Need to specify simulation starting datetime for continue run?

[nmizukami]

It seems like simulation start time (simStart) needs to be specified for continue run for standalone run. This is to read runoff input at correct time. Similarly, restart file the needs to be specified in fname_state_in in control file.

For cesm coupled run, the simulation start time is coming from the coupler, so it is ok. and fname_state_in is read in from rpointer file.

Erik (@ekluzek), do you think it is better to have the code figure out start time for continue run?

[ekluzek]

To have the code figure out the start time for a continue run, you'll need to add some similar infrastructure as in CESM to handle this. I think it is worthwhile, but might take some effort. I suppose maybe the easiest/best thing to do would be to use the CESM restart infrastructure for both cesm-coupled and standalone. It would take some work to figure it out, but then it would be something that's the same between standalone and coupled and that would be a good thing.

[nmizukami]

Just looking into the codes, fixing this needs some moving around of several subroutines, sosimStart and fname_state_in are populated accordingly from rpointer file (and restart file to get simStart information)