Emerging Microbes & InfectionsPathologyPeer-reviewed journal article
Investigating high pathogenicity avian influenza virus incursions to remote islands: Detection of H5N1 on Gough Island in the South Atlantic Ocean
Antje SteinfurthiD, Joshua G. Lynton‐JenkinsiD, Jaimie B. CleelandiD, Benjamin C. MollettiD, Holly A. CoombesiD, Andrea Moores, Robyn Neal, Ben Clifton, Marco Falchieri, Christopher W. JonesiD, Michelle M. RisiiD, Susannah GoldiD, Joe JamesiD, Peter G. RyaniD, Jenny Solís, Ashley C. BanyardiD
Abstract
Understanding the mechanisms underlying the emergence and spread of high pathogenicity avian influenza virus (HPAIV) is critical for tracking its global dissemination, particularly via migratory seabirds, given their role in transmission over long distances. Scavenging seabirds, such as skuas, may act as both reservoirs and vectors, and have been linked to multiple outbreaks since 2021. Here, we report the detection of HPAIV H5N1 clade 2.3.4.4b in three Tristan skua (Stercorarius antarcticus hamiltoni) carcasses on Gough Island in the central South Atlantic Ocean. To investigate potential incursion routes, we combined genomic analyses with year-round tracking data from global location sensors. Although migratory movement patterns suggested southern Africa as the most obvious pathway, the strain detected on Gough Island was more closely related to that identified in South Georgia, indicating that infection may have occurred during the pre-laying exodus, when skuas disperse into frontal waters south of the island. No further cases have been confirmed for Gough, but more systematic monitoring is needed to understand the dynamics of virus infection. The detection of HPAIV H5N1 in skuas on Gough Island highlights the importance of continued vigilance, proactive and geographically inclusive surveillance strategies, and biosecurity measures globally, alongside efforts to reduce other pressures on globally important seabird populations to help strengthen their resilience.
- organism: virus
- organism: multiple species
- assay: whole-genome sequencing
- research focus: phylogenetics / evolution
- approach: Bayesian inference
- approach: statistical model
- approach: experimental assay
In-depth summary
Claude Opus 5.5 · xhigh effort · 5 Nov 2026 · from the full text of version 2 (22 Jan 2026) · CC BY-ND 4.0
Whole-genome sequencing and geolocation tracking indicate that high pathogenicity avian influenza H5N1 reached Tristan skuas on remote Gough Island via sub-Antarctic marine networks linked to South Georgia rather than Africa.
High pathogenicity avian influenza virus (HPAIV) H5N1 threatens isolated seabird populations, yet transmission pathways across remote Southern Ocean islands are difficult to resolve from outbreak surveillance alone. Combining viral whole-genome phylogenetics with year-round seabird geolocation tracking distinguishes sub-Antarctic pelagic incursion routes from continental wintering flyways.
Key results
ShownHigh pathogenicity avian influenza virus (HPAIV) H5N1 clade 2.3.4.4b, genotype B3.2, was detected in three dead Tristan skuas on Gough Island.
All 3 sampled carcasses in September 2024 tested positive for HPAIV H5N1 by rRT-PCR in brain (Ct 19.5-23.8) and oropharyngeal swabs while testing negative in cloacal swabs; 2/2 brain swabs yielded live virus in SPF eggs and produced whole genome sequences sharing >99% identity. · Results; Diagnostic evaluation of samples; Tables
ShownPhylogenetic and phylogeographic analyses link the Gough Island H5N1 strains to the South American and sub-Antarctic lineage via South Georgia rather than Africa.
Maximum-likelihood and TreeTime time-scaled phylogenies (944 reference HA sequences plus 2 Gough Island sequences) placed the Gough viruses in a monophyletic Antarctic/sub-Antarctic clade alongside ancestral South Georgia strains and Crozet/Kerguelen descendants, with South Georgia as the primary regional source. · Phylogenetic Analysis; Diagnostic evaluation of samples; Figure legends
ShownYear-round geolocation tracking shows that although Tristan skuas winter off southern Africa, outbreak timing aligns with their pre-laying sub-Antarctic exodus.
