Monitoring marine biodiversity in hard-bottom habitats is challenging as it typically involves resource-intensive, non-standardized, and often destructive sampling methods that limit its scalability. Differences in monitoring approaches furthermore hinder inter-comparison among monitoring programs. Standardised collectors such as Artificial Reef Monitoring Structures (ARMS) can be used to monitor status and changes of hard substrate communities in coastal environments. In addition, ARMS constitute an early-warning system for marine biological invasions by identifying newly introduced Non-Indigenous Species (NIS) and track the migration patterns of already known NIS in European continental waters.

In the framework of ASSEMBLE+ project and as part of the European ARMS programme (ARMS-MBON) (Obst et al. 2020), ARMS were deployed in two locations in Greece and, more specifically, in the marina of the Old Venetian Harbour of Heraklion (1HERP) and in the Underwater Biotechnological Park of the Hellenic Centre for Marine Research (2UBPC).

The ARMS deployment and retrieval dates are shown in Table 1; deployment and retrieval were done according to the standards and protocols established by the Smithsonian Institution. Upon retrieval, the plates from the ARMS were disassembled, photographed (Fig. 1a) and samples of both the motile and sessile communities were collected for molecular analysis. Each sampling event produced three fractions (sessile, motile 90–500 μm and motile 500 μm−2 mm) as well as a stack of plate and specimen images.

a

b

Figure 1.

a. Example of an ARMS plate with bryozoan formations (Reteporella sp.) b. Comparison of image-based and eDNA metabarcoding based results, for the taxa identified at the species level, from the ARMS that was retrieved on 2019-01-28.

DNA was extracted from the sampled fractions and amplified by PCR, targeting different molecular markers (18S rRNA, COI and ITS). Resulting amplicons were sequenced using Illumina MiSeq Reagent Kit v3 (2 × 300 bp) and analyzed using PEMA (Zafeiropoulos et al. 2020). All raw sequence files of this study were submitted to the European Nucleotide Archive (ENA) (Harrison et al. 2021) with the study accession number PRJEB33796.

Images were analyzed using photoQuad image processing software (Trygonis and Sini 2012), which is specialised for the analysis of sessile biodiversity on photoquadrats. Cover of sessile taxa was measured by superimposing on every image 100 randomly distributed points. Each point was manually assigned to the corresponding taxon or morpho-functional category based on external morphological characters.

Repeatable workflow procedures for integrated processing of image and sequence data are currently under development as part of the LifeWatch-ERIC Internal Joint Initiative on NIS. In addition, all ARMS-related data are stored in the ASSEMBLE Plus data collection of the Marine Data Archive (MDA) using a Darwin Core Archive (DwC-A) format, including the linkages to the images and sequences (Exter et al. 2020).

Comparison of traditional biodiversity assessment methods, such as image-based identifications, complemented by the eDNA metabarcoding results, will shed light on the investigation of marine biodiversity patterns in the Eastern Mediterranean Sea (Fig. 1b). Furthermore, the results will provide crucial information on the importance of ARMS for biodiversity assessment and as an efficient tool to monitor community shifts and invasion events in marine ecosystems undegoing fast change.

Christina Pavloudi

1st DNAQUA International Conference (March 9-11, 2021)