Do impacts from forest harvesting spatially accumulate in stream networks?

Abstract

Despite a suite of provincial guidelines working in concert with federal policy to promote sustainable forest management in Canada, legacies of ecosystem degradation from forestry persist, particularly within nearby streams. Harvesting-induced impacts to small headwaters tend to be well-studied and reasonably predictable, but it is lesser known how impacts manifest in larger downstream areas and if best management practices (BMPs) designed to protect against such impacts are effective at broader spatial scales. To address this uncertainty, I examined ecosystem indicators affected by selection-based harvesting under BMPs within mixed hardwood forest catchments along a spatial gradient (1st to 4th order streams, drainage areas 49 to 1856 ha). Indicators included water chemistry, dissolved organic matter (DOM) quality, sediment deposition and organic matter content, leaf litter decomposition and associated macroinvertebrate community structures, analysis of the algal contribution to stream consumer diet, and mercury (Hg) in water and biota. Indicator responses along the spatial gradient for streams within paired harvested and reference catchments were compared using two-factor ANOVAs. A significant interaction between treatment (reference vs. harvested) and sample site location (upstream, middle-reach, and downstream) was used as evidence for spatially cumulative trends within the harvested catchment (i.e., indicator response over spatial range differed in harvested catchment compared to reference catchment). A treatment effect with potential to have adverse implications at downstream sites in harvested catchments indicated that BMPs were not effective at preventing impacts at broader spatial scales. Overall, I found strong evidence for forest harvesting-induced spatially cumulative trends for % organic content of coarse sediment and various endpoints for Hg in water and biota. BMPs were effective at preventing an adverse impact from harvesting at downstream sites for all indicators except sediment deposition, % dietary algae, Hg(II) in biota, and bioaccumulation factor for MeHg in biota, but they were not effective at upstream sites (i.e., local spatial scale) for conductivity, total nitrogen, % organic content of sediment, % shredders, % dietary algae, and MeHg and Hg(II) in water and biota. My study contributes to a comprehensive and predictive understanding of the potential for spatially cumulative trends within harvested catchments, which is critical to help forest managers maintain healthy forest streams and their provisioning of aquatic ecosystem services for future generations of humans and wildlife alike.