NEON biodiversity explorer

Drought is a nationwide threat to biodiversity.

Addressing how drought affects the natural heritage of the United States is a primary challenges motivating the National Ecological Observatory Network (NEON).

Long-term consequences of drought are unpredictable because each organism responds not only to drought, but also other organisms. Even species insensitive to drought will be influenced if the plants it feeds on or the predator that consumes it respond to drought.

Ecologists could better anticipate drought effects by developing tools that integrate information from many species as they respond both to climate and each other.

We employ a novel generalized joint attribute modeling (gjam) approach to identify how interactions between different species determine the effects of drought.

Detailed descriptions of this project are available in the links at the top of this page.

Coming Soon: downloadable predictive distribution maps




jointly modeling species responses to drought

Large networks generate high-dimensional data measured on many scales. This poses a challenge for analysis because data are multivariate, diverse in form, and contain many zeros. Generalized joint attribute modeling (gjam) allows us to analyze such data because it is joint distribution of all species, and considers all data forms simultaneously (Clark 2016).

Unlike descriptive ordinations, it is fully probabilistic. Unlike structural equation models it accommodates each of the three challenges mentioned above.

First analyses have been applied to traits across eastern North America (Clark 2016). The full model applied to trees and microbiome data is in revision (Clark et al. 2016). Dimension reduction - which allows us to apply the model to hundreds of species simultaneously - is in revision (Taylor-Rodriquez et al. 2016).

  • NEON sites

  • plots

  • species


James S. Clark

Duke University

Robert Dunn

North Carolina State University

Roland Kays

North Carolina State University
N.C. Museum of Natural Science

Allan Gelfand

Duke University

Brad Tomasek

Duke University

Daniel Taylor-Rodriguez

Duke University

Chase Nuñez

Duke University

Bénédicte Bachelot

Duke University