A Global Eddy Splitting and Merging Trajectory Dataset Based on Satellite Altimetry Utilizing Eddygroup, Eddytree and Eddygraph

Developed from satellite altimetry spanning 1993 – 2023 with enhanced algorithms, this suite of products provides global records of oceanic eddies and eddygroups, the topological relationships between them, and objectively tracked eddy split-and-merge events generated from the identification results using an objective tracking algorithm. The global mesoscale eddy and eddygroup dataset contains identification data for eddies and eddygroups from January 1993 to December 2023. Using Absolute Dynamic Topography (ADT) data from the satellite-altimetry product SEALEVEL_GLO_PHY_L4_MY_008_047, we apply a sea-surface-height-anomaly–based Eulerian algorithm to identify mesoscale eddies and eddygroups on a daily basis. After identification, a multi-level topological structure between eddies and eddygroups is established to generate a global mesoscale eddy and eddygroup identification data product based on satellite altimeters.Based on the eddy and eddygroup identification data, the particle drift algorithm is adopted for the objective splitting and merging tracking of eddies, and the global mesoscale eddy objective splitting and merging tracking data product is obtained.
This dataset comprises 11,353 JSON files (11,322 identification files and one tracking file per year for 31 years), utilising an azimuthal equal-area projection. The file naming convention is as follows: (1) XXXX (compressed folder) where XXXX denotes the year. Each folder contains 365 or 366 daily eddy and eddy group identification records, named as: eddy_grouptree_XXXXXXXX.json; (2) JSON files represent tracking trajectory data, named as: xxxxxxxx_yyyyyyyy_net_tracks_.json. For detailed usage instructions, please refer to the ‘Data Usage Guide’ within the download section.(3) merge7_an_cy_parent.json and split7_an_cy_parent.json contain data on typical splitting and merging events, extracting all occurrences between 1993 and 2023 where a seven-day period immediately preceding or following the event did not involve splitting or merging with other vortices.

本数据集基于1993年至2023年的卫星测高（satellite altimetry）数据，结合改进算法构建，该系列产品提供全球海洋中尺度涡旋（mesoscale eddies）与涡旋群（eddygroups）的时空记录、二者间的拓扑关系，以及基于客观追踪算法从识别结果中生成的涡旋客观分裂与合并事件记录。

全球中尺度涡旋与涡旋群数据集涵盖1993年1月至2023年12月的涡旋及涡旋群识别数据。本数据集采用卫星测高产品SEALEVEL_GLO_PHY_L4_MY_008_047提供的绝对动态地形（Absolute Dynamic Topography, ADT）数据，基于海面高度异常（sea-surface-height-anomaly）构建欧拉（Eulerian）算法，逐日识别中尺度涡旋与涡旋群。完成识别后，我们建立涡旋与涡旋群间的多层级拓扑结构，生成基于卫星测高数据的全球中尺度涡旋与涡旋群识别数据产品。基于上述涡旋与涡旋群识别数据，本数据集采用粒子漂移算法实现涡旋的客观分裂与合并追踪，进而得到全球中尺度涡旋客观分裂与合并追踪数据产品。

本数据集共包含11353个JSON文件（31年中，每年含11322个识别文件与1个追踪文件），采用方位等面积投影。文件命名规则如下：
(1) 以年份命名的压缩文件夹（XXXX），其中XXXX代表年份，每个文件夹内包含365或366条每日涡旋与涡旋群识别记录，文件名为eddy_grouptree_XXXXXXXX.json；
(2) 代表追踪轨迹数据的JSON文件，命名格式为xxxxxxxx_yyyyyyyy_net_tracks_.json；
(3) merge7_an_cy_parent.json与split7_an_cy_parent.json存储典型分裂与合并事件数据，提取了1993年至2023年间所有事件前后7天内未与其他涡旋发生分裂或合并的案例。如需详细使用说明，请参阅下载区域内的《数据使用指南》。