Published November 1, 2022 | Version v3

Data and Code Supplement for "A Mountain-Induced Moist Baroclinic Wave Test Case for the Dynamical Cores of Atmospheric General Circulation Models"

  • 1. University of Michigan

Description

Code and Data Supplement for "A Mountain-Induced Moist Baroclinic Wave Test Case for the Dynamical Cores of Atmospheric General Circulation Models"
===========================================================

This directory contains the data and scripts used to create the plots from our publication as well as the source
code modifications necessary to run this test case within the CESM and MPAS models.

Generating Plots
---------------

The `netcdf` directory contains the nominal half-degree runs necessary to generate nearly all of the plots from the paper. The one plot which is not reproducible from these data is the volume-integrated Eddy Kinetic Energy in the Spectral Element model. Storing high-resolution 4D wind fields requires a prohibitive amount of space. These data can be provided by the corresponding author, O.K. Hughes (owhughes@umich.edu). However, because this is several hundred GB of data I would strongly recommend generating these high-resolution runs yourself on your local system if you need them. Using 288 Intel Skylake cores (that is, 8 nodes each with two 18C processors) ran on the order of an hour.

In order to generate the plots from the paper, you need only install NCL and then run run.bash. Instructions for installing NCL
can be found in the `run.bash` script.

Source Code Modifications
----------------

CESM
The `src` subdirectory contains the files `user_nl_cam` and `ic_baroclinic.F90`. Create a case using `--compset=FKESSLER` and `--run-unsupported` options when running `create_newcase`. If your case is located at `${CASE_DIR}`, then from within the directory containing this README, run `cp user_nl_cam ${CASE_DIR}/user_nl_cam`, and then run `cp ic_baroclinic.F90 ${CASE_DIR}/SourceMods/src.cam/`. Then build and run the model using the usual workflow.

MPAS

The MPAS code was run using a branch of the MPAS model provided by the model developers to the authors. While the source code modifications are provided in the `src` directory, I would strongly recommend contacting the corresponding author if you wish to run this test case in the MPAS codebase.

Files

data.zip

Files (36.1 GB)

Name Size
md5:804faca5ffc6d03f9b7849fc16eafbb1
36.1 GB Preview Download
md5:dad4efc27b5552da1e861467ec5ad051
60.4 kB Preview Download
md5:4b51df58d9bf9eea9aeda5ace43faa73
2.3 kB Preview Download
md5:59ee7576ea564827af812a3d2c644ebb
64.0 kB Preview Download