search

MIT-LAE / APCEMM

Aircraft Plume Chemistry, Emissions, and Microphysics Model
MIT License
7 stars 16 forks source link

Vertical velocities from netCDF input #17

Closed thabbott closed 1 month ago

thabbott commented 3 months ago

I've been running some APCEMM simulations with met data read from a netCDF file and have noticed what seems like strange behavior: I get different results from a simulation that doesn't read vertical velocities and from a simulation that reads a vertical velocity field that's zero everywhere.

This zip archive includes a sample netCDF file and two YAML files that reproduce this issue:

input.zip

When I run without reading vertical velocities (w_off.yaml), I get this IWC field after 1 hour:

image

When I run a simulation that reads vertical velocities equal to zero everywhere (w_on.yaml), I get a different result:

image

(I'm open to the possibility that this is an issue with the way I've set up my APCEMM input files--if that's the case please let me know!)

sdeastham commented 1 month ago

Thanks @thabbott ! I've now had a chance to look into this in more detail, and I think that the vertical velocity code basically needs an overhaul. A re-write of vertical velocity handling is now slated for v1.2.0; you can follow this on issue https://github.com/MIT-LAE/APCEMM/issues/26. I'll close this issue for now though as the fix you develop in https://github.com/contrailcirrus/pycontrails/commit/c697345d76968c8f06e1cbd3324c9e619cdb7717 is also the recommended approach.