The Montana Cooperative Wildlife Research Unit (CRU) has been providing science support to Montana Fish, Wildlife, and Parks (MFWP), the Nez Perce Tribe, and Idaho Fish and Game on gray wolf populations since 2006. Most of this research focuses on the monitoring of wolves over large spatial scales. Over the past 10 years, the growing wolf population of the northern Rockies exceeds the capacity of traditional monitoring methods (capturing and collaring wolves, radio-tracking) to provide reliable estimates. Abundant, wide-ranging carnivores that exist at low densities are notoriously difficult to monitor, and no means of generating reliable estimates of population size or breeding pairs for wolves on a statewide scale existed. To provide more accurate and reliable statewide estimates of the number of packs and abundance, the Montana CRU adapted occupancy modeling to the unique application of monitoring wolves as well as an estimator for the number of breeding pairs that is equally robust.

Occupancy modeling is a statistical framework that uses observations of presence and absence to infer whether animals occupy "patches" on a landscape. The CRU defined patches for wolves as a grid of squares overlaid across Montana and Idaho, each the size of average wolf territories in each state. With repeated sampling of many patches (e.g., hunter observations, genetic sampling, and telemetry locations), researchers can determine which sampled patches are occupied and estimate the likelihood that unsampled patches are occupied. Because the squares within the grid approximate territory size of wolves, it is assumed that occupied patches contain distinct wolf packs. Summing the number of occupied patches estimates the number of wolf packs in a state. Multiplying the number of packs by average pack size estimates the number of wolves in a state.

Estimating the number of breeding pairs of wolves (defined as one adult male, one adult female, and two pups at the end of the year) is both necessary and difficult. The detailed data collected on wolves prior to their delisting as endangered species was used to estimate the likelihood that packs of different sizes contained breeding pairs. Results show that small packs have lower probabilities of containing a breeding pair than larger packs, and these probabilities differ geographically according to the intensity of management actions and the growth rate for the wolf population. Knowing these probabilities allows managers to estimate the number of breeding pairs based on knowledge of pack size under a variety of conditions.

This approach to monitoring abundance based on occupancy estimation is being integrated into post-delisting wolf management strategies for both Montana and Idaho. Since delisting, the CRU has worked closely with MFWP to adapt estimation procedures to the changing circumstances of wolf management, where harvest is now used to manage wolf population size. Much uncertainty about the effects of harvest on wolf populations exists, including what dynamics contribute most to population growth for wolves and how those dynamics are affected by harvest. Without understanding these processes, it is difficult to design harvest programs that can achieve desired population size ? one that minimizes impacts of wolves on ungulate populations and livestock but maintains a viable and connected wolf population.

The CRU's ongoing work with MFWP supports an adaptive harvest management program for wolves. An adaptive harvest management program works to reduce critical uncertainties about population dynamics of game animals and how they are affected by management through the repeated implementation of harvest and monitoring of its effects. In other words, managers design an annual harvest plan to achieve specific goals (e.g., a certain population size, reduction of livestock depredation, etc.). After implementation of that plan, monitoring is conducted to evaluate how well goals were achieved. Over multiple iterations of this process, the extent to which management produced desired results provides important information for better understanding and refining management decisions.

This approach allows managers to learn about the effects of alternative harvest approaches on actual achievement of management goals. Importantly, it will improve the ability of MFWP to set harvests for wolves that will ensure an appropriate balance between the abundance of ungulates, the maintenance of livestock operations, and a viable and connected wolf population.

Each month, the ONB features articles from Cooperative Fish and Wildlife Research Units across the country. Working with key cooperators, including WMI, Units are leading exciting, new fish and wildlife research projects that we believe our readers will appreciate reading about. This month's article was written by Mike Mitchell, PhD, mike.mitchell@umontana.edu with the Montana Cooperative Wildlife Research Unit at the University of Montana.