cmpdata package can handle and analyze raw CMIP6 data.
- GitHub repo: https://github.com/TaufiqHassan/cmpdata
- Documentation: https://cmpdata.readthedocs.io/en/latest
- Handle raw CMIP6 data
- Analyze any specific models/experiments/variables
- Perform regridding for model ensemble means
- Data preprocessing
- Statistical analysis
Installation via conda -
conda install -c thassan cmpdata
Requires python v3.6 or v3.7. Make sure you have added the conda-forge channel in your environment.
conda create --name cmpdata python=3.7 (or 3.6) conda activate cmpdata conda config --env --add channels conda-forge conda install -c thassan cmpdata
Alternatively, use the YAML file to create a virtual conda enviroment (cmpdata)
conda env create -f environment.yml
And then activate cmpdata to use cmpdata
conda activate cmpdata
Supports both OSX and Linux. Windows users can use Windows Subsystem.
cmpdata can be used to handle and analyze raw CMIP6 data. A lot of the options available in acccmip6 is available in cmpdata, especially for selecting models, experiments and variables.
cmpdata also tries to be a good command-line interface (CLI). Run cmpdata -h to see a help message with all the arguments you can pass.
cmpdata -h
usage: cmpdata.py [-h] -o {info,rm,mm,stats,ts} [-dir DIR] [-m M] [-e E] [-v V] [-r R] [-out OUT] [-f F] [-init INIT] [-end END] [-t] [-z] [-s S] [-mm]
[-std] [-clim] [-anom] [-manom] [-trend] [-aggr] [-freq FREQ] [-reg REG] [-rm] [-a] [-curve] [-w] [-ci CI] [-regrid]
options:
-h, --help show this help message and exit
-o {info,rm,mm,stats,ts}, --output-options {info,rm,mm,stats,ts}
Select an output option
-dir DIR Select directory.
-m M Model names
-e E Experiment names
-v V Variable names
-r R Realization
-out OUT Output file name
-f F Input filenames for stats
-init INIT Initial year
-end END Ending year
-t Temporal mean option
-z Zonal mean option
-s S Seasonal mean option
-mm Calculate model ensemble mean
-std Calculate model std
-clim Calculate monthly climatology
-anom Calculate monthly anomaly
-manom Calculate model anomaly
-trend Calculate variable grid-by-grid trends
-aggr Calculate model aggreement
-freq FREQ Temporal mean frequency: annual/daily/monthly
-reg REG Select region for timeseries output (choices: NH/SH/NH-mid/SH-mid/NH-pole/SH-pole/NA/NAT/CONUS)
-rm Use the realization means
-a Use cell areas for spatial mean calculations
-curve Regridding to curvilinear grids
-w All model means as ens dim (used by -std, -mm, -aggr stats options)
-ci CI confidence interval used in -trend and -aggr options
-regrid regridding on/off
All demonstrations are done using a small number of test data files, but these are applicable for any number of files/models/variables/experiments and so on.
- Get info: Get available CMIP6 data information in a directory with
cmpdata -o info.
This code is licensed under the MIT License.