Estimates of how many species the planet loses per hour range from a fraction of one to roughly four, depending on which assumptions go into the calculation. The most widely repeated figure over the past three decades has been “up to 100 species a day,” which works out to about four per hour, but that number was extrapolated from models of tropical-forest loss and has been challenged as difficult to reconcile with the far smaller count of extinctions formally documented by scientists. The gap between modeled losses and confirmed ones is enormous, and understanding why it exists tells you more about the extinction crisis than any single hourly number can.
Where the Headline Numbers Come From
The claim that dozens or even hundreds of species vanish daily traces back to a modeling approach called the species-area relationship. The basic idea is straightforward: if you know how many species live in a given area of habitat, you can reverse the math and predict how many species should disappear when that habitat shrinks. Researchers have used this method for decades to project the consequences of tropical deforestation, and it can generate startlingly large numbers because tropical forests harbor such extraordinary concentrations of life.
The trouble is that the method tends to overestimate actual extinctions. The species-area curve was originally designed to predict how many species you would find when sampling a new area, not how many you would lose when the area shrank. Reversing it does not account for the fact that species cling to habitat remnants, shift their ranges, or survive in degraded patches far longer than the model expects. One analysis in the journal Ecology argued that the numbers simply do not add up when you compare species-area projections to real-world extinction records.1PubMed. Estimating extinction from species–area relationships: why the numbers do not add up That does not mean the losses are small. It means the most commonly cited per-hour figures rest on a foundation that most ecologists now consider too blunt.
The Mismatch Between Models and Confirmed Losses
For more than 30 years, some projections have suggested extinctions through tropical forest loss alone are running as high as 100 species a day, yet fewer than 1,200 extinctions have been formally recorded over the past 400 years.2Biodiversity and Conservation. Re-assessing current extinction rates That discrepancy is jarring and has fueled skepticism in some quarters. But treating confirmed extinctions as the real total would be a mistake for several reasons.
Declaring a species extinct is extremely conservative. The IUCN Red List, the most comprehensive global inventory of species status, applies stringent criteria before moving a species into the “Extinct” column. A species has to have been the target of exhaustive surveys across its entire known range, with no confirmed sightings, before it qualifies. For many poorly studied organisms, especially invertebrates, fungi, and tropical plants, those surveys have simply never happened.3Proceedings of the Royal Society B Biological Sciences. Three Decades of Classifying Threatened Species: Lessons Learned from and about the IUCN Red List Criteria for Quantifying Extinction Risk The Red List criteria were also originally designed around population declines measured over fairly short timescales, which raises questions about their ability to detect slower-burning threats like climate change.4PubMed. Detecting extinction risk from climate change by IUCN Red List criteria
So the confirmed count is almost certainly a serious undercount, while the headline model-based figures are almost certainly too high. The real number lies somewhere in between, and pinning it down per hour with any confidence is beyond what current data allow.
What We Can Say About the Pace
Even when researchers use cautious assumptions, the picture is alarming. A 2015 study used a conservative background rate of 2 mammal extinctions per 10,000 species per 100 years and compared it against documented vertebrate losses. Even with assumptions deliberately tilted to minimize the signal, the average rate of vertebrate species loss over the past century was up to 100 times higher than what would be expected naturally. The extinctions that did occur in a single century would have taken, depending on the group, somewhere between 800 and 10,000 years under normal background conditions.5PubMed Central. Accelerated modern human-induced species losses: Entering the sixth mass extinction
The numbers are even more extreme for specific groups. Amphibians may be disappearing at roughly 211 times the natural background rate, and if species currently on the brink are included, the figure could climb to tens of thousands of times higher.6Journal of Herpetology. Amphibian Decline or Extinction? Current Declines Dwarf Background Extinction Rate Freshwater fish tell a similar story: their modern extinction rate has been estimated at about 33 extinctions per million species-years, more than 100 times the natural background for that group.7PubMed Central. Global Patterns and Drivers of Freshwater Fish Extinctions: Can We Learn From Our Losses? These “times the background rate” comparisons are more scientifically grounded than any per-hour claim because they anchor the current pace to a well-studied baseline rather than extrapolating from habitat area alone.
