Use restarts for CL4 and CL5 => CL6

[luyssaert]

The default simulations CL4 and CL5 should use a restart from OOL_SEC_STO_FG2. The trunk takes some decades to develop mature vegetation. When tested without restart files, the roughness length will probably be too low for most of the simulation.

First test to be done is using the files produced by Nicolas Vuichard to run a offline spinup. At ciclad :
/data/nvuichard/CMIP6_forcing_ORCHIDEE/piControl/
See also ticket #802

[jgipsl]

First test was done using the files IGCM/SRF/METEO/CRUJRA/v2.0_CMIP6_144x143 but there is a problem in the land-sea mask, see #827

Now tests are ongoing with the files CMIP6_forcing_ORCHIDEE/piControl/modifaxis/cmip6_piControl_XXXX.nc from Nicolas, described here #802. Problem when running LMDZOR using restart files created by forcing of these files gives instability problem in the atmosphere (Stopping in physiq, Plantage hgardfou)

[nvuilsce]

Work is planned on that ticket the week of March, 9.

[luyssaert]

Following discussion it was decided to introduce two approaches to create forcing files from LMDzOR/IPSL-CM/WRF for ORCHIDEE offline:

• Run a script on existing coupled simulations to create forcing files that can be used to run ORCHIDEE in offline mode. This is a temporary solution as we need the CMIP6 coupled simulations to run an ORCHIDEE spinup (ticket #900). This method is to be used with past simulations (CMIP6)
• Add the functionality to ORCHIDEE to write the climate variables from the atmospheric model to a correctly formatted output file that can be used by ORCHIDEE. Adjust the driver to read these forcing files (ticket #899). This method is to be used with future coupled simulation

[jgipsl]

A new proposition of spinup for CL6 is described here: #901