New paper stirs up controversy over how scientists estimate extinction rates

by Jeremy Hance
from mongabay.com

A new paper in Nature negating how scientists estimate extinction rates has struck a nerve across the scientific community. The new paper clearly states that a mass extinction crisis is underway, however it argues that due to an incorrect method of determining extinction rates the crisis isn't as severe as has been reported. But other experts in the field contacted disagree, telling mongabay.com that the new the paper is 'plain wrong'. In fact, a number of well-known researchers are currently drafting a response to the day-old, but controversial paper.

Estimating extinction rates has always been full of uncertainty. Given the lack of knowledge over how many species survive on Earth—is it 5 million or 100 million?—it is equally difficult to know how quickly species (the majority of which haven't even been described) vanish in the face of habitat loss and other impacts. Scientists have predicted that the current rate of extinction loss is between 100 to 1,000 times above the average background rate of extinctions, but the new paper says such predications are overestimations.

Paper adds controversy
The new paper, by Fangliang He and Stephen Hubbell, has aggravated the debate over just how many species are vanishing. He and Hubbell argue that the major way in which researchers estimate species extinction—the 'species-area relationship'—has overestimated extinction rates by up to 160 percent.

Hubbell, a tropical forest ecologist and a scientist at the Smithsonian Tropical Research Institute (STRI), says in a press release that although it's true "we are losing habitat faster than at any time over the last 65 million years" the method researchers have used to estimate extinctions is "erroneous".

"The good news is that we are not in quite as serious trouble right now as people had thought, but that is no reason for complacency. I don't want this research to be misconstrued as saying we don't have anything to worry about when nothing is further from the truth," Hubbell says, obviously wanting to make certain the media doesn't turn this into a denial of the mass extinction crisis, which experts agree is currently occurring or very near on the horizon.

But according to Hubbell and He, using species-area relationship does not come up with an accurate answer as to the rate of extinction. Species-area relationship estimates loss of species by looking at the loss of habitat. Surveying how many species are in a given area, researchers than extrapolate how many species are in the wider ecosystem. Working backwards, researchers may estimate the number of species that go extinct as the habitat is destroyed. For example if 90% of a forest is clearcut, the species-area relationship predicts about half the species in the ecosystem would vanish. These species may not vanish right away, small populations may still persist, but ecologists argue these populations suffer from 'extinction debt'. In other words, they are doomed to extinction over time, what some researchers describe as the living dead'. According to research it can take species many generations to go extinct even after a population tipping-point has been reached.

But, the paper argues that this mathematical method suffers from overestimating extinctions because most species still survive, albeit in significantly reduced populations, in the habitat remaining.

"We show that this surrogate measure is fundamentally flawed," says Hubbell, "you can't just turn [the species-area relationship] around to calculate how many species should be left when the area is reduced; the area you need to sample to first locate a species is always less than the area you have to sample to eliminate the last member of the species."

Hubbell says he agrees that 'extinction debt' still exists, but that it is not as prevalent as has been estimated. Instead, the paper argues that researchers should predict species extinction related to habitat loss only when an entire species' habitat is gone (i.e. the species is endemic to the destroyed area).

As an example of where the species-area relationship has overblown predictions, the study points to forecasts in the early 1980s that half of the species on Earth would be extinct by 2000. However, that has not occurred among known species.

"Nothing like that has happened," Hubbell says.

Muddying the waters, of course, is the fact that while habitat loss is the greatest threat to species worldwide, other impacts, such as hunting and poaching, overconsumption, pollution, disease, and invasive species also imperil species. In addition, climate change could become a major in causing future extinctions, especially in the face of tiny populations surviving in habitat fragments.

Complicating the picture even further is conservation efforts of known species. Some species should have gone extinct, but continue to hang-on due largely to long-term and active conservation efforts. However, cryptic species—those not yet known to researchers—don't have such a luxury.

Response
The response to Hubbell and He's paper from other experts in the field has been rapid and negative.

