Changes between Version 2 and Version 3 of model_comparison


Ignore:
Timestamp:
07/26/19 16:37:32 (4 years ago)
Author:
snguyen
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • model_comparison

    v2 v3  
    33== Algorithm == 
    44 
     5It would be possible to implement different parameterizations and compute them at the same time. 
     6 
     7Roughly the algorithm would be: 
     8 
     9* storage of the state 
     10 
     11* Beginning of loop 
     12 
     13  * if different from first loop : restore the state 
     14  
     15  * if loop n compute nth model tendencies (the last model must be the model used for actual time stepping) 
     16 
     17  * output the nth model variables values to file 
     18 
     19* end of loop 
     20 
     21The storing and restoring is already implemented in the Schwarz loop. 
     22Clearly the Schwarz loop can be modified to do this. 
     23 
     24There are several problems in the case of coupling ! 
     25 
     26You will need to store the fields sent and received instead of going through OASIS for exchanging between ocean and atmosphere. 
     27 
     28If you have several physical time steps during a Schwarz window you will compute a different trajectory for each parameterization instead of a single point. 
     29 
     30In the case of forced LMDZ you can set the Schwarz loop to be for a single physical time step. 
     31 
     32Another solution is to store the ocean forcing of a reference coupling simulation and use this to run a forced LMDZ run. 
     33 
    534== modification of OASIS communications == 
    635 
     36It would be needed to implement variables to store and restore the coupling instead of sending and receiving through OASIS 
     37 
    738== XIOS modifications == 
     39 
     40It would be needed to implement XIOS calls to write the values for the different models in a single file. 
     41A temporal filter to append all the values at each loop repetition could be used.