| 45 | We propose to output variables with the same units as they are used. One exception is regarding the time unit. We don't want to ouput variables expressed per time step unit, like a flux per 30min. The unit "per second" should be the standard. |
| 46 | |
| 47 | ''Example 1:''[[BR]] |
| 48 | For instance, in intersurf.f90 line [http://forge.ipsl.jussieu.fr/orchidee/browser/trunk/ORCHIDEE/src_sechiba/intersurf.f90#L1701 1701]: |
| 49 | The instruction |
| 50 | {{{ |
| 51 | CALL xios_orchidee_send_field("evap",zvevapp*one_day/dt_sechiba) |
| 52 | }}} |
| 53 | should be modified as follows: |
| 54 | {{{ |
| 55 | CALL xios_orchidee_send_field("zvevapp",zvevapp/dt_sechiba) |
| 56 | }}} |
| 57 | |
| 58 | and in the field_def file, instead of having[[BR]] |
| 59 | {{{ |
| 60 | <field id="evap" name="evap" long_name="Evaporation" unit="mm/d"/> |
| 61 | }}} |
| 62 | one proposes to have: |
| 63 | {{{ |
| 64 | <field id="zvevapp" name="zvevapp" long_name="Evaporation" unit="mm/s"/> |
| 65 | <field id="evap" name="evap" field_ref="zvevapp" long_name="Evaporation" unit="mm/d"/> zvevapp*86400 |
| 66 | }}} |
| 67 | |
| 68 | ''Example 2:''[[BR]] |
| 69 | Some outputs are only the sum of elementary outputs. For instance, in slowproc.f90, the following section can be deleted: |
| 70 | {{{ |
| 71 | ! 4.2.2 Compute the net primary production as the diff from |
| 72 | ! Gross primary productin and the growth and maintenance |
| 73 | ! respirations |
| 74 | npp(:,1)=zero |
| 75 | DO j = 2,nvm |
| 76 | npp(:,j) = gpp(:,j) - resp_growth(:,j) - resp_maint(:,j) |
| 77 | ENDDO |
| 78 | |
| 79 | CALL xios_orchidee_send_field("npp",npp/dt_sechiba) |
| 80 | }}} |
| 81 | beacause, gpp, resp_growth and resp_maint are already ouptuts.[[BR]] |
| 82 | |
| 83 | npp will be now only set in the field_def file as follows: |
| 84 | |
| 85 | |
| 86 | |