search

E3SM-Project / E3SM

Energy Exascale Earth System Model source code. NOTE: use "maint" branches for your work. Head of master is not validated.
https://docs.e3sm.org/E3SM
Other
352 stars 363 forks source link

SCM runs retain constituents from initial condition #5177

Open quantheory opened 2 years ago

quantheory commented 2 years ago

I've been experimenting with DYCOMS-II RF02 single-column simulations to get a feel for how drizzle is behaving in E3SMv2 (and its sensitivity to resolution), and I noticed some weird behaviors in the first few time steps that turned out to be due to extra data from the initial conditions file that I wasn't expecting. The cloud liquid field from the initial conditions was ignored because it was specified in the the IOP file, but the ice cloud and rain fields were read in from the ncdata file, and at shorter time steps there was enough there to "rain out" most of the initial cloud, which then takes some time to recover. This is a pretty mild effect at the default time step, but still lengthens the transient "spinup" behavior of the simulation a bit.

I think that what should happen here is that the constituent array should be initialized to zero (or qmin), except for those constituents that are specified in the IOP file. I implemented such a change, and the weird spinup behavior went away in the DYCOMS case.

I will try to put a PR together for this later today, adding a namelist variable to zero out these fields. Not sure what to name it; maybe scm_ignore_non_iop_tracers?

quantheory commented 2 years ago

Tagging @bogensch