1.12 Gradients and hierarchies in sampling
There has been debate about appropriate sampling designs to use for analyses of
influences along an environmental or other gradient. For example, when testing a
hypothesis about the influence of discharge of fresh water from an estuary, it is
appropriate to sample along a gradient from the mouth of the estuary. Similarly
34 Methods for the Study of Marine Benthos
and more obviously, sampling down a depth gradient requires the sites to be along
the gradient.
In other cases, the situation is not so clear-cut and whether or not sampling
along a gradient is the best option depends entirely on the hypothesis being tested.
Furthermore, even if a gradient is to be sampled, the spacing of samples is dependent
on the precise issue under examination.
To provide an illustration of the issues, consider a simple case of discharges
of contaminants from an outfall on the coast (see Underwood, 2000, for more
details). In all three situations considered here, it is proposed that pollution due to
the contaminants will be revealed by differences in the assemblages of animals in
sediments near to, as opposed to far away from, the outfall.