- Shallow subsurface flow and surface runoff are spatially and temporally variable in forested environments. The location and timing of this runoff depends on both site characteristics (e.g., vegetation, soil texture, geology, and topography) and on time-varying conditions (e.g., soil water content, precipitation type, duration, magnitude, and intensity). Forest management activities (e.g., road and skid-trail building, harvesting) are known to increase the potential for more surface runoff, promoting increased erosion and sediment delivery to streams. While riparian buffers are widely accepted as effective in capturing runoff and sediment generated from timber harvest areas, uncertainty remains of the degree to which the harvest area influences the natural delivery pathways of water from the shallow subsurface and surface to the stream. In this study, we quantified shallow subsurface flow, surface runoff, and sediment concentrations generated from a harvested cutblock in comparison to a riparian buffer during the spring-fall directly following timber harvesting of a steep, Rocky Mountain headwater catchment (Alberta, Canada).
In Chapter 2, we quantified surface runoff, shallow subsurface flow, and sediment concentrations during precipitation events at the edge of a riparian buffer, in the riparian buffer, and along an unharvested hillslope. We instrumented a steep, planar hillslope with 40 unbounded collection troughs to collect shallow subsurface flow and surface runoff. We compared the spatial and temporal variability in runoff volumes and sediment concentrations to site-specific conditions, including proximity to harvest disturbance, local slope, foliar and basal cover, and soil moisture. The lowest runoff volumes were observed at the harvest-riparian edge while the largest runoff volumes were collected along the unharvested, reference hillslope. Runoff volumes in the riparian buffer were most similar to runoff volumes of the unharvested hillslope. However, we found no evidence for differences (α = 0.05) in average runoff volume between trough transects. Comparatively, we found strong evidence for greater sediment concentrations at the harvest-riparian edge compared to sites in the riparian buffer or on the unharvested hillslope. Analysis of relationships between runoff, sediment, and site conditions, indicated: 1) runoff volumes increased with increasing local slope and with increasing total precipitation, 2) runoff and sediment concentrations were poorly correlated with foliar and basal cover, depth of the O-horizon, and topographic wetness indexes, and 3) soil moisture content within the upper 0–6 cm of mineral soil in the general harvest area was greater than in the riparian buffer and unharvested hillslope during 4 soil moisture field campaigns.
In Chapter 3, we simulated high-intensity, short-duration rainfall (~75 mm hr-1) on 15 plots (1 m2) to test for differences in shallow subsurface and surface runoff response and resulting sediment concentrations. Differences in runoff rates and sediment concentrations were compared between the general harvest area, the harvest-riparian buffer edge, and in the riparian buffer under contrasting antecedent moisture conditions. We observed the lowest runoff rates in the general harvest area, increasing at the harvest-riparian buffer edge, with the greatest runoff rates in the riparian buffer. The greatest runoff volumes occurred during dry antecedent moisture conditions with a significant difference in runoff rates between the general harvest area and the riparian buffer. Differences in runoff response between plot locations decreased under wet antecedent moisture conditions. Sediment concentrations were significantly greater within the general harvest area than the harvest-riparian edge and in the riparian buffer, with the greatest variability in sediment concentrations observed at the harvest edge.
The magnitude of runoff response following forest harvesting was low to moderate in comparison to other disturbances (e.g., roads, severe wildfire). We conclude that localized redistribution of runoff and sediment dominates on the steep, vegetated, planar slopes tested in this study, where delivery of sediment generated in the general harvest area through the riparian buffer to the stream remains unlikely.