
>>>> 
>>>> _______________________________________________
>>>> MITgcm-devel mailing list
>>>> MITgcm-devel at mitgcm.org
>>>> http://mitgcm.org/mailman/listinfo/mitgcm-devel
>>> 
>> 
>