Limitations:
The inhibiting factors within this analysis include fragmented data due to the federal and provincial jurisdictions that are involved in managing national parks and their data. Therefore, we were unable to obtain substantial amounts of data on the Banff crossing locations and their spatial relationships with the ecological variables that are surely to be factors in the decision making process when implementing this type of infrastructure. Furthermore, when obtaining data on road networks, a key variable we were not able to include in our final analysis was the relationship of traffic densities and wildlife collisions within the inter-provincial mountain pass of Crowsnest, AB. Data for this variable was obtained via employees of the Alberta Dept. of Transportation, however the lack of organizational meta data and suitable format of attributes within the dataset completely prevented us from using the data as anything more than spatial references for road networks.
A further issue we encountered was the volume of data available, specifically for the locations of wildlife collisions throughout the province. Even though the data covered an a span of 4 years, when filtered into our desired area of interest, the data became only satisfactory in volume but was still usable in conducting our point density in the Crowsnest region, this in turn, costed a loss of precision in on our point density analysis.
The final limitation we faced in our proposal of potential locations for wildlife crossings in the Crowsnest region was the inability to confidently recommend the specific crossing type within the proposed locations. This is due to limitations in resources and intention of the analysis. A recommendation of this nature would require in-person inspection and surveying of the precise road segments in order to determine the suitability for the specific type of crossing infrastructure.
Suggestions for Further Analysis:
It is by our recommendation that the process of our analysis is extended in the areas that it is limited, specifically in a manner that incorporates a higher volume of wildlife collision data, the addition of traffic volumes and densities as a variable within the suitability model, as well as finalizing the analysis by conducting on-the-ground surveys of the road segments highlighted by our proposed locations in order to determine the most appropriate crossing type to incorporate in the already existing highway infrastructure. Additionally, as a supplement to its incorporation into the suitability model, a Generalized Linear Regression of the relationship between traffic density and wildlife collisions would further enhance this analysis. Finally, our process of determining suitable locations for potential highway crossings can be extrapolated to mountain passes throughout the rest of the province and into British Columbia in order to densify the network of wildlife crossings available in the Canadian Rockies. These crossings are effective mitigation tools for preventing damage to people and wildlife. Moreover, this infrastructure not only prevents person-wildlife incidents but also increases the connectivity and accessibility of continually fragmented habitats.