Six Amphibian Species Extinct in 18 Months. Scientists Warn the Pace Is Accelerating.

Mariana Mira holds the photograph like a holy relic. It shows a golden toad, small enough to fit in a child's palm, its skin the colour of fire against wet leaves in a Costa Rican cloud forest. The photograph was taken in 1987. The toad was declared extinct in 2004. But Mira, a conservation biologist at the Federal University of Rio de Janeiro, keeps the image pinned above her desk because it reminds her of what she has spent fifteen years trying to prevent: the moment when a species becomes a photograph.

That moment arrived six times between October 2024 and March 2026. Six amphibian species — four frogs, one salamander, one caecilian — were formally declared extinct by the International Union for Conservation of Nature after exhaustive field surveys found no living specimens. The declarations were not surprising. What surprised scientists was the speed. The pace of amphibian extinction, according to data published in Nature in February 2026, is now 45,000 times faster than the background extinction rate — the natural rate at which species disappear absent human activity. The data suggests we are in the opening phase of a mass extinction event comparable to the asteroid strike that ended the dinosaurs. Except this time, we are the asteroid.

The thing is, we have known this was coming. The United Nations Convention on Biological Diversity has held fifteen conferences since 1992. Governments have set twenty biodiversity targets. They have missed every single one. Meanwhile, the planet is haemorrhaging species at a rate that makes climate models look conservative. And unlike carbon emissions, extinction is irreversible. When the last golden toad died, it took 190 million years of evolutionary history with it.

What the Data Actually Shows

The February 2026 study, led by researchers at Yale University and the Zoological Society of London, compiled extinction records for 8,565 amphibian species tracked since 1500. The baseline data comes from fossil records, which show that in a stable ecosystem, roughly one species per million goes extinct each year. That is the background rate. It is the hum of evolutionary turnover — species appear, adapt, and occasionally vanish as environments shift over geological time.

The modern rate is nothing like that. Between 1500 and 1900, amphibian extinctions tracked roughly 100 times the background rate — elevated, but within historical variation. After 1980, the rate began to spike. After 2000, it went vertical. The study found that since 2004, at least 37 amphibian species have been confirmed extinct. Another 484 species are listed as Critically Endangered and have not been seen in the wild for more than a decade. If those species are gone, the true extinction rate is over 200 times higher than current estimates.

Here is what this means: biodiversity loss is not gradual. It is not a slow fade. It is a cascade. Species do not disappear one at a time in neat succession. They vanish in clusters, pulled down by interlocking crises — habitat destruction, disease, climate disruption, invasive species, pollution. Amphibians are particularly vulnerable because they live in two worlds. They need water to breed and land to hunt. Destroy either, and the species collapses.

The Fungus That Rewrote the Rules

The single deadliest threat is invisible. Batrachochytrium dendrobatidis, known as Bd, is a chytrid fungus that infects amphibian skin. The skin is not cosmetic for a frog. It is how they breathe, regulate water, and fight infection. Bd clogs the skin's pores, causing cardiac arrest. Infected frogs die within weeks.

The fungus was first identified in 1998. Genetic analysis published in Science in 2018 traced its origin to the Korean Peninsula, where it lived harmlessly in native amphibians for centuries. Then the global wildlife trade exploded. Bd hitched a ride on shipments of live frogs destined for zoos, laboratories, and the pet trade. By the 1980s, it had reached every continent except Antarctica. It spread faster than scientists could map it.

A 2019 study in Science estimated that Bd has caused the decline of at least 501 amphibian species and the extinction of 90. No other disease in recorded history has caused the extinction of so many species. Not Dutch elm disease. Not white-nose syndrome in bats. Not even the plagues that reshaped human civilisations.

The wildlife trade remains largely unregulated. The Convention on International Trade in Endangered Species (CITES) lists 1,082 species for which international trade is restricted or banned. Enforcement is inconsistent. A 2025 investigation by the Environmental Investigation Agency found that only 38 of 184 CITES signatories have functional border inspection systems capable of detecting smuggled amphibians. Seizures have increased 220 per cent since 2020, but prosecutions have not kept pace. Most traffickers pay administrative fines and resume operations within months.

The Targets Governments Keep Missing

In December 2022, governments gathered in Montreal and agreed to the Kunming-Montreal Global Biodiversity Framework. The centrepiece was a commitment to protect 30 per cent of the planet's land and oceans by 2030. The target is known as 30x30. It was celebrated as historic. It was also, according to data released in March 2026 by the UN Environment Programme, already failing.

As of April 2026, 17.3 per cent of terrestrial ecosystems and 8.4 per cent of marine ecosystems are under some form of legal protection. But legal protection does not mean ecological integrity. A 2024 study published in Nature Sustainability found that only 42 per cent of the world's protected areas meet the minimum standards for effective management. The rest exist on paper. They have boundaries, but no rangers. They have laws, but no enforcement. They are what conservationists call 'paper parks.'

The problem is funding. The Kunming-Montreal Framework calls for $200 billion per year in biodiversity finance by 2030. Current spending is $154 billion, according to the OECD. Most of that comes from domestic government budgets in wealthy nations. International finance — the money that would flow to biodiverse but cash-poor countries like Brazil, Indonesia, and the Democratic Republic of Congo — totalled $11 billion in 2025. That is less than the annual revenue of a mid-sized European airline.

