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

[szy21]

The Climate Modeling Alliance

Software Design Issue 📜

Purpose

Demonstrate that diagnostic EDMF scheme captures key cloud regimes (GABLS, BOMEX, DYCOMS, RICO, TRMM LBA) with explicit timestepping

Cost/Benefits/Risks

People and Personnel

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

Components

Inputs

Results and Deliverables

• Box simulations comparable to LES and old EDMF

Task Breakdown And Schedule

Closures

Beta Give feedback

1. Fix the closures with topography (it should always use z - z_sfc) #2028
   +0
   
2. Add nonhydrostatic pressure drag for diagnostic edmfx #1800
   EDMF 🌪️ +1
   
3. Add prandtl number calculation #1955
   EDMF 🌪️ +1
   
4. Check consistency in thermo state domain decomposition #1994
   +0
   
5. Reduce sgs maximum iteration for saturation adjustment #1985
   +0
   
6. Couple 0-moment microphysics with diagnostic explicit EDMF #2003
   EDMF 🌪️ +1
   

Box test cases

Beta Give feedback

1. Bomex
2. Dycoms
3. Rico
4. Add GABLS for diagnostic EDMF #1960
   EDMF 🌪️ +1
   
5. Add TRMM for diagnostic EDMF #1957
   EDMF 🌪️ +1
   

SDI Revision Log

CC

@tapios @simonbyrne @cmbengue

[szy21]

All the test cases have been added to ci, under EDMFX, see the latest build for example.

In summary, DYCOMS and GABLS look good. TRMM looks ok although the area fraction is too small. Bomex and Rico look noisy. The results are likely due to an idealized entrainment and detrainment closure. We think the results are good enough to show that we have explicit diagnostic edmf working for box test cases. I have moved the remaining tasks to #1975, as some of them are less important for now, or depend on other milestones (implicit solver and covariant derivative). I will close this SDI and milestone after the last task is closed.