Wildfire evidence study · British Columbia, Canada · 2003
Documentary review · Sources checked 1 November 2026 · Historical case, not live incident information.
The Okanagan Mountain Park fire offers a compact example of a common archive error: assigning every fact in an image caption to the moment the satellite passed overhead. This review separates image acquisition, dated damage reporting and later municipal recovery accounts.
What the sources record
NASA Earth Observatory published an IKONOS image feature on 29 August 2003. The image was acquired on 21 August. Its caption reports 247 homes destroyed or damaged as of 26 August and an area exceeding 200 square kilometres as of 27 August. The caption places the northern damage area outside the displayed image. [1]
Kelowna’s municipal retrospective reports a lightning start on 16 August, more than 25,000 hectares affected, more than 27,000 evacuees and 239 homes destroyed. It also describes the later recovery organization and assistance fund. [2]
Four dates cannot be compressed into one snapshot
| Date in 2003 | Meaning |
|---|---|
| 21 August | Satellite acquisition |
| 26 August | Damage figure’s stated reference date |
| 27 August | Area figure’s stated reference date |
| 29 August | Feature publication |
The image cannot independently verify later losses merely because they appear in its caption. An analyst should store those as attributed documentary claims, while treating the image as an observation with its own footprint and acquisition time. The same rule applies when an article is updated long after its original publication.
Why 247 and 239 are not directly contradictory
Destroyed-or-damaged homes and destroyed homes are different categories. Subtracting 239 from 247 would give eight arithmetically, but would not establish that exactly eight homes were damaged. The records also differ in timing and potentially in geographic coverage. A defensible reconciliation needs matching property lists or a source that explicitly explains the categories.
Likewise, a dated evacuation count is not necessarily a count of everyone displaced during the whole incident. It may represent orders active at that point, residents who actually left or a cumulative total. Without a definition, adding counts from different announcements risks counting the same people twice.
Recovery finance is a separate measure
The city states that its recovery society distributed $1,402,000 and ended its operations in April 2004. [2] That is a funding and organizational milestone. It cannot establish that all homes had been rebuilt, all losses compensated or all households recovered by that date.
A more complete recovery assessment would connect assistance categories to beneficiaries and dates, then separately track housing completion and service restoration. The administrative closure of one organization should not become a substitute for those outcomes.
Limits and reuse
No new satellite classification, property inventory or evacuation reconstruction was produced. This is a documentary comparison of two public sources. The commercial imagery credited in the NASA feature is linked rather than republished; its presence on a NASA page should not be taken as blanket permission to reuse it.
Primary sources
- NASA Earth Observatory, Close-up of Okanagan Fire, 29 August 2003
- City of Kelowna, historical fire and recovery account
Continue the investigation
Evidence-review methodology · Why wildfire acreage changes · Canada case studies
Cite this record
Tecnobosque. Okanagan Mountain Park Fire 2003: Satellite Dates, Housing Loss and Recovery Records. Documentary review, 1 November 2026. Permanent record. Cite the original sources for their findings.