Is the Car Park Polluting the Creek?
A Year 7 class samples invertebrates at three points along an urban creek. Site A is 50 m upstream of the stormwater outflow from a new car park. Site B is 20 m below the outflow. Site C is 200 m further downstream, where the creek runs through a park. At each site the students take one 30-second kick sample and identify animals to order. Counts - Site A: mayfly 18, stonefly 12, caddisfly 20, worm 6. Site B: mayfly 2, stonefly 0, caddisfly 5, worm 61. Site C: mayfly 9, stonefly 4, caddisfly 14, worm 21. Dissolved oxygen measured at each site at the same time of day: A 9.1 mg/L, B 5.2 mg/L, C 7.4 mg/L. Water temperature was similar at all three sites and flow was steady on the sampling day. A student reports that "Site B is the healthiest because it has the most animals - 68 compared with 56 and 48".
1. Judge which site is healthiest and explain why the student's count-based reasoning is wrong.
Read the worked answer
Site A is the healthiest, despite having fewer animals in total. The evidence is composition and oxygen, not abundance.
Mayfly, stonefly and caddisfly nymphs are oxygen-sensitive and cannot survive prolonged low oxygen or heavy sediment, so they are useful indicators. At A they make up 50 of 56 animals and oxygen is 9.1 mg/L - a healthy, clean riffle community. At B the sensitive groups collapse (mayfly 2, stonefly 0, caddisfly 5) and worms rise to 61 of 68, with oxygen down to 5.2 mg/L. Worms tolerate organic enrichment and low oxygen, so their numbers increase as conditions worsen. Site C shows partial recovery: sensitive animals return but are still below A, and oxygen at 7.4 mg/L is intermediate.
The student has confused abundance with health. A polluted site can carry more individual animals because a few tolerant species multiply, while sensitive species disappear. Richness and the proportion of sensitive taxa matter far more than the total count - which is exactly what a biotic index such as SIGNAL or EPT richness is designed to capture.
2. Explain how runoff from the car park could produce the pattern seen at Site B.
Read the worked answer
The outflow delivers several pollutants at once, and the oxygen drop at B is the fingerprint.
Runoff carries fine sediment, oil and fuel residues, tyre and brake wear particles, and nutrients washed off leaves and fertiliser. Sediment settles into the gaps between stones, smothering the gravel beds that mayfly and stonefly nymphs need and clogging their gills. Oil and organic matter are broken down by bacteria, and that decomposition consumes dissolved oxygen - which matches the fall from 9.1 to 5.2 mg/L. Nutrients feed algae; when the algae die they are decomposed by the same oxygen-consuming bacteria. Warm summer runoff also holds less oxygen and raises metabolic demand in the animals.
Downstream at C the creek has had 200 m to re-aerate over riffles, sediment has partly settled out, and pollutants are diluted, which fits oxygen recovering to 7.4 mg/L and sensitive taxa partly returning. Note that the pattern is consistent with the car park, but consistent is not the same as proven - a single sampling day cannot rule out other sources upstream of B.
3. Assess the reliability of the sampling and describe two changes that would make the conclusion stronger.
Read the worked answer
Reliability is weak in three ways. Each site was sampled once, with a single 30-second kick sample, so there is no way to tell whether the counts reflect the site or just that patch of stream bed - kick sampling is notoriously variable between spots. The sites may also differ in substrate and flow, so the comparison mixes habitat effects with pollution effects. And one day's sampling ignores season, recent rainfall and time since the outflow was built; a single storm can change results for weeks.
Two strong improvements: (1) take three replicate kick samples at each site, from the same habitat type (riffle, similar depth and flow), with the same net mesh size, and report the mean and range - non-overlapping ranges are much more convincing than single counts. (2) Repeat the survey monthly for at least a season and calculate a biotic index score each time, so the pattern at B can be shown to be persistent rather than a one-off. Adding continuous dissolved oxygen logging near the outflow would also link runoff events to oxygen dips.