Scripts used for the publication "An assessment of basal melt parameterisations for Antarctic ice shelves" ========================================================================================================
These are the scripts that were developed and used for the publication: Burgard, C., Jourdain, N. C., Reese, R., Jenkins, A., and Mathiot, P.: An assessment of basal melt parameterisations for Antarctic ice shelves, The Cryosphere, https://doi.org/10.5194/tc-2022-32, in press, 2022.
If you want to play around with the parameterisations and your own data, have a look at the package multimelt: https://github.com/ClimateClara/multimelt
Useful functions are grouped in the package assess_param_funcs. To install them and use them in further scripts, don't forget to run
pip install .The scripts to format the data and produce the figures can be found in the folder scripts_and_notebooks.
Note - In the scripts, the NEMO runs are called 'OPM+number'. Here are the corresponding names given in the manuscript: OPM006=HIGHGETZ, OPM016=WARMROSS, OPM018=COLDAMU and OPM021=REALISTIC.
Initial data formatting (from raw NEMO output to interesting variables gridded on stereographic grid)
The scripts for the initial formatting of the data can be found in scripts_and_notebooks/data_formatting. Start with prepare_data_NEMO.sh, then move to custom_lsmask.ipynb and finally to regridding_vars_cdo.ipynb. At this point you have the relevant NEMO fields on a stereographic grid.
The scripts to prepare the ice-shelf masks, the box and plume characteristics, and the temperature and salinity profiles can be found in scripts_and_notebooks/pre_processing.
isf_mask_NEMO.ipynb and isf_mask_BedMachine.ipynb prepare masks of ice shelves, and plume and box characteristics, on the NEMO and Bedmachine grid (the latter is needed for Fig. 8) respectively.
prepare_reference_melt_file.ipynb prepares 2D and 1D metrics of the melt in NEMO for future comparison to the results of the parameterisations.
T_S_profile_formatting_with_conversion.ipynb converts the 3D fields from conservative temperature to potential temperature and from absolute salinity to practical salinity.
T_S_profiles_front.ipynb prepares the average temperature and salinity profiles in front of the ice shelf.
T_S_profiles_Dutrieux14.ipynb converts the profiles given in Dutrieux et al. 2014 to the format needed for my formulation of the parameterisations (for Figure 8).
The scripts to conduct the cross-validation, the best-estimate tuning and the tuning on different bootstrap samples can be found in scripts_and_notebooks/tuning.
For the simple parameterisations, run prepare_2D_thermal_forcing_simple.ipynb, prepare_1D_thermal_forcing_term_simple_for_linreg.ipynb, and tuning_cluster_ALL_CV_BT.ipynb.
For the more complex ones, usee the bash script run_generalized_tuning_script.sh to call the python script run_generalized_tuning_from_bash_crossval.py and run the tuning on the different samples (either leave-one-ice-shelf-out or leave-one-time-block-out or the whole sample or a random bootstrap sample). Then group the tuneed parameters with group_CV_parameters.ipynb (cross-validation) and/or group_BT_parameters.ipynb (bootstrap).
The names of the parameterisations in the files and scripts are:
- 'linear_local' for the linear, local parameterisation;
- 'quadratic_local' for the quadratic, local parameterisation using a constant slope for all Antarctica; 'quadratic_local_cavslope' for the quadratic, local parameterisation using one slope on the cavity level of each ice shelf; 'quadratic_local_locslope' for the quadratic, local parameterisation using a slope on the grid-cell level;
- 'quadratic_mixed_mean' for the quadratic, semilocal parameterisation using a constant slope for all Antarctica; 'quadratic_mixed_cavslope' for the quadratic, semilocal parameterisation using one slope on the cavity level of each ice shelf; 'quadratic_mixed_locslope' for the quadratic, semilocal parameterisation using a slope on the grid-cell level;
- 'lazero19_2' for the plume parameterisation as suggested by Lazeroms et al. (2019); 'lazero19_modif2' for the modified plume parameterisation as suggested in this paper;
- 'boxes$n_pism$i_picop$j' are the box and PICOP parameterisations. $n = number of the configuration, where '1' = 2 boxes, '2' = 5 boxes, '3' = 10 boxes, '4' = PICO boxes; $i is yes or no, where yes is heterogeneous boxes and no is homogeneous boxes; $j is yes or no, where yes is PICOP (using the plume parameterisation to infer the melt) and no is "normal" box parameterisation.
The scripts to run the parameterisations can be found in scripts_and_notebooks/apply_params.
evalmetrics_results_CV.ipynb computes the integrated melt and the melt near the grounding line, applying the parameters of the cross-validation on the corresponding left out time block or ice shelf and applyong the original parameters.
script_to_apply_all_param_basic_application.ipynb computes several 2D and 1D metrics resulting from the parameterisations for a given set of parameters for one NEMO run (useful for spatial patterns for example).
apply_param_PIGL_dutrieux_BedMachine.ipynb computes the parameterisations using best-estimate parameters applied to BedMachine output and the Dutrieux profiles. Only for Pine Island. For Figure 8.
apply_pointbypointRMSE_box1_forFigF1.ipynb computes the difference between parameterised and reference melt in box1 point by point (for the comparison in Fig. F1).
The scripts to finalise the figures can be found in scripts_and_notebooks/figures.
Figures 2 and 3 are done with Figures_2_and_3.ipynb.
Figures 4, 7, E1, E2, E3 and values for Tables 3, 5, 7, 9 are done with Figures_4_7_E1_E2_E3.ipynb and check_RMSE_orig_parameters.ipynb.
Figure 5 is done with prepare_data_Figures_5_6.ipynb and Figure_5.ipynb.
Figure 6 is done with prepare_data_Figures_5_6.ipynb and Figure_6.ipynb.
Figure 8 is done with Figure_8a.ipynb and Figure_8b.ipynb.
Figure 9 is done with Figure_9.ipynb.
Figure F1 is composed of the left panel of the figure created with Figure_E1_leftpanel.ipynb and of the right panel of Fig. 7.
Figure B1 is done with Figure_B1.bash and scripts found in tools_fig_B1B2B3/VALSO/ (this is the version downloaded from https://github.com/pmathiot/VALSO on October 11th 2022).
Figure B2 is done with Figure_B2.bash and scripts found in tools_fig_B1B2B3/PyChart/ (this is the version downloaded from https://github.com/pmathiot/PyChart on October 11th 2022).
Figure B3 is done with Figure_B3.bash and scripts found in tools_fig_B1B2B3/PyChart/ (this is the version downloaded from https://github.com/pmathiot/PyChart on October 11th 2022).
Quick figures (not looking as good as in the paper) B1, B2 and B3 can be made with reproduce_plots_appendix.ipynb