Risk Frontiers deployed a damage survey team in early December which helped collect data for our natural catastrophe modelling. The team travelled to bushfire-affected communities in northern NSW to make observations and report on impacted areas. The two fires concerned behaved differently and were influenced by weather conditions and terrain. The role of wind conditions and embers in the Busbys Flat Fire were significant factors in the location and distribution of destroyed buildings and their proximity to bushland. Industries/Infrastructure affected: Sawmill, pine plantations, railway.
Long Gully Fire (Drake Fire)
A survey of Long Gully Road from the Bruxner Highway confirmed the area had dense vegetation, which was severely burnt during the Long Gully Fire (LGF). The southern end of Long Gully Road (close to the fire ignition point) is remote and steep, which would likely have limited initial fire control efforts. Losses from early stages (September) of the LGF appear to be limited to private holdings and farms, with no evidence of any commercial or industrial enterprises being impacted. The team located 15 buildings (and one vehicle) impacted by the fire along Long Gully Road – most were totally destroyed or damaged enough to require demolition. The buildings were a combination of residences (of varying size and construction) and out-buildings. Destroyed properties varied in distance from the road from 10-15 metres to 1.9 km and were on either side of the road. Visible debris revealed no consistency to the construction materials of destroyed buildings, with corrugated sheeting, brickwork, timber and fibro (or similar) sheeting evident and several examples of water tanks remaining. There was minimal variation in distance to adjacent bushland. Most destroyed structures were no more than 20 metres from significant bush. At several locations, only a concrete slab remained after debris was removed. Over 7.5 weeks, the LGF burnt more than 74,000 hectares of bush and farmland. Conditions on Tuesday 8 October caused the LGF to reintensify and join with other local fires (including the Busbys Flat Fire). Together, the new combined fire was responsible for extensive damage to Rappville and two fatalities.
The Busbys Flat Fire (BFF), Rappville and wider area
An act of arson on Friday 4th October in the Busbys Flat area is the suspected cause of the Busbys Flat Fire (BFF). High temperatures and ferocious winds on Tuesday 8th October caused the BFF to intensify and merge with other major fires burning in the area, including the still active Long Gully Fire. This combined fire destroyed an estimated 30 homes and commercial properties as it travelled from its ignition point east toward Rappville (population 170). A noteworthy aspect of this fire, as reported by witnesses and volunteers, was the quantity of embers it generated, which were then carried by strong wind over large distances. Within the town, 16 buildings, mainly dwellings, were burnt. Where debris had not been cleared, the most common construction materials were evidently timber and fibro, with corrugated sheeting and brickwork. At least eight of the 16 destroyed building sites had ‘‘asbestos” warning signs posted and were secured by fences. There was an apparently random distribution of destroyed buildings, and the lack of substantial bushland within the village demonstrated how embers in high winds can propagate fires over long distances. The fire that impacted Rappville and surrounding areas was responsible for significant commercial losses, consisting of 200 claims costing an estimated $25 million. Significant infrastructure damage included a large sawmill located on Old Tenterfield Road, distorted steel tracks and destroyed hardwood sleepers of the Rappville Rail Bridge, and extensive fire damage to numerous large pine plantations.
Statistical dependence of bushfire risk on distance to bush and the influence of ember attack
Proximity to bushland is a significant factor in determining a building’s vulnerability. Figure 1 depicts bushfire damage based on aggregated data from recent major bushfires and shows the percentile of destroyed buildings in relation to nearby bushland (i.e: an ignition source). However, the Rappville (2019) and Duffy (2003) examples suggest that in cases where ember attack is a major element of a fire’s behaviour, this dependence may be less important. At Rappville ~55% destroyed structures occurred between 9 – 100 metres of bushland with the remaining ~45% occurring outside 100 metres. These distances were significantly greater at Duffy. Notably, weather conditions prior to both fires were starkly similar. At Duffy, the Bushfire CRC reported that “unusual severity of the fire was generated by extreme weather conditions” (a combination of particularly strong wind, temperatures near 40°C and drought conditions) and that “most houses were ignited by either ember attack or house-to-house ignition.” 1
1Bushfire CRC Report – INVESTIGATION OF BUSHFIRE ATTACK MECHANISMS RESULTING IN HOUSE LOSS IN THE ACT BUSHFIRE 2003 (2005). http://www.bushfirecrc.com/sites/default/files/downloads/act_bushfire_crc_report.pdf
About the author/s
Salomé Hussein
Salomé is our expert in machine vision, radar analysis, and robotic process automation (RPA). She holds a PhD in Physics and specialises in modelling hail and agriculture.
Jacob Evans
Jacob has been involved in the development of our FloodAUS, CyclAUS and QuakeAUS Cat-Loss models. He specialises in data science and mathematics.
Jacob received his PhD in Condensed Matter Physics from Macquarie University. Jacob’s interests include data science, numerical modelling, physics and mathematics.
Joining Risk Frontiers in 2017, Jacob has worked across a range of projects from model development and climate risk management to resilience and portfolio modelling.
Notably, Jacob has employed techniques such as machine learning and statistical analysis to understand the vulnerability of risk across Risk Frontiers catastrophe suite, as well as applying machine learning algorithms for flood, cyclone and earthquake models. He also works on hazard and loss modelling.