8. Graphics

Note

Graphics plotting is not a supported feature for the MRW release. The instructins provided here are for informational purposes only. Users may experiment with the scripts at their own risk.

The MRW Application comes with two Python scripts for users to visually check the results of their forecast experiments. This script plots 2-m temperature, hourly precipitation, cloud cover, and 10-m wind at a user inputted time range. In addition, it creates a forecast loop GIF for each forecast product.

Users will need to load Anaconda first. Here are some examples of how to do this:

On an HPC system, you can use modules to load Anaconda. Otherwise, use the following instructions. Find download link here: https://www.anaconda.com/products/individual. Scroll to the bottom to the “Anaconda Installers” section and copy the download link for the appropriate version:

cd $HOME
wget <download_link>
bash <Anaconda_excecutable_name>
export PATH=$HOME/anaconda3/bin:$PATH

Install Libraries:

conda install -c conda-forge -y cartopy
conda install -y netCDF4

Download Files:

cd $SCRATCH

Get natural earth files

wget https://ftp.emc.ncep.noaa.gov/EIB/UFS/SRW/v1p0/natural_earth/natural_earth_ufs-srw-release-v1.0.0.tar.gz
tar -xzf natural_earth_ufs-srw-release-v1.0.0.tar.gz
wget <TODO_new_URL>

Run Script: Navigate to the run directory of your experiment

cd $SCRATCH/<expt_name>/run

Create symbolic link to plotting script:

ln -sf dir/of/plot_mrw.py .

Run script:

python3 plot_mrw.py (Start time)<YYYYMMDDHH> (Start Forecast hour)<HHH> (End Forecast hour)<HHH> <Natural Earth Directory>

For example:

python3 plot_mrw.py 2019082900 000 048 $SCRATCH/natural_earth

View Results: A utility typically used to visualize the resulting images in png format is display. If it is available on your platform, you can use the command:

display *.png

Users working on a remote platform can use the scp command to transfer files to their local system to view the GIFs and files.

Sample Output The sample plots are in the sample_output.pdf document in the plotting_scripts directory. They are consistent with the Hurricane Dorian initial conditions and tag ufs-v1.1.0. Users results may look different if they are using a different branch or tag. Their results may also look different if they are running on a different platform than the one used to generate the plots.

Additional Options After completing the graphics plotting successfully, users may be interested in trying to change a namelist option and run a second test to check their understanding of how results will change. If you are interested in doing that, please visit the UFS Portal to take our graduate student test, which will give you instructions to take that leap, and will also provide important information for our development work

8.1. Plot Results from MRW Graduate Student Test (GST)

8.1.1. plot_mrw_cloud_diff.py

The plot_mrw_cloud_diff.py generates plots of cloud cover from the UFS MRW App Graduate Student Test (GST). Users plot a control, experiment, and the difference between the control and the experiment.

Make sure all the necessary modules can be imported.

Five command line arguments are required:

  1. Cycle date/time in YYYYMMDDHH format

  2. Forecast hour in HHH format

  3. CTRL_DIR: Control directory. Postprocessed data should be found in the directory: CTRL_DIR/run/

  4. EXPT_DIR: Experiment directory. Postprocessed data should be found in the directory: EXPT_DIR/run/

  5. CARTOPY_DIR: Base directory of cartopy shapefiles. Shapefiles cannot be directly downloaded to NOAA machines from the internet, so shapefiles need to be downloaded if geopolitical boundaries are desired on the maps. The file structure should be: CARTOPY_DIR/shapefiles/natural_earth/cultural/*.shp. For more information regarding files needed to set up display maps in Cartopy, see the SRW App Users’ Guide.

Plots are saved in the plotting_scripts as YYYYMMDDHH<variable>_<domain>_fHHH.png. The variable domains in this script can be set to either ‘conus’ for a CONUS map or manually adjusted in the Domain section

8.1.2. plot_mrw.py

This script generates plots of several output variables from the UFS MRW App Graduate Student Test.

Make sure all the necessary modules can be imported. Five command line arguments are needed: #. Cycle date/time in YYYYMMDDHH format #. Start forecast hour in HHH format #. End forecast hour in HHH format #. CARTOPY_DIR: Base directory of cartopy shapefiles. Shapefiles cannot be directly downloaded to NOAA machines from the internet, so shapefiles need to be downloaded if geopolitical boundaries are desired on the maps. The file structure should be: CARTOPY_DIR/shapefiles/natural_earth/cultural/*.shp. For more information regarding files needed to setup display maps in Cartopy, see the SRW App Users’ Guide.

Plots are saved in the plotting_scripts as YYYYMMDDHH<variable>_<domain>_fHHH.png. The variable domains in this script can be set to either ‘conus’ for a CONUS map or manually adjusted in the Domain section.