Measuring Biodiversity

Measuring biodiversity accurately is essential for monitoring ecosystems, assessing the impact of human activities, and making informed conservation decisions. In this lesson, we examine the practical techniques used to sample organisms and the mathematical index used to quantify species diversity.

---

Why Measure Biodiversity?

Quantifying biodiversity allows scientists and conservationists to:

• Monitor changes in ecosystems over time (e.g., is biodiversity declining in a particular habitat?)
• Compare habitats — which areas are most biodiverse and therefore priorities for conservation?
• Assess the impact of human activities such as pollution, deforestation, or urbanisation
• Evaluate conservation efforts — are management strategies (e.g., habitat restoration, reintroductions) working?
• Provide baseline data for future studies and policy decisions

---

Sampling Techniques

It is usually impossible to count every individual of every species in an area (a complete census). Instead, biologists take samples — collecting data from representative portions of the habitat and using these to estimate the overall biodiversity.

Key Principles of Sampling

1. Random sampling — sampling points must be chosen randomly (e.g., using random number generators to produce coordinates) to avoid bias
2. Representative sampling — the sample must be large enough and spread across the habitat to be representative of the whole area
3. Repeatability — methods must be standardised so that results can be compared between different times, places, and investigators

Sampling Equipment and Methods

Method	Description	Best For	Limitations
Quadrat	A square frame (typically 0.25 m² or 1 m²) placed on the ground; organisms within are identified and counted	Sessile (non-moving) organisms: plants, lichens, barnacles	Not suitable for mobile animals; size of quadrat must be appropriate for organism size
Point quadrat	A frame with pins that are lowered vertically; each species touched by a pin is recorded	Estimating percentage cover of plant species in grassland	Time-consuming; requires skill in species identification
Transect (belt)	Quadrats are placed at regular intervals along a line (tape measure) across a habitat	Studying how species distribution changes along an environmental gradient (e.g., from shore to inland)	Labour-intensive; may miss patchy distributions
Transect (line)	A tape measure is laid across the habitat; species touching the line are recorded at regular intervals	Quick survey of species distribution across a gradient	Less detailed than belt transect
Sweep net	A strong net swept through vegetation to catch invertebrates	Flying or vegetation-dwelling invertebrates	Catch depends on sweeping technique and speed; not quantitative
Pitfall trap	A container sunk into the ground with its rim level with the surface; organisms fall in and are trapped	Ground-dwelling invertebrates (beetles, spiders)	Catches depend on activity of organisms, not abundance; predation within trap
Kick sampling	The riverbed is disturbed by kicking while a net is held downstream to catch dislodged organisms	Freshwater invertebrates in streams and rivers	Semi-quantitative; depends on effort and technique
Light trap	A bright light source attracts nocturnal flying insects into a collecting chamber	Nocturnal flying insects (moths)	Biased towards species attracted to light; weather affects catches
Pooter	A small suction device used to collect individual small invertebrates by mouth	Small, individual invertebrates that need careful collection	Very time-consuming; only suitable for small numbers
Tullgren funnel	A sample of soil or leaf litter is placed on a mesh above a funnel, with a heat/light source above; organisms move away from heat and fall into a collecting jar	Soil and leaf-litter invertebrates	Only catches organisms that respond to heat/light gradient

Exam Tip: You must be able to describe and evaluate specific sampling methods. Always discuss advantages and limitations, and explain why random sampling is essential to avoid bias.

---

Random Sampling with Quadrats — Core Practical

Required Practical: Investigating the Distribution and Abundance of Organisms

This is a required practical for Edexcel A-Level Biology. You must be able to describe and explain this procedure in detail.

Method

1. Define the study area — mark out the area to be sampled (e.g., a field, woodland floor)
2. Generate random coordinates — use a random number generator to produce pairs of coordinates within the study area. These determine where each quadrat is placed.
3. Place quadrats — lay the quadrat at each random coordinate
4. Record data — within each quadrat, identify all species present and record either:
   • Frequency — the number of quadrats in which a species appears (out of total quadrats)
   • Percentage cover — the estimated percentage of the quadrat area covered by each species
   • Density — the number of individuals of each species per unit area
5. Repeat — use a sufficiently large number of quadrats (typically 10–30 or more) to ensure the data is representative
6. Calculate mean values — calculate the mean density, frequency, or percentage cover for each species

Why Random Sampling?

