Anthropogenic Warming Assessment

The Indicators of Global Climate Change (IGCC) project provides annual updates to key IPCC assessments. This repository contains the results and generating code for the assessment of the level and rate of global warming, and the attributed contributions to them.

IGCC Releases and Citations

Summary Table

Indicator Year	Code Release	Paper Reference	Paper DOI	Dataset Reference	Dataset DOI
2022	IGCC-2022	Forster et al. (2023) - Section 7		Smith et al. (2023)
2023	IGCC-2023	Forster et al. (2024) - Section 7		Smith et al. (2024)
2024	IGCC-2024	Forster et al. (2025) - Section 8		Smith et al. (2025)

Note:

• The GitHub Code Release links to the specific release version of the code and results in this repository used to generate results and figures. The generating code here does not have a DOI.
• The Paper Reference links to the peer-reviewed publication that presents the results and figures of the IGCC assessment. This has a permanent DOI, and can be cited.
• The Dataset Reference is the data repository that contains the results of the IGCC assessment. This has a permanent DOI, and can be cited.

Accessing historical release versions of this repository

Each annual update of the IGCC will be available as a separate release in this repository.

The release version is identified by the year of the assessed indicator, and not the year of publication, e.g. results for the levels of attributed warming in 2022 (as published in Forster et al. (2023)) are available in the release named IGCC-2022, not IGCC-2023.

By default, the version of the results and code displayed on the main page of the repository is the most recent version.

The code and results for a specific iteration of the IGCC can be accessed in a number of ways:

1. Access the results directly on GitHub:
   1. Navigate to the code for a specific version directly on GitHub at https://github.com/ClimateIndicator/anthropogenic-warming-assessment/tree/<IGCC-version>, where <IGCC-version> should be replaced by the required version, for example IGCC-2022.
   2. Equivalently, you can display a given release by selecting the required version from the tags list in the branch drop down menu on this page (which will say master by default).
2. Download the zip containing the code and results:
   1. Navigate to the releases page
   2. Download the zip file associated with the required release version, which is listed in the 'Assets' section of a given release.
3. Use the command line locally to clone the repository and checkout the required version:
   1. git clone <repo url>: clone the repository
   2. cd anthropogenic-warming-assessment: move into the newly cloned local repository
   3. git checkout <IGCC-version>: checkout the required version, replacing <IGCC-version> with, for example, IGCC-2022 - the local workspace will now contain the historical repo for the required version.

Methods

Multi-method Assessment

The multi-method assessment of anthropogenic warming is based on the results from three individual attribution methods. The assessment approach is detailed in section 7 of the IGCC 2022 report (Forster et al. (2023)), and again in each subsequent iteration.

Running the code

The code for the multi-method assessment is contained in the script multi_method_assessment.py. It takes as input the results from all three individual attribution methods, and produces the overall multi-method assessment.

Plots of the results can be created by running python multi_method_assessemnt.py inside the anthropogenic-warming-assessment directory. The plots will be saved in a newly created plots/ directory. All the required data for this is contained in the results/ directory.

Conda Environment

The conda environment used to generate the results in this repository is available in the file environment.yml to aid reproducibility.

To keep the environment file as cross-platform as possible, only the directly installed packages and their version during the analysis are included. Note that the analysis was only carried out and tested on a Linux computing cluster.

The environment can be installed by running conda env create --name <env-name> --file environment.yml where you can specify your preferred environment name in place of <env-name>.

Note, you may need to rebuild the matplotlib font cache using rm -rf ~/.cache/matplotlib/ to ensure that matplotlib can find the fonts that are installed through the open-fonts conda package included in the environment file.

Individual Attribution Methods

Three attribution methods are used for the multi-method assessment of anthropogenic warming in the IGCC papers. These methods are detailed in the supplement of the IGCC papers. They are:

1. (GWI) Global Warming Index, with results available in the files results/Walsh_*.csv.
2. (KCC) Kriging for Climate Change, with results available in the files results/Ribes_*.csv.
3. (ROF) Regularised Optimal Fingerprinting, with results available in the files results/Gillett_*.csv.

There is no pipeline that automatically pulls in the results for each method; the results files were simply manually copied into the results/ directory, and pushed to this GitHub repository.

Global Warming Index

The code for the GWI attribution is available at global-warming-index (github.com).

Regularised Optimal Fingerprinting

The esmvaltool code (which generates ROF attribution results based on CMIP6 outout) is available on GitHub at ESMValGroup/ESMValTool at forster23 (github.com). The actual python diagnostics code is available here: ESMValTool/esmvaltool/diag_scripts/attribute at forster23 · ESMValGroup/ESMValTool (github.com).

Kriging for Climate Change

The code for the KCC attribution in the IGCC 2022 report based on code contained in the following GitLab repo: Global temperature constraint (gitlab.com).

Results

The results from all three individual attribution methods, and the overall multi-method assessment, are provided in the results/ directory in this repository, which contains the following:

• Overall multi-method assessment
  • Assessment-6thIPCC_headlines.csv: assessment results directly quoted from the IPCC's sixth assessment cycle (including both AR6 and SR1.5).
  • Assessment-Update-<indicator-year>_GMST_headlines.csv: the results from the multi-method assessment for the level of warming in the <indicator-year> update of the IGCC.
  • Assessment-Update-<indicator-year>_GMST_rates.csv: the results for the rates of warming in the <indicator-year> update of the IGCC.
  • Assessment-Extrapolation-<indicator-year+1>_GMST_headlines.csv: a preliminary estimate for the next year of anthropogenic warming based on a linear extrapolation of each attribution method.
• Indivudual attribution methods
  • <Surname>_GMST_headlines.csv: the results from the individual attribution methods that feed into the multi-method assessment.
  • <Surname>_GMST_timeseries.csv: the single-year timeseries from the individual attribution methods.
  • <Surname>_GMST_rates.csv: the decadal rates for the individual attribution methods.
  • Note: <Surname> is replaced in each case by the surname of the lead author of each method, i.e. Walsh, Ribes, and Gillett for the GWI, KCC, and ROF methods, respectively.

The results from this anthropogenic_warming_assessment GitHub repository are formally available with metadata in the ClimateIndicator/data GitHub repository, and with a citable DOI in the Zenodo repository - note the specific versioning for each release differs from this feeder repository.