Winter habitat selection of lynx (Lynx canadensis) in northern Washington

Benjamin Thomas. Maletzke

I investigated habitat selection of lynx on a 211 km2 portion of the Okanogan National Forest in north-central Washington. I completed two winter field seasons, 2002- 2003 and 2003-2004, snow-tracking lynx for a total of 155 km using precise-positioning Global Positioning Systems to record movements and behaviors of lynx. A Geographic Information Systems (GIS) analyses of these movements was used to determine winter habitat selection on a study area comprised of Engelmann spruce/subalpine fir and Douglas-fir/ponderosa pine forest with very little lodgepole pine. The area was fragmented by recent wildfire and timber harvest. Habitat variables included forest vegetation type, overstory canopy cover, understory cover, slope, aspect, and elevation. I used t-tests and selection ratios (S) to compare differences in proportions of lynx use trails versus random availability trails. Lynx selected P < 0.05 for Engelmann spruce/subalpine fir forest types (S=1.66), canopy cover of 11-39% (S=1.29), and understory of 11-39% (S=1.18), but avoided forest openings (S=0.28), recent burns (S=0.38), Douglas-fir/ponderosa pine (S=0.41), canopy cover 30° (S=0.62). From these results, I determined significant variables (P< 0.05 for Engelmann spruce/subalpine fir forest types (S=1.66), canopy cover of 11-39% (S=1.29), and understory of 11-39% (S=1.18), but avoided forest openings (S=0.28), recent burns (S=0.38), Douglas-fir/ponderosa pine (S=0.41), canopy cover 30° (S=0.62). From these results, I determined significant variables (P< 0.05) for inclusion in a logistic regression model of vegetative and physiographic variables important to lynx. Logistic Regression indicated lynx selected (P< 0.05) for Engelmann spruce/subalpine fir forest types, slopes< 30°, elevations of 1524 m – 1828 m, and canopy cover of 11-39%. My results suggest that lynx do not always require and select lodgepole pine forests, but that Engelmann spruce/subalpine fir forest types are important to and selected by lynx. The logistic regression model I developed will enable forest managers to predict the relative probability of lynx presence for most areas of Washington and the North Cascades region, which share similar characteristics of vegetation and elevation found in the Black Pine Basin study area.

Winter habitat selection of lynx (Lynx canadensis) in northern Washington

Files and links (1)

Abstract

Metrics

Details