Variable impacts of an invasive marine predator on a threatened native prey: The role of non-linear species interactions

Background/Question/Methods

Ecologists and managers often assume that the impact of an invasive species is proportional to its abundance, but the per capita effect of an invader may vary across the landscape in response to biotic and abiotic context.  If per capita effects vary, then the sites where the invader is most abundant may not be the most heavily impacted, and sites may respond differently to invasion control efforts.  In the case of introduced predators, variation in per capita effect might arise from nonlinear functional responses, interference among predator individuals, or variation in behavioral parameters such as attack rates and handling times.  We examined variation in the impact of an introduced predatory marine snail, the Japanese oyster drill (Ocinebrina inornata), on a threatened native prey species, the Olympia oyster (Ostreola conchaphila), in Puget Sound, Washington State, USA.  Introduced predators have been suggested as a potential factor in Olympia oysters’ failure to recover from historic declines, yet at some sites relatively dense oyster beds appear to coexist with dense drill populations.
Results/Conclusions

In a tethering study conducted at a total of seven sites in two years, drill predation was locally intense (up to 99% mortality in 5 months, comprising up to 100% of total mortality), but was highly variable within and among sites.  Drill density alone explained only 26% of the variation in the per capita rate of oyster mortality due to drill predation.  Model selection indicated that drills have a Type II functional response, which leads to depensation, or inversely density-dependent oyster mortality.  Per capita effects of drills on oysters were greatest where oysters or alternative prey, particularly barnacles, were rare.  These patterns were consistent with the results of field enclosure experiments, which showed Type II functional responses for drills feeding on oysters and barnacles.  Drills strongly preferred barnacles as prey (Manly’s α = 0.92).  There was no evidence of interference among drills, as modeled by a ratio-dependent functional response.  We applied these results to a larger-scale experiment designed to evaluate drill control methods in the context of an ongoing oyster restoration project.  Although removal treatments were ineffective at reducing drill density because of reinvasion from neighboring areas, variation in drill impact was consistent with depensatory predation and the importance of barnacles as alternative prey.  Understanding the processes that drive variation in invasive species impacts may help in prioritizing sites for control efforts and designing effective mitigation strategies.