Informing cumulative-effects assessment of data-centre expansion: comparative electricity and location-based carbon evidence from the United States, Great Britain, Ireland and the Netherlands (replication package)

This replication package accompanies the manuscript “Informing cumulative-effects assessment of data-centre expansion: comparative electricity and location-based carbon evidence from the United States, Great Britain, Ireland and the Netherlands”. The package contains archived or documented source data, processed datasets, Python analysis scripts, source inventories, a data dictionary, reference-verification records, policy-document coding, an illustrative project-file audit, sensitivity analyses, tables, figures, execution logs and SHA-256 checksums. The supplied workflow reproduces the reported annual data-centre electricity series, allocated electricity-related lifecycle GHG proxies, exact symmetric two-factor accounting decomposition, no-increase accounting comparator, alternative-baseline results, illustrative policy-path sensitivities and the Great Britain boundary and factor-definition sensitivity. The four cases use source-defined data and are not treated as fully harmonised national statistics. The cross-case carbon factors are annual generation-mix lifecycle GHG intensities and the calculated values are allocation proxies, not project carbon footprints, marginal emissions estimates or complete GHG Protocol Scope 2 accounts. Great Britain electricity consumption is paired with a United Kingdom lifecycle factor in the cross-case calculation and is therefore identified as boundary-sensitive; a separate Great Britain operational-CO2 sensitivity is included. The project-file audit is purposive and illustrative rather than representative or systematic. From the package root, the results can be regenerated using: PYTHONDONTWRITEBYTECODE=1 python run_all.py Third-party data retain the licences and reuse conditions of their original providers, as documented in the source inventory.