Interactive effects of wave exposure and tidal height on population structure of the mussel Perna perna

The influence of wave exposure and of tidal height on mussel (Perna perna Linnaeus) population structure (size, density, biomass and adult/juvenile correlations) was examined at 18 sites along the south coast of South Africa. Sites were classified as exposed or sheltered prior to sampling, without reference to the biota, on the basis of aspect, topography and wave regime. A single set of samples was collected from each site during three spring tide cycles. Adult mussels on these shores almost always attach directly to the rocks, and layering of mussels is virtually absent. Shore height always had a strong influence on population structure, but exposure had significant effects only lower on the shore, and almost exclusively on mussel sizes. Principal component analysis (PCA), based on size distribution data for each population, revealed a general upshore decrease in the modal size of the adult cohort. The effects of exposure on size distribution, however, varied with tidal height. PCA separated exposed zones, with larger mussels, from sheltered zones on the low-shore. Farther upshore the two shore types were increasingly confounded. The maximum size of mussels showed a similar pattern, with significant differences (ANOVA, p < 0.05) between exposed and sheltered sites only on the low- and mid-shores. Density was calculated from randomly placed quadrats (i.e. not necessarily from areas of 100% cover) and showed a different pattern. Adult (>15 mm) densities decreased up the shore, with low-, mid- and high-shore zones being significantly different from one another (ANOVA, p < 0.0001; followed by multiple range tests). However, exposure had no significant effect on density, nor was there a significant interaction with zone. Recruit (<15 mm) densities were positively correlated with adult (>15 mm) densities in all zones and for both exposure regimes ( p < 0.05 in all cases), but there was considerable variability and extremely low predictability in these relationships (r ² generally <0.2). Predictability tended to be greater towards the high-shore, where adults were more clumped. As with density, biomass was not affected by exposure, but decreased upshore as mean size and density decreased. A reduction in the influence of exposure farther upshore may be caused by greater emersion overriding the effects of exposure. The presence of free space within mussel beds and significant correlations between recruit and adult densities suggest that these mussel populations are recruit limited. Received: 7 January 2000 / Accepted: 6 July 2000