Our analysis identified 13 corridors connecting our core areas of ideal wolf habitat. Although the corridors avoided roads whenever possible, crossings were unavoidable if the core areas were to be linked; this reality threatens wolf use of corridors in their dispersion. Luckily our corridors generally narrowed at these crossings (Figure 8, a). We therefore recommend the construction of wildlife overpasses at the locations illustrated below (Figure 8, b). Wolves are shown to utilize wildlife overpasses at a much higher rate than wildlife underpasses (Mysłajek, Olkowska, Wronka-Tomulewicz, M. et al., 2020). The construction of these structures will enable the movement of wolves across the landscape, increasing the resilience of wolves to habitat fragmentation and destruction.
Figure 8. (a) The narrowing of connectivity corridors when crossing major roads (roads in thin yellow). (b) Suggested locations for the construction of wildlife overpasses (blue dots).
Unfortunately, the extreme prevalence of logging across Vancouver island means that corridors and core areas still often occur in areas that continue to be heavily logged (Figure 9). Since deer and elk require forest cover (Quayle & Brunt, 2003; Reid, 2003), addressing rampant logging on Vancouver Island is the most effective way to maintain high carrying capacities across the landscape. Wolf populations are strongly influenced by the presence (or absence) of these preferred prey species (Darimont &Paquet, 2000); therefore protecting forested habitat greatly improves the chances of survival for wolves. This is supported by the fact that our identified corridors to travel through and converge around existing protected areas where logging is minimized (Figure 10); core areas also tend to overlap with existing protected areas) (Figure 11).
Figure 9. Despite the fact that the most suitable habitat and connectivity corridors were identified, these areas are still marred by heavy logging. Connectivity corridors and core areas tend to lie where FSR density is lowest. White indicates more heavily logged areas. Corridors are in green, core areas are outlined in yellow.
Figure 10. Corridors tend to favour existing protected areas (green).
Figure 11. Core areas (purple) often overlap existing protected areas (green).
In order to improve both habitat retention and habitat connectivity for Vancouver Island wolves, we recommend the prioritization of management decisions that protect core areas and connectivity corridors that do not already overlap with protected areas from clearcut logging.
These new protected areas would provide secure links between existing protected areas on Vancouver Island, ensuring that ideal habitat is available for use by wolf packs and protecting overall ecosystem function.
Sources of Error and Limitations
The following are some of many possible sources of error in the execution of this project:
- Conversion of data (from polygon to raster or vice versa, edges of polygons and lines diminish in accuracy).
- Only very general overall range data was available to guide the bounds of our analysis, this may not fully represent the range of Vancouver Island wolves.
- Shrinking cell size (resolution diminishes as cell size is increased, in our case converting from cell size 100 to 400).
- We assumed a linear distribution of the factors, when this may not be the case (i.e. as stand age increases, preference may also steadily increase).
- NoData values were given Boolean values, in order to use the Weighted Overlay and Sum tools (i.e., an area where no forest data was collected was assumed to be of least preference, when in fact it is unknown).
- By making the decision to not put a limitation on the maximum size of the core areas (there was no data available), ArcGIS software (‘Locate Regions’) arbitrarily decided the cutoff.
- By limiting our core areas to 10 (a limitation of the ‘Linkage Mapper’ tool) we may have excluded suitable wolf habitat from our connectivity analysis.
- The wolf habitat suitability analysis included a pseudo-duplication of input factors. While we needed to consider forest age data in the creation of our prey habitat suitability raster, we also needed to consider FSR density in our wolf suitability raster. FSR and forest age are likely correlated. This decision was unavoidable and was based on the methods of a reputable study (Cleland, 2013).
- Available forestry data cut off the Northern tip of Vancouver Island, a remote and protected forest area (Cape Scott Provincial Park) which likely would have increased the suitability of the surrounding area.
- Logging operations are in constant flux across Vancouver Island. This analysis is likely already out of date.
This project is limited in the sense that much of the data needed to be assumed or extrapolated, and results will vary highly on the choices of the ArcGIS spatial analyst. Additionally, weights concerning the core habitat areas, resistance areas, and corridor locations were not calculated in collaboration with a species expert or local biologist. As stated by the creators of Linkage Mapper, “connectivity modeling involves a great deal of research, data compilation, GIS analyses, and careful interpretation of results” (McRae & Kavanagh, 2011).