From the Columbia River estuary to the Columbia Plateau area of central Washington, a distance of more than 500 miles, fish-eating birds have been preying for decades on juvenile salmon and steelhead as they migrate to the ocean in the spring and early summer. While some fish species, such as Upper Columbia Steelhead, continue to be severely impacted by this predation, overall the amount of predation has declined over the years, a hopeful sign, researchers say. However, in the Columbia River estuary, a large colony of Double-crested Cormorants, harassed off an island where they nested previously, have taken up residence on the Columbia River bridge between Astoria, Oregon, and Megler, Washington, and continue to prey on juvenile fish in the river.
Last September, Dan Roby, a retired Oregon State University professor, and Allen Evans of RealTime Research, a Bend, Oregon, consulting firm, presented to the Council a summary of 11 years of their research on fish-eating birds in the Columbia River, 2008-19. Roby said they “discovered to our surprise that avian predation is a major source of smolt mortality” on juvenile, Upper Columbia Steelhead, an ESA-listed threatened species, accounting for almost half of the deaths of that species between Rock Island Dam on the middle Columbia River and Bonneville Dam on lower Columbia River, a distance of 307 miles. The predation rate on that species is higher than for other ESA-listed stocks, including Chinook salmon, Roby said.
Juvenile steelhead in general have been shown to be more susceptible to avian predation, especially predation by Caspian terns, than other species of Columbia Basin salmonids. This may be because of their size, roughly six inches or so when they migrate, larger than some other juvenile fish, and their tendency to migrate near the surface of the river, both of which may make them more visible from the air.
In a synopsis of its 2019 annual report on bird predation on juvenile fish in the Columbia Plateau region of Washington, Real Time Research writes that avian predation “… is often the single greatest source of mortality for [Upper Columbia] steelhead during out-migration from Rock Island Dam to Bonneville Dam, with bird predation accounting for more than 50 percent of all mortality sources in 10 of the last 12 years (2008-2019), including in 2019.” According to the report, the estimated Upper Columbia Steelhead smolt losses to birds, primarily Caspian Terns, “were greater than the direct losses associated with dam passage, predation from piscivorous fish, mortality from disease, and all other remaining mortality factors combined.”
Juvenile fish migrating down the Columbia and lower Snake rivers to the ocean every spring and early summer pass through the foraging areas of as many as 14 colonies of fish-eating birds, including gulls, Caspian Terns, and Double-crested Cormorants. The annual migration happens to coincide with the annual hatching of baby birds, and so adult birds are not only feeding themselves from the river, they also are taking fish for their chicks.
Fish that survive downriver past Bonneville Dam face even more predation, according to the report. This “substantial” predation, as the report describes it, primarily by terns and cormorants, amounts to “upwards of 28 percent of available steelhead smolts in the estuary,” according to the report.
But, there is a bit of good news, at least for smolt survival. In an email, Roby wrote that the breeding populations of Caspian Terns and Double-crested Cormorants in the Columbia Plateau region and the Columbia River estuary are both in decline, not increasing. In the Columbia Plateau region, the number of breeding Caspian Terns in 2019 was about 445 pairs, a 49-percent decrease from the average in the same region prior to implementation of the Inland Avian Predation Management Plan. The plan was developed by the U.S. Army Corps of Engineers and the U.S. Bureau of Reclamation in response to direction in the 2008 Biological Opinion on Operations of the Federal Columbia River Power System. Among other actions, the Opinion directed federal agencies to develop the Inland Avian Predation Management Plan, monitor avian predators in the mid-Columbia River area, and evaluate impacts on ESA-listed juvenile salmon and steelhead.
The numbers of breeding pairs of Double-crested Cormorants in the Columbia Plateau region also has been in decline, he said. The largest colony of cormorants in the Columbia Plateau, numbering over 1,000 breeding pairs, had been at the north end of Potholes Reservoir near Moses Lake, Roby wrote.
“That colony has nearly disappeared, and no other colonies in the Plateau region have grown to a commensurate extent,” he wrote. “So even though the management plan did not target Double-crested Cormorants, the regional population has declined to a similar extent as the regional population of Caspian Terns.”
