Woodland Caribou populations are under threat in Alberta and across Canada. Recovery strategies have identified linear feature management as a priority action to recover caribou habitat and populations. Given the priority of caribou conservation and the diversity of human land uses occurring within caribou ranges, effective management requires the best available information on the status of caribou habitat. This project focused on caribou ranges in northwestern Alberta, and sought to address the following questions:
What is the accuracy of existing human footprint data?
What is the state of vegetation growth on vegetated footprint types?
Where should restoration efforts on seismic lines be targeted?
We collected high-resolution aerial imagery and associated ground-truthed data from four sampling blocks in two caribou ranges to assess human footprint accuracy and state of vegetation recovery. We used existing geospatial datasets and approaches developed for northeastern Alberta to address targeting restoration efforts. The collected aerial imagery resulted in a large geospatial dataset including orthophoto mosaics and 3D point clouds, which create three-dimensional visualizations and allow the identification of individual trees and their heights.
We found that the overall accuracy of the existing human footprint data was high (93.44%), with variations in accuracy across footprint types and blocks. Seismic lines, wells, and harvest areas had similar, high levels of overall accuracy; accuracy was lowest for “other human footprint” types such as pipelines. Our results indicate that regrowth is occurring not only on seismic lines, but on other types of human footprint as well. This regrowth varies locally in the type, height, and density of vegetation.
The township prioritization analysis identified that high priority restoration zones within the Chinchaga and Bistcho caribou ranges tended to be spatially clumped, suggesting that habitat restoration should be prioritized in these regions. Conversely, high priority zones in the Caribou Mountains and Yates caribou ranges tended to be more dispersed, making it less efficient to target the high priority zones within these ranges. A key next step for this
work would be incorporating the information on vegetation recovery into this prioritization analysis.
We discuss a number of potential next steps ranging from further exploration and analysis of the data, opportunities for ongoing data collection, and the opportunity to create an information portal that facilitates the use of this type of data for seismic line restoration management. These results are an exciting demonstration of the level of precision possible in discriminating vegetation cover types, heights, and density on human footprint features. Using the photogrammetry approach tested in this project, we are able to detect these vegetation regeneration variables to a high level of spatial resolution. This data has useful applications for developing and subsequently tracking vegetation recovery thresholds used for sub-regional or range planning, such as defining critical caribou habitat, which ultimately inform land management decisions.