Search

Understanding species’ roles in food webs requires an accurate assessment of their trophic niche. However, it is challenging to delineate potential trophic interactions across an ecosystem, and a paucity of empirical information often leads to inconsistent definitions of trophic guilds based on expert opinion, especially when applied to hyperdiverse ecosystems. Using coral reef fishes as a model group, we show that experts disagree on the assignment of broad trophic guilds for more than 20% of species, which hampers comparability across studies. Here, we propose a quantitative, unbiased, and reproducible approach to define trophic guilds and apply recent advances in machine learning to predict probabilities of pairwise trophic interactions with high accuracy. We synthesize data from community-wide gut content analyses of tropical coral reef fishes worldwide, resulting in diet information from 13,961 individuals belonging to 615 reef fish. We then use network analysis to identify 8 trophic guilds and Bayesian phylogenetic modeling to show that trophic guilds can be predicted based on phylogeny and maximum body size. Finally, we use machine learning to test whether pairwise trophic interactions can be predicted with accuracy. Our models achieved a misclassification error of less than 5%, indicating that our approach results in a quantitative and reproducible trophic categorization scheme, as well as high-resolution probabilities of trophic interactions. By applying our framework to the most diverse vertebrate consumer group, we show that it can be applied to other organismal groups to advance reproducibility in trait-based ecology. Our work thus provides a viable approach to account for the complexity of predator–prey interactions in highly diverse ecosystems. Copyright

Few studies assess the effects of recreational fishing in isolation from commercial fishing. We used meta-analysis to synthesise 4444 samples from 30 years (1987–2017) of fish surveys inside and outside a large network of highly protected reserves in the Ningaloo Marine Park, Western Australia, where the major fishing activity is recreational. Data were collected by different agencies, using varied survey designs and sampling methods. We contrasted the relative abundance and biomass of target and non-target fish groups between fished and reserve locations. We considered the influence of, and possible interactions between, seven additional variables: age and size of reserve, one of two reserve network configurations, reef habitat type, recreational fishing activity, shore-based fishing regulations and survey method. Taxa responded differently: the abundance and biomass inside reserves relative to outside was higher for targeted lethrinids, while other targeted (and non-targeted) fish groups were indistinguishable. Reef habitat was important for explaining lethrinid response to protection, and this factor interacted with reserve size, such that larger reserves were demonstrably more effective in the back reef and lagoon habitats. There was little evidence of changes in relative abundance and biomass of fishes with reserve age, or after rezoning and expansion of the reserve network. Our study demonstrates the complexities in quantifying fishing effects, highlighting some of the key factors and interactions that likely underlie the varied results in reserve assessments that should be considered in future reserve design and assessment.

Baselines and benchmarks (B&Bs) are needed to evaluate the ecological status and fisheries potential of coral reefs. B&Bs may depend on habitat features and energetic limitations that constrain biomass within the natural variability of the environment and fish behaviors. To evaluate if broad B&Bs exist, we compiled data on the biomass of fishes in ~1000 reefs with no recent history of fishing in 19 ecoregions. These reefs spanned the full longitude and latitude of Indian and Pacific Ocean reefs and included older high-compliance fisheries closures (>15 yr closure) and remote reef areas (>9 h travel time from fisheries markets). There was no significant change in biomass over the 15 to 48 yr closure period but closures had only ~40% of the biomass (740 kg ha −1 , lower confidence interval [LCI] = 660 kg ha −1 , upper confidence interval [UCI] = 810 kg ha −1 , n = 157) of remote tropical reefs (1870 [1730, 2000] kg ha −1 , n = 503). Remote subtropical reefs had lower biomass (950 [860, 1040] kg ha −1 , n = 329) than tropical reefs. Closures and remote reef fish biomass responded differently to environmental variables of coral cover, net primary productivity, and light, indicating that remote reefs are more limited by productivity and habitat than closures. Closures in fished seascapes are unlikely to achieve the biomass and community composition of remote reefs, which suggests fisheries benchmarks will differ substantially from wilderness baselines. A fishery benchmark (B 0 ) of ~1000 kg ha −1 adjusted for geography is suggested for fisheries purposes. For ecological purposes, a wilderness baseline of ~1900 kg ha −1 is appropriate for including large and mobile species not well protected by closures.