Downloadable data files shared across apps in this repo, rather than duplicated per-app.
| file | used by | contents |
|---|---|---|
fire-santafe-sample.geojson |
Progression | a saved run of Fire’s made-up cellular spread model — 120 MultiLineString perimeters, keyed by tick, from fire.save() |
palisades-2025-progression.geojson |
Progression | CAL FIRE’s official progression mapping for the January 2025 Palisades Fire (LA) — 11 MultiPolygon snapshots, keyed by PROG_DATETIME, 770 → 23,717 acres over 5 days |
camp-fire-2018-progression.geojson |
Progression | NIFC’s historic GeoMAC archive for the November 2018 Camp Fire (CA) — 28 MultiPolygon snapshots, keyed by perimeterdatetime, 54,586 → 153,336 acres |
las-conchas-2011-progression.geojson |
Progression | NIFC’s historic GeoMAC archive for the June–July 2011 Las Conchas Fire (Jemez Mountains, NM) — 29 MultiPolygon snapshots, keyed by perimeterdatetime, 43,641 → 156,656 acres |
dog-head-2016-progression.geojson |
Progression | NIFC’s historic GeoMAC archive for the June 2016 Dog Head Fire (NM) — 18 MultiPolygon snapshots, keyed by perimeterdatetime, 682 → 17,911 acres |
powder-fox-tartar-2026-progression.geojson |
Progression | SimTable’s time-of-arrival raster for the Jul–Aug 2026 Powder/Fox/Tartar Fire Complex (OR), converted to polygons — 58 MultiPolygon frames, keyed by utcMs, one per real report time |
The real ones (everything but the Fire sample) are downloaded and cleaned
up by fetch-progression.js:
deno -A fetch-progression.js --list # known dataset names
deno -A fetch-progression.js camp-fire-2018
deno -A fetch-progression.js --all # refetch everything
Two source shapes feed it. Most are an ArcGIS FeatureServer of dated
perimeter vectors (NIFC’s historic GeoMAC archive, or a one-off agency
progression service) — the script handles the two things a plain REST
query leaves for you to clean up by hand: oversized geometry (the
server-side maxAllowableOffset param — cut Camp Fire from 24.6 MB to
1.5 MB) and incidents that report several disjoint polygon pieces per date
instead of one (dissolveByDate, tested against Hermits Peak/Calf Canyon —
see the script’s SOURCES entry for why that one’s output isn’t shipped
here despite the mechanics working). The other shape, kind: "toa-raster",
is a single PNG where each pixel’s color packs a seconds-since-ignition
value (SimTable’s format — see Powder/Fox/Tartar’s SOURCES entry for the
decoding and downsampling); the script thresholds it at each of the
incident’s real report times to produce the same per-frame MultiPolygon
shape as everything else, so the Progression app never has to know a
raster was involved.
See Progression/README.md for how each dataset is shaped and where it was found.