How scientists are tracking salmon with air samples

A team of researchers has shown that you don’t need to dip a jar into a river to learn who’s swimming by. You can just sample the air. 

The team, led by the University of Washington (UW), set out simple filters along Issaquah Creek near the Issaquah Salmon Hatchery during last fall’s salmon run.

The researchers captured trace genetic material shed by Coho salmon leaping and thrashing at the surface. 

Those airborne fragments, known as environmental DNA, or eDNA, rose and fell in step with the hatchery’s daily fish counts – pointing to a new, low impact way to monitor migrations. 

Yin Cheong Aden Ip, the study’s lead author, said he saw the fish jumping and started thinking – could we recover their genetic material from the air?

Filters to catch salmon DNA

Environmental DNA has become a workhorse of conservation. Animals leave behind tiny bits of skin, mucus, and other cells, and scientists detect those traces in water or soil to map who’s there without ever catching or sighting a creature. 

What almost no one has tried is tracking an aquatic species by sampling air next to the water. The UW team designed a test that did exactly that. 

The researchers placed different kinds of passive filters 10-12 feet from Issaquah Creek – three vertical filter types and, for comparison, an open 2-liter tub of ultra-pure water to catch settling particles.

The filters were used on six days between August and October. Each deployment ran for 24 hours.

Back in the lab, the team rinsed the filters, extracted whatever DNA they had gathered, and used a Coho-specific genetic tag with PCR (polymerase chain reaction) to measure how much salmon signal was present. 

The approach doesn’t yield a literal headcount, so they paired the air eDNA data with water eDNA samples and the hatchery’s visual tallies, building a model that treats each method as imperfect but collectively informative about true abundance.

DNA moves from water to air

The result surprised even seasoned eDNA practitioners. Despite being roughly 25,000 times lower in concentration than what they measured in the creek itself, the salmon DNA floating in air still tracked the same daily rise-and-fall pattern as observed fish passing through the river. 

That temporal sync suggests the filters were picking up fresh material shed by fish at the surface rather than random background noise. 

Study senior author Ryan Kelly is a professor of marine and environmental affairs at UW and director of the eDNA Collaborative.

“This work is at the edge of what is possible with eDNA. It pushes the boundaries way further than I thought we could,” said Professor Kelly.

The study also makes a bigger conceptual point: eDNA doesn’t respect the neat categories in our field notebooks.

The team’s data show that DNA can move between water and air, which helps explain why aquatic species sometimes pop up in airborne eDNA surveys.

Why airborne eDNA matters

Traditional salmon monitoring depends on people standing at chokepoints to count fish, a labor-intensive work that’s not always feasible in remote or hazardous terrain.

Autonomous cameras and sonar help, but they require power, maintenance, and calm water. 

By contrast, the method tested here is refreshingly low-tech: a filter on a stand and time. That simplicity could make it attractive for short-term deployments in places without electricity or for augmenting existing counts during peak migration weeks. 

“This is Aden’s baby,” Kelly said. “He arrived saying ‘I know you can get eDNA from the water, but I want to do something nobody has done before.’”

Because eDNA acts like a relative abundance signal rather than a direct census, the team built their model to weave air, water, and visual data together. 

The fit wasn’t perfect, but the airborne traces rose when more salmon were observed and fell when fewer came through. 

“This technique quantitatively links air, water and fish,” said Yin Cheong. “Airborne eDNA doesn’t give us a headcount, but it does tell us where salmon are and what their relative abundance is in different streams.”

Practical limitations and next steps

Collecting DNA out of thin air comes with certain problems. Rain can scrub particles from the air, wind can disperse them, and humidity and temperature may influence how long DNA persists aloft. 

The scientists acknowledge those variables and plan to probe them in follow-up tests, along with dialing in which filter types and placements work best under different conditions. 

For now, the data show that even very sparse airborne signals carry useful information when analyzed alongside other streams of evidence.

The work also hints at broader applications. If salmon can be tracked from the riverbank, perhaps other surface-active aquatic species – from spawning suckers to surfacing amphibians – could be, too.

In places where water sampling is restricted or impractical, air filters might offer a complementary window into aquatic life.

A new tool for an old challenge

Salmon management is ultimately about timing and numbers: when fish arrive, how many pass, whether restoration is working.

Adding a passive, power-free way to “listen” for fish DNA could give biologists another tool, especially in migration bottlenecks across large watersheds. 

“Right now, we’re pushing the boundaries of possibility,” Kelly said. “Eventually, we will develop the technique, as we have for waterborne eDNA, into something that can help guide management and policy.”

The idea that you can stand a few strides from a river and collect the genetic echo of a migration still feels a bit like science fiction. But as this team shows, when salmon churn the surface, they don’t just leave ripples. They leave a molecular trail in the air, and that trail can be read.

The study is published in the journal Scientific Reports.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–