Biodiversity is the variety of life in a place. It is measured at three levels: species, genetic, and ecosystem. At the community level, two measures count: species richness (how many species), and the index of diversity (how those species are distributed across individuals).
Biodiversity is measured at three levels and quantified by richness and the diversity index.
Biodiversity is examined at three levels. Species diversity is the variety of species in a community. Genetic diversity is the range of alleles in a single species' gene pool. Ecosystem diversity is the variety of habitats across a region. The three are independent: a region can have high species diversity and low genetic diversity within those species, or the reverse.
Species richness is the number of different species in a community.
Write
community, notpopulation. Species richness is across a multi-species community; a population is one species. "Species richness in a population" is rejected on every mark scheme that tests the definition.
D = N(N − 1) / Σn(n − 1). N is the total number of individuals of all species combined. n is the number of individuals of each individual species. A higher D indicates a more diverse community. The interpretation: D reflects the probability that two randomly chosen individuals from the sample belong to different species. The formula structure does not change when biomass replaces individual counts — n becomes the biomass of each species and N the total biomass.
Nis the total number of individuals of all species, not the number of species. Inverting N and the species count flips the formula. On 2017 P3 Q04.1 the inversion was the dominant failure mode.
Pitfall — The low-diversity paradox
A low index of diversity means one or a few species dominate — not that all species are present in small numbers.
A low D value is produced when one or a few species account for a disproportionately large share of all individuals; the dominance inflates Σn(n − 1) and pulls D down. Many different species each present in small numbers would produce a high D, the opposite of what students often write. The mark-credited phrase is "dominated by one or a few species in very large numbers." On the 2023 P3 Q05.2 interpretation question, only 7% scored both marks.
Sampling methods estimate species abundance without counting every individual.
Counting every organism in a community is impractical, so sampling techniques estimate species numbers and abundance from a subset of the habitat. The appropriate technique depends on the organisms and the habitat: quadrats and transects for plants and sessile or slow-moving organisms, mark-release-recapture for mobile animals.
Quadrats versus belt transects.
| Method | When to use | How abundance is measured | Standard caveat |
|---|---|---|---|
| Quadrats | Uniform habitat with sessile or slow-moving organisms | Percentage cover or frequency of occurrence | Placed at random coordinates; multiple samples until the running mean stabilises |
| Belt transect | Gradient or boundary habitat (shore, woodland edge, altitude) | Quadrats placed at regular intervals along a line across the gradient | Systematic placement, not random; captures change with the gradient |
Choose the method that matches the habitat. Quadrats placed in a gradient, or a transect run through a uniform habitat, are wrong-method answers. "Random coordinates" does not rescue the wrong choice. On 2017 P3 Q04.3 the woodland question rejected both quadrats and random coordinates because the correct method was a directional transect.
Population size ≈ (first sample × second sample) / marked individuals in the second sample. A first sample is captured, individually marked without harm, and released. After enough time for marked individuals to mix randomly with the unmarked population, a second sample is captured. The marked proportion in the second sample reflects the marked proportion in the whole population.
Mark-release-recapture relies on four assumptions about the population.
The mark-release-recapture estimate is valid only if four assumptions hold. Each is examined as a separate mark point in mark schemes, and violating any one of them breaks the calculation.
- The population is closed between the two captures — no significant immigration, emigration, births, or deaths.
- Marks do not harm the animals or alter their behaviour — and do not change their visibility to predators, do not wash off, and do not fade between captures.
- Marked individuals mix freely and randomly with the unmarked population before the second sample is taken.
- Every individual has an equal probability of being captured in both samples.
Evaluate-validity questions in this topic credit specific design flaws: a single sampling site, a single time of year, no baseline measurement before an intervention, environmental conditions that varied between treatment years. Rote criticisms of statistics or sample size are not credited.
For an evaluate-validity question, name a specific design flaw. "No statistical test was done", "sample size was too small", and "only X years of data" are explicitly rejected on mark schemes for this topic. The credited flaws are single-site, single-time-of-year, missing baseline, or between-year variation in treatment.
Statistical interpretation requires difference, chance, and representative as fixed vocabulary.
When biodiversity measures are compared between communities or over time, a statistical test asks whether the observed difference is likely to be due to chance. The credited interpretation language is fixed across mark schemes: it names the probability of getting the difference by chance, paired with the direction of change.
A full-credit P value answer names three things. (1) Direction of change in the data (which community is more diverse, or how the index changed over time). (2) Significant or not significant. (3) "The probability of getting this difference by chance is less than 5%" (significant) or "greater than 5%" (not significant). All three are required. The word difference is mandatory; "results are due to chance" is rejected because it conflates the data with the gap between groups. A bare P value with no probability language is also rejected.
Write
probability of getting this difference by chance.Differenceis mandatory. "Results are due to chance" is rejected. A bare P value with no probability language is rejected. The full sentence is the mark, not the number on its own.
Sample-adequacy answers also have fixed phrasing. The credited word is representative or accurate, not "reliable" — reliable describes consistency of measurement, not whether the sample reflects the population. If a specific sample-size number is given, it must be at least 10; "several" or "a few" is not credited. The alternative criterion is a stable running mean that no longer changes appreciably as more samples are added.
Write
representativeoraccurate, notreliable. Reliable describes measurement consistency, not whether the sample reflects the population. The two words are not interchangeable in this topic.
Farming reduces biodiversity; conservation measures pair biodiversity features with economic outcomes.
Intensive farming reduces biodiversity through several mechanisms operating at different levels. Monoculture — growing a single crop across a large area — eliminates non-crop habitat. Selective breeding of crops and livestock narrows the allelic diversity within those species over generations. Pesticide and herbicide use kills non-target invertebrates (including pollinators and natural predators of pests) and removes the plant food webs that support them. Removal of hedgerows to enlarge field units eliminates the semi-natural habitat strips that shelter farmland species and connect habitat patches.
Maintain or restore hedgerows — shelter, food, nesting sites, and corridors for farmland wildlife. Intercropping — alternating crops in the same field restores structural diversity. Reduced or targeted pesticide and herbicide use — allows non-crop species to persist. Preserve wetlands and other high-biodiversity habitats — maintains specialist species that cannot survive in drained farmland. These are modifications to agricultural practice, not alternatives to productive agriculture.
For Suggest or Explain questions about the farmer's advantage of biodiversity, the credited answer pairs a biodiversity feature with an economic outcome. Pollinators raise yield. Predators of pests reduce crop damage and pesticide cost. Hedgerows shelter both, which raises yield and reduces pest-control spending. "More biodiversity" or "better habitat" alone earns nothing because the economic word is the mark. Suggest / Explain in this topic uses 1+1 marking: every suggestion needs an explanation naming the experimental or ecological outcome.
In a farmer-advantage answer, name the economic outcome. Biodiversity feature → economic outcome. "Pollinators raise yield." "Predators reduce pesticide cost." The economic word — yield, cost, damage — is the mark. "More biodiversity" on its own scores zero.
Key terms
- species richness
- index of diversity
- community
- chance
- probability
- representative
- quadrat
- random sampling
- farming
- yield