Victoria’s new draft feral horse management plan, released on the last working day before Christmas, will be open for comment until February 2. But will it protect the Alpine National Park? The answers are yes on the Bogong High Plains, and no in the eastern Alps.
The government deserves congratulations for planning to remove all horses from the most sensitive alpine areas around Falls Creek by 2020. These areas of the Bogong High Plains have fewer than 100 horses, but also rare snow-patch and bog communities that are extremely vulnerable.
But elsewhere, the goal of removing 400 horses a year from the eastern Alps doesn’t seem to go far enough. And by refusing to countenance the idea of culling, the state government is passing up the only realistic chance of getting feral horse numbers under control.
The bulk of the plan provides grounds for cautious optimism. It acknowledges that feral horses threaten a range of native mammals, frogs and lizards, as well as displacing kangaroos and wallabies. Horses have enormous impacts on vegetation in alpine bogs and streams, and in many other ecosystems too.
The plan also makes clear that reducing horse numbers is a legal requirement. Victoria’s Flora and Fauna Guarantee Act 1988 lists “degradation and loss of habitats caused by feral horses” as a threatening process. The Victorian National Parks Act 1975 calls for “exotic species” such as horses to be exterminated or controlled within national parks.
The plan also sets a realistic time frame for review (annual reviews and major review after three years), and suggests that management plans will be altered if adequate environmental protection is not achieved. All of this is extremely promising, suggesting the state government is genuinely interested in delivering tangible environmental benefits.
But while the aspirations are good, the details present some problems. The draft plan promises to “explore all possible control options” to deliver a low horse population in the eastern Alps.
But the proposed reliance on trapping and removal, rather than culling, suggests the government is reluctant to enter what would be a tough debate against the often vocal pro-brumby lobby groups. This reluctance is to the detriment of our native species and apparently at odds with legislation.
The problem is that the New South Wales government has already tried trapping and removing horses in Kosciuszko National Park, and it hasn’t worked. Horses have continued to spread northward onto the main range, where environmentally sensitive alpine tarn and snow-patch communities occur.
It is unclear whether Victoria’s “aspirational goal” of removing 400 horses each year over three years will actually be enough to reduce horse numbers, or even to stabilise them. The report mentions modelling showing that the population can be stabilised by taking 200 horses per year, and that it would start to decline if 400 were taken per year.
But none of this modelling is published, so it can’t be evaluated in detail. And simple calculations suggest that these figures are incredibly optimistic.
The report says there were 2,350 horses in the eastern Victorian Alps in 2014. Horse populations can increase at up to 20% per year, so by now there could be more than 4,000 feral horses.
This means that even if the government does manage to remove the full quota of 400 horses each year, it would only take a 10% population growth rate for the numbers to keep rising. At a rate of 20%, there could be well over 5,000 horses by 2020, even with trapping and removal.
Based on this rough calculation, the plan needs to eradicate many more horses. The draft plan claims that feral horses in the eastern Alps are “well established and are considered beyond eradication using currently available control tools”. Yet this claim ignores aerial culling, which is the cheapest, most effective, and most ethical way to reduce feral horse numbers.
Highly trained sharp-shooters and helicopter pilot teams can destroy more than 50 horses per day (based on previous culls in NSW, in which three teams of three people destroyed 606 horses over three days). Three teams could solve the feral horse problem in the Victorian alpine country in a month, and at lower cost.
It cost taxpayers more than A$1,000 for each horse trapped and removed from Kosciuszko National Park. Using the NSW cull as a guide to the resources required, and assuming A$300 per day per person, and A$10,000 per day per helicopter, it might have cost around A$150 per horse using aerial culling. That’s roughly 15% of the cost of trapping and removal.
Despite the risks to wildlife canvassed in the draft plan, and similar reports from NSW, there is no peer-reviewed research that defines the threats to native animals. A revised plan must include research to understand both the impacts of feral horses on native animal populations and their welfare.
