RAÄ Exports

RAÄ exports provide Sweden-specific archaeology context from Riksantikvarieämbetet/Fornsök Open Data. The public map uses an aggregated density layer so national context remains legible without rendering hundreds of thousands of point markers. This is a Sweden-scoped source and product surface; it supplies no equivalent registry coverage for the other Nordic countries.

Current Governed Surface

data/raa/normalized/sweden_archaeology_layer.json records:

Measure	Value
all published sites represented by the source	761,917
records classified Fornlämning	318,265
Fornlämning or possible ancient remains	416,913
one-degree density features rendered	106

The public density GeoJSON is data/raa/normalized/sweden_archaeology_density.geojson. Its cells summarize the Fornlämning selection for national-scale rendering. A density feature is therefore an aggregate display object, not an archaeological site and not a sample.

flowchart LR
    OpenData["published RAÄ site records"] --> Classify["heritage classification"]
    Classify --> Select["Fornlämning selection"]
    Select --> Aggregate["one-degree cell counts"]
    Aggregate --> Context["Sweden archaeology density layer"]
    Context --> Product["map and report context"]

Three Counts, Three Meanings

The 761,917 total describes the broad published source population represented by the layer metadata. The 318,265 count describes the Fornlämning subset used for density rendering. The 106 count describes grid cells that carry aggregated map features. Comparing those numbers as though they shared an observation unit would confuse source records, selected records, and rendered geometry.

Worked Aggregate: The Densest Published Cell

The largest checked-in density feature covers the one-degree cell from longitude 17 to 18 and latitude 59 to 60. Its published count is 27,450 selected Fornlämning records.

That feature establishes one aggregate statement: 27,450 source records in the selected classification were assigned to this grid cell under the current normalization. It does not identify 27,450 coordinates in the export, and the cell centroid or polygon must not be treated as the location of any individual record.

Reuse operation	Defensible result	Information lost or invented
preserve the cell polygon and count	comparable aggregate under the same grid and selection	none beyond source-level detail already abstracted by the layer
convert the cell to one point	coarse visualization only, if labelled as an aggregate	polygon extent and within-cell distribution
expand the count into repeated points	no defensible scientific result	synthetic coordinates and false independence
compare with another grid size	only after recomputing both populations	direct count comparability across unequal areas

The grid definition is therefore part of the observation. A density count without its cell geometry, selection class, and source population cannot be reconstructed or compared responsibly.

Supported Interpretation

The layer supports questions about the spatial density of published Swedish archaeology records under the declared classification and grid. It helps identify where admitted aDNA, pollen, or lake candidates sit relative to that registry context.

It does not establish:

• a sample-to-site relationship;
• contemporaneity between a registry record and another layer;
• historical abundance from modern registry density;
• equivalent archaeology coverage outside Sweden; or
• exact site chronology from a density cell.

The current normalized RAÄ layer carries no numeric temporal intervals. Spatial co-occurrence with a cell must therefore remain archaeology context, not a time-aligned event claim.

Reuse Contract

Carry the layer metadata, selection class, source population counts, one-degree grid definition, density GeoJSON, source identity, and non-numeric temporal posture together. Preserve cell counts as aggregates and do not expand a cell into synthetic site points.

Continue to RAÄ source guidance for source semantics, maps for role-aware spatial reading, and publication limits for comparison boundaries.