"[The study] is widely inaccurate, not because it is wrong, but because it considers only a very small part of the problem," Stuart Pimm professor of Conservation Ecology at Duke University told mongabay.com. "The claims in the title are, simply false, and constitute nothing more than arrogant posturing on part of authors who would not have got their paper published had they not seen fit to ignore and misquote an extensive body of literature published in prestigious international journals. "

Pimm points out that the title of the Hubbell and He's paper, Species–area relationships always overestimate extinction rates from habitat loss, is undercut immediately by studies that have found that species-area relationship has resulted in correct predictions. For example, Pimm points to a study of his where the species-area relationship correctly predicted the extinction of 4.5 birds in Eastern North America. In fact, four birds have gone extinct, and one (the 0.5) is threatened with extinction.

According to Pimm, the problem lies in the paper's conflating immediate species extinction after habitat loss with eventual species extinctions over time, in other words it completely ignores the extinction debt, those living dead populations that will wink out in time.

"Imagine destruction that wipes out 95% of the habitat in an area metaphorically 'overnight'. How many species have disappeared 'the following morning'? The paper tells you. It is not many, just those wholly restricted to the 95% (and absent from the 5% where they would survive)," Pimm says, agreeing that the initial extinction rate would be low, because it would only affect endemics, i.e. those found only in the habitat destroyed. But in time, says Pimm, extinction rates would worsen even if no additional habitat loss occurs.

"The important question is…how many of additional species living lonely lives in their isolated patches (the 5%) would become extinct eventually because their population sizes are too small to be viable?"

Lian Pin Koh, a tropical ecologist at ETH Zurich, agrees with Pimm that the paper's findings are incorrect. Koh says the authors "are simply barking up the wrong tree" because they have confused how researchers estimate the number of species in an ecosystem and how they estimate extinction rates.

"The basic premise of the He and Hubbell paper is plain wrong," Koh told mongabay.com, "The species-area relationships they refer to are actually species accumulation curves—the relationship between sample size and number of species encountered. On the other hand, the species-area models that ecologists use to predict species extinctions are a different set of relationships altogether, one that is based on empirical evidence of the relationship between the size of islands and number of species inhabiting those islands."

In other words, according to Koh, the paper does not take-on the correct model.

Koh and Pimm are apart of a group of experts who are currently drafting a formal response to the Nature paper.

"The paper is a sham," Pimm continues, "it does not report extinction rates or the numbers of species that are threatened. Despite its posturing, it deals with a different issue."

However, Hubbell said in a press release that he has '100 percent' confident in his findings. He sees his paper as 'good news' in that it gives humanity more time to prevent a mass extinction crisis. But, at least, in this researchers agree: life on Earth is threatened worldwide.

The evidence for this is undeniable: according to the IUCN Red List, the world authority on species' threat levels, 41% of the world's amphibians are threatened with extinction, 33% of cartilaginous fishes, 33% of coral reefs, 25% of mammals, 22% of reptiles, 15% of bony fishes, 14% of sea grasses, and 13% of birds. Even these are likely to be underestimates. Studies have shown that recently discovered species have a higher risk of extinction than species known for centuries. Therefore, cryptic species, those still undiscovered (which outnumber known species) are more likely to be weighted towards threatened rather than thriving.

Whatever the exact rate is at which species are going extinct, a biodiversity crisis is occurring that portends a global extinction not seen since the dinosaurs fell. Though not as important to mainstream media as Arnold Schwarzenegger's infidelity and low on the list of priorities for world leaders, ecologists—even when they disagree on the details—still agree the world is in the middle of a full-blown crisis.

"The next mass extinction may be upon us or just around the corner," says Hubbell. "There have been five mass extinctions in the history of the Earth, and we could be entering the sixth mass extinction."

CITATION: Fangliang He and Stephen P. Hubbell. Species–area relationships always overestimate extinction rates from habitat loss. Nature. Vol. 473. May 19, 2011. doi:10.1038/nature09985.