Carrying capacity is the maximum population size an environment can sustain indefinitely given its available resources, and one of the most vivid examples comes from a remote island in the Bering Sea. In 1944, 29 reindeer were released on St. Matthew Island; by 1963 the herd had exploded to 6,000, and by 1966 it had crashed to just 42 survivors. That boom-and-bust arc is a textbook illustration of what happens when a population overshoots the limit its habitat can support. But carrying capacity shows up in settings far beyond arctic islands, from laboratory flasks of single-celled organisms to urban water systems and global food production, and its real-world behavior is messier and more interesting than the tidy curves found in introductory ecology textbooks.
The St. Matthew Island Reindeer
The St. Matthew Island case is probably the most frequently cited carrying capacity example in ecology, and for good reason: it played out with dramatic speed and almost no confounding variables. Twenty-nine yearling reindeer were introduced to the uninhabited island in the Bering Sea in 1944 as part of a U.S. military wildlife program. With abundant lichen, no predators, and no human harvest, the population grew to roughly 1,350 by 1957 and then to about 6,000 by 1963. Over the following three years, the herd collapsed by 99%, leaving only 42 animals by the summer of 1966.1Rangifer. St. Matthew Island reindeer crash revisited: Their demise was not nigh—but then, why did they die?
What happened was a classic overshoot. The reindeer consumed lichen far faster than the slow-growing plants could regenerate. When a harsh winter hit in 1963-64, the food supply simply was not there. Starvation killed almost the entire herd. The island’s carrying capacity for reindeer had been far lower than 6,000, probably somewhere in the low thousands at most, but without predators or disease to regulate growth, the population blew past that limit and kept climbing until the resource base collapsed beneath it. The case is useful precisely because it shows that carrying capacity is not a wall a population bumps into and levels off at. It is a threshold that, when overshot, can result in a crash well below the original sustainable level, because the overshoot itself degrades the habitat.
Moose, Wolves, and Isle Royale
If St. Matthew Island shows what happens without predators, Isle Royale National Park in Lake Superior shows what happens with them. The island’s moose and wolf populations have been studied continuously since 1958, making them one of the longest-running predator-prey datasets in ecology. Modeling work on the Isle Royale system has shown that increasing predation rates led to lower peak moose densities and higher forage availability, which in turn raised the landscape’s carrying capacity for moose in the long run.2PubMed. Modelling moose-forest interactions under different predation scenarios at Isle Royale National Park, USA
That sounds counterintuitive: wolves eating moose actually increased how many moose the island could eventually support. The mechanism is that predation prevented the moose from overgrazing their browse species, especially balsam fir. With more forage remaining on the landscape, the habitat stayed healthier and could sustain more moose across decades. Simulations showed that after about 40 years, moose density was similar across different predation scenarios, but available forage biomass and carrying capacity continued to diverge. Heavier predation preserved the food base better.3PubMed. Modelling moose-forest interactions under different predation scenarios at Isle Royale National Park, USA Where the St. Matthew reindeer illustrate overshoot in the absence of regulation, Isle Royale illustrates how natural regulation by predators can maintain or even increase carrying capacity by protecting the resource base.
Carrying Capacity in a Lab Flask
Long before the St. Matthew reindeer, the Russian ecologist G.F. Gause demonstrated carrying capacity in laboratory experiments with tiny single-celled organisms called paramecia. In the 1930s, Gause grew populations of Paramecium aurelia and P. caudatum in controlled flasks with measured food supplies. The populations grew rapidly at first, then slowed and leveled off as resources became scarce, tracing the S-shaped (logistic) curve that has since become the standard textbook illustration of population growth approaching carrying capacity.
Modern re-analysis of Gause’s original time-series data has confirmed that his results fit logistic growth models well. Researchers testing several population dynamics models against Gause’s data found that the Verhulst and Gompertz models both provided strong fits, supporting the classic interpretation that the paramecia populations were regulated by density-dependent resource limitation.4Europe PMC. Analysis of G.F. Gause experimental time-series by means of continuous-time models The flask experiments are valuable because everything is controlled: food input, temperature, waste removal. That makes it possible to isolate density-dependent regulation from the confounding factors that complicate field studies. A paramecium flask is about as clean an example of carrying capacity as you will ever find.
