O1.5.1b Run implicit diagnostic EDMF for canonical test cases in a box

[szy21]

The Climate Modeling Alliance

Software Design Issue 📜

Purpose

Demonstrate diagnostic EDMF scheme with implicit (JFNK with preconditioner) timestepping for diffusion terms in grid-scale equations (GABLS, BOMEX, DYCOMS, RICO, TRMM LBA)

Cost/Benefits/Risks

People and Personnel

• Lead: @szy21
• Collaborators: @dennisYatunin @trontrytel
• Reviewers: @tapios

Components

Inputs

Results and Deliverables

Task Breakdown And Schedule

Implementation

Beta Give feedback

1. Check physical vertical velocity used in closures
2. Use a better detrainment closure #2027
   +0
   
3. Limit mixing length in dissipation #1986
   EDMF 🌪️ +1
4. Add shear to precomputed quantities #1880
   EDMF 🌪️ +1
   
5. Run diagnostic edmf in a box with implicit diffusion #2397
   EDMF 🌪️ +1
   
6. Set u^3 at the boundary to 0 in diagnostic edmf #2055
   EDMF 🌪️ +1
   
7. Add different options for entrainment and detrainment #2110
   EDMF 🌪️ +1
   

SGS fluxes

Beta Give feedback

1. Change vertical diffusion of grid-scale horizontal momentum #1859
   EDMF 🌪️ enhancement +2
   
2. Add edmfx momentum flux to grid-mean tendencies #1736
   +0
   

Documentation

Beta Give feedback

No tasks being tracked yet.

Test cases

Beta Give feedback

1. Bomex
2. Dycoms
3. Rico
4. GABLS
5. TRMM
6. Add comparison with main and LES for edmfx cases #2033
   EDMF 🌪️ +1

Miscellaneous

Beta Give feedback

1. Add energy source term associated with momentum drag #2267
   EDMF 🌪️ +1
   
2. check kinetic energy calculation in EDMF #2347
   +0
   

SDI Revision Log

CC

@tapios @simonbyrne @cmbengue

[szy21]

All the box test cases are run stably with implicit diffusion with 200 secs timestep. The simulations have been added to ci. I think we can close this issue. The remaining tasks are not important for now, and I will move them to the next milestone.