Reproducible CMIP6 Effective Radiative Forcing Analysis

Scientific context

CMIP6 multi-model experiments provide estimates of Effective Radiative Forcing (ERF) from various anthropogenic drivers.
Understanding ERF variability is key to constraining climate sensitivity.

Internship objectives

1. Ingest raw CMIP6 NetCDF outputs (ERF_EMI_DMS, siconc, …).
   
2. Compute regional and global ERF statistics.
   
3. Visualise spatial patterns and temporal trends with colour-blind-safe maps.
   
4. Publish fully reproducible code, data-retrieval instructions, and archived outputs.

Quick-start (reproduction instructions)

Clone the repository & create the Conda environment

1. git clone https://github.com/IGE-OpenReproLab2025/ShahabM2internship2025Project.git
2. cd ShahabM2internship2025Project

• #### (mamba is faster; conda works too)
  • mamba env create -f environment.yml
  • mamba activate ShahabM2internship

Point the code to your NetCDF folder ── choose one:

• Environment variable (preferred)

  export DMSDATA=/mnt/data-summer-shared/SampleData_Shahab/data/raw

• Hard-code RAW in src/load_data.py

• RAW = Path("/mnt/data-summer-shared/SampleDataShahab/data/raw")

Launch JupyterLab and run the notebooks

 jupyter lab

Download data (~2 GB)

python src/load_data.py --download-full  # ESGF credentials required

Run the notebook

jupyter lab notebooks/ Main.ipynb

Contributors

| Role | Name | ORCID | Contribution | |------|------|-------|--------------| | Developer | Shahab Shahlaei | 0000-0003-3513-2727 | Conceptualisation, coding, data analysis, visualisation, writing | | Scientific supervisor | Dr. Jennie L. Thomas | 0000-0002-5986-7026 | Methodology review, climate-science guidance | | Internship mentor | Dr. Ruth Price | 0000-0003-1981-9860 | Project framing, feedback on narrative | | IT / ESGF support | MikeMYKAELVIGO AurélieAlbert | – | Infrastructure, data-transfer troubleshooting |

Material

This repository ships everything needed to reproduce the CMIP6 ERF_MMSo4 analysis:

1. notebooks/ – executable Jupyter notebooks with clear objectives, narrative, captions and conclusions.
   
2. src/ – reusable Python package (load_cmip6, plot_global_map, compute_erf_statistics, …) with full doc-strings.
   
3. docs/ –
   
   • DATA.md: input/output dataset catalogue, download instructions, licences, preservation plan.
     
   • this README.
     
4. environment.yml – pinned Conda environment (Python 3.11, xarray 2024.5 +, etc.).
   
5. tests/ – pytest/nbval smoke test that runs the main notebook in mock-data mode.
   
6. .zenodo.json + CHANGELOG.md – metadata for long-term archiving and version history.
   
7. data/ – folder created on first download; remains empty in the repo to keep size small.

Licence

| Scope | Licence | File(s) | Terms | |-------|---------|---------|-------| | Source code & notebooks | MIT | Entire repo (see LICENSE) | Permissive: free use, modification & distribution with attribution. | | Generated output data & figures | CC-BY-4.0 | NetCDFs in data/processed/, PNG/SVG plots | Share-alike not required; attribution to this repo. | | Upstream CMIP6 input NetCDFs | CMIP6 Terms of Use | Downloaded via ESGF (load_data.py --download-full) | Free for non-commercial research; cite original modelling centres. |

By cloning, using or contributing to this project you agree to abide by these licences.
If you need a different licensing arrangement, open an issue to discuss before using the material commercially.