MCR LTER: Coral Reef: Coupled Natural-Human Systems: GPS Benthic Data

This dataset includes cover of benthic organisms and substrates in more than 600,000 photographic samples of Moorea’s lagoon habitats, collected between 2018 and 2021. In summer of each year swimmers towed downward facing cameras, taking 1.5 photos per second of habitats they passed over. The resulting pictures were processed through a computer vision algorithm (CoralNet), resulting in percent cover estimates in each photo of organisms such as coral and algae and substrates such as sand and rubble. These data were georeferenced based on simultaneously collecting GPS data. Data are presented here on the scale of the photograph, and are also aggregated to the scale of a minute of sampling and a transect of swimming. The latter scales are chosen to coincide with fish counting data which was simultaneously collected but is presented in a different data set (EDI data package ID: knb-lter-mcr.4013.1) doi:10.6073/pasta/ff3de88334edda65fef26de45ff26c1d These data were collected as part of CNH-L: Multiscale Dynamics of Coral Reef Fisheries: Feedbacks Between Fishing Practices, Livelihood Strategies, and Shifting Dominance of Coral and Algae (BCS-1714704) with additional support from the Moorea Coral Reef LTER (OCE- 1637396).

本数据集涵盖了2018至2021年间采集的超过60万份莫雷阿岛泻湖生境摄影样本中的底栖生物与底质盖度数据。每年夏季，研究人员会让游泳者牵引朝下安装的相机，以每秒1.5张的速率对途经的生境进行拍摄。所得影像经计算机视觉算法（CoralNet）处理后，可得到每张照片中珊瑚、藻类等生物以及沙、碎石等底质的盖度百分比估算值。这些数据基于同步采集的全球定位系统（Global Positioning System, GPS）数据完成了地理配准。本数据集以单张照片为尺度呈现数据，同时也将数据聚合至采样分钟级与游泳样线级尺度。选择后两种尺度是为了与同期采集、但存于另一数据集（EDI数据包ID: knb-lter-mcr.4013.1，DOI:10.6073/pasta/ff3de88334edda65fef26de45ff26c1d）中的鱼类计数数据保持匹配。本数据集作为CNH-L项目“珊瑚礁渔业的多尺度动态：捕捞实践、生计策略与珊瑚和藻类优势种的动态转换”（项目编号BCS-1714704）的一部分采集所得，另得到莫雷阿珊瑚礁长期生态研究（Long-Term Ecological Research, LTER）项目（项目编号OCE-1637396）的额外支持。