Extinctions Overestimated by 160 Percent?

Method for calculating animal and plant die-offs flawed, study says.
Brian Handwerk National Geographic News 18 May 11;

Global extinction rates may have been overestimated by as much as 160 percent, according to a new analysis.

In recent decades numerous studies have predicted that habitat destruction will doom some 20 to 50 percent of Earth's species within 500 years.

It's true that many species are still dying off, but the decline is happening at a slower pace than generally feared, according to study co-author Stephen Hubbell, an ecologist at the University of California, Los Angeles.

"The good news is that we may have a little more time in terms of saving some species," Hubbell said.

The bad news, he stressed, is that surging extinctions driven by habitat loss remain the critical conservation problem of the 21st century.

Method for Measuring Extinctions Flawed?

There's no proven, direct method for verifying extinction rates, so most scientists have relied on an indirect method to estimate how quickly plants and animals are disappearing.

That method calculates the rate at which new species are found when a new habitat area is sampled—called the species-area relationship (SAR)—and simply reverses that curve to predict the number of species that will go extinct as similarly sized areas of habitat are destroyed.

But Hubbell said the method is flawed, because much more land area must be lost to cause an extinction than is required to find a new species.

That's because only one individual of a species needs to be found in an area for scientists to deem it a new population, but extinction requires every member of a species to disappear.

"It's equivalent to saying a species is committed to extinction if you find the first individual and destroy its habitat, and that's clearly not true," Hubbell said.

"You have to destroy all of the habitat that has all the individuals of a species in it before that species goes extinct."

Hubbell and colleague Fangliang He of Sun Yat-sen University in Guangzhou, China, analyzed data from eight previously mapped forest areas from around the world. Each plot was between about 50 and 125 acres (20 and 50 hectares). The team also looked at ranges of several bird species in the continental United States.

Based on this real-life data and a mathematical model—in which the hypothetical destruction of habitat always resulted in fewer extinctions than predicted by SAR—the scientists calculated that the SAR-derived extinction rates had been overstated by as much as 160 percent.

The team also suggested that future studies could reveal even higher overestimates in some places.

Habitat Loss Still a Threat to Species

Yet ecologist Eric Dinerstein, who wasn't involved in the new study, said that examining how extinction rates are calculated is a bit of an academic argument for many conservationists.

"If it's a 160 percent overestimate or an 80 percent overestimate or a 20 percent overestimate, [comparing] which model of extinction rates is more accurate isn't the most important question," said Dinerstein, vice president of conservation science for WWF, a global conservation group.

"The overpowering message is that habitat loss and fragmentation are still the greatest threat to the future of species, and they are only increasing."

Dinerstein added that it's hard to determine when a species has gone extinct, as evidenced by numerous animals once thought gone but later found alive in small numbers.

And the final extinction of a species may be beside the point, Dinerstein said. What really matters is ecological extinction.

"That's when a population drops below a certain number of individuals and is no longer playing an ecological role in the ecosystem," Dinerstein said.

At this point the diminished species has so little interaction with the other plants and animals in the habitat that the species might as well be gone, from the point of view of the ecosystem.

Ecological extinction is of "much more concern to conservationists than [identifying] the last one or two individuals of some species which are still [alive] but functionally extinct."

Extinction Rates Critical for Conservation

Hubbell and He stressed that their research doesn't change the big picture, which isn't particularly rosy for species survival.

"I think [scientists and conservationists] are right in saying that we're really on the cusp of a sixth mass extinction or that it's actually in progress. We certainly don't disagree with that assessment," said Hubbell, whose study appears May 19 in the journal Nature.

But he also noted that learning how to calculate extinction rates properly is critical for conservation.

Take extinction-rate estimates by major initiatives, such as the Intergovernmental Panel on Climate Change reports and the U.N.'s Millennium Ecosystem Assessment.

If such estimates "are going to have consequences for billions of dollars in conservation efforts, don't you think we ought to know better why we're spending money and what the actual numbers are?" Hubbell said.

Co-author He also told reporters during a press briefing that no other scientific activity is arguably more important than understanding the causes and consequences of species extinctions.

However, He and Hubbell added that determining extinction rates has a long way to go.

"The bad news is that we really don't have good methods for estimating extinction yet," Hubbell explained.

"The precise answer depends on the precise pattern of habitat destruction in relation to the precise distribution of species.

