Macroinvertebrate responses to differing riparian treatments following forest harvest in the headwaters of Trask River watershed

Abstract

Listen

Translate article

Although it is well-established that macroinvertebrates are an integral part of a healthy stream ecosystem, studies designed to inform forest harvest best management practices for sustainability of macroinvertebrate communities in small streams have been limited. Current riparian management practices for headwater streams generally focus on providing adequate wood supply, reducing debris flows and transport of sediment, and maintaining water quality in larger streams that contain fish. Here we studied the effects of contemporary forest harvest practices on benthic macroinvertebrate density and community composition in 12 small headwater streams as part of the Trask River Watershed Study. To account for seasonal and year-to-year variability in macroinvertebrate densities, we sampled each site at the same time of year for six years before and four years after harvest. Three watersheds were clearcut with variable riparian buffers, three were clearcut with 12 m riparian buffers, and one was thinned with 15 m buffers. Five adjacent watersheds were not treated and were studied as reference sites. All sites had approximately 50 yr old conifer and alder forests, which had revegetated following fires and harvest in prior decades.In these steep mountain streams, benthic macroinvertebrate density and community composition were generally similar across watersheds prior to harvest. Collectors were the most abundant functional feeding group, followed by shredders; scrapers were the least abundant. After harvest, in two of the three clearcut watersheds with variable buffers, the total benthic density and percentage of collectors increased, primarily due to greater numbers of Chironomidae. We also observed more abundant emergence post-harvest at these two watersheds. In the third clearcut watershed with a variable buffer and in the three clearcut watersheds with uniform 12 m riparian buffers, we did not see changes in density or functional feeding groups after harvest. We explored the spatial and temporal aspects of the data using both non-parametric and parametric statistical methods. High variability swamped the treatment effect when watersheds were grouped by riparian treatment for the repeated measure analyses, while NMDS showed a significant shift post-harvest for the watersheds with variable buffers. In the clearcut watersheds with uniform buffers, macroinvertebrates densities and community composition did not shift. At two of the clearcut watersheds with variable buffers, we observed increased density for the first two years after harvest, and chironomids comprised up to 80% of the abundance in samples. In the subsequent years, we documented mixed responses for two watersheds. It was invaluable to have multiple pre- and post-harvest sampling periods in both harvested and reference watersheds in order to better identify site specific responses to treatment and to separate background variability from responses to harvest. We show that retention of riparian vegetation on headwater streams during clearcut harvesting minimized changes in total macroinvertebrate densities and community composition. Small buffers may be key to maintenance of diverse stream invertebrate communities in headwaters.