Opened 11 years ago

Closed 11 years ago

#9 closed task (fixed)

SECHIBA CMOR outputs

Reported by: dsolyga Owned by: dsolyga
Priority: critical Milestone: orchidee_1_9_6
Component: Physical processes Version: orchidee_1_9_5_2
Keywords: Cc:

Description

(From Agnès Ducharne, Frédérique Cheruy and Marie-Alice Foujols)

The 6 following CMOR variables mrros, mrr, prveg, evspsblveg, evspsblsoi, tran are underestimated by a multiplicative factor 48.
The solution consists in two steps :

  • to delete in the modules hydrolc and sechiba the division by one_day (=86400). Example :
      histvar(:)=run_off_tot(:)/one_day
      CALL histwrite(hist_id, 'mrros', kjit, histvar, kjpindex, index)

is replaced by :

     CALL histwrite(hist_id, 'mrros', kjit, run_off_tot, kjpindex, index)

  • to change the operators avescatter by fluxop_scinsec in intersurf. Example :
   CALL histdef(hist_id, 'mrros', "Surface Runoff", "kg m-2 s-1", &
   & iim,jjm, hori_id, 1, 1, 1, -99, 32, avescatter(11), dt,dw)

is replaced by :

   CALL histdef(hist_id, 'mrros', "Surface Runoff", "kg m-2 s-1", &
   & iim,jjm, hori_id, 1, 1, 1, -99, 32, fluxop_scinsec(11), dt,dw)

With this correction, we will have the same operators for the CMOR and the ALMA outputs.

Change History (2)

comment:1 Changed 11 years ago by dsolyga

  • Status changed from new to accepted

comment:2 Changed 11 years ago by dsolyga

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

The CMOR outputs are corrected according Agnès and Marie-Alice proposal : see revision [734]

When I make the ratio between the standard outputs and the CMOR outputs, I found 86331 and not
86400 because the operator was in simple precision :

    WRITE(flux_scinsec,'("ave(scatter(X*",F8.6,"))")') un/dt 

So when dt = 1800s, we got 1/1800 = 0,000556 and 1/0,000556=1798,56. If we multiply by 48, we obtain approximatively 86331! We set the operator flux_scinsec in double precision. See revision [740].

The last correction was made for the units of bqsb and gqsb : they are now in kg/m2. (see revision [737]).

Note: See TracTickets for help on using tickets.