Scientists have a powerful new tool to help them “see” fish in the Chesapeake Bay’s murky tributaries, and it’s yielding some surprisingly good news about two of the estuary’s most troubled species.

“Imaging sonar” uses sound to help them view, and count, passing fish in dark or cloudy water. For the past few years, scientists with the Smithsonian Environmental Research Center have been deploying one of these underwater sound cameras in some of the Bay’s rivers to monitor spawning runs of alewife and blueback herring, collectively known as river herring.

One of the top catches in the Chesapeake Bay in the 1930s, river herring spawning runs have dwindled in most places, prompting serious concern that they may be disappearing across much of their range, which covers virtually the entire Atlantic Coast. Though their commercial importance has faded, alewives and blueback herring are still considered important forage for larger fish, including striped bass and bluefish. So, their demise could have a ripple effect on other species.

No one knows for sure how many river herring are in the Bay, as fisheries managers lack the staff and resources to do a comprehensive assessment. But a SERC-led team of scientists deployed an imaging sonar device in the Choptank River in 2014 that captured images of the fish as they swam by. Based on the rate at which scientists saw the shadowy blips cross their computer screens, they estimated that as many as 1.3 million river herring swam upriver that spring to spawn.

That’s more than expected, and way more than state biologists had figured were there in the early 1970s, the last time anyone looked intensively at the Choptank’s herring runs.

“It was very good news that we saw as many herring as we did,” said Matthew Ogburn, a Smithsonian biologist leading the team that’s been studying the species’ spawning runs. They published their findings in November in the journal Transactions of the American Fisheries Society.

Indeed, the scientists said their estimates of 581,000 alewives in the river and 726,000 blueback herring suggest that the Choptank’s river herring run is on par with relatively healthy spawning migrations still seen in some major Northeast rivers.

That’s very good news for the Choptank, if verified and sustained. But it doesn’t mean the rest of the Bay’s river herring are out of trouble.

“There are a lot more in the Choptank than there are in some other places,” Ogburn said.

River herring are anadromous, meaning they spend most of their lives in the ocean but migrate to freshwater each spring to spawn. Alewives begin their annual migration in late February or early March, while bluebacks typically show up in early April. They return to the river or stream where they were born, which means the health of each tributary’s run may differ, depending on a variety of factors.

Over the last several decades, though, scientists estimate that coastwide river herring populations have declined by more than 90 percent because of habitat loss, overfishing, pollution and other threats, including dams that block spawning migrations, physical alterations of streams and climate change.

The commercial catch has plummeted from a peak of 140 million pounds in 1969 to a low of 823,000 in 2006. It’s recovered a bit since then, but remains less than 4 million pounds annually, according to the Atlantic States Marine Fisheries Commission. To preserve its remaining stocks, Maryland imposed a commercial harvest moratorium in December 2011, though recreational anglers may still catch and release them. Other East Coast states followed with their own catch bans in 2012, though six still allow some fishing for them.

So depleted are some stocks that the National Oceanic and Atmospheric Administration recently considered listing them as endangered or threatened, but ultimately rejected an environmental group’s petition to do so as “not warranted at this time.” Since then, the federal fisheries agency has been working with states and other partners to restore spawning runs. (A federal court recently ordered the National Marine Fisheries Service to reconsider its decision not to list blueback herrring.)

One of the hurdles is the lack of information about their status in individual rivers. Fish are hard to count — like most wildlife, they scatter at the sight of humans, and they’re underwater, which makes spotting and identifying them that much harder. Biologists traditionally have used nets to intercept, identify and count them. In modern times, they’ve also used electronic fish counters, underwater video cameras and even electrofishing gear, which temporarily stuns fish, prompting them to float to the surface long enough for a quick census.

The DNR biologists who last surveyed the Choptank in the early 1970s installed an electronic fish counter at a weir, or low dam, that fish had to pass over to get upriver. They monitored the species of fish moving upstream by trapping some at the weir, checking the trap at least daily, and more often when run counts were heavy. But a quarter of the time they couldn’t sample because river flows were too high, the Smithsonian team said.

Those methods have other handicaps as well. For electronic counters and video to work, the fish have to swim through narrow passages, which may discourage some from continuing their spawning run. Video is also less effective in turbid streams or at night.

