The upper Great Lakes region in central North America contains a forest transition zone where temperate and boreal tree species reach their northern and southern range limits, respectively. It is only within this narrow latitudinal band (~3 degrees), that relatively warm-adapted temperate and cold-adapted boreal tree species are found growing together in upland mesic sites. If climate is a main driver of forest dynamics within this region, recent and predicted climate change should result in major forest shifts, including the expansion of temperate species and range contraction of boreal species. Such changes should first be manifest in growth and abundance trends of tree species in the understory regeneration layers. In addition to climate, numerous other factors such as overstory composition, understory abiotic environment, competition with shrub and herbaceous layers, and browse pressure drive tree regeneration trends. Interrelationships and interactions among these drivers will ultimately determine the direction and rate of forest change. We explored these research topics through field studies of naturally established seedlings and saplings at 124 upland mesic forest sites across a three state (Minnesota, Wisconsin, and Michigan U.S.A) 170,000 km2 area of the temperate-boreal transition zone.
Chapter 1 examined relative abundance shifts of temperate and boreal tree regeneration at two spatial scales: local ecotonal boundaries between temperate and boreal dominated stands and across the regional temperate-boreal transition zone. Because we compared understory performance across locally changing overstory composition, we calculated species regeneration success as the difference in relative abundance between the understory and overstory layers. At the local scale, both shade-tolerant temperate and boreal species exhibited positive tree regeneration success across ecotonal boundaries. However, across the region, regeneration performance varied with mean summer temperature and to a lesser extent mean annual precipitation. Changes in regeneration success were generally greatest at the warm end of the transition zone, with temperate broadleaf Acer saccharum, Fraxinus nigra, and Ostrya virginiana responding positively and boreal Abies balsamea showing significantly reduced performance. For the most frequent temperate species, Acer rubrum, regeneration success was greatest in boreal neighborhoods and at cool and dry sites. Other species did not exhibit detectable shifts in regeneration success, potentially due to traits such as shade-tolerance, palatability, and mode of reproduction. Overall we found that numerous tree species growing across the temperate-boreal transition zone are likely sensitive to climate at early stages of development, with observed shifts in regeneration success concomitant with the direction predicted in response to climate change.
Chapter 2 assessed the relative importance of and interrelationships among explanatory variable sets in explaining the composition of the tree regeneration layer. We used redundancy analysis (RDA) and variation partitioning to quantify the unique, shared, and total explanatory power of four sets of drivers: climate, understory abiotic environment, overstory composition, and understory biota. The results showed that all four driver sets individually explained a significant portion of tree regeneration compositional variation and additionally that there were strong relationships among explanatory variables. Overstory composition, which directly influences seed availability and also was found to be closely associated with understory environmental conditions and biota, had approximately twice the explanatory power of any of the other three driver sets. Some of the strongest individual drivers were overstory Acer saccharum and Populus tremuloides, soil pH, mean summer temperature, and mean annual precipitation. Suites of associated drivers included cool, moist, sandy, and acidic conditions; overstory boreal broadleaf species, light availability, shrub abundance, and forb cover; and warm temperatures and graminoid cover. Due to the strong interrelationships among drivers, the direction and rate of forest change will likely depend on how the importance of drivers shifts with climate and, for the biotic drivers, on the rate and magnitude of their own responses to climate change.
Chapter 3 investigated sapling height and radial growth rates of five temperate and boreal species. This study included over 1700 stems of naturally established, competing saplings growing at 14 sites across the temperate-boreal transition zone. Top performing linear mixed-effects models typically included two-way interactions among mean summer temperature, browse pressure, understory light levels, and initial sapling size. As hypothesized, temperate sapling growth increased and boreal growth decreased with increasing temperatures. However, the relative performance of competing species shifted depending on the level of browse pressure. Positive temperate growth responses to temperature were eliminated by heavy browse pressure, tilting growth rates in favor of less palatable boreal conifers at all but the warmest sites. Spatial variations in browse pressure levels across the region suggest that temperate expansion may proceed most rapidly in areas where browsing is least intense. Growth responses to temperature also varied with sapling size and, for the least shade-tolerant species in the study, Quercus rubra, light availability. Enhanced growth by temperate species in response to warmer temperatures was most detectable under favorable conditions including low browse pressure and high understory light availability, suggesting that any efforts to facilitate forest compositional changes will need to take into account these trends.