Random sampling is essential to ensure the data is unbiased. If the investigator chooses where to place quadrats, they might (consciously or unconsciously) select areas that are particularly species-rich or poor, giving a misleading picture of the habitat.

Estimating Population Size

To estimate the total population of a species in an area using quadrat data:

$\text{Estimated population} = \text{mean number per quadrat} \times \frac{\text{total area}}{\text{quadrat area}}$Estimated population=mean number per quadrat×quadrat areatotal area​

For example, if the mean number of daisy plants per 0.25 m² quadrat is 6, and the total field area is 500 m²:

$\text{Estimated population} = 6 \times \frac{500}{0.25} = 6 \times 2000 = 12{,}000 \text{ daisies}$Estimated population=6×0.25500​=6×2000=12,000 daisies

---

The Mark-Release-Recapture Method

For mobile animals, quadrats are impractical. The mark-release-recapture (Lincoln Index) method is used instead.

Method

1. Capture a sample of animals from the population and count them — this is M (number marked)
2. Mark each captured individual in a way that does not harm them or make them more visible to predators (e.g., a small dot of non-toxic paint on a beetle's shell)
3. Release the marked animals back into the habitat
4. Allow time for the marked individuals to mix randomly with the rest of the population (usually at least 24 hours)
5. Recapture a second sample and count the total caught (n) and the number that are marked (m)
6. Estimate the population size using the Lincoln Index:

$N = \frac{M \times n}{m}$N=mM×n​

Where:

• N = estimated total population size
• M = number of individuals marked and released in the first sample
• n = total number of individuals captured in the second sample
• m = number of marked individuals recaptured in the second sample

Worked Example

A biologist captures 40 woodlice from a garden, marks them with a dot of paint, and releases them. The next day, 50 woodlice are captured, of which 8 are marked.

$N = \frac{40 \times 50}{8} = \frac{2000}{8} = 250$N=840×50​=82000​=250

The estimated population is 250 woodlice.

Assumptions of Mark-Release-Recapture

The method is only valid if several assumptions are met:

Assumption	Explanation	If Violated
No births, deaths, immigration or emigration between sampling events	The population size must remain constant	Estimate will be inaccurate
Marked individuals mix randomly with the rest of the population	Ensures the recapture sample is representative	If marked individuals cluster together, the method is unreliable
Marks do not affect survival	Marking must not make individuals more visible to predators or cause harm	Marked individuals may die at higher rates, inflating the population estimate
Marks do not rub off or fade	All marked individuals must still be identifiable at recapture	Lost marks reduce m, inflating the estimate
Equal catchability	Marked and unmarked individuals must have an equal probability of being caught	If marked individuals become "trap-shy" or "trap-happy," the estimate is biased

Exam Tip: Mark-release-recapture questions are very common. Always state the formula AND list the assumptions. If asked to evaluate the method, explain what happens if each assumption is violated.

---

Simpson's Index of Diversity (D)

Why Use a Diversity Index?

Species richness alone does not fully capture biodiversity because it ignores relative abundance (evenness). A diversity index combines both richness and evenness into a single numerical value.

The Formula

The Simpson's Index of Diversity is calculated as:

$D = 1 - \frac{\sum n(n-1)}{N(N-1)}$D=1−N(N−1)∑n(n−1)​

Where:

• D = Simpson's Index of Diversity
• n = the number of individuals of each species
• N = the total number of all individuals of all species
• Σ = sum of (calculated for each species)

Interpreting Simpson's Index

• D ranges from 0 to 1
• D = 0 means no diversity (only one species present, or one species completely dominates)
• D = 1 means infinite diversity (every individual belongs to a different species) — in practice, D approaches 1 but never reaches it
• Higher D = higher biodiversity

Worked Example

A pond contains three species of invertebrate:

Species	Number of Individuals (n)	n(n-1)
Mayfly larva	15	15 × 14 = 210
Freshwater shrimp	10	10 × 9 = 90
Caddisfly larva	5	5 × 4 = 20
Totals	N = 30	Σn(n-1) = 320

$D = 1 - \frac{320}{30 \times 29} = 1 - \frac{320}{870} = 1 - 0.368 = 0.632$D=1−30×29320​=1−870320​=1−0.368=0.632

The Simpson's Index of Diversity is 0.632, indicating moderate diversity.

