Cross-shelf investigation of coral reef cryptic benthic organisms reveals diversity patterns of the hidden majority

Abstract

Coral reefs harbor diverse assemblages of organisms yet the majority of this diversity is hidden within the three dimensional structure of the reef and neglected using standard visual surveys. This study uses Autonomous Reef Monitoring Structures (ARMS) and amplicon sequencing methodologies, targeting mitochondrial cytochrome oxidase I and 18S rRNA genes, to investigate changes in the cryptic reef biodiversity. ARMS, deployed at 11 sites across a near- to off-shore gradient in the Red Sea were dominated by Porifera (sessile fraction), Arthropoda and Annelida (mobile fractions). The two primer sets detected different taxa lists, but patterns in community composition and structure were similar. While the microhabitat of the ARMS deployment affected the community structure, a clear cross-shelf gradient was observed for all fractions investigated. The partitioning of beta-diversity revealed that replacement (i.e. the substitution of species) made the highest contribution with richness playing a smaller role. Hence, different reef habitats across the shelf are relevant to regional diversity, as they harbor different communities, a result with clear implications for the design of Marine Protected Areas. ARMS can be vital tools to assess biodiversity patterns in the generally neglected but species-rich cryptic benthos, providing invaluable information for the management and conservation of hard-bottomed habitats over local and global scales.

Figure 1

Chlorophyll a concentrations (mean of composite monthly averages over the time frame of the study) for the study regions derived from NASA’s Oceancolor website (https://oceancolor.gsfc.nasa.gov/) derived from the MODIS A satellites at a 4 km resolution.

Figure 2

Methodological approach to ARMS sampling in the central Red Sea.

Figure 3

Species accumulation curves computed per size fractions (Sessile, 106–500 μm, 500 μm) and shelf positions (Near-shore, Mid-shore, Off-shore). Note that we used DNA metabarcoding to characterize communities of organisms <2000 um and DNA barcoding for communities >2000 um.

Figure 4

Differences in average number of OTUs per ARMS among size fractions (Sessile, 106–500 μm, 500–2000 μm) and shelf positions (Near-shore, Mid-shore, Off-shore). The average for the shelf positions was taken from those ARMS that had all size fractions present. The median is shown by the solid line across the box whilst the mean is indicated by the black diamond. Small dots represent outliers. Letters denote those categories, which are based on the DunnTest. n is the number of samples per shelf or size fraction.

Figure 5

The structure (# of OTUs; top row) and composition (abundance of reads; bottom row) of communities at each reef. Taxa were filtered at 1% for the structure and 0.5% for the composition, which means that proportions do not sum to 100%.

Figure 6

PCoA analysis illustrating dissimilarities in community composition based on Jaccard and Bray Curtis dissimilarity matrices of COI. Analysis was undertaken on the full ARMS unit as well as the different fractions (Sessile, 106–500 μm and 500–2000 μm and >2000 μm). Points are coloured according to shelf position.

Figure 7

Ternary plots of similarity [1-D (Dissimilarity obtained using Jaccard)] and the partitions of beta diversity (replacement and richness) for the full ARMS community and the various fractions obtained from the COI metabarcoding and barcoding. Ternary plots are shown for the total experiment as well as within and among sites and within and among shelf positions. Numbers in brackets on the axis labels represent the mean value.

Figure 8

Local Contribution to Beta Diversity (LCBD) values. (A) The individual replicates. Values are depicted for the COI metabarcoding of the full ARMS and the various fractions. The size of the circle is proportional to the LCBD value and red indicates a significant difference. (B) The average LCBD for the full ARMS per reef. The dotted line in (B) indicates the value of LCBD if all reefs contributed equally.