FireAUS

FireAUS is Risk Frontiers' probabilistic catastrophe loss model for bushfire and grassfire across Australia. Bushfire losses rank among Australia's top five natural disasters by insurance and economic impact; the 2019/20 Black Summer fires caused more than $2.4 billion in insured losses. The model's hazard component is built on MODIS satellite data and machine learning. The MODIS burnt area product, combined with Risk Frontiers' 120 years of fire damage data, trains machine-learning models of how fire occurrence and spread relate to weather, topography and the built environment. Fire propagation is simulated with a stochastic cellular-automata model whose rules describe transitions between areas identified by satellite data as burned and their unburned neighbours. FireAUS simulates fires at 1 km resolution across a 50,000-year event catalogue and estimates damage at 100 m with a machine-learning model trained on loss data from past fires, giving address-level risk for every property on mainland Australia and Tasmania (100% of G-NAF addresses). It covers residential, commercial and industrial lines including business interruption, and reports losses at address, SA1, postcode and CRESTA level. Ignition rules driven by weather variables let it be adapted to future-climate analysis under a given emissions scenario. Insurers and reinsurers use its outputs for capital management and risk pricing, electricity distribution network service providers model their liability for causing bushfires, and governments use its address-level risk metrics for planning and policy. Risk Frontiers delivers its catastrophe loss models to clients through its Multi-Peril Workbench loss modelling software.