Russell Buzby

For most of the last twenty years the conversation between climate scientists and fire agencies has been slightly awkward. Researchers would present evidence that climate change was making fire weather more intense. Agencies would acknowledge the trend, and then, reasonably enough, point out that attribution at the level of any specific event or season remained uncertain. The uncertainty was genuine and the caution was appropriate. It also created a policy opening that governments found useful: climate change might be making things worse, but it was hard to say by how much, and the cost of acting on the assumption was real.

That conversation is now effectively over, at least as a question of science. A paper published in Science Advances in mid-March 2026 by Turco and colleagues analysed global fire weather days from 1980 to 2023 using climate model ensembles and fingerprint detection methods. Their finding is as sharp as attribution science tends to get: the anthropogenic signal in the rising frequency of extreme fire weather is detectable above natural variability at 99 per cent confidence (Turco et al., 2026). Extreme fire weather days have risen by roughly 65 per cent globally since 1980, and the cause is greenhouse gas emissions rather than natural cycles or land-use change alone.

A second Science Advances paper by Yin and colleagues, published in February 2026, looked at the same question from a different angle and reached a more operationally alarming conclusion. Between 1979 and 2024, synchronous extreme fire weather increased by more than twofold in most regions (Yin et al., 2026), with over half of that rise attributable to human-induced climate forcing. It is no longer unusual for southern Europe and western North America to experience extreme fire weather at the same time as south-eastern Australia. The climatic conditions that used to rotate around the globe in a rough seasonal sequence are now overlapping with increasing frequency, and the overlap is not random noise.

Taken together, these two findings close the attribution debate for practical purposes. The first tells you the trend is real and human-caused. The second tells you the trend has an operational signature: fire seasons are converging in time as well as intensifying in magnitude. Neither of these is a surprise to anyone who has been on the ground during the last decade. Both are now cleanly quantified in the peer-reviewed literature at a level of confidence that makes them difficult to brush aside in policy conversations.

For anyone working in emergency management, the practical implication of that shift is not subtle. The scientific question has stopped being the bottleneck. The bottleneck is now resourcing.

That change of bottleneck is worth taking seriously, because it is harder to solve than the debate it replaces. While climate attribution was uncertain, the policy argument could be framed as a knowledge problem: we should invest more in science, we should run more models, we should commission more studies. That framing was comfortable because it deferred the expensive decisions. A knowledge problem can be solved by universities and CSIRO on grant timelines. A resourcing problem can only be solved by treasury, and treasury departments do not enjoy being presented with new compounding liabilities.

Victoria’s 2025 to 2026 bushfire season gave a local illustration of what that bill looks like when it comes due. It was the worst season the state had experienced since Black Summer. More than 400,000 hectares burned. A State of Disaster was declared across 18 local government areas. Temperatures reached 48.9 degrees at several Bureau of Meteorology stations. The World Weather Attribution analysis of the January heatwave found that human-induced climate change had made the driving conditions around five times more likely than they would have been in a cooler world. The CFA and MFB responded at a scale that stretched available crews and assets, and the state relied on mutual aid from other jurisdictions where the season allowed.

That last point is the one the Yin paper sharpens. Australian mutual aid arrangements and our access to international firefighting resources are built on an assumption that fire seasons across the Pacific and the Mediterranean rotate through the year. Strike teams fly north in July and south in December. Aerial assets cycle between continents. If synchronous extreme fire weather continues to trend the way Yin and colleagues document, that rotation stops being reliable, and any preparedness plan that tacitly assumes access to off-season international capacity is running on a premise the atmosphere is no longer willing to honour. This is the subject of a separate article on concurrent fire weather and the end of sequential mutual aid, but the resourcing implication bleeds into everything else.

There is an instinctive response to this kind of argument which is worth acknowledging before dismissing. It says that resource questions at the state level are important, but the real leverage is community preparedness, because even a fully resourced service cannot be everywhere at once. That instinct is correct as far as it goes. The CFA’s Get Fire Ready initiative, evaluated in March 2026, reported that 96 per cent of attendees either took or planned to take preparedness action after the program (CFA Victoria, 2026). Those numbers are genuinely encouraging, and the program is clearly doing something useful at the community level.

But community preparedness programs cannot do the work that the resourcing question actually asks of the system. Household-level preparedness changes what happens to individual properties on a bad day. It does not change the availability of strike teams and aircraft, nor the radio coverage and water supply that let those assets do their job, nor the basic capacity of the service to hold a fire line under extreme conditions. A Get Fire Ready participant who has cleared their gutters and put together a go-bag is better off than one who has not. The same participant, facing a 48.9 degree day with no available appliances and a fire front moving faster than anything the state can project, is nonetheless facing a state-level capacity problem. Community engagement does not substitute for investment in the assets and workforce that make those programs effective in the first place.

This is where the policy conversation will get uncomfortable over the next few years. The attribution science does not generate the money for the resourcing solution, and the resourcing solution is not cheap. Fire agencies around the country will be asking for larger aerial fleets and bigger rural fire workforces. They will also be asking for modern communications kit, better fuel management funding, and longer-term infrastructure hardening at the wildland-urban interface. Treasury departments will quite reasonably want to know how much of this is truly new and how much has always been needed but was previously deferred on the assumption that the climate context was going to stay broadly where it was. The honest answer to that question, after the Turco paper and the Yin paper, is that the old assumption is no longer defensible and that the liability has genuinely grown.

Emergency management practitioners have known this for years without being able to prove it. What the last few weeks of the scientific literature has done is remove the last plausible reason for treating that practitioner judgement as contested. The science is settled. The politics of paying for the response that the settled science implies have not even properly started. That is the conversation the next Victorian or New South Wales fire season is likely to force open, whether we are ready for it or not.

The attribution debate is the easier argument to have. The resourcing argument is the one that actually decides what happens in the next bad summer.

References

CFA Victoria. (2026, March 12). Get Fire Ready program evaluation. Country Fire Authority, Victoria.

Turco, M. and colleagues. (2026). The emerging human fingerprint on global extreme fire weather. Science Advances, 12, eadx9845. https://doi.org/10.1126/sciadv.adx9845

World Weather Attribution. (2026, January 22). Human-induced climate change made the January 2026 Victorian heatwave around five times more likely. World Weather Attribution.

Yin, C. and colleagues. (2026, February 18). Increasing synchronicity of global extreme fire weather. Science Advances. https://doi.org/10.1126/sciadv.adx8813