Ontogenetic shifts in predator diet drive tradeoffs between fisheries yield and strength of predator-prey interactions

Abstract In some marine ecosystems, overharvesting marine predators has triggered major changes in trophic structure and ecosystem function. However, harvest levels that are deemed sustainable for one species may still lead to unexpected impacts elsewhere in the ecosystem. For example, by imposing an additional source of mortality, even sustainable harvesting can lead to a reduction in the number of large individuals within a population, and this truncation in size structure is typically more severe when the largest, most valuable size classes are targeted. Often small and large individuals within a species differ in important ways, including in what they consume, so a loss in predator-prey interactions could occur even without changes in overall predator biomass. Here we explore whether a truncation in predator size structure alone can reduce or functionally eliminate linkages between predator and prey. For this outcome to occur, a predator’s diet must change as it grows in size. We examined evidence for changes in diets with size among predators in three large marine ecosystems, and used a size-structured population dynamics model to evaluate the extent to which otherwise sustainable fishing results in disproportionate reductions in predation. Modelling suggests that diet shifts occurring late in life history (diet mid-point >45% of the maximum size) led to losses in predation that were more severe than would have been expected from losses in predator biomass. Further, the form of the fishery selectivity was less important than the degree of reduction in biomass within each size class relative to the timing of diet shifts. Empirical diet information demonstrates that piscivores vary widely in their onset to piscivory, and this may buffer the potential impacts of truncation in size structure. However, over half of the piscivores had diet shifts toward specific fish taxa at sizes at or above that which would lead to disproportionate reductions in prey consumption. Information about when and how diets change with predator size could identify ecosystems where harvest may lead to unexpected losses in predator-prey interactions.