Forests and the Self-Thinning Rule
Carrying capacity applies to plants just as much as animals, though it plays out differently. In dense forest stands, individual trees compete for light, water, and soil nutrients. As trees grow larger, fewer of them can fit in a given area, and the weakest individuals die off. Ecologists call this process self-thinning, and it follows a remarkably predictable pattern: average plant size increases as density decreases, tracking a consistent slope when plotted on a log-log scale.
This pattern emerges because a stand that has reached its carrying capacity settles into a state where the average intensity of competitive interactions among individuals becomes roughly constant. When a stand hits that ceiling, further growth by surviving trees can only happen if some trees die and free up resources.5Ecological Modelling. Self-thinning rule: a causal interpretation from ecological field theory If you have ever walked through a young pine plantation and noticed that many of the smaller trees are dead or dying while the larger ones thrive, you have seen self-thinning in action. The forest is sorting itself out against its carrying capacity, shedding individuals until the remaining trees can be supported by the available resources.
Fisheries and Maximum Sustainable Yield
In fisheries management, carrying capacity takes on a direct economic dimension. The concept of maximum sustainable yield, or MSY, is built on the idea that a fish population held below its natural carrying capacity grows faster because there is less competition for food. By harvesting just the right amount, managers can theoretically skim off the surplus growth indefinitely without depleting the stock. MSY is essentially the point at which the gap between a population’s current size and its carrying capacity produces the fastest growth.
In practice, MSY has been controversial because it oversimplifies real fisheries. Recent modeling work has expanded the classic framework to account for multiple fishing sites and the way fishing effort distributes itself across space. When fishing effort follows a pattern where boats move toward the most productive areas, the effective MSY across the fishery can be depressed below what a simpler model would predict.6Natural Resource Modeling. Maximum sustainable yield as a reference point in the presence of fishing effort that follows an ideal free distribution In other words, the spatial behavior of fishers changes the carrying capacity equation. A stock’s theoretical sustainable harvest depends not just on biology but on where and how intensely people fish. This is one reason why many fisheries scientists now treat MSY as a rough reference point rather than a hard target.
Rangeland Stocking and Overshoot on the Ground
Carrying capacity is not just an academic idea. For pastoralists and ranchers, it is the number that determines how many animals their land can feed without degrading. Get it wrong, and you lose both your pasture and your livestock. A quantitative assessment of rangelands in Harshin district in eastern Ethiopia found that the land’s carrying capacity was about 0.3 tropical livestock units per hectare per year. The actual stocking rate at the time was 5.4 livestock units per hectare, roughly 18 times higher than what the land could sustain.7PubMed Central. Estimating carrying capacity and stocking rates of rangelands in Harshin District, Eastern Somali Region, Ethiopia
That degree of overshoot, running at 18 times carrying capacity, signals severe overgrazing and land degradation. Soil erodes, perennial grasses are replaced by inedible scrub, and the land’s future ability to support livestock drops even further. It mirrors the St. Matthew Island dynamic on a human timescale: exceeding carrying capacity degrades the resource base, which lowers carrying capacity further, which intensifies the overshoot. For communities that depend on pastoralism, carrying capacity assessments are not theoretical exercises. They are the difference between sustainable livelihoods and desertification.
