Opposing states of subtidal habitat across temperate Australia: consistency and predictability in kelp canopy-benthic associations

Abstract

We tested the view that few generalizations are possible about the structure of subtidal algae assemblages and that the situation worsens or does not change as more sites are examined. We quantified the percentage cover of 4 morphological groups of benthic algae (encrusting and articu- lated coralline algae, and foliose and turfing algae) under canopies of kelp Ecklonia radiata and in areas without kelp (macroalgae >4 cm high). This was repeated over 4 spatial scales ranging from quadrats (separated by 10s of m), sites (separated by km), locations (separated by 100s of km) to regions (separated by 1000s of km) across the southern coastline of Australia: Western Australia (WA), Southern Australia (SA) and Eastern Australia (EA). The key result was that while comparison among sites revealed substantial and often inconsistent variation in abundance of benthic algae between habitats (kelp vs open), clear patterns emerged when locations and regions were compared. In EA, where grazers are effective in maintaining extensive areas of encrusting corallines, patterns of algal cover between habitats (kelp vs open) were generally reversed to WA and SA, where grazers are sub- stantially less effective. These results indicate a large distinction in the ecology of these regions (WA = SA ≠ EA) and how lack of understanding of this pattern at the regional scale tends to suggest overwhelming variation when single studies are compared among regions. These differences also highlight that comparisons of studies done at small scales, even if done at several sites in a locality, provide a difficult basis to understand the generality of pattern in algal assemblage structure due to large variation at this scale. While we acknowledge that these broad patterns were not possible to validate at the scale of sites, it was possible to increase the scale of observation to encompass broader patterns that might be organized around a relatively simple set of ecological predictions.