Broad-scale analysis of fish community data suggests critical need to support regional connectivity of coral reefs

Graeme Cumming & David Bellwood

Abstract:

Connectivity is vital for the biodiversity and functioning of marine ecosystems. It is known to be important for coral reefs, but the scales at which connectivity effects matter – and correspondingly, the scales at which management responses are needed - are poorly understood in marine systems. We used 23 years of fish monitoring data collected from ~50 different coral reefs by the Australian Institute of Marine Science, together with a range of geographic data layers (including the Allen Coral Atlas) and additional network analysis, to explore the balance of local and regional influence on fish communities. Variance partitioning indicated that 42% of the variance in fish community composition can be explained by regional effects or their interaction with coarse-grained local influences (habitat). The variance explained by regional influences was divided evenly between measures that capture location on environmental gradients (e.g., proximity to coastal shelf, latitude) and cross-scale centrality measures of reef location within a broader reef network. A total of 11% of variance could be directly or indirectly attributed to management. Our results provide clear evidence that management and restoration of reefs across the globe must consider both local and regional influences on reef-associated organisms, and highlight potential benefits of improving connectivity in human-dominated coastal seascapes.