**Note** from Bead: Fire Progression · [canonical source](https://redfish.acequia.io/guerin/.agents/182bd1a7-a376-4dd0-b445-578adc6635b9/2026-06-16/notes/01-sandy-fire-worked-example.md) · session 2026-06-16 · discussion: Talk: Fire Progression
## Reported ignition origin - **Location:** 2600 block of **Rudolph Drive, Simi Valley, CA** - **Coordinates:** `34.25235, -118.746484` (lat, lon) - **Time:** ~10:17 a.m. PDT, Mon **2026-05-18** = **17:17 Z** - **Cause:** a tractor struck a rock; the spark ignited dry brush. - **Source:** njsr.com.ng announcement #187 (see `names.json` → `sandy-narrative-source`; full text imported to `../artifacts/sandy-fire-narrative.md`). This is the **origin point** for the TOA algorithm (`notes/00`, origin rule #1). It supersedes the synthesized centroid-interior origin the existing raster used.
## Mounted data — Sandy Fire STAC (`cbf8b027`) `sandy-stac` → `…/cbf8b027-…/2026-06-11/artifacts/stac/2026SandyFire/` **Time-stamped heat perimeters** (`perimeters/`, AEVEX, from N42Z + N461DF sorties): | time (UTC) | acres | item | |-------------------------|-------|------| | 2026-05-18T18:02:42.369Z | 183 | `CA-VNC-SANDY-42Z_PERIM_180242369_183acres` | | 2026-05-18T22:09:26.956Z | 836 | `CA-VNC-SANDY_PERIM_220926956_836acres` | | 2026-05-19T01:56:00.493Z | 1364 | `CA-VNC-SANDY_PERIM_015600493_1364acres` | | 2026-05-19T11:22:49.873Z | 1385 | `CA-VNC-SANDY_PERIM_112249873_1385acres` | | 2026-05-19T23:20:00Z | — | `CA-VNC-SANDY_20260519_232000Z_…N461DF_HeatPerimeter` | | 2026-05-20T23:13:00Z | — | `CA-VNC-SANDY_20260520_231300Z_…N461DF_HeatPerimeter` | **Existing TOA item** (`sandy-toa-item`, `progression/items/toa-2026_CAVNC_042326_sandy.json`): - `toa:encoding: "simtable-flamesim-rgb"`, `toa:source: "levelset interpolation of NIFC perimeter KMLs (progressionMaker)"` — i.e. the exact algorithm in `notes/00`, already run. - 8 timesteps, `start_datetime 2026-05-18T23:49Z` → `end_datetime 2026-05-23T05:29Z`; `acres` 0 → 2179. - RGB PNG + worldfile + metadata hosted at `simtable.com/apps/fireProgression/output2026/`.
## The t0 discrepancy (flagged for re-anchoring) Three different "starts" are in play: 1. **Reported ignition:** 2026-05-18 **17:17 Z** (the true origin time). 2. **First observed heat perimeter:** **18:02 Z**, 183 acres (≈ "180 acres within 30 min" in the narrative — consistent with ignition at 17:17 Z). 3. **Existing TOA `start_datetime`:** **23:49 Z** — its 8 NIFC perimeters begin ~5.5 h *after* ignition and ~5.75 h after the first AEVEX heat perimeter. So the existing raster's minute-0 is **not** the ignition point/time. The reported origin (`34.25235, -118.746484`) also lies **just north of** the existing raster bbox top (`34.25173`) — the front swept south/downhill from a ridge ignition. **Open action:** recompute the TOA anchored on the true ignition origin + 17:17 Z, optionally folding the early AEVEX heat perimeters (18:02 / 22:09 Z) ahead of the NIFC set so the gradient reaches back to ignition.