Dark Extinctions and the Species We Never Knew
One of the most unsettling dimensions of this crisis involves species that vanished before anyone cataloged them. Ecologists call these “dark extinctions.” When European navigators reached oceanic islands and introduced rats, goats, and other destructive animals, ecosystems that had evolved in isolation were transformed before any naturalist arrived to record what lived there. For birds alone, statistical methods suggest roughly 56 species went extinct during the modern taxonomic period (since about 1800) without ever being formally described, and about 180 more disappeared in the centuries before systematic recording began.8PubMed Central. Dark extinction: the problem of unknown historical extinctions
The problem extends well beyond islands and history. Scientists have formally described roughly 2 million species, but estimates of total species richness on Earth range from 8 million to over 10 million. Most of the undescribed species are invertebrates, fungi, and microorganisms concentrated in the tropics. Research published in Conservation Letters found that undescribed species face higher extinction risk than known ones, likely because they tend to have small ranges and occur in under-surveyed, rapidly degrading habitats.9Conservation Letters. Undescribed species have higher extinction risk than known species Because every previous model of total species loss implicitly assumed undescribed extinctions could be ignored, the actual toll has been severely underestimated.10PubMed. Estimating how many undescribed species have gone extinct
This is part of why the per-hour question resists a clean answer. You cannot count what you never named. A comprehensive two-century analysis of Singapore’s biodiversity attempted to address this by applying statistical models that explicitly account for dark extinctions across more than 3,000 species in 10 taxonomic groups, and found that the loss picture looked considerably worse once undetected disappearances were factored in.11PubMed Central. Two centuries of biodiversity discovery and loss in Singapore
Which Groups Are Losing the Most
Extinction does not fall evenly across the tree of life. Among the roughly 163,000 species assessed on the IUCN Red List, about 0.6% are classified as extinct. But that average obscures steep variation. Mollusks, for instance, have lost nearly 3% of assessed species. Turtles, birds, and mammals show losses around 1 to 3%. Amphibians, ray-finned fishes, lizards, and snakes fall lower, between 0.2% and 0.5%. Among plants, assessed mosses and red algae have lost more than 1% of species, while vascular plants sit at about 0.2%.12Proceedings of the Royal Society B: Biological Sciences. Unpacking the extinction crisis: rates, patterns and causes of recent extinctions in plants and animals
Freshwater ecosystems are disproportionately affected. Nearly a third of the world’s freshwater fish species are at risk, and about 30% of freshwater mammals are threatened, with population declines of roughly 3.6% over a 20-year window.13PubMed. Extinction risk of the world’s freshwater mammals Freshwater habitats cover a tiny fraction of the Earth’s surface but support a wildly outsized share of vertebrate diversity, so losses there ripple widely.
At the genus level, documented extinctions look rarer. Among the thousands of assessed arthropod genera, only about 0.3% are extinct, and among plant genera, the figure is roughly 0.2%. Chordates and mollusks show significantly higher genus-level extinction than arthropods and plants.14PLOS Biology. Recent extinctions of plant and animal genera are rare, localized, and decelerated Losing a genus means losing an entire branch of evolutionary history, so even small percentages represent major losses of biological distinctiveness.
Where Extinctions Are Concentrated
Geography matters as much as taxonomy. Nearly half the world’s vascular plant species and a third of terrestrial vertebrates are confined to just 25 biodiversity hotspots. These are regions that combine high concentrations of species found nowhere else with severe habitat loss. More than half of all threatened terrestrial vertebrates are endemic to these hotspots, and the mismatch between the proportion of threatened species and the proportion that have actually been assessed suggests the real toll in poorly surveyed tropical hotspots is far higher than the Red List currently reflects.15Conservation Biology. Habitat Loss and Extinction in the Hotspots of Biodiversity
Land-use change, particularly the conversion of forests and wetlands to agriculture, remains the primary driver of species extinctions globally.16Conservation Biology. Future habitat loss and extinctions driven by land-use change in biodiversity hotspots under four scenarios of climate-change mitigation But no single threat operates in isolation. Climate change, overexploitation, invasive species, and habitat degradation interact in ways that are often greater than the sum of their parts. One threat can magnify another: fishing pressure may amplify the effects of warming oceans on intertidal species, and habitat fragmentation makes populations more vulnerable to climate shifts they might otherwise survive.17PubMed. Synergies among extinction drivers under global change These synergistic interactions make forecasting species loss harder and make the outcome worse than any single-threat analysis would predict.
