Species composition in a central hardwood forest in Kentucky 11 years after clear-cutting.

INTRODUCTION

The vast majority of the forests of Kentucky are second-growth, most having been harvested by the late 1940s (Parker and Merritt, 1995; Smith, 1995). These second-growth central hardwood forests recover rapidly from disturbance, resulting in forests that reflect the species composition and structure of the forests that preceded them (Bougher and Winstead, 1974; Muller, 1983). Rapid forest recovery following disturbance probably occurs for two reasons: (1) stump sprouting following forest harvest favors regeneration of species that were already on the site, and (2) low levels of soil disturbance associated with forest clear-cut leave the belowground ecosystem intact, permitting vigorous and healthy regeneration (Muller, 1983; Burns and Honkala, 1990). Five yr after a clear-cut in southeastern Kentucky, Muller (1990) found that height and diameter growth of stump sprouts were twice those of seedlings and root sprouts. He suggested that rapid growth of stump sprouts, resulting from the competitive edge provided by pre-existing root systems, is an important mechanism by which this forest type recovers from disturbance while retaining much of the original species composition. However, at 5 yr postclear-cut, stand density was dominated by understory arborescent species, especially flowering dogwood (Cornus florida) and red maple (Acer rubrum), which were not important components of the predisturbance mature forest. Thus, if forest species composition is to eventually approximate that of the predisturbance forest, we should see a successional shift from dominance by understory species and red maple to dominance by species that were most important in the original stand. This successional process should be the product of sprouting ability (or the number of sprouts each species produces), intraspecific competition among seedlings, stump sprouts and root sprouts within a species, and interspecific competition. The relative importance of seedling versus sprout regeneration 11 yr after clear-cutting and the further progression of this successional recovery are the focus of this study.

METHODS

Measurements of tree growth and regeneration were made in an 11.2-ha clear-cut watershed at the University of Kentucky Robinson Forest in Breathitt, Knott and Perry counties, southeastern Kentucky. These measurements were obtained: (1) immediately before clear-cut in 1983-1984, (2) 5 yr after clear-cut in 1988-1989 (Muller, 1990), and (3) in 1994, 11 growing seasons after the clear-cut. Before the clear-cut, the site was occupied by a 70-yr-old, second-growth forest dominated by oaks (Quercus alba, Q. coccinea, Q. prinus and Q. velutina), hickories (Carya cordiformis, C. glabra, C. ovata and C. tomentosa) and yellow-poplar (Liriodendron tulipifera) which comprised 39%, 17%, and 15%, respectively, of the volume of trees [greater than or equal to]25.4 cm dbh (Overstreet, 1984). In the autumn and winter of 1983-1984 all stems were cut at stump height (0.3 m) and marketable stems were removed from the site. With the exception of haul road and logyard construction, there was little disturbance of the soil. Residual logging debris was left in place.

In 1989, 50 circular plots with radius of 1 m were established along four transects located …