Increasing fishing activity on coral reefs threatens both their fisheries and biodiversity . Marine fishery reserves, areas in which fishing is permanently prohibited, offer potential economic and conservation benefits. Despite their strong potential, we lack a fundamental understanding of the design of marine fishery reserves. Computer models -- in which fish life history, movement dynamics, and fishing pressure are included -- predict that reserves can maintain productive fisheries even if they encompass large proportions of management areas. Moreover, the models suggest that reserves will increase the persistence of easily over-harvested species and will also decrease year-to-year variations in catches. The reserve design that will maximize long-term fish catches depends on the life history, larval dispersal, and adult movement dynamics of the target species, as well as the fishing effort in the management area. Nevertheless, these computer models predict that the use of marine fishery reserves is aneffective general fishery management technique, useful in particular for multi-species fisheries, and which also provides significant conservation benefits.

Marine reserves have come under criticism because of the short-term fishery losses likely to be associated with them. At the same time, marine reserves have been touted as a tool to rehabilitate depleted populations, at least for species with relatively limited adult movement. I used models to compare the short- and long-term fisheries consequences of three fishery-management tools for depleted populations. These management tools included temporary closure of the entire fishery, minimum size limits, and marine reserves. I compared them using three short-term indices: magnitude of initial drop in catches relative to those prior to new management, years until catches reached prior levels, and cumulative loss during those years. I examined a single long-term index: long-term sustainable yields. Results highlighted the potential of reserves as an efficient and effective fishery-management tool for species that will remain within reserve boundaries. Reserves created few short-term losses beyond those associated with other management measures, yet produced the highest stable catch levels. Moreover, peak catches with reserves occurred with less restriction than peak catches with other management measures. These results were consistent across two species that matured before entering the fishery but did not apply to one species that was fished while immature. In that latter case, minimum size limits produced more substantial benefits than reserves could. Nevertheless, these analyses suggest that a wide range of circumstances exist where reserves, if properly designed to minimize adult spillover while allowing abundant larval transport, can maximize fisheries harvests with a minimum of total restrictions.

We used an existing model in our initial examination of the effects of marine harvest refugia, i.e. protected areas where fishing is prohibited, on the bocaccio rockfish (Sebastes paucispinis) population off central and northern California. We incorporated size-specific life history information into the model, including growth, survival, and fecundity, and examined the long-term fisheries consequences of refugia. The key assumptions of the model were that larvae dispersed widely from their areas of origin and adults remained inthe areas where they settled. Using the model based on these assumptions, we predict moderate-to-great potential for enhancement of bocaccio catch if refugia are established, depending on the magnitude of fishing mortality outside the proteced area. From this model, we also suggest that refugia could decrease variability in annual catches. In the future, we will expand this study to include rockfish species with different life history characteristics and to examine in greater detail the effects of adult movements on the model output.

Ecologists have paid increasing attention to the design of marine protected areas (MPAs), and their design advice consistently recommends representing all habitat types within MPAs or MPA networks as a means to provide protection to all parts of the natural ocean system. Recent developments of new habitat-mapping techniques make this advice more achievable, but the success of such an approach depends largely on our ability to define habitat types in a way that is ecologically relevant. We devised and tested the ecological relevance of a set of habitat-type definitions through our participation in a stakeholder-driven process to design a network of MPAs, focusing on no-take marine reserves in the Seaflower Biosphere Reserve, San Andrés Archipelago, Colombia. A priori definitions of habitat types were ecologically relevant, in that our habitat-type definitions corresponded to identifiable and unique characteristics in the ecological communities found there. The identification of ecological pathways and connectivity among habitats also helped in designing ecologically relevant reserve boundaries. Our findings contributed to the overall design process, along with our summary of other general principles of marine reserve design. Extensive stakeholder input provided information concerning the resources and their patterns of use. These inputs also contributed to the reserve design process. We anticipate success for the Seaflower Biosphere Reserve at achieving conservation and social goals because its zoning process includes detailed yet flexible scientific advice and the participation of stakeholders at every step.

We used fishery population models to assess the potential for marine fishery reserves, areas permanently closed to fishing, to enhance long-term fishery yields. Our models included detailed life history data. They also included the key assumptions that adults did not cross reserve boundaries and that larvae mixed thoroughly across the boundary but were retained sufficiently to produce a stock-recruitment relationship for the management area. We analyzed the results of these models to determine how reserve size, fishing mortality, and life history traits, particularly population growth potential, affected the fisheries benefits from reserves. We predict that reserves will enhance catches from any overfished population that meets our assumptions, particularly heavily overfished populations with low population growth potential. We further predict that reserves can enhance catches when they make up 40% or more of fisheries management areas, significantly higher proportions than are ty reserve systems. Finally, we predict that reserves in systems that meet our assumptions will reduce annual catch variation in surrounding fishing grounds. The fisheries benefits and optimal design of marine reserves in any situation depended on the life history of the species of interest as well as its rate of fishing mortality. However, the generality of our results across a range of species suggest that marine reserves are a viable fisheries management alternative.