Beyond forest gap dynamics: Canopy tree senescence and advance regeneration

Background/Question/Methods Losses of mid-tolerant tree species in selection-managed stands have been observed widely. One hypothesis explaining this decline states that large gaps are necessary for creating sufficiently high light levels for regeneration of mid-tolerant species. Selection management, however, tends to create gaps that are too small and ensuing light levels too low for successful regeneration of mid-tolerant species. Nevertheless, such species can be found in closed, old-growth forests despite absence of large gaps or disturbance-related canopy declines. An alternative hypothesis states that large senescing trees experience strong decreases in leaf area index, leading to increased forest floor light levels that are utilized by regenerating mid-tolerant species. Selection management, however, tends to harvest trees before senescence and mid-tolerant regeneration, thus promoting tolerant tree species. In line with this latter hypothesis, we looked for evidence of regeneration of mid-tolerant species under large senescing trees. Therefore we mapped a large, old-growth forest plot in Haliburton, Ontario, in the mixed deciduous-coniferous Great Lakes-St. Lawrence Forest, and quantified the relationship between regeneration abundance and abundance of large, senescing trees for intolerant, mid-tolerant, and tolerant tree species. We expected positive effects of large tree abundance for intolerant and mid-tolerant regeneration, but smaller or negative effects for tolerant species.

Results/Conclusions The forest canopy was dominated by tolerant species, but several intolerant and mid-tolerant species were also present, though rare. Despite the rarity of intolerant and mid-tolerant species in the canopy, regeneration of species of all tolerance levels was present throughout the plot. Despite some variation in the pattern, most intolerant and mid-tolerant species showed a positive response of regeneration abundance to large tree abundance. Most tolerant species, however, showed no or a negative response of regeneration abundance to large tree abundance. Our findings lend support for the hypothesis that gradual declines in leaf area index as trees senesce, may play an important but so far overlooked role in maintaining mid-tolerant species and tree species diversity. We suggest that retention of senescing trees or delay of harvest may aid in maintaining species composition in selection-managed stands.