Biology › Section R · The biosphere, populations, and communities › Topic R.6
Population growth, limits, and regulation
Any population with food to spare can multiply astonishingly fast, yet no population grows for ever. Two simple equations capture both facts, and the factors that slow growth down explain why real numbers rise, level off, overshoot and crash.
Why growth cannot last
In 1798 Thomas Malthus argued that a population with unlimited resources multiplies faster and faster until supplies run short. Darwin built on that idea: when more offspring are born than can survive, the best suited leave the most descendants. Ecologists capture this in two classic models, exponential and logistic growth, then ask which real factors hold populations in check.
Births, deaths, and the rate r
Treat a population as closed, with no immigration or emigration. Over a time interval, its change in size is births minus deaths: \(\frac{\Delta N}{\Delta t} = B - D\). If each individual produces \(b\) offspring per year and has a chance \(d\) of dying per year, then \(B = bN\) and \(D = dN\). Shrinking the interval to an instant gives
where \(r = b - d\) is the intrinsic rate of increase. When \(r > 0\) the population grows, when \(r < 0\) it shrinks, and when \(r = 0\) it holds steady, a state called zero population growth.
Exponential growth and the J-curve
In \(\frac{dN}{dt} = rN\), the number added per unit time is proportional to the number already present. A constant \(r\) therefore produces an ever-steeper rise: twice as many individuals add twice as many newcomers. Plotted against time, \(N\) traces a J-shaped curve. Each species has a maximum possible \(r\), reached only under ideal conditions, called its biotic potential, \(r_{max}\). A bacterium that divides every half-hour has a vastly higher \(r_{max}\) than an elephant. Exponential growth appears when a few individuals meet abundant resources, such as bacteria in fresh broth.
Worked example: exponential growth in numbers
1. A flask starts with 500 bacteria that divide every 30 minutes, with nutrients to spare and no deaths. In 4 hours there are 8 doublings, so \(N = 500 \times 2^8 = 500 \times 256 = 128{,}000\). 2. Compare the increments: the first half-hour adds 500 cells, but the last adds 64,000, because the population doubled from 64,000 to 128,000. 3. Now a vole population of 2,000 with \(b = 0.6\) and \(d = 0.4\) per individual per year. Then \(r = 0.2\) per year and \(\frac{dN}{dt} = 0.2 \times 2{,}000 = 400\) voles per year. 4. At 4,000 voles the same \(r\) gives 800 per year. Check: \(r\) never changed, yet the growth rate doubled with \(N\).
Carrying capacity and the logistic model
Resources are finite, so ecologists add a brake. The carrying capacity, \(K\), is the largest population a particular environment can support over time. The logistic equation multiplies exponential growth by the fraction of that capacity still unused:
Here \(K - N\) is how many more individuals the habitat can hold, and dividing by \(K\) turns it into a fraction. The same letters name the two life-history extremes of the previous topic. K-selected species, such as elephants and oaks, live in stable habitats where numbers stay near K, so they mature late and invest in a few large offspring. r-selected species, such as dandelions, suit unpredictable habitats where numbers are often far below K, so fast breeding pays. These are ends of a continuum, and newer models explain life histories mainly through age-specific mortality.
Reading the S-shaped curve
The braking fraction changes as \(N\) changes. When \(N\) is small, \(\frac{K - N}{K}\) is close to 1 and growth is almost exponential. As \(N\) climbs, the fraction shrinks, so each individual adds less. At \(N = K\) the fraction is zero and growth stops. If \(N\) rises above \(K\), the fraction turns negative and the population declines back towards \(K\). The result is an S-shaped (sigmoid) curve with three phases: a slow start, a steep middle, and a plateau. Because \(N(K - N)\) is largest when \(N = \frac{K}{2}\), the population adds individuals fastest at half the carrying capacity, not near it.
