Metapopulation and source–sink dynamics are increasingly considered within spatially explicit management of wildlife populations, yet the application of these concepts has generally been limited to comparisons of the performance (e.g., demographic rates or dispersal) inside vs. outside protected areas, and at spatial scales that do not encompass an entire metapopulation. In the present study, a spatially explicit, size-structured matrix model was applied to simulate the dynamics of an Eastern oyster (Crassostrea virginica) metapopulation in the second largest estuary in the United States—the Albemarle-Pamlico Estuarine System in North Carolina. The model integrated larval dispersal simulations with empirical measures of oyster demographic rates to simulate the dynamics of the entire oyster metapopulation consisting of 646 reefs and five reef types: (1) restored subtidal reefs closed to harvest (i.e., sanctuaries or protected areas; n = 14), (2) restored subtidal reefs open to harvest (n = 53), (3) natural subtidal reefs open to harvest (n = 301), (4) natural intertidal reefs open to harvest (n = 129), and (5) oyster reefs on manmade, hard structures such as seawalls (n = 149). Key findings included (1) an overall stable, yet slightly declining oyster metapopulation, (2) variable reef type-specific population trajectories, largely dependent on spatiotemporal variation in larval recruitment, (3) a greater relative importance of inter-reef larval connectivity on metapopulation dynamics than local larval retention processes, and (4) spatiotemporal variation in the source–sink status of reef subpopulations wherein subtidal sanctuaries and reefs located in the northeastern portion of the estuary were frequent sources. From a management perspective, continued protection of oyster sanctuaries is warranted. Sanctuaries represented only 6.2% of the total reef area, however, they harbored 19% (± 2%) of all oysters and produced 25% (± 6%) of all larvae settling within the metapopulation. Additional management priorities should focus on restoration or conservation of subpopulations that serve as frequent source subpopulations (including those with poor demographic rates, but high connectivity potential), and management of harvest from sink subpopulations. The application of a source–sink framework and similar integrated modeling approach could inform management of oysters in other systems, as well as other species that exhibit similar metapopulation characteristics.

中文翻译：

---

牡蛎的集合种群动态：来源、汇以及对保护和恢复的影响

在野生动物种群的空间明确管理中越来越多地考虑元种群和源汇动态，但这些概念的应用通常仅限于比较保护区内与保护区外的表现（例如，人口率或扩散），以及在空间不包含整个元群的尺度。在本研究中，应用空间明确的、大小结构的矩阵模型来模拟东部牡蛎（Crassostrea virginica) 美国第二大河口——北卡罗来纳州的 Albemarle-Pamlico 河口系统的元种群。该模型将幼虫扩散模拟与牡蛎人口统计率的经验测量相结合，以模拟由 646 个珊瑚礁和五种珊瑚礁类型组成的整个牡蛎综合种群的动态：(1) 恢复捕捞的潮下珊瑚礁（即保护区或保护区；n  = 14), (2) 恢复的潮下礁开放采伐 ( n  = 53), (3) 天然潮下礁开放采伐 ( n  = 301), (4) 天然潮间礁开放采伐 ( n  = 129), 和 ( 5) 人造硬质结构上的牡蛎礁，例如海堤 ( n = 149）。主要发现包括（1）总体稳定但略有下降的牡蛎种群，（2）不同的珊瑚礁类型特定的种群轨迹，主要取决于幼虫招募的时空变化，（3）珊瑚礁间幼虫连通性的相对重要性更大种群动态比局部幼虫滞留过程，以及（4）珊瑚礁亚种群源汇状态的时空变化，其中位于河口东北部的潮下保护区和珊瑚礁是常见的来源。从管理的角度来看，有必要继续保护牡蛎保护区。保护区仅占珊瑚礁总面积的 6.2%，然而，它们栖息着 19% (± 2%) 的所有牡蛎，并产生了 25% (± 6%) 的幼虫在复合种群中定居。其他管理优先事项应侧重于恢复或保护作为频繁来源亚种群的亚种群（包括那些人口率低但连通性潜力高的亚种群），以及对汇亚种群的收获进行管理。源汇框架和类似的综合建模方法的应用可以为其他系统中的牡蛎管理以及表现出类似集合种群特征的其他物种提供信息。