Take samples everywhere and you will blow the budget. Take them in the wrong places and you will miss what you came for. This is how to get the balance right.
Where to put your samples: designing an eDNA sampling strategy
Most of the value of an eDNA survey is decided before anyone reaches the water. Deciding where to sample, where to take your subsamples, and how many samples to take, is project specific. It depends on your budget, your goals, and the habitat you are working in.
The general rule is simple. The more samples you take, the more representative your data will be. But samples cost money, and taking them everywhere is rarely the best use of a budget. Placing them strategically is what keeps costs down while still giving you results you can trust. This post walks through how we think about that.
Still water behaves differently from moving water
The first thing to settle is what kind of waterbody you are sampling, because eDNA behaves very differently in still and flowing conditions. That single difference drives most of the design.
In still water, such as ponds and lakes, eDNA stays where it is shed. It forms localised pockets rather than spreading evenly, so your sampling has to go and find it. In flowing water, eDNA travels and mixes as the current carries it downstream. That changes where you place your points, and how far apart.
Ponds and lakes
Because eDNA is more localised in still water, the aim is to cover the range of habitats present.
In ponds and small lakes, try to filter water from every distinct microhabitat, and space your samples roughly every 50 metres around the perimeter.
For larger lakes, aim for at least one sample per 200 metres of shoreline. That density gives you a fair chance of picking up rare or elusive species and of capturing data that represents the whole waterbody rather than one corner of it. It also pays to think about where eDNA is likely to gather. Sheltered bays and water inflows tend to accumulate DNA, so they are worth a sample.
Thermal stratification means that in warmer months, DNA tends to remain closer to the depth/area at which it was shed. Conversely, in cooler months, thermal mixing pushes water, from the centre, out towards the edge. Therefore, sampling in open water, and at varying depth, can yield more data in summer. Whereas in winter, sampling around the perimeter works sufficiently.
If you are working to a tight budget, or the lake is particularly large, subsamples are the way to stretch your effort further. More on those below.
Rivers and streams
In faster moving water, eDNA is carried along and mixed, so DNA from one spot quickly becomes diluted and shared downstream. That means your sample sites should be spread further apart than on still water, to give you a better chance of detecting species that are unique to particular stretches.
How far apart depends on how fast the river is flowing. As a rough guide:
Slow moving streams, with a flow velocity under 0.3 m/s: space samples roughly 100 to 200 metres apart.
Medium rivers, at 0.3 to 0.8 m/s: space samples 200 to 500 metres apart.
Fast rivers, above 0.8 m/s: space samples 500 metres to 1 kilometre apart.
Wherever you are on a river, it is worth sampling above and below any significant tributaries. A tributary can dilute the eDNA already in the main channel, and it can introduce new DNA of its own, so bracketing it helps you make sense of both.
If you are specifically targeting a species you know to be rare or sparse, such as white-clawed crayfish or water voles, then tighten the spacing between samples. Rare species leave a fainter signal, and closer sampling improves your odds of catching it.
Why subsamples matter
Subsampling is one of the simplest ways to get more out of each sampling point. It matters most for biodiversity work. If you are targeting bacteria or tracking pollution it is less critical, but it is still good practice.
The idea is straightforward. Instead of taking one grab of water at a site, you collect several small samples and pool them together into your main sample. Doing this maximises the amount of DNA you gather, and it captures a more representative picture by including the spots where DNA tends to become trapped, such as the pools and riffles of a river.
As a rule of thumb, aim for around 10 to 15 small samples pooled into one sampling bag. You can do this by dipping the bag into the water at several points, or by using a sterilised container or bottle to collect water and decant it into the bag. Sampling the water directly works too: draw up water with the syringe, move along, and draw again from a new spot each time. If you are wading in to sample, always start downstream and work upstream, so you are not carrying DNA off your boots and waders into water you have yet to sample.
Where you take your subsamples depends on your goals and where you can safely reach. In rivers, collect them across the width, taking some from the left bank, some from the middle, and some from the right, and include any nearby riffles or pools. In lakes and ponds, a subsample every 10 to 20 metres along the shoreline works well.
A note on cleaning your kit
eDNA methods are sensitive, which is what makes them powerful and also what makes contamination a real risk. If you are reusing equipment, such as sampling poles, between sites, disinfect them first with a 10 per cent solution of household bleach in tap water. It is a small step, and it stops one waterbody’s DNA turning up in another’s results.
Getting the design right
There is no single correct sampling strategy. The right one balances your budget against your goals and the habitat in front of you. Match your sample density to how much the biology is likely to vary, spread your effort to cover the catchment, and use subsamples to make each point count.
Get those choices right and the sampling itself becomes routine. If you would like a hand designing the approach for your project, get in touch at hello@trace-biomonitoring.com and we will help you plan it.
