No abstract

Summary1. Ecosystem-based management of coral reef fisheries aims to sustainably deliver a diverse portfolio of ecosystem services. This goal can be undermined if the ecosystem shifts into a different state, with altered ecosystem functions and benefits to people. If levels of drivers that cause transitions between states are identified, management measures could be aimed at maintaining drivers below these levels to avoid ecosystem shifts. 2. Analysing data from a large number of Caribbean coral reefs (N = 2001), suites of nonlinear thresholds were identified between metrics of coral reef processes and structure along a gradient of total fish biomass (a proxy for fishing pressure). Several metrics (macroalgal cover, invertivorous fishes and fish species richness) associated with coral-dominated reefs exhibited thresholds at relatively high fish biomass levels (50-88% of unfished biomass). Other metrics (urchin biomass, ratio of macroalgal to coral cover, herbivorous fishes and coral cover) showed thresholds at lower fish biomass levels (28-37% of unfished biomass). 3. Ratios of total fish biomass in fishing areas to closed areas (unfished biomass) in the Caribbean indicate that reefs may generally be at risk for change at ratios between 0Á5 (coral dominated) and 0Á3 (macroalgal dominated). Similar relationships were found for coral reefs in the Indian Ocean. While these results illustrate thresholds at the scale of the entire Caribbean, assessing local reefs is advisable because biomass levels vary within the region, and reef trajectories depend on past, present and future local conditions. 4. Synthesis and applications. If the thresholds in this study are generalizable to scales relevant to management, it may be possible to produce sustainable yield while simultaneously maintaining coral-dominated reefs by restricting fishing mortality to levels that result in biomass ratios near 0Á5. Fishing down to biomass ratios near 0Á3 may increase the risk of overfishing (resulting in lower long-term yields) and transition to macroalgal-dominated reefs. Thresholds offer a simple and powerful way for managers to operationalize precautionary ecosystem-based fishery management by adaptively limiting fishing pressure in order to (i) maintain desirable coral reef conditions, (ii) establish a system-specific target for generating pretty good yield and (iii) maintain sustainable multi-species fishery yields.

Coral reefs support numerous ornamental fisheries, but there are concerns about stock sustainability due to the volume of animals caught. Such impacts are difficult to quantify and manage because fishery data are often lacking. Here, we suggest a framework that integrates several data‐poor assessment and management methods in order to provide management guidance for fisheries that differ widely in the kinds and amounts of data available. First, a resource manager could assess the status of the ecosystem (using quantitative metrics where data are available and semi‐quantitative risk assessment where they are not) and determine whether overall fishing mortality should be reduced. Next, productivity susceptibility analysis can be used to estimate vulnerability to fishing using basic information on life history and the nature of the fishery. Information on the relative degree of exploitation (e.g. export data or ratios of fish density inside and outside no‐take marine reserves) is then combined with the vulnerability ranks to prioritize species for precautionary management and further analysis. For example, species that are both highly exploited and vulnerable are good candidates for precautionary reductions in allowable capture. Species that appear to be less vulnerable could be managed on a stock‐specific basis to prevent over‐exploitation of some species resulting from the use of aggregate catch limits. The framework could be applied to coral reef ornamental fisheries which typically lack landings, catch‐per‐unit‐effort and age‐size data to generate management guidance to reduce overfishing risk. We illustrate the application of this framework to an ornamental fishery in Indonesia.

Coral reefs worldwide face an uncertain future with many reefs reported to transition from being dominated by corals to macroalgae. However, given the complexity and diversity of the ecosystem, research on how regimes vary spatially and temporally is needed. Reef regimes are most often characterised by their benthic components; however, complex dynamics are associated with losses and gains in both fish and benthic assemblages. To capture this complexity, we synthesised 3,345 surveys from Hawai‘i to define reef regimes in terms of both fish and benthic assemblages. Model-based clustering revealed five distinct regimes that varied ecologically, and were spatially heterogeneous by island, depth and exposure. We identified a regime characteristic of a degraded state with low coral cover and fish biomass, one that had low coral but high fish biomass, as well as three other regimes that varied significantly in their ecology but were previously considered a single coral dominated regime. Analyses of time series data reflected complex system dynamics, with multiple transitions among regimes that were a function of both local and global stressors. Coupling fish and benthic communities into reef regimes to capture complex dynamics holds promise for monitoring reef change and guiding ecosystem-based management of coral reefs.