[MITgcm-devel] another bug in growth.F ?

[Dimitris Menemenlis]

Jinlun, thank you for confirmation.

Patrick, Martin, and Jean-Michel, this will break lab-sea experiment ... and 
possibly the adjoint, but these are bug fixes so they are urgent.  Can I go 
ahead and submit, along with Martin and Jinlun's changes for HSNOW treatment?

A version of the modified growth.F that I would check in is here:
http://mitgcm.org/cgi-bin/viewcvs.cgi/MITgcm_contrib/high_res_cube/code-mods/growth.F

Dimitris


> Dimitris, go ahead removing it from 1 and 2. Jinlun
> 
> Dimitris Menemenlis wrote:
> 
>> Jinlun, thanks for getting back.  Can the A22 (0.15) minimum be removed
>> from all computations in growth.F and groatb.F?  There are three places
>> that are affected by A22:
>> 
>> 1. Computation of hsnow, precip, and evap in growth.F.  It definitely needs
>> to be removed from there as this introduces an artificial forcing bias in
>> the solution, even over fully-open-water regions.
>> 
>> 2. Computation of GAREA in growth.F.  I think that it should be removed
>> from there too but am not 100% sure.
>> 
>> 3. Computation of ice thickness HICE in groatb.F.  Removing it from there
>> would cause code to blow up since AREA(I,J,2,bi,bj) is in the denominator.
>> 
>> Should I go ahead and remove from 1 and 2 above but leave as is in 3?
>> 
>> Dimitris