sdm-scale-replication

This release tests whether the scale dependence of range-map species richness reported by Hurlbert & Jetz (2007) — that mapping ranges too finely can mislabel up to two-thirds of biodiversity hotspots — holds for Iberian birds using 61.7 million GBIF occurrences across six HEALPix-NESTED resolution levels. The original direction replicates (finer scales mislabel more hotspots), but the magnitude is far larger; an independent check against EU Birds Directive Article 12 expert range maps shows the gap is dominated by observer-effort bias in occurrence data, not by the underlying claim. Outcome: partially supported, qualifies the original.

Reference paper: Hurlbert & Jetz (2007), PNAS — https://doi.org/10.1073/pnas.0704469104

This release adds, on top of v0.1.0:

• Completed six-step FORRT nanopublication chain (Quote → AIDA → Claim → Replication Study → Replication Outcome → CiTO Citation), all signed and published.
• Replication Outcome confidence characterised as Moderate.

Also included: the reproducible four-notebook pipeline (data download → clean → analysis → figures), pixi environment + lockfile, Dockerfile, Snakefile, and CI/Jupyter Book workflows.

Prior archive: v0.1.0 — https://doi.org/10.5281/zenodo.20363556 Concept DOI (latest): https://doi.org/10.5281/zenodo.20363555

To cite, see CITATION.cff in the repository.