More on the effects of hunting on wolf populations

I’ve heard back from some colleagues concerning the new paper by Creel and Rotella, a meta-analysis of wolf mortality, that I described in a recent post. The respondents felt that the Creel and Rotella meta-analysis is generally sound and the conclusions that harvest mortality is not (or at least is only partially) compensatory to natural mortality has significant implications for state management, especially management that assumes that hunting of wolves can be used to reduce livestock conflicts. Both Creel and Rotella (2010) and a new meta-analysis paper by Dennis Murray et al. (below) reject the compensatory mortality hypothesis (which, as described in the previous post, proposes that increase in mortality caused by hunting is compensated for by a decrease in natural mortality), and conclude that hunting and lethal control (e.g., in response to depredation) kill different wolves. This makes sense because hunters kill wolves mostly on public lands in the fall where there are no cattle, and wolves kill livestock in spring/summer (and also early fall) on private lands. Both studies generally agree that a human-associated mortality (hunting, lethal control) rate of around 22% is the highest that the Northern Rocky Mountain (NRM) wolf population can sustain without population decline, which is a lower rate than has been previously estimated for other wolf populations.
Some readers have pointed out two (relatively minor) problems with the conclusions of Creel and Rotella:
1) In the calculation of the hunting mortality as a proportion of the wolf population for 2010, Creel divided harvest the next fall by total wolves in Jan 2009, but this ignores reproduction and pup recruitment so is an overestimate of hunting mortality as a proportion of the population.
2) Other researchers analyzing a similar data set, have, unlike Creel and Rotella, found evidence for weak density dependence in wolf numbers, with the NRM wolf population stabilizing at a carrying capacity of around 2500 wolves.
Coincidentally, in a 2006 paper by myself and colleagues published in Bioscience, we used a very different type of model to estimate potential wolf population levels for the western US, and found the NRM could hold about 2800 wolves under current conditions. I would caution, however, that for a species such as the wolf, ‘carrying capacity’ is strongly affected by the level of tolerance by humans. The region can be divided into three general zones. In the first, which encompasses areas of low road access and human population, wolves are likely to persist under a range of management. In the second zone, which has high levels of human settlement, it would be difficult for packs to persist under most plausible scenarios. However, the number of wolves that inhabit an intermediate zone of lightly-settled lands is highly dependent on wolf management policy. In the NRM, this intermediate zone lies primarily in western Montana. Any exact estimates of ‘carrying capacity’ are thus less useful than a comparison of potential wolf population levels under a range of policy options.

(Murray et al. 2010 – full article)