The debate over culling horses typically ignores the unseen suffering that horses cause to native animals. Quantifying that suffering will be crucial for making informed decisions around feral horse management.
It is great that we have a plan for managing horses in the Victorian Alpine National Park – albeit one that seems unlikely to work in the eastern Alps. But the Victorian government needs to show courage and leadership on the issue of culling feral horses. Our alpine natural heritage will continue to decline until horses are taken out of our national parks, and that will only happen when managers can include culling among their suite of management tools.
In NSW, the feral horses in Kosciuszko National Park are growing in number, and doing real damage to Australia’s highest mountains. Hopefully both states can take back the reins of feral horse management from single-issue lobby groups and exercise some real control over their feral horses.
Over the past year the global media has been full of reports of catastrophic fires in California, the Mediterranean, Chile and elsewhere. One suggested reason for increases in catastrophic wildfires has been human-induced climate change. Higher temperatures, drier weather and windier conditions all increase the impact of fires.
While climate change indeed raises the risk of wildfires, our research shows that another way humans can change patterns of fire activity is by introducing flammable plants to new environments.
Plantations of highly flammable exotic species, such as pines and eucalypts, probably helped to fuel the recent catastrophic fires in Portugal and in Chile. In arid regions, such as parts of the US southwest, the introduction of exotic grasses has transformed shrublands, as fires increase in severity.
Invasive plants and fire
How do invasive plants change fire patterns? We burned species mixtures (aka “mixed grills”) on our plant barbecue to help find out.
One of the main ways flammable invasive plants can have long-lasting impacts on an ecosystem comes from positive fire-vegetation feedbacks. Such feedbacks can occur when a flammable weed invades a less fire-prone ecosystem. By changing the available fuel the invader makes fires more likely and often hotter.
If the invading species has characteristics that allow it to outcompete native species after a fire, then it will further dominate the ecosystem. Such traits include thick bark, the ability to resprout following fire, or seeds that survive burning. This invasion will likely lead to more fires, changing the species composition and function of the ecosystem in a “fire begets fire” cycle. Extreme examples of this dynamic are where flammable grasses or shrubs invade forests, leading to loss of the forest ecosystems.
We wanted to understand how invasive plants interact with other species when burned in combination. To explore the mechanisms underpinning such feedbacks, we examined how invasive plants might change the nature of a fire when burned together with native species.
We collected 70cm shoots of four globally invasive species (of both high and low flammability) and burned them in pairwise combinations with New Zealand native trees and shrubs to determine which characteristics of a fire could be attributed to the invasive plants.
We found that overall flammability was largely driven by the most flammable species in the mixture, showing how highly flammable weeds could set in motion fire-vegetation feedbacks.
We established that a greater difference in flammability between the two species led to a larger influence of the more flammable species on overall flammability. This outcome suggests weeds that are much more flammable than the invaded community can have larger impacts on fire patterns.
Importantly, we also showed the influence of the highly flammable species was independent of its biomass, meaning highly flammable weeds may impact community flammability even at low abundances.
When we looked closer at the different components of flammability (combustibility, ignitability, consumability and sustainability) we found some important nuances in our results.
While the maximum temperature reached in our burns (combustibility) and the ignition speed (ignitability) were both most influenced by the more flammable species, consumability (the amount of biomass burned) and sustainability (how long the fire burns) were equally influenced by both the more flammable and less flammable species.
In short, more flammable weeds will cause a fire to ignite more quickly and burn hotter.
However, less flammable species can reduce the duration of a fire compared to when a more flammable species is burnt alone. These results could have important ecological implications, as the longer a fire burns the more likely it is to kill plants: low-flammability plants could reduce this impact.
Managing weeds to reduce fire impacts
Even low abundances of highly flammable invasive weeds could set in motion positive fire-vegetation feedbacks that lead to drastic changes to ecosystems. If this result holds when our shoot-scale experiments are repeated using field trials, then land managers should work quickly to remove even small infestations of highly flammable species, such as gorse (Ulex europaeus) and prickly hakea (Hakea sericea).