Carrying Capacity in Cities
Ecologists developed the concept for wildlife and ecosystems, but urban planners have borrowed it to ask a parallel question: how many people can a city support given its water, sanitation, transport, and land? The answer can be startling. A framework applied to an Indian hill city found that while its physical land area could theoretically accommodate over 834,000 people per day, the real carrying capacity based on actual infrastructure dropped to about 149,000, and the effective carrying capacity given resource and service delivery constraints was limited to roughly 8,400.8PubMed Central. A framework for assessing urban carrying capacity and its impact on water and sanitation service delivery in a hill city
Those three tiers illustrate something important: carrying capacity depends on which resource you measure. Physically, a city has space. But when you account for water supply, sewage treatment, and road capacity, the number shrinks by orders of magnitude. This layered approach gives city managers concrete guidance on where investment is needed. If the bottleneck is water treatment, expanding housing without expanding water infrastructure will produce an urban version of the reindeer overshoot: overcrowding, service failure, and eventual decline in livability.9Habitat International. Sustainable urban development: A review on urban carrying capacity assessment
Invasive Species and the Shrinking Ceiling
Carrying capacity is not a fixed number stamped on a landscape. It rises and falls as the environment changes, and one of the fastest ways to drive it down is through invasive species. A clear example comes from research on the West Virginia white butterfly (Pieris virginiensis), a native species whose host plants are being displaced by garlic mustard, an invasive weed from Europe. Stochastic population models showed that as garlic mustard spread and host plants disappeared, both the probability of butterfly population persistence and mean population size dropped sharply. At around 50% garlic mustard cover, the probability of population persistence fell to zero.10PubMed. Modelling the impacts of two exotic invasive species on a native butterfly: top-down vs. bottom-up effects
The butterfly’s carrying capacity did not change because the climate shifted or the landscape shrank. It changed because a key resource, the host plant, was eliminated from below. This “bottom-up” force, the loss of habitat and food, was more damaging than the “top-down” force of predation by an invasive parasitoid wasp that also attacks the butterfly. The lesson generalizes: carrying capacity is only as strong as the weakest link in a species’ resource chain, and invasive species frequently target exactly that link.
Disease as a Density-Dependent Brake
When populations approach carrying capacity, crowding creates conditions that favor disease transmission. Research on wildlife host-pathogen dynamics has found that infection rates in animal populations follow a saturating function of host density: as animals become more densely packed, contact rates and transmission increase, but the relationship is not simply linear. At very high densities, the rate at which new infections are acquired plateaus, partly because susceptible individuals become harder to find when most of the population is already infected or immune.11PubMed Central. Host-pathogen time series data in wildlife support a transmission function between density and frequency dependence
This means disease acts as a density-dependent regulator, kicking in harder as populations grow and easing off when they decline. For many wildlife populations, disease outbreaks near carrying capacity are what prevent the kind of dramatic overshoot seen on St. Matthew Island. The reindeer lacked both predators and significant pathogens; most real-world populations face both, which tends to produce a bumpier approach to carrying capacity rather than a clean crash. Seasonal variation in social behavior and susceptibility also modulates transmission, adding another layer of complexity to how carrying capacity is actually enforced in the wild.
Climate Change and Shifting Carrying Capacities
If disease and invasive species can lower carrying capacity on short timescales, climate change does it on longer ones and across much broader areas. Species’ use of resources is spatially complex and temporally dynamic, changing seasonally with fluctuating population sizes and shifting directionally over time as habitats are altered and ecological communities respond to new climatic conditions.12Diversity and Distributions. Conditions for successful range shifts under climate change: The role of species dispersal and landscape configuration When the climate warms, some habitats become unsuitable and others open up. A species’ ability to track these shifts depends heavily on how well it disperses and how the suitable habitat patches are arranged across the landscape.
Research on range shifts under climate change has found that in landscapes with evenly distributed suitable habitats, most species can shift their ranges successfully. But in patchy, heterogeneous landscapes, the outcome depends heavily on habitat configuration and the speed of climate change. Species with high dispersal ability manage to move but often end up at lower population densities after shifting. These nonlinear effects on population size mean that even if a species survives a range shift, the carrying capacity of its new range may be lower than what it left behind.13Diversity and Distributions. Conditions for successful range shifts under climate change: The role of species dispersal and landscape configuration For conservation managers, this means protecting habitat corridors and connectivity is not just about giving species room to move. It is about preserving carrying capacity in the receiving landscape.
