ccpp-physics
ccpp-physics copied to clipboard
NCAR
Ferrier-Aligo microphysics implementation needs to be revisited
Description
This issue replaces https://github.com/NCAR/ccpp-physics/issues/438. For more than a year now, Ferrier Aligo microphysics is known not to function correctly in the UFS. A few issues with the implementation in CCPP are noted in https://github.com/NCAR/ccpp-physics/pull/749:
- missing microphysics cloud interaction in case the bulk condensates (fractions) are used
- bulk condensates (fractions) were stored in the wrong (non-persistent) DDTs and not used anywhere in the model except in Ferrier-Aligo MP itself
Because of these deficiencies, the fractions are removed from the non-persistent DDTs in FV3 and made local variables in FA MP in PR https://github.com/NCAR/ccpp-physics/pull/749.
Should the capability to run FA MP with bulk condensates be needed in the future, the scheme implementation must be revisited, the microphysics-cloud interaction added, and the variables added correctly to the host model(s).
Dom, it appears the issues have been identify based on your description of this ticket. How much an effort is required to fix the issues ? (sorry for asking you a question that might be difficult to answer) @ericaligo-NOAA
Dom, it appears the issues have been identify based on your description of this ticket. How much an effort is required to fix the issues ? (sorry for asking you a question that might be difficult to answer) @ericaligo-NOAA
Fanglin, the issues described here are relatively easy and quick to fix. Maybe a couple of days of work, including the mp-radiation interaction. The problem is (was in the past) that the code crashed with a memory corruption after the adiabatic init of the dycore, i.e. before even calling the physics the first time. There must (have been) a bug in the way the different tracers for FA were coded into the dycore. I spent a week or so trying to figure this out about a year ago. Chances are that with the recent updates to the dycore and improvements to the build system (debugging flags etc.), we will be able to figure this out, but to me this is a big unknown. Could be only a day or two, could be a few weeks.
Do we need to run with total condensate advection? I would say no as that might be a step backwards in terms of the science.
On 10/22/2021 3:46 PM, Dom Heinzeller wrote:
Description
This issue replaces #438 https://github.com/NCAR/ccpp-physics/issues/438. For more than a year now, Ferrier Aligo microphysics is known not to function correctly in the UFS. A few issues with the implementation in CCPP are noted in #749 https://github.com/NCAR/ccpp-physics/pull/749:
- missing microphysics cloud interaction in case the bulk condensates (fractions) are used
- bulk condensates (fractions) were stored in the wrong (non-persistent) DDTs and not used anywhere in the model except in Ferrier-Aligo MP itself
Because of these deficiencies, the fractions are removed from the non-persistent DDTs in FV3 and made local variables in FA MP in PR #749 https://github.com/NCAR/ccpp-physics/pull/749.
Should the capability to run FA MP with bulk condensates be needed in the future, the scheme implementation must be revisited, the microphysics-cloud interaction added, and the variables added correctly to the host model(s).
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/NCAR/ccpp-physics/issues/762, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALQ75MMCHYFZMQ54OAPK2XLUIG5Q3ANCNFSM5GRII5NA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
@ericaligo-NOAA @yangfanglin What is the status of the FA mp scheme, is it working in the UFS now? I am going through old issues and PRs and wonder if this can be closed. Thanks!