Creek Dissolved Oxygen and Pasture Runoff
A secondary class studied a 2.4 km creek reach in June. Site A is 300 m upstream of a 12-ha cattle pasture with stream access; site B is 250 m downstream of the pasture. At both sites, students measured dissolved oxygen (DO), pH, nitrate, and temperature at 06:00, 10:00, 14:00, 18:00, 22:00, and 02:00 on one day. At site A, DO ranged from 7.2 to 8.4 mg/L and pH from 7.4 to 7.7. At site B, DO was 4.1 mg/L at 06:00, rose to 11.8 mg/L at 14:00, and fell to 5.0 mg/L by 02:00; pH was 7.3 at dawn and 7.9 at 14:00. Nitrate was 1.2 mg/L at A and 8.9 mg/L at B; temperature at B was 19°C at dawn and 24°C at 14:00. One sample was collected per time point at each site. A student concludes, 'The pasture is killing the creek because oxygen is low in the morning.'
1. Explain the diurnal DO and pH pattern at site B and evaluate the student's conclusion.
Read the worked answer
At site B, DO peaks in mid-afternoon and is lowest just before dawn. During daylight, algae and plants photosynthesize and produce O2 faster than respiration consumes it; at night, photosynthesis stops but respiration continues, so DO falls. The afternoon pH rise is consistent with photosynthesis removing CO2 and shifting the carbonate equilibrium. The pattern is real, but the conclusion is overstated. One pre-dawn value of 4.1 mg/L does not establish that the pasture is killing the creek: there is only one sample per time point and no continuous record. Low dawn DO can also be worsened by high temperature, low flow, and organic matter. The nitrate gradient (1.2 vs 8.9 mg/L) and the larger diurnal swing point toward nutrient enrichment, but causation requires flow, BOD, chlorophyll, and repeated-day data.
2. Identify two confounding variables in the upstream/downstream comparison and describe a sampling design change that would reduce confounding.
Read the worked answer
Two confounders are temperature and flow: site B was warmer at dawn and may have had different discharge, depth, or light exposure. A third is sampling frequency, because one grab sample cannot represent a diurnal minimum. A stronger design would sample A and B within minutes of each other at each time point, repeat over several days, and add a third reference site with similar temperature and canopy but no pasture. Continuous DO and temperature loggers at both sites for 48 hours would capture the pre-dawn minimum and let students compare minima at matched temperatures and flows. Sampling before and after storms, and measuring turbidity and benthic organic matter, would further separate pasture effects from within-day variation.
3. What additional measurements would test whether the low dawn DO is driven by excess algal production rather than by temperature alone?
Read the worked answer
Measure chlorophyll a in the water column and on rocks, benthic algal cover, BOD or COD, total phosphorus, and total nitrogen. If site B has much higher algal biomass and BOD than a temperature-matched upstream reference, excess production and subsequent respiration are supported. Continuous DO and temperature data can show whether B is supersaturated in late afternoon and crashes before dawn beyond what temperature alone predicts. If temperature were the main driver, DO should track temperature similarly at both sites and the afternoon supersaturation should be small. Repeated pre-dawn sampling at matched temperatures, plus flow measurements, would strengthen the inference; a single chlorophyll value would not.