There is not a direct way to do this.
The way I would approach this (assuming I understand what you are trying to do) is:
(1) run the compset B_RCP8.5_CAM5_CN from 2005-2060
(2) build a compset B2060 (you can use the compset B1850 as a template)
(3) start an hybrid from the end of (1)
Thanks, Hannay! For "step (2): build a compset B2060", my understanding is that:
a) I need to create the forcing files (aerosol, ozone, land PFT, etc.) for the year-2060 using the default RCP8.5 scenario inputs. For example, I need to interpolate the aerosol emission data to the year-2060;
b) Then I need to create a new case using the default B_1850_C5CN compset as a template;
c) For the new case, use restart files from step (1) as initial conditions, and change the Namelists to force the model to read the new year-2060 forcing;
b) I am not sure if the step (1) is necessary, but I think it might help the model to reach equilibrium in a shorter time.
Am I right?
------
Another question I have is how to do a hybrid run of the RCP scenarios?
For example, I first run the RCP4.5 scenario from the year 2005 to 2060, and then I want to do a hybrid run using the 2050 climate as initial conditions but using forcing from the RCP8.5 scenario to run from 2060 to 2100. After creating a new case using B_RCP8.5_CAM5_CN, I need to change the initial condition (RUN_REFCASE and RUN_REFDATE).
My question is how do I force the model to read the RCP8.5 forcing from the year 2060? Do I only need to change the RUN_STARTDATE to 2060 in the env_run.xml?
------
My last question is, the 20th case run from 1850 to 2000, and all RCP scenarios start from 2005. Since all the input data for 20th case extend at least to 2005. I just want to confirm that I can simply continue running the 20th case from 2000 to 2005?
Thanks very much!
Lei