Eight GLS-tracked skuas (345 +/- 3.3 days) wintered on the South African and Namibian shelves for 141 +/- 19.5 days but returned to pre-laying foraging grounds near the Subantarctic front (95% KUD: 977,900 km^2) between 13 June and 1 August, weeks before the mid-September mortality event (including tracked bird SAFRING 888666). · Ringing and tracking data; Discussion; Figure legends
Methods, data and reproducibility
Brain, tracheal/oropharyngeal, and cloacal swabs from three dead Tristan skuas on Gough Island were analyzed by rRT-PCR, egg virus isolation, Oxford Nanopore GridION whole-genome sequencing, IQ-Tree phylogenetics, and TreeTime phylogeography. Movement tracks and kernel utilization distributions (SGAT, BAStag, adehabitatHR) were analyzed from 8 geolocator-tagged adult skuas.
- Brain, tracheal/oropharyngeal, and cloacal swabs from dead adult Tristan skuas (Stercorarius antarcticus hamiltoni) · virus · n = 3 sampled carcasses (out of 4 found); 2 WGS obtained
- EpiFlu reference HPAIV H5N1 clade 2.3.4.4b sequences (2020-2025) from Europe, Africa, Antarctica, and the Americas · virus · n = 944 HA-gene sequences plus 2 Gough Island sequences in HA tree
- Year-round light-level geolocation sensor (GLS) tracking data from adult Tristan skuas on Gough Island · multiple species · n = 8 recovered tags (10 deployed), 345 +/- 3.3 days, 687 +/- 7.8 locations/bird
- Methods & tools
- R, SGAT, BAStag, adehabitatHR, PARNAS, Mafft, Aliview, IQ-Tree, GenoFLU-multi, TreeTime
- Software & environment
- R version 4.4.1, SGAT (v0.1.3), BAStag (v0.1.3), adehabitatHR (v0.4.22), sp, sf, raster, Mafft v7.525, Aliview v.2021, IQ-Tree v2.4.0, TreeTime v0.11.4
- Reproducibility gaps
- EpiFlu accession numbers for the Gough Island viral genomes and raw geolocation tracking files are not listed in the text; the custom assembly pipeline has no public URL.
Limitations (3)
- Geolocation tracking data were collected in 2017-2018, up to six years prior to the September 2024 HPAIV H5N1 detection, rather than concurrently with the outbreak. · stated by the authors
- The exact timing and location of HPAIV transmission to the Tristan skuas remain unknown, as does whether skuas or other avian or marine mammal species first introduced the virus to Gough Island. · stated by the authors
- Large portions of Gough Island and the neighboring Tristan da Cunha Archipelago are inaccessible and lack systematic wildlife disease surveillance, meaning additional mortalities and regional viral spread may be underestimated. · stated by the authors
Read this if you track high pathogenicity avian influenza H5N1 spread in wildlife or combine viral phylodynamics with animal movement ecology in pelagic seabirds.
- Citations
- 7
- FWCI
- 27.75
- Field percentile
- top 1%
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Detection and spread of high pathogenicity avian influenza virus H5N1 in the Antarctic Region
A. Bennison, A. M. Byrne, … A. C. Banyard·bioRxiv·Nov 23, 2023
- Skuas as sentinels of high pathogenicity avian influenza H5N1 on the Antarctic Peninsula in the 2024/2025 austral summer
Michelle Wille, Will Abbott, … Meagan Dewar·Microbial GenomicsPeer-reviewed·Aug 14, 2026
- First detected incursions of avian influenza H5N1 clade 2.3.4.4b into mainland Australia from the Southern Ocean
Matthew J. Neave, Sam Hair, … Frank Y. K. Wong·bioRxiv·Aug 10, 2026
- Pathology of naturally acquired high pathogenicity avian influenza virus H5N1 infection in seabirds
F. Z. Lean, M. Falchieri, … A. Nunez·bioRxiv·Feb 17, 2023
- Impacts of high pathogenicity avian influenza H5N1 2.3.4.4b south of the Antarctic Circle
Simon B. Z. Gorta, Víctor Neira, … Cagan Hakki Sekercioglu·Ecological Solutions and EvidencePeer-reviewed·Jun 24, 2026
- Transcontinental Spread of HPAI H5N1 from South America to Antarctica via Avian Vectors
Xu Ruifeng, Gao Minhao, … Wei Wang·bioRxiv·Sep 6, 2025