Conversely, the role of low flammability plants in extinguishing fires further supports the suggestion that the strategic planting of such species across the landscape as “green firebreaks” could be a useful fire management tool.
In any case, our “mixed grill” study further highlights the role of exotic plants in fuelling hotter wildfires.
I am walking quietly through the forest. As I reach the edge of the trees there is a snort and a staccato of hoofbeats, and four horses materialise only metres in front of me: a foal, two mares and a dark stallion. The stallion, ears pricked, tosses his head and prances forward. As I crouch to pick up a branch, the stallion wheels and gallops off with the group. They hurdle an old stock fence, and almost as soon as their hoofs touch down, another big grey stallion comes towards them over the hill.
The next minutes are completely mesmerising. The two stallions fight, 50 metres from me. Dust hangs in the air around them, their screams echo off the hills, the impact of their hoof strikes reverberates in my belly. They rear, scream; snake heads out to bite, whirl and kick. Eventually, bleeding and bruised, the dark stallion breaks and runs. The grey makes a show of chasing, then canters back to the mares, arching his neck, prancing with lifted tail.
This is one of many times I have seen horses, called brumbies in Australia, in the mountains. While cross-country skiing in the south I have watched them in the snow – ragged manes flying, galloping through a mist of ice crystals – and many times while driving and bushwalking in both the north and south of Kosciuszko National Park. I have also watched them cantering in clouds of dust in central Australia, and grazing in the swamps of Kakadu. Each of these wild horse encounters has been deeply visceral and emotional, elemental expressions of life in dramatic and beautiful landscapes.
Horses are large, powerful and charismatic animals, and humans have ancient connections to them. Wild horses are dominant among the 13 species painted on the caves of Chauvet in France 30,000 years ago, and while there continues to be debate, archaeologists suggest evidence for horse domestication is at least 5,500 years old. And like the oldest human-animal relationship outside hunting – with dogs – the horse relationship is unique because we now mostly do not eat this animal.
Like dogs, horses now occur on every continent except Antarctica, and humans have been the primary agent for their dispersal. In North America, where the first true horses evolved and then died out, they were reintroduced by Columbus in 1493. Horses are the most recent of the main species humans domesticated, and the least different (with cats) from their wild counterparts.
Australia has the largest wild horse herd in the world, maybe 400,000 or more horses, spread across nearly every bioregion from the tropical north to the arid centre to the alpine areas. That sounds like a dramatically large number, but Australia also has around one million domestic horses, about 100 million cattle and sheep, maybe 20 million feral pigs and 25 million kangaroos. But the presence of wild horses here is deeply controversial.
Six thousand of these horses are in Kosciuszko National Park. Ongoing controversy around these wild horses encompasses debate about their impact and their cultural meaning. There is very little systematic research and a large amount of emotive and anecdotal argument, from both sides. There is circularity and self-referencing in government wild horse management plans, very little reference to studies from Australia and almost no peer-reviewed research on horse impacts in the Snowy Mountains, despite decades of argument that they cause environmental degradation.
And Kosciuszko is right next to Canberra and the Australian Capital Territory, which has the highest per capita horse-ownership of anywhere in Australia. Several enterprises run horse-trekking trips into the Snowy Mountains, often interacting with brumbies. The Dalgety and Corryong annual shows on the boundaries of the park highlight horse skills, including catching and gentling brumbies. In many places mountain cattle properties are increasingly using horses instead of motorbikes to handle stock.
The Kosciuszko wild horses are also tangled within the embedded idiosyncrasies and contradictions of the largest national park in New South Wales. Here there are protected populations of two species of invasive fish (brown and rainbow trout) that are demonstrably responsible for local extinctions of native fish and frog species; a gigantic hydro-electric scheme with dominant infrastructure across large areas of the park; and expanding ski resorts where it is possible to buy lodges. Much of the landscape that is now part of the park has a long history of summer grazing by sheep and cattle, with stockworkers’ huts scattered across the high country. This “wilderness” has been home to Aboriginal people for millennia, as well as well-known grazing grounds for more than a century.
