Limitations on Regeneration Potential Following Even-aged Harvests in Mixed-oak Stands
Open Access
- Author:
- Campbell, Anthony J
- Graduate Program:
- Forest Resources
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- February 15, 2013
- Committee Members:
- Kim C Steiner, Thesis Advisor/Co-Advisor
James Craig Finley, Thesis Advisor/Co-Advisor
Laura P Leites, Thesis Advisor/Co-Advisor - Keywords:
- stochastic frontier analysis
oak regeneration
interfering vegetation
mixed-oak forests - Abstract:
- Forest managers are interested in factors that inhibit oak (Quercus spp.) success, because of wide-spread regeneration failures in the forests of the eastern United States. Using stochastic frontier analysis, this project investigated the effects of pre-harvest levels of rhizomatous fern, mountain-laurel (Kalmia latifolia L.), and tall shrub cover on oak, red maple (Acer rubrum L.), and total regeneration abundance in the decade after even-aged harvest. The data used were collected from 18 mixed-oak stands in central Pennsylvania. Stands were harvested between 1997 and 2001, and were followed through their tenth growing season. Parameter estimates were compared to determine if regeneration differed in its response to the three classes of interfering vegetation and non-modeled factors. Oak and total regeneration responded similarly to rhizomatous fern and tall shrub cover, and were inhibited least by mountain-laurel. No significant relationship was found between tall shrub cover and abundance of red maple regeneration, but this species was significantly inhibited by rhizomatous fern and mountain-laurel cover, the former having the most severe effect. Oak and red maple responded very similarly to rhizomatous fern and mountain-laurel cover, but red maple exhibited significantly more resiliency to non-modeled inhibitory factors regardless of cover type. Stochastic frontier analysis was also compared with boundary analysis to demonstrate the superiority of the former as a tool for estimating maximum values.