https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/Head
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://www.nanopub.org/nschema#hasAssertion
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/assertion
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://www.nanopub.org/nschema#hasProvenance
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/provenance
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://www.nanopub.org/nschema#hasPublicationInfo
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/pubinfo
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://www.w3.org/1999/02/22-rdf-syntax-ns#type
http://www.nanopub.org/nschema#Nanopublication
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/assertion
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
http://schema.org/endDate
2026-05-23
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
http://www.w3.org/1999/02/22-rdf-syntax-ns#type
https://w3id.org/sciencelive/o/terms/FORRT-Replication-Outcome
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
http://www.w3.org/2000/01/rdf-schema#label
Hurlbert & Jetz 2007 hotspot scale-dependence — Iberian-birds × HEALPix-NESTED replication outcome
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasConclusionDescription
The replication reproduces Hurlbert & Jetz 2007's qualitative finding that biodiversity-hotspot identity is strongly scale-dependent: at every
HEALPix-NESTED resolution tested (Nside 16 to 512, cell side ≈ 407 km down to ≈ 13 km), hotspot non-overlap rises monotonically with grid
refinement and dissolves only at the coarsest resolutions, matching the pattern reported for Australia and southern Africa. At the 0.25° reference scale (≈ 25 km, HEALPix Nside 256) this replication finds 89.9 % misidentification on a museum-only basis-of-record subset and 97.8 % on an all-observations subset, versus Hurlbert & Jetz's 47.8 % (Australia) and 68.6 % (southern Africa). A gold-standard test using the EU Birds Directive Article 12 expert rangemap polygons (2013–2018, 260 Iberian species) does not close the gap (85–95 % misidentification on the matched-species subset), attributing the magnitude offset predominantly to observer-effort bias in the modern citizen-science GBIF atlas itself (log-log Pearson r ≈ 0.96 between per-cell record count and per-cell species count for the all-observations strategy at Nside 256), rather than to a refutation of Hurlbert & Jetz's underlying claim — hence Partially Supported.
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasConfidenceLevel
https://w3id.org/sciencelive/o/terms/Moderate
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasEvidenceDescription
Hotspot misidentification (symmetric set non-overlap of top-5 % richest cells) was computed at six HEALPix-NESTED resolutions for two GBIF
basis-of-record strategies on Iberian birds, year-split 2000 (atlas = post-2000 occurrences; rangemap-equivalent = pre-2000 species convex
hulls).
Headline numbers at Nside 256 (≈ 25 km, the 0.25° equivalent of Hurlbert & Jetz Table 2):
- museum strategy (PRESERVED_SPECIMEN + MACHINE_OBSERVATION): 89.9 % misidentified (Wilcoxon signed-rank p < 10⁻¹⁰⁰).
- allbor strategy (+ HUMAN_OBSERVATION, citizen science included): 97.8 % misidentified (Wilcoxon signed-rank p < 10⁻¹⁰⁰).
- Hurlbert & Jetz 2007 Australia: 47.8 %.
- Hurlbert & Jetz 2007 Southern Africa: 68.6 %.
Direction-of-effect agreement is monotone across all six Nsides for both strategies, replicating the central scale-dependence finding.
Gold-standard rangemap test at Nside 256 (matched species subset, 260 Iberian breeding species also present in our post-2000 GBIF set): EU Birds Directive Article 12 expert polygons (2013–2018, 10 × 10 km grid cells, EEA CC-BY 4.0) yield misidentification of 94.9 % for museum (n = 213 matched species) and 85.2 % for allbor (n = 226).
The expert rangemap does NOT close the gap with Hurlbert & Jetz's 47.8–68.6 % range — refuting the hypothesis that hull-as-rangemap substitution is the dominant cause of the magnitude offset.
Mechanism decomposition at Nside 256:
- Atlas-vs-observer-effort per-cell correlation (log-log Pearson r at Nside 256): r = 0.48 (museum), r = 0.96 (allbor); the all-observations atlas is essentially a log-linear function of per-cell record count, so for allbor the "atlas hotspots" predominantly index citizen-science observer effort (cities, accessible protected areas) rather than biological richness. This is the dominant residual: under any rangemap substitute including the gold-standard Article 12 expert polygons, the atlas top-5 % set consistently disagrees with the rangemap top-5 % set, because the atlas top-5 % cells are observer hotspots while the rangemap top-5 % cells are biological hotspots.
- Top-K sweep: at top-25 % (versus Hurlbert & Jetz's top-5 %), misidentification drops to 74.2 % (museum) and 71.6 % (allbor), inside the Hurlbert & Jetz Australia–southern-Africa range.
- Concave-hull substitute (shapely 2.0, ratio = 0.3) closes 6.3 percentage points for museum (89.9 → 83.0) and 13.5 percentage points for allbor (97.8 → 83.8) on the full-species set, confirming that convex hulls over-predict presence — a secondary contributor relative to the observer-effort dominant.
Land-mask sensitivity (peninsula-only cells via NaturalEarth 10m) shifts the Nside 256 number by ≤ 4.3 percentage points and is therefore not a material confounder. Per-species temporal hull drift (Jaccard distance between pre-2000 and post-2000 convex hulls) is large for museum (median 0.81) and moderate for allbor (median 0.21); this does not materially shift the aggregate top-5 % misidentification at Nside 256 (Δ < 1 percentage point), because the pre-2000 and post-2000 hulls, though individually differently shaped, are similarly inflated in total
area.
