Fish observation and length data derived from stereo-baited remote underwater videos collected around Saint Lucia in 2019

Analysed stereo-Baited Remote Underwater Videos (BRUVs) from the 2019 survey in the north west of St. Lucia. In total 87 1-hour BRUV stations were successfully deployed to a maximum depth of 100m. Video classification assessed fish MaxN (number of visible individuals in a single frame) by species and measured all fish lengths at MaxN. This data holding contains: The exported files from Eventmeasure for all 87 stations (3DPoints, ImagePtPair, Info, Lengths, MovieSeq, Period, Points and Source); Metadata for each deployment as a csv file (including station summary statistic of: Number of taxa, total MaxN, diversity metrics, biomass and cluster groups; and a GIS shapefile containing the same summary information. Data were collected as part of the Commonwealth Marine Economies Programme in collaboration with the St. Lucia Department of Fisheries. The survey was conducted in accordance with the standardised field and video annotation guide (DOI: 10.1111/2041-210X.13470). Videos were classified by either of two analysts with 10% duplicated for quality control and to ensure consistency. Errors or issues raised were addressed for all stations. Footage were analysed within the SeaGIS Eventmeasure software, classifying all fish species based on MaxN (number of visible individuals in a single frame), and measuring all individual fish at MaxN.

本数据集涵盖2019年圣卢西亚西北部调查所获取并经分析的立体诱饵式远程水下视频（stereo-Baited Remote Underwater Videos, BRUVs）。本次调查总计成功布设87个时长1小时的BRUV站位，最大布设水深达100米。
视频分类环节按物种统计了鱼类MaxN（单帧可见个体数），并在MaxN时刻测量了所有鱼类的体长。本数据集包含以下内容：针对全部87个站位导出的Eventmeasure软件输出文件（包括3DPoints、ImagePtPair、Info、Lengths、MovieSeq、Period、Points及Source子文件）；各布设站位的元数据CSV文件（包含站位汇总统计信息：类群数量、总MaxN、多样性指标、生物量及聚类群组），同时附带包含相同汇总信息的GIS形状文件（shapefile）。
本数据为英联邦海洋经济计划（Commonwealth Marine Economies Programme）与圣卢西亚渔业部门合作开展的调查成果。本次调查严格遵循标准化野外作业与视频标注指南（DOI: 10.1111/2041-210X.13470）实施。视频分类工作由两名分析人员分别完成，其中10%的视频进行重复标注以开展质量控制并确保标注一致性。所有站位发现的错误与问题均已完成修正。所有影像均通过SeaGIS Eventmeasure软件进行分析，基于MaxN（单帧可见个体数）完成全部鱼类物种分类，并在MaxN时刻测量所有个体的体长。