​Current Research

• Legume-rhizobia symbioses: Bacteria (rhizobia) live in the root nodules of legumes and fix atmospheric nitrogen, converting it into bioavailable forms that legumes can utilize. In exchange, legumes provide shelter and sugars produced during photosynthesis to rhizobia. I'm exploring the magnitude of the contribution of symbiotic nitrogen fixation to the nitrogen cycle, and the potential for its use as a mechanism of restoration in the enhancement of understory productivity and biodiversity in LLP savannas. I will establish how much variation is seen in nodule characteristics and rhizobial symbiont identity across multiple legume species and sites within the LLP savannas of the North Carolina Sandhills ecoregion. Furthermore, I will address host specificity and how nitrogen deposition may impact legume-rhizobia symbioses and how they may be used as a tool to help aid in restoration​.

​

• NPKDNet: I'm exploring how drought and nutrient enrichment interact to affect:
  1. plant communities (diversity & community composition)​
  2. ecosystem function (ANPP & arthropod communities)

 

• DRAGNet: I'm studying how anthropogenic disturbance and recovery from disturbance interact with nutrient enrichment to affect:
  1. plant communities (diversity & community composition)
  2. ecosystem function (ANPP)

---

Previous Research

Throughout the southeast United States, the long-term overharvest of longleaf pine (Pinus palustris) trees together with decades of fire suppression has reduced the once vast and biodiverse longleaf pine (LLP) ecosystem to a small fraction of its previous distribution and health. The high density and diversity of understory plant species in the LLP ecosystem makes it a priority for conservation and restoration. In this study, I used a multi-scale approach to identify key factors controlling the variability in biodiversity and understory dynamics in North Carolina (NC) LLP savanna, and address the potential for the dominant grass species, Aristida stricta, to structure the understory plant communities and drive biodiversity. There were 15 study plots across North Carolina spanning from the Sandhills, eastward to the Outer Coastal Plain.

---

A single plot (625m2) was established at each of the 15 sites in LLP-dominated stands. Within each 625m2 study plot, 25 subplots (each 25m2) were arranged in a grid. Inside each subplot (n=25), a 1m2 species composition quadrat was established. All herbaceous understory plant species rooted within the 1m2 quadrats identified and percent cover estimated. Late in the growing season, ANPP was estimated by clipping all aboveground herbaceous biomass in one 0.1m2 quadrat nested within each 1m2 quadrat (n=25). All plants within the nested quadrat were clipped at ground level with scissors and sorted by growth form and live versus dead tissue (grass, woody, forb, past year’s dead, and LLP needle litter).

Overall, I found large variability in plant community characteristics and function across the 15 sites. The relative cover of the dominant grass species, A. stricta, varied greatly within and across plots, ranging from 0% to 100% relative cover at the subplot level and 0% to 85.46% at the plot level, with most subplots having ~30% relative cover. Ecosystem function, measured here as ANPP, ranged from 0g/m-2 to 493.8g/m-2. Richness was also variable, ranging from 1 to 17 species per subplot (Figure 2c), and from 10 to 49 total species per plot

This study also found a unimodal productivity-biodiversity relationship, with richness peaking at ~150g/m2. The unimodal direction of this productivity-biodiversity relationship may be particular to LLP-A. stricta communities, as studies in LLP-A. beyrichiana communities have reported a positive relationship (Kirkman et al. 2001). Subplots containing wiregrass were on average more biodiverse than subplots lacking wiregrass, however, too much wiregrass cover can have a negative effect on biodiversity. In LLP understory plant communities, the grass and forb species are what contribute to the high diversity of the understory, and so when addressing the richness of just graminoid and forb species, the ideal cover of wiregrass at a small scale is ~16%. Managing for wiregrass cover is an important component for restoring LLP understories to high levels of diversity. Wiregrass must be present in the understory; however, the level of wiregrass must be kept relatively low (~16%) in order to ensure space and resource availability for other grass species and forbs which contribute greatly to biodiversity. Land managers can potentially use wiregrass presence and abundance as a predictor of biodiversity dynamics on their land. This will lead to more frequent understory monitoring as managers will be able to eliminate the exhaustive and high-skill requiring, biodiversity surveys.
​
​

​

---

​The effect of ANPP on graminoid and forb richness (a), the effect of relative wiregrass cover on graminoid and forb richness (b), and the effect of relative wiregrass cover on ANPP using polynomial regression (c). Thin black lines represent the trend for each plot (n=15) and the bold lines represent the overall trend across plots.