In a rapidly changing world it is essential to understand how biotic communities may respond to environmental disturbances. Being able to predict how these disturbances may affect system-level attributes, such as the overall stability of biotic communities, is of particular interest. Recently, network modeling has gained traction as a powerful tool for studying complex patterns of species' interconnections, but it is underutilized by freshwater ecologists. I am therefore beginning new research on fish co-occurrence networks within dendritic river ecosystems.

Network graphs, which represent species as ‘nodes’ and interconnections as ‘edges’, are an intuitive way to document and visualize species’ connections at the system-level. And graph theory provides a rigorous mathematical framework for quantifying network structure and testing hypotheses on network stability. However, two key knowledge gaps currently limit the utility of network tools in management applications. First, most network studies have focused on discrete mutualistic plant-pollinator or host-parasite interactions, or on antagonistic predator-prey interactions. But associations that are less easily classified (e.g., commensalism) or diffuse (i.e., realized through an intermediary) can also influence community structure and methods to incorporate them are now needed. Second, ecological network theory can predict how an acute extinction event (i.e., node removal) will influence the overall network, but it does not yet predict how ecological networks will respond to chronic environmental disturbances that do not cause acute extinctions.

Recently, I teamed up with ace bio- statistician Joseph Veech (Texas State University) to address these two knowledge gaps – accounting for multiple types of species’ associations and predicting network responses to chronic disturbances – using North American freshwater fishes as a model system. We received a SERDP (DoD) grant to study fish co-occurrence networks in the three regions shown at right.

The project will incorporate an exceptionally large, standardized, pre-existing database of fish co-occurrences (the combined Environmental Monitoring and Assessment Program, and National Rivers and Streams Assessment datasets) within Mid-Atlantic, Mississippi basin, and Pacific Northwest rivers. In each region, fish co-occurrence networks will be built for sampling sites that have been objectively classified, using a standardized scoring system based on a suite of physicochemical habitat parameters, as ‘least’, ‘moderately’, or ‘severely’ disturbed. Different metrics of network stability, including connectivity and modularity will then be used to gauge network responses to environmental disturbance (i.e., comparisons among least, moderately, and severely disturbed sites). A spatially-explicit framework to study and model fish species associations within dendritic river networks will also be created by superimposing spatially-implicit networks on spatially-explicit, digital stream/river maps. (A conceptual diagram of our approach is shown below.) Stay tuned for the results!

Three network graphs built from the same presence-absence matrix (shown as an inset at lower left, with species represented as rows, sites represented as columns, and species’ presences at a given site represented by an ‘x’). The non-weighted, undirected, unipartite graph at upper-left (a) shows spatially-implicit links (co-occurrences) among species (gray circles). A weighted and directed, unipartite graph is shown for the same co-occurrence data at center (b); solid lines indicate (+) links, dashed lines reflect (-) links, and line thickness indicates link strength. The graph in (b) is also spatially-explicit; species’ nodes are located at the centroids of their regional distributions (i.e., locations within the shaded river basin). A spatially-explicit, bipartite graph is shown at right (c); sampling sites (black numbered boxes) are located along the dendritic river network (heavy gray lines) and superimposed species’ nodes are positioned as in (b), but links are only drawn between species and their known locations.