Simulation Modeling of Population Expansion for Introduced Mountain Goats in the Olympic Mountains of Washington State

Mountain goats (Oreamnos americanus) are not native to the Olympic Peninsula as they are to other regions of Washington State. A total of eleven or twelve animals were translocated from Alaska and British Columbia between 1925 and 1929 then released in the foothills of Mount Storm King. By 1970 these founding goats had colonized the entire Olympic range and concerns about the management of this introduced species developed as damage to alpine soil and vegetation was noted. An aerial census of the Olympic range conducted in July 1983 estimated the mountain goat population at 1,175 (95% CI 840 – 1510). A series of removals reduced the population to 389 (95% CI 181 – 597) goats by 1990, with a period of stasis occurring during the following decade. The most recent two aerial surveys (2011 and 2016) indicate positive growth, and a variety of efforts to mitigate damage to fragile alpine ecosystems are again under consideration. I parameterized an existing population model, CDPOP, for use with mountain goats. CDPOP is a simulation program that uses individual-based movement (including dispersal), reproduction, and mortality to predict the influence of landscape heterogeneity on population dynamics and genetic exchange. Population parameters for the model were derived from published literature. I successfully calibrated the model and simulated the population trajectory for Olympic mountain goats from establishment through the 1983 census. Modeled population dispersal closely tracked anecdotal reports of dispersal. However, observed heterozygosity for the modeled population did not align with previous research. I suspect genetic diversity for the true founding goats was not as great as that of the individuals used to initialize the model. Sensitivity analyses showed that changes in annual reproductive rate had the greatest influence on population trajectories, followed by juvenile mortality and v adult female mortality, respectively. These findings differ from those in two related studies, likely due to the early primiparity within the modeled population. I validated the model by simulating the period from 1990 to 2016. The modeled population showed that approximately 75% to 80% of the total animals removed during the 1980’s needed to be female in order for the observed population stasis to occur. Finally, I discuss avenues for future model development and applications. This model could be utilized to inform current management decisions regarding the impact of removals from the Olympic mountain goat population and proposals to use these animals to augment reduced native populations in the Cascade Mountain Range