Roleystone–Kelmscott Fire 2011: Timeline Evidence and Community Preparedness

Bushfire case study · Western Australia · 6 February 2011

The Roleystone–Kelmscott fire destroyed homes on Perth’s urban fringe. Its documentary record is especially useful for understanding what a reconstructed fire timeline can establish—and where apparent precision exceeds the observations.

Evidence reviewed: 27 October 2026. This analysis compares the government inquiry, a departmental fire-spread reconstruction and published community research. It is a documentary assessment, not a new simulation or a property-level forensic investigation.

What happened in Roleystone and Kelmscott?

The Perth Hills inquiry records 71 homes destroyed and 39 damaged on 6 February 2011, with no lives lost. It reports the evacuation of 517 families, including some moved between multiple locations. These are the inquiry’s figures; families are not a count of individual evacuees. [1]

AIDR places the incident within a weekend that also included a separate fire around Red Hill and neighbouring communities. Its Roleystone narrative reports approximately 440 hectares burned. The two incidents should not be merged into a single fire footprint. [2]

Why a fire report time is not necessarily front arrival

Lachlan McCaw’s reconstruction, dated 23 March 2011, used communications logs, weather observations, mapping, aerial imagery and a field visit. It explicitly treats logged fire-presence times as indicative because a report might describe a spot fire ahead of the main front. [3, Appendix 1]

Tecnobosque analysis: drawing a continuous front through every reported location would erase this distinction. A reproducible timeline should retain the observation type, location, timestamp and uncertainty. A record of fire at a property cannot, by itself, tell us when every neighbouring property became exposed.

Why do area estimates differ?

The reconstruction describes about 475 hectares and explains the difficulty of defining an exact area where burned and unburned residential properties sit beside one another. It also distinguishes weather stations on the escarpment from those on the coastal plain. [3]

The approximate 440-hectare figure in AIDR and the reconstruction’s estimate remain separately attributed here. This comparison does not establish that one supersedes the other. Before calculating a damage rate per hectare, an analyst would need a consistent boundary and counting method. A denominator chosen from another assessment can make a precise-looking result misleading.

Risk awareness was not the same as personal preparedness

The Bushfire CRC’s October 2011 newsletter describes research involving interviews with residents of 425 households and 900 postal responses. It reports that many residents recognised bushfire as a possibility in their area without recognising a threat to their own household. [4]

This distinction helps explain why general awareness is an incomplete preparedness measure. However, these research samples are not a census of everyone affected, and this summary does not establish that attendance at an information session alone caused a particular household outcome.

What this case adds to the archive

The inquiry’s housing discussion drew on an interim house-loss assessment and noted that the final version was not available when it was written. [1, section 5.3] That timing matters: a later investigation may answer questions a contemporaneous review could only flag.

For readers comparing incidents, keep three records separate: the fire’s physical development, what people understood at the time, and what investigators established afterward. Their clocks and evidence are connected, but they are not interchangeable.

Sources and scope

1. Government of Western Australia: A Shared Responsibility, Perth Hills Bushfire February 2011 Review, introduction and section 5.3.
2. AIDR: Perth Hills bushfire, 2011, incident narrative.
3. DEC submission and Appendix 1: McCaw’s reconstruction of the Kelmscott–Roleystone bushfire, 23 March 2011; Appendix 1 begins on PDF page 39.
4. Bushfire CRC: Fire Update, October 2011, Perth Hills community research summary.