Skip to content

Nordic source capability and country dimensions

The retained source/country replay joins the governed 11-family source-stage matrix to the 198-cell country coverage ledger. It requires every family to have one row for each of six country partitions and three distinct dimensions: source-reported, governed assignment, and publication. The resulting 66 family/partition entries retain the original availability, lifecycle, reason codes, source snapshot and all nullable counts separately for each dimension. The four Nordic countries and the UNASSIGNED and OUTSIDE partitions are all present.

The parent ledger's 20 input artifacts were rechecked against their byte counts and SHA-256 digests. The source-stage matrix has 11 rows and a 187-row capability/materialization audit; the country ledger has 198 cells. This consumer does not reinterpret a null count as zero, sum incompatible families, or promote capture to scientific admission.

The combined source reconciliation also checks the three historical Rootsi and OpenLand intake receipts and their 12 four-country claims. Those claims remain separate from this 11-family cube; their original archives do not create new admitted country observations.

The Neotoma dataset ledger adds source-bound entity decisions for one family: 85 datasets occur in the selected direct observed-pollen study and 124 do not. Its separate sample prerequisites do not admit any dataset to LRA or measured-vegetation validation. Entity decisions for the other source families still require their own source evidence.

The used-dataset publication ledger joins those 85 Neotoma study datasets to exact provider citation responses. Seventy-nine have one provider-flagged primary publication; six need primary-link review. Thirteen carry distinct source DOI names, with 18 datasets in the union of citation-review cases. Neotoma's policy-level CC BY 4.0 statement and these metadata do not resolve sample-specific scientific admission or vegetation truth.

The existing SEAD site ledger already records all 2,195 discovered site IDs and UUIDs. Governed boundaries assign 2,069 to SE/DK/NO/FI and retain 103 for review plus 23 unassigned; only assigned sites occur in the retrieved and normalized site consumers. This is archaeological site context and country accountability. It does not admit a SEAD pollen series, an LRA cohort, or vegetation truth. The two linked entity ledgers therefore cover different scientific roles and denominators.

LandClim's governed sequence inventory at data/landclim/normalized/nordic_pollen_site_sequences.csv supplies 490 context rows: Sweden 198, Denmark 127, Norway 93 and Finland 72. These rows already feed the atlas and describe source site coverage, not authenticated historical counts. The separate 1,128-row LandClim II study crosswalk leaves every historical primary input identity unresolved; its Nordic country labels are publication metadata. Neither inventory supplies measured vegetation truth or a qualified LRA cohort for the current Neotoma sites.

The human and animal DNA source-scope review keeps two more entity populations distinct. AADR v66 has 51,005 annotation rows across its 1240K and HO captures, representing 27,755 Genetic IDs; these are human-source accountability units. The animal point surface has 170 features backed by 331 distinct sample IDs, of which eight Nordic locality features represent 12 sample IDs. Its 1,468-row foundation still has eleven unresolved paper-link rows. Neither population supplies pollen counts or an independent vegetation reference, and the eight animal locality features are not a Nordic pollen cohort.

The Nordic AADR entity account joins all 1,325 exact four-country source Genetic IDs to their country sample products: 1,231 appear and 94 lack source coordinates. The animal entity account checks the existing 331 mapped and 1,137 map-excluded sample decisions against all 1,468 normalized IDs without copying the original decision rows. These are source-to-product membership results, not scientific or source-country admission for either family.

Family Governed raw stage Governed published stage Four-country publication availability Qualification
AADR present present collected in SE/DK/NO/FI authority and source review required
animal aDNA present present collected in SE/DK/FI; zero observations in NO authority and biological sample review required
boundaries present present collected in SE/DK/NO/FI boundary inclusion review required
EMPD2 missing missing blocked in SE/DK/NO/FI source-reported partial coverage is not an admitted reference
HYDE missing missing blocked in SE/DK/NO/FI private custody is separate from governed source admission
LandClim present present collected in SE/DK/NO/FI context publication is not vegetation truth
Neotoma present present collected in SE/DK/NO/FI point context is not LRA admission
NFI missing missing blocked in SE/DK/NO/FI source and country reference review required
RAÄ present refused blocked in SE; source unavailable elsewhere authority and review unresolved
SEAD present present collected in SE/DK/NO/FI source-reported country cells are zero, governed site assignments are nonzero
SVAR present present blocked in SE; source unavailable elsewhere source authority refused

“Collected” in this table is the exact available_collected publication availability code, not a claim that every source family has an approved analytical use. The full report supplies the separate lifecycle status and count denominator for each cell. In particular, SEAD's source-reported zero country cells coexist with governed and published site counts because those dimensions answer different questions. The source-stage matrix's missing HYDE row describes governed repository materialization; it does not deny the separately retained private archive.

Reproduce

From the repository root, use a fresh output directory:

PYTHONDONTWRITEBYTECODE=1 .venv/bin/python -m \
  bijux_pollenomics_dev.source_completion.source_capability \
  --repository "$PWD" \
  --output "$PWD/artifacts/jobs/source-capability/reproduction"

The producer refuses a missing or duplicate source/country cell, changed parent artifact, negative count, incomplete source audit or changed input bytes. It emits no new country assignment, ecological classification, cross-source comparison, or release decision.