Following the Trail of eDNA

Environmental DNA, or eDNA, is helping scientists locate elusive wildlife—from rare and endangered species to emerging invasives

• By Peter Andrey Smith
• Conservation
• Sep 26, 2024

Illustration by Ibrahim Rayintakath

A COLLEAGUE IN JULIE LOCKWOOD’S LAB FIRST ENVISIONED IT: a supersized swab. Five years ago, Lockwood, a Rutgers University ecologist who often surveys habitat for elusive wildlife, was brainstorming with her team about a new search tool that would borrow from forensic crime scene investigations—but on a much larger scale. “We were honestly just sitting around at lunch and trying to come up with ways of sampling more terrestrial spaces. And someone was like, ‘Oh, we just need, like, a giant swab for CSI.’” Unfortunately, Q-tips with softball-sized heads do not really exist. “And then I can’t remember who it was, but they’re like, ‘Well, what about a paint roller?’” she recalls. “So we tried it, and it worked.”

Since then, Lockwood and her colleagues have honed a method called tree rolling. Using a standard-issue paint roller, slightly dampened with water, the researchers attach a long handle and roll its cylindrical nap along the bark and foliage of trees to swab for evidence of wildlife. The tool functions a bit like a lint roller, picking up the remnants of DNA that animals leave behind in the environment. (They’ve also used the tool to collect DNA from the surfaces of soil and wooden boards.)

Back at the lab, technicians rinse these DNA fragments off the rollers and use computers to analyze the results. In one study that surveyed 21 trees in two New Jersey oak-hickory forests, Lockwood says she and her team were able to estimate about 90 percent of the mammal species living in an area of more than 400 acres—all in a single day.

Long-term monitoring of forest-dwelling mammals is vital to their conservation, but traditional habitat assessment surveys can be both expensive and labor intensive. Even a single snapshot in time can require multiple methods: Motion-activated cameras may record bears, cougars and other large mammals; checking nighttime transects could reveal flying squirrels and other nocturnal animals dwelling in the canopy; and researchers deploy acoustic devices, such as LIDAR, to detect bats.

Yet in her New Jersey study, Lockwood gathered evidence of all these organisms at once by swabbing for traces of DNA. “One method, one way, done,” she says. Her results, published in 2023 in Scientific Reports, found evidence of 16 mammal species, including the southern flying squirrel, white-tailed deer and seven other species commonly seen in the region, as well as the endangered and rarely observed eastern small-footed bat.

Groundbreaking and fast-growing field

Every living thing, it turns out, leaves a detectable trace. In recent years, researchers around the globe increasingly have been searching for and finding this so-called environmental DNA, or eDNA, in virtually any imaginable location. Indeed, within the last two decades, the fast-growing field has firmly established itself, with both a dedicated journal and a society that hosts annual meetings. Outside academia, state and federal agencies also have started using eDNA to monitor wildlife.

Beyond the technique’s convenience and relatively low cost, taking a sample—whether a tube full of water, soil, snow or air—requires little or no expertise. As Lockwood’s research suggests, analyses of such samples can pick up signs of wildlife covering a broad taxonomic range. And critical to scientists trying to find evidence of rare or endangered species, the noninvasive method does not involve physically capturing an organism in a trap, net or vacuum-powered bug collector, meaning it minimizes potential harm to vulnerable wildlife.

Like many in the field, Lockwood learned about eDNA when she heard of its use in aquatic environments. The first study that came to her attention, in 2008, was conducted by a French team that, by examining a shot glass’s worth of pond water, detected the presence of American bullfrogs, which are invasive across southern Europe. Environmental DNA did something that physically tromping through the muck and mud did not: It allowed researchers to find elusive individual frogs when the population remained small and potentially easier to control.

Today, aquatic environments remain the habitats most often surveyed for eDNA, largely because collecting samples is so straightforward. By contrast, eDNA’s use in terrestrial habitats has been limited by the lack of simple collection methods, hence Lockwood’s efforts to come up with something new. Once she and her colleagues developed their tree-rolling methodology, Lockwood—who’d long thought of herself as a dyed-in-the-wool, muddy-boots ecologist—quickly became immersed in the field of environmental DNA. “I don’t think it’s overstating it to say it’s sort of a revolution, in terms of how we think about how to study wildlife, especially rare wildlife or invasive species,” she says. “It solved so many of the problems that I spent most of my career trying to overcome.”

The field owes a small debt to an unlikely source: the fictional detective Sherlock Holmes, who in the early 20th century inspired a real-life French criminologist named Edmond Locard. The Holmes stories described tracking down criminal suspects by analyzing specks of mud from their shoes, and Locard proposed a more general principle: When two things come in contact, he said, they inevitably leave a trace of themselves behind. He distilled the idea into an indelible phrase: “Every contact leaves a trace.”

