Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Forest Resources

Committee Chair/Advisor

Wang, Gaofeng

Committee Member

Walker , Joan L

Committee Member

Rieck , James R

Committee Member

Imm , Donald W

Committee Member

Baldwin , Robert F


Throughout the southeastern United States, upland pine sites that were once dominated by longleaf pine (Pinus palustris Mill.) have been converted to faster growing species such as loblolly pine (Pinus taeda L.). This study was designed to determine optimal silvicultural techniques for restoring longleaf pine in existing loblolly pine stands while retaining canopy trees and enhancing desirable characteristics of the ground layer vegetation. We applied seven harvesting treatments to six loblolly pine stands at Fort Benning, GA, with treatments including four that created uniform canopy distribution (Control: uncut, with basal area > 14 m2/ha; MedBA: residual basal area of ~ 9 m2/ha; LowBA: residual basal area of ~ 5 m2/ha; and Clearcut: complete canopy removal) and three that used group selection to create gaps of different sizes (SG: small-gap, ~ 0.12 ha; MG: medium-gap, ~ 0.25 ha; and LG: large-gap, ~ 0.50 ha). Additional cultural treatments were applied in a split-plot design, including an untreated control (NT), herbicide control of woody and herbaceous vegetation (H), and the herbicide treatment plus fertilizer (H+F). We monitored artificially regenerated longleaf pine seedling mortality and growth, measured the response of ground layer vegetation, and quantified resource availability over the first three growing seasons after harvesting.
Longleaf pine seedling mortality was highest in the first growing season, but by the end of three growing seasons mortality averaged 55%. Mortality was highest on Clearcut plots and increased from the forest edge to the gap interior in gap plots, demonstrating a facilitation effect of canopy trees on seedling survival in the first year after planting. Canopy trees showed a competition effect on seedling root collar diameter and the percentage of trees in height growth, with the highest growth rates on Clearcut plots and very little growth over three years on Control plots. Average seedling root collar diameter in gap plots did not differ from that in Clearcut plots, although seedling size increased from the forest edge to the gap interior. The H and H+F treatments did not affect cumulative mortality or seedling size after two or three growing seasons.
Canopy manipulation strongly affected light availability at the forest floor, with a negative exponential relationship between canopy density and canopy light transmittance and increasing light availability from the canopy edge to the gap center in each gap. Light transmittance was higher on the northern half of gaps than on the southern half of gaps, and average light transmittance increased with gap size. Soil moisture at 6 cm in the soil did not differ among canopy treatments or by within-gap position, but soil moisture at 60 and 100 cm in the soil profile was greater within canopy openings than beneath the canopy. Total soil nitrogen was highest north of gap center in LG plots, but there were no effects of within-gap position on foliar nitrogen of longleaf pine seedlings. Our results suggest that light is the most limiting resource for longleaf pine seedlings but that competition for below-ground resources, which is temporally more variable, also affects seedling response.
The ground layer was dominated by herbaceous vegetation in each growing season, and the study treatments did not affect the relative dominance of functional groups. In the uniform plots, vegetation cover increased following canopy removal, with the lowest cover of vegetation on Control plots and the highest cover on Clearcut and LowBA plots in each growing season. Similar patterns were observed for woody vegetation, but cover of herbaceous vegetation was not significantly different among canopy treatments in the third growing season. Harvesting released woody stems into the midstory, with higher stem densities in Clearcut and LowBA plots than in MedBA and Control plots. The herbicide treatments reduced woody stems in the second growing season, but the effect was not significant after three years. Species richness did not differ among the study treatments, and patterns of species composition were most strongly affected by site-specific factors. At a local scale, species composition shifted to early successional species following harvesting.
To reduce the development of a woody midstory and retain canopy trees for other ecological services, we recommend using single-tree selection to thin stands to a residual basal area between 5 and 8 m2/ha. Additionally, small gaps (0.1 ha) may be used to distribute local patches of longleaf pine regeneration throughout existing loblolly pine stands. In stands with high densities of woody stems, herbicides can be used to target hardwoods to improve the structure of the ground layer vegetation. Ultimately, frequent fire will be necessary to maintain the desired ecosystem structure over the long time periods.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.