Model-Data multi-variable intercomparison


You can download a PDF version of the diagram below by clicking on it.

See the commands that produce these diagrams

 Restored fields 

 Non-restored fields 

Total space-time field Multi-component diagram

We compute comparison statistics and show them via two Taylor diagrams:

1) variables with restored fields:

2) variables with non-restored fields

The commands that produce these diagrams:

1st step: The commands that compute the statistics NetCDF files

  1. Levitus/SSS statistics: (file LEV_SSS_comp_stat/LEV_SSS_comp_stat.nc)

    >>> python model_vs_data_comp_stat.py -oLEV_SSS_comp_stat/LEV_SSS_comp_stat.nc -tLEV_SSS_comp_stat/LEV_SSS_comp_stat.txt /home/geocean/ocmip/data/gridded/LEV1998_monthly.nc TEMPERATURE "/home/geocean/ocmip/phase2/%(MODEL)/Abiotic/equil/%(model)_LEV_SSS.nc" SSS


  2. Levitus/SST statistics:

    >>> same: replace SSS with SST

  3. Takahashi/CO2 flux: (file TAKA_FCO2/comp_stat/TAKA_FCO2_comp_stat.nc)

    >>> python model_vs_data_comp_stat.py -oTAKA_FCO2/comp_stat/TAKA_FCO2_comp_stat.nc -tTAKA_FCO2/comp_stat/TAKA_FCO2_comp_stat.txt /home/geocean/ocmip/data/gridded/taka2001/TAKA_fco2_2002.nc FCO2 "/home/geocean/ocmip/phase2/%(MODEL)/Abiotic/hist/%(model)_TAKA_Ftot_1995_drift.nc" FTOT

  4. World Ocean Atlas 2001/Phosphate concentration

    >>> python model_vs_data_comp_stat.py -oWOA_PO4_comp_stat/WOA_PO4_comp_stat.nc -s1000 -tWOA_PO4_comp_stat/WOA_PO4_comp_stat.txt -2 /home/geocean/ocmip/data/gridded/louanchi/po4_louanchi.nc,PO4 /home/geocean/ocmip/data/gridded/WOA2001/po4_lev.nc po4 "/home/geocean/ocmip/phase2/%(model)/Biotic/%(model)_LEV_PHOSPHATE.nc" PHOSPHATE

  5. to 7.:

    >>> python model_vs_data_comp_stat.py ...

    same command with appropriate parameter list specifying model-output and data files and variables

2nd step: The commands that read the statistics files and produce the Taylor diagrams, above

Restored fields

>>> python plot_taylor_diagrams.py -i32 -o ../Multivariate/T_S_PO4.ps -pv ../LEV_SSS_comp_stat/LEV_SSS_comp_stat.nc ../WOA_PO4_comp_stat/WOA_PO4_comp_stat.nc ../LEV_SST_comp_stat/LEV_SST_comp_stat.nc

Non-restored fields

>>> python plot_taylor_diagrams.py -i32 -o ../Multivariate/C14_CFC_DIC_Flux.ps -pv ../TAKA_FCO2/comp_stat/TAKA_FCO2_comp_stat.nc ../Anthropic_CO2/comp_stat/high_lat_masked/comp_stat.nc ../C-14/comp_stat/comp_stat.nc ../CFC11/comp_stat/comp_stat.nc
The argument list for these commands simply list all the NetCDF files computed in the first step, one for each variable.

Option '-v' requests a multi-variate Taylor diagram
Option '-i' sets the size of code-letter for each model on the diagram (size ranges from 1 to 100)
Option '-o' allows to specify output image file
Option '-p' selects 'Portrait' output format