One of the big issues with all types of sampling, Ogburn said, is that if not frequently checked, they can miss a big chunk of the spawning run. Fish tend to move in spurts, prompted by weather conditions and water flows. During a 2013 survey, for instance, he said, the sonar detected 100,000 fish surging upriver in a single four-hour period, and a similar pulse the next night. Other sampling methods might have missed those fluxes.

The Smithsonian team, joined by a scientist with the University of Maryland, placed its dual-frequency identification sonar acoustic camera, or DIDSON, in the Choptank near Greensboro, not far upriver from where the DNR biologists had conducted their surveys in the 1970s. They put orange plastic fencing in the water to herd fish through an area covered by the sonar beams, so none would be missed in the count.

The gadget was programmed to take readings for 10 minutes every hour. Researchers screened all of the fish “shadows” picked up by the sonar to identify the images that were the size of typical of river herring. The scientists periodically checked fish from the river to see what species the sonar was picking up using electrofishing gear, and found that 80 percent of the those in the targeted size range were river herring.

“When we started this project,” Ogburn said, “we really didn’t know what we would see...We are getting a lot more information about river herring, which are a really important species for forage for a lot of different fish species, and birds as well.”

Based on their series of 10-minute sonar samples and the electrofishing checks, the researchers extrapolated the numbers of each species making the spawning run. They estimated there were roughly as many blueback herring as had been tallied decades earlier — 495,000 in spring 2014, compared with 531,000 in 1972 and 271,000 in 1973. But the estimate of alewives was an order of magnitude higher than the old survey — 377,000 in 2014 versus 39,000 in ’72 and 34,000 in ‘73.

Nick Carter, a retired DNR biologist who worked on the 1970s survey, said that he likes the new technology the Smithsonian team used. But after reviewing the journal article and communicating with Ogburn, he said he has doubts about their population estimates.

“I don’t believe there are that many,” Carter said. He acknowledged that the DNR’s 1970s survey may have missed some fish, particularly during heavy spring flows. But he questioned the researchers’ ability to accurately identify species based on sonar images and periodic electrofishing samples. Other fish, notably white perch, make spawning runs about the same time, he noted. And because the new population estimates are based on extrapolations of 10-minute sonar samples, Carter said small errors could be magnified.

Ogburn, though, said his team is “pretty confident” that their sonar scans were able to tell river herring from other fish, particularly white perch, which he said have a different swimming pattern.

Environmental conditions could account for some of the differences. As the journal article noted, historical landings data have varied by an order of magnitude in a few years’ time, suggesting wide swings in spawning runs — or in fishing pressure. More than 2.1 million alewives were reported caught in the Choptank and its tributaries in 1909, followed by a 336,000 fish harvest in 1915, according to federal records.

Maryland’s DNR now only monitors river herring runs in the North East and Nanticoke rivers, not the Choptank. Though the agency does some fish sampling there, it’s not consistent enough to estimate abundance, according to DNR biologist Genine Lipkey. Even so, Lipkey said that she finds the SERC estimates plausible.

“The Choptank is a large river that could host [a] high abundance of river herring,” she said in an email.

The sonar survey also picked up some intriguing shifts in the makeup of the Choptank spawning run, when compared with the results of the earlier survey. There were far more males than females of both species, which runs counter to the assumptions biologists have used in modeling herring stocks that the sexes are more evenly apportioned.

The much greater relative strength of the alewife run also raises questions. In the 1980s, DNR biologists considered it to be “a remnant population that is at a very low level of abundance and may be declining,” the paper noted. The scientists said it’s hard to tell with limited data whether the shift represents a rebound for alewives or a decline for blueback herring. But they noted that other research indicates bluebacks are more prone to bycatch in offshore fisheries than alewives.

Ogburn said that to him, the most significant outcome of the Choptank survey is the refinement of a new method for surveying spawning runs that can be used throughout the Bay to track progress of ramped-up river herring restoration efforts.

The new technology does have limitations — the sonar beam can only see fish as small as herring with precision through about 10 meters of water, Ogburn said. For that reason, it won’t be any help in trying to monitor herring runs in the Bay’s largest tributary, the Susquehanna River, which has an ambitious goal of restoring 5 million fish to now-barren spawning habitat above four dams.

But in smaller rivers and streams, Ogburn said, the technology holds real promise. He and colleagues have been doing sonar surveys for four years, repeating in the Choptank and trying it out in other rivers and creeks, including the Patapsco and Deer Creek, a tributary of the Susquehanna.

With the data they’re compiling, he said, “We actually have a starting point [so] we can move forward to assess how the various conservation actions are helping river herring.”