Early development of subtidal macrofaunal assemblages: relationships to period and timing of colonization

Colonization and successional development of very diverse subtidal assemblages on rocky surfaces are not clearly understood. Artificial units of habitat (AUHs) made of nylon pot-scourers were used to test predictions from various models of succession. An experiment was designed in an attempt to unconfound the period of deployment (equals age of succession) from the time-period during which AUHs were deployed. AUHs were deployed in two sites, 100 m apart, for 1 month, starting at 0, 1, 2 and 3 months, for 2 months, starting 0 and 2 months and for 4 months from 0 month. Ninety-nine taxa were recorded in the AUHs. Successional change was not due to nett accumulation of taxa, nor simply to longer-term AUHs sampling successive different periods of time. Assemblages developing over the same period were different, but only a small amount of the variability was seasonal. Assemblages converged as period of deployment increased. There was less change from one to two months than from two to four months in the development of assemblages, but some of this was due to seasonal difference between the first and last two months. There were no differences between sites in any of the analyses of structure of assemblages. Few individual taxa showed consistent patterns of changing abundance with length of deployment. Different types of organisms showed markedly different patterns of arrival. The increase in number of species of gastropods was much smaller than the corresponding increase in number of taxa of polychaetes. Succession in these assemblages is complex and variable, but shows some repeated patterns. Fitting these to models of succession is only partially successful and new models are needed for very diverse assemblages.