In a first, US scientists have monitored a spring fish migration through DNA tests on water samples, a study said on Wednesday.

"Environmental DNA" or eDNA, strained from one-litre samples drawn weekly from New York's East and Hudson rivers over six months last year, revealed the presence or absence of several key fish species passing through the water on each test day.

The weekly data snapshots created a moving picture that largely reinforced and correlated with knowledge from migration studies conducted over many years with fishnet trawls.

The Rockefeller University study, published on Wednesday in PLoS ONE, pioneers a way to monitor fish migration that involves a fraction of the effort and cost of trawling, all without harming the fish.

eDNA is now also being assessed as an easy way to estimate the abundance of diverse fish species and other forms of marine life in the dark waters of rivers, lakes, and seas.

The paper, "Aquatic Environmental DNA Detects Seasonal Fish Abundance and Habitat Preference in an Urban Estuary", is authored by Mark Stoeckle and Director Jesse Ausubel of the Program for the Human Environment of The Rockefeller University.

John Bowne high school student Lyubov Soboleva and Rockefeller University scientist Zachary Charlop-Powers co-authored the paper.

As they swim, the fish leave traces of their DNA in the water, sloughed off their slimy, gelatinous outer coating or in excretions, for example.

"Researchers in Europe first demonstrated that relatively small volumes of freshwater and seawater environments have enough invisible bits of DNA floating in them to detect dozens of fish species," Stoeckle said.

"By conducting a series of tests over time, collecting surface water from the same point on both the Hudson and East rivers once a week for six months, wea¿ve successfully demonstrated a novel way to record fish migration."

"Our work also offers clear new insight into the durability of DNA in the water, which persists despite currents and tides with a goldilocks quality just right for research. If the DNA disappeared too quickly, we couldn't obtain an informative sample; if it persisted for too long, there would be too much DNA in the water ' title=' DNA in the water '>DNA in the water to yield useful, timely insights."

In all, Stoeckle and colleagues obtained the DNA of 42 fish species, including most (81 percent) of the species known to be locally abundant or common and relatively few (23 percent) of the uncommon ones.

"We didn't find anything shocking about the fish migration -- the seasonal movements and the species we found are known already," said Stoeckle.

"That's actually good news, adding to evidence that eDNA is a good proxy. It amazes me that we can get the same information from a small cup of water and a large net full of fish."

According to him, some species couldn't yet be distinguished, notably some in the herring family, which have identical sequences in the region of DNA used for testing.

As well, some of the DNA obtained couldn't be identified because the DNA reference library, while steadily growing, is incomplete.

"We knew that we had DNA from a fish, but couldn't pinpoint the species," Stoeckle added.

The work on aquatic DNA was carried out as part of the Marine Science and Policy Initiative of the Program for the Human Environment of The Rockefeller University and the Urban Coast Institute of Monmouth University.