urs_beyerle@env_ethz_ch
New Member
Hi,
We normally do a micro perturbation using pertlim for a startup or hybrid run of CESM2. But since a hybrid run is not bit-for-bit equal to the original run, all the disturbed runs including the run with pertlim=0 can significantly differ from the original run during the first days (see the figure below). Because we are interested particular in the first days after the perturbation, this is not ideal.
Is there a way to do a perturbation of a branched run? One idea I had was to modify the restart files (cam.r, cam.rh0, cam.rs) of cam to mimic such perturbation. But which file should I change and which variable/field in order to do something similar as the atmospheric temperature field perturbation?
A related question, what was the original reason to implement pertlim for startup and hybrid runs only?
Any feedback or suggestions would be very welcome.
Cheers,
Urs
We normally do a micro perturbation using pertlim for a startup or hybrid run of CESM2. But since a hybrid run is not bit-for-bit equal to the original run, all the disturbed runs including the run with pertlim=0 can significantly differ from the original run during the first days (see the figure below). Because we are interested particular in the first days after the perturbation, this is not ideal.
Is there a way to do a perturbation of a branched run? One idea I had was to modify the restart files (cam.r, cam.rh0, cam.rs) of cam to mimic such perturbation. But which file should I change and which variable/field in order to do something similar as the atmospheric temperature field perturbation?
A related question, what was the original reason to implement pertlim for startup and hybrid runs only?
Any feedback or suggestions would be very welcome.
Cheers,
Urs