Indirect effects and prey behavior mediate interactions between an endangered prey and recovering predator

Managing for simultaneous recovery of interacting species, particularly top predators and their prey, is a longstanding challenge in applied ecology and conservation. The effects of sea otters (Enhydra lutris kenyoni) on abalone (Haliotis spp.) is a salient example along North America's west coast where sea otters are recovering from 18th- and 19th-century fur trade while efforts are being made to recover abalone from more recent overfishing. To understand the direct and indirect effects of sea otters on northern abalone (H. kamtschatkana) and the relative influence of biotic and abiotic conditions, we surveyed subtidal rocky reef sites varying in otter occupation time in three regions of British Columbia, Canada. Sites occupied by sea otters for over 30 years had 16 times lower densities of exposed abalone than sites where otters have yet to recover (0.46 +/- 0.08/20 m(2) vs. 7.56 +/- 0.98/20 m(2)), but they also had higher densities of cryptic abalone (2.17 +/- 1.31/20 m(2) vs. 1.31 +/- 0.20/20 m(2)). Abalone densities were greater in deeper vs. shallower habitats at sites with sea otters compared to sites without otters. Sea otter effects on exposed abalone density were three times greater in magnitude than those of any other factor, whereas substrate and wave exposure effects on cryptic abalone were six times greater than those of sea otters. While higher substrate complexity may benefit abalone by providing refugia from sea otter predation, laboratory experiments revealed that it may also lead to higher capture efficiency by sunflower stars (Pycnopodia helianthoides), a ubiquitous mesopredator, compared to habitat with lower complexity. Sea otter recovery indirectly benefitted abalone by decreasing biomass of predatory sunflower stars and competitive grazing sea urchins, while increasing stipe density and depth of kelp that provides food and protective habitat. Importantly, abalone persisted in the face of sea otter recovery, albeit at lower densities of smaller and more cryptic individuals. We provide empirical evidence of how complex ecological interactions influence the effects of recovering predators on their recovering prey. This ecosystem-based understanding can inform conservation trade-offs when balancing multifaceted ecological, cultural, and socio-economic objectives for species at risk.