Benthic dinoflagellates: Testing the reliability of the artificial substrate method in the Macaronesian region

The suitability of the 'artificial substrate' method, i.e. standardized surfaces of fiberglass screens, for the quantification of four benthic harmful algal bloom (BHAB) dinoflagellates (Gambierdiscus, Ostreopsis, Prorocentrum and Coolia) was tested relative to estimates from natural macroalgal substrates. Sampling took place in a variety of intertidal and subtidal coastal habitats under different water motion conditions, at depths from 1 to 7 m, in two archipelagos of the Macaronesia region: The Canary Islands and Cape Verde. An immersion time of 24 h was sufficient to adequately estimate dinoflagellate abundances. Seven replicates were established as the ideal replication level, considering both reproducibility and sampling effort. In most cases, cell abundances of the four dinoflagellate genera showed lower variability on artificial substrates than on macroalgae, leading to more reliable estimates of abundances. The ratio of mean cell abundances on artificial substrates to mean cell abundances on macroalgae highly varied among sampling sites for each genus. This was especially true for Ostreopsis and Coolia. Thus, given the potentially harmful nature of benthic dinoflagellates, the transformation of abundances expressed as cells g^-1 of macroalgae to abundances expressed as cells cm^-2 is risky, and it should not be attempted in monitoring and management programs of harmful microalgae. In summary, results of this study support the use of artificial substrates in monitoring programs of BHAB dinoflagellates, while the risks of using macroalgae are stressed.