Material Insights · Understanding the Material World

Ecology: populations, interactions and energy flow: Populations: how numbers grow, level off and swing

All courses / Understanding Science / Module 17: Ecology: populations, interactions and energy flow Part 2 of 6

Populations: how numbers grow, level off and swing

A population changes by four things only: births, deaths, arrivals and departures. This part builds the exponential and logistic models from that count, tests them with tiger and human numbers, and ends with India's population story.

2.1 Growth in proportion to numbers is exponential, and no population can keep it up

Money in a bank grows by interest on interest. A population does the same, because each member adds offspring, and those offspring soon add more. If each individual contributes a net rate r per year (births minus deaths per head), growth each year is r times the number present. That gives N = N0 x e^(rt), the exponential curve. The doubling time is about 70 divided by the percentage rate: at 7 per cent a year, 10 years; at 2 per cent, 35 years.

Tiger counts show it. The national estimate was 1,411 in 2006 and 2,967 in 2018. The ratio is 2.10 over 12 years, so the yearly factor is 2.10 to the power one-twelfth, about 1.064, a rate of 6.4 per cent a year (calculated). The doubling time is 70 / 6.4, or about 11 years (calculated). The National Tiger Conservation Authority itself reports growth of about 6 per cent a year, and 6.1 per cent a year in its 2022 report.

The curve cannot last. A bacterium doubling every 20 minutes starts at about a trillionth of a gram. Earth weighs about 6 x 10^27 grams, a ratio of 6 x 10^39, near 2^132 (calculated). At 20 minutes per doubling, 132 doublings take 2,640 minutes, about 44 hours (calculated). Food, space or waste stops it within hours.

Data Worked estimate: yearly growth of India's tiger count
StepWorkingResult
Counts2006: 1,411 tigers. 2018: 2,967 tigersRatio 2,967 / 1,411 = 2.10 (calculated)
Years between2018 minus 200612 years
Yearly factor2.10 to the power 1/12about 1.064 per year, so 6.4 per cent a year (calculated)
Doubling time70 / 6.4 per cent a yearabout 11 years (calculated)
Check: next four yearsIf 6.4 per cent a year continued, 2022 would be 2,967 x 1.064^4 = 2,967 x 1.28 tigersabout 3,800 tigers (calculated); the 2022 average was 3,682, a rate of (3,682 / 2,967)^(1/4) = 1.055, about 5.5 per cent a year (calculated)

Survey methods improved between counts, so the rates are approximate.

Source: National Tiger Conservation Authority, Counting tigers in India (monitoring page), https://ntca.gov.in/monitoring (2018: 2,967 tigers, standard error range 2,603 to 3,346; 2,461 photographed; 26,838 camera locations; 381,400 square km of forest surveyed; growth of about 6 per cent a year from 2006 to 2018)

2.2 Crowding slows growth to a ceiling, the carrying capacity, and the curve becomes an S

Add more tigers to a forest and each has fewer deer. Births per head fall and deaths per head rise. So the net rate r is not constant: it shrinks as numbers rise and reaches zero when the population hits what the habitat can support. That level is the carrying capacity, K. A simple model multiplies the exponential growth by (1 minus N/K), which is near 1 when the population is small and near 0 when it is near K. This is logistic growth, and it traces an S-shaped curve.

Gause saw it in his test tubes: Paramecium numbers rose, bent over and levelled off as food and waste set the limit. A numerical walk shows the shape. Take K = 1,000 and a population of 20, with a potential growth of 50 per cent per generation, and let each generation grow by 0.5 x N x (1 minus N/K).

The fastest growth is not at the start or the end, but at half of K, where there are many parents and still plenty of room. Managers use this: a fishery or a forest kept near K/2 gives the largest harvest it can sustain, because that is where it adds most each year. Growth limits come in two kinds. Density-dependent factors, such as competition, disease and predation, bite harder as numbers rise. Density-independent factors, such as a flood, a frost or a cyclone, kill a share of the population whatever its density.

Data Worked estimate: logistic growth with K = 1,000 animals, starting at 20, growth rate 0.5 per generation
GenerationPopulation NIncrease in next generation = 0.5 x N x (1 minus N/K)
02010
49643
619980
8380118
9497125 (the largest)
10622118
1283668
169867
209990

Each row is calculated by repeating N(next) = N + 0.5 x N x (1 minus N/1,000), rounded to whole animals. Growth peaks near N = K/2 = 500, then falls to zero at K (calculated).

