Date of Award

Fall 12-18-2020

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)


Forest Resources


Shawn Fraver

Second Committee Member

Laura Kenefic

Third Committee Member

Pascal Berrill

Additional Committee Members

Jay Wason


Declines in stands of northern white-cedar (Thuja occidentalis L., hereafter cedar) have been observed as both shifts in species composition and reductions in cedar densities, particularly those stands in lowland sites (Curtis 1946, Boulfroy 2012). While several factors inhibiting cedar regeneration have been identified, a thorough understanding of the conditions that best promote regeneration is lacking. Our objectives for the first chapter were to characterize the site conditions associated with successful regeneration in lowland cedar stands and to describe how spatial patterns of various cedar size classes relate to site preference and to regeneration dynamics. These objectives were achieved by mapping the seedling, sapling, and overstory communities at 15 lowland cedar stands at five sites in Maine, USA, and examining the fine-scale site conditions (microsites) in which cedar seedlings and saplings occur. Our analyses demonstrated that cedar regeneration tended to establish on elevated microtopographic features (i.e., mounds) while failing to establish in large numbers in small wet depressions (i.e., pits) and interspaces between mounds and pits (i.e., flats); however, this trend was more pronounced for seedlings than for saplings. Volumetric moisture content in these three features was shown to decrease in the order pits > flats > mounds. Logistic regressions using regeneration status (live vs. dead) as the response variable further supported the importance of microtopography, as well as canopy openness (greater openness associated with live stems). Neighborhood crowding was also associated with status among saplings (greater crowding associated with live stems), while seedlings were not affected by differences in crowding intensity. In addition, browse on seedlings was associated with dead status. The distinctly clustered spatial patterning found among both seedlings and saplings suggests a dependence on favorable microsites, which create small-scale patchiness within stands. These findings can aid land managers in developing informed plans to promote viable cedar populations in these ecologically and economically important forests.

Our overall objective in the second chapter was to assess browsing pressure on tree regeneration in lowland cedar stands. We used observations of seedling and sapling browse from these same 15 stands to explore the following objectives: 1) determine if cedar seedlings shorter than expected snow depths experience less intense browse, and 2) assess relative browse impact on common woody species in these stands due to three common herbivores: white-tailed deer (Odocoileus virginianus Zimmermann), snowshoe hare (Lepus americanus Erxleben), and moose (Alces alces Gray). To accomplish the first objective, we determined the percent of browsed cedar seedlings by height class (to 140 cm), and tested these values against a uniform distribution (equal browse among all classes). Our results showed that browse increased with increasing height class, with seedlings below ca. 50 cm showing a clear positive relationship between increasing height and increasing browse pressure. We conclude that snow cover in these stands, which ranges from ca. 25 to 85 cm maximum winter depth, offers cedar seedlings some protection from browse during winter. The second objective was accomplished by conducting a series of chi-squared goodness-of-fit tests analyzing the abundance of eight common tree species and one shrub species against browse frequency, and plotting the results for visual interpretation. Results clearly showed hardwood and deciduous shrub species were preferred overall, but elevated preference for cedar was seen in plots assumed to be used primarily by deer, and elevated Abies balsamea (L.) Mill preference in plots with greater regional moose densities. These results show hare and moose do not substantially impact cedar regeneration, but deer browse likely contributes to the poor recruitment of cedar seedlings and sapling in stands classified as habitat suitable for use as deer wintering areas.