Tree Species Composition and Structure in an Old Bottomland Hardwood Forest in South-Central Arkansas

Tree species composition and structure was determined for an old bottomland hardwood forest located in the Moro Creek Bottoms Natural Area in south-central Arkansas. Diversity for this forest was high with species richness ranging from 33 for the overstory and sapling strata to 26 for the seedling stratum and Shannon-Weiner values of 2.54 to 1.02 for the overstory and seedling strata. Sweetgum (Liquidambar styraciflua L.) and oaks (Quercus spp.) dominated the overstory stratum with 66 percent of the importance values. American hornbeam (Carpinus caroliniana Walter) dominated the midstory and sapling strata with 32 and 42 percent of the importance values. Despite the large tree sizes and ages of several trees, comparison of the forest with models of bottomland hardwood succession and eastern United States old-growth definitions indicate the forest can best be described as old pioneer in transition to changing old growth. INTRODUCTION Interest in locating and understanding the floral and faunal composition of eastern United States old-growth forests has increased considerably over the past two decades (Davis 1993, 1996; Nowacki and Trianosky 1993; Tyrrell et al. 1998 Mitchell et al. 2009). With the aid of new technology, such as geographical information systems and landscape simulation models (Scheller and Mladenoff 2007), previously unknown eastern old-growth forests are identified with each passing year (Stahle and Chaney 1994, Therrell and Stahle 1998, Foti 2001). Concurrent with the need for locating these forests is the necessity to understand their composition and structure (Sheppard and Cook 1988, Martin 1992). This need is particularly acute with the possible rediscovery of the ivory-billed woodpecker (Campephilus principalis) in eastern Arkansas (Fitzpatrick et al. 2005), whose habitat requirements include large, decadent trees in bottomland hardwood forests for foraging (Tanner 1942). Such trees are an integral part of old-growth forests (Davis 1996, Oliver and Larson 1996). Old-growth composition and structure information can also be used to develop strategies for conserving current old-growth forests (White and Lloyd 1995, Mitchell et al. 2009) and for creating structural characteristics that can promote the development of ‘‘new’’ oldgrowth forests (Guldin 1991, Runkle 1991, National Council on Science and Sustainable Forestry 2008, Bauhus et al. 2009). This information is critical for southern United States bottomland hardwood forests where flooding and subsequent sediment deposition patterns result in a variety of topographic features or microtopography (Gosselink et al. 1990, Hodges 1997). These features increase ground surface and soil variability, leading to more complicated compositions and structures relative to upland forests (Wharton et al. 1982, Richardson 1994). *email address: [email protected] Received May 8, 2008; Accepted November 18, 2009. CASTANEA 75(3): 315–329. SEPTEMBER 2010