search

NCAR / ccpp-physics

GFS physics for CCPP
Other
58 stars 145 forks source link

Addition of total flux calculation for wet deposition in Thompson #926

Closed bbakernoaa closed 2 years ago

bbakernoaa commented 2 years ago

When switching to to Thompson in P8b with the coupled aerosols the values were not calculated by default and therefore zeros were passed to the GOCART model through the NUOPC layer causing no large scale wet deposition. This PR adds back in the capability that was added to the GFDL MP for the coupled GOCART model for the calculation of the large scale wet deposition. Here we use the instantaneous tendency to calculate the flux. This has no impact on the meteorology and only adds in the needed variables for the GOCART model.

For comparison see here: https://docs.google.com/presentation/d/1DzPGs85respsvhLe64rZm3I8V4MxM0q3/edit?usp=sharing&ouid=111407383710090576995&rtpof=true&sd=true

Particularly look at the v7 (this fix) and in comparison with the GFDL MP. They should not be identical because of different model physics but it is in the right ballpark.

@zhanglikate @yangfanglin @AnningCheng-NOAA

AnningCheng-NOAA commented 2 years ago

They are the tendencies in Model%avg_max_length bucket (an hour in default setting). More physical meaningful approach are preferred in the future.

AnningCheng-NOAA commented 2 years ago

Attached are the the ppt file for seven experiment designs here. Version 2 and version 3 are physical making meaningful. The other versions are for the results meeting expectation for a few cases, which could be likely fail for other cases. prcp_flux.pdf

AnningCheng-NOAA commented 2 years ago

Here are the slides for the results from version 1 to version 6 C96_UFS8bGBTH1356_201608.pptx.pdf .

AnningCheng-NOAA commented 2 years ago

The difference between version 6 and version 7 is that version 6 accidently turned on the normalization of rain-fluxes in the mp_thompson_post.F90.

AnningCheng-NOAA commented 2 years ago

Version 7 gives the same results as version 1.