These complexities and contradictions reflect our often unconscious modern propensity for hubris: we insist we are in charge of what happens on the planet, including in its “wild” places and “wild” species. Terms like “land management”, “natural resource management”, and “conservation management”, all reflect this assumption of superiority and control.
The United States has similar controversies over the management of mustangs across large areas of the west. New Zealand has the Kaimanawa horses, a special and isolated herd on army land. In both of those countries, as in Australia, there is a unique history of horse interactions with Indigenous communities. The great Native American horse cultures are well known and extraordinary, as Indians had no introduction to equestrian skills from the Spanish invaders, they learnt extremely quickly from scratch.
The first horses in New Zealand were a gift to Maori communities from missionary Samuel Marsden in 1814, and a Waitangi Tribunal Claim has been brought to protect the Kaimanawa horses as Maori taonga (treasures). Aboriginal stockmen and stockwomen were the mainstay of the pastoral industry all over Australia until the equal wage ruling of 1968 resulted in the wholesale expulsion of Aboriginal stockworkers in north and central Australia.
Peter Mitchell’s recent book Horse Nations uses that term to describe the people-animal relationship in certain Indigenous communities. Both Native American and Aboriginal cosmologies often place animals including horses, as their own “nations”, with whom they have a responsibility to respectfully interact.
The wild horses of the Australian Alps are arguably the strongest cultural icons. The enduring legacy of The Man from Snowy River, both the iconic Banjo Paterson poem and the 1980s film, but also the Silver Brumby series of novels by Elyne Mitchell, still in print after nearly 70 years, idealise the strength, beauty and spirit of wild mountain horses. At least one source suggests that “the man” from Paterson’s poem was in fact a young Aboriginal rider.
This is not at all implausible – there is much documentation, as well as strong oral histories, of Aboriginal men and women working stock on horseback across the Snowy Mountains. The Aboriginal mountain missions at Brungle and Delegate both have many stories of earlier generations working as stock riders and also mustering wild mountain horses. David Dixon, Ngarigo elder, says
Our old people were animal lovers. They would have had great respect for these powerful horse spirits. Our people have always been accepting of visitors to our lands and quite capable of adapting to change so that our visitors can also belong, and have their place.
While the iconic figure of the cowboy and stockman is masculine, amongst Aboriginal stockworkers women and girls were likely as common as men and boys. In contemporary times, women far outnumber men in equestrian participation, and brumby defenders are equally represented by men and women. Four Australian horsewomen generously shared their knowledge and skills in the research that backgrounds this essay.
In the mid 1970s, I worked as a ranger in Kosciuszko National Park. In those days rangering was a seat-of-the-pants enterprise: we used to buy at least part of our uniforms out of our own money because the issued items were so inadequate, we taught ourselves to cross-country ski, we drank socially with the brumby-runners and other people from the surrounding rural communities.
In many places rangers were and are intimately part of the community, not seen as “public servants”. There is a complex and interesting relationship between university-educated national parks staff and local rural workers with deeply embodied knowledge and skills, with rangers acknowledging that they need the skills of these locals to carry out much animal-related work in the parks, including trapping and mustering wild horses. Recent proposals to helicopter shoot large numbers of wild horses in Kosciuszko would potentially sever this link. Helicopter shooting requires specific marksmanship skills not common in rural communities.
While we debate how to reduce our wild horse numbers, other countries are working to re-establish wild horse herds in Europe and Asia. It is often argued that domestication saved horses (and many other species) from extinction, aiding their establishment all over the planet while their wild ancestors diminished or disappeared. Creating populations of newly wild species is termed both “rewilding’ and ”de-domestication“, and there are numerous and increasing examples around the world. Some of these proposals include the reestablishment of species long extinct, or their ecological equivalents.
