DOI-Lands boundary mask feed for PROMETHEUS (situational awareness for DOI Emergency Management / EHSD). Keyless federal sources → slim GeoJSON + JSON on the GitHub CDN, weekly cron.
Phase 1 covers NPS only. BLM / FWS / BIA / BOR / BOEM added in later phases.
data/lands-nps.geojson— 422 NPS unit polygons, Douglas-Peucker simplified (0.001° ≈ 111m). Properties:bureau,unit_code,unit_name,unit_type,region,state,acres(authoritative unsimplified),acres_simplified(derived).data/lands-index.json— fast-filter index: national totals + per-unit{unit_code, unit_name, bbox, centroid, acres, unit_type, region, state}. No geometry — for panels that only need the roll-up + a click-to-drilldown.data/lands-nps-benchmark.json— acreage-error benchmark under tolerance sweep (transparency: operators can see what precision they're getting).
prometheus-hazard-lands-data— intersects hazard polygons (fire/weather/quake) against this mask and publishes the DOI-Lands-scoped hazard feeds.prometheus-sa(the app) — reads only the intersected hazard feeds; never touches this raw mask directly (would be a 20MB payload).
- NPS Land Resources Division ArcGIS FeatureServer:
services1.arcgis.com/fBc8EJBxQRMcHlei/.../National_Park_Service_Boundaries/FeatureServer/0 - Keyless,
f=geojson, paginated at 2000/request. - Attribution: National Park Service — Land Resources Division.
- 422 features (matches published NPS unit count).
- 85,191,005 total acres (matches published NPS ~85M figure).
- 0.013% national roll-up error at the locked simplification tolerance.
- 20 MB raw / ~4 MB gzipped over the CDN. Refresh weekly.