Regeneration response to repeated prescribed burning in Appalachian hardwood forests


Tara L. Keyser, Mary A. Arthur, Heather D. Alexander, and David L. Loftis

Research Forester (TLK), USDA Forest Service, Southern Research Station, Bent Creek Experimental Forest, 1577 Brevard Rd., Asheville, NC 28806; Professor of Ecology (MAA), Department of Forestry, University of Kentucky, 105 T.P. Cooper Building, Lexington, KY 40546; Assistant Professor (HDA), Department of Biological Sciences, One West University Blvd. University of Texas, Brownsville, TX 78520; Research Forester (DLL), Emeritus, USDA Forest Service, Southern Research Station, Bent Creek Experimental Forest, 1577 Brevard Rd., Asheville, NC 28806. TLK is corresponding author. To contact, call 828.667.5261 ext. 202 or email

Abstract – Management on public lands across the eastern US is increasingly focused on the restoration of resilient structures and species compositions, with prescribed burning being the primary tool by which many landscape-level restoration efforts are implemented. In Appalachian hardwood forests, where altered disturbance regimes have contributed to an increase in shade-tolerant species and concomitant decrease in the relative abundance and competitiveness of oak species in the forest understory, prescribed fire is often recommended as a treatment that enhances the development of desirable oak species. In 2002, three study sites (200 - 300 ha) were established on the Cumberland Plateau in eastern KY. Each site contained three treatment areas (58 - 116 ha) which were randomly assigned to one of three treatments: Frequent Fire (FF); Infrequent Fire (IF); Fire-Excluded (FE). Within each treatment area, 8 - 12, 0.04 ha subplots were established, with each subplot classified by landscape position (subxeric, intermediate, or submesic). At each subplot, advance reproduction (<3.8 cm dbh) was enumerated by species into five size classes using a 0.004 ha plot. Prescribed fires were conducted in 2003 and 2009 (IF treatment) and 2003, 2004, 2006, 2008, and 2011 (FF treatment). Regeneration inventories were conducted in 2002, 2004, 2008, and 2015. A randomized complete block experimental design with a split-plot treatment structure was used to test the effects of treatment and landscape position on the regeneration pool. Long-term changes in the abundance of understory tree species in response to fire frequency are presented, and potential effects on future species composition are discussed.