GitHub repository: https://github.com/annefou/sdm-scale-replication
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasLimitationsDescription
The magnitude offset relative to Hurlbert & Jetz 2007 is dominated by a single methodological factor that is documented and characterised but cannot be removed inside this replication. The direction of the scale-dependence finding is unaffected.
1. Atlas-axis observer-effort confounding (DOMINANT). Per-cell GBIF record count and per-cell species count correlate at log-log Pearson r ≈ 0.96 for the all-observations (allbor) strategy at the headline scale, and r ≈ 0.48 for the museum-only strategy. The "atlas hotspots" derived from modern Iberian GBIF therefore predominantly index where citizen-science observers go (cities, accessible protected areas, well-birded corridors) rather than where birds are biologically densest. The gold-standard rangemap test using EU Birds Directive Article 12 expert polygons (Test 6, matched subset, 213–226 species) yields 85–95 % misidentification at the headline scale — confirming that the residual is not a rangemap-substitute artefact but is genuinely the observer-effort distortion of the modern atlas. Hurlbert & Jetz's 1990s atlases had observer bias too, but to a much smaller degree than the current citizen-science-dominated GBIF record.
2. Top-K hotspot-threshold choice. Hurlbert & Jetz's top-5 % is a convention, not a biologically grounded threshold. At top-25 %, the misidentification numbers in this replication (74.2 % museum, 71.6 % allbor) sit inside the Hurlbert & Jetz Australia–southern-Africa reference range (47.8–68.6 %). The specific magnitude of the gap is K-sensitive in a way the original paper did not surface.
3. Hull-as-rangemap substitute (SECONDARY). The canonical pipeline uses convex hulls of pre-2000 GBIF occurrences as a rangemap surrogate where Hurlbert & Jetz used expert BirdLife polygons. Convex hulls over-predict presence; concave hulls close 6.3–13.5 percentage points of the magnitude gap at the headline scale on the full-species set. The gold-standard Article 12 test (above) shows this is a real but secondary contributor — the expert rangemap does not close the residual once observer-effort confounding is in play.
Two further caveats. At the coarsest resolutions tested (Nside 16 with 8 Iberian cells, Nside 32 with 29) the top-5 % hotspot set rounds to one or two cells, so per-Nside numbers at those resolutions are noise-dominated; the conclusion does not rely on them. Cell geometry differs from Hurlbert & Jetz (HEALPix equal-area NESTED here, lat-lon graticule there); at Iberia's mid-latitude this is unlikely to shift the comparison materially but is not directly tested.
Despite these limitations, the direction of Hurlbert & Jetz's finding — that hotspot identification is unstable across resolutions and stabilises only at very coarse cells — is robustly replicated in modern Iberian bird data with HEALPix-NESTED indexing. The observer-effort confound is itself an empirical extension of Hurlbert & Jetz's underlying message that the choice of data source materially changes which cells qualify as hotspots: with modern citizen-science atlases, the observer-effort axis becomes the dominant data-source distortion rather than the range-map polygon choice that Hurlbert & Jetz centred on.
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasOutcomeRepository
https://doi.org/10.5281/zenodo.20363556
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/hasValidationStatus
https://w3id.org/sciencelive/o/terms/PartiallySupported
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/o/terms/isOutcomeOf
https://w3id.org/sciencelive/np/RA-H2_b7MJxxUg9nJH8McxccG_6Kr3s8xlXQM8vrnamnU/hj2007-iberia-healpix-study
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/provenance
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/assertion
http://www.w3.org/ns/prov#wasAttributedTo
https://orcid.org/0000-0002-1784-2920
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/pubinfo
https://orcid.org/0000-0002-1784-2920
http://xmlns.com/foaf/0.1/name
Anne Fouilloux
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://purl.org/dc/terms/created
2026-05-24T07:42:57.784Z
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://purl.org/dc/terms/creator
https://orcid.org/0000-0002-1784-2920
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://purl.org/dc/terms/license
https://creativecommons.org/licenses/by/4.0/
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://purl.org/nanopub/x/introduces
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/hj2007-iberia-scale-outcome
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://purl.org/nanopub/x/wasCreatedAt
https://platform.sciencelive4all.org
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
http://www.w3.org/2000/01/rdf-schema#label
Hurlbert & Jetz 2007 hotspot scale-dependence — Iberian-birds × HEALPix-NESTED replication outcome
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
https://w3id.org/np/o/ntemplate/wasCreatedFromTemplate
https://w3id.org/np/RA2zljn0Nw9SadppOyxZoh-_Rxosslrq-vYG-p9SttnJE
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/sig
http://purl.org/nanopub/x/hasAlgorithm
RSA
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/sig
http://purl.org/nanopub/x/hasPublicKey
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
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/sig
http://purl.org/nanopub/x/hasSignature
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
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/sig
http://purl.org/nanopub/x/hasSignatureTarget
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI
https://w3id.org/sciencelive/np/RAzeZKbUCEMXZXDc-WzgHZ4K5mOMwotYhS2uCKDDmdcHI/sig
http://purl.org/nanopub/x/signedBy
https://orcid.org/0000-0002-1784-2920