Bugfix and optimization of prognostic closure in cumulus convection

[lisa-bengtsson]

PR Checklist

• This PR is up-to-date with the top of all sub-component repositories except for those sub-components which are the subject of this PR. Please consult the ufs-weather-model wiki if you are unsure how to do this.

• This PR has been tested using a branch which is up-to-date with the top of all sub-component repositories except for those sub-components which are the subject of this PR

• An Issue describing the work contained in this PR has been created either in the subcomponent(s) or in the ufs-weather-model. The Issue should be created in the repository that is most relevant to the changes in contained in the PR. The Issue and the dependent sub-component PR
  are specified below.

• Results for one or more of the regression tests change and the reasons for the changes are understood and explained below.

• New or updated input data is required by this PR. If checked, please work with the code managers to update input data sets on all platforms.

Instructions: All subsequent sections of text should be filled in as appropriate.

Description

This PR corrects a bug when using the prognostic closure (progsigma = true) to compute the q-tendency due to microphysics even if the diagnostic flags are false. It also optimizes the cloudbase massflux value for shallow and deep convection to work with the overall P8 physics suite.

New baselines are needed for the test: control_c384_progsigma

This update does not change the results in the control tests.

Issue(s) addressed

It addresses the issue: NCAR/ccpp-physics#960

Testing

Tested on Hera.
New baselines are needed for the test: control_c384_progsigma
This update does not change the results in the control tests.

• hera.intel
• hera.gnu
• orion.intel
• cheyenne.intel
• cheyenne.gnu
• gaea.intel
• jet.intel
• wcoss2.intel
• acorn.intel
• opnReqTest for newly added/changed feature
• CI

Dependencies

NCAR/ccpp-physics#967
NOAA-EMC/ufsatm#582