How many bears are there?

[MarvB]

By BRODIE FARQUHAR
Star-Tribune correspondent

LANDER -- No one, not federal biologists or conservation activists, knows how many grizzly bears there are in the greater Yellowstone ecosystem.

That could be a problem, because federal and state officials have seized upon the number of 588 grizzly bears in the drive to remove the animal's protection under the Endangered Species Act.

The number used to justify delisting the grizzly was generated in 2004, based on the sightings of 49 bear sows with cubs, then processed through a much-tweaked formula to come up with 588 -- a widely mentioned number that some say “proves” the species is recovered in the Yellowstone area.

Yet in 2005, there were only 31 sow/cub sightings -- a 40 percent decline -- the lowest number of female-with-cubs-of-the-year sightings since 1997.

If that sightings number is processed exactly as it had been the year before, it would produce a population number of roughly 350 bears. While U.S. Fish and Wildlife Service and Interior Department officials have repeatedly trumpeted the 588 figure, there has been no public discussion of the 350-bear figure.

So which number is right? After all, federal biologists point out there were only 11 adult female bear deaths recorded for that 2005 time frame, not hundreds.

Critics of the grizzly bear delisting decision contend that using females with cubs of the year is an inherently unreliable method for assessing population size and trends.

Doug Honnold, lead attorney for the Bozeman, Mont., office of Earthjustice, said the Fish and Wildlife Service faces an internal, logical contradiction.

If the agency’s preferred method for assessing total population is accurate, then the Yellowstone grizzly bear population has just suffered a tremendous decline that clearly puts the population at risk of extinction in the near future, he said.

If, alternately, the decline in observations of females with cubs is not indicative of a crashing Yellowstone grizzly bear population, then the methods on which the Fish and Wildlife Service relies to justify population size, population trend, and allowable mortality levels is a failure, and a new method needs to be developed, he said.

“This is a critical issue that goes to the heart of the delisting proposal,” Honnold said.

But Chris Servheen, the Fish and Wildlife Service’s grizzly bear recovery coordinator since 1981, said the annual female-with-cub count produces an annual index, or approximation of the population -- not a census figure.

“This is a classic example of why lawyers shouldn’t interpret science,” Servheen said.

Servheen said that if the grizzly recovery program had millions of dollars, it could launch a DNA analysis program that would give higher-quality data. Chuck Schwartz, head of the Interagency Grizzly Bear Study Team, goes further, saying that in hindsight, the team should have presented a range of numbers rather than a fixed number, “because we don’t know precisely what the population is.”

Schwartz couldn’t say what that range would be, but he said 500 to 600 is a realistic, yet conservative estimate of the number of grizzlies out there.

Five hundred bears is generally regarded as the key threshold, above which bears are doing fine and below which the bears might be in trouble. In November, the Billings Gazette reported that in the previous five years, an average of 43.6 females with cubs were observed, including a high of 52 in 2002 and 49 in 2004. The last time the number was as low as 31 was in 1997.

Members of the Interagency Grizzly Bear Study Team explained at the time that the lower number could be attributed to a "bumper crop" of cubs in recent years that left many females unable to breed. Female bears have a three-year cycle: bearing a cub, then caring for a yearling, then back into breeding on the third year. Another factor could have been persistent snow at some moth feeding sites, which kept the bear count down.

Cherry picking?

Sampling error or variability is a critical issue, said Steve Stringham, an Alaska bear biologist. He said he doesn’t have a huge problem with reliance on female-with-cub sightings, saying it is “not an unreasonable method.”

What Stringham finds troublesome is when federal biologists refuse to release the data that underpins their methods and their results.

“Science is supposed to be an open process,” said Stringham, who has a doctorate in wildlife biology. Federal scientists have refused, so far, to release their raw data or computer models. “It all has to be taken on faith,” Stringham said.

Honnold noted that peer review is usually limited to fellow scientists checking procedures, methods and accuracy in results n only rarely does it drill down to the raw data level. Yet when you have critical issues, he said, that’s what is needed.

Craig Pease, a demographer on the faculty of Vermont Law School, said he’s “astonished by the failure of the delisting process to expose the underlying uncertainties.”

He charged the Fish and Wildlife Service with more than cherry picking, with federal officials publicizing the 2004 figure because it gave a reason to delist the bear, then ignoring the 2005 data because it might be difficult to explain.

“This is selective manipulation of the data and is blatantly deceptive,” Pease said. The government really should have a population estimate protocol, not an ad hoc scheme that isn’t used elsewhere, he said.

Gold standard

So what’s the alternative to using female-with-cub sightings?

Servheen and Honnold agree that the gold standard of counting bears would be a system of hair snares and DNA analysis. That’s where bears are lured into scent stations and have to step over or crawl under a strand of barbed wire. Snagged hair can be analyzed to identify bears and entire family tree structures.

The limiting factor is money, they both acknowledge.

That same limit applies to capture, mark and release techniques, which were used in 1998 and 1999 but abandoned when budgets were cut in 2000. Marked or collared bears don't get double-counted under that system, giving scientists a higher confidence in their sightings and subsequent head counts.

Science sees the world in shades of gray, given a wide range of variables and uncertainties. Yet other people want to understand that world in black-and-white terms, Schwartz said. Communicating the nuances, the uncertainties and probabilities of science “is darn hard to do."

 

[Ten Bears]

What a load of bunk. I'd go with the 500 bears estimate, and I understand why there'd be different numbers of sow and cubs from year to year.

 

[mtmiller]

...and I understand why there'd be different numbers of sow and cubs from year to year.

There are tons of variables involved. There is no perfect system.

 

[JoseCuervo]

It would be amusing for Ten Beers to share his knowledge on this....

 

[Bambistew]

Don't they count the bears as they walk out the gates at Yellowstone? I would think that all it would take is a little clicker, like the ones you can get at a golf shop...

 

[MarvB]

Bambi...those are only the bears with the pic-a-nic baskets there booboo!

 

[Ten Bears]

Jose' Queervo, what part of it would you like explained? The part about observability, or the part that sows only have cubs about every third year, and that by doing so the number of sows with cubs would vary each year.

 

[A-con]

I think the most accurate way to count the bears would be to count all the bears at a Burger King drive through, then multiply that number by the number of Burger Kings in the greater yellowstone area. Same way they count the wolves.