Silent Immunity: How Waterfowl at a California Refuge Are Weathering the H5N1 Panzootic

When clade 2.3.4.4b highly pathogenic avian influenza (HPAI) H5N1 arrived in North America in 2021, it triggered one of the most dramatic shifts in wild bird disease dynamics ever recorded. The virus spread rapidly across the continent, propelled largely by infection and migration in wild waterfowl. As of August 2026, more than 18,500 individual wild birds from over 170 species have been confirmed positive for the virus in the United States alone. Yet the toll has been strikingly uneven: some dabbling ducks carry the virus without visible illness, while sea ducks, diving ducks, and certain geese have suffered mass mortality events. Understanding why some species weather the panzootic while others collapse has become a central question for wildlife disease ecologists.

A new study led by researchers from the U.S. Geological Survey, the University of Georgia’s Southeastern Cooperative Wildlife Disease Study, the U.S. Fish and Wildlife Service, and the California Waterfowl Association offers a detailed portrait of infection and immunity in the waterfowl and waterbirds of northern California’s Sacramento National Wildlife Refuge Complex (SNWRC). Published in Frontiers in Cellular and Infection Microbiology, the research leveraged ongoing capture efforts for movement ecology studies between October 2022 and July 2025 to sample wild birds for both active viral infection and serological evidence of previous exposure. The results reveal a population largely carrying the immunological fingerprints of H5N1 exposure, alongside surprising evidence that some sympatric waterbird species may be far less engaged with the virus than their duck and goose neighbors.

The SNWRC spans roughly 70,000 acres of managed wetlands in the Sacramento Valley and serves as critical habitat for migratory and wintering waterfowl in the Pacific Flyway. Sacramento National Wildlife Refuge alone has historically hosted an estimated 16 percent of California’s wintering ducks, or 10 to 15 percent of the Pacific Flyway’s population. This extraordinary density of birds makes the complex a natural laboratory for avian influenza surveillance—and a place where the consequences of HPAI incursion have been painfully visible. A mortality event from June 2022 through January 2023 killed an estimated 100 Snow Geese, with smaller losses among Mallards, Gadwall, Canada Geese, and even American White Pelicans. A subsequent outbreak from December 2023 through January 2024 killed over 1,100 Snow Geese and 200 Ross’s Geese. Beyond mortality, some infected Snow Geese have failed to migrate or delayed their departure, extending the ecological footprint of the virus.

Between October 2022 and July 2025, the team captured assorted waterbirds—ducks, geese, coots, and stilts—using rocket nets and air boats with dip nets during targeted sampling windows in July, October, and December. Each bird received a federal metal band, and paired oropharyngeal and cloacal swabs were collected using sterile polyester-tipped applicators and placed in cold virus transport medium. When study parameters allowed, blood was drawn from the jugular vein, centrifuged to separate serum, and archived for antibody testing. All birds appeared healthy at the time of processing, with no clinical signs typical of HPAI infection. Laboratory work proceeded on two fronts: viral RNA was extracted from swabs and screened by real-time reverse transcriptase PCR targeting the influenza A virus matrix gene and the clade 2.3.4.4b H5 hemagglutinin, while serum samples were heat-inactivated and tested with a commercial blocking enzyme-linked immunosorbent assay (bELISA) for antibodies to influenza A viruses (IAVs).

The scale of the effort was substantial: 384 swab samples and 204 sera from assorted waterfowl and waterbird species were analyzed. Yet active HPAI H5N1 infection was confirmed in only three individuals—two Snow Geese and one Greater White-fronted Goose—all asymptomatic at capture. The National Veterinary Services Laboratory confirmed all three detections as clade 2.3.4.4b HPAI H5N1. A B3.6 genotype virus was isolated from the Greater White-fronted Goose sample, while the Snow Goose samples failed to meet the cycle threshold at which virus isolation is attempted. The near absence of detectable infection across three years of sampling, in a refuge that has experienced repeated HPAI mortality events, immediately raised the question of what the serological data might explain.

The antibody results were striking. Of 164 waterfowl tested, 107 (65 percent) tested positive for antibodies to IAVs; among 40 additional waterbirds, 12 (30 percent) were positive. Antibodies were detected in all duck and goose species sampled, and the majority of positive samples also carried antibodies specific to both the H5 hemagglutinin and the N1 neuraminidase—the two surface proteins of the circulating H5N1 strain. In ducks sampled during October, the proportion of IAV antibody-positive birds carrying both H5 and N1 antibodies remained high and relatively constant, at 64 percent in 2023 and 80 percent in 2024. A remarkable 88 percent of ducks sampled in July 2025 also carried antibodies to both proteins. This pattern indicates a population-level history of substantial exposure to H5N1 and subsequent survival, consistent with findings from numerous duck species in the Mississippi Flyway, where H5 and N1 antibodies increased dramatically following the 2.3.4.4b incursion.

