Marine Science Institute, University of the Philippines Diliman, Quezon City, Philippines.
PLoS One. 2019 Jul 23;14(7):e0219913. doi: 10.1371/journal.pone.0219913. eCollection 2019.
Coral reefs of the North Indo-West Pacific provide important ecosystem services to the region but are subjected to multiple local and global threats. Strengthening management measures necessitate understanding the variability of larval connectivity and bridging global connectivity models to local scales. An individual-based Lagrangian biophysical model was used to simulate connectivity between coral reefs for three organisms with different early life history characteristics: a coral (Acropora millepora), a sea urchin (Tripneustes gratilla), and a reef fish (Epinephelus sp). Connectivity metrics and reef clusters were computed from the settlement probability matrices. Fitted power law functions derived from the dispersal kernels provided relative probabilities of connection given only the distance between reefs, and demonstrated that 95% of the larvae across organisms settled within a third of their maximum settlement distances. The magnitude of the connectivity metric values of reef cells were sensitive to differences both in the type of organism and temporal variability. Seasonal variability of connections was more dominant than interannual variability. However, despite these differences, the moderate to high correlation of metrics between organisms and seasonal matrices suggest that the spatial patterns are relatively similar between reefs. A cluster analysis based on the Bray-Curtis Dissimilarity of sink and source connections synthesized the inherent variability of these multiple large connectivity matrices. Through this, similarities in regional connectivity patterns were determined at various cluster sizes depending on the scale of interest. The validity of the model is supported by 1) the simulated dispersal kernels being within the range of reported parentage analysis estimates; and, 2) the clusters that emerged reflect the dispersal barriers implied by previously published population genetics studies. The tools presented here (dispersal kernels, temporal variability maps and reef clustering) can be used to include regional patterns of connectivity into the spatial management of coral reefs.
北印度洋-西太平洋的珊瑚礁为该地区提供了重要的生态系统服务，但也受到多种本地和全球威胁的影响。加强管理措施需要了解幼虫连通性的可变性，并将全球连通性模型与本地尺度联系起来。本文使用基于个体的拉格朗日生物物理模型来模拟三种具有不同早期生活史特征的生物之间的珊瑚礁连通性：珊瑚（Acropora millepora）、海胆（Tripneustes gratilla）和珊瑚鱼（Epinephelus sp.）。从定居概率矩阵中计算连通性指标和珊瑚礁群。从扩散核导出的拟合幂律函数提供了仅根据珊瑚礁之间的距离给出连接的相对概率，并表明 95%的幼虫在其最大定居距离的三分之一内定居。珊瑚礁单元的连通性指标值的大小对生物类型和时间可变性的差异都很敏感。连接的季节性变化比年际变化更为重要。然而，尽管存在这些差异，生物之间和季节性矩阵之间的指标具有中度到高度相关性表明珊瑚礁之间的空间模式相对相似。基于源和汇连接的 Bray-Curtis 不相似性的聚类分析综合了这些多个大型连通性矩阵的固有可变性。通过这种方式，可以根据感兴趣的尺度，在不同的聚类大小下确定区域连通性模式的相似性。模型的有效性得到以下两方面的支持：1）模拟的扩散核在已报道的亲子分析估计范围内；2）出现的聚类反映了先前发表的种群遗传学研究中暗示的扩散障碍。本文提出的工具（扩散核、时间可变性图和珊瑚礁聚类）可用于将连通性的区域模式纳入珊瑚礁的空间管理。
