Dieback in the Native Shrub, Lindera benzoin: A Subtle Effect of Forest Fragmentation

I report on a documented dieback in spicebush, Lindera benzoin, after the winter of 1993-1994 in two mature deciduous forest stands in southwestern Ohio. The study compares differences between 67 ha Hueston Woods and 5.2 ha Kramer Woods and examines spatial patterns of decline in Kramer Woods. Local daily minimum temperatures in January 1994 were among the coldest on record since 1889The shrub is one of the most important understory species in both forests, but relative basal area measures of dead stems were significantly less in Hueston Woods (mean = 54% of the total) than in Kramer Woods (86%). Spicebush shows a clustered distribution in Kramer Woods, present on a flat upland area along the northern edge. Only two points with spicebush showed no decline and they are >60 m from the north edge and along streams. Dieback was up to 100% for plots with mean shrub heights >1.5 m, resulting in a significant understory gap. These findings identify important differences between and within forest patches that appear attributable to fragmentation effects. Forest conservation plans need to better capture a diversity of topographic settings or incorporate buffer zones along edges to better ensure the protection of native flora in human-dominated landscapes. OHIO J. SCI. 96 (4/5): 76-80, 1996 INTRODUCTION Southwestern Ohio is in the southern portion of the beech-maple association in the eastern deciduous forest (Braun 1950). The potential vegetation varies in response to topography and soil conditions, but on optimal settings north of the glacial boundary, beech (Fagus grandifolid) and sugar maple (Acer saccharuni) are dominant in the forest canopy (Braun 1950, Vankat et al. 1975). Spicebush (Lindera benzoin) is characteristically the dominant understory shrub (Braun 1950). Post-European forest clearing of the eastern deciduous forest was nearly complete in its magnitude before the early 1900s (Williams 1989, Whitney 1994). In the beech-maple association, intensive row-crop production with the rise of the corn belt isolated small forest fragments in a matrix of agricultural lands (Hart 1986). Curtis (1956) in southeastern Wisconsin, Whitney and Somerlot (1985) in central Ohio, and Medley et al. (1995) in southwestern Ohio document >90% loss with settlement and only a slight increase to less than 20% forest cover. Current forest cover in the southwestern part of the state is estimated at 14% (Griffith and Wharton 1994). Intensive and highly productive agriculture, coupled with urban expansion, maintain a very low distribution of forest (remnant or emergent) within the original mapped boundaries of the beech-maple association. The ecological consequences of fragmentation are profound as large contiguous forests are converted to small isolated patches (see papers in Burgess and Sharpe 1981, Godron and Forman 1983). Abrupt forest edges alter microclimatic conditions and the relative competitive abilities of shade intolerant and tolerant Manuscript received 27 March 1996 and in revised form 10 September 1996 (#96-06). species (Saunders et al. 1991). For instance, the greater light availability of small forest fragments or disturbed mid-successional woodlands may explain the success of the Asian shrub Lonicera maackii (Amur honeysuckle) in southwestern Ohio after about I960 (Luken 1993) and European herb Alliariapetiolata (garlic mustard) in forests throughout the midwest (Nuzzo 1993). Humanrelated disturbances may provide window's for invasion and available habitats for establishment by nonnative plants (Hobbs and Huenneke 1992). The actual creation of an edge and adjacent land-use activities may be viewed as direct human-related disturbances. A more subtle concern, however, is the degree to which fragmentation may influence the "natural" disturbance regime of beechmaple forests. I documented dieback in spicebush, Lindera benzoin, after the winter of 1993-1994 at two locations in southwestern Ohio. Spicebush is known to be susceptible to winter-minimum temperatures (Braun 1961), so its decline after a harsh winter may be viewed as a natural disturbance. The study first compares differences between the relative importance and dieback of spicebush in two forest patches: Hueston Woods, a 67 ha remnant stand, and Kramer Woods, a 5.2 ha regrowth stand that is at least 100 yrs old. Second, it examines spatial patterns of abundance and dieback in spicebush within a forest patch, Kramer Woods. Significant differences in the level of dieback between these two forests or spatial irregularities within Kramer Woods may suggest fragmentation effects. Land conservation plans need to carefully consider the impacts imposed by fragmentation when determining the size and position of a nature reserve in an otherwise human-modified landscape (National Research Council 1993). By focusing on the potential impacts of low-temperature extremes, this study examines one disturbance that may be indirectly influenced by the landscape position of a forest fragment. OHIO JOURNAL OF SCIENCE K. E. MEDLEY 77 MATERIALS AND METHODS Study Area Two protected forest areas were selected that are less than 10 km apart and occupy similar physical settings on upland till-plain in southwestern Ohio (Fig. 1). Hueston Woods is an approximate 67 ha uncut forest stand protected in Hueston Woods State Park, Preble County, Ohio. The forest, especially its eastern sector, has remained essentially undisturbed since its purchase by the Hueston family in 1797 and acquisition by the State of Ohio in 1940 (Runkle et al. 1984). Kramer Woods is a much smaller, about 5.2 ha, forest stand that is protected as a natural area on the campus of Miami University, Butler County, Ohio. The forest is an old regrowth stand, at least 100 years, that was protected by the Kramer family and then donated to the university in 1989. These two forest stands differ significantly in size, but both have a long history of protection and now serve as nature reserves in the beech-maple association. Climatological Record The study focused on vegetation response after the winter of 1993-1994. Minimum winter temperatures were recorded at the Ecology Research Center, adjacent