UBC Research Data

How does vegetation fragmentation affect urban carbon stock in City of Vancouver, Canada Li, Derek


Urban vegetation acts as an extremely important carbon stock to store carbon. Although many studies have investigated the relationship between urban vegetation and carbon stocks, few of them closely examine the phenomenon of vegetation fragmentation in urban areas and how it affects the carbon stock. Therefore, this project investigates how vegetation fragmentation affects carbon stocks in the City of Vancouver using GIS analytic tools. To do so, ArcGIS Pro was used to modify and analyze remotely sensed data of Vancouver. Land cover data were reclassified and clipped within the Vancouver dissemination area (DA) boundary to generate a new raster contain only the vegetation data of Vancouver. This raster was then inputted into FRAGSTATS to calculate aggregation metrics that quantify the fragmentation. After that, the aggregation metrics were joined with the carbon data and put back to the Vancouver dissemination area boundary. Geographically weighted regression was then used to compute a statistical model to analyze the relationship between carbon and fragmentation of each DA. The statistical model shows a strong positive relationship between carbon and fragmentation and has an overall adjusted R squared value of 0.77, suggesting a strong relationship. The results suggest that higher fragmentation is associated with lower carbon density and thus smaller carbon stock. Therefore, it is strongly recommended that city planning takes the spatial distribution and composition of urban vegetation into consideration. Furthermore, the study also found that some smaller vegetation patches have very high carbon density even though their area is small. Therefore, mid to large-scale parks and other infrastructure that provide decent areas of closely aggregated trees may be promoted as their contribution to the urban carbon stock are effective and significant.

Item Media

Item Citations and Data

Usage Statistics