"And although we can look at habitat destruction from satellites, we often just don't know where species live on the ground."

Plant, Animal Extinctions Often Exaggerated: Study
Alister Doyle PlanetArk 19 May 11;

A projected spate of extinctions of animals and plants this century may be less drastic than feared because the most widely used scientific method can exaggerate losses by more than 160 percent, a study said on Wednesday.

"Extinctions caused by habitat loss require greater loss of habitat than previously thought," two experts, based in China and the United States, wrote in the journal Nature.

Despite that good news, the report also endorsed past findings that human activities are wrecking habitats from the tropics to the Arctic, threatening the worst losses of species since the dinosaurs.

"Our results must not lead to complacency about extinction due to habitat loss, which is a real and growing threat," Fangliang He and Stephen Hubbell wrote.

The study, based on a survey of birds in the United States and forests, suggested the most commonly used method can exaggerate losses by more than 160 percent.

"The method has to be revised," Hubbell, of the University of California, told a news conference.

Scientists have long struggled to project extinctions as a rising human population shrinks habitats, for instance by felling forests to clear land for farms or cities. Pollution and global warming are also adding to threats.

The scientists stoked controversy by saying there was "reason to question" a U.N.-led Millennium Ecosystem Assessment that projected future extinctions at 1,000 to 10,000 times current rates, and a 2004 study saying that 18 to 35 percent of all species could be set on a path toward extinction by 2050.

Chris Thomas, the lead author of the latter study at the University of York in England, said he had published an update later in 2004 with a less severe extinction projection, broadly using techniques advocated in Wednesday's report.

CLIMATE CHANGE

"It is a pity that the authors did not realize this," he said. "And currently there is no reason for complacency that the extinction risk from climate change will necessarily be lower" than originally projected, he told Reuters.

Wednesday's report did not question findings by the U.N. panel of climate scientists in 2007 -- used by governments to guide climate policies -- that said 20 to 30 percent of species may be "at increased risk of extinction" as temperatures rise.

For scientists, the problem is they can fairly easily count species in an area -- adding one for each new bird, flower or mammal they find. It is far harder to count extinctions since that requires a judgment that the last individual has died.

Some studies in the 1970s, for instance, wrongly projected that half of all species could be lost by 2000.

More recent studies have added the idea of an "extinction debt," that species are doomed to die out once their habitat shrinks beyond a critical point. The study said estimations used by that technique were mathematically flawed.

Still, it said there was "no doubt whatsoever that the Millennium Ecosystem Assessment has correctly identified habitat loss as the primary threat to conserving the Earth's biodiversity, and the sixth mass extinction might already be upon us or imminent."

Scientists count five mass extinctions in the fossil record, the most recent 65 million years ago when dinosaurs vanished.

(Editing by Mark Trevelyan)

Species loss far less severe than feared: study
Marlowe Hood Yahoo News 18 May 11;

PARIS (AFP) – The pace at which humans are driving animal and plant species toward extinction through habitat destruction is at least twice as slow as previously thought, according to a study released Wednesday.

Earth's biodiversity continues to dwindle due to deforestation, climate change, over-exploitation and chemical runoff into rivers and oceans, said the study, published in Nature.

"The evidence is in -- humans really are causing extreme extinction rates," said co-author Stephen Hubbell, a professor of ecology and evolutionary biology at the University of California at Los Angeles.

But key measures of species loss in the 2005 UN Millennium Ecosystem Assessment and the 2007 Intergovernmental Panel on Climate Change (IPCC) report are based on "fundamentally flawed" methods that exaggerate the threat of extinction, the researchers said.

The International Union for the Conservation of Nature (IUCN) "Red List" of endangered species -- likewise a benchmark for policy makers -- is now also subject to review, they said.

"Based on a mathematical proof and empirical data, we show that previous estimates should be divided roughly by 2.5," Hubbell told journalists by phone.

"This is welcome news in that we have bought a little time for saving species. But it is unwelcome news because we have to redo a whole lot of research that was done incorrectly."

Up to now, scientists have asserted that species are currently dying out at 100 to 1,000 times the so-called "background rate," the average pace of extinctions over the history of life on Earth.

UN reports have predicted these rates will accelerate tenfold in the coming centuries.

