Spatio-temporal patterns in the diversity of demersal fish communities off the south coast of South Africa

The diversity of ecological communities has been the focus of many studies. Because biodiversity provides several indicators used in an Ecosystem Approach to Fisheries (EAF) to track changes in fish communities, we investigated the spatial and temporal patterns in the diversity of some demersal fish communities subjected to varying fishing pressure. Depth and catch rate were the most important predictors in explaining changes in diversity followed by longitude and survey year. Diversity, as measured by the various indices except for taxonomic distinctness (a dagger*), initially declined with increasing depth to about a depth of 80 m, then increased to about 150 m after which it declined. Taxonomic distinctness index (a dagger*) showed an increase in the taxonomic heterogeneity of the demersal community below the 300-m isobath. Diversity remained relatively constant with increase in longitude to around 24A degrees E (which has the lowest diversity) after which it increased. The assessment of the temporal trend in diversity indicates that survey year has a significant effect on all diversity indices except for a dagger*. Diversity increased and dominance declined with time. This may be result of a decline in the abundance of dominant species or an increase in the abundance less dominant species, or a combination of both effects. Multivariate analysis of the set of diversity indices showed three groups of indices: those reflecting species richness (S, Margalef's d), those measuring mainly taxonomic relatedness (a dagger*), and those balancing the richness and evenness components of diversity (J', H', lambda, a dagger, Hill's N1, and Hill's N2). The relationship between evenness, catch rate, and size was also investigated. Size classes with highest evenness were found to have lowest catch rate and vice versa. This highlights the need to consider the size and trophic level of species when linking diversity to the functioning of ecosystems.