• SCC code. This code takes in damage function coefficients estimated previously, and outputs SCC values.
• /functions/: this folder contains parameters from the FAIR climate model (See Step 5 of Methods in the paper) and functions called by the SCC code.

You can start a jupyter session in the browser by running jupyter lab and navigate to the notebooks in this directory.

Details of the calculations done in this code can be found in Appendix Section F, and in comments inside the code itself.

• This code can be run for any of the three econometric models for which we project impacts under all 33 climate projections in the SMME and for which we present SCC values in the paper. Please select the model you wish to run at the top of this code, by changing the model variable in the first cell of the code.
• These are defined as in the readme for 3_post_projection:
  • main, which can run for SSP2, SSP3, or SSP4.
  • lininter and lininter_double,lininter_half, which are run only for SSP3.
• The main-SSP3 SCC is calculated for 8 different price scenarios (Appendix Section D). All other permutations are calculated only for the price014 price scenario (i.e. 1.4% annual price growth).
• To choose the SSP to run (for the main model), edit the ssp variable in the first cell.
• Every model x scenario can be run with either hold_2100_damages_fixed= FALSE (the default and main option used in the paper), or hold_2100_damages_fixed = TRUE. Holding 2100 damages fixed means that we fix the post-2100 damage functions to be the same as the end-of-century damage function (See Appendix Section G.2), rather than allow them to evolve using our usual extrapolation approach (See Appendix Section E). You can change this option by editing the first cell of the code.
• There is an option to generate plots that may help users visualise the calculation process. Set generate_plots = True in the first cell of the code to produce extra plots using this code (these are not directly used in the paper).

• Damage function coefficients, which are stored in: <yourDATA>/OUTPUT/projection_system_outputs/damage_function_estimation/.

• The output of this code are csv files containing SCC values for different scenarios. These csv files are manually processed into the latex tables shown in the paper.
• The output csvs from this code allow users to look at values for other RCPs than are discussed in the paper, and for a broader range of discount rates. They also include a breakdown of the damages by whether they are incurred pre or post 2100.
  • To see the values relevant to the paper, filter the output csv files such that time_cut = all, discrate is in the set (2.5,3,5), and only consider the columns rcp45 and rcp85.
• This code generates values used in:
  • Figure 4E. These central estimates are outputted to <yourDATA>/OUTPUT/figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-False.csv.
  • Table Appendix F.2. The central estimates in this table are taken from /figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-False.csv.
  • Table Appendix G.1. The central estimates in this table are taken from /figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-False.csv.
  • Table Appendix G.2. The values in this table are taken from /figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-True.csv.
  • Table Appendix G.3. The values in this table are taken from the following three csvs:
    • /figures/scc_values/main/scc_energy_SSP2_all_prices_2100-fixed-False.csv
    • /figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-False.csv
    • /figures/scc_values/main/scc_energy_SSP4_all_prices_2100-fixed-False.csv
  • Table Appendix I.1. The values in this table are taken from the following three csvs.
    • /figures/scc_values/main/scc_energy_SSP3_all_prices_2100-fixed-False.csv
    • /figures/scc_values/lininter/scc_energy_SSP3_all_prices_2100-fixed-False.csv
    • /figures/scc_values/lininter_double/scc_energy_SSP3_all_prices_2100-fixed-False.csv
    • /figures/scc_values/lininter_half/scc_energy_SSP3_all_prices_2100-fixed-False.csv