Designing a downstream river fieldwork investigation
A class of 24 students investigates how channel characteristics change downstream along a 3.4 km reach of a small river. They have one day, two 50 m tapes, a flow meter, a clinometer and a 1:25,000 topographic map. The map shows the river from grid 462 318 (spot height 84 m) to 498 302 (spot height 31 m), crossing a road at 478 311 and a housing estate at 486 306. Access is permitted only at the road bridge and a footpath at the downstream end; the middle section is marshy. Pilot data show width 3.1–8.6 m, mean depth 0.18–0.94 m, velocity 0.14–0.72 m s⁻¹ and bedload size 4–65 mm. The teacher wants a systematic sampling strategy but is worried about safety and time.
1. Using the map description and pilot data, justify a sampling strategy that is both systematic and safe.
Read the worked answer
A purely systematic strategy along the whole reach is not safe or practical because the middle section is marshy and access is limited. A stratified systematic strategy is better.
Use the map to stratify the reach:
- The river falls from 84 m at 462 318 to 31 m at 498 302, a drop of 53 m over 3.4 km. This confirms a meaningful downstream gradient, so sites should be spread along the reach.
- The road bridge at 478 311 and the footpath near the downstream end are the only legal access points. The housing estate at 486 306 is a possible disturbance, so at least one site should be upstream and one downstream of it.
Proposed strategy:
- Divide the reach into four strata: above the bridge, bridge to housing estate, estate to footpath, and below the footpath if accessible.
- Within each accessible stratum, sample at fixed intervals, for example every 250 m, but replace inaccessible marsh sites with the nearest safe alternative and record the substitution.
- At each site, measure width, depth and velocity at five equidistant points across the channel. Measure bedload at three points.
Safety and time:
- Avoid depths above about 0.8 m and use a buddy system and rope. With 24 students, four groups can cover about eight sites in one day. The flow meter and tapes are shared, so the timetable must be realistic.
Judgement: the strategy is systematic within accessible strata, not across the whole river. That is defensible because safety and access are real constraints, provided the substitutions are recorded and the conclusions are limited to the accessible reach.
2. Identify two sources of error in the discharge measurement and explain how they would affect the result.
Read the worked answer
1. Measuring velocity at only one depth or one point in the cross-section.
Discharge is the product of cross-sectional area and mean velocity. Velocity is not uniform: it is usually slower near the bed and banks because of friction. If students measure only at the surface or only at the centre, they may record a velocity that is higher than the true mean. This would overestimate discharge. Measuring at 0.6 of the depth from the surface is a standard approximation, but it still assumes a typical velocity profile.
2. Measuring width at the water surface rather than the channel bed.
If the banks are undercut or the water surface extends into vegetation, the surface width may be larger than the effective flow width. That would overestimate the cross-sectional area and therefore discharge. Conversely, if the tape sags or is not horizontal, the width could be underestimated.
Other possible errors: the flow meter may be uncalibrated or placed too close to the bed; the depth may be measured at the bank rather than at the velocity point; and the cross-section may be assumed rectangular when it is not.
Judgement: the direction of error matters. Surface-only velocity tends to overestimate discharge; a sagging tape or an unrepresentative cross-section can bias it either way. Students should take multiple measurements and report a mean with a range, not a single value.
3. Explain how the students should present and analyse the data to test the hypothesis that channel characteristics change downstream.
Read the worked answer
The hypothesis predicts a trend with distance downstream, so the analysis should show both the trend and the variation around it.
Presentation:
- Plot scatter graphs with distance downstream on the x-axis and width, mean depth, mean velocity, discharge and bedload size on the y-axis. Use the same distance scale so the graphs can be compared.
- Draw cross-sections for each site to show channel shape, not just width and depth.
- Use the map to mark sites and any tributaries, the housing estate and the road bridge, because these may explain anomalies.
Analysis:
- Calculate discharge (width × mean depth × mean velocity) at each site and plot it downstream.
- Use Spearman's rank correlation to test whether each variable changes systematically with distance. With about eight sites, a scatter graph alone is weak evidence.
- Identify anomalies and explain them using the map and field notes. For example, a site below the housing estate may have a modified channel; a tributary may increase discharge; a marsh may reduce velocity.
Judgement: the conclusion should distinguish a general downstream trend from local controls. If width and discharge increase downstream but velocity does not, the hypothesis is only partly supported. The students should also acknowledge that the inaccessible marsh section limits the representativeness of the sample.