By 2085, a Third of All Animal Habitats Could Be Hit by Multiple Climate Disasters at Once

In the summer of 2019, a heatwave moving across eastern Australia killed more than 72,000 flying foxes. The bats dropped from their roost trees mid-flight, or died clinging to branches, cooked alive by temperatures that exceeded anything in the historical record for those regions. That same season, wildfires ripped through the Pantanal and killed an estimated 17 million vertebrates. Conservation biologists have spent years studying each of these disasters in isolation. A new analysis from the Potsdam Institute for Climate Impact Research asks what happens when you stop treating them as separate problems.

The issue isn’t exposure to any single disaster. It’s the compounding. By 2085, more than a third of the land area within current species ranges could be hit by two or more types of extreme events in the same year, or in quick succession, and almost nothing in conventional conservation planning accounts for that.

The study, published today in Nature Ecology & Evolution, analysed projected extreme event exposure for 33,936 terrestrial vertebrate species: amphibians, birds, mammals and reptiles. The team drew on climate impact models from the Inter-Sectoral Impact Model Intercomparison Project, a framework that goes beyond simple temperature projections to simulate burned area, soil moisture deficits, and river flood depth. This distinction matters more than it might seem. A fire weather index tells you conditions are ripe for burning; a burned-area model tells you how much of a given species’ habitat actually goes up. The difference, for a Carnaby’s Black Cockatoo trying to survive in Western Australia, is the difference between a theoretical risk and a real one.

That cockatoo, incidentally, lost roughly 60 percent of its population following a single heatwave in 2011. Not across decades. One event, one season.

What the new analysis makes vivid is how rapidly the arithmetic of catastrophe is changing. In 2050, already, 22 ecoregions have more than half their area exposed to two or more types of extreme events simultaneously. By 2085, that figure rises to 236 ecoregions. The hotspots are exactly where you’d least want them: the Amazon basin, tropical Africa, Southeast Asia. These are the places on Earth most densely packed with species, many of them found nowhere else, and they’re projected to face the steepest increases in wildfire frequency alongside heatwaves that are already intensifying.

“I think climate change, and in particular extreme events, are still really being underestimated when it comes to conservation planning,” said lead author Stefanie Heinicke, a postdoctoral researcher at PIK. “It’s not just going to be a gradual shift of temperature over many years.”

The Wildfire Blind Spot

The research is perhaps most striking on the wildfire question. Previous work on species’ climate vulnerability has tended to focus on temperature extremes, partly because those figures fall out naturally from climate models, and partly because the data on how fire exposure translates into species harm has been murkier. Katja Frieler, who leads the ISIMIP project and co-authored the paper, was direct about the gap: “The wildfire projections being so significant is really notable. I don’t know of another study that has projected wildfire exposure for animals yet, so seeing that there is a bigger threat from fires than drought for example; this was a significant blind spot.” By 2050, wildfire already overtakes drought as the second-largest exposure category for vertebrates globally, affecting 16% of species’ range area compared with 8% for drought. Amphibians, notably, face disproportionate drought exposure relative to other groups, probably because of their reliance on moist microhabitats and their limited capacity to seek shelter or migrate.

When Disasters Stack

The mechanisms driving multi-hazard risk are not just additive. Fire preceded by drought burns hotter, spreads further and leaves animals with fewer refugia. Analysis of the 2019-20 Australian megafire season showed species declines 27-40% greater in areas where drought had come first. The Pantanal fires of 2020, separately, killed an estimated 17 million vertebrates. These aren’t anomalies to be averaged out; they’re becoming, in parts of the world, the expected rhythm of the calendar.

The study’s authors are careful about what they have and haven’t measured. Exposure is not the same as impact. Some species are physiologically plastic enough to cope with conditions well outside historical norms; others depend on the very disturbances the models flag as threats. The riffian skink, for instance, thrives in post-fire habitat where shrub clearance has opened up the terrain it prefers. A moderate wildfire regime isn’t necessarily disaster for every species in its path. What the analysis can’t yet do, and says so explicitly, is connect the exposure data to population-level sensitivity or adaptive capacity. That knowledge gap is more or less the next frontier of the field.

