As noted on the Ignition page, there are many different human and non-human sources of ignition. The measure we use here is an ignition likelihood model that includes the influence of weather (a fire danger index). Therefore, it can be calculated directly from climate models. It is important to note that potential future changes in other drivers of ignition, such as population density and vegetation, are not represented.
The data shown here is based on an ignition likelihood model that includes multiple inputs. The only input for which we have assessed climate change impacts is the Forest Fire Danger Index (FFDI), based on projections from the ESCI climate change project. No changes in population, vegetation or other model inputs were represented. It represents the multi-model mean i.e. there are projections from multiple climate models and this shows the average across them all.
See what the modelling says to understand more about these projections.
To learn more about ignition likelihood models read:
Clarke, H. et al (2019):
Developing and testing models of the drivers of anthropogenic and lightning-caused wildfire ignitions in south-eastern Australia.
Journal of Environmental Management.
This data shows us the projected change in average ignition likelihood across Victoria from 1980 to 2100 under an intermediate emissions scenario, and a very high emissions scenario. The ignition probability shown here is a complex function of fire weather, house density, distance to roads, mean annual rainfall and vegetation type. We use a fire year rather than a calendar year e.g. the 2020 fire year runs from July 2019 to June 2020.
Hover over the map and graphs to get more info.