Long-term recovery of ecosystem services following forest harvest in coastal temperate rainforests of Vancouver Island, British Columbia, Canada

All ecosystems, and the ecosystem services (ES) they provide, are susceptible to potentially lasting impacts of resource extraction. For example, in forests, timber harvesting provides near-term services (such as wood products), but can be responsible for declines in other services such as carbon storage or wild edible foods, which may take decades or even centuries to recover. Some ES may recover quickly with forest regrowth, while others recover either slowly or not at all. However, the long-term recovery of multiple forest ES has rarely been quantified. My fundamental goal in this thesis is to build an improved understanding of how multiple ES recover following forest harvest, using the heavily harvested coastal temperate forests of western Vancouver Island, British Columbia, Canada, as my study system. First, I used a forest chronosequence to estimate the recovery trajectories for eight ES over a 212-year period. I used changes in key forest structures to estimate the provision of the following services: wood volume, carbon storage, potential nesting platforms used by an emblematic old-growth associate bird species the marbled murrelet (Brachyramphus marmoratus), habitat services provided by coarse woody debris, habitat services provided by dead trees, large heritage trees, wild edible berries, and large redcedar (Thuja plicata) used in traditional First Nations carving. ES recovered along varying non-linear trajectories and within markedly different timeframes. Wood volume stocks, dead tree biomass, and carbon storage recovered the fastest, reaching their maximum rates of recovery at around 65 years. In contrast, recovery of wild edible berries, heritage large trees, and habitat for marbled murrelet did not even commence until 70-100 years. Large heritage trees and large redcedar did not recover to old-growth baseline (forests >250 years old) during the 212-year period of my chronosequence. Second, I examined how ES recovery differed in two forest types: riparian and upland forests. With field assistance from a local First Nations crew, I estimated ten ES in old-growth (late seral stands >250 years in age) and second-growth stands (~35 years age) within each forest type. In addition to those services sampled in Chapter 1, I also estimated cedar bark for use in traditional First Nations weaving, salal (Gualtheria shallon) merchantable greenery, and fish habitat provided by instream woody debris. The abundance of ES differed significantly by forest type and forest age. For example, large cedar and potential nesting platforms for marbled murrelets were absent in second-growth stands, and significantly higher in riparian sites relative to their presence in upland forests. Old-growth riparian forests were hotspots of many ES, providing the highest levels of all services except salal merchantable greenery. The long timeframes and varying trajectories of recovery highlight the need to avoid ES declines proactively, for example by preserving sites with high levels of ES or working with First Nations to identify key areas with high levels of desired ES. Forest age and forest type have significant and major effects on multiple ES, and are thus two key variables for managing multiple ES in forested landscapes. Overall, this thesis provides insights into the effects of forest harvesting on multiple ES of ecological, cultural and economic importance. By applying forest ecological understanding to track changes in a bundle of ES, I identify influences of site conditions, long timeframes of successional recovery, and impacts from management to gain a broader understanding of the factors shaping forest ES. By building an improved empirical and conceptual understanding of multiple ES and their change through time, I have provided novel insights as well as practical solutions towards the challenge of long-term forest planning to sustain multiple ES.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.