Processed data and analysis code underlying Figure 2 of "Artificial Intelligence Pathways from Weather to Climate"
Year: 2026
Type: dataset
Abstract: This record contains the processed data and analysis code underlying Figure 2 of the manuscript Artificial Intelligence Pathways from Weather to Climate. Figure 2 presents a CliMA-S2S week-3 probabilistic forecast case initialized on 9 July 1997. The figure shows seven-day-mean 300-hPa geopotential height (z300) and 2-m temperature (t2m), together with ERA5 verification fields and ranked probability skill score (RPSS) and continuous ranked probability skill score (CRPSS). The archive contains processed CliMA-S2S and ERA5 fields for six initialization dates, climatology data for 1981–2022, cached leave-one-initialization-out calibration products, and the Python analysis scripts required to reproduce the Figure 2 probabilistic panels. For the 9 July 1997 test initialization, predictive uncertainty is obtained using statistical calibration of the single deterministic model trajectory rather than an ensemble of perturbed initial conditions. The included README.md documents the archive structure, dependencies, methodology, and command required to reproduce the figure. ERA5 data are derived from the Copernicus Climate Change Service; attribution and licensing information are provided in NOTICE.
Author: Christopoulos, Costa
Institution: California Institute of Technology
Language: English
Subfield: Atmospheric Science
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDG: Climate action
Open Access status: green
OpenAlex ID: W7211990706