I am a quantitative ecologist interested in building mechanistic models to forecast population and community responses to environmental change. I try to balance mathematically rigorous analyses with biological sensibilities to produce models that synthesize a variety of data sources and assumptions, while highlighting which parts of our inference derive from each.
Some Current and Recent Projects
Demographic distribution models
I'm using demographic information to predict species ranges in a variety of projects. To understand a species' response to a changing environment, it is critical to relate demographic heterogeneity back to its environmental drivers. This includes proteas in South Africa (paper already out), invasive plants in New England (see my talk at ESA this year), trees in the western US (this one turns out to be pretty tricky), and trees in the eastern US (just getting started).
Developing new tools for range modeling
I'm developing new tools for distribution modeling on a variety of fronts. With the BIEN (Botanical Information and Ecology Network) group, I'm developing rule sets to build the highest quality range models possible when working with large numbers of species (~200,000), where individual tuning is impractical. We'll be building range models for all plants in the New World. With colleagues at UConn, I're developing methods to included spatially explicit prior information into Maxent models, for which we've come up with about seven new applications. With colleagues at Microsoft Research , I'm helping to develop active learning methods to inform occurrence sampling strategies with optimal information content. Our goal is to connect this to crowd-sourced data collection.
Maximum entropy and ecological theory
I've always been intrigued by the predictive success possible with maximum entropy models in ecology. In the last decade, maximum entropy models have been applied to ecological patterns in a variety of ways: macroecology (cf. John Harte), community assembly (cf. Bill Shipley), ranges (cf. Steven Phillips), and food webs (cf. Richard Williams). At the moment, I'm working on the most thorough tests of the maximum entropy theory of ecology within a single system (the Cape Floristic Region of South Africa). We're also developing an R package to facilitate other's using the theory. I've been ruminating for some time on alternative theoretical applications of maximum entropy models in ecology, which I'd love to discuss with anyone interested.
Promoting understanding of ecological models
Collaborators: Johan P. Dahlgren, C.J.E. Metcalf, Dylan Childs, M.E.K. Evans, Eelke Jongejans, Sydne Record, Mark Rees, Roberto Salguero-Gomez, Sean McMahon, Matthew Smith, John Silander, Jane Elith, Wilfried Thuiller, Niklaus Zimmerman, Tom Edwards, Antoine Guisan, Signe Normand, Rafael Wuest
As I've learned the intricacies of many of the models that I've used, I've translated that understanding in to a variety of published guides. These projects dive into the decision making process inherent in all models to help researchers understand the biological implications of different modeling choices. These include a handful of papers already out: Advancing demography in ecology with integral projection models: a practical guide, A practical guide to Maxent: What it does, and why inputs and settings matter, A comparison of Maxent and Maxlike for modeling species distributions. Others are forthcoming: 'Back to the basics of species distribution modeling: what do we learn from complex versus simple response curves?' and 'Confusing geographic aggregation with complex environmental response and overfitting in species' range models.'
Community abundance patterns in South African fynbos
This work is part of a larger collaborative effort focused on understanding the different aspects of biodiversity in plant communities in the Greater Cape Floristic Region (GCFR) of South Africa. The larger project seeks to understand the relationship between functional traits and patterns of genetic, functional and taxonomic diversity. Our goal is to integrate knowledge collected at these different scales to predict species and community responses, enhanced by understanding their evolutionary past. I'm focusing on determining how community-level distributions of functional traits vary along ecological gradients and how this variation can predict species' abundance patterns. Here are some brief articles about the project: link, link.