The Wind Farm and the Bat Count
A regional council is deciding whether to approve a wind farm of 12 turbines on a ridge. A consultant's report says bat activity is 'low' based on two nights of acoustic surveys in winter. A local wildlife group says the ridge is a known migration corridor and provides three years of citizen-science records showing bat calls in spring and autumn. The council must decide whether to require further survey before approval. The developer argues that delaying costs money and that the consultant used 'standard methods'.
1. Compare the two sources of evidence and explain which is more reliable for the council's decision, giving reasons.
Read the worked answer
Consultant's evidence: two nights in winter, using 'standard methods'. Its strength is that it was collected by a professional using a defined protocol. Its weakness is sampling: two nights cannot capture seasonal variation, and winter is precisely when many bat species are least active or hibernating. A 'low' result in winter says little about spring and autumn migration. The report gives no indication of detection probability, weather conditions, or whether the two nights were representative.
Wildlife group's evidence: three years of records across seasons. Its strength is temporal coverage and the fact that it spans the migration periods the consultant missed. Its weakness is that citizen-science records vary in observer skill and effort, and absence of records is not proof of absence.
Which is more reliable for this decision: neither alone is sufficient, but the wildlife group's data are more relevant to the specific risk (collision during migration), because they cover the seasons when risk is highest. The consultant's two winter nights are too small a sample to support a general claim of 'low activity'. The council should treat the consultant's conclusion as unsupported rather than false, and require a survey designed around the seasons and species of concern.
2. Design a survey program that would give the council defensible evidence, specifying timing, methods and how to handle uncertainty.
Read the worked answer
Timing: Survey across at least one full year, with intensive sampling in spring and autumn migration and during peak summer activity. Sample on a stratified random schedule of nights, including calm and windy nights, so weather is not confounded with season.
Methods:
- Fixed acoustic detectors at multiple points along the ridge, including proposed turbine locations and control points away from the ridge.
- Automated call identification with manual verification of a random subsample, and report detection probability.
- Carcass searches beneath turbines once operating, with searcher-efficiency correction, to measure actual collision rates.
- Record weather, moon phase and temperature at each sampling event.
Handling uncertainty: Pre-register the decision rule (e.g. if estimated collision risk exceeds a threshold, require curtailment during migration). Report confidence intervals, not just 'low' or 'high'. Combine the consultant's and citizen-science data where methods allow, and state explicitly what the data cannot show.
This gives the council evidence matched to the season and species at risk, with quantified uncertainty rather than a single label.