The network geometry of rivers can be classified as dendritic, with a greater number of branches encountered as you move up through the network. This physical structure regulates the dispersal pathways of resident biota. Additionally, functional connectivity along the network is temporally dynamic due to natural (intermittence of flow, environmental variability) and anthropogenic (dams, pollution, water extraction) processes. Both the branching structure of rivers and temporal variability in connectivity can influence patterns of biodiversity by changing the relative influence of metacommunity organising processes such as mass effects and species sorting. Here we asked how variability in temporal connectivity affects diversity across a gradient of network topologies?
We use a graph-theoretic approach to stochastically build dendritic networks where nodes represent local discrete habitat patches (e.g. pools or riffles) connected by edges. We developed a discrete-time niche-weighted model of a competitive metacommunity to simulate community dynamics across a range of network structures (linear – bifurcating) and with a range of temporal connectivity scenarios (barrier imperviousness, barrier location and length of barrier placement).
Results/Conclusions
A loss of temporal connectivity was associated with a reduction in mass effects resulting in lower alpha diversity. The results associated with a loss of temporal connectivity were least pronounced in bifurcating networks and greatest in linear networks. When barriers were placed on least connected edges (typically those in headwaters) there was a greater loss of alpha diversity than when barriers were placed on most connected nodes (typically high order channels). Our findings suggest that consideration needs to be given to both network structure and temporal dynamics when considering how connectivity influences freshwater metacommunities. Our findings are of particular relevance to restoration practitioners where the “field of dreams” approach has often been employed with varying success.