Mesoscale alongshore and cross-shore transport and settlement of invertebrate larvae on the south east coast of South Africa

Pelagic larval stages of most marine benthic species are important in maintaining coastal populations of adults. Several physio-chemical factors such as currents, winds, larval behaviour and time have an influence on the dispersal and transport of larvae to the adult habitat but their role is however still poorly understood. The aim of this study was to investigate the alongshore and cross-shore transport and temporal delivery of invertebrate larvae at four sites (Kenton on Sea, Cannon Rocks, Schoenmakerskop and Cape St Francis) along the south east coast of Eastern Cape, South Africa. Larval distribution of several taxa was determined during two sampling periods, by collecting water samples at nearshore line transects (3 distances: 900m, 1500m and 2400m – perpendicular to the shore) and at three depths (surface, thermocline/middle, bottom), which ranged from 15m (inshore) to 50m (offshore). Physical properties (current speed and direction, dissolved oxygen, fluorescence, turbidity, temperature, salinity, pH, pressure, density and conductivity) were measured and coupled in order to further understand larval distribution. The larvae were analysed as both total abundance and separately as the abundances of a variety of taxa which were: Perna perna, Mytilus galloprovincialis, oysters, early and late nauplii and cyprids. Larval settlement and recruitment on the rocky shores were measured by monthly deployment and collection of 20 (10 each for barnacles and mussels) artificial collectors at each site, which were preserved in ethanol or frozen for further processing. Multiple Permutational Multivariate Analysis of Variance (PERMANOVA) analyses were used to test the effects of site, depth and distance from the shore for the nearshore larvae (taxa analysed separately). In addition, a distance based linear model (distLM) was performed to analyse the relationship between the total larval abundance and the above mentioned physical variables. Multiple two-way analyses of variance (ANOVA) were performed to test the effects of months and sites on the settlement and recruitment of the larvae (P. perna, M. galloprovincialis, other bivalves, cyprids and juvenile barnacles) arriving on the shore. For the nearshore larval distribution, results from the PERMANOVAs revealed that most taxa showed a significant site and depth interactions with the exception of ‘early nauplii’ taxon. Also nearly all taxa were found within the thermocline, besides ‘oyster’ and ‘cyprids’ which were located at thermocline or bottom. Larvae were also located at variable distances from the shore, with most occurring at the offshore stations. Furthermore, there was a geographical separation of larval abundance according to sites, with most larvae located at Cannon Rocks and Kenton on Sea and least at Schoenmakerskop and Cape St Francis. For the settlement and recruitment, most taxa showed a seasonal trend, with the highest abundance of settlers and recruits expectedly appearing during the summer months of the sampling period. Additionally there was a site effect for most taxa (P. perna, M. galloprovincialis, other bivalves and juvenile barnacles), where settlers and recruits were mostly found at Cannon Rocks. Significant differences in abundance of settlers and recruits amongst the four sites indicate spatial and temporal variability for the targeted 180km stretch of coast. Overall for this study, taxon and ontogenetic stage of larvae were important in the distribution and abundance of larvae. Throughout the time frame of nearshore and intertidal sampling, Cannon Rocks consistently resulted as a ‘hot spot’ for larval abundance, settlement and recruitment, while a broad west to east separation was also observed. These results hence highlight that within this stretch of c.180km coast, time, taxon, ontogeny and post-settlement factors influence early dynamics of benthic populations