How effective are MPAs? Predation control and 'spill-in effects' in seagrass-coral reef lagoons under contrasting fishery management

Marine protected areas (MPAs) are heavily promoted as a panacea for marine conservation, but lagging and sometimes idiosyncratic protection effects bring their overall effectiveness into question. In Kenyan lagoons, seagrass overgrazing by the sea urchin Tripneustes gratilla has been linked to removal of predators, but overgrazing has also been observed within well-protected MPAs. In this study we investigated the effectiveness of Kenyan MPAs in facilitating predation control over sea urchins, particularly T gratilla, in relation to system (seagrass or coral reef), distance to reefs, and seagrass presence. A strong protection effect on urchin densities on reefs and a negative correlation between T gratilla density and predation pressure (from sea stars, fish and gastropods) in seagrass beds (r(2) = 0.345) confirmed the importance of top-down control. Yet there were no clear effects of protection or distance to reefs in seagrass beds, most likely due to (1) low predator densities in the recently established Mombasa MPA; (2) 'spill-in' of aggregated T gratilla into the older Watamu MPA (potentially facilitated by low predation pressure on the large urchins and nutrient enrichment); and (3) a potential buffering effect of seagrass canopies on predation, regardless of distance to reefs. Effects of seagrass presence differed between areas, but indicated that overgrazing in some areas could be self-regulated by inducing higher urchin mortality. As MPA effects appear to be system-, time- and site-specific, managers should also assess other more holistic approaches (e.g. banned fishing of urchin predators and reduced nutrient input from land runoff) to protect seagrasses.