In the period increasingly accepted as the Anthropocene, species are both declining and flourishing. Domesticated species have been moved all over the world; other introduced species flourish in new landscapes, and many of these are escaped or released domesticates. In the oceans, as large predators have declined all the cephalopods (octopus, squid and cuttlefish) are increasing. Highly specialised species that evolved on isolated islands have declined precipitously, while generalist species are flourishing.
Global conservation management attempts to work against both of these trends: we attempt to suppress populations of flourishing species, while supporting or increasing populations of declining ones, including through translocations and captive breeding programs. These activities call into question the nature of nature in the 21st century: what is the “wild” in all this management and manipulation?
In these questions, the lives and cosmologies of Indigenous peoples, and the lives of other species, offer us serious teachings. The agency and intelligence of animals, the increasing discoveries of distinct cultures amongst animal populations, the agency of planetary systems in continually reorganising around changing inputs, all stand against the modern human insistence on control, stability and stasis.
While hiking mountain grasslands looking for wild horse bands, I have several times come across horse skeletons whitening in the sunlight, their energy and power transmuted back into the source from which new lives will spring. In a world where human societies are increasingly narcissistic, where our dominant concern is ourselves, recognising the agency and intelligence of other species can be deeply humbling.
Perhaps our task is to harmonise ourselves with these old and new environments, not continually attempt to “manage” them into some other state that we in our hubris think is more desirable, whether ecologically, economically or culturally.
Thanks to Adrienne Corradini, Jen Owens, Blaire Carlon and Tonia Gray for improving my understanding of horse and brumby issues.
Cats kill more than a million birds every day across Australia, according to our new estimate – the first robust attempt to quantify the problem on a nationwide scale.
By combining data on the cat population, hunting rates and spatial distribution, we calculate that they kill 377 million birds a year. Rates are highest in Australia’s dry interior, suggesting that feral cats pose a serious and largely unseen threat to native bird species.
This has been a contentious issue for more than 100 years, since the spread of feral cats encompassed the entire Australian mainland. In 1906 the ornithologist A.J. Campbell noted that the arrival of feral cats in a location often immediately preceded the decline of many native bird species, and he campaigned vigorously for action:
Undoubtedly, if many of our highly interesting and beautiful birds, especially ground-loving species, are to be preserved from total extinction, we must as a bird-lovers’ union, at no distant date face squarely a wildcat destruction scheme.
His call produced little response, and there has been no successful and enduring reduction in cat numbers since. Nor, until now, has there been a concerted effort to find out exactly how many birds are being killed by cats.
Counting the cost
To provide a first national assessment of the toll taken by cats on Australian birds, we have compiled almost 100 studies detailing the diets of Australia’s feral cats. The results show that the average feral cat eats about two birds every five days.
We conclude that, on average, feral cats in Australia’s largely natural landscapes kill 272 million birds per year. Bird-kill rates are highest in arid Australia (up to 330 birds per square km per year) and on islands, where rates can vary greatly depending on size.
We also estimate (albeit with fewer data) that feral cats in human-modified landscapes, such as the areas surrounding cities, kill a further 44 million birds each year. Pet cats, meanwhile, kill about 61 million birds per year.
Overall, this amounts to more than 377 million birds killed by cats per year in Australia – more than a million every day.
Which species are suffering?
In a related study, we also compiled records of the bird species being killed by cats in Australia. We found records of cats killing more than 330 native bird species – about half of all Australia’s resident bird species. In natural and remote landscapes, 99% of the cat-killed birds are native species. Our results also show that cats are known to kill 71 of Australia’s 117 threatened bird species.
Birds that feed or nest on the ground, live on islands, and are medium-sized (60-300g) are most likely to be killed by cats.
It is difficult to put a million-plus daily bird deaths in context without a reliable estimate of the total number of birds in Australia. But our coarse assessment from many published estimates of local bird density suggests that there are about 11 billion land birds in Australia,
suggesting that cats kill about 3-4% of Australia’s birds each year.
However, particular species are hit much harder than others, and the population viability of some species (such as quail-thrushes, button-quails and ground-feeding pigeons and doves) is likely to be especially threatened.
