Metabarcoding hyperdiverse kelp holdfast communities on temperate reefs: An experimental approach to inform future studies

Abstract

Classical taxonomic approaches to quantifying biodiversity can be notoriously laborious and restrictive. Instead, molecular metabarcoding is emerging as a rapid, high‐throughput, and cost‐effective tool to catalog biodiversity. Despite the appeal of metabarcoding, methodological and procedural biases must be understood before robust biodiversity inferences can be made. Here, we use CO1 metabarcoding to characterize marine eukaryote communities associated with Ecklonia radiata, the dominant eco‐engineering kelp of temperate Australasia. To establish a standardized and reproducible community metabarcoding protocol, we examined the influence of different sample preparation, laboratory, and bioinformatic steps on inferences of species richness and composition of communities associated with E. radiata holdfasts (the root‐like structure anchoring the kelp to the substratum) sampled from northeastern New Zealand. Specifically, we examined the effect of sieving the community into different size fractions and the replicability of results across DNA extractions, polymerase chain reactions and sequencing. Overall, we found that sieving the community into two size fractions before DNA extraction enabled detection of a greater diversity of taxa than not sieving samples. When compared with traditional morphology‐based inventories of kelp holdfast biodiversity, we found that although the taxonomic precision of our metabarcoding approach at the species and genus level was limited by the availability of reference sequences in public repositories, we recovered ~40% more taxa and a greater taxonomic breadth of organisms than morphological surveys (e.g., 18 phyla as compared with 14 phyla). On the basis of our findings, we provide methodological guidelines for the use of metabarcoding as a tool for surveying and monitoring the hyperdiverse species assemblages associated with kelp holdfasts.