IMPACT OF BEAVER DAM FLOODING ON VEGETATIVE SPECIES COMPOSITION IN CENTRAL PIEDMONT FLOODPLAIN FORESTS

Abstract

The North American beaver is a rodent that alters habitats by creating dams which stop the flow of water in rivers and creeks. This water diverts up and over the adjacent floodplain, flooding the soil and creating a disturbance that can be withstood by hydrophytic plant species with alternative ways to gain oxygen, replacing most species that have no such adaptations. When the dam structure fails after the site is abandoned, fast-propagating r-type species (such as invasives) colonize the area. This species turnover caused by beaver modification is expected to create an increase in both community (or beta) diversity and landscape (or gamma) diversity. While this change in species composition has been studied in the Adirondack Mountains of New York and the foothills of North Carolina, it is yet to be studied in the central Piedmont of North Carolina. I hypothesized that both community and landscape diversity would increase due to beaver modification, as well as invasive species having the highest abundance in those dam-abandoned sites. After narrowing down the study sites to two separate sub-watersheds, I defined the three treatments as active dam (DAM), which had dam and flooding indicators presently and in the past decade, abandoned (ABN), which had no dam or flooding indicators presently but did in the past decade, and reference (REF), which has had no beaver modification as far back as aerial imagery will show. Within the plots of each treatment, species cover was measured at three separate strata: canopy and sapling/shrub/vine (using a transect method), and herbaceous (using a subplot). This species cover data was converted to relative abundance, entered into a Bray-Curtis dissimilarity index, and the composition difference was analyzed statistically and visually. The abundance data was entered into a Chao 1 model for simulated ecosystems to create replications of homogenous and heterogenous landscapes. I then estimated the amount of beaver modification in the two sub watersheds and plugged that value into the heterogenous model to gain a rough estimate of landscape species richness. I analyzed the prevalence index (wetland indicators) of all three treatments and compared them with a Kruskal-Wallis and Dunn’s Test. I used the same statistical tests for the abundance of species classified as invasive between the three treatments. Key Findings:

There was evidence of a significant difference between the plant species compositions of the three treatments. This was most likely between the DAM and REF and/or the ABN and REF sites, as the visual analysis showed no overlap between those pairs. This was translated to a significant increase in community diversity. The estimated species richness of the two sub watersheds was higher than the median species richness of the unmodified landscapes. This indicated signs of an increase in landscape diversity, though more plots would be needed to determine any significance. The prevalence index of the DAM sites was significantly lower than the REF sites. The lower the index, the more flood tolerant species tend to be. This was consistent with the plants in DAM sites having higher flood tolerance than plants in the REF sites. The abundance of invasive species was significantly higher in the abandoned sites compared to the other two treatments. This was consistent with the research of Japanese stiltgrass preferring open-canopy, post-disturbance patches with lots of nitrates and urban runoff.

Key Recommendations:

More plots should be completed among more sub-watersheds in the central Piedmont. This would allow for a statistical analysis to be done on landscape diversity, as each replication could have its own independent group of plots, satisfying the independent sampling requirement of statistics. Dam restoration efforts should be completed at sites that have shown evidence of being abandoned. This would sustain the high-water table and prevent invasive species from establishing and spreading out along the bare, moist soil. This stabilization would also prevent the decreased plant and wildlife diversity that would result from invasive species spread. Dam creation efforts should be considered in the next two decades as the effects of climate change continue. The temperature of the Piedmont is projected to increase by 0.86 decreases Celsius by 2040. This would likely cause a change in species composition along the temperature gradient, as competitive species may have more success against flood tolerant species in environments less wet than the present. Dam creation as a management strategy can artificially create inundated ecosystems where not only flood-adapted species, but also habitat diversity can sustain.