Abstract
Marine reserves have increasingly been recognised for their potential to address the pervasive problem of unsustainable harvest of fisheries worldwide. Biologists advocate the benefits of increased spawning biomass, larger modal sizes and greater densities of fish within marine reserves, and the possibility of spillover to adjacent fishable areas.
Bioeconomic studies, however, find that pay-offs from stand-alone marine reserves rarely compete with sustainable yield management schemes, but that they can be beneficial when stocks are heavily exploited. Most of these bioeconomic models are analytical and deterministic in nature, and therefore ignore the redistribution of effort in response to closure and the inherent uncertainty of the marine environment.
We present a bioeconomic analysis of a network of no-take areas around Stewart Island in New Zealand applied to the shellfish species paua (abalone) that incorporates both predicted redistribution and reduction in effort, as well as stochastic recruitment. A nested logit model is applied to spatially recorded catch and effort data by the Ministry of Fisheries between 1998 and 2003 to capture the two level decision-making process of divers. On any given day, divers decide whether to go diving at all, and if so, which of the 16 statistical areas around Stewart Island to visit. Weather conditions, spatially varying levels of catch per unit of effort and distance are used as explanatory variables to select areas for closure according to the 'least economic impact' in terms of loss of diving trips. An age-structured biological model is developed with parameters specifically applied to paua stocks around Stewart Island. Virgin paua biomass as of 1974 is estimated on the basis of growth, survival, post-larval recruitment and egg production in the absence of fishing. Historic catch rates are then applied to find overall and area-specific levels of exploitation rates, spawning biomass, egg production, legal biomass and numbers of paua. In a final step, the economic model is linked to the biological model to simulate the imposition of no-take areas when taking account of the initial disproportional shift of harvest to fished areas in the first year, and the increase in overall pressure on legal biomass in the years thereafter.
We contribute to the marine reserve debate by showing that in the very long run, the overall yield under closure of a relatively small area approaches and even slightly surpasses the yield under no closure for an assumed spillover gradient of 40% despite the redistribution of effort. The most important benefits of marine reserves emerge when stochastic recruitment is included in the recruitment function. In practice, predictions about the stock status and the impact of different harvest levels become much more difficult when acknowledging the inherent variability of the marine environment. The likelihood of stock collapse depends on the assumed value of two recruitment parameters, which highlights the effects of parameter uncertainty and emphasizes the role of marine reserves for population persistence. We also show that under uncertainty average yields under a management regime of a network of notake areas in addition to the quota system can equal yields under no closure for an assumed spillover gradient of 40%, despite the increased pressure on areas adjacent to the closed areas.
Our findings have significant implications for the management of the paua fishery at Stewart Island. For a heterogeneously abundant species, such as paua, spatial management in addition to quota limits could be vital in ensuring the long-term sustainability of the fishery given the inherent variability of the marine environment.