Environmental DNA (eDNA)

What is eDNA?

eDNA is genetic material collected directly from the environment rather than from a specific organism. A water sample from a river, a scoop of soil from a forest floor, or even a cubic meter of filtered air can contain DNA fragments which can be sequenced to identify which species have been present, within the past few days or weeks.

The method works because organisms constantly shed biological material into their surroundings. A fish releases skin cells, mucus, and waste into the water it swims through. A mammal leaves hair, footprints, and fecal matter on the land it travels across. A bat exhales into the air. Each of these traces carries DNA that can be extracted, amplified, and matched against reference databases of known species.

The DNA degrades relatively quickly, typically within two to four weeks in water, depending on temperature, UV exposure, and microbial activity. This means a positive detection is generally a sign of recent presence, not just a historical trace. It also means sampling needs to be timely, and the logistics of collection, preservation, and laboratory analysis matter a great deal to the quality of results.¹

Why does eDNA matter for conservation and rewilding?

Traditional wildlife surveys are expensive, slow, and limited by what observers can see, hear, or trap. They tend to miss species that are rare, nocturnal, aquatic, or simply good at hiding. eDNA analysis changes that. A single water sample from a pond can confirm whether great crested newts are present with a 99.3% detection rate, compared to 75-76% for torch counts or bottle trapping. You do not need to see the animal. It just needs to have been there.²

For rewilding, this is particularly useful at two moments: before a project starts, to establish what is already present in an area, and after, to measure what has returned. Both are difficult to do well with conventional survey methods at landscape scale. eDNA can survey an entire river catchment's fish community from a handful of water samples. It can detect whether beavers are using a particular stretch of river before their lodges or dams are visible. It can confirm the presence of elusive predators, for example wildcats, pine martens, water voles, in terrain where camera trapping could require hundreds of units over months.

A 2024 systematic review published in Frontiers in Conservation Science found that eDNA metabarcoding is particularly promising for monitoring terrestrial rewilding projects because it can generate whole-community datasets across multiple species groups from a single sampling effort. No conventional survey method can also do this. The same study noted that most eDNA rewilding research so far has focused on aquatic systems, and that terrestrial applications are still catching up.³

How eDNA surveys work

The basic process is the same whether the sample comes from water, soil, or air. Samples are collected and concentrated, DNA is extracted, specific DNA sequences are amplified using PCR (polymerase chain reaction), and then the amplified product is sequenced. The resulting data is compared to a reference database to identify the organisms the DNA came from.

There are two main approaches. Targeted eDNA uses species-specific primers to look for one species at a time. This method is useful when you need high-sensitivity confirmation that a particular animal is present or absent. eDNA metabarcoding uses broader primers to capture and sequence everything in the sample simultaneously, producing a community snapshot that can identify dozens or hundreds of species at once. The tradeoff is that metabarcoding can miss rare species that targeted qPCR would catch, because the rare signal could get drowned out in the noise of more abundant DNA.²

The quality of results depends heavily on two things: how well the samples are collected and preserved, and how complete the reference database is. If a species has never had its DNA sequenced and deposited in a reference library, eDNA cannot identify it. The sequence will show up as an unknown species. This is a genuine limitation, particularly for invertebrates, fungi, and microorganisms, where reference databases remain patchy.

Where eDNA is being used

Freshwater systems are where eDNA has seen the most uptake, partly because water is such an efficient collector and carrier of biological traces. In the UK, eDNA testing for great crested newts became routine enough that Natural England approved it as a standard survey method in 2014. This is the first time any country formally adopted eDNA analysis for identification of a protected species. Detection rates in peer-reviewed trials ran above 99%, compared to 44 to 76% for traditional field survey methods. The same ponds could be surveyed outside the traditional survey window, and by volunteers rather than biologists.¹

River surveys now routinely use eDNA to map fish communities, track invasive species like signal crayfish and topmouth gudgeon, and monitor the spread or recovery of salmon and lamprey populations. In the ocean, water samples from coral reef systems can identify hundreds of fish species in a single analysis. The same survey would take a team of divers months to conduct visually.

Soil eDNA is just catching up. Samples from forest floors and grasslands can reveal the presence of soil fungi, invertebrates, and burrowing mammals. Groundwater eDNA has been used to detect stygobiont species, the cave-adapted invertebrates that live in underground aquifer systems and are among the hardest organisms in the world to survey by any conventional method.

Airborne eDNA is the newest frontier. A 2025 paper in Nature Communications showed that continuous air sampling over a nature reserve could reconstruct changes in vertebrate diversity across several decades from archived filter samples. Pollen filters from meteorological stations, it turns out, are unintentional eDNA archives, a largely untapped record of ecological change.⁴

Limitations and open questions

eDNA is a powerful tool but not an infallible one. DNA moves. In aquatic systems, water currents can carry eDNA downstream from where an animal actually is, producing detections in locations the animal has never visited. This is well-documented and needs to be accounted for in survey design. A detection in a river tells you the species is somewhere upstream, not necessarily at that exact sampling point.

Additionally, environmental conditions affect how quickly DNA degrades. Cold, dark, low-UV environments preserve eDNA longer. Warm, sunlit, acidic environments break it down faster. A negative result does not always mean a species is absent. It may mean the conditions degraded the DNA before sampling, or that sampling intensity was insufficient.

Contamination is a real risk at every stage, from sample collection through to laboratory analysis. Rigorous protocols, field blanks, and laboratory controls are essential. A single stray hair from a researcher can produce a false positive. This is not a reason to distrust the method, but it is a reason to take quality control seriously.

Finally, eDNA tells you that a species was present. It does not currently tell you how many individuals, what condition they were in, or what they were doing. Abundance estimation from eDNA concentration is an active area of research and is improving, but the method remains primarily a presence-absence tool for most applications.⁵

eDNA and citizen science

One of the less-discussed advantages of eDNA is that sample collection does not require specialist training. Collecting a water sample from a pond is something a volunteer can do reliably, which opens up the possibility of large-scale citizen science monitoring at a cost that professional surveys cannot match.

The great crested newt citizen science program in the UK demonstrated this at national scale. Volunteers collected water samples from thousands of ponds across the country, producing a distribution map that no professional survey program could have afforded to generate. The limiting factor was not collection quality but laboratory throughput and reference database completeness.¹

As sequencing costs fall, this combination of citizen-collected samples feeding into laboratory analysis is likely to become a standard approach for environmental biodiversity monitoring. For rewilding projects specifically, it offers a way to track ecosystem recovery across large areas without the cost of continuous professional survey effort.

Frequently asked questions