PERSISTENCE AND SPACE OCCUPANCY BY SUBTIDAL BLUE MUSSEL PATCHES

At a subtidal, soft-bottom site in the western Baltic Sea, mussel (Myt ilus edulis) patches co-occur with high predator abundances. Sea star (Asterias rubens) biomasses, in particular, exceed reported values con sidered sufficient for restricting mussels to the intertidal zone. To determine how mussels can persist in the face of intense predation, we decomposed patch space occupancy into the relative contributions of n ewly arriving individuals (recruitment) and of increases in body size of the individuals already present in the patch over 13 mo. Sea stars, as major predators, were only able to control 77% of the potential pe r capita recruitment rate of 91 individuals/yr in 2 m depth. The remai ning recruitment rate of 21 individuals/yr was sufficient to allow pat ches to occupy 1.6 times more space per year. Transplantation of patch es to 6 m depth, where recruitment is negligible, revealed that sea st ars were also ineffective in controlling mussel coverage through consu mption of larger mussels (>1 yr, >30 mm shell length). In deeper water , space occupancy of patches through increases in mussel body size was able to balance predation mortality, demonstrating that mussels attai ned a relative refuge in size at only 33 mm shell length. Based on the measured shell growth rates, mussels attain this size after approxima te to 15 mo. In situ observations of Asterias feeding activity, the ra tios between necessary predator sizes to attack prey of a given size, and predator size distributions suggest that sea stars were on average too small to feed effectively on adult (>1 yr) mussels. Probably, Ast erias cannot respond to abundant prey and increase its maximal body si ze at the site because salinities are at its lower tolerable limit (12 -18 g/kg). Thus, bottom-up factors such as high prey productivity in c oncert with subtle size-based ineffectiveness of the predator populati on allow otherwise unstable predator-prey populations of a generalist predator and its preferred prey to coexist. Although mussel predators were unable to decimate mussels to local extinction, a release of expe rimental mussel patches from predation with strong recruitment (2 m de pth) resulted in an approximately sevenfold yearly areal increase in s hallow treatments, which would lead to a 100% mussel cover at the site within 1 yr. Given that mussels can dominate both rocky substratum an d soft sediment, we also studied the effect of substratum quality in f actorial combination with presence/absence of predation and water dept h on mussel abundance. Attachment to stable substratum did not affect recruitment to the patches or patch space occupancy, but it completely prevented patch dislodgment and subsequent drift. In contrast to rock y shores, mussel patch dislodgment may represent the major mode of pat ch dispersal and new patch formation in soft-bottom environments as de monstrated by a drift collector fence.