Data from: Bayesian inference reveals positive but subtle effects of experimental fishery closures on marine predator demographics

Global forage-fish landings are increasing, with potentially grave consequences for marine ecosystems. Predators of forage fish may be influenced by this harvest, but the nature of these effects is contentious. Experimental fishery manipulations offer the best solution to quantify population-level impacts, but are rare. We used Bayesian inference to examine changes in chick survival, body condition and population growth rate of endangered African penguins Spheniscus demersus in response to eight years of alternating time-area closures around two pairs of colonies. Our results demonstrate that fishing closures improved chick survival and condition, after controlling for changing prey availability. However, this effect was inconsistent across sites and years, highlighting the difficultly of assessing management interventions in marine ecosystems. Nevertheless, modelled increases in population growth rates exceeded 1% at one colony; i.e. the threshold considered biologically meaningful by fisheries management in South Africa. Fishing closures evidently can improve the population trend of a forage-fish dependent predator – we therefore recommend they continue in South Africa and support their application elsewhere. However, detecting demographic gains for mobile marine predators from small no-take zones requires experimental time-frames and scales that will often exceed those desired by decision-makers.

全球饵料鱼（forage fish）捕捞量持续增长，对海洋生态系统可能造成严重后果。饵料鱼捕食者或受此类捕捞活动的影响，但此类影响的具体机制尚存争议。通过渔业操控实验可量化其对种群水平的影响，这是最佳研究方案，但此类实验较为罕见。针对两处成对聚居地周边实施的8年轮换式时间-区域禁渔措施（time-area closures），我们采用贝叶斯推断（Bayesian inference）方法，分析了濒危非洲企鹅（Spheniscus demersus）的雏鸟存活率、身体状况与种群增长率的变化情况。研究结果显示，在控制猎物可获得性变化的干扰后，禁渔措施可提升雏鸟存活率与身体状况。但该效应在不同研究地点与年份间并不一致，凸显了评估海洋生态系统管理干预措施的难度。不过，在其中一处聚居地，模型模拟的种群增长率提升幅度超过了1%——这一阈值被南非渔业管理部门认定为具备生物学意义的临界值。显然，禁渔措施可改善依赖饵料鱼的捕食者的种群发展趋势，因此我们建议南非继续推行此类措施，并支持在其他地区推广应用。但要从小型禁渔区中检测到移动性海洋捕食者的种群统计学收益，所需的实验周期与实验规模往往超出决策者的预期。