Assessment of Fisher Habitat in Washington State: Tier 1 and Tier 2 Final Report

This document reviews activities involved with the assessment of habitat for the possible reintroduction of fishers (Martes pennanti) to Washington State. It details the refinement of previous Tier 1 analyses that used Interagency Vegetation Mapping Project (IVMP) data to model ‘suitable habitat’, which only included denning and resting habitat. Although fishers use other types of habitat, den and rest sites are the most important life requisites for a fisher population. This report also describes the analyses conducted for the Tier 2 effort that further defines suitable habitat for fishers.

The Fisher Science Team recommended that certain aspects of the Tier 1 effort be refined to possibly improve results that would be used as a base for Tier 2 activities. There was an overestimation of suitable habitat within the East Cascades using 10 inches and greater Quadratic Mean Diameter (QMD) in the habitat model. The alternative of using 20 inch and greater QMD in the habitat model for the East Cascades resulted in a very sparse distribution of suitable habitat, so late seral forests were interpreted from year 2000 Landsat 7 data. Also, fixed upper elevation boundaries in the Cascades in particular were not be sensitive to ecological variations throughout this mountain range. The upper limit of the silver fir (Abies amabilis) zone was chosen as the elevation boundary for suitable fisher habitat, and this improved ecological sensitivity throughout the study area, especially for the southern Cascades. Furthermore, the 1000 km2 and 1500 km2 ‘moving window’ areas representing the approximate area of a small fisher population resulted in too much generalization of the concentration areas of suitable habitat. A concentration area analysis using an average fisher home range window of 25 km2 demonstrated a greater sensitivity to local distributions of suitable habitat. Overall, these refinements improved the overall characterization of suitable habitat analyzed in the Tier 1 effort.

About 31% of the extent of home ranges of female and male fishers in a southern Oregon population was identified as suitable habitat. This information was used to generate concentration areas in Washington that represented suitable habitat concentrations of less than 31%, or greater than or equal to 31%. Interestingly, this population did not use large and contiguous areas of suitable habitat north of its observed location. The largest concentration area in the Olympics had twice the combined area of the three largest concentration areas in the Cascades. The West Cascades had numerous concentration areas, but most of these were relatively small and linear in shape. However, investigation of the North Cascades indicated that corridors of mid and late seral forest appear to exist between currently identified concentration and connectivity areas if you reduce the importance of the silver fir zone boundary.

The potential for movement of fishers between patches of suitable habitat is an important factor to consider, so an analysis was conducted that demonstrated the connectivity between the patches of suitable habitat throughout the study area. The largest connectivity area in the Olympics had twice the combined area of the three largest connectivity areas in the Cascades. The concentration and connectivity areas have a synergistic effect in helping to identify potential fisher reintroduction sites.

The Fisher Science Team reviewed the results from the refinement of the Tier 1 analyses and formed a concensus to concentrate on the Olympics during the Tier 2 analysis, as this region contained the greatest amount and the most contiguous extent of suitable habitat.

It was important to validate the results of the IVMP model to provide a greater confidence in its value for identifying focus areas of fisher habitat. There was a 62% agreement overall between the IVMP suitable habitat and the Olympic National Forest (ONF) and Olympic National Park (ONP) late seral model data for the Olympics. Significant areas of disagreement between the models were investigated by analyzing the same areas within orthophotos. Overall, the IVMP model seemed conservative in identifying late seral forests throughout most of the ONP and ONF except for the western side. The models complimented one another in many areas based on orthophoto review, and this suggests that substantially more suitable habitat exists over what was indicated by just the IVMP model results.

There was a 68% agreement between the IVMP suitable habitat and the Mount Baker- Snoqualmie National Forest late seral model data, and an 81% agreement between the IVMP model and the Gifford-Pinchot National Forest (GPNF) late seral model. However, nearly 62% of this agreement within the GPNF was for areas where late seral forests were considered absent in both models. The differences between late seral forest models in the Cascades were not investigated using orthophotos.

The results of the various late seral model analyses helped to identify ‘focus areas’ within the Olympics that have the most optimal size, shape, and density of suitable habitat to possibly establish a reintroduced fisher population. However, the Fisher Science Team identified the need to describe the habitat quality of these areas to assure they could satisfy other life requisites of a population. Current Vegetation Survey (CVS) plot data for the ONF and Pacific Meridian Resources (PMR) plot data for the ONP were used to assess the habitat quality on these federal properties that contained most of the focus areas. Three definitions of old growth forests were used to determine if the selected plots that corresponded with the IVMP suitable habitat had characteristics of old growth forests that are important for fishers. This comparison indicates that the CVS and PMR plots possess forest component characteristics that are similar to old growth, but may not match these definitions in every respect. The IVMP suitable habitat model was developed to identify late seral forests, which includes both old growth and mature forests.

The Cascades are considered by members of the Fisher Science Team and the Northwest Ecosystem Alliance to be important as a corridor for connectivity of fisher populations throughout the North American Cascade Range. Local corridors between concentration areas in the North and South Cascades could be investigated further by using orthophotos and/or field visits to determine their viability to fishers. Additional analyses using stereo pairs of color orthophotos or field visits could identify more detailed ‘core areas’ of fisher habitat in the Olympics that would have the optimal forest conditions to support and maintain a fisher population. Additional information on how fishers move from one area of suitable habitat to another would be beneficial in more accurately modeling the connectivity between suitable habitat. Also, it is important to have a more sophisticated model for determining how many fishers a given focus or core area could support, and this level of modeling is currently being pursued at the Washington Department of Fish and Wildlife using a software program called PATCH.