H5N1 is in Australian resident birds, and overseas the first wave does most of the killing

Australian government reports show H5N1 circulating in resident birds, and hundreds of dead birds are now being reported as the strain spreads ^[1]. The overseas record says this is the start of the curve rather than the peak, and that the first wave tends to be by far the worst ^[4], which puts a hard deadline on any spending meant to change the outcome.

The lead time was long. The current outbreak emerged in 2020 and was spreading widely by late 2021, leaving a trail of devastation among wildlife across Europe and the Americas ^[2]. Australia and New Zealand stayed free of the strain even as Antarctic birds and mammals began dying through 2024 and 2025 ^[3].

What arrived is not the 1996 poultry virus, which produced periodic outbreaks, mass culls and sporadic wild bird die-offs ^[5]. H5N1 clade 2.3.4.4b, detected in 2020, replicates in both the gastrointestinal and respiratory tracts, where earlier strains tended to favour one or the other, and it can infect the central nervous system, causing neurological disease and death ^[6][7]. Its variants can infect more than 500 wild bird species, over half of which had no previously recorded infection ^[8]. It has killed foxes, deer and sea lions and infected cattle; human infections remain rare, with no reports of spread between people ^[9].

The first-pass numbers are the ones to plan against. Across western Europe and North America, mortality among breeding-age adults ran from 25% to 76% in the worst-hit species, and many populations had near-complete breeding failures ^[10]. When the virus reached South America in 2023, some bird species lost up to half their global population, and 17,000 elephant seals died in a single Argentine colony ^[11].

Then the curve bends. Mass deaths do not appear to strike the same populations year after year, even with the virus still circulating ^[12]. Adult sandwich terns died in droves in western Europe in 2022; in 2023 adult mortality fell by 96%, to roughly a twenty-fifth of the first-wave rate ^[13][21], because the older and more vulnerable birds had already died, survivors gained some immune resistance, younger birds moved into vacated breeding slots, and chicks faced less competition for food ^[14]. Sandwich tern numbers fell only modestly overall, and the species is still projected to need almost 20 years to return to pre-outbreak levels ^[15]. That is the asymmetry worth budgeting around: the loss is booked in a season or two, the repayment takes decades, and for species that shed a large share of their global population recovery is likely to take decades if it happens at all ^[20].

Exposure is uneven. Atlantic seabirds breeding in large, dense colonies, including gannets, terns and skuas, were hit hard, while neighbouring petrels, puffins and shearwaters nesting in burrows suffered far fewer deaths ^[16]. That structure may give some protection to Australia's burrow-nesting little penguins, and authorities are moving to vaccinate thousands of them ^[17]. Compounding pressure is the aggravating factor: Adelie penguins in Antarctica faced bird flu alongside avian cholera in 2024, and African penguins were hit while already in trouble from the collapse of sardine fisheries ^[18]. Extrapolation has limits, since many Australian birds and mammals are unique and some, such as black swans, lack crucial immune defences against influenza ^[19].

Three things are worth watching. Whether the little penguin vaccination program finishes before the wave reaches those colonies rather than during it ^[17]; whether dense-colony breeders here record the near-total breeding failures seen overseas ^[10][16]; and the second-season adult mortality figure, which overseas was the clearest read on acquired resistance ^[13].