Do surrogates describe patterns in marine macroalgal diversity in the Recherche Archipelago, temperate Australia?

Abstract

Abstract Surrogates aim to predict species diversity and to minimize sampling effort. Here the value of surrogates for marine macroalgae is tested. Higher taxonomic levels and dominant taxa were evaluated as surrogates for detecting patterns in macroalgal species diversity (derived from species‐level biomass data) in the Recherche Archipelago, Western Australia. Data were stratified by two island groups (inshore and offshore), three depth intervals (<10 m, 10–20 m, >20 m), and two exposures to wave energy. Correlations between similarity matrices from surrogate and species diversity data matrices were analysed using a modified Spearman rank correlation (rs). The ability of surrogates to detect differences between exposures to wave energy was also investigated using analysis of similarity. Species diversity data were aggregated to higher taxonomic levels and were either fourth‐root or presence/absence (richness) transformed. Species richness was the most consistent surrogate (rs‐values>0.69, P=0.001). Genus‐level richness was also strongly correlated to species diversity in the Recherche Archipelago, except in depths <10 m where patterns in diversity were characterized by species turnover. At offshore islands in depths 10–20 m and >20 m, family‐ and order‐level biomass data were suitable surrogates owing to the abundance of Alariaceae (Order Laminariales) represented by a single genus and species. Surrogates using biomass data from the orders Laminariales, Ceramiales, Dictyotales and Fucales were also successful at offshore islands in middle to deeper depths. At inshore islands in depths 10–20 m and >20 m, genera from the orders Gigartinales, Fucales and Ceramiales were strongly correlated to species biomass data. Surrogates, like species richness, that were strongly correlated to species biomass data were also able to distinguish between exposures to wave energy, based on analysis of similarity tests. Sampling effort would be reduced if collecting richness data in species‐rich macroalgal assemblages such as those found in temperate Australia. Copyright © 2006 John Wiley & Sons, Ltd.