In Australia, cats are likely to significantly increase the extinction risk faced by some bird species. In many locations, birds face a range of interacting threats, with cat abundance and hunting success shown to increase in fragmented bushland, in areas with high stocking rates, and in places with poorly managed fire regimes, so cat impacts compound these other threats.
The threatened species strategy also prioritised efforts to control feral cats more intensively, eradicate them from islands with important biodiversity values, and to expand a national network of fenced areas that excludes feral cats and foxes.
But while fences can create important havens for many threatened mammals, they are much less effective for protecting birds. To save birds, cats will need to be controlled on a much broader scale.
We acknowledge the contribution of Russell Palmer (WA Department of Biodiversity Conservation and Attractions), Chris Dickman (University of Sydney), David Paton (University of Adelaide), Alex Nankivell (Nature Foundation SA Inc.), Mike Lawes (University of KwaZulu-Natal), and Glenn Edwards (Department of Environment and Natural Resources) to this article.
Throughout history, humans have taken plants and animals with them as they travelled the world. Those that survived the journey to establish populations in the diaspora have found new opportunities as they integrate into new ecosystems.
These immigrant populations have come to be regarded as “invaders” and “aliens” that threaten pristine nature. But for many species, migration may just be a way to survive the global extinction crisis.
In our recently published study, we found that one of the Earth’s most imperilled group of species is hanging on in part thanks to introduced populations.
Megafauna – plant-eating terrestrial mammals weighing more than 100kg – have established in new and unexpected places. These “feral” populations are rewilding the world with unique and fascinating ecological functions that had been lost for thousands of years.
Today’s world of giants is only a shadow of its former glory. Around 50,000 years ago, giant kangaroos, rhino-like diprotodons, and other unimaginable animals were lost from Australia.
Formal conservation distribution maps show that much of Earth is empty of megafauna. But this is only a part of the picture.
Many megafauna are now found outside their historic native ranges. In fact, thanks to introduced populations, regional megafauna species richness is substantially higher today than at any other time during the past 10,000 years.
Worldwide introductions have increased the number of megafauna by 11% in Africa and Asia, by 33% in Europe, by 57% in North America, by 62% in South America, and by 100% in Australia.
Australia lost all of its native megafauna tens of thousands of years ago, but today has eight introduced megafauna species, including the world’s only wild population of dromedary camels.
These immigrant megafauna have found critical sanctuary. Overall, 64% of introduced megafauna species are either threatened, extinct, or declining in their native ranges.
Some megafauna have survived thanks to domestication and subsequent “feralisation”, forming a bridge between the wild pre-agricultural landscapes of the early Holocene almost 10,000 years ago, to the wild post-industrial ecosystems of the Anthropocene today.
Wild cattle, for example, are descendants of the extinct aurochs. Meanwhile, the wild camels of Australia have brought back a species extinct in the wild for thousands of years. Likewise, the vast majority of the world’s wild horses and wild donkeys are feral.
There have been global calls to rewild the world, but rewilding has already been happening, often with little intention and in unexpected ways.
A small population of wild hippopotamuses has recently established in South America. The nicknamed “cocaine hippos” are the offspring of animals who escaped the abandoned hacienda of Colombian drug lord Pablo Escobar.
By insisting that only idealised pre-human ecosystems are worth conserving, we overlook the fact that these emerging new forms of wilderness are not only common but critical to the survival of many existing ecosystems.
Megafauna are Earth’s tree-breakers, wood-eaters, hole-diggers, trailblazers, wallowers, nutrient-movers, and seed-carriers. By consuming coarse, fibrous plant matter they drive nutrient cycles that enrich soils, restructure plant communities, and help other species to survive.
The wide wanderings of megafauna move nutrients uphill that would otherwise wash downstream and into the oceans. These animals can be thought of as “nutrient pumps” that help maintain soil fertility. Megafauna also sustain communities of scavengers and predators.
