New method may provide better wolf estimates

Monitoring wolf populations has never been easy and it’s gotten more difficult in recent years with an expanding wolf population in Montana, but now there’s interest in putting a new method to work, possibly as soon as this year.

Montana Fish, Wildlife and Parks officials have been working with the Montana Cooperative Wildlife Research Unit at the University of Montana for the last few years to develop a “patch occupancy modeling” for wolves, a method that relies heavily on an annual hunter survey.

Currently, population estimates depends on the work of state wolf management specialists such as Kent Laudon, who covers an area of about 10,000 square miles in Northwest Montana.

That work involves trapping and fitting wolves with radio collars and tracking those collars to monitor packs. Visual sightings result in annual minimum population estimates that no longer accurately reflect the actual number of wolves on the landscape.

“We’ve always had this minimum count but it’s kind of outlived its utility,” said Laudon, who has been monitoring wolves in Northwest Montana since 2004. “When there weren’t so many packs on the landscape, it was practical at that time.”

But as of last spring, there were 39 wolf packs in Northwest Montana, and Laudon can no longer keep up with all of them, much less keep track of pack reproduction and mortality data.

“Really it just gets beyond human effort thresholds,” Laudon said.

Enter patch occupancy modeling, a statistical population monitoring method that was first applied to amphibians. In this case, it would be used to determine the number of wolves, packs and breeding pairs.

“It’s relatively new and it’s based on some simple notions,” explained Mike Mitchell, leader of the cooperative research unit. “If you go looking for animals in an area, you’re not always going to see them even if they are there.”

Occupancy modeling involves estimating the probability of detection and compensating for animals that may be missed by field observations.

“Wolves are helpful because they are highly territorial,” Mitchell said. Pack territories of about 250 square miles have been applied as patches in the model.

Laudon said the model also accounts for prey bases and terrain, and it creates probabilities for areas to be colonized by wolves as well as probabilities for occupied areas to become unoccupied.

The observations of Montana hunters also are plugged into the model.

Montana Fish, Wildlife and Parks has long conducted telephone surveys after every hunting season, but in 2007 the survey included three additional questions: Did you see a wolf or wolves? How many? And where was it seen?

In 2008 and 2009, about 80,000 hunters were contacted, a healthy sample for statistical purposes, and each year there were about 2,400 wolf sightings statewide, Mitchell said, stressing that the number includes repeat sightings where the same animals are seen by more than one hunter.

Mitchell contends that hunter statistics are invaluable.

“This is a level of observation that there’s just no way we could replicate in any other way on the landscape,” he added.

The data collected by Laudon and his counterparts also is an important component that may not always be available. The whole modeling project was started partly in anticipation that wolf delisting would eventually lead to less money to maintain the current level of monitoring.

“You need some field validation of what the model is telling you ... while we still have pretty darned detailed information, we’re using that to calibrate our model to make sure it is as representative of reality as we can get,” Mitchell said.

Laudon and others involved with wolf management have long stressed that the minimum count does not account for all the wolves on the landscape, and he anticipates the model will provide a more accurate picture of the population.

“It doesn’t necessarily mean we are abandoning the minimum count,” Laudon said. “The hope is that it will become better, estimating closer to an actual population rather than a minimum.”

Mitchell said the model does reflect more wolves than the most recent minimum count, but he declined to provide numbers, because the work was recently submitted for a peer-reviewed publication and that process is not complete.

Once published, he anticipates it will be influential in the arena of the Endangered Species Act, where “the best available science” is an important standard.

“Whenever something comes out in peer-review literature that goes into the pool of available science,” Mitchell said, adding that U.S. Fish and Wildlife Service officials have supported the modeling work.

If it’s implemented in Montana, Laudon expects he and his colleagues will see some practical benefits, starting with less radio collaring work.

“Instead of me being married to a trapline, I can be more flexible and mobile,” he said. “It would give me more time to search for new packs.”

Fewer radio collars means less expensive monitoring flights. And he pointed out that patch occupancy modeling also is expected to improve the state’s ability to set appropriate hunting quotas when state management over wolves is restored.

Quotas vary from one wolf management unit to the next, and from one year to the next, depending on how populations are faring in different parts of the state.

Reporter Jim Mann may be reached at 758-4407 or by email at jmann@dailyinterlake.com.