As we continue to use eDNA in marine ecosystems, one large, lingering question remains: where, and how often, should we sample? The ocean is vast, covering most of the globe and spanning across immense spatial scales and almost unfathomable depths. Even relatively tiny areas of interest, such as 1 km x 1 km marine protected areas (MPAs), contain several different habitat types across hundreds of meters of depth. If you sample 1 liter in one corner of an MPA at one depth, is that eDNA sample truly representative of the biodiversity within the entire MPA? While eDNA can improve the scale of monitoring efforts because of its sensitivity (1-3), it is unclear how much of a region’s biodiversity is reflected in one eDNA sample. Therefore, knowledge about the spatial and depth variation of eDNA signatures is crucial to figure out how many bottles of water to take and where we should collect to best maximize biodiversity monitoring efforts.
We set out to fill this gap in knowledge in our recently published study where we collected eDNA samples along a fine-scale depth gradient (0 m at the surface to 10 m at the ocean bottom) in a kelp forest ecosystem (i.e., where you SCUBA dive) to assess how eDNA signatures vary across depth. We also took samples from an adjacent surf zone habitat (i.e., where you surf) to see how eDNA varies across horizontally distinct habitats.
We detected 71 species of fish, including many of my favorite marine species like the Shovelnose Guitarfish and California Butterfly Ray. Importantly, we found distinct differences in fish communities across depth. eDNA from surface water communities had fewer total species, and the composition of the communities differed between surface and depth. Most interestingly, this depth variation matched the known ecologies of the fishes: eDNA from pelagic, surface-dwelling species proportionally dominated the surface stations, and eDNA from demersal, bottom-dwelling species proportionally dominated the deep stations. For example, California Grunion and Pacific Anchovy were more abundant at the surface, where they typically like to spend time feeding, whereas Pacific Angel Shark and sanddabs were more abundant at the sea floor, where they like to hang out and bury in the sand.
We also found similar results across horizontally distinct habitats, where species who like to hang out in the surf zone, like the California Corbina and Yellowfin Croaker, were more abundant there, and kelp-dwelling species, like the Kelp Bass and Barred Sand Bass, were more abundant in the samples from the kelp forest. Overall, our results show that eDNA captures ecologically relevant depth and spatial variation of fish species, indicating that eDNA is an accurate snapshot of local marine biodiversity.
Additionally, we found differences in species assemblages across days. The biggest differences were in some of the migratory, highly mobile fishes that swim in and out of nearshore ecosystems, such as Pacific Sardine and Pacific Anchovy. However, it’s important to note that the depth and habitat associations were largely consistent – demersal fishes were more abundant in demersal samples and surf zone fishes were more abundant in surf zone samples across the three sampling days. So, ultimately, the depth and spatial variation we observed was largely consistent over time.
Although our study strongly suggests narrow transport of eDNA in marine ecosystems, there is still a lot to be learned about the fate and transport of eDNA. In particular, we need to reconcile the fact that “fish in bucket” experiments suggest eDNA lasts for days to weeks, while our results from the field suggest eDNA isn’t even transporting across 5 m of depth or 150 m of space. We hypothesize that this is possible if eDNA is constantly being generated, and this recently generated eDNA swamps out the signals of past-generated, degrading eDNA. This is definitely a potentially fascinating line of research going forward.
So, what does this mean for sampling? Well, firstly, it adds to a growing list of evidence that eDNA is a reliable tool to survey hyper-local biodiversity. Our results suggest eDNA is biased towards the species’ whose eDNA was captured within or near that location. On the other hand, this strong depth and spatial variation does mean that samples collected at one location may only reflect the species at that location and may not reflect the biodiversity of other depths or habitats. If you sample in the surf zone, then you are sampling fish immediately within that habitat, not so much the fishes in the kelp forest 150 m offshore. To get a full picture of biodiversity in areas such as MPAs, we would need to sample across depths (at least surface vs. bottom), across habitat types (subtidal kelp forest, sandy bottoms, surf zones, etc.), and across multiple time points, especially if there are big seasonal biodiversity patterns. This increases the scope of sampling needed for eDNA-based assessments – 1 or 3 L of water at knee-depth height is not reflective of the total biodiversity even within a small MPA. However, one advantage of eDNA is that sampling costs are low enough (~$50 a sample) that increased sampling efforts are likely not cost prohibitive. This is especially true compared to the costs of big boats, personnel, and equipment needed for SCUBA surveys. An added bonus is that the cost of sequencing continues to become cheaper over time, which will only make eDNA more affordable in the future.
So, back to our original question, how frequently and where should we sample with eDNA? The answer, as is the response to basically every question in ecology, is “it depends”. If the goal is to maximize biodiversity with 100 total samples, then you will need to sample multiple depths, multiple habitat types, and across multiple days. If you especially want to capture all the rare species, you should definitely keep on sampling harder. However, if the goal is to compare the dominant rocky reef fish communities inside and outside an MPA, then the most important factor is to be consistent across depth and directly sample the rocky reef, on SCUBA or with a Niskin. Ultimately, our results, and others, suggest eDNA is a snapshot local biodiversity across depth, space, and time.
References:
1. Ely T, Barber PH, Man L, Gold Z. Short-lived detection of an introduced vertebrate eDNA signal in a nearshore rocky reef environment. PLOS ONE. 2021;16: e0245314. pmid:34086697
2. Jeunen G-J, Lamare MD, Knapp M, Spencer HG, Taylor HR, Stat M, et al. Water stratification in the marine biome restricts vertical environmental DNA (eDNA) signal dispersal. Environmental DNA. 2020;2: 99–111.
3. Port JA, O’Donnell JL, Romero‐Maraccini OC, Leary PR, Litvin SY, Nickols KJ, et al. Assessing vertebrate biodiversity in a kelp forest ecosystem using environmental DNA. Molecular Ecology. 2016;25: 527–541. pmid:26586544