The study also reports the first known observation of H5 antibodies in the American Coot (Fulica americana), a waterbird species not traditionally associated with influenza A virus circulation. However, both overall IAV seroprevalence and H5-specific antibodies in coots were notably lower than in sympatric waterfowl: IAV antibody prevalence fell from 54 percent in 2023 to 24 percent in 2024, and only 14 to 20 percent of antibody-positive coots carried both H5 and N1 antibodies. Black-necked Stilts showed similarly minimal antibody detection, though sampling was especially limited for this species. The disparity between coots and the ducks and geese sharing the same aquatic habitats cannot be directly explained by the data, but the authors suggest two possibilities: coots may have lower innate susceptibility to infection, or they may exhibit behaviors that limit exposure. A survivorship bias—in which most exposed individuals died—seems unlikely given the low number of American Coot mortalities reported during the outbreak.

Translating antibody prevalence into population immunity is complicated, but experimental studies offer a framework. Sequential challenges with low pathogenic (LP) and highly pathogenic (HP) IAVs have shown that previous exposure can reduce infection, mortality, and viral shedding during subsequent infections with both LP and HP H5 viruses. The relatively high level of subtype-specific immunity observed in ducks—up to 80 percent carrying both H5 and N1 antibodies—may partially explain the near absence of detectable HPAI H5N1 infection during the study years. The authors caution, however, that the serologic techniques used cannot determine whether H5 or N1 antibodies result solely from HP H5N1 exposure; they likely represent a history of exposure to both HP and LP H5 viruses, and LP H5 viruses are known to circulate commonly in these wetlands, potentially maintained locally by non-migratory waterfowl. Cross-reactive antibodies and antigenic imprinting may also mask antigenic differences between LP and HP H5 viruses, preventing definitive estimation of specific exposure history.

The findings also validate the bELISA screening threshold used in the study. A serum-to-negative control absorbance ratio below 0.7 proved an effective guide for further serotyping: while some individuals above the threshold tested positive for H5 and N1 antibodies, this represented only a small percentage of samples. The authors note that test sensitivity may vary over time and among testing formats, and that the timing of infection before sampling cannot be known, but the threshold should not introduce significant error in detecting subtype-specific antibodies. This methodological validation matters for surveillance programs considering serology as a complementary tool to the U.S. national active surveillance strategy, which focuses on viral detection in waterfowl and does not include species such as coots.

Overall, the study underscores the value of serologic surveillance as a supportive strategy for informing HPAI H5N1 risk management in wild birds. The high level of IAV immunity in waterfowl suggests, based on experimental evidence, a reduced potential for widespread or seasonally concentrated mortality in these populations. Results from coots and stilts highlight the species differences that exist in exposure and potential risk at a given site and time—and the possibility that some species remain essentially immunologically naïve and susceptible, even when the virus circulates intensively among sympatric waterfowl. The authors emphasize that additional sampling across subsequent seasons is needed to understand how these immunological baselines shift, but the data provide a critical portrait of the years immediately following the clade 2.3.4.4b incursion in the Pacific Flyway. As HPAI H5N1 continues to evolve and expand its host range across North America, studies like this one—integrating virology, serology, and ecology in a single surveillance framework—will be essential for anticipating where the virus will strike next and which species will bear the burden.

Subject of Research: Highly pathogenic H5N1 avian influenza infection and exposure in waterfowl and waterbirds of the Sacramento National Wildlife Refuge Complex, 2022-2025

Article Title: Highly pathogenic H5N1 avian influenza infection and exposure in waterfowl and waterbirds of the Sacramento National Wildlife Refuge Complex, 2022-2025

Article References: Sullivan, J. D., Poulson, R. L., Casazza, M. L., Matchett, E. L., D’Errico, M., Bundy, A., Chappell, J., Mackell, D., Mott, A., Overton, C. T., Stallknecht, D. E., & Prosser, D. J. (2026). Highly pathogenic H5N1 avian influenza infection and exposure in waterfowl and waterbirds of the Sacramento National Wildlife Refuge Complex, 2022-2025. Frontiers in Cellular and Infection Microbiology, 16, Article 1993811. https://doi.org/10.3389/fcimb.2026.1993811

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1993811

Keywords: Highly, pathogenic, H5N1, avian, influenza, infection, exposure, waterfowl, waterbirds, Sacramento, National, Wildlife