Versions Compared

Old Version 2

changes.mady.by.user Chris Golaz

Saved on

compared with

New Version 3

changes.mady.by.user Chris Golaz

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  1. Account on crunchy.llnl.gov
  2. Climatology files generated by ncclimo (Generating and Regridding Climatologies (climo files) with NCO and ncclimo). Sample files are available on crunchy for the step-by-step guide below.

Step-by-step guide

Get ACME PMP utility scripts included in ACME PreAndPostProcessingScripts repository

For illustration, we clone the repository under ~/ACME_code

mkdir -p ~/ACME_code

cd ~/ACME_code

git clone git@github.com:ACME-Climate/PreAndPostProcessingScripts.git

Set-up working directory and environment

Create working directory and copy utility scripts

mkdir ~/mytest

cd ~/mytest

cp ~/ACME_code/PreAndPostProcessingScripts/create_model_diagnostics/pcmdi_metrics_package

Load conda environment

export PATH=/usr/local/anaconda2/bin:${PATH}

source activate latest

Use cdscan to create xml describing input files

mkdir xmls

cdscan -x xmls/alpha6_1850_0070-0099.xml /work/golaz1/ACME_simulations/20160520.A_WCYCL1850.ne30_oEC.edison.alpha6_01/pp/clim_rgr/0070-0099/*.nc

The output file be a file named xmls/alpha6_1850_0070-0099.xml that describes the input netcdf files.

 

Process climatologies to be consistent with CMIP data structure

mkdir data

python transform_acme_clims_to_CMIP5_structure.py

The result will be a number of individual netcdf files in the sub-directory data/alpha6_1850_0070-0099:

...

pr.nc  prw.nc  psl.nc  rlutcs.nc  rlut.nc  rsutcs.nc  rsut.nc  ta.nc  tas.nc  tauu.nc  tauv.nc  ua.nc  va.nc  zg.nc

 

Compute derived fields (currently only CRE fields)

python compute_acme_derived_vars.py -b data

Two additional netcdf files for LW and SW cloud radiative effects will be created under data/alpha6_1850_0070-0099:

ls data/alpha6_1850_0070-0099/
pr.nc  prw.nc  psl.nc  rltcre.nc  rlutcs.nc  rlut.nc  rstcre.nc  rsutcs.nc  rsut.nc  ta.nc  tas.nc  tauu.nc  tauv.nc  ua.nc  va.nc  zg.nc

...

Run PMP mean climate statistics

mkdir results

pcmdi_metrics_driver.py -p acme_pmp_parameterfile.py

This is the core of PMP. It will compute climate statistics comparing the model output with various observational datasets. The statistics are stored in json files located under the subdirectory results/pmp:

...

There is one json file for every variable. Additionally, re-gridded netcdf files are under the subdirectory interpolated-output.

Create portrait plots

mkdir plots

python portrait_plots.py

The resulting output should be a png image file under the subdirectory plots:

...

acme_pmp-control_tests_rms_GLB.png

...

This image should look like

...

To generate portrait plots for other regions (tropics, nothern extra-tropics, souther extra-tropics), edit portrait_plots.py and change definition of the region from global to another option and then rerun the script.

reg = 'GLB'

#reg = 'NHEX'

#reg = 'SHEX'

#reg = 'TROPICS'

 

Questions?

Chris Golaz may be able to help.

...