In North America, we have found that introduced wild donkeys, locally known as “burros”, dig wells more than a metre deep to reach groundwater. At least 31 species use these wells, and in certain conditions they become nurseries for germinating trees.
The removal of donkeys and other introduced megafauna to protect desert springs in North America and Australia seems to have led to an exuberant growth of wetland vegetation that constricted open water habitat, dried some springs, and ultimately resulted in the extinction of native fish. Ironically, land managers now simulate megafauna by manually removing vegetation.
It is likely that introduced megafauna are doing much more that remains unknown because we have yet to accept these organisms as having ecological value.
Living in a feral world
Like any other species, the presence of megafauna benefits some species while challenging others. Introduced megafauna can put huge pressure on plant communities, but this is also true of native megafauna.
Whether we consider the ecological roles of introduced species like burros and brumbies as desirable or not depends primarily on our own values. But one thing is certain: no species operates in isolation.
Visions of protected dingoes hunting introduced donkeys and Sambar deer in Australia, or protected wolves hunting introduced Oryx and horses in the American West, can give us a new perspective on conserving both native and introduced species.
Can a snake bring down a forest? If we’re talking about the Pacific island of Guam, the answer may well be yes.
Our research adds to mounting evidence that the killing of many of the island’s bird species by an invasive species of snake is having severe knock-on effects for Guam’s trees, which rely on the birds to spread their seeds.
Invasive predators are known to wreak havoc on native animal populations, but our study shows how the knock-on effects can be bad news for native forests too.
The brown tree snake was accidentally introduced to Guam in the mid-1940s and rapidly spread across the island. At the same time, bird populations on Guam mysteriously began to decline. For years, no one knew why.
But the ecological damage doesn’t stop there. The loss of native bird species has triggered some unexpected changes in Guam’s forests. Both the establishment of new trees and the diversity of those trees is falling. These changes show how an invasive predator can indirectly yet significantly alter an entire ecosystem.
Birds and trees
Birds are very important to trees. In the tropics, up to 90% of tree species rely on animals, often birds, to spread their seeds. Birds eat fruit from the trees and then defecate the undigested seeds far away from the parent tree’s canopy, where there are fewer predators and pathogens that specialise on that species, where competition for light, water and nutrients is less intense, and where seeds can take advantage of promising new real estate when old trees die.
Without birds, roughly 95% of seeds of two common tree species on Guam (Psychotria mariana and Premna serratifolia) land directly beneath their parent tree. Compare that with the nearby islands of Saipan, Tinian and Rota – none of which have brown tree snakes – where less than 40% of seeds land near their parent tree. On Saipan, seeds that escape their parent tree are five times more likely to survive.
Overall, for the roughly 70% of tree species on Guam that rely on birds to spread their seeds, research suggests that the bird deaths caused by the brown tree snake have reduced the establishment of new tree seedlings by 61-92%, depending on the species.
Forests’ future threatened
These numbers suggest that many tree species in Guam are under serious threat, which in turn threatens the species diversity of the island’s forests.
Treefall gaps appear when an adult tree dies, opening up the canopy and increasing the light that reaches the forest floor. Many species rely on this increased light for germination and early growth, so these gaps are hotspots for new seedlings.
We found that Saipan and Rota had roughly double the number of species of seedlings growing in these gaps, compared with Guam. What’s more, seedling species on Guam tended to be clumped together, as you might expect if more than 90% of seeds are falling beneath their parent trees.
We also found that birds are important in moving the seeds of certain types of species to gaps. In forests, “pioneer species” are those that rapidly colonise gaps, exploiting the increased light to grow fast and reproduce young. Crucially, we found pioneer species in all gaps on islands with birds, but in very few gaps on Guam, where these species could be at risk of being lost entirely.
Invasive predators are a reality for many ecosystems, particularly on islands, and the situation on Guam is particularly extreme. Perhaps nowhere else in the world has experienced such dramatic losses of native fauna as a result of invasion.
While these direct impacts of invasion are astounding, the indirect impacts cascading through the ecosystem are just starting to unfold, and may prove to be similarly catastrophic.