The Emissions Lever

There is, though, one very clear signal in the numbers. The difference between high and low emissions scenarios is not marginal. Under aggressive emissions cuts (SSP1-2.6, aligned with Paris Agreement targets), the share of vertebrate habitat facing multiple extreme event types by 2085 drops from 36% to just 9%. That’s not a rounding error. Roughly three-quarters of the multi-hazard exposure projected for the high-emissions world simply doesn’t materialise if warming is kept in check. “There’s still a lot of difference we can make by cutting emissions as fast as we can from today,” Heinicke said.

For conservation practitioners, the study is a provocation. Protected areas are typically designated on the basis of where species are now, and managed against threats that operate on human timescales. A reserve that protects a rainforest species today may, by 2060, be sitting inside an ecoregion hit by heatwaves every few years, with wildfire increasing and seasonal flooding amplifying the pressure. Whether the reserve boundary, or the management plan, or the species itself, can adapt quickly enough to that kind of compound, accelerating pressure is a question the modelling can’t answer. But it’s now possible to identify where, and roughly when, those questions are going to become urgent.

The 236 ecoregions that are expected to face multi-hazard exposure by 2085 are, in that sense, a list of pending decisions for conservation planning. Some of them are in mid-latitudes that might seem less obviously imperilled than tropical biodiversity hotspots. The mathematics of compounding risk, it turns out, doesn’t spare the temperate world either.

https://doi.org/10.1038/s41559-026-03050-0

Frequently Asked Questions

Why is being hit by multiple climate disasters worse than just one?

When extreme events stack on top of each other, the damage compounds in ways that individual events don’t predict. A fire that follows a drought burns more intensely and leaves animals with fewer places to shelter; a heatwave in a habitat already stressed by flooding offers fewer opportunities for recovery. Research on the 2019-20 Australian megafires found that plant and animal species declined 27-40% more severely in areas where drought had immediately preceded the fires. Species that might survive one catastrophe often can’t rebound fast enough before the next one arrives.

Is it true that some animals actually benefit from events like fire and drought?

Yes, and the researchers are careful to acknowledge this. Some species depend on the open habitat that fire creates, and certain droughts reduce fungal pressure on amphibian populations by slowing the growth of harmful fungi. The real concern is the intensity and frequency of these events under climate change: even species adapted to periodic disturbance can be overwhelmed when those disturbances become more severe, more frequent, or arrive in combinations they haven’t evolved to handle. The study measures exposure, not impact, precisely because the relationship between events and outcomes varies so much by species.

Why did wildfire come out as a bigger threat to animals than drought?

Partly because this study is one of the first to model wildfire exposure for vertebrates directly, using burned-area projections rather than fire weather indices. Previous analyses tended to focus on temperature or aridity, where the data was easier to extract from climate models. When burned area is modelled explicitly, fires emerge as the second-largest source of extreme exposure by 2050, affecting roughly 16% of vertebrate habitat globally compared with about 8% for drought. The Amazon basin, southern Africa and Southeast Asia are projected to see the steepest increases in fire frequency, which overlap almost exactly with the highest concentrations of species richness on Earth.

Could cutting emissions now actually make a meaningful difference to wildlife at this point?

The study suggests it would make an enormous difference. Under a high-emissions scenario, 36% of vertebrate habitat could face multiple types of extreme events by 2085. Under a low-emissions path aligned with the Paris Agreement, that figure falls to just 9%. That gap represents hundreds of ecoregions and tens of thousands of species. The researchers are clear that mitigation isn’t a half-measure here: the trajectory of emissions over the coming decades will determine whether multi-hazard exposure becomes a global norm or stays a regional exception.