The new study challenges these estimates. "The method has got to be revised. It is not right," said Hubbell.

How did science get it wrong for so long?

Because it is difficult to directly measure extinction rates, scientists used an indirect approach called a "species-area relationship."

This method starts with the number of species found in a given area and then estimates how that number grows as the area expands.

To figure out how many species will remain when the amount of land decreases due to habitat loss, researchers simply reversed the calculations.

But the study, co-authored by Fangliang He of Sun Yat-sen University in Guangzhou, shows that the area required to remove the entire population is always larger -- usually much larger -- than the area needed to make contact with a species for the first time.

"You can't just turn it around to calculate how many species should be left when the area is reduced," said Hubbell.

That, however, is precisely what scientists have done for nearly three decades, giving rise to a glaring discrepancy between what models predicted and what was observed on the ground or in the sea.

Dire forecasts in the early 1980s said that as many as half of species on Earth would disappear by 2000. "Obviously that didn't happen," Hubbell said.

But rather than question the methods, scientists developed a concept called "extinction debt" to explain the gap.

Species in decline, according to this logic, are doomed to disappear even if it takes decades or longer for the last individuals to die out.

But extinction debt, it turns out, almost certainly does not exist.

"It is kind of shocking" that no one spotted the error earlier, said Hubbell. "What this shows is that many scientists can be led away from the right answer by thinking about the problem in the wrong way."

Human encroachment is the main driver of species extinction. Only 20 percent of forests are still in a wild state, and nearly 40 percent of the planet's ice-free land is now given over to agriculture.

Some three-quarters of all species are thought to live in rain forests, which are disappearing at the rate of about half-a-percent per year.

Calculations may have overestimated extinction rates
Debora MacKenzie New Scientist 18 May 11;

THE destruction of nature is driving species to extinction - but perhaps not as rapidly as has been thought. While the most widely publicised estimates predict the loss of natural habitat will condemn 18 to 35 per cent of all species to extinction by 2050, these figures could be about twice as high as the actual number - all because of a mathematical error that has gone unnoticed for decades.

We still face an extinction crisis, warn Stephen Hubbell of the University of California, Los Angeles, and Fangliang He of Sun Yat-sen University in Guangzhou, China. But the pair's work will allow biologists to more precisely define how habitat destruction leads to extinction.

It is impossible to accurately measure extinction rates. Dozens of new species are identified each year, and counting those that disappear is hard because many are small and live in poorly studied, mainly tropical environments.

Instead, extinction rates are often predicted from a mathematical model based on habitat loss, which is more easily measured. The larger the area you survey, the more species you encounter. Ecologists calculate a curve called the species area relationship (SAR) for an ecosystem by measuring the area they must survey to encounter the first individual of each successive species. To establish the number of extinctions caused by habitat destruction, they run the SAR calculation in reverse.

"We had a feeling there were problems with this, but we could not say why mathematically," Hubbell says. So Hubbell and He checked the method using data from forest plots located all over the world. The pair could calculate the SAR for each plot, and also see what happened to species unique to these plots if they "destroyed" a certain area of each plot in their mathematical model. As the area of destruction widened, these species began to die out. But after each simulated loss of habitat, "more species always remained than were expected from the SAR", says Hubbell.

The pair's analysis explains why. Using the reverse SAR method, biologists have assumed that a species is lost with the destruction of an area of habitat equivalent to the area needed to first encounter it. But in reality, the species is lost only with destruction of the habitat area that includes every individual of the species, which is always larger. Consequently, the SAR method loses species too fast.

The duo developed a model relating extinction rate instead to the entire area occupied by a species. Using the forest data, and extensive data sets on birds, they found that the SAR gave extinction rates that were between 83 and 165 per cent higher than those their method produced (Nature, DOI: 10.1038/nature09985).

Similarly detailed information does not exist for most of the world's species, making it difficult to apply Hubbell and He's model more generally. "As a rule of thumb, we might correct traditional extinction rates by dividing them by factor of 2 to 2.5," says He.

Jean-Christophe Vié, deputy head of species survival at the International Union for the Conservation of Nature, agrees better baseline data on species is badly needed. He says IUCN doesn't use the SAR method. But, he points out, "a twofold miscalculation doesn't make much difference to an extinction rate now 100 to 1000 times the natural background".

Hubbell and He agree: "Mass extinction might already be upon us."