In more recent decades, wildlife scientists have combined this concept with the genetic-sequencing tools developed by molecular biologists. Every organism, in the sum of all its genes, carries what amounts to a metaphoric ID card: its genome, also called its DNA barcode. When animals come in contact with their environment, they inevitably leave traces of their genomes, dropping their ID cards in the form of urine, scat, skin cells and other cellular debris. Using rapid-gene-sequencing technology, researchers can scan for the DNA left behind by wildlife in virtually any environment—from oceans and lakes to sand, soil and even dust particles carried aloft in the global jet stream.

One of eDNA’s greatest assets is that this indirect means of detection can be more accurate than direct observation. Some herpetologists, for example, put plywood boards down on the forest floor to attract reptiles drawn to the moist hiding places underneath that may also harbor invertebrate prey. The method allows scientists to carry out visual surveys.

In a study conducted in the New Jersey Pine Barrens, Lockwood and her colleagues paired this traditional technique with eDNA collected underneath the so-called coverboards. While her team rarely laid eyes on Scincella lateralis—a cryptic little skink of “possible conservation concern”—they readily found traces of its DNA. In a paper published in 2022 in Conversation Biology, they reported detecting skink DNA as many as 16 times more often than they saw the animal firsthand by physically lifting the boards.

Exposing invisible invaders

Scanning for eDNA also is helping scientists detect signs of invasive species more accurately—and hopefully sooner—than searching by sight. Demonstrated in aquatic environments by the French researchers who found bullfrogs more than a decade ago, the method has been adopted by biologists doing similar work in terrestrial habitats. Lockwood is among them, looking specifically for evidence of the invasive spotted lanternfly.

Native to Asia, the lanternfly, a planthopper, was first reported on this continent in Pennsylvania in 2014. Since then, it has spread to at least 17 states, and its range continues to expand. A voracious eater—which dines on everything from maples, oaks and some 70 additional tree species to peaches, grapes and other crops—the insect can cause wilting, defoliation, low yields and even plant death. Already, it is costing farmers tens of millions of dollars a year, primarily by killing grapevines.

Given the importance of detecting lanternflies before their populations grow too large to control, the New York State Department of Transportation has hired Lockwood’s team to roll trees near highway rest stops as a “first pass” to winnow down the scope of its search—from the 55,000 square miles that make up the entire state to its most critical areas. (In a previous study, Lockwood’s lab showed eDNA was more than twice as likely as visual surveys to detect lanternflies.) The state targets highways because the insects are believed to be hitchhikers, traveling by way of roads and jumping off vehicles when they stop. The idea, Lockwood says, is to avoid having to search everywhere, but “look here, bring in all your conventional sort of trapping tools and look right here, because we think we’re getting an early detection signal.”

Beyond locating specific target species, whether wanted or unwanted, eDNA is allowing researchers to characterize many different organisms in a given sample—as in Lockwood’s forest mammal work. Known as metabarcoding, the technique’s scope is potentially unlimited. A single drop of seawater, for instance, can contain DNA shed by everything from bacteria to whales. Earlier this year, an Australian team used the method to scan eDNA plucked out of more than 40 spider webs. In one analysis, the researchers detected DNA from 13 native species of birds, the motorbike frog and the snake-eyed skink, suggesting the naturally occurring webs could double as biodiversity monitors.

In New Zealand, scientists from government, industry, communities and beyond are scaling up that approach to build a biodiversity surveillance tool for the nation’s rivers. Championed by Michael Bunce, former chief science advisor at New Zealand’s Department of Conservation, the project has created a database of more than 50,000 aquatic eDNA samples profiling everything from microbes to mammals. The results can be displayed on interactive aerial maps so scientists and others can explore and share the data.

Bunce, who has been involved with eDNA research practically since its inception, says the tool can help set conservation priorities. In a world with so many environmental challenges, he believes the most pressing question comes down to allocating resources: “Where do you go and why?” With resources limited, more boots on the ground are not enough, says Bunce, who sees eDNA as a way of pointing us in the right direction, putting the right foot forward.

Environmental DNA does have its limits, of course. It provides, at best, suggestive evidence, and like the direct observation of wildlife, its results cannot authoritatively exclude the presence or absence of a species. The method also makes it difficult to assess health or measure population abundance—critical factors in managing and protecting wildlife. The field faces broader concerns as well, including the impact it has on researchers. As one marine biologist wrote in an editorial published in Nature, “The more time we spend analysing at the bench, not the beach, the less connected we are to the ecosystems we are trying to protect.”

Still, to most scientists who’ve adopted eDNA methods, the advantages outweigh the shortcomings. Those researchers include Lockwood. Early in her career, she recalls, a project to monitor the endangered Cape Sable seaside sparrow meant she had to helicopter into the Everglades and camp in remote areas for a week at a time—all with limited success. The bird is a ground-dwelling species that camouflages itself in inaccessible, muddy grasslands. “So, it’s like finding a needle in a haystack,” she says. But if eDNA technology had been available back then, Lockwood adds, it would have been more “like taking a metal detector into the haystack with you.”

---

Peter Andrey Smith is a science reporter based in Maine.

---