The 2019/20 Black Summer (September 2019-February 2020) bushfire disaster in southeast Australia was unprecedented in historical experience. The bushfires resulted in 33 direct fatalities, 417 excess deaths and exposed over 80% of the Australian population to dangerous air pollution.  The extreme bushfire activity over the Black Summer was the result of the compounding and alignment of multiple factors including, a severe multi-year drought, climate variability and climate change. 

However, disentangling the relative influence of natural variability from anthropogenic climate change over southeast Australia remains challenging due to the large and complex interannual climate variability in this region combined with a short (post 1950) observational record. Past periods of enhanced bushfire activity, drought and elevated temperatures from paleoclimate archives like tree rings, speleothems and sediment cores provide valuable long term context to fire activity. But the temporal and spatial spread of suitable paleoclimate archives for reconstructing fire weather extremes are limited across Australia. The majority of Australian rainfall and temperature proxy records with annual temporal resolution are from tree rings and coral cores. In Australia these only extend back ~500 years and are predominately found in Tasmania (tree rings) and northern Australia (coral cores).  Lower temporal resolution records, including speleothems and lake sediments suggest that centuries covered by Australian tree rings and coral cores are biased to wetter conditions, and therefore potentially underestimate past and present bushfire hazard. 

In this study we used the summer concentration of sea-salt aerosols preserved in the Law Dome ice core (Wilkes Land, East Antarctica) to reconstruct past bushfire weather in southeast Australia over the past 2000 years. Our previous research has found that small changes in the amount of  sea-salt in the snowfall during summer in the Law Dome region are closely linked to weather conditions in southeast Australia because the enormous weather systems in the Southern Ocean create a 'weather bridge’ linking the two continents. As these weather systems shift from year to year, they lift differing amounts of sea salt aerosol from the ocean surface, which are carried to Antarctica and deposited in snowfall. By drilling ice cores in places like Law Dome that receive high amounts of annual snowfall (an average of 1.5m per year), we can extract this preserved record over many centuries. 

We discovered a relationship between the ice core record and the Forest Fire Danger Index (indicator of fire weather) over the region of Australia most impacted by the Black Summer bushfires. We then used a synoptic typing dataset (a method of grouping similar weather patterns together over time) to uncover the direct weather-scale connection between sea salt in the ice core and Australian fire weather. This weather scale connection give us confidence in the types of synoptic-scale weather conditions that we have confidence in the remote proxy representing in the reconstruction, while also acknowledging the conditions that we likely miss. 

Many of Australia’s bushfire disasters have been linked to intense summer cold fronts, including Ash Wednesday (1983), the Canberra bushfires (2003), Black Saturday (2009) and the recent Black Summer. Using the synoptic typing dataset and the ice core record, we found the same weather conditions are behind periods of lower levels of sea salt in the ice and increased fire weather in southeast Australia. These conditions are caused by the strong westerly winds circling Antarctica moving north towards Australia, bringing summer cold fronts. This is referred to as the negative phase of the Southern Annular Mode. 

 In the last 2000 years, our reconstruction suggest that fire weather was the same or even worse than the 2019-2020 Black Summer, in the summers of 485, 683, 709, 760, 862, 885 and 1108 CE.

Climate change is increasing the likelihood of extreme fire weather, but our research shows that natural climate variability can also produce fire weather that is equivalent to (and worse than) that experienced during the Black Summer. This highlights the need for bushfire risk estimation and mitigation that considers the full range of plausible natural variability in Australia’s fire weather as well as the impacts of anthropogenic climate change.

One way to do this is via a method called  "climate storylines" that use multiple lines of evidence (e.g. observations, paleoclimate, climate model projections, traditional knowledge, expert opinion etc.) to develop scenarios describing the plausible impacts of climate variability and change that are in turn used to inform climate risk assessments.   

This Behind the Paper post is co-authored by Danielle Udy, Anthony Kiem, Tessa Vance, Neil Holbrook and Nerilie Abram. This research was supported by the Australian Research Council Centre of Excellence for Climate Extremes, Australian Centre for Excellence in Antarctic Science, the  Australian Antarctic Program Partnership and Institute for Marine and Antarctic Studies at the University of Tasmania, the University of Newcastle and Australian National University.