Rewilding: The Projects That Work and the Ones That Do Not

Some conservation efforts are succeeding. The Iberian lynx, Europe's most endangered cat, was down to 94 individuals in 2002. Intensive captive breeding and habitat restoration brought the population to 1,668 by January 2026. The species was downlisted from Critically Endangered to Endangered in 2024. It is one of the few examples of a mammal pulled back from the brink.

Rewilding — the large-scale restoration of ecosystems by reintroducing apex predators, removing livestock, and letting natural processes reassert themselves — has become a flagship strategy. The most famous case is Yellowstone, where the reintroduction of grey wolves in 1995 triggered a trophic cascade that restored vegetation, stabilised riverbanks, and increased biodiversity across the entire park.

But rewilding is expensive, politically contentious, and hard to scale. A 2025 meta-analysis in Conservation Biology reviewed 127 rewilding projects across six continents. Forty-three per cent showed measurable increases in biodiversity. Thirty-one per cent showed no significant change. Twenty-six per cent failed outright, usually because of conflict with local communities over land use or livestock predation.

The projects that succeeded had three things in common: secure long-term funding, legal protection for restored areas, and genuine buy-in from local populations. The projects that failed lacked at least one of those. In Kenya's Laikipia Plateau, attempts to reintroduce lions collapsed in 2023 after pastoralists, excluded from planning, killed six of the reintroduced animals in retaliation for livestock losses. In Romania, a project to restore bison to the Carpathian Mountains succeeded because local governments created economic incentives — ecotourism revenue, employment as park rangers — that gave communities a stake in the animals' survival.

Ocean Dead Zones: The Crisis We Do Not See

Terrestrial biodiversity gets the headlines. Marine biodiversity is collapsing just as fast, but underwater. Dead zones — areas of ocean with oxygen levels too low to support most marine life — have quadrupled in size since 1950. As of March 2026, there are 1,476 documented dead zones worldwide, covering a combined area larger than the United Kingdom.

The cause is nutrient pollution — nitrogen and phosphorus runoff from agriculture. Fertilisers wash into rivers, flow into coastal waters, and trigger algae blooms. When the algae die and decompose, bacteria consume the oxygen. Fish suffocate. Coral reefs bleach. Entire ecosystems collapse.

The largest dead zone is in the Gulf of Mexico, fed by runoff from the Mississippi River basin. It covers 16,000 square kilometres during peak seasons. A 2024 study by the National Oceanic and Atmospheric Administration found that shrimp catches in the Gulf have declined 34 per cent since 2000, even as fishing effort has increased. The shrimp are not overfished. They are being asphyxiated.

The solution is straightforward: reduce fertiliser use, restore wetlands that filter runoff, and regulate agricultural discharge. But fertiliser is cheap and wetlands are being drained for farmland. The U.S. Environmental Protection Agency's 2021 target to reduce nutrient pollution in the Mississippi basin by 45 per cent by 2035 is, as of April 2026, on track to miss by at least two decades. Voluntary compliance has not worked. Mandatory regulation is politically impossible in states where agribusiness controls legislatures.

The Uncomfortable Data About What Comes Next

Scientists do not agree on how many species we are losing. Estimates range from 150 species per day to 24,000 per year. The disagreement is not about methodology. It is about what counts as known. The IUCN Red List has assessed 150,388 species. There are an estimated 8.7 million species on Earth. We have catalogued less than 2 per cent. Most species will go extinct before we know they existed.

What scientists do agree on is that we are approaching tipping points. Ecosystems are not linear. They absorb pressure until a threshold is crossed, and then they collapse suddenly. The Amazon rainforest is approaching that threshold. A 2023 study in Nature Climate Change found that 17 per cent of the Amazon has already been deforested. At 20 to 25 per cent, the forest's ability to generate its own rainfall will fail. The rainforest will become savannah. That transition would release 90 billion tonnes of carbon dioxide and cause the extinction of tens of thousands of species found nowhere else.

The same dynamics apply to coral reefs, Arctic permafrost, and the West Antarctic Ice Sheet. Each system has a point of no return. We do not know exactly where those points are. We know we are getting close.

What We Still Do Not Know

Mariana Mira still photographs frogs. She spends three months a year in the Atlantic Forest of Brazil, documenting populations of the golden lion tamarin frog, a species that exists in four isolated patches of forest totalling less than 300 square kilometres. The species is listed as Endangered. Mira thinks it should be Critically Endangered. The populations are too small, too fragmented, too vulnerable to a single fire or disease outbreak.

She does not know if her work will matter. She does not know if the species will still exist in ten years. What she does know is this: the photograph of the golden toad above her desk is not a memorial. It is a warning. It is what happens when we wait for certainty before we act. Extinction does not wait. It does not negotiate. It simply arrives, irreversible and final.

The question is not whether we can stop every extinction. We cannot. The question is whether we can stop the cascade before it takes down the ecosystems that make human civilisation possible. We know what needs to be done. We have the science. We have the tools. What we do not have is the political will to treat this crisis as if our survival depends on it. Because it does.