Counting wild animals is no easy task

Four of us drove a twisted, grueling, rainy 26 miles from Corvallis to Alsea over Mary’s Peak Pass to work, crammed like sardines into an old Datsun pickup. Yes, four of us, every single day! (See my December column on the consequences of exceeding carrying capacity.)

To ignore the physical agony, we tried to keep our minds focused on other problems. Early one boring morning Bob broke the silence: "People that drink Billy Beer must be a bunch of slobs." John said, "That’s an interesting conclusion. How do you know?" Bob said he’d counted 1,000 Billy cans along the roadside in the last mile and only one Coors. "Billy drinkers are littering slobs."

Andy, the engineer in the group, said, "Your census does not prove your hypothesis. It is entirely possible that Billy drinkers have weaker arms than Coors drinkers, who can get their cans into the bushes." We all muttered in acknowledgment of the wisdom in Andy’s alternative hypothesis. But there was no way we could validate either hypothesis from the Datsun.

One last hypothesis was that Billy Beer was the only beer sold in the area, the only discriminatory factor being money. We confidently rejected this hypothesis based on sampling, the cornerstone of good science. But we still had no data on arm strength or attitude, nor could we prove that our census was proportional to sales.

How do we know how many deer, lynx, cougar, birds, fish, or bigfeet (plural for bigfoot?) live in the Bear Creek watershed? Censuses.

Advertisement

Typically, we count what we can see. Biologists use cameras, tracking, spotlighting and hunter success, but warn that "failing to detect" does not necessarily mean "absent." Sounds a little like the reliability of counting beer cans. Many conclusions are possible but validation is impossible.

Birds are netted, counted and banded (not appropriate for most humans) and the history of capture increases validity. The long-term, internationally coordinated Audubon Christmas Bird Count is an excellent example (see the Klamath Bird Observatory home page, www.klamathbird.org, to participate).

For humans, the "household" provides control. A census form is sent to each household. Still there are problems. Remember the controversy over the numbers of homeless following the last census?

Fish are forced through counting stations, like airport security. But not all streams have stations and some fish may not make it that far. The point is that validity is expected to increase with control, history, standard methods, standardized times and the extent of the survey.

Most ecologists are used to dealing with "messy data" and uncertainty. (There is now even a branch of statistics dealing with the "analysis of messy data.") Most of us feel comfortable using relative probabilities.

Even after 30 years of study, we do not know the actual number of northern spotted owls, but it is probable our estimates are close and we can detect trends. The recent discovery of the ivory- billed woodpecker, thought to be extinct, is surprising but attests to the difficulty of counting populations (see www.npr.org/templates/story/story.php?storyId=4622633).

Bigfoot? I doubt they exist, but I am not certain. Remember, "failing to detect" does not necessarily mean "absent."

I have spent much of my life in the woods and have seen fewer than half a dozen cougars. A more agile and intelligent creature living in rough terrain could evade even the CIA for years.

Tom Atzet is a retired Forest Service ecologist living in Merlin. He can be reached at P.O. Box 1226, Merlin, OR 97532.