Changes between Version 5 and Version 6 of ticket/1658/General


Ignore:
Timestamp:
2016-06-09T15:06:51+02:00 (4 years ago)
Author:
frrh
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • ticket/1658/General

    v5 v6  
    9595 
    9696  
     97   So now I modify oce.F90 to add the four fields indicated above and allocate them via a CPP condition on key_oasis3 and key_medusa.  
     98   I also introduce new RTL defined in oce.F90 for now : ln_medusa to enable us to test things at run time without littering the code with #ifdefs.  
     99 
     100   This compiles and fails - we have no values in DMS. The routine where they seem to be available is trcbio_medusa. But we have to intercept values before the arrays are deallocated. It's made extra difficult by the fact that trcbio_medusa basically contains one massive 3 level loop: over k,j,i. It's huge and it's not clear at what stage we should populate our arrays - for now I do it just prior to the deallocation. That's all jolly well but it doesn't solve the issue of where the values come from at startup because this code will not be called until after the first coupling exchange. For now I just initialise DMS_out_cpl and  CO2Flux_out_cpl to zero in oce.F90, but that is purely for testing purposes and must not be seen as a solution because it clearly does not cater for restartability.  
     101 
    97102 
    98103