Amphibian Declines - Global Warming or Beer to Blame?

Bill's forum was the first! All subjects are welcome. Participation by all encouraged.

Moderator: Available

Post Reply
User avatar
Glenn
Posts: 2199
Joined: Thu Dec 20, 2001 6:01 am
Location: Lincoln, Nebraska

Amphibian Declines - Global Warming or Beer to Blame?

Post by Glenn »

You'll like this one Bill.
Amphibian Extinctions: Is Global Warming Off the Hook?
Amphibian Extinctions: Is Global Warming Off the Hook?
Brian Handwerk
for National Geographic News
December 1, 2008

The world's amphibians are in dire straits—but global warming may not be the problem, a new study suggests.

Previous research has pinned steep declines in amphibian species on rising global temperatures, which are said to be fueling the growth of a deadly fungus.

Most experts agree that the disease-causing chytrid fungus Batrachochytrium dendrobatidis is taking a terrible toll on frogs and toads. One in three species worldwide is threatened with extinction.

"There seems to be convincing evidence that chytrid fungus is the bullet killing amphibians," said University of South Florida biologist Jason Rohr, lead author of the study, published in a recent issue of the journal Proceedings of the National Academy of Sciences.

"But the evidence that climate change is pulling the trigger is weak at this point."

Beer to Blame?

Rohr and colleagues don't completely discount the role of global warming in amphibian declines.

But they say decades of data show only that some correlation exists between rising air temperatures and Latin American amphibian extinctions—and that data are well short of proving causation.

In fact, the researchers found that, in the Latin American countries they studied, beer and banana production were actually better predictors of amphibian extinctions than tropical air temperature.

While beer and bananas are certainly not to blame, the whimsical comparison makes a point.

"We can't jump to conclusions of causality based on a correlation—especially when we're talking about 60 or 70 species," Rohr said.

"The major message is that we need to be careful."

But J. Alan Pounds, resident scientist at Costa Rica's Monteverde Cloud Forest Preserve and an author of previous research on the topic, maintains climate change is a key factor in the extinction crisis.

"There is a clear link between global warming and amphibian declines," he said, citing a growing body of evidence, including a recent study in Yellowstone National Park.

"The analysis by Rohr et al. is seriously flawed, as we will demonstrate in due course."

Heated Debate

Some scientists say the fungus is an invasive species, entering an ecosystem and wreaking havoc on species with no natural defenses.

Pounds and others believe that rising daytime and nocturnal temperatures are narrowing the gap between daily highs and lows.

Fewer temperature extremes are advantageous for chytrid fungus, which grows and reproduces best at temperatures between 17 and 25 degrees Celsius (63 and 77 degrees Fahrenheit).

The new research suggests neither theory has it exactly right.

For example, Rohr explained that, although extinctions increased significantly in the 1980s, that period did not show smaller swings between average daily high and low temps that would theoretically produce more fungus growth.

In the 1990s, daily temperature swings were smaller, but amphibian extinctions declined during that period, the study concludes.

Rohr and colleagues conclude that an unknown mix of factors is likely endangering amphibians.

Sorting it all out remains a high priority—and for many amphibians, time could be running out.
Glenn
User avatar
Glenn
Posts: 2199
Joined: Thu Dec 20, 2001 6:01 am
Location: Lincoln, Nebraska

Post by Glenn »

"We can't jump to conclusions of causality based on a correlation—especially when we're talking about 60 or 70 species," Rohr said.

"The major message is that we need to be careful."
Confusing causality with correlation is common. I have encountered it frequently with students, particularly in their term papers, and as a result I developed an exercise where they analyze a study claiming a relationship between the presence of Walmarts and declining income in Nebraska counties. The study is so full of holes (it only looks at income-change in counties with Walmarts, not how income was concurrently changing in counties without Walmarts; only looks at income-change in Walmart counties during the time they have a Walmart, without consideration of any long-term trends in income-change including before Walmart moved in; only shows a rough correlation and not causality or consideration of other factors) that it makes a relatively easy learning tool for freshman students to evaluate quality of the research and logical fallicies. When we discuss the exercise after I have graded it, the take-away message I leave is the same as above: We need to be careful about evaluating evidence and the conclusions we reach, and critically evaluate our sources of info.

Could the presence of a Walmart negatively affect the average income of rural Nebraska counties? Possibly, but this particular study certainly does not prove anything. And there are other more likely suspect causes, like the depopulation of Nebraska's rural counties, due to limited opportunities, resulting in the loss of a lot of higher-paying professional jobs (doctors, lawyers, etc) in those counties because there is not enough population base to support them.
Glenn
Valkenar
Posts: 1316
Joined: Mon Aug 21, 2000 6:01 am
Location: Somerville, ma.

Post by Valkenar »

Glenn wrote: Could the presence of a Walmart negatively affect the average income of rural Nebraska counties? Possibly, but this particular study certainly does not prove anything.
I would tend to expect that poverty causes Walmart, something that would also create a positive correlation between the two.
User avatar
Glenn
Posts: 2199
Joined: Thu Dec 20, 2001 6:01 am
Location: Lincoln, Nebraska

Post by Glenn »

Valkenar wrote: I would tend to expect that poverty causes Walmart, something that would also create a positive correlation between the two.
And that is a possibility too, other factors are causing average income to decline in counties creating a situation Walmart can capitalize on. And there are other possibilities. The complexity of the issue with numerous possible interactions prevents easy assumption of causality from a simple correlation, but people will try it.

A favorite example in statistics course/textbooks is the positive correlation between ice cream sales and crime rates, when ice cream sales go up so do crime rates. So ice cream sales cause crime? Maybe if it is successful ice cream stores that are the only ones being robbed. More likely there are other factors, such as nice weather being conducive to both ice cream sales and criminal activity independently.
Glenn
User avatar
Bill Glasheen
Posts: 17299
Joined: Thu Mar 11, 1999 6:01 am
Location: Richmond, VA --- Louisville, KY

Post by Bill Glasheen »

Good points, Glenn. This is why many scientists constantly seek the holy grail of the randomized, controlled trial (RCT). Better yet, a double-blinded one. But they are expensive, and in some cases highly unethical.

That doesn't mean that other designs aren't useful. But they must be done very carefully so that the causality issue can properly be addressed.

Bill
cxt
Posts: 1230
Joined: Wed Sep 10, 2003 5:29 pm

Post by cxt »

Couple of (common) ones......for fun :)

"Hot chacolate reduces crime"

Works like the reverse of the ice cream one---cold temps...the kind where people stay in and drink more HC led to a decrease of some kinds of crime.

"the larger a childs feet the better their handwriting"

Older kids usually have larger feet.....and older kids usually have better handwriting than younger ones.

I sometimes have students work the raw numbers and graph these just so they can see what it looks like......before we take it apart and explain the error/s in confusing corellation and causation....sometimes it helps.

I'd kinda like to know what exactly it was/is about "beer and banna production" that they reffered too...but the article does not seem to spell it out.......oh well, gives me something to do this weekend.
Forget #6, you are now serving nonsense.

HH
User avatar
Glenn
Posts: 2199
Joined: Thu Dec 20, 2001 6:01 am
Location: Lincoln, Nebraska

Post by Glenn »

cxt wrote: I'd kinda like to know what exactly it was/is about "beer and banna production" that they reffered too...but the article does not seem to spell it out
I don't know, but it provides a different entertaining descriptive example from the crime/ice cream one I'm always using in class :D
Glenn
Post Reply

Return to “Bill Glasheen's Dojo Roundtable”