What if there was a humane, targeted way to wipe out alien pest species such as mice, rats and rabbits, by turning their own genes on themselves so they can no longer reproduce and their population collapses?
Gene drives – a technique that involves deliberately spreading a faulty gene throughout a population – promises to do exactly that.
Our study, published today in the journal Proceedings of the Royal Society B, suggests that under certain circumstances, genome editing could work.
Good and bad genes
The simplest way to construct a gene drive aimed at suppressing a pest population is to identify a gene that is essential for the pest species’ reproduction or embryonic development. A new DNA sequence – the gene-drive “cassette” – is then inserted into that gene to disrupt its function, creating a faulty version (or “allele”) of that gene.
Typically, faulty alleles would not spread through populations, because the evolutionary fitness of individuals carrying them is reduced, meaning they will be less likely than non-faulty alleles to be passed on to the next generation. But the newly developed CRISPR gene-editing technology can cheat natural selection by creating gene-drive sequences that are much more likely to be passed on to the next generation.
Here’s how the trick works. The gene-drive cassette contains the genetic information to make two new products: an enzyme that cuts DNA, and a molecule called a guide RNA. These products act together as a tiny pair of molecular scissors that cuts the second (normal) copy of the target gene.
To fix the cut, the cell uses the gene drive sequence as a repair template. This results in a copy of the gene drive (and therefore the faulty gene) on both chromosomes.
This process is called “homing” and, when switched on in the egg- or sperm-producing cells of an animal, it should guarantee that almost all of their offspring inherit the gene-drive sequence.
As the gene-drive sequence spreads, mating between carriers becomes more likely, producing offspring that possess two faulty alleles and are therefore sterile or fail to develop past the embryonic stage.
Will it work?
Initial attempts to develop suppression drives will likely focus on invasive species with rapid life cycles that allow gene drives to spread rapidly. House mice are an obvious candidate because they have lots of offspring, they have been studied in great detail by biologists, and have colonised vast areas of the world, including islands.
In our study we developed a mathematical model to predict whether gene drives can realistically be used to eradicate invasive mice from islands.
Our results show that this strategy can work. We predict that a single introduction of just 100 mice carrying a gene drive could eradicate a population of 50,000 mice within four to five years.
But it will only work if the process of genetic homing – which acts to overcome natural selection – functions as planned.
Experiments with non-vertebrate species show that homing can fail in some circumstances. For example, the DNA break can be repaired by an alternative mechanism that stitches the broken DNA sequence back together without copying the gene-drive template. This also destroys the DNA sequence targeted by the guide RNA, producing a “resistance allele” that can never receive the gene drive.
A recent study in mosquitos estimated that resistance alleles were formed in at least 2% of homing attempts. Our simulation experiments for mice confirm this presents a serious problem.
After accounting for low failure rates during homing, the creation and spread of resistance alleles allowed the modelled populations to rebound after an initial decline in abundance. Imperfect homing therefore threatens the ability of gene drives to eradicate or even suppress pest populations.
One potential solution to this problem is to encode multiple guide RNAs within the gene-drive cassette, each targeting a different DNA sequence. This should reduce homing failure rates by allowing “multiple shots on goal”, and avoiding the creation of resistance alleles in more cases.
To wipe out a population of 200,000 mice living on an island, we calculate that the gene-drive sequences would need to contain at least three different guide RNA sequences, to avoid the mice ultimately getting the better of our attempts to eradicate them.
From hype to reality
Are gene drives a hyperdrive to pest control, or just hype? Part of the answer will come from experiments with gene drives on laboratory mice (with appropriate containment). That will help to provide crucial data to inform the debate about their possible deployment.
We also need more sophisticated computer modelling to predict the impacts on non-target populations if introduced gene drives were to spread beyond the populations targeted for management. Using simulation, it will be possible to test the performance and safety of different gene-drive strategies, including strategies that involve multiple drives operating on multiple genes.