Functional Extinction Happens First
Counting species that have formally vanished misses a dimension of the crisis that may matter more for ecosystems. A species can still exist, technically, while having already lost its ecological role. When a pollinator’s population crashes below the level needed to sustain the plants that depend on it, or when a predator becomes too rare to regulate its prey, the species is functionally extinct even though individuals remain alive somewhere. Research across natural and theoretical food webs found that when one species is pushed toward decline, up to 80% of the time, the first species to actually go extinct is a different one, typically something else in the network whose survival depended on the declining species doing its job.18PubMed. High frequency of functional extinctions in ecological networks
A broader review confirmed that the loss of ecological interactions is often decoupled from, and runs ahead of, the loss of species themselves. Pollination relationships, seed-dispersal partnerships, and predator-prey dynamics can collapse well before any participant species vanishes from the Red List.19Functional Ecology. Beyond species loss: the extinction of ecological interactions in a changing world This matters because ecosystem services people depend on, like crop pollination, pest control, and water purification, are delivered by interactions, not by species in isolation. Counting formal extinctions alone underestimates how much ecological function has already been lost.
Extinction Debt and Losses Already Locked In
Some of the extinctions we should be counting per hour have not happened yet but are essentially guaranteed. When a forest is fragmented, the species living in the remaining patches do not all die immediately. Long-lived organisms like trees may persist for decades or centuries in habitat that can no longer sustain their populations over the long run. This delayed reckoning is called an extinction debt. Studies of fragmented European landscapes found that the number of specialist plant species alive today correlates better with the amount of habitat that existed in the mid-20th century than with the amount that exists now, meaning many species still present are living on borrowed time.20PubMed Central. Habitat fragmentation causes immediate and time-delayed biodiversity loss at different trophic levels
A continent-wide analysis across 22 European countries reached a striking conclusion: the number of threatened species today matches socioeconomic pressures from the early to mid-1900s better than it matches current pressures. In other words, the biodiversity consequences of past human activity have not yet fully materialized. The negative effects of what we have already done to habitats will continue playing out for decades to come, regardless of what conservation measures are implemented starting today.21PubMed Central. Europe’s other debt crisis caused by the long legacy of future extinctions When people ask how many species we are losing per hour, the honest framing includes species whose doom is sealed but whose last individual has not yet died.
Conservation Action Does Work
The scale of the crisis can make the situation feel hopeless, but targeted conservation has a measurable track record. An analysis of bird and mammal extinctions since 1993 estimated that conservation efforts prevented between 21 and 32 bird extinctions and between 7 and 16 mammal extinctions. Given that 10 bird and 5 mammal species did go extinct (or are strongly suspected to have) in that same window, extinction rates would have been roughly 3 to 4 times greater without intervention.22Conservation Letters. How many bird and mammal extinctions has recent conservation action prevented?
Evidence from China reinforces the point: targeted conservation actions, such as habitat restoration and breeding programs, significantly improved the population status of threatened species, while factors like legal protection status or media attention on their own did not move the needle.23Biological Conservation. Targeted conservation actions are required to save species from extinction The lesson is that conservation works when it is specific and well-resourced, but broad designations on paper are not enough.
Finding Species We Thought Were Gone
New monitoring tools are reshaping what we know about which species still exist. Environmental DNA, or eDNA, involves filtering traces of genetic material from water or soil samples to detect the organisms living in or passing through an area. In Brazil’s Atlantic Forest, one of the most threatened ecosystems on Earth, eDNA analysis detected DNA from frog species that had not been seen for years, including one not recorded since 1968. The technique confirmed the presence of species that traditional survey methods had missed entirely, suggesting that some populations written off as locally vanished were still hanging on at very low densities.24PubMed. Lost and found: Frogs in a biodiversity hotspot rediscovered with environmental DNA
A similar approach was applied to the Vietnamese pond turtle, one of the most endangered turtles in the world. Researchers detected its DNA in aquatic habitats across several Vietnamese provinces, including a protected area where the species had never been documented in the wild before.25Environmental DNA. In Search of One of the Most Endangered Turtles in the World, the Vietnamese Pond Turtle (Mauremys annamensis), Using Environmental DNA These rediscoveries do not undo the broader trend, but they illustrate that our extinction counts are shaped partly by how hard we look. Better surveillance tools could both sharpen the per-hour estimates and, occasionally, offer a reprieve by finding populations that were never truly gone.
What Extinction Costs Economically
Behind every species lost is a web of ecological services that the economy depends on, even if markets do not price them. A model developed through the National Bureau of Economic Research explored how individual species contribute to ecosystem productivity and, through it, to economic output. The economic value of losing any one species depends on how many similar species remain to fill the same ecological role, how important that role is for overall ecosystem function, and how tightly a given country’s economy relies on the ecosystem services in question.26NBER. The Economics of Biodiversity Loss Countries in the tropics, where species richness is highest but so is the rate of loss, face the steepest potential costs because they tend to depend more directly on natural-resource-based economic activity. The framing shifts the conversation from an abstract species tally to a more concrete question: what does each lost species cost the communities that depended on what it did?