BAMOS Vol 30 No. 3 2017 | Page 20

20 BAMOS Sept 2017 Figure 6. Wind speed (m s -1 ) for January 4, 2013 at 0400hrs. A red square in the left panel highlights the Giblin River fire location. The right panel shows wind speed for the region within the red rectangle based on data from the Initial Analysis (at 1.5 km resolution) for Tasmania. Cobbler Road fire New South Wales The Cobbler Road grassfire during January 2013 in New South Wales (Figure 7) is one of the fires recently examined to assess fire spread simulators used in Australia. The New South Wales Rural Fire Service (2013) describes the details around the Cobbler Road fire under the title “Speed and fury”: “The Cobbler Road fire, which started under Extreme conditions, burnt quickly, travelling 35 kms and covering 14,000 hectares within six hours. It caused significant damage to farming country including extensive livestock losses. Much of the activity took place overnight on 8–9 January 2013. The strong westerly wind did not ease overnight and nearly 150 firefighters worked intensively to protect properties in the path of this remarkably fast moving grass fire.” Fire spread simulators are used to predict the surface spread of a fire, and can be valuable tools in operational decision making and planning. These simulators have been assessed by comparison with the observed burnt areas (Faggian et al 2017). Preliminary studies incorporating the reanalysis data (Chris Bridge, Bureau of Meteorology, pers. comm.), compared with data from the Bureau of Meteorology Australian Digital Forecast Database, show that for the small set of case studies examined, incorporating the reanalysis data generally led to better overall results using the same evaluation metrics as used in the fire project. This can vary somewhat depending on the type of simulator used and the circumstances of the fire events. This is an exciting potential area of study for the reanalysis data. Making use of the reanalysis data The reanalysis data set will be especially valuable in providing climatologies of weather extremes across the nation. Due to the high spatial and temporal resolution of the reanalysis data, mapping of the extremes of variables, such as the 95th percentile of daily maximum wind speed (Figure 8), is made possible. This is information required for example in infrastructure design and to assess the intermittency and covariability of resources in renewable power generation. Understanding of short-lived or fast-developing atmospheric phenomena such as thunderstorms will be improved with the detailed temporal and spatial information on atmospheric conditions. Processing of a reanalysis dataset can provide an indication of thunderstorm environments, and assist in studies of those environments conducive to massive lightning occurrence. Severe thunderstorm environments are reasonably well-understood, so a comprehensive climatology of these environments can be generated from a reanalysis dataset. The vast amount of data currently being generated will permit studies leading to an unprecedented understanding of local weather, particularly in areas that are currently poorly served by weather observations such as the Tasmanian Wilderness World Heritage Area that suffered significant bushfires in January 2016. Bushfires are one of the most costly natural disasters in Australia in terms of loss of life and damage to property and the environment (Crompton, 2011). BARRA will enable studies of the passage of cold fronts over southern Australia during summertime, identifying characteristics of such fronts that often lead to extreme fire or fire weather activity, deriving variables such as the Forest Fire Danger Index (FFDI, Figure 9) and permitting the development of “climatologies” of fuel load for grass fire danger studies. This will allow for greater resilience in the future, especially under future climate projections that predict an increase in the frequency of severe fire weather (FFDI over 50, CSIRO & Australian Bureau of Meteorology 2015).