Patterns of mussel recruitment in southern Africa: a caution about using artificial substrata to approximate natural recruitment

Quantifying sessile marine invertebrate recruitment often requires destructive sampling or extrapolation from artificial substrata, the latter introducing the danger of artifacts. We measured intertidal mussel recruitment into mussel beds and into brushes at three-month intervals for five years across 3,200 km of southern Africa and determined substrata effects on recruitment rate. Recruitment into mussel beds showed a strong, coast-wide gradient, with high recruitment on the West coast, diminishing on the South coast, and increasing slightly on the East coast. At scales of 10 s of km, brushes reflected natural temporal recruitment variability, with a strong significant linear correlation between recruitment into brushes and into mussel beds. However, the relationship became semi-logarithmic when comparing among locations at a scale of 100 s of km. Artificial substrata thus reflect local natural settlement well but may be a poor indicator of it when spatial scales are large, particularly when mussel bed topography is complex, or localities have very different recruitment densities.