by Colin C Ferguson, Centre for Research into the Built Environment, Nottingham Polytechnic.
This paper was first published in GE’s June 1992 issue.
A properly formulated spatial sampling strategy should be a key component in any contaminated site investigation. A poorly designed or too sparse strategy may fail to provide adequate assurance that critically contaminated areas have not been missed. This in turn may result in unacceptable health risks to site workers or subsequent site users, or in expensive unplanned remedial work. Conversely, oversampling is needlessly costly and, without an appropriate sampling design, may still be ineffective. This paper reviews currently available guidance on spatial sampling strategies; uses modern spatial sampling theory to select four sampling plans, all of which are expected to perform reasonably well; and presents a report on a programme of computer experiments designed to evaluate detailed performance. Results show that a herringbone sampling plan is likely to be optimal when little or no prior information is available on the spatial distribution ofcontaminants.