In general terms as mentioned above the macro infauna will provide the best indicator to the health of the systems. Reference to baseline data should be made to define a set of stations for routine monitoring the numbers of which will be dictated by financial and time considerations. In intertidal areas core samples on transects taken down the intertidal zone should be taken. There is no clearly defined time of year for sampling as summer months which provide ample daylight and amenable weather conditions are subject to ephemeral populations of invertebrates and recruitment of juveniles. Established winter populations may fluctuate due to storm events, heavy rainfall and freezing, the limited hours of daylight is also a problem.

If samples are to be compared over time it is important that the sample time must not vary from year to year and major weather events noted (Dalkin & Barnett, 1998). Subtidally at representative stations grab sampling should be undertaken. In areas where the epifaunal component is important use of trawls and towed or remote operated video along set transects may be necessary. Subtidally the time of sampling is best carried out around May as weather conditions are generally best from May to September and recruitment generally occurs from February to May (Thomas, 1998). However weather conditions may preclude this but it is important that samples are taken at as close to the same time of year as possible.

The high cost of benthic sampling may mean that it is more practical to only work up a reduced number of replicates or stations with the remainder being carried out to allow full statistical analysis if some change is evident.

The frequency of monitoring will be largely determined by financial and time constraints. Hiscock (1998a) recommends following a six year cycle for statutory sites with a proportion of the sites e.g. those which have a particular scientific or conservation interest monitored more regularly at least every three years. These guidelines seem reasonable for the monitoring of the intertidal and subtidal SACs dealt with here but this will depend on the size of the areas involved.

The characterisation of the community structure will involve separation of the main trends within a multivariate data set. Each species and each physico-chemical parameter constitutes a variable within the statistical analysis. It is often possible to rationalise the species list to around ten important species which control the dynamics of the community particularly if manipulation experiments are feasible or have been carried out (Gray, 1981). Multivariate analysis of sites within the SACs e.g. techniques such as TWINSPAN (Gauch, 1982) or the MDS and SIMPER routine in the PRIMER package (Warwick & Clarke, 1994) identify groups of sites with similar composition and the indicator species which are particularly well correlated to the assemblages found within the SAC. Replicate samples will usually be required to allow detailed statistical analysis and significance testing.

References