Before-After Control Impact (BACI) sampling designs are commonly used in environmental impact assessment and are considered the most effective for detecting changes due to anthropogenic disturbances. These designs handle local spatial variability through randomized placement of samples into a treatment stratum and a control stratum. When a contaminant disperses with distance from a point source it is suggested that a ’gradient‘ design will be more sensitive to change than randomized placement of samples. This requires allocating samples according to distance, rather than by random placement within randomly placed blocks. In this paper gradient versus random sampling designs were compared using data from an oil field in the North Sea. The gradient sampling design was more powerful than a randomized CI sampling design.