Changes between Version 1 and Version 2 of Tags/196/ExternalisationParameters/ValidateExternalisation


Ignore:
Timestamp:
11/28/11 16:16:36 (9 years ago)
Author:
dsolyga
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • Tags/196/ExternalisationParameters/ValidateExternalisation

    v1 v2  
    1 == Validate the externalized version == 
    2  
    3 === fluxnet site === 
     1= Validate the externalized version = 
     2 
     3 
     4== fluxnet site (MARCH 2011) == 
    45 
    56First of all, we validate the externalization of parameters on a fluxnet site and with IMPOSE_VEG = y. 
     
    328329}}} 
    329330 
    330 ==== Results ==== 
     331=== Results === 
    331332 
    332333The comparisons are OK. 
     
    343344of PFTs. [[BR]] 
    344345 
    345  
    346  
     346== NEW VALIDATION TESTS (OCT-NOV 2011) == 
     347 
     348After merging the tag 1.9.5.2 into the externalized version, I launched some new validation tests. [[BR]] 
     349 
     350There are three kind of tests : Fluxnet (local scale), regional and global. [[BR]] 
     351 
     352These tests are not part of a scientific validation of the model but only technical ones. As I have only reorganized the code, I have to check if 
     353I obtain the same results in both cases. 
     354 
     355I used the same forcing files for all tests : ncc forcing ( for the year 1982) except for the test with the weather generator (CRU for the year 1985).[[BR]] 
     356 
     357My referenced version was the TAG 1.9.5.2 with corrections of the two following bugs : black_carbon and herbivorory. 
     358[[BR]] 
     359The tools used are cdo (cdo diffv command), a validation script (ksh + ferret) and ferret (difference maps).[[BR]] 
     360 
     361=== FLUXNET sites === 
     362 
     363I launched the following tests : 
     364  1. FLUXNET_REF_OCT2011 (launched with the referenced version) 
     365  2. VAL_FLUXNET_2PFTS_OCT 
     366  3. VAL_FLUXNET_C3crop_to_C3grass 
     367  4. VAL_FLUXNET_C3GRASS_in_14 
     368  5. VAL_FLUXNET_C3GRASS_SPLIT 
     369The four last ones were launched with the externalized version. The setup of these jobs are the same defined in March (see above) : 
     370 ||VAL_FLUXNET_2PFTS_OCT   ||setup : 2 PFTS 
     371 ||VAL_FLUXNET_C3crop_to_C3grass||   setup : 13 PFTs with PFT_12 reparametrized as PFT 10 
     372 ||VAL_FLUXNET_C3GRASS_in_14 ||      setup : 14 PFTs with PFT 14 as C3 grass 
     373 ||VAL_FLUXNET_C3GRASS_SPLIT ||      setup : 14 PFTs with C3grass shared  between PFT 10 and  PFT 14  
     374 
     375In order to check in teststomate was correctly adapted to the externalized version, I launched and compared the two following jobs : 
     376 1. FLUXNET_TESTSTOMATE (referenced version) : 13 PFTs with vegetation only on PFT1 and PFT 10 
     377 2. EXT_TESTSTOMATE_FLUXNET : 2PFTS  
     378Both have the following setup : 
     379{{{ 
     380# Initialisation for spin-up : 
     381# orchidee with sechiba alone (!!! if ok_stomate == n !!!) 
     382duree_nostomate=0 
     383# orchidee with stomate 
     384duree_inistomate=1 
     385# teststomate (only if duree_inistomate > 0) 
     386duree_offlineini=20 
     387 
     388# Loop over ORCHIDEE runs (used for spin-up) 
     389# The whole job is restarted n_iter times 
     390n_iter=1 
     391# orchidee with sechiba (and stomate if ok_stomate=y below) 
     392duree_sechiba=20 
     393# teststomate 
     394duree_stomate=0 
     395# forcesoil 
     396duree_carbonsol=0 
     397 
     398# Final run (full ORCHIDEE) 
     399# This last parameter must be non-zero. 
     400duree_final=20 
     401     
     402}}} 
     403 
     404The last configuration to test was FORCESOIL. So I launched the following jobs : 
     405 1. TAG_FLUX_FOR (referenced version) : 13 PFTs 
     406 2. EXT_FLUX_FOR : 13 PFTs 
     407The setup was : 
     408 
     409{{{ 
     410# Initialisation for spin-up : 
     411# orchidee with sechiba alone (!!! if ok_stomate == n !!!) 
     412duree_nostomate=0 
     413# orchidee with stomate 
     414duree_inistomate=1 
     415# teststomate (only if duree_inistomate > 0) 
     416duree_offlineini=0 
     417 
     418# Loop over ORCHIDEE runs (used for spin-up) 
     419# The whole job is restarted n_iter times 
     420n_iter=1 
     421# orchidee with sechiba (and stomate if ok_stomate=y below) 
     422duree_sechiba=20 
     423# teststomate 
     424duree_stomate=0 
     425# forcesoil 
     426duree_carbonsol=10000 
     427 
     428# Final run (full ORCHIDEE) 
     429# This last parameter must be non-zero. 
     430duree_final=20 
     431}}} 
     432 
     433 
     434 
     435 
     436=== Comparisons === 
     437 
     438The comparisons by the three tools were all OK.  
     439All results are available in a directory called VALIDATE_OCTOBRE_2011 on obelix. 
     440  
     441 
     442 
     443== Regional scale == 
     444 
     445Only one test was made over a region of Europe (France I think). 
     446