- Forested landscapes in the Pacific Northwest have changed drastically in the last Century with large-scale wildfires, and increased extent and frequency of commercial timber harvests. Thus, landscapes once dominated by older forest cover types became a matrix of younger forests with patches of older forests dispersed throughout. These younger forests lack the structural, functional and compositional characteristics of older forests that support a more diverse array of fauna and flora ecosystems. Current federal land management policies seek to restore historic forest compositions and conserve species associated with older forests. However, current forest management practices that meet long-term restoration goals may be detrimental in the short-term, especially for low-vagility wildlife species.
The red tree vole (Arborimus longicaudus) is a highly specialized small mammal, endemic to the forests of western Oregon and northwest California. Tree voles nest in tree canopies and forage on conifer needles. They are important prey for the northern spotted owl (Strix occidentalis caurina) and are currently a candidate for the endangered species list. Tree voles have been reported as associated with older forests (> 120 years), while younger forests (< 80 years) are categorized as less-suitable habitat. Understanding the factors limiting tree vole occupancy and activity in younger forests is therefore important, especially when most forest management practices, including basal area removal (thinning), are implemented in forests less than 80 years old. Under certain conditions augmenting younger forest canopies with additional structure (i.e. artificial nest platforms) can increase tree vole occupancy, but limiting factors such as predation risks are less understood.
In younger forests, structural characteristics and predation risks not present in older forests may present limitations to occupancy and survival of tree voles. To test these hypotheses, I quantified detections of tree voles and their predators, and estimated apparent tree vole survival at artificial nest platforms in adjacent older and younger forest stands in the central Oregon Coast Range. I used multi-season occupancy models to quantify differences in tree vole occupancy patterns and examine tree vole response to increased predator visitations. Additionally, I used Cormack-Jolly-Seber models to estimate apparent tree vole survival at nest platforms in adjacent older and younger forests. From August 2017 – December 2018 I observed more sporadic tree vole use of older forest nest platforms (weekly extinction rate, older forest = 0.27, 95% CI = (0.12 to 0.51), younger forest = 0.21 (0.09 to 0.44), β = 0.37 (0.05 to 0.68); weekly colonization rate, older forest = 0.16 (0.10 to 0.31), younger forest = 0.13 (0.08 to 0.25), β = 0.42 (0.15 to 0.70)). I observed increased tree vole extinction rates as weasel (tree vole extinction β = 0.47 (0.11 to 0.83)) and squirrel (tree vole extinction β = 0.06, ( -0.002 to 0.11)) detections increased at nest platforms in both forest types.
Further, there were proportionately more younger forest nest platforms with squirrel (younger forest = 0.35 ± 0.19, older forest = 0.07 ± 0.05) and avian predator (younger forest = 0.08 ± 0.07, older forest = 0.03 ± 0.003) detections. Annual apparent survival estimates (phi = 0.03) did not differ between forest types, yet my sample size was very small (n = 12) and conclusions are therefore limited. My results support previous reports of nest structure availability as a major limiting factor for tree vole occupancy in younger forests. Predators were detected at higher proportions in younger forests while higher detections of weasels and squirrels were related to increased tree vole extinctions at nest platforms. Though nest platforms alleviate the structural limitations of younger forests, increased predation risks may discourage tree vole persistence, even when additional structure is available.
In the Pacific Northwest, thinning is used to promote the structural development of dense regularly-spaced younger forests over time; however, there may be a short-term (< 5 years) reduction in inter-tree connections leading to increased tree vole extinctions at nest structures and reduced subsequent colonization of nest structures. To test these hypotheses, I removed inter-tree connections from around 26 trees with nest platforms at 2 younger forest sites in the Oregon Coast Range. Tree vole occupancy patterns were quantified using multi-season occupancy models and tree vole detections at nest platforms. During March 2018 – December 2018, after all treatments were implemented, I observed no difference in tree vole extinction probabilities at isolated vs. interconnected trees. However, tree voles at isolated trees were detected with greater consistency than tree voles at interconnected trees (tree vole detectability at isolated trees, β = 0.12 (0.01 to 0.24). Additionally, once an extinction event occurred at a nest platform, tree voles were less likely to recolonize an isolated tree compared to an interconnected tree (tree vole colonization rate at isolated trees = 0.035 (0.025 to 0.048), interconnected trees = .052 (0.040 to 0.069), β = - 0.42, (- 0.82 to - 0.03)). My results support the hypothesis that a reduction in inter-tree connections reduce tree vole colonization of isolated trees in younger forests. Tree voles appeared to be initially resilient to localized disturbances around nest structures, but subsequent recolonization of residual nest structures was reduced at trees lacking inter-tree connections. Non-uniform thinning techniques that promote residual tree structure and canopy connectiveness in the short-term, may better mitigate the immediate negative effects of thinning in younger forests.