Release Locations for Grizzly Bears in the North Cascades Ecosystem
And Potential Areas of Impact One Year After Release
Grizzly bears (Ursus arctos horribilis) can be found in much of the northern hemisphere of the world. However, they presently occupy a much smaller territory in North America than they once roamed, limited to only 2% of their former range in the contiguous United States (Singleton et al. 2004). The International Union for Conservation of Nature lists the remnant population in the North Cascades Ecosystem (NCE) of Washington State as critically endangered with a complete degree of isolation (McLellan et al. 2016). In a study using DNA from hair samples, the North Cascades Ecosystem (NCE) was found to have only 0.15 bears per square kilometer and a total population of just six bears (Romain-Bondi et al. 2004). In January 2017, a draft environmental impact statement was released for a plan to restore the grizzly bear population in the NCE. The NCE is particularly suited to grizzly bear restoration not only because it is part of their historic range, but also because land management practices have maintained the habitat as a grizzly bear recovery zone for decades. Additionally, most of the land in the NCE is federally owned as national forests and the North Cascades National Park. Altogether, this ecosystem extends over 25,000 square kilometers with abundant wilderness. This restoration plan would provide many benefits to the region, namely balancing the ecosystem, avoiding permanent population loss, and providing residents and visitors with the opportunity to observe grizzly bears in the wild, which is a potential boon to the economy (Swanson et al. 1994). In my geospatial analysis, I wanted to identify ten release sites for grizzly bear transplants, taking into account factors for high quality bear habitat as well as a low likelihood of negative human interaction. After determining these locations, I considered the potential impact of the released bears on hiking trails and hunting areas, as well as any human settlements affected. My final priority is to begin raising bear awareness in Washington State in preparation for the population restoration.
North Cascades Ecosystem within Washington State
Parameters
In order to determine where the best quality grizzly bear habitat is, it is important to set the parameters for the definition. The factors considered in this analysis were as follows:
​
Wilderness: Plentiful in the NCE
Elevation: Optimal between 1000 and 6000 feet
Forest Edges: Optimal <1000 meters, marginal within 4000 meters, poor >4000 meters
Fish: Rivers where fish are known or assumed to be present. Optimal <1000 meters, marginal within 3000 meters, poor >3000 meters
Habitat Type: Optimal in forests, grasslands, and shrublands, poor in others
Avalanche Chutes: Optimal <1000 meters, marginal within 3000 meters, poor >3000 meters
Away from Roads: Optimal >500 meters from primary or secondary roads
Away from Humans: Optimal >1500 meters from human populations
Methods
Analysis was split into four phases.
​
Phase One – Optimal grizzly bear habitat using environmental factors:
Fish/rivers and avalanche chutes were buffered according to the set parameters, then rasterized along with the sorted optimal/poor habitat types. All environmental factors were reclassified (elevation 0,1; distance to forest edge 1,2,3; distance to fish 1,2,3; distance to avalanche chutes 1,2,3; suitable habitat type 0,1). These were added and multiplied in the raster calculator resulting in values 0-9 (higher being better habitat), then reclassified into optimal (3), marginal (2), and poor (1) habitat.
Phase Two – Optimal grizzly bear habitat avoiding human influence:
Census blocks and total population for each in the NCE were used to interpolate population density throughout the region with the IDW method. Populations >100 people/km² were buffered to the 1500-meter sphere of influence, then reconverted into raster, which was reclassified as 0 and 1. Roads were buffered to 500 meters, converted to raster, and also reclassified as 0 and 1. These were multiplied by the 0-9 valued grizzly habitat raster, then reclassified into optimal (3), marginal (2), and poor (1) habitat.
Phase Three – Accessibility and Release Sites:
Service roads, under the categories of 4-wheel drive, unimproved, and light duty roads, were selected from the state roads data, along with either the ‘active’ or ‘unknown’ attribute parameter, as abandoned service roads would not be accessible and using only active roads eliminated too many accessibility options. These roads were exported as a new layer and then overlaid on the grizzly habitat with human avoidance polygons to use in creating a new point feature class. Ten points were plotted for grizzly release sites on the service roads that also ran through class 3 (optimal) grizzly habitat. An Esri imagery basemap was used to verify the roads were correct.
​
Phase Four – Areas of Impact:
The release point features were buffered to 30 kilometers and exported into a new polygon layer. These polygons were then used to select other layers of human recreation and population located within the grizzly buffer zone. These layers included game management units (hunting), the North Cascades National Park, campgrounds in the park, and populated places. The selected features were exported into new, separate feature classes in order to properly visualize the extent of potential impact.
Results and Discussion
Phases One and Two produced expected results, with optimal bear habitat following the Cascades mountain range, and larger human influence on the outskirts and to the south of the NCE.
Ten points were successfully located for Phase Three. They were dispersed toward the north in order to avoid the most human influence as possible, in addition to being closer to the North Cascades National Park and the British Columbia border, and thus closer to other bear populations.
Phase Four found game management units, campgrounds, trails, and populated places within the grizzly bear impact zones. The release site numbers on the table correspond to those on the release site and impact maps. The trails data did not contain distinguishing fields, thus the only method for conveying impacted trails is through the map visual. The most ideal release sites regarding the number of impacted human recreational areas and settlements are 1 and 3, running into fewer than 10 settlements as well as lower numbers of campgrounds and GMUs. They also cross well into the national park, which is free from all hunting activities and thus excellent land for bears.
