Patterns of plant community structure within and among primary and second-growth northern hardwood forest stands. doi:10.1016/j.foreco.2009.09.012

Forest scientists advocate the use of natural disturbance-based forest management for restoring the characteristics of old-growth forests to younger second-growth northern hardwood stands. However, prescriptions rely upon studies that have (1) not spanned the full range of conditions and species assemblages, and (2) focused primarily on contrasting old-growth andmature second-growth stands at a single scale. To examine how the legacy of historical logging activities influences forest structure and function, we compared and contrasted patterns of plant community structurewithin and among secondgrowth and primary stands on the north shore of Lake Superior in Minnesota, USA — near the current range limits of the dominant species, sugar maple (Acer saccharum). We expected second-growth stands to be in younger developmental stages, and structurally less heterogeneous both within and among stands. Furthermore, we expected those differences to be associated with patterns of plant community composition and diversity. Three of the four primary stands and one of the eight second-growth stands were in the old-growth stage of development. Yellow birch (Betula alleghaniensis) and conifers as a group (Thuja occidentalis, Picea glauca and Abies balsamea) were more abundant, and yellow birch was more variable, within primary stands than second-growth stands. The volume and heterogeneity of coarse woody debris in intermediate decayand size classes was also greater within and among primary stands relative to second-growth stands. While mean subplot richness of overstory tree species was greater in primary stands, mean quadrat richness, and rates of species accumulation for forest herbs as well as total herbaceous cover, and graminoid cover were greater in second-growth stands. Furthermore, total basal area (BA), the BA of conifer species, the density of yellow birch trees, understory vegetation and light transmittance were more variable among second-growth stands. At the multivariate level, primary stands were distinguished from second-growth stands not by differences in stand structure, but by a greater abundance of yellow birch and conifer species in the canopy, whichwas also related to O-horizon depth and understory plant species composition and structure. Differences in community structure between primary and second-growth standsmay have resulted from the original cutover aswell as highgrade logging, which together may have disrupted the mechanisms that maintain populations of important co-dominant tree species and associated understory plant communities in northern