Changes in plant species composition along an elevation gradient in an old-growth bottomland hardwood-Pinus taeda forest in southern Arkansas1

Journal of the Torrey Botanical Society, Jan-Mar 2005 by Grell, Adrian G, Shelton, Michael G, Heitzman, Eric

Indirect gradient analysis. The variation explained by the two axes in the overstory (86.4%), seedling (90%), and herbaceous (85.5%) ordinations was much higher in this study than in others. Grace et al. (2000), studying influences of environmental conditions on vegetation in the Texas coastal tallgrass prairie, found only 42% of the variation explained by a two-dimensional ordination. McCune et al. (1997), examining lichen communities throughout southeastern United States, identified only 59% of the variation explained by a two-dimensional ordination. The considerable differences observed between our study and other studies most likely are attributed to using elevation as a criterion for randomly locating sampling plots, the small area represented by the Lost Forty, and the distinct difference in community types. Our technique assured that all cells (1.5 m) within the overstory plots were within the specified elevation class.

In contrast to other strata, no useful ordination was found for the sapling stratum. Only eight species were reported for saplings after elimination of rare species. Moreover, the sapling stratum was dominated by Carpinus caroliniana, Ilex opaca, and Nyssa sylvatica, which accounted for 70% of the importance in the low class, 85% in the mid-elevation class, and 45% in the high class. The dominance of these species, coupled with the overall low species diversity, suggest that the sapling stratum is biologically dissimilar to the other strata across all elevations. The uniqueness of the sapling layer is also supported by the direct ordination. There were only two species with significant differences among elevation classes for saplings, compared to seven species for both overstory and herbaceous layers and 13 species for seedlings.

Results of the indirect gradient analysis generally support findings from the direct gradient analysis. For example, overstory species characteristic of higher elevations in the direct gradient analysis, such as Tilia americana and Ostrya virginiana, were found clustered together on the right side of the graph of the indirect gradient analysis, which was indicative of higher elevations. Species such as Taxodium distlchum and Diospyros virginiana were clustered on the left side of the graph, which reflects lower elevations. Interestingly, environmental variables in the seedling and herbaceous ordinations revealed much weaker correlations with elevation than in the overstory ordination. Heitzman et al. (2004) propose that the closed canopy of the Lost Forty limits the successful establishment of many species in the understory, especially those that are shade intolerant. Furthermore, flood duration and frequency have been identified as the cause of mortality in ground flora layers because of saturated soil conditions (Barnes 1978, Titus 1990). While some understory plants might establish themselves in marginal microsites during periods with low flooding, mortality of these species would likely occur when major flooding resumes. Conversely, the overstory species present at the Lost Forty have survived to maturity over many years of variation in environmental conditions, which could explain the higher correlations of the environmental variables for this stratum.