Worked example: logistic growth rates
A deer herd has \(r_{max} = 0.5\) per year and \(K = 1{,}000\). 1. At \(N = 100\): \(0.5 \times 100 \times \frac{900}{1{,}000} = 45\) deer per year. 2. At \(N = 500\): \(0.5 \times 500 \times \frac{500}{1{,}000} = 125\) deer per year, the maximum, equal to \(\frac{r_{max}K}{4}\). 3. At \(N = 900\): \(0.5 \times 900 \times \frac{100}{1{,}000} = 45\) deer per year, the same as at 100. 4. At \(N = 1{,}200\): \(0.5 \times 1{,}200 \times \frac{-200}{1{,}000} = -120\) deer per year. Check: the rate is symmetric about 500, peaks there, and turns negative above \(K\).
Real populations: competition, overshoot, and crash
The logistic model assumes every individual has equal access to resources. In reality, members of a species compete for food, light, nest sites and mates, and this intraspecific competition intensifies as numbers rise. Waste build-up can lower \(K\) as well. A Paramecium culture with a fixed food supply traces a neat S-curve, but wild populations often overshoot \(K\) and oscillate around it. In 1944, 29 reindeer were released on St. Matthew Island in the Bering Sea, which had no predators. By 1963 there were about 6,000. The herd had overgrazed its lichen food, and after a severe winter only 42 animals remained in 1966.
What sets the limits
The logistic model also assumes \(K\) is constant, but winter supports fewer animals than summer, and a fire or eruption can reset a habitat. Species sharing a habitat also compete for the same resources, called interspecific competition. Limiting factors fall into two groups. Density-dependent factors, mostly biotic, act more strongly as density rises: competition, predation, parasites, disease and waste. Density-independent factors, mostly abiotic, kill the same proportion whatever the density: storms, floods, fires, cold snaps and pollution. The two interact. After a harsh winter, a large surviving herd rebounds faster than a small one, simply because more animals remain to breed.
Worked example: classifying a limiting factor
An ecologist plants oak seedlings in plots at three densities and records the percentage that die. 1. Fungal infection killed 5% of seedlings at 10 per square metre, 20% at 40 per square metre, and 45% at 80 per square metre. 2. The death rate rises with density, so the fungus acts as a density-dependent factor, plausibly because spores pass more easily between close neighbours. 3. A late frost killed 30% of seedlings in every plot. 4. The proportion killed does not change with density, so the frost is density-independent. 5. Check: classify by the proportion affected, not the number killed, which is always larger in a bigger population.
Common mistakes to avoid
First, \(r\) is not the growth rate itself: \(r\) is a per-individual rate, and \(\frac{dN}{dt} = rN\) is the number added per unit time. Second, exponential growth does not add the same number each period; it adds the same proportion, so the increments keep growing. Third, logistic growth is fastest at \(\frac{K}{2}\), not just below \(K\). Fourth, \(K\) is not fixed: it changes with seasons and disturbances. Fifth, judge density dependence by the proportion of the population affected, not by the head count.
Key takeaways
With no migration, a population changes by births minus deaths, giving \(\frac{dN}{dt} = rN\), where \(r = b - d\). Constant \(r\) yields exponential, J-shaped growth up to the biotic potential \(r_{max}\). Finite resources impose a carrying capacity, and the logistic equation \(\frac{dN}{dt} = r_{max}N\frac{K - N}{K}\) produces an S-shaped curve: nearly exponential when small, fastest at \(\frac{K}{2}\), zero at \(K\), and negative above it. Real populations overshoot and fluctuate as \(K\) shifts. Density-dependent factors strengthen with crowding; density-independent factors do not; and the two interact.
Flashcards
Flashcard
Growth by a constant proportion per unit time: dN/dt = rN
Test your understanding
Question 1 of 5
A closed population of 1,500 has a per capita birth rate of 0.30 per year and a per capita death rate of 0.10 per year. At what rate is it growing?
Question 2 of 5
A population follows the logistic model with r max = 0.4 per year and K = 2,000. How fast is it growing when N = 1,500?
Question 3 of 5
In the logistic model, at what population size does the population add individuals fastest?
Question 4 of 5
Which of these acts as a density-independent factor?
Question 5 of 5
According to the logistic model, what happens if a population rises above its carrying capacity?