Earth’s Carrying Capacity for Humans
The question people eventually arrive at when thinking about carrying capacity is: what is Earth’s carrying capacity for humans? The honest answer is that nobody knows, because it depends on choices as much as on ecology. Human carrying capacity is determined by natural constraints and by human decisions about economics, environment, culture, values, politics, and demography. Unlike nonhuman populations, human choice plays a central role, and simple ecological models of carrying capacity do not capture this.14PubMed. Population growth and earth’s human carrying capacity
Estimates vary enormously depending on assumptions. A 2025 study using a Ricker logistic model fitted to global population data found two strikingly different numbers depending on which phase of human history was used for calibration. The model predicted that global population could peak at 11.7 to 12.4 billion people between 2067 and 2076, but also estimated that the number of people Earth might sustain long-term is closer to 2.5 billion.15Environmental Research Letters. Global human population has surpassed Earth’s sustainable carrying capacity The gap between those numbers reflects the difference between what technology and fossil fuels temporarily allow and what the biosphere can support indefinitely. Humans are, as ecologists put it, the ultimate ecosystem engineers: we modify our environment to expand carrying capacity through agriculture, sanitation, medicine, and energy extraction. But those modifications have costs, including soil depletion, aquifer drawdown, biodiversity loss, and climate change, that may reduce carrying capacity over time.
This is where carrying capacity gets philosophically tricky for humans. A person living at a high-consumption standard in a wealthy country draws down resources at a fundamentally different rate than a person living at subsistence level. Earth’s carrying capacity for 8 billion people living modestly is very different from its carrying capacity for 8 billion people living like the average resident of a high-income country. That is why researchers increasingly frame the question not just in terms of population size but in terms of the sustainability gap between current environmental footprints and associated capacity thresholds.16Ecological Economics. Understanding the complementary linkages between environmental footprints and planetary boundaries in a footprint–boundary environmental sustainability assessment framework The number that matters is not just “how many people” but “how many people at what level of resource use.”
Why r- and K-Selected Species Hit the Ceiling Differently
Not all species interact with carrying capacity the same way. Ecologists have long distinguished between organisms that reproduce rapidly in unstable environments, often called r-selected species, and those that reproduce slowly in stable ones, sometimes called K-selected species. The “K” in K-selection literally refers to carrying capacity: these are species whose populations tend to hover near the environmental limit, investing in fewer, higher-quality offspring that compete well in crowded conditions. Think elephants, whales, and old-growth trees. R-selected species, like insects and annual weeds, tend to boom and bust, colonizing new habitats fast and rarely sitting near carrying capacity for long.
This framework has been criticized and refined over the decades. Research on guppies in Trinidad has shown that life-history differences between populations can be driven by predator-specific mortality patterns, density-dependent regulation, and resource availability all at once, which does not fit neatly into a simple r-vs-K dichotomy.17Wiley Online Library (Ecology). r- AND K-SELECTION REVISITED: THE ROLE OF POPULATION REGULATION IN LIFE-HISTORY EVOLUTION Still, the core intuition holds: species adapted to stable, crowded environments invest in traits that help them compete at carrying capacity, while species adapted to unpredictable environments invest in traits that help them grow fast when resources are temporarily abundant. Reindeer on St. Matthew Island behaved like an r-selected species released from all constraints; moose on Isle Royale, held in check by wolves, behaved more like a K-selected population fluctuating near a regulated ceiling.
When Carrying Capacity Is a Moving Target
Across all of these examples, one pattern stands out: carrying capacity is almost never a fixed number. It shifts with seasons, weather, disease prevalence, predator abundance, invasive species, human management, and long-term climate trends. Even in Gause’s controlled flasks, slight changes in nutrient input altered where the population leveled off. In real ecosystems, the ceiling is always moving.
This matters for how carrying capacity is used in conservation and management. A wildlife agency estimating how many deer a forest can support needs to account for the fact that a wet year and a drought year yield very different answers. A fisheries manager setting catch limits needs to update those limits as ocean temperatures shift and fish distributions change. A rancher in eastern Ethiopia needs to adjust stocking rates not just to this year’s forage production but to its trajectory under ongoing grazing pressure. And a city planner sizing water infrastructure needs to plan for a population that may grow beyond what current resources allow.
The common thread is that carrying capacity is a relationship, not a number. It describes the interaction between a population’s demands and its environment’s ability to meet them. Change either side of that relationship, whether through technology, consumption patterns, habitat loss, or climate, and the carrying capacity changes with it. The reindeer on St. Matthew Island did not just exceed a number. They consumed the resource that produced the number, and the number dropped out from under them.