(LONG) Lynx redux
Daniel B. Wheeler
Science, politics uneasy partners
Policy based on research for public isues such as the Endangered
Species Act is tricky because there are few absolutes
By MICHAEL MILSTEIN, The Oregonian
Two words - "sound science" - have become a catchphrase for those on
all sides of contenious decisions over logging, water, wildlife and
other natural resource issues that define the destiny of the West.
It echoes through the Klamath Basin, the snowy slopes of the Cascades
and the Northwest's sentinel forests.
Politicians toss the term like confetti at a party, while others use
it as an angry catchall for everything that a bad decision lacks. Land
managers may treat it as their Holy Grail, a silver bullet that will
tell them what decision to make.
To scientists, however, the only true science is sound science.
The key, they say, is whether the science is applied soundly and
whether the public and the land managers who rely on it understand
what it says and - just as important - what it doesn't.
Science may have its own quality controls, but that does not mean it
should go unquestioned.
"Don't let someone tell you, 'Science made me do something,'" said
Thomas Mills, director of the U.S. Forest Service's Pacific Northwest
Research Station in Portland. "Science helps you understand the
consequences of a decision, and who wouldn't want to know the
consequences of their actions? But people make decisions, and that's
the bottom line."
And those who make decisions should know that in science, as in life,
there ar efew absolutes. Rarely, if ever, will research prove that if
a forest is logged or a reservoir lowered, an endangered species will
go extinct. Instead, it will suggest a range of possible likely
consequences that decision-makers must then balance with public - and
sometimes political - values to decide which way to go.
"It really sin't a science question in that case; it's a value
qustion," Mills said. "People need to decide at what point the risk is
too great. That's a value choice."
The biggest blowup recently in the realm of land and wildlife science
did not involve scientific findings as much as scientific procedure.
But it's a good example of how science, and even much-heralded tools
such as DNA analysis, may be no more foolproof than human nature.
It erupted from a federal attempt to chart the range of lynx by
checking for ahir from the threatened cat on scratching posts
scattered through Western forests.
A pilot project had yielded false reports of lynx in Oregon and
Washington after fur samples were inadvertently contaminated. It is
still unclear what happened in that case, but "something went very
badly wrong," acknolwedged Scott Mills, a University of Montana
professor who is overseeing a broader four-year lynx survey for the
U.S. Forest Service.
His survey differed from the earlier faulty project. Mills developed
and published a protocol - a set of scientific guidelines - for the
survey in a journal reviewed by other researchers, a process known as
peer review. He also tested the protocol before and during the survey.
But seven state and federal biologists gathering hair samples for
Mills in oregon and Washington also sent him tufts of ahir from
captive or stuffed lynx to see for themselves whether this lab would
get it right. They did not tell the lab, making it a "blind" test, or
what scientists sometimes call a "blind control."
A Jan. 10 editorial in the scientific journal Nature concluded that
the scintists were trying to correct for a poorly designed study that
should have incorporated such blind controls in the first place.
Mills argued that his study had plenty of built-in controls. If no
one had caught the outside interference, he said, it could hav
eundermined his protocol and results. "If there hadn't been an
investigation, we wouldn't have known (the hair) was pulled off a wall
mount."
Either way, politicans and critics of the federal Endangerd Species
Act jumped the biologists in what Nature called a "lunch mob," blaming
them for "biofraud" and claiming that the biologists were trying to
trigger protections for wildlife that didn't exist. Some questioned
the reliability of unrelated research, shaking the roots of wildlife
science to a dgree that surprised even Scott Mills.
"I'm surprised at how polarized it's become, that politicians are
going to extreme lengths to portray that all biologists everywhere are
fraudulent," he said.
"Two different languages"
The speed with which science can go from a quiet academic pursuit to
a political battering ram may reflect the gulf separating scientists
and politicians, lawyers, land managers and others who decide the
future of public lands and resources. All have different goals, which
may color the way they apply the work of others.
"Often the expectations aren't clear on either side," said Deborah
Brosnan, president of Sustainable Ecosystems Institute, a Portland
nonprofit gorup that mediates environmental disputes. "Often the
manager wan'ts the scientists to solve a problem, but they speak two
different languages."
"Just because a scientific paper is peer-reviewed, for instance, does
not mean it's unassailable fact. Rather it means that other scientists
have found the reasoning and methods to be sound, a kind of "wuality
control in the scientific profession," Brosnan said.
Two peer-reviewed papers may draw opposit conclusions, said Steven
Courtney, vice president of SEI. That does not mean either is wrong.
It does mean science is not a clear-cut, yes-or-no pursuit - a real
but frustrating limitation in a world where public land decisions can
affect lives and livelihoods.
The Endangered Species Act calls for using the "best available"
science, which may not be enough science to satisfy everyone.
Brosnan has called for remaking the peer review process into a more
open system, a kind of impartial group discussion, so those involved
in environmental disputes better grasp the realities of science and do
not view it merely as a tool to get their way.
Much is not yet known
Indeed, some environmental mandates hinge on unknowns.
A provision of the Northwest Forest Paln, which governs logging in
the region, requires costly surveys for obscure slugs, snails and
other creatures at logging sites because no one knows how common or
rare they are.
Federal agencies withheld water from farmers in the Klamath Basin
last year not because they knew it would help protected fish, but
because they could not be sure that releasing the water would not hurt
the fish.
Sometimes the heated atmosphere of environmental disputes might limit
the progress of science itself. Stan Geiger, a Portland wetland
ecologist who has studied the Klamath Basin for years, recalls that
one scientists refused to review another's study because of the threat
of lawsuits over water use in the basin.
Such divides keep sicentists from sharing information, so neither the
public nor decision-makers learn everything that researchers have
learned about the vast basin and its wildlife. Different agencies with
roles in the basin may rely on only a slice of science when making
critical choices.
"There is no research plan that anyone is following, no quality
assurance protocols everyone has agreed to, so it opens the door to
suspicion," Geiger said. "It's comforting to say we need 'sound
science,' because we do. But it's another thing to work through it
when there are all these reports out there and nobody is looking at
the whole picture."
Comment by poster: It appears from this article that the planting of
hairs in the study was actually a double-blind control, i.e. a method
of insuring whether the study would stand up to peer-reviewed
standards. For that reason hairs of known lynx were included in the
study, to see whether they would be detected as the frauds they were.
They frauds were detected, but the results were published before the
author's commented on the results _or methodology_. It's interesting
to see who else was duped in this by reading the early posts. It is
also revealing to see what axes were being sharpened...
I know mine got a little dulled by it. But no more than anyone else's.
Daniel B. Wheeler
www.oregonwhitetruffles.com
mhagen
one hung up in review for a full year now, getting ever finer and finer
dents pounded out of it. The hazard - or maybe the price paid for a good
one - is that by adding the expense of a pilot study, one can find the
defects and one can usually learn enough about the subject to make the final
study better. The pilot study isn't wasted expense. I've used the same
method to obtain pre-cruise estimates. If your pre samples are done
according to a uniform procedure, you can use them as extra sample points on
the larger inventory.
snip