In the Columbia River estuary, the breeding population of Caspian Terns was about 3,860 pairs in 2019, all nesting at the colony on East Sand Island, a sand island near the mouth of the river built by the U.S. Army Corps of Engineers from material dredged from the river’s navigation channel. That is a 64-percent decrease from the number that nested on East Sand Island in 2008, the first year of implementation of the Caspian Tern Management Plan for the Columbia River Estuary, and within the target size for the tern colony stipulated by the management plan (3,125 - 4,375 pairs). In addition to the declines in the Columbia Plateau and the estuary, the number of Caspian Terns is declining throughout the Pacific Flyway, the bird migration corridor that stretches from Alaska to southern Chile.
Why the decline?
While ‘uncertainty’ is the keyword in attempting to explain a decline over such a broad area, several factors are evident.
“Based on some of the Flyway-wide survey results for these two species, compiled by Migratory Birds and Habitat Programs of the U.S. Fish and Wildlife Service, plus our monitoring of the Caspian tern colony on East Sand Island,” Roby wrote, “we can draw some strong inferences.”
First, he said, more than 65 percent of the Pacific Flyway population of Caspian Terns formerly nested on East Sand Island, in the decade of the 2000s. During that time, tern nesting success at the East Sand colony was comparatively high. But in the most recent decade, 2010-2019, the numbers declined as the size of the available nesting habitat on East Sand was steadily decreased by wildlife managers, consistent with direction in the management plan for terns in the estuary. At the same time, the nesting success of those birds that used the reduced habitat declined, significantly, particularly in 2011 and 2017. Terns failed to produce any chicks on East Sand in those years.
Second, there was a drought in the Southern Oregon/Northeast California area, where the U.S. Army Corps of Engineers built eight islands as alternative nesting habitat for terns that formerly nested at East Sand. That’s a long way from the Columbia River, but for terns it’s not unusual to fly hundreds of miles to find new nesting sites. However, the drought made the islands accessible from land, and thus for land-based predators, and also reduced the amount of forage fish for terns in the remaining waters, and so many of the birds nested elsewhere. One colony, described by Roby as"moderate sized" formed on a warehouse rooftop in South Seattle.
And third, beginning in 2014, the success of the inland avian predation management plan reduced the size of the tern population in the Columbia Plateau. As a result, the tern population lost nesting habitat in all three regions: in the Columbia River estuary, in the Southern Oregon/Northern California area, and in the Columbia Plateau.
“No major new colonies of Caspian Terns formed in the Pacific Flyway during this period, except for the one on the Seattle warehouse roof,” Roby said.
Additionally, those terns that moved back to East Sand from the desiccated islands to the south did not do well. There were not enough chicks to replace birds that died.
Again, it’s hard to say definitively, but several factors apparently contributed, he said. Bald eagles chased terns away from their nests; gulls swooped in and ate tern eggs and killed chicks; and the availability of forage fish – anchovies, surfperch, eulachon (smelt), and herring – declined as the result of high river volumes and poor ocean conditions for the fish.
A long history of predation by terns in the estuary
The problem of birds eating juvenile salmon and steelhead in the Columbia River estuary first became apparent in the mid-1990s, when fisheries biologists noticed that the population of birds nesting on Rice Island, a huge sand pile created from material dredged from the shipping channel of the Columbia River by the U.S. Army Corps of Engineers, was increasing and apparently beginning to impact migrating fish.
Rice Island, upstream from Astoria, is owned by Oregon. The island originally was created in 1962, and Caspian terns began colonizing it in 1986. The colony grew rapidly from about 1,000 breeding pairs to 8,700 by 1998. As the population grew, biologists began to worry that the birds were consuming ESA-listed species. The first ESA listings of salmon and steelhead in the Columbia River Basin occurred in the early 1990s.
It was estimated that Caspian terns nesting on Rice Island consumed 5.4 - 14.2 million juvenile salmonids – salmon and steelhead – in both 1997 and 1998. This represented 5 – 15 percent of all salmonid smolts reaching the estuary in those two years, bird researchers reported. In 1999 regional fish and wildlife managers declared that something needed to be done to manage the birds and reduce the losses.
That year, the effort began to relocate the Caspian tern colony from Rice Island, where the colony had grown to become the largest in the world, to a site on East Sand Island, about 10 miles west, and therefore much closer to the ocean. Nearer the ocean, there are larger numbers of ocean forage-fish species for birds to prey on.
Moving the colony meant reducing the size of the nesting area by erecting snow fences to surround a designated area, placing stick-mounted streamers that waved in the wind outside the nesting area, and chasing the birds away from no-nesting areas using all-terrain vehicles, sometimes blasting loud music. This kind of harassment was necessary because the term “nest” is a bit of a misnomer for terns, as a tern nest is more a spot on an open beach than the familiar nest of sticks and feathers. Terns lay their eggs on open sand in a depression they scrape with a foot. They are spooked by distractions, including visual ones, noise, and motion – hence the streamers, snow fencing, ATVs, and music. The harassment worked. By 2000, more than 94 percent of the terns nesting on Rice Island had relocated to East Sand Island.
From 2001 to 2014, all Caspian terns nesting in the Columbia River estuary used East Sand Island, except for three nesting pairs that laid a total of four eggs on Rice Island in 2011, biologists reported. From 2015 to 2018, terns attempted to re-establish a breeding colony on Rice Island, but harassment successfully drove them away.
In all, this had a positive result for fish. From 2001 to 2015, the estimated consumption of juvenile salmonids by Caspian terns nesting on East Sand Island averaged 5.1 million smolts per year, which is an almost 59-percent reduction in annual consumption compared to when the Caspian tern colony was on Rice Island, where an estimated 12.4 million smolts were consumed in 1998 alone.
The management plan for terns in the estuary was developed by the U.S. Fish and Wildlife Service, adopted in 2008 and updated in 2015. The plan calls for encouraging the “redistribution” of the East Sand Island tern colony to alternative nesting sites outside the Columbia River Basin, in southeastern Oregon and near San Francisco Bay. The alternative nesting sites were constructed, though not well-used by the birds. Concurrently, tern nesting habitat on East Sand Island was reduced from five acres in 2008 to one acre for the 2015-2019 breeding seasons.
In 2019, terns arrived at the East Sand Island colony and began nesting in early to mid-April, the usual time of year. But gulls ate a large number of tern eggs initially, and this delayed the eventual timing of hatching and chick rearing. The tern colony reached its peak in mid-June – 3,860 breeding pairs in the one-acre nesting colony area, which was less than in 2018 (4,959 pairs) and was within the range of desired colony size in the management plan. The colony had dispersed by September, as usual.
During the 2019 nesting season, observers with binoculars recorded terns capturing fish 4,050 times. The observations showed that marine forage fish made up a larger percentage of the total than salmonids. That has been the case each nesting season since 2000.
Cormorants in the estuary
Over time, East Sand Island also has been quite popular with Double-crested Cormorants. The cormorant colony on East Sand grew from about 100 nesting pairs in 1989 to a peak of 14,916 pairs in 2013, according information on the Corps of Engineers’ cormorant website.
This colony alone accounted for more than 40 percent of the cormorant population in the West and was the largest breeding colony in North America. Eventually, concern over the impact the birds were having on juvenile salmon and steelhead, particularly ESA-listed species, led to the development of the management plan for cormorants in the estuary, adopted by the Corps of Engineers in 2015.
Like the plan for Caspian Terns in the Columbia Plateau, the plan for cormorants in the estuary responded to direction in NOAA Fisheries’ 2008 Biological Opinion on operations of the Federal Columbia River Power System to protect ESA-listed juvenile salmon and steelhead. In the Bi-Op, NOAA recommended a number of “reasonable and prudent alternatives” (RPAs) to protect the fish and their habitat, including in the estuary. Two of the RPAs focused on reducing the impacts of Double-crested Cormorants. RPA 46 required reducing the cormorant population in the estuary, particularly the East Sand colony, which accounted for 98 percent of the breeding population there. RPA 67 required monitoring and evaluation of the management plan and the cormorant population.
In response to the BiOp, the Corps created an interagency working group that developed conceptual alternatives for reducing the size of the East Sand colony. From their work, a specific management objective of no more than 5,380 to 5,939 breeding pairs was identified as ideal and included in a supplemental BiOp issued in 2014. The Corps adopted this goal and included it in the 2015 management plan.
Before the management plan, however, studies of non-lethal ways to discourage nesting – the use of barrier fences, for example – had been conducted. These helped inform decisions once the management plan was adopted, and it was quickly decided that, similar to Caspian Terns on Rice Island a decade earlier, a useful action would be to discourage the birds from nesting on East Sand. This involved chasing birds away from potential nesting sites, in this case with bright lights and “human hazers,” and also reducing the amount of nesting habitat on the island. Barrier fences were installed around the designated nesting habitat, and each year the fences were moved to make the area smaller. By 2013 the nesting area was reduced to 4.4 acres, and by 2019 it was down to just over one acre. Some birds were caught and fitted with radio tags, and their movements followed to determine where they were nesting instead of on the island.
In an email, Jake Macdonald, a physical scientist with the Portland District of the Corps, said hazing experiments in the early 2000s on terns and late 2000s on cormorants showed that both species could be hazed from areas where they aren’t wanted, but that cormorants were more likely to relocate within the estuary when hazed off East Sand Island, while terns were more likely to relocate to another basin. He said that appears to be because there is more habitat for cormorants in the estuary than for terns, and the long-flying terns are more likely to just fly somewhere else to nest.
That created a dilemma: how to effectively reduce the number of cormorants in the estuary if they were likely to remain there?
The Corps, in consultation with state and federal fish and wildlife agencies, chose a two-phased management approach. The first phase included reducing the size of the colony to about 5,600 pairs between 2015 and 2018 - about the mid-point of the management plan goal – with methods that included shooting birds and spreading oil on their eggs to prevent them from hatching.
Macdonald said adding lethal removal to the non-lethal hazing was seen as the best way to successfully reduce the cormorant population in the estuary. According to the Record of Decision <https://protect-us.mimecast.com/s/KvzOCkRMm6IYEKqf2Ha4I> on the final environmental impact statement (FEIS to reduce predation on juvenile salmon and steelhead in the estuary), this option, a combination of passive harassment and lethal removal "...best balances the competing needs of the biological resources considered in the FEIS and represents the widest range of benefits to ESA-listed juvenile salmonids while reducing risk to the sustainability of the western population of [cormorants] in the long-term." The FEIS also prescribed extensive monitoring of cormorants throughout the estuary and the Western Flyway, and set thresholds to halt the hazing and killing at East Sand if they were seen to be reducing the size of the regional cormorant population.
In the second phase, much of the western side of the island was bulldozed, allowing it to flood and thus reducing the available nesting habitat. This, in combination with continued hazing, was intended to ensure the colony would meet the management plan guideline.
Shooting cormorants and oiling their eggs began in 2015 and continued through 2017. In all, 5,576 birds were killed and 6,181 nests were destroyed by oiling eggs. Once the shooting began, the birds would disperse. In 2016 the entire colony dispersed, and in In 2017 the cormorant colony did not arrive in its usual numbers or at the usual time, in the spring. In fact, the birds didn’t arrive on the island until June, late in the usual breeding season, and only about 1,100 returned. Many birds nested elsewhere in the area. This mass desertion of the island might have resulted from the shooting and egg-oiling, or it might also have been caused by eagles attacking the nests.
Environmental groups were outraged by three years of killing and nest destruction, and also by a proposed federal rule change in 2020 that would allow an increase in the number of cormorants that could be killed across the nation. Fish farming businesses in the Midwest and South asked for the rule change, and initially the proposal applied only to states east of the Continental Divide. But later, it was changed to apply nationwide. The National Audubon Society, in a February 2020 news release, said cormorants were scapegoats for the real problems for fish and fisheries across the nation, including dams, water pollution, and aquatic invasive species.
Meanwhile, cormorants began nesting elsewhere than East Sand Island, most notably on the girders and concrete pilings of two U.S. Highway 101 bridges in Astoria, Oregon – one that connects Astoria to Megler, Washington over the Columbia, and the other from Astoria to Warrenton over Youngs Bay. The number of cormorants on the Astoria/Megler Bridge, which were just a few as long ago as 2004, jumped to 1,700 pairs in 2018 and 3,542 pairs in 2019. The Oregon Department of Transportation is monitoring the bridge birds, and the damage their acidic excrement is doing to the structure's paint. On the Astoria/Megler bridge the excrement has accumulated to such a depth that it is now difficult for the state to adequately inspect and evaluate the bridge, an important maintenance task for a steel structure so close to salt water.
“In early May, we completed the bi-annual, federally required inspection of the bridge section where most of the cormorants are living and nesting,” said ODOT spokesman Lou Torres. “While we were able to do the necessary inspection, it was difficult with the all birds, nests and guano.”
Torres said ODOT hazed cormorants from a boat on the Washington side of the bridge during the three weeks preceding inspection, under the guidance of a federal wildlife biologist with the U.S. Department of Agriculture. The hazing used a paintball marker, which provided a visual and auditory deterrent, and a green laser during low-light conditions. This helped minimize nesting on that side of the bridge, although eggs were still discovered during inspection, Torres said. ODOT also installed two speakers on the Oregon side that blasted predator noises.
“We could not tell how effective the speakers were, although recent surveys show an increase in the number of birds on the Oregon side of the bridge,” Torres said. “The problem is growing worse for ODOT.”
He said ODOT is working with the Oregon Department of Fish and Wildlife to develop a plan to deter cormorants. The plan needs to be in place well before that side of the bridge is repainted beginning in 2022, he said.
As cormorants began nesting on the bridges, they also began nesting farther upriver in the estuary and lower Columbia area. That worries state salmon and steelhead managers and the Corps of Engineers for the same reason they worried about terns nesting on Rice Island in the late 1990s – the concentration of salmonids among all small fish in the river is greater farther away from the ocean. Thus, the effort to disperse and manage the cormorant population and reduce its consumption of juvenile salmon and steelhead may have had precisely the opposite effect of what was intended.
“An important thing to consider is that the available data suggest cormorants consume far more salmonids per individual bird at colony sites upriver of East Sand Island,” wrote James Lawonn, a wildlife biologist for the Oregon Department of Fish and Wildlife (ODFW), in an email. Lawonn monitors cormorants in the estuary. “So, although there are fewer birds in the estuary overall, the predation rate on salmonids can be the same, or even worse.”
That may be intuitive, but in fact there is little data to back up such an assertion, he said.
“Unfortunately, the federal agencies have not funded sufficient PIT-tag monitoring at colony sites besides East Sand Island, and so there is no way to determine whether cormorant management on East Sand actually reduced predation on an estuary-wide scale, which was the ultimate goal of the management plan,” Lawonn wrote. He is working on a paper that will explore the available data to try to determine whether predation changed in response to the management actions. That paper has been delayed and may be done by late this summer or early in the fall, he said, adding it’s no secret ODFW would like to see cormorants leave the bridges and nest again on East Sand.
Meanwhile, the Corps of Engineers considers its implementation of the cormorant management plan to be nearly done, Macdonald said.
“Overall Corps funding of avian research and monitoring is decreasing as we wrap up implementation of the two management plans,” Macdonald wrote in an email. “The Caspian Tern plan has been fully implemented and the Caspian Tern colony on East Sand is being operated and maintained as a Corps project with minimal funding for annual vegetation removal and colony size monitoring.” He said the Corps will continue to monitor the size of bird colonies annually.
Reason for optimism
While cormorants have taken to the bridges in Astoria and may be consuming as many salmon and steelhead smolts as ever, the fact that predation by terns in the estuary appears to be waning is a bit of good news.
“I am generally positive about the reduction of avian predation rates in the Columbia Plateau region and the Columbia River estuary,” Roby wrote in an email. “The [estuary and Columbia Plateau] management plans have both achieved significant reductions in Caspian tern predation rates on juvenile salmonids from the Columbia Basin.”
But he believes critical uncertainties remain, regarding the continuing impacts on smolt survival from the Caspian Tern colony in the Blalock Islands in Lake Umatilla, the reservoir behind John Day Dam, and from gull colonies in The Dalles, John Day, and McNary reservoirs. Caspian terns in that area, harassed off their usual breeding island, simply moved downriver to the Blalock Islands, which are within the Umatilla National Wildlife Refuge. There, they cannot be harassed because the islands are protected habitat.
Additionally, there is some uncertainty about whether efforts to maintain the Caspian Tern breeding colony on East Sand Island and prevent terns from nesting elsewhere in the Columbia River estuary will be successful in the long term. With the Corps of Engineers scaling back its work, reduced funding may prove to be an important matter, as the ongoing research and monitoring of avian predation has a significant annual budget. In Fiscal Year 2020, the Bonneville Power Administration awarded $560,341 to Oregon State University and Real Time Research for its monitoring and evaluation work. Grant County Public Utility District and the Priest Rapids Coordinating Committee (PRCC), which oversees anadromous fish activities associated with Grant PUD’s Priest Rapids Dam, contributed $382,987 for monitoring in 2020 in the Columbia Plateau region. In all, Grant and the PRCC have awarded about $5.8 million for this research over the last seven years, said Curtis Dotson, Grant’s fisheries program manager. In 2019, the Corps of Engineers spent about $150,000 recovering and analyzing PIT tags on East Sand Island. That’s less than in previous years because there were significantly smaller tern and cormorant colonies on East Sand Island to recover tags from and to analyze data with compared to previous years. The Bureau of Reclamation is also involved. Michael Coffey, Bureau regional public affairs officer in Boise, said the Bureau’s work on avian predation in the Columbia Plateau includes constructing a network of ropes and flags on Goose Island and other islands nearby in the Potholes Reservoir to dissuade terns from nesting; hazing terns and, if necessary, removing eggs; and monitoring habitat use, colony numbers, and bird behavior.
“My greatest concern is over the critical uncertainties with regard to the management plan for Double-crested Cormorants in the estuary,” Roby wrote. “Implementation of this management plan seems to have gone seriously awry, and the largest breeding colony of Double-crested Cormorants anywhere in the Pacific Flyway is now on the Astoria-Megler Bridge, instead of East Sand Island.”
He said it now appears that the cormorant colony on East Sand has been abandoned and a new colony is forming on the bridge and could reach at least 5,600 pairs – the average size of the colony envisioned in the management plan for East Sand. The numbers are right, but the location is wrong, and if other cormorants are nesting farther upriver in the estuary, where there are more salmon and steelhead in the mix of fish in the river, then predation could be worse than before.
What should be done?
Roby believes the bridge birds need to be persuaded to return to nest on East Sand. If that doesn’t happen, more salmon and steelhead will die, and the messy bridge will need repeated and, probably, expensive maintenance.
The Council also has weighed in on the avian predation issue. In a March 2020 letter to the U.S. Fish and Wildlife Service, the Council wrote, “the ecology of the river system is clearly out of balance.” What is needed, the Council wrote, is a better alignment of federal agency missions and responsibilities, better interagency collaboration, and a basin-wide strategy to reduce predation by all species of fish-eating birds. These birds present an “existential threat to Columbia River Basin salmon and steelhead.”
Guy Norman, a Washington member of the Council and Chair of the Council’s Fish and Wildlife Committee, said the Council’s Columbia River Basin Fish and Wildlife Program supports on-going efforts to reduce bird, fish, and marine mammal predation on salmon and steelhead, and to increase or revise those efforts as necessary.
“The region recognizes that avian predation continues to have a significant impact on ESA-listed juvenile salmon and steelhead, and there is a need to further address their impacts in the estuary and in the interior Columbia Basin,” Norman said.