Deep Learning Resolves Representative Movement Patterns in a Marine Predator Species

• Published in: Applied Sciences
• Main Authors: Peng, Chengbin, Duarte, Carlos M., Costa, Daniel P., Guinet, Christophe, Harcourt, R., Hindell, Mark A., Sequeira, Ana M. M., Muelbert, Monica, Thums, Michele, Wong, Ka-Chun, Zhang, Xiangliang
• Other Authors: Biological and Environmental Sciences and Engineering (BESE) Division, Computer Science Program, Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division, Machine Intelligence & kNowledge Engineering Lab, Marine Science Program, Red Sea Research Center (