Spatiotemporal analysis of ecosystem change and landscape connectivity using satellite imagery in west-central British Columbia, Canada

• Main Author: Morley, Ilythia Dunbar
• Format: Text
• Language: English
• Published: University of British Columbia 2021
• Subjects: Canada; British Columbia; Omni; Alces alces
• Online Access: https://dx.doi.org/10.14288/1.0402358; https://doi.library.ubc.ca/10.14288/1.0402358

Environmental change poses a significant threat to landscape connectivity and threatened wildlife populations. There is a growing need to understand the impacts of broad-scale change across dynamic landscapes on wildlife movement to inform conservation strategies and landscape management plans. As the availability of satellite imagery time series increases, there are new opportunities to monitor broad-scale changes in ecosystems and landscape connectivity. The objective of this thesis is to apply novel methodological approaches using a time series of Landsat imagery to quantify spatiotemporal changes in ecosystem cover and landscape connectivity from 1997 to 2019 in west-central British Columbia, Canada. Using Time-Weighted Dynamic Time Warping (TWDTW) and Landsat imagery, spatiotemporal changes in biogeoclimatic ecosystem classification (BEC) zones were quantified to summarize complex patterns of change that reflect the influence of landscape disturbance. The TWDTW classification showed a transition of the IDF Dry ecosystem to MS Dry and SPBS Dry in the north and northeast of the study area in response to large wildfires in the region. Reduced IDF Dry cover signifies a loss in ungulate habitat, variation in the “green-up” date of vegetation, and increased low productivity forest cover. Omniscape, a circuit theory approach for omni-directional landscape connectivity modelling, was then used to quantify and map landscape connectivity for moose (Alces alces) populations in 1997, 2009, and 2019 and assess the impacts of ecosystem change on potential wildlife movement. Overall landscape connectivity for moose reduced by 70.23% between 1997 and 2019, during which time broad-scale disturbance resulted in ecosystem change. Results represent a novel spatiotemporal analysis of landscape connectivity, reveal variation in overall connectivity for moose across the region in response to disturbance, and predict the location of potential movement corridors. In summary, this thesis demonstrates the application of the TWDTW approach to classify spatiotemporal changes in ecosystem cover across heterogeneous landscapes and the suitability of the Omniscape method for quantifying changing patterns of potential landscape connectivity in the context of ongoing ecosystem change. The combination of time series ecosystem change monitoring and connectivity modelling provides the opportunity to examine the important spatiotemporal relationship between ecosystem cover, disturbance, and wildlife movement.