Scientists know that climate change is putting species around the globe in peril, but just how much peril?
After all, when evolution failed to keep pace with a major climatic event 65 million years ago, half the planet's species went extinct and dinosaurs were reduced to jittery feathered creatures that get bullied by squirrels on bird-feeders.
A new study suggests that our current era of climate change won't just exceed the rate of evolution, but will do so by a factor of thousands.
Although the work doesn't go so far as predicting an extinction rate, it doesn't bode well for the near future of global biodiversity.
The world has warmed 0.6 degrees Celsius in the past few decades, and climate models say that we could see another 4 degrees by century's end.
"We want to know if species will be able to adapt to climate change quickly enough based on how they adapted to climate change in the past," says evolutionary ecologist John Wiens, of the University of Arizona, and lead author of the new study. Wiens decided to investigate by looking at the top branches of family trees.
When two living species are closely related, scientists can estimate how long ago they diverged, thus providing an age for their common ancestor.
Researchers can also estimate temperature and precipitation in that ancestor's habitat, using evolutionary models.
With help from Yale University biology student Ignacio Quintero, Wiens calculated such estimates for 540 species in 17 groups of living vertebrates.
They studied reptiles, amphibians, birds, and mammals primarily native to North and Central America, but with some European, Asian, Australian, South American, and African species as well.
Then they used global climate models to determine how the local climate of each species is expected to change by the end of this century.
Despite differences in local climate and in the vertebrates themselves, the results were consistent.
The average rate of adaptation for 15 of the 17 groups was less than 1C per million years. Two groups adapted slightly faster, but still below 2C per million years.
So if a frog breeds in autumn because the temperature is right, it might adapt to warmer temperatures by breeding in December, January, or February. And lizards that survive on those eggs might have to change their diet. But the study found that such adaptations typically occur about 10,000 to 100,000 times too slowly to keep pace with global warming projections for the year 2100.
The researchers reached the same conclusion for the expected regional increases and decreases in rainfall: Again, the species adapted 10,000 to 100,000 times too slowly.
Adapting too slowly does not mean certain death. A species can relocate. But due to habitat destruction and other factors, not all species can move.
If a rodent lives on a mountain and warmer temperatures compel the animal to climb higher, it may run out of mountain while temperatures keep rising.
Wiens was surprised by the results because they suggest that the studied species, which typically adapt to less than 1C of change per million years, now must adapt to 4C between now and the year 2100.
"It's almost crazy to think that they're going to, in just a few decades, be more different than they've become over millions of years," he says.
Wiens cautions that the study - which will be published in the August issue of Ecology Letters - looked at only hundreds of species, not the millions in real-life ecosystems, and does not attempt to estimate an extinction rate.
"If you extrapolated from our sample of species, it might be about 50 per cent, but it could be more or it could be less," he says.
"I think this is a very interesting and worthy study, which will certainly stimulate a lot of discussion," says evolutionary biologist Michael Donoghue of Yale University, who wasn't involved in the study.
But the paper itself notes that animals could be considerably more adaptable than its methods suggest.
"Somebody could reasonably argue that they've been evolving at this rate because climates have been changing slowly," Donoghue says.
Other studies have found remarkably high rates of adaptation in some species, such as Galápagos finches, he says.
For example, a 2006 report found that when a nonnative species of ground finch arrived on one Galápagos island, it began feeding on large seeds that were preferred by a smaller native ground finch.
Unable to compete with the bigger bird, the smaller species instead evolved a smaller beak to specialise in smaller seeds - all in about 20 years.
"I think they have done a good job of addressing potential minor and major sources of error," Donoghue says.
"This will help with the reception of these findings, but will not completely eliminate skepticism."
Another possible source of error is the estimated time scales.
If the researchers incorrectly estimated a 200,000-year-old salamander species to be 2 million years old, then their result was off by a factor of 10, making the salamander's rate of evolution 20,000 times slower than climate change instead of 200,000 times slower.
In other words, the sources of error are important but probably not significant enough to overwhelm the conclusions, says Lauren Buckley, a University of Washington, Seattle, climate change ecologist who wasn't involved in the study.
"They've made a good argument that the point still stands," Buckley says.
"I think my main take-home from the article is that the techniques are pretty coarse, but it does give us a back-of-the-envelope estimation."
- SCIENCE NOW
What will be the main motivation for humanity's future space endeavours?Related story: (See story)
The cost of losing nature