|
House loss rate in severe bushfire attacks
Denis O'Bryan 2011 This is an abridged version of another paper Introduction Fire agencies have a preference to remove people from the path of severe bushfire attack - whether by compulsory evacuations or encouraging early departure. Eg, fire agencies ask people to prepare a bushfire survival plan, which is essentially a voluntary evacuation plan. Whatever the terminology, if people leave, their house is vacant during the bushfire attack. The author has concerns that historical facts have not been properly considered in the campaigns by fire agencies to remove people from their houses in the path of severe bushfire attack. For example, Wilson and Ferguson (1983) found strong evidence that vacant houses have a high loss rate (approx 60%), whereas occupied houses had a very high survival rate (80% +). This paper documents some recent statistics about the influence of occupancy on house survival rates. The author assessed the bushfire attacks in all areas as essentially ember attacks, noting that there was no moving flame within the residential areas. The flame near houses was stationary within dry garden beds or flammable material, and remained stationary because the fuel bed on the ground was discontinuous due to the road network and pathways. Hypothesis The paper tests the following hypothesis: During severe bushfire attack in residential areas (suburban or semi urban) - The chance of house survival is very high if the house is occupied - The chance of house destruction falls as the house occupation rate increases. Method Evidence is collected from the aftermath of three severe bushfire attacks, - Ash Wednesday (Macedon area) 1983, survey of 450 properties - Canberra fires 2003, Duffy area, survey of 49 properties - Black Saturday, 2009, survey of 1100 properties All three areas that were hit by a severe bushfire at the peak of its severity All areas had substantial urban residential areas and featured substantial house losses. All areas were examined by the author, two during the bushfire attack and immediately after, and two soon after the bushfire attack. This gives consistency of assessment Surveys of survivors were conducted in all areas. All days were similar - very hot (40*C or more), very dry (RH 10% or less) and very strong winds (40 kph or more), and would have been classified as CATASTROPHIC. Results Ash Wednesday: - 14% of all houses were occupied, 51% of all houses were destroyed Canberra fires: - 47% of all houses were occupied, 33% of all houses were destroyed Black Saturday: - 46% of all houses were occupied, 35% of all houses were destroyed These findings were compared to the Marysville situation. Marysville was almost totally evacuated and the town was almost totally destroyed. Estimated figures are as follows: Marysville: - 8% of all houses were occupied, 96% of all houses were destroyed Discussion The data paints a consistent picture of the inverse correlation between occupation rate and house destruction rate. This evidence supports the hypothesis. There is a distinction between occupying a property and occupying with intent to self defend. The Black Saturday survey found that 42% of people stayed and self defended, whereas 4% stayed for shelter or other reasons. This is confirmed by other data that found that 83% of people who stayed were determined to self defend. This leads to two suggestions of how to increase the survival rate of occupied houses, firstly, attitude or physical ability to self defend and secondly, knowledge and skill about how to self defend properly. Increasing the survival rate of occupied houses A The Ash Wednesday survey found that 65 houses were occupied during the bushfire attack but 12 (18%) of these were destroyed, which means a survival rate of occupied houses 82%. Six of these were occupied by people unable to self defend, eg, due to physical incapacity. When they reanalysed their data, they found if houses were occupied by able bodied residents, the survival rate was 90%. B The Victorian Bushfire Royal Commission (VBRC) found that many properties were occupied during the bushfire attack, but suffered deaths. Figures for the Kilmore East and Murrindindi fires show that of 63 properties where deaths occurred, investigators found 38 (60%) were well equipped and prepared in accordance with fire agency literature and advice. These questions arise - was the advice incorrect or did the residents have operator error? - but the VBRC did not explore them. The suggestion is that if residents know what they are doing and do it properly, self defence should raise the survival rate of occupied houses to almost 100%. Further analysis leads to suggestions of how to decrease the destruction rate of vacant houses. Decreasing the destruction rate of vacant houses A … The Ash Wednesday survey found that 385 houses were vacant at the time of the bushfire attack, and 217 (56%) of these were destroyed. It also found that another 54 were ignited by the ember storm, but were extinguished by residents who remained on site. For example, the author is aware of one man who stayed in a residential part of Macedon and extinguished his own house plus four vacant ones before being overcome by exhaustion. He helplessly watched as other vacant houses burnt down. If sufficient residents had not stayed, the house loss would have been 217 + 54 = 271 (70%). B … The vacancy rate of the Ash Wednesday study area was significantly high at 86% compared but the destruction rate of all houses was 51%. The vacancy rate (estimated) at Marysville was slightly higher at 92%, yet the destruction rate was some 96% of all houses within the township. This suggests there may be a vacancy rate above which the destruction rate rises rapidly. This paper speculates that the threshold vacancy rate may be 90%. The reason may be as simple as there are not enough residents within the attack area to save significant numbers of other houses. But there may be other factors such as poor property preparation of other houses, or distance to adjacent houses may be too great for the resident to leave their own place with confidence to extinguish their spot fires. Whether there is a threshold rate or not, the above data provides support for this hypothesis, revised as follows: During severe bushfire attack in residential areas (suburban or semi urban) - The chance of house survival is very high if the house is occupied, particularly by able bodied people, determined and knowledgeable self defenders. - The probability of house destruction falls as the house occupation rate increases. - The probability of house survival rises as the house vacancy rate increases. When government preference to evacuate areas in the path of severe bushfire attack is assessed against this hypothesis, the expected house loss rate can be expected to rise as more properties are evacuated. Conclusion The data paints a consistent picture of the inverse correlation between occupation rate and house destruction rate, which supports the hypothesis. Understanding the hypothesis leads to suggestions about how to decrease the destruction rate of vacant houses and increase the survival rate of occupied houses. The consequence of the government's policy preference of evacuation can be clearly seen. | |
 ....Return to Home PageCopyright Red Eagle |