Using net-zero carbon debt to track climate overshoot responsibility

• Paper: https://doi.org/10.1073/pnas.2409316122
• Archive of record: https://zenodo.org/badge/latestdoi/734527284 (every release is archived on Zenodo)
• Interactive explorer: https://setupelz.com/carbondebt/
• Summary page: https://setupelz.com/work/carbon-debt.html
• Earlier deposit of this archive: https://doi.org/10.5281/zenodo.14915596

Replication archive for:

Pelz, S., Ganti, G., Lamboll, R., Grant, L., Smith, C., Pachauri, S., Rogelj, J., Riahi, K., Thiery, W. and Gidden, M.J. (2025). Using net-zero carbon debt to track climate overshoot responsibility. Proceedings of the National Academy of Sciences 122(13), e2409316122. https://doi.org/10.1073/pnas.2409316122

Archive releases: https://github.com/setupelz/repl_2023_carbondebt/releases (the large inputs and the pipeline outputs are attached there, see Data). Interactive explorer: https://setupelz.com/carbondebt/

What is here

Folder	Content
Code/	The analysis pipeline, R scripts and two Jupyter notebooks, numbered in run order (below).
Code/explorer/	Build script for the interactive explorer (Python, fair-shares).
Data/	Inputs by source (countrygroups/, equity_data/, fairparams/, impacts/, pathways/), explorer inputs (explorer/), and MANIFEST.md5 for the release-attached files. Pipeline outputs go to Data/processed/.
Manuscript/Figures/	Main and SI figures as written by the scripts.
site/	The net-zero carbon debt explorer, a static page.

Pipeline

Run in numeric order from the project root (replication_nzcdebt.Rproj).

Script	Does
00_prep_ar6_filter.R	Filter and pre-process the AR6 scenario database.
00_tempassessment.ipynb, 00_tempassessment_impren.ipynb	FaIR temperature assessment of the AR6 and NDC/LTS pathways (Python, environment.yml).
01_remainingcarbonbudget.R, 01a_remainingcarbonbudget_SI.R	Remaining carbon budgets for countries and regions; SI variants of the warming-contribution allocations.
02_carbondebt_systematic.R	Net-zero carbon debt across the AR6 database.
03_nztargetassessment.R	Net-zero targets for the analysis countries.
04_ndcltspathwayprocessing.R	Pre-processing of the assessed NDC and LTS pathways.
05_heatwaveexposure.R	Lifetime exposure to extreme heatwaves under the assessed pathways.
06_ndcltsassessment.R	Evaluation of the assessed pathways.

R packages are loaded with pacman at the top of each script. The FaIR notebooks use the conda environment in environment.yml (fair==2.2.0).

Data

Everything under Data/ is in the repository except the files listed in Data/MANIFEST.md5, which are attached to the release instead because of their size: the five largest inputs (SSP database, GMST response fields, historical emissions, heatwave and cohort netCDFs) and the fourteen pipeline outputs in Data/processed/. Download them from the release page into the folders named in the manifest and verify:

md5sum -c Data/MANIFEST.md5

With the outputs in place the figure scripts run without rerunning the FaIR notebooks. Sources and licences of the third-party inputs: Data/README.md.

Explorer

site/ is the net-zero carbon debt explorer (originally a 2024 Shiny app for stakeholder engagement in the ELEVATE project, rebuilt as a static page). It applies the carbon-debt accounting of the paper, debt = cumulative emissions minus fair-share allocation, across three temperature goals, six accounting start years, two future-emissions scenarios and two capability adjustments for the IAMC R10 regions. Remaining carbon budgets come from the fair-shares library (Lamboll et al. 2023). Inputs are the three CSVs in Data/explorer/; Code/explorer/build.py writes site/assets/data.json:

cd Code/explorer && uv run python build.py
cd ../../site && python3 -m http.server     # local preview

Citation

See CITATION.cff. Cite the paper for the method and results, and the archive release for the code and data.

Licence

MIT for the code and figures in this archive (LICENSE). Third-party inputs keep their own licences, listed in Data/README.md.