Opened 9 years ago

Closed 7 years ago

#10 closed enhancement (fixed)

CMOR Soil depth axis output

Reported by: dsolyga Owned by: jgipsl
Priority: minor Milestone: ORCHIDEE 2.0
Component: Validation Version: orchidee_1_9_5_2
Keywords: CMOR outputs Cc: nvuilsce

Description

(Reported by C.Koven)

Here is description of the problem (sorry, in french) :

(from Nicolas Vuichard mail)

Concernant le point sur la mauvaise définition de la dimension 'depth' utilisée pour renseigner les horizons de sol (tu as reçu un mail de Charlie Koven à ce sujet, je crois).

Pour rappel, normalement les valeurs de depth devraient être exprimées en mètre et correspondre aux horizons de sol. Dans les données produites par orchidee, depth ne contient que les indices de ces horizons (ie 1,2,3,4,5,6,7).

Les deux variables définies sur la dimension depth sont tsl et mrlsl mais mrlsl n'est pas produite par orchidee, seul tsl est donc à corriger.

Voici donc les horizons de sol utilisés dans orchidee (en m) qui sont les valeurs que doit prendre la dimension 'depth':

0.04
0.13    
0.30    
0.64    
1.34    
2.72    
5.47   
Et les bounds associés:
0 0.04
0.04 0.13
....

Dernier point, plus anecdotique, j'ai pu voir dans le document sur les variables cmor, que normalement la variable associée à la dimension pour la profondeur du sol devrait être nommée 'sdepth' et non 'depth' comme c'est le cas actuellement.

Change History (5)

comment:1 Changed 9 years ago by dsolyga

  • Milestone changed from orchidee_1_9_6 to ORCHIDEE 2.0

comment:2 Changed 9 years ago by dsolyga

  • Owner changed from dsolyga to jgipsl
  • Status changed from new to assigned

comment:3 Changed 7 years ago by aducharne

The orignal comment relates to the CMOR protocol.

This problem occurs in intersurf.f90, it relates to the axis called solth, which is the vertical axis for "Deep ground temperatures", called ptn in thermosoil.f90.

Note that ptn is the only exported variable using this axis solth.

Instead of filling solth with sol=(1:ngrnd), we should use instead zz(1:ngrnd) (in meters), the one from thermosoil.90, which needs to be recomputed in intersurf.90, as thermosoil is called later.

comment:4 Changed 7 years ago by jgipsl

The numbers of the levels in the begining of this ticket does not seems to be correct. Looking into a CMORised CMIP5 variable (for exempel at ciclad : /prodigfs/esg/CMIP5/output/IPSL/IPSL-CM5A-LR/amip/mon/land/tsl/r6i1p1/tsl_Lmon_IPSL-CM5A-LR_amip_r6i1p1_195001-195912.nc), the variable depth is :

 depth = 0.017852796241641, 0.0788060575723648, 0.200712606310844, 
    0.444525688886642, 0.932151675224304, 1.90740394592285, 3.85790801048279 ;

The zz variable in ORCHIDEE_1_9_5, tag used for CMIP5, the values are the same:

zz=  1.78527975403299857E-002  7.88060610288256469E-002  0.20071258800581698
     0.44452564195979960       0.93215174986776483       1.9074039656836959
     3.8579083973155575 

In the trunk ORCHIDEE rev 1943, the values have changed and are now :

zz=  1.41896498930979102E-002  6.26361450034387396E-002  0.15952913522412038
     0.35331511566548363       0.74088707654821018       1.5160309983136635  
     3.0663188418445699

Conclusion : I'll write a function in thermsoil that will return the zz vector. This funcion will be called in intersurf and in xios_orchidee for creating the vertical axe solth. The name of the axe will not be changed at this moment.

comment:5 Changed 7 years ago by jgipsl

  • Resolution set to fixed
  • Status changed from assigned to closed

Done in rev [1947]

Note: See TracTickets for help on using tickets.