Using machine learning models to predict the distribution of a cryptic marine species: the sperm whale

Implementation of effective conservation planning relies on a robust understanding of the spatio-temporal distribution of the target species. In the marine realm, this is even more challenging for cryptic species with extreme diving behaviour like the sperm whales. Our study aims at investigating the movements and predicting suitable habitat maps for this species in the Mascarene Archipelago in the South-West Indian Ocean. Using 21 satellite tracks of sperm whale and 8 environmental predictors, 14 supervised machine learning algorithms were tested and compared to predict the whales’ distribution during the wet and dry season, separately. Fourteen of the whales remained in close proximity to Mauritius while a migratory pattern was evidenced with a synchronized departure for 8 females that headed towards Rodrigues Island. The best performing algorithm was the random forest, showing a strong affinity for Sea Surface Height during the wet season and for bottom temperature during the dry sea...

高效的保护规划实施，有赖于对目标物种时空分布的扎实认知。在海洋环境中，对于诸如抹香鲸这类具备极端潜水行为的隐秘物种而言，这项任务的难度更甚。本研究旨在探究西南印度洋马斯克林群岛海域该物种的活动规律，并预测其适宜生境分布图。本研究利用21条抹香鲸卫星追踪数据与8项环境预测因子，测试并对比了14种监督式机器学习算法，以分别预测湿季与干季的鲸类分布。其中14头抹香鲸始终活动于毛里求斯附近海域，而研究还证实了该种群存在迁徙模式：8头雌性抹香鲸同步启程，前往罗德里格斯岛。表现最优的算法为随机森林（Random Forest），该算法在湿季对海面高度（Sea Surface Height）表现出极强的关联性，在干季则与底层水温高度相关……