SESYNC-LTER post-docs selected

New metrics to identify compositional changes using rank abundance curves 

Meghan Avolio, University of Utah

As a plant ecologist, I am deeply interested in patterns of community change, particularly the impacts of global change drivers on communities and its consequences for ecosystem function. Ecologists typically track community changes by simplifying community structure to single variable indices of diversity that can easily be compared across systems, but result in much of the information being lost. Alternatively, ecologists can use abstract measures that take into account detailed compositional changes, but lack sufficient information to explain how the community changed. Lastly, studies that detail changes in community composition are highly species-specific, which limits their generality across ecosystems.

During my fellowship at SESYNC I will develop new metrics for identifying compositional changes in generalizable ways using rank abundance curves (RACs), which display the abundance of each species in the community. RACs can change in four ways: curve shape, species re-ordering, and gain or loss of species. Other than methods comparing shapes of RACs, there are no rigorous statistical tests available to compare RACs, limiting their usefulness and our understanding of community changes. For example, studying the degree to which global change drivers result in species re-ordering gives essential information that species richness cannot detect.

I previously received funding from the LTER Network Office to lead a working group with Dr. Kimberly La Pierre examining patterns of community convergence or divergence in response to global change drivers. Our first manuscript, in press at Ecosphere, conceptualizes a framework to integrate study community changes among and within treatments with changes in RACs. We also hypothesize how changes in RACs might be reflected in other ways community patterns are measured, which I will also directly test at SESYNC.

To accomplish my objectives, I will run simulations and analyze data from 82 global change experiments, including 32 LTER experiments—products of the previously-funded working group. I will be working with Drs. Greg Houseman and Scott Collins, and a statistician who is yet to be determined, and my goal is to ultimately produce an R package that will enable ecologists to rigorously test changes in RACs and detect changes in the entire plant community.

Effects of functional neighborhood, climate and geographical location on tree population dynamics

Jenny Zambrano, University of Illinois at Chicago

Identifying the mechanisms that shape natural communities is a major challenge in community ecology. Climatic conditions and local neighborhoods have been described as important filters for selecting a subset of species with traits that are fit for a particular site, thereby driving community assembly and dynamics. Evidence for both filters comes from quantifying how the traits of neighboring species influence individual performance.

While there have been investigations of neighborhoods and traits at local sites and across regions with and without trait data, there is no information on how individual species respond and whether those responses differ across species ranges. The relative influence of climate and biotic neighborhoods varies across a species’ range and should be considered when interpreting relationships between trait neighborhood and performance. However, this has not been happening.

We propose to develop neighborhood models that locate not only the individual, but also processes affecting its performance. These models will allow us to understand the interaction between individual plant traits and the environment at small and large scales by using data from Long Term Ecological Research (LTER) sites and the US Department of Agriculture’s Forest Inventory and Analysis (FIA) Program from across the eastern United States and Puerto Rico. We are guided by the following questions:

 1) Are trait neighborhood-performance relationships related to the position of an individual in its geographic or climatic range?

We will quantify whether the neighborhood-performance relationship of a species in a LTER site is related to where exactly that population is located in its geographic or climatic range. For example, the population of species 1 at the LTER site may be near the edge of its range, while the population of species 2 in the same site may be near the middle of its range. Our expectations for how the trait similarity of neighboring individuals will influence growth and mortality rates of focal individuals from those two species are different.

2) Is inter-specific trait-based density dependence stronger in the tropics?

The presence of temperate and tropical sites in the LTER network also allows us to address whether neighborhood-performance interactions are related to latitude. Previous work has shown that intra-specific (i.e., within the same species) density dependence increases in a negative direction towards the tropics, thereby allowing more species to co-exist, but we have no clear tests of whether trait-based density dependence is stronger in the tropics within the same species. We will quantify tree neighborhood-performance relationships in northeast US LTER sites and in the Luquillo (Puerto Rico) site to determine whether individual tree performance is more strongly related to local trait differences in the Luquillo site compared to the northeastern temperate sites.

3) How do trait neighborhood-performance relationships change within species across their geographic or climatic ranges?

To address this question we will expand the scope of our research to incorporate the USDA FIA dataset, which will enable us to quantify neighborhood-performance relationships for species across tens to hundreds of populations across its geographic range.

In light of future global change, it is imperative to understand the effects of biotic (i.e., living) and abiotic (non-living) conditions on the relationships between trait neighborhood and performance to predict future changes in community dynamics. Results of this study will provide a general understanding of the mechanisms shaping plant communities, allowing us to assess potential changes in community dynamics relevant for developing future management strategies.