Date of Award


Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)


Forest Resources


Robert S. Seymour

Second Committee Member

Alan S. White

Third Committee Member

Malcolm L. Hunter


Silvicultural systems based on old-growth forest disturbance regimes have been proposed as a way to harvest forest products while maintaining natural stand structures and biodiversity. In northeastern North America's Acadian forest ecosystem, data from old-growth preserves indicate that disturbance regimes have historically been dominated by small canopy gap events, leading to complex multi-aged stand structures. To evaluate the outcomes of silviculture intended to resemble these gap-driven processes we investigated forest age and structure in mature, apparently even-aged second-growth stands in the Scientific Forest Management Area of Baxter State Park, Maine. Silvicultural treatment effects in managed stands were quantified by pairing these stands with untreated benchmark stands in a complete-block experimental design. Benchmarks had no known history of management since the period of widespread sawlog harvesting in this region during the last decades of the 19th century. Managed stands were treated approximately 15-20 years ago with uniform shelterwood establishment cutting to harvest mature growing stock and allow development of regeneration. Recent gap-based silviculture took the form of group overstory removals meant to approximate old-growth rates of canopy disturbance through creating discrete gaps in the forest canopy. Rather than being even-aged, benchmark forests in fact showed a multi-aged population structure dominated by two cohorts. The first corresponds with intense sawlog harvesting during the late 19th century, the second with the aftermath of the 1913-1919 spruce budworm (Choristoneura fumiferana Clem.) outbreak. These two cohorts associated with large known disturbances dominated stand age structures, but trees were recruited into the stand in nearly all decades from 1870 onwards, indicating that overstory tree mortality commonly opens up growing space. In the managed stands the presence of a recent cohort indicated that establishment cutting was a severe enough disturbance to create an increased rate of sapling establishment. The oldest age classes were much reduced in these stands, probably as a result of their preferential removal during establishment cutting. Despite similar origins, reconstructed disturbance histories of the two stand types were markedly different, even in decades before management began. This raises questions about the impact of tree mortality on the results of dendrochronological studies. The rate of canopy removal in managed stands over a twenty-year cutting cycle averaged 22.9%, similar to previous old-growth studies and to twenty-year rates of canopy loss in benchmark stands. Individual harvest gaps were much larger than natural treefall gaps, though, meaning that disturbed canopy area was concentrated in fewer gaps. We found no difference in species composition between benchmark and treated stands, but both silvicultural treatments affected stand structure. As expected, establishment cutting reduced stand basal area while preserving numbers of cavity trees and snags, and inputting high amounts of small-diameter coarse woody debris, and allowed the pool of seedlings and saplings in the stands to grow and self-thin. Subsequent gap creation removed all cavity trees and snags while leaving only scattered legacy overstory trees. The taller saplings in gaps appeared to be damaged, and within harvester trails used to create the gaps, nearly all regeneration was eliminated.