Forest regeneration in gaps seven years after partial harvesting in riparian buffers of boreal mixedwood streams

• We examined tree regeneration in gaps created by partial harvesting in riparian buffers.
• Tree regeneration in small, medium and large gaps were compared and contrasted with uncut buffers.
• We found dense tree regeneration in medium to large gaps with favorable light and temperature.
• Partial harvesting creates desirable habitat complexity by increased seedling growth in gaps.

Partial harvesting in boreal forest riparian buffers has been proposed as a management tool to emulate natural disturbance (END) along streams and lakes to increase shoreline habitat complexity through ensuing forest regeneration. We investigated the effect of partial harvesting in stream-side riparian buffers on regeneration of canopy species (Abies balsamea, Betula papyrifera, Picea glauca, Picea mariana and Populus tremuloides) by testing the hypothesis that juvenile trees would be more abundant, species-rich, and larger in gaps than in non-harvested buffers, and that those differences would be proportional to gap size. Our main objective was to determine what gaps sizes are conducive to promote tree and woody shrub regeneration in stream-side riparian areas of boreal mixedwood forests. We compared woody plant regeneration among harvest gaps varying in size (small, 10–20 m2; medium, 21–100 m2; and large, >100 m2), and contrasted with regeneration under closed canopy in unharvested buffers and undisturbed reference forests. We found that woody plant communities in partial harvesting gaps were denser and more diverse than unharvested buffers. They were more so in medium to large gaps and consisted of shade-intolerant and early successional species. Favorable light and soil temperature in the large gaps appear to be responsible for this. We concluded that intentional shoreline disturbance aiming to achieve increased riparian habitat complexity and early successional forest community is possible by partial harvesting at up to 50% basal area removal. However, as END principles are increasingly applied to riparian forest management, it will be necessary to test and monitor the effectiveness and longer-term ecological responses of riparian communities to such management at catchment and landscape levels.