Source: Hewitt, Lyons, Suchocki and Yeh, Conceptual Integrated Science, 2nd edition (Pearson, 2013), ch. 21, 'Ecology' (Math Connection: exponential and logistic growth)
  • Hewitt, Lyons, Suchocki and Yeh, Conceptual Integrated Science, 2nd edition (Pearson, 2013) · ch. 21, 'Ecology' (population ecology; Math Connection)
  • Standard histories of ecology (Encyclopaedia Britannica and the original papers named in the text) · Haeckel (1866), Tansley (1935), Liebig (1840), Shelford (1913), Connell on intermediate disturbance (Science, 1978), Carson (1962), Elton (1958), Williamson and Fitter on the tens rule (1996), the Hudson's Bay Company pelt records, Surtsey (formed 1963; UNESCO World Heritage listing https://whc.unesco.org/en/list/1267, which could not be opened in this check), and Rockstrom and colleagues (Nature, 2009), Steffen and colleagues (Science, 2015), Richardson and colleagues (Science Advances, 2023): standard dates, not re-fetched in this check

2.3 Species trade the number of offspring against the care of each, and survival curves show the bargain

A hilsa swimming up a river lays hundreds of thousands of eggs, and almost every one dies. An Indian rhinoceros carries one calf for about 16 months, has a calf only every three years or so, and guards it for years. Both strategies work, because each fits a different world. This is a life history: when an organism starts to breed, how many offspring it makes, how big they are and how long it lives. Energy spent on one thing cannot be spent on another, so a trade-off is forced.

Species near the many-small-offspring end are called r-selected: they suit unstable places where numbers crash and rebuild. Species near the few-large-offspring end are K-selected: they suit crowded places near the ceiling, where competing for resources counts most. Most lie between, and ecologists now treat the r and K labels as a rough sketch rather than two fixed types.

A survivorship curve plots how many of a group born together remain alive at each age. Type I species, such as humans and rhinos, lose few young and die mostly in old age. Type II species, such as hydra and many lizards, face a roughly constant risk at every age. Type III species, such as hilsa, oysters and most trees, lose almost all offspring early, and only a few survive to breed. A tiger lies toward Type I with high cub mortality: litters of two to four cubs, often a female breeding every two years or so after reaching maturity at three to four.

  • Hewitt, Lyons, Suchocki and Yeh, Conceptual Integrated Science, 2nd edition (Pearson, 2013) · ch. 21, 'Ecology' (population ecology: life history and survivorship)

2.4 Human numbers grew exponentially, then slowed, and India's story shows the pattern

Around 1800 there were 1 billion people; the UN estimates nearly 8.2 billion in mid-2024. The UN's World Population Prospects 2024 projects a peak near 10.3 billion in the mid-2080s, then a gentle decline. The pattern is the demographic transition. First deaths fall, as sanitation, vaccines and more food arrive. Births stay high for a while, so numbers surge. Then births fall too, as children survive, women become more educated, and families choose to have fewer.

India is partway through. Its population was about 238 million in the 1901 Census, 361 million in 1951 and about 1.21 billion in 2011. The UN puts it at about 1.45 billion in 2024, and its World Population Prospects 2024 projects a peak of about 1.7 billion in the early 2060s, followed by a fall of about 12 per cent by 2100. Fertility has fallen to near 2 children per woman, close to the 2.1 needed to replace the parents, yet numbers still rise because India has a large share of young people. An age pyramid, with a wide base of young and a narrow top of old, carries built-in momentum: even at replacement fertility, many young adults mean many births for a generation.

Pressure on nature is numbers times use per person. An ecological footprint expresses use as the land and sea needed to supply a person's food, fibre, energy and waste. It differs several times between rich and poor, so population and consumption both matter. When humanity as a whole uses more each year than the Earth can regrow in that year, it is in ecological overshoot, drawing down stocks such as soil, groundwater and fish.

Data Worked estimate: India's average yearly growth rate
PeriodPopulation start and endYearly rate from ln(end / start) / yearsDoubling time (70 / rate)
1901 to 1951238.4 million to 361.1 millionln(1.515) / 50 years = 0.83 per cent a year (calculated)about 84 years (calculated)
1951 to 2011361.1 million to 1,210.9 millionln(3.353) / 60 years = 2.0 per cent a year (calculated)about 35 years (calculated)
2011 to 20241,210.9 million to 1,450 millionln(1.197) / 13 years = 1.4 per cent a year (calculated)about 50 years (calculated)

The 1901, 1951 and 2011 figures are census counts; the 2024 figure is a UN WPP 2024 estimate for mid-year, so the last rate is approximate.

Source: United Nations, Department of Economic and Social Affairs, Population Division, World Population Prospects 2024: Summary of Results (2024) and WPP 2024 Key Messages, https://desapublications.un.org/file/20622/download (India 'will likely decline by 12 per cent after reaching its peak in the early 2060s at about 1.7 billion' and is expected to remain the world's largest population throughout the century) and https://population.un.org/WPP/Publications/Files/WPP2024_Key-Messages.pdf (world peak of around 10.3 billion in the mid-2080s, up from 8.2 billion in 2024); India's 2024 estimate of about 1.45 billion is the WPP 2024 figure

See it in the table

Eg Eggs

Check your understanding

A population grows at 3.5 per cent a year. About how long does it take to double?

The rule of 70 gives 70 / 3.5 = 20 years.

In the logistic model, when is a population's absolute growth per generation greatest?

Growth is N x r x (1 minus N/K). That product is largest when N is K/2, because there are many parents and still much room.

India's population growth rate has fallen, yet the population still rises. Why?

Growth is the rate times the number. With fertility near replacement but a young age structure, births stay high for a generation.
← Part 1: Life in its surroundings: levels, limits, niches and biomesMark read · Part 3: Interactions and change: predators, partners, webs and succession →