In marine waterbodies, less is currently known about how hydrological systems affect eDNA transportation and distribution. However, it has been shown that communities obtained from marine eDNA metabarcoding are highly representative of the immediate local habitat where the sample was collected, both on horizontal and vertical planes. This means that when samples are collected in a transect going from shore to offshore, different communities will be detected, sometimes even within the range of tens of meters. As in lakes, vertical stratification of the water (as a result of thermoclines) restricts mixing of eDNA, meaning that water samples should be collected from each depth zone of interest to fully characterise the marine communities at the sampling location.
eDNA in marine systems is generally much more dilute compared to that in freshwater systems. This is in part dependent on the target group or species but is particularly pertinent for larger vertebrates (extra-organismal vertebrate eDNA). Planktonic or microbial taxa usually require smaller volumes. Therefore, sample volume should be maximized to be representative of the environment and the taxa that are targeted. Each sample should be at least 2 L volume, and the volume of water filtered should be in the range of 2-5 L (more is always better). Turbidity is usually less of a problem in marine water, although inshore areas (e.g. mangrove forests, marinas, areas with high population density) can become turbid due to coastal run-off and wave action disturbing the sea floor. In this case, filter as much water as possible from each sample until the filter completely clogs.
Sample number will depend on the spatial scale of the study or monitoring project. In order to characterize a community or compare sites, a minimum of 20 samples, even for relatively small areas, is strongly advised. This usually involves collecting independent samples (rather than subsamples) spread out across the sampling area.
Sampling design will equally be dependent on the size of the sampling area, and also on the type of ecosystem (coral reef, mangrove forest, pelagic, etc.) that is targeted. Season (and even time of day) may need to be taken into consideration, as many fish species move to (in)shore areas for mating and spawning and move to deeper or warmer waters in winter, while other species may prefer cold, deep water during summer. Thus, it is important to consider migration patterns as well as mating and spawning sites.
When sampling estuaries, seas, or oceans: