Description
AN ABSTRACT OF THE THESIS OF
Katelyn A. Courtot for the degree of Master of Science in Biology
presented on
Title: Co-Occurrence and Habitat Use Overlap of Snowshoe Hares (Lepus
americanus) and American Martens (Martes americana) in the Boreal – Temperate
Ecotone of the Northeastern U.S.
Abstract approved:
___________
Leonard R. Reitsma, Ph.D.
Managing forests for wildlife with different habitat preferences in addition to
providing economic-based outcomes related to forest products poses a unique set of
challenges for natural resource professionals. Snowshoe hares (Lepus americanus) are
a keystone species influencing community dynamics, and American martens (Martes
americana) are an umbrella species indicating ecosystem health. Hares prefer early successional forest habitat in contrast to martens, which are typically associated with
older forests. To evaluate the shared habitat preferences of both species, we collected
forest structure data between May and November 2020 in 14 forested stands located in
the Nulhegan Basin (NB), Vermont (n = 7) and the White Mountain National Forest
(WMNF), New Hampshire (n = 7). My objective was to test the hypothesis that the
co-occurrence of hares and martens is contingent on habitat availability at several
scales. Specifically, I predicted that patches of early-successional forest within a
matrix of mid- to late-successional forest allowed for habitat overlap of hares and
martens. We measured basal area, canopy closure, diameter at breast height and
counted stems at 45-50 plots within each of the 14 stands. We also measured coarse
woody debris along three randomly selected transects in each stand. We compared
these data with 2 years (2016 and 2017) of live-trapping data to determine what forest
structure variables were associated with the habitat use overlap of hares and martens.
The NB has a robust population of hares with few martens and the opposite pattern is
observed in the WMNF, except that hares are relatively common, although at lower
abundance in these forests. First, we fit a univariate, two-level categorical model (NB
vs WMNF) to evaluate differences in second-order habitat selection. To evaluate
fourth-order habitat selection, we fit 7 additive occupancy models for each species
using forest structure variables and live-capture data. To further evaluate fourth-order
habitat selection, we used three Principal Component Analysis (PCA) models to assess
habitat use overlap. Results from the additive models suggest that hares selected areas
with high stems counts and against areas with high basal area. Martens selected
against areas with high stem counts, although basal area had only a weak effect.
Visual-exploratory analysis using PCA suggested that deciduous stems accounted for the habitat use overlap between hares and martens at a finer spatial scale. The WMNF
experiences moderate amounts of disturbance that create small patches, typically
resulting from windthrow events. This study suggests that these early-successional
patches of forest within a later-successional landscape allow for the co-occurrence of
both species, with deciduous stems driving the habitat use overlap. Deciduous stems
have three phases that contribute to the habitat use overlap of hares and martens;
seedlings/saplings, mature trees and coarse woody debris. The presence of keystone
and umbrella species within the same community may increase biodiversity, and is
consistent with the Intermediate Disturbance Hypothesis, however, more empirical
support is needed to correctly guide and solidify future management.
