The role of canopy gaps in maintaining vascular plant diversity at a forested wetland in New York State

We studied species diversity and composition of herbaceous layer vascular plant communities under closed canopies and within gaps in a central New York conifer swamp. Our study site, Nelson Swamp, has high documented species richness and supports a large number of state-protected species. Understory cover and species richness at both the quadrat and gap scale were compared among closed canopy areas and gaps in three size classes. To assess differences in the understory environment among the three size classes, we compared light, hydrology, and microtopography. We also tested for correlations between gap species richness and light, wetness, and microsite heterogeneity to determine which aspects of the gap environment might influence vascular plant diversity therein. There was higher quadrat- and gap-level species richness in gaps than in closed canopy areas. Small gaps did not appear to offer a physical environment substantially different from non-gaps, as no differences in their understory communities were apparent. Mid-sized and large gaps had higher species richness, substrate heterogeneity, and water table fluctuation than small gaps. Based on the correlations, the individual factors influencing species richness at the gap scale were depth to water, water level fluctuation, and microtopographic variability, indicating the importance of hydrology over light as a species control. Although gaps and non-gaps shared many of the same species, there were distinct subsets of species that were more important in either type of community. Based on these subsets, closed canopy areas and large gaps can be viewed as two ends of a continuum corresponding to light and hydrology gradients. Out of all the species encountered, 10 percent were found exclusively in gaps, and none were found only under closed canopy. Thus, these gaps are not only sites of higher overall plant growth, but also areas that allow rare species to persist.