Comparing Two Habitats

By calculating D for different habitats, we can objectively compare their biodiversity. For example:

• Unpolluted stream: D = 0.85 (high diversity)
• Polluted stream: D = 0.30 (low diversity — dominated by pollution-tolerant species)

This comparison provides quantitative evidence that pollution reduces biodiversity.

Exam Tip: You MUST be able to calculate Simpson's Index of Diversity from raw data. Practise the calculation with different datasets. Show all your working in the exam — even if your final answer is wrong, you can earn method marks.

---

Using Sampling Data to Assess Habitat Quality

Biodiversity measurements can indicate habitat quality:

Indicator	High Biodiversity	Low Biodiversity
Water quality	Clean, unpolluted water supports many species of invertebrate (mayfly, stonefly, caddisfly larvae)	Polluted water is dominated by few tolerant species (bloodworm, tubifex, rat-tailed maggot)
Air quality	Clean air supports many lichen species	Polluted air supports only a few tolerant lichens (or none)
Soil health	Healthy soil with high organic matter supports diverse invertebrate communities	Degraded soil with heavy metal contamination supports few species

Indicator species are organisms that are particularly sensitive to environmental changes and whose presence or absence indicates specific conditions. For example:

• Stonefly larvae indicate clean, well-oxygenated water (they are intolerant of pollution)
• Tubifex worms indicate polluted, low-oxygen water (they tolerate these conditions)
• Lichens indicate air quality — different species tolerate different levels of sulfur dioxide pollution

---

Sources of Error and Improving Reliability

Source of Error	Solution
Non-random sampling	Use random number generator for coordinates
Too few samples	Increase the number of quadrats or recaptures
Identification errors	Use identification keys; train investigators
Seasonal variation	Sample at the same time of year for comparisons
Observer bias	Use standardised protocols; multiple observers
Disturbing the habitat	Use minimally invasive techniques

---

Summary

Key Concept	Detail
Sampling	Taking representative portions of a habitat to estimate biodiversity
Quadrats	For sessile organisms; placed randomly
Mark-release-recapture	For mobile animals; Lincoln Index: N = Mn/m
Simpson's Index (D)	D = 1 − Σn(n−1) / N(N−1); ranges 0 to 1; higher = more diverse
Species richness	Number of species only
Species diversity	Richness + evenness
Indicator species	Organisms whose presence indicates environmental conditions
Required practical	Random sampling with quadrats to investigate distribution and abundance

Exam Tip: This lesson contains one of the most important required practicals. Be prepared to describe the full method, explain why random sampling is essential, calculate Simpson's Index, and evaluate the limitations of each technique.

---

A-Level Deep Dive: Measuring Biodiversity

Spec mapping

The Edexcel 9BI0 specification places measurement of biodiversity within Topic 4: Biodiversity and Natural Resources, with synoptic overlap into the previous lesson on biodiversity and species richness (which defined biodiversity — this lesson operationalises the species level), the next lesson on natural selection and evolution (selection changes the abundance distribution Simpson's index summarises), Topic 5: On the Wild Side (succession and the biodiversity–stability relationship use the same quadrat / transect / Simpson tooling), Topic 7: Run for your Life (representative sampling reappears in the cardiac-output and respirometer practicals), and Paper 3: General and Practical Principles in Biology (sampling, statistical analysis and methodology evaluation are tested across topics). The relevant statements concern: random and systematic sampling using quadrats, point quadrats and transects; mark–release–recapture (Lincoln Index) for mobile organisms; calculating Simpson's $D = 1 - \sum (n/N)^2$D=1−∑(n/N)2; comparing biodiversity between habitats; and evaluating sources of error (refer to the official Pearson Edexcel 9BI0 specification document for exact wording).

---

Worked example with full mark scheme

Question (8 marks):

A field biologist surveys a $400\,\text{m}^2$400m2 grassland for invertebrates using $0.25\,\text{m}^2$0.25m2 random quadrats. The aggregated counts are:

Species	Number of individuals (n)
Common ground beetle	20
Yellow meadow ant (workers)	15
Garden snail	5
Total	N = 40

(a) Calculate Simpson's index of diversity using the proportional form $D = 1 - \sum (n/N)^2$D=1−∑(n/N)2, showing working to three significant figures. (3)

(b) The same area is resurveyed five years later after intensive grazing. Richness drops to two species and Simpson's $D$D falls to $0.18$0.18. Interpret these results and explain why $D$D has fallen further than the richness loss alone would suggest. (3)

(c) Evaluate one design weakness of the original survey and one specific improvement. (2)

Solution with mark scheme:

(a) $\sum (n/N)^2 = (20/40)^2 + (15/40)^2 + (5/40)^2 = 0.250 + 0.141 + 0.0156 = 0.407$∑(n/N)2=(20/40)2+(15/40)2+(5/40)2=0.250+0.141+0.0156=0.407, so $D = 1 - 0.407 = 0.593$D=1−0.407=0.593.

M1 (AO2.1) — proportions correct ($0.500$0.500, $0.375$0.375, $0.125$0.125).

M1 (AO2.1) — squares and sums to $0.407$0.407.

A1 (AO2.1) — final answer $D = 0.593$D=0.593 to three significant figures.

(b) M1 (AO3.1a) — richness has fallen from three species to two, and Simpson's $D$D from $0.593$0.593 to $0.18$0.18.

M1 (AO3.1a) — a fall in $D$D of $\sim 0.41$∼0.41 is much larger than the loss of one species from three would suggest; it signals a combined loss of richness and evenness. $D = 0.18$D=0.18 implies $\sum p^2 \approx 0.82$∑p2≈0.82, consistent with one species at roughly $90\%$90%.

A1 (AO3.2a) — Simpson's $D$D is dominated by the most abundant species because its term in $\sum p^2$∑p2 is squared. Grazing has not merely removed a species — it has re-engineered the dominance structure, and the index is sensitive to both effects.

(c) M1 (AO3.1a) — design weakness: a sample of forty individuals is small enough that rare-species detection is unreliable. The species–area relationship $S = c A^z$S=cAz predicts observed richness scales sublinearly with sampled area, so rare species are under-recorded at low effort.

M1 (AO3.2a) — matched improvement: increase quadrat number until a species–area accumulation curve plateaus (e.g., < $1$1 new species per $5$5 additional quadrats), and report a bootstrap confidence interval on $D$D.

Total: 8 marks.

---

Specimen question modelled on the Edexcel 9BI0 paper format

Question (6 marks): Describe how a researcher would use random quadrat sampling to estimate Simpson's index of diversity for plants in a $50 \times 50\,\text{m}$50×50m meadow, and evaluate the validity of conclusions drawn from a fixed sample size of ten quadrats.

Mark scheme decomposition by AO:

Marking point	AO	Credit-worthy content
1	AO1.1	States quadrats must be placed at coordinates from a random number generator — not "thrown over the shoulder" — to remove observer bias.
2	AO1.2	Describes the protocol — place quadrat, identify every plant species inside, count individuals (or estimate percentage cover for mat-forming species), tally, repeat.
3	AO2.1	Applies the calculation: aggregate counts across quadrats, compute $n/N$n/N, square and sum to obtain $\sum (n/N)^2$∑(n/N)2, then $D = 1 - \sum (n/N)^2$D=1−∑(n/N)2.
4	AO2.1	Sample-size reasoning: ten quadrats is likely too few — plot running $D$D against quadrat number and continue until the species–area accumulation curve plateaus.
5	AO3.1a	Evaluates: with ten quadrats, rare species are under-detected, observed $D$D is biased low, and the CI on $D$D is wide. Fixed-effort comparisons between meadows can be unreliable if richness is high.
6	AO3.2a	Concludes that valid inference requires standardised effort, a plateaued accumulation curve and a reported CI on $D$D — not a single point estimate.

Total: 6 marks split AO1 = 2, AO2 = 2, AO3 = 2. This is a typical Section B "describe and evaluate" question — Edexcel rewards candidates who describe the protocol (AO1) and justify the sample-size and confidence-interval moves (AO3) rather than merely listing equipment.

---

Synoptic links