With the increasing amount of spatial-temporal~(ST) ocean data, numerous spatial-temporal data mining (STDM) studies have been conducted to address various oceanic issues, e.g., climate forecasting and disaster warning. Compared with typical ST data (e.g., traffic data), ST ocean data is more complicated with some unique characteristics, e.g., diverse regionality and high sparsity. These characteristics make it difficult to design and train STDM models. Unfortunately, an overview of these studies is still missing, hindering computer scientists to identify the research issues in ocean while discouraging researchers in ocean science from applying advanced STDM techniques. To remedy this situation, we provide a comprehensive survey to summarize existing STDM studies in ocean. Concretely, we first summarize the widely-used ST ocean datasets and identify their unique characteristics. Then, typical ST ocean data quality enhancement techniques are discussed. Next, we classify existing STDM studies for ocean into four types of tasks, i.e., prediction, event detection, pattern mining, and anomaly detection, and elaborate the techniques for these tasks. Finally, promising research opportunities are highlighted. This survey will help scientists from the fields of both computer science and ocean science have a better understanding of the fundamental concepts, key techniques, and open challenges of STDM in ocean.

翻译：随着海洋时空数据的日益增多，大量时空数据挖掘研究已被用于解决各类海洋问题，例如气候预测和灾害预警。与典型的时空数据（如交通数据）相比，海洋时空数据因其独特的区域多样性和高稀疏性等特点而更为复杂。这些特性使得设计和训练时空数据挖掘模型变得困难。然而，目前尚缺乏对这些研究的系统综述，这既阻碍了计算机科学家识别海洋领域的研究问题，也降低了海洋科学研究者应用先进时空数据挖掘技术的积极性。为弥补这一空白，我们提供了一份全面的综述，总结了海洋领域的现有时空数据挖掘研究。具体而言，我们首先归纳了广泛使用的海洋时空数据集并识别其独特特性；随后讨论了典型的海洋时空数据质量增强技术；接着将现有海洋时空数据挖掘研究分为四类任务——预测、事件检测、模式挖掘和异常检测，并详细阐述了这些任务的技术方法；最后强调了具有前景的研究方向。本综述将帮助计算机科学和海洋科学领域的研究者更好地理解海洋时空数据挖掘的基本概念、关键技术和开放挑战。