Highly Pathogenic Avian Influenza (HPAI H5N1) viruses of clade 2.3.4.4b have driven substantial global impacts in recent years, affecting wild birds, poultry and mammals. Wild birds are central to the panzootic by mediating both large-scale and regional viral dissemination. The authors used France as a case study—citing strong epizootic impacts and high genomic sequencing coverage—to investigate how different H5N1 genotypes circulated in Europe during 2021–2023 and to identify ecological and environmental factors associated with spread.
The study applied continuous phylogeographic analyses to reconstruct the spatiotemporal spread of multiple viral lineages. These reconstructions were then combined with environmental and ecological variables to evaluate associations between virus movements and host, habitat, and bird-group characteristics. The approach aimed to compare patterns across the main H5N1 genotypes observed in Europe during 2021–2023.
Genotypes exhibited distinct spatial and host dynamics in France. The genotype labeled EA-2021-AB displayed widespread dissemination and involved multiple host species across the country. EA-2022-BB was primarily associated with Laridae (gulls), showing a stronger link to this bird group. The secondary wave of EA-2020-C circulated mainly among northern gannets and carried a pronounced coastal signature. These genotype-specific differences indicate that viral dissemination patterns were not uniform across strains and that particular genotypes were linked to particular ecological contexts.
Across genotypes and lineages, ecological associations were heterogeneous. The analyses identified five species-level variables and three bird-group variables as key ecological drivers, underscoring the role of species ecology and group-level behaviors in HPAI circulation. Importantly, no consistent single host emerged as a universal driver across all lineages.
Many identified associations involved species that had not been reported as infected by the corresponding viral lineage in surveillance data. The authors interpret this in two non-exclusive ways: these associations may reflect shared habitat use and ecological interfaces that facilitate indirect transmission, or they may point to undetected infections in some species. The latter possibility supports the need for targeted active surveillance in taxa and locations suggested by ecological associations but not yet documented as infected.
The heterogeneous pattern of associations highlights that a combination of host, environmental and virological factors shapes H5N1 dissemination in wild birds, rather than any single dominant factor.
Using the ecological associations derived from phylogeographic and environmental analyses, the study produced ecological risk maps. These maps identified additional areas of elevated risk that were not included within France’s current HPAI risk zones, while also accurately capturing recent outbreak dynamics. The spatial mismatch between emergent high-risk areas and existing zones supports the authors’ argument that risk zoning should be updated to reflect evolving epidemiological patterns and the expanding range of affected hosts.
The results show that H5N1 spread in wild birds in France during 2021–2023 was highly heterogeneous across genotypes and shaped by multiple interacting factors. Because genotype-specific host associations and environmental signatures differed, the authors recommend that risk zones and surveillance strategies be reviewed and updated frequently. They stress the importance of considering shared ecological interfaces—and not solely documented infections—when designing surveillance and control measures. Targeted active surveillance in species and locations indicated by ecological associations may reveal undetected infections and improve early detection.
The source reports that many ecological associations involved species not reported as infected by the corresponding lineage, but it does not specify which surveillance gaps, sampling biases, or detection limits might account for these discrepancies. The preprint does not report additional methodological limitations or detailed numerical results in the provided abstract text. Therefore, specifics on sample sizes, statistical effect sizes, model performance metrics and exact species- or variable-level results were not reported in the source abstract and would require consultation of the full preprint for complete details.