Since 1959 the avian flu virus H5N1 has been popping up around the globe. Now scientists believe it could spark the next pandemic.
Along the untamed coast of southern Victoria, where thyme rice-flower and coast daisy hustle for space and mud flats melt into aqua ocean, a team of birdwatchers is searching for an invisible enemy.
For almost 45 years, Rosalind Jessop has tracked the health of communities of wild birds, gathering data on flock sizes and numbers of young, to draw conclusions about species health and breeding patterns.
But three summers ago, the focus of her volunteer work shifted.
It is on beaches just like these that scientists and nature lovers including Jessop now fear Australia's next pandemic will be born, flown in on board one of these migrating birds, or carried into local waters inside a seal or sea lion.
The virus causing most concern right now is not a coronavirus like COVID-19's SARS-Cov-2 that emerged in bats. This time biosecurity experts have their eye on H5N1, a strain of the many influenza "A" viruses that can cause the flu in humans and often originate in birds.
But this avian flu virus — with the mundane-sounding subtype of 2.3.4.4b — is not so mundane after all. It has experts whispering about "pandemic potential".
There are two big problems with avian influenza H5N1 2.3.4.4b.
Firstly, it is classified as "highly pathogenic", meaning it makes the animals it infects very sick. Most die.
More significant, perhaps, is that H5N1 2.3.4.4b is transforming before our eyes. Already devastating wild birds and farmed poultry, the virus is rapidly learning to infect more and more species, including mammals.
Experts describe its evolution as like watching a new pandemic emerging in real time.
"We stand on the brink of an avian influenza pandemic that could be significantly worse than COVID-19," wrote Matthew S Miller from Canada's MG DeGroote Institute for Infectious Disease Research last week, as medical journal Nature Medicine wrote starkly that the virus poses "a risk for a human pandemic".
The United Nations' Food and Agriculture Organisation has warned the spread of H5N1 has already reached an "unprecedented" scale that threatens food security and global economics.
"It's a very, very well-trodden path," says Edward Holmes, Professor of Virology at the University of Sydney, listing off the route from isolated virus to pandemic via millions of infections and mutations in a process we all grew to understand after taking a crash cause in amateur virology living through COVID-19.
"It's a numbers game," says Professor Holmes. "Every day we roll the dice, then we roll it again. And one day, our number could come up."
So far Australia's number has not come up: the only region on earth along with New Zealand and the Pacific not infected with H5N1 2.3.4.4b.
The question is: How long can we outlast it?
Tracking a serial killer
You might wonder what the difference is between H5N1 and the dozens of other avian influenza "A" strains that circulate the globe at any point in time. Why we are worrying about a virus that hasn't even arrived in Australia yet when several other avian flus are impacting our poultry industry, leading to mass culls of infected birds and egg shortages at the supermarket.
Notwithstanding this dire impact, the avian viruses present in Australia come from strains and are currently under control. All but one is considered low pathogenic. They are unlikely to pose a wider threat.
But H5N1 is different. While not yet a pandemic, it is already considered a panzootic, the animal equivalent.
From its quiet debut in Scotland in 1959, H5N1 gained global attention when it infected geese in southern China in 1996 and 18 humans in Hong Kong a year later, leading to six deaths.
"From the moment we knew about H5N1 in a goose in China we knew that given the constellation of that virus it could become a problematic one," says Deakin University's Chair in Ecology, Marcel Klaassen. "And it has proven to be a problematic one."
Virologist Holmes sees it similarly: "There's been a lot of attention on this virus. It's a kind of hardy perennial to worry about."
H5N1 went on to infect wild birds and then spread easily via the birds' migratory pathways that criss-crossed the globe, mutating as it went. By 2015, it was found throughout Asia, the Middle East, North Africa and Europe.
Then, like losing the trail of a serial killer, H5N1 went silent. Scientists hoped they had seen the last of it. But in 2021 H5N1 re-emerged in its current mutated subtype, or clade, and its evolution began to gain pace.
"What happened in late 2021 is that the ecology of the virus changed," Holmes says, explaining that the way H5N1 interacted with its hosts, the environment and ecosystems changed the way it was able to spread and evolve.
Klaassen puts it more bluntly: "The shit truly hit the fan," he says, capturing the scientific community's sense of alarm.
Most significantly, the virus showed rapid global expansion in wild birds and farmed poultry. By the end of the year H5N1 2.3.4.4b had reached the US.
When sea mammals such as seals and sea lions ate sick or dead wild birds, they too became infected. On land, foxes and raccoon dogs caught the virus. By the end of 2022 H5N1 was also found in farmed mink that had come in contact with infected waterfowl and significantly, the virus began to spread in the US.
Then, in March last year, another important turning point: H5N1 was found in dairy cattle, probably first transmitted by contact with an infected bird's mucus or faeces, then spread from cow-to-cow via infected milking equipment.
Fast-forward to 2025 and H5N1 is now rampant in Europe, Asia, South America and the US, where it's running wild in the poultry and dairy industries — not helped by Donald Trump's new secretary of health and human services, Robert F Kennedy Jr, who advocates letting the virus rip.
Danger ahead
The danger zone for the development of a pandemic occurs when a virus infects mammal hosts that are in close contact with humans. Pigs have historically served this purpose. But scientists are concerned that dairy cattle are already acting as the new mammalian gateway for H5N1.
"This is why people are now so worried," says Holmes. "Being a bird-to-human virus was a big evolutionary gap but now it's in these other mammals that gap has narrowed. It's like 'Oh my God, that's a big concern'."
And last week a new, ominous, development: In the UK H5N1 was found in a sheep on a Yorkshire farm, also infecting its milk.
Unpasteurised milk is bursting with virus. In the US the disease has been found in cats that roam nearby dairy herds scavenging a sneaky drop here and there. Raw pet food has also been implicated.
"The milk has just got so much virus in it. It's got more virus in it than we can make in the lab. It's actually quite unbelievable," says Dr Michelle Wille, an expert in the ecology and evolution of avian flu viruses at the Doherty Institute's Centre for Pathogen Genomics at the University of Melbourne.
And yes, humans are being infected, too.
In January, a US man died from an H5N1 infection picked up from his backyard chickens. When his blood was analysed scientists discovered something alarming: the virus had already begun to mutate at two sites that would allow H5N1 to adapt to better replicate in humans. In February, three US vets working with dairy cattle contracted H5N1. Since 2003 almost 1000 human cases have been reported, including one in Australia carried here by a child travelling from overseas.
The global mortality rate from known human infection with H5N1 is extremely alarming — reported officially as almost 50 per cent. While the virus has proven it can be deadly and should not be underestimated Holmes also advises logic, arguing the figure is misleading because their sufferers recovered and many cases were never identified and reported. Of 70 recent positive cases known of in the US, only one person has died.
It is suspected that previous flu infection will offer some protection but how that works in reality can't be yet predicted and is influenced by how the virus mutates.
It can feel tricky to reconcile all this uncertainty with the advice not to panic. For now, the saving grace is that like other avian influenzas, H5N1 does not spread easily to or between humans. Unlike COVID-19 that spreads through the air, H5N1 2.3.4.4b spreads most effectively (at least at this stage) by direct contact with secretions like saliva, mucus or faeces as well as ingesting dead animals or unpasteurised milk, itself a form of secretion. The good news for humans is these are all avenues that can largely be predicted and avoided.
Yet the question keeping scientists up at night is this: as H5N1 2.3.4.4b infects more and more mammals, how many mutations will it take to develop the ability to spread between humans or through the air like COVID-19?
It's something a virologist like Holmes is watching closely.
"In almost 30 years since we've known about H5N1, it's not really ever yet evolved human-to-human transmission," he says. "It's had a lot of opportunities because it spills over into humans all the time so that's the key question. Will it become a human-transmitted pathogen?"
To do it, H5N1 needs a new set of mutations, Holmes says. Signs of such mutations in the pathology of the US man who passed away after contracting H5N1 is unsettling and suggests the virus is trying its best to build them.
"Significantly for humans there is evidence of transmission between mammals and it's picking up some mutations that we know allow it to become mammal adapted," he says.
Examples of these adaptations might be mutations to allow the virus to tolerate different pH levels or temperatures or similarly to COV-Sars-2, the virus might develop a little hinge that allows it to sneak in some amino acids that make human transmission so much easier.
"They allow critically better adaptation to human cell receptors," Holmes says. "As it spreads between mammals it's going to pick up more of those mammal adaptive mutations."
Strange as it seems, at the moment H5N1 is entering mammals using avian receptors, an ancient biological hangover that remains with some modern mammals such as cattle.
And some positive news to hang on to, Holmes says, is that not every animal virus mutates to affect humans. He points out equine flu that infects horses has never made the leap despite close contact with humans. But regardless, he says, it must not be forgotten that the biological motivation of H5N1 will continue to drive it towards adapting to as many species as possible. That includes humans.
"Because that's what evolution is designed to do," he says.
The more birds and mammals it infects the more opportunities it has to work out a solution.
Fighting on the beaches
While scientists wait for H5N1 to make its next microscopic move, back on the Victorian coast Rosalind Jessop is fighting on the frontline in the battle to detect the virus's arrival in Australia, and to prevent it spreading here.
Flocks of wild shore birds migrate to and from Australia each spring using what's known as the East Asian-Australasian Flyway, a kind of global super-highway for migratory birds. It stretches from Siberia inside the Arctic Circle to Australia, branching in different directions through Asia as the birds stop to rest and feed on their journey south, and later, back north again.
Land reclamation caused by industrial and urban development has devastated native habitats along the flyway and for years Jessop's work has focused on charting that impact, recording declining population numbers along with fewer young birds reflecting breeding struggles.
But now there is a new threat. Everywhere the birds travel along the flyway has outbreaks of H5N1. Not only is the virus killing birds that are already under threat but the presence of H5N1 on the route to Australia increases the risk of the virus arriving here.
As well as tagging birds with little flags holding tiny geo-locators, Jessop — who has a PhD in conservation biology — is helping wage a kind of guerilla war against the virus. With members of the Victorian Wader Study Group — that also includes scientists like Deakin University ecologist Marcel Klaassen, and the Doherty Institute's avian flu expert Michelle Wille who invest their own research funds into the work — Jessop monitors bird arrivals and movements every year, peaking between spring and the end of summer.
Wild birds leave the northern hemisphere summer after breeding and feeding 24 hours a day under the midnight sun, to spend a second summer here where weather is better and food more plentiful than toughing out a Siberian winter. The Waders set traps while the birds are busy feeding on the mud flats and then when they return to their roosts, Jessop and her team members spring into action.
They fire canon nets to confine as many as several hundred birds at a time and then set to work. Animal welfare standards give the researchers four hours to gather blood and other samples to determine if the flock is harbouring an H5N1 outbreak, or antibodies to the virus that suggest a bird has had a previous infection and recovered. If a recovered bird makes it to Australia that flags concern — increasing the risk birds might arrive while still infectious. The data helps Klaassen, Wille and Jessop deduce how H5N1 is spreading and most importantly of all — if it has arrived in Australia.
It's good news so far.
"Testing allows us to get an impression of what the exposure to this nasty virus is of the birds that arrive to Australia," Klaasson says. "I can assure you it is a lot of work and we have really done our utmost. Of the birds we have caught and sampled there is not a single indication of this nasty lineage."
Similar groups to the Victorian Waders are working on coastlines around Australia, guided by a complex algorithm of risk that helps to pinpoint the species most likely to bring the virus to Australia and the locations where it's most likely to happen.
The red-necked stint — a long-distance flyer — is one species on the radar. And the coast adjacent to Broome in Western Australia is considered one of the highest risk locations in the country, given its proximity to Asia — increasing the chance a migratory bird may arrive with an H5N1 infection rather than dying before it gets here — along with the huge number of birds it attracts to the beaches where they gather.
Yet a game plan that relies on chance sightings of sick or dead birds along with capture and testing of random individuals might feel rudimentary, fluky even, given what is at stake. Dr Tiggy Grillo heads Wildlife Health Australia, a coordinating body that supports the health of Australian wildlife including a surveillance program that supports the work of Klaassen, Wille and bird groups like the Victorian Waders. She says surveillance is the best strategy Australia has right now to prevent H5N1 taking hold.
"I'm continuing to say 'if' rather than 'when' the virus arrives," she says, optimistically. "We can't stop it arriving. We can't put up boundaries for wildlife. But there are a lot of variables that need to align for the virus to get here. We know the risk pathway is there but the message to Australians is keep a look out. Know what risk looks like and identify sick or dead wildlife that is new or unusual."
This strategy, Klaassen says, requires all of us to keep an eye out along our coastlines and areas where birds and wildlife gather and then quickly report any sightings of groups of sick or dead wildlife. "If H5N1 arrives here I think it will show itself very quickly," he says, explaining that the highly pathogenic nature of the virus will rapidly kill any wildlife it infects and dead birds or seals for example will start to wash up on shore.
Yet with millions of birds flying in and out on the Flyway after passing through infected countries — why hasn't H5N1 yet been detected? One reason is that Australia's isolation is offering crucial protection: most infected birds have died before arriving. Because the virus is not yet present in the Pacific, birds who fly shorter distances are not carriers.
But things can change quickly.
"What we are seeing at the moment is that the evolutionary rate of the virus is really high," Klaassen says. "That means that when the virus develops quickly you get new variants and maybe then there will also be a variant that happily sits on board a shore board for 10 days and arrives into Australia."
Then just when scientists were beginning to feel they understood the shape of that risk, late last year a dramatic and unwelcome shift took place that has completely changed the game.
A new threat from the south
Towards the end of 2023 H5N1 was found on the subantarctic islands of South Georgia, South Sandwich and also the Falklands after jumping from South America where more than 100,000 wild birds were reported dead in Peru alone, along with farmed poultry and marine mammals.
Then, a year ago two brown skua — a kind of predatory sea bird — washed up, dead, on Antarctica's western coast. Testing showed the birds had died from H5N1.
With H5N1 confirmed as circulating in Antarctica, where recent testing has found it in 24 of 27 sites investigated, affecting 13 bird and animal species, Australia's risk profile has also changed.
Distance can no longer protect us.
The virus has been found on Marion, Crozet and also Kerguelen Island which is only 440 kilometres from Australia's Heard and McDonald Islands and 4000 kilometres from Perth. This is a distance that is easily flown by brown skua which regularly hunt off the Australian coastline. Australian fur seals are also active in the Southern Ocean, another potential carrier of the virus if they eat dead birds or interact with Antarctic seals that are H5N1 carriers.
Wille, from the Doherty Institute, has spent countless hours on Australia's coast with researchers like Jessop and has co-authored academic articles with Klaassen that investigate the role of wild birds in spreading avian influenza. She is like a detective mapping criminal activity and analysing the virus's escape routes.
In recent months a new suspect has emerged on Wille's radar: the northern giant petrel. This bird breeds on subantarctic islands including Australia's Macquarie Island and spends time in the waters of Australia's south-east and Tasmania during winter, arriving as early as May. It feeds on fish, squid and krill but is also an aggressive hunter of other birds. It loves to scavenge carrion, a behaviour that puts the petrel at high risk of picking up H5N1.
With winter months approaching in the southern hemisphere attention has shifted from the northern route via the Asian-Australiasian Flyway to the more recently discovered southern route.
Wille is concerned that this Australian winter, just a couple of months from now, could see the first cases of H5N1 reach Australia very soon.
Australia's H5N1 risk assessments — written in 2022 — need urgent updates to take into account this new risk pathway which she says "is becoming more likely".
"We need to re-evaluate how we approach that risk," she says, "probably by doing a little bit more surveillance of our sea birds and particularly species like the Northern Giant Petrels for example which are moving really vast distances."
A catastrophe lurks
With so much focus on the H5N1's potential impact on Australian agricultural producers and the threat of a pandemic to humans, the reality of what this virus could do to Australia's unique native species can feel overlooked.
"We are all focused on humans," says virologist Holmes, "but in fact the biggest impacts may be on native wildlife."
There is little doubt that if H5N1 takes hold here, our native animals are at extreme risk — from fur seals to black swans, that lack some of the genes that help fight viruses, and predatory marsupials like Tasmanian devils, Australia has many wildlife species that could be devastated by an H5N1 outbreak. Not to mention the icons like kangaroos, koalas, kookaburras and sulphur-crested cockatoos.
When Wille lists what is so far known about H5N1's effects on native species elsewhere in the world the thought of its effect on Australia's wildlife is chilling.
"This virus has been really catastrophic to wildlife everywhere it's showed up," she says, listing with resignation the known impacts on a range of species:
- 500,000 wild birds dead in South America;
- 40 per cent of Peruvian pelicans dead in a matter of months;
- Populations of great skuas that breed in Scotland down by 60 per cent;
- 16 per cent of Europe's Sandwich Terns have died in one year;
- 30,000 South American sea lions have died equal to around 10 per cent of the population;
- 18,000 southern elephant seals have died
The mass death and even extinction of any animal species is devastating and Wille says the risk to Australian species is real and "it's something that deserves our attention".
"If this virus arrives in Australia it's probably going to have the same impact here as it has had everywhere else," she says. "This is not something to be trifled with."
Insurance populations of threatened Australian wildlife already exist — notably Tasmanian devils. Similar programs are in place across Northern Australia, for example, to protect the Mala, or Rufous hare-wallaby, as well as threatened species such as greater bilbies and burrowing bettongs
Grillo, from Wildlife Health Australia, says this strategy may need to be expanded, creating a sort of contemporary Noah's Ark.
"The threat of this virus is new. It is unprecedented," she says. "We've been preparing for the potential for a pathogen to impact wildlife for a very long time. If this virus was to arrive, that preparation means hopefully we will be ready."
If you see a sick or dead bird in Australia, report it to the emergency disease watch hotline 1800 675 888.
Credits
Words and production:
Illustrations: Lindsay Dunbar
