data · 11 raw datasets + derived JSONs

Data

All raw datasets, derived JSONs, and reproducible code.

Every dataset used to build this atlas is openly available. Below are direct download links, citations, and where each one lives in the project’s data pipeline.

Raw datasets (downloaded)

DatasetProviderURLDOIUsed for
Berkeley Earth land+ocean 1° × 1° (1750–2024)Berkeley Earthhttps://berkeley-earth-temperature.s3.us-west-1.amazonaws.com/Global/Gridded/Complete_TAVG_LatLong1.ncRohde & Hausfather 2020All stripes, all trends
HadCRUT5 ensemble mean (1850–2026)UK Met Office / UEAhttps://crudata.uea.ac.uk/cru/data/temperature/HadCRUT.5.1.0.0.analysis.anomalies.ensemble_mean.ncMorice et al. 2021Cross-validation
HadSST4 medianUK Met Office / UEAhttps://crudata.uea.ac.uk/cru/data/temperature/HadSST.4.2.0.0_median.ncSST component
CRUTEM5 land anomaliesUK Met Office / UEAhttps://crudata.uea.ac.uk/cru/data/temperature/CRUTEM.5.1.0.0.anomalies.ncLand-only cross-check
NOAA GHCN-Monthly v3.3.0NOAA NCEIhttps://www.ncei.noaa.gov/pub/data/ghcn/v3/ghcnm.tavg.latest.qcu.tar.gzMenne et al.Station-level cross-check
NOAA Global Surface Temp v6.1NOAA NCEIhttps://www.ncei.noaa.gov/data/noaa-global-surface-temperature/v6.1/access/gridded/Yin et al. 2024Independent gridded estimate
GISTEMP v4 (NASA)NASA GISShttps://data.giss.nasa.gov/gistemp/tabledata_v4/Lenssen et al. 2019Cross-validation
RAPID 26°N AMOC (2004–2024)NOC / BODChttps://rapid.ac.uk/sites/default/files/rapid_data/moc_transports.nc10.5285/48d0bf43-0598-ceb2-e063-7086abc062f1AMOC subsection
OSNAP 2014–2022 MOCGeorgia Techhttps://doi.org/10.35090/gatech/78023Fu et al. 2025AMOC cross-validation
Jomelli 2022 AMOC reconstruction (10k–2k BP)NOAA Paleoclimatehttps://www.ncei.noaa.gov/pub/data/paleo/reconstructions/vincent2022/amoc_recon_vincent2022.txt10.1038/s41467-022-28939-9AMOC Holocene context
Mjell 2019 sortable siltPangaeahttps://doi.pangaea.de/10.1594/PANGAEA.899459Mjell et al. 2016Deep-current proxy

Derived data (downloadable from this site)

  • /data/stripes_index.json (6.3 MB) — Index of all 9,685 locations with metadata, trends, and stripe-SVG paths
  • /data/amoc_analysis.json (2 KB) — Summary of RAPID + OSNAP trend analysis + EWS results
  • /data/rapid_ts.json (13 KB) — RAPID monthly AMOC time series, 240 months
  • /data/osnap_ts.json (5 KB) — OSNAP monthly AMOC time series, 96 months
  • /data/recon_ts.json (5 KB) — Jomelli 2022 AMOC reconstruction, 161 samples
  • /data/ews_windows.json (70 KB) — EWS variance/autocorrelation in 2/3/4/5-year sliding windows

Reference datasets (downloaded for context, not used in atlas)

  • GeoNames cities500 — 234,668 populated places worldwide
  • GeoNames admin1CodesASCII — 3,865 state/province codes

Reproducible code

The full pipeline lives in the project’s code/ directory. Each script is standalone and can be re-run from raw data:

  1. code/build_locations.py — extracts the 11,608 canonical locations from GeoNames + Berkeley Earth grid
  2. code/extract_series.py — extracts monthly temperature series for each location, computes annual means + OLS trends
  3. code/generate_stripes.py — generates per-location SVG stripes (annual + monthly-detail versions)
  4. code/amoc_analysis.py — RAPID/OSNAP/AMOC-recon trend + EWS analysis

Total pipeline runtime: ~5 minutes on the VM. Total disk usage: ~1.5 GB raw + 180 MB processed + 180 MB SVGs.

Citation for this work

Herman. (2026). "Climate Stripes Atlas: Local Warming for 11,608 Locations,
1850-2024, with an Investigation of AMOC Collapse Timing." 
climate.hermanity.dev.

The full paper-style write-up with the 19-citation table is on the Paper page.

License

All raw datasets are used under their respective provider licenses (Berkeley Earth: CC-BY-NC, NOAA: public domain, UK Met Office: open government license, Pangaea: CC-BY). The derived SVGs, JSONs, and code in this project are released under CC-BY 4.0.