Permanent technical record · TB-WBB-CAN-2016-HORSE-RIVER
A versioned reconstruction of the Horse River fire and Fort McMurray evacuation that preserves source differences instead of hiding them.
Record summary. The Horse River wildfire began outside Fort McMurray on 1 May 2016 and entered the urban area within two days. By the evening of 3 May, mandatory evacuation covered the city. This record separates satellite observations, official reporting, technical interpretation and unresolved details; it makes no finding of legal responsibility.
Executive evidence summary
The incident was first reported on 1 May at approximately 2 hectares, about 7 km outside Fort McMurray.
90 · Very high · ReportedLandsat observed active fire southwest of downtown at 12:34 p.m. on 3 May; evacuation expanded across the city that evening.
93 · Very high · Observed / ReportedAuthoritative sources report more than 80,000, almost 88,000, or approximately 90,000 people, depending on scope and definition.
64 · Moderate · DisputedNo retrievable final official cause determination has been established in the source set processed for this version.
40 · Low · Unknown
Incident overview
| Field | Documented value | Evidence note |
|---|---|---|
| Incident | Horse River wildfire / Fort McMurray wildfire | Record ID TB-WBB-CAN-2016-HORSE-RIVER |
| Country | Canada | Province of Alberta |
| Administrative area | Regional Municipality of Wood Buffalo | Fort McMurray urban service area and surroundings |
| First reported | 1 May 2016 | Approximately 2 ha and about 7 km outside the community in the official post-incident summary |
| Citywide evacuation | By evening, 3 May 2016 | NASA summary based on contemporary official updates |
| Evacuation estimates | >80,000; almost 88,000; approximately 90,000 | Source definitions include residents, people, or residents and workers |
| Reported structural loss | 1,595 structures containing 2,579 dwelling units | NRC Canada table citing RMWB GIS / KPMG material |
| Reported insured loss | CAD 3.811 billion | 2017 Canadian dollars in the NRC table |
| Record status | Reference record under development | Reviewed 2 September 2026 |
Core chronology
Weather, fuels and exposure
The official review describes severe northern-Alberta fire conditions after a dry previous year, low snowpack and a hot, dry spring. Contemporary NASA reporting also records hot, dry conditions and province-wide fire restrictions. These factors establish the operating environment; they do not by themselves establish the ignition cause.
Fort McMurray was a high-consequence wildland–urban interface event. NRC Canada notes that WUI fires expose structures through flame, radiant heat and embers, and can require large evacuations while overwhelming available protection capacity. Tecnobosque treats that general mechanism as context and keeps event-specific attribution tied to the incident evidence.
Fire progression and mapped area
This record deliberately avoids one context-free burned-area number. NASA provides three distinct snapshots: almost 77 km² by 4 May; at least 2,410 km² across fires in the region by 12 May; and 5,229 km² consumed by the Fort McMurray fire at the 24 May MODIS observation. Different dates, geographic scopes and reporting purposes make these values non-interchangeable.
Data rule
An observation-time footprint is not a final perimeter. The Black Box dataset will add a final burned-area field only after an authoritative incident perimeter and its definition are versioned in the source register.
Evacuation and information systems
Government of Canada reporting describes the event as the largest mass evacuation in Alberta history and one of Canada’s largest. It also identifies minute-by-minute situational awareness as a central operational challenge.
- MASAS shared fire locations, road closures, hazards, safe routes, command posts and staging areas in real time.
- Users could view weather information including wind direction, a critical safety variable during the incident.
- The Operations Centre Interconnectivity Portal supported federal coordination around residents and critical infrastructure.
- BlueSky Canada smoke forecasts supported precautionary decisions for emergency managers and industrial facilities.
- Digital volunteers and social-media monitoring helped separate actionable information from rumour and speculation.
Why the evacuation totals differ
The reviewed authoritative sources use three defensible figures: more than 80,000 residents, almost 88,000 people, and approximately 90,000 residents and workers. These are not averaged into a synthetic total. The wording shows that population scope and estimation method differ, so the claim is marked Disputed at the exact-number level while the mass-evacuation finding remains strongly supported.
Impacts and recovery evidence
NRC Canada reports 1,595 structures containing 2,579 dwelling units lost.
88 · Very high · ReportedNRC Canada reports CAD 3.811 billion in 2017 dollars, preserving the currency-year basis.
86 · Very high · ReportedAlberta and RMWB monitored air, ash and soil, with a final quality-controlled technical synthesis published in 2018.
87 · Very high · ReportedThe final Alberta summary found no further wildfire-related air or soil monitoring or remediation was required.
87 · Very high · ReportedSystems analysis
The official Alberta/MNP review says detection performed within expectations and the initial attack was immediate, while also identifying improvements in how strong fire-weather science should feed operations and planning. This distinction matters: a rapid initial response can coexist with an event that exceeds assumptions, resources or decision windows.
The record therefore avoids a single-cause narrative. The documented outcome emerged from the interaction of extreme seasonal conditions, rapid WUI transition, evacuation at urban scale, changing road and hazard information, smoke exposure, critical-infrastructure risk and multi-agency coordination.
Origin and cause
The NASA article published during the event states that the cause was still under investigation as of 4 May 2016. This record has not yet located a retrievable final official determination in the authoritative material processed for version 1.0. The ignition cause is therefore labelled Unknown here. That label describes the evidence available to this record; it is not a claim that no later investigation exists.
Legal boundary
An ignition-cause finding, an operational review and a legal allocation of responsibility are different products. Tecnobosque documents evidence and uncertainty and makes no determination of negligence, civil liability or criminal responsibility.
TB-ECI claim register
Scores are provisional manual applications of Wildfire Black Box methodology v1.0 and TB-ECI beta 0.1. They measure evidence strength and auditability—not the probability that a claim is true.
| ID | Material claim | State | TB-ECI | Band |
|---|---|---|---|---|
| C001 | First reported May 1 at approximately 2 ha, about 7 km outside Fort McMurray | Reported | 90/100 | Very high |
| C002 | Dry prior year, low snowpack and hot, dry spring created severe conditions | Reported / Interpreted | 88/100 | Very high |
| C003 | Landsat observation southwest of downtown at 12:34 p.m. on May 3 | Observed | 95/100 | Very high |
| C004 | Mandatory evacuation covered Fort McMurray by May 3 evening | Reported | 91/100 | Very high |
| C005 | Exact evacuation count: >80,000, almost 88,000, or approximately 90,000 | Disputed | 64/100 | Moderate |
| C006 | 5,229 km² consumed by the May 24 MODIS observation, not a final-area value | Observed / Reported | 90/100 | Very high |
| C007 | 1,595 structures containing 2,579 dwelling units lost | Reported | 88/100 | Very high |
| C008 | CAD 3.811 billion insured loss in 2017 dollars | Reported | 86/100 | Very high |
| C009 | Shared maps, safe-route data, weather and smoke forecasting supported response | Reported | 84/100 | High |
| C010 | Final official ignition cause in this source set | Unknown | 40/100 | Low |
| C011 | Final environmental review found no further wildfire-related air or soil work required | Reported | 87/100 | Very high |
Conflicts, unknowns and open modules
- Evacuation total: source-defined estimates are preserved rather than blended.
- Final burned area: not yet entered because dated satellite and regional values are not interchangeable with a final incident perimeter.
- Ignition cause: no retrievable final official determination has yet been processed for this version.
- Human impact: fatalities, injuries and medical outcomes require a dedicated definition-controlled claim module before scoring.
- Exact progression: this beta record does not yet reproduce an hour-by-hour geospatial perimeter.
- Indigenous and industrial impacts: threatened First Nations communities, oil-sands facilities and worker movements require separate sourced modules.
- Recovery costs: insured loss is not the same as total economic, public-sector or social cost.
Primary source register
- Government of Alberta — Wildfire maps, data and historical reviews
- Government of Alberta / MNP — Review of the 2016 Horse River Wildfire
- NASA Earth Observatory — Wildfire in Fort McMurray
- NASA Earth Observatory — Fort McMurray Burn Scar
- NASA Earth Observatory — Fort McMurray Wildfire Still Burns
- National Research Council Canada — Protecting homes from wildland fire
- Defence Research and Development Canada — Emergency-management support during the wildfire
- Government of Alberta — Environmental monitoring in Fort McMurray
- Government of Alberta — Hourly ambient air-quality data
- Regional Municipality of Wood Buffalo — Post-incident assessment summary
Record governance
Version 1.0 · 2 September 2026. Initial Canadian reference record. Added dated satellite observations, source-defined evacuation estimates, structure and insured-loss values, response-system evidence, environmental monitoring and provisional TB-ECI scoring.
Next review trigger: publication or recovery of a final official ignition-cause determination, a definition-controlled final incident perimeter, or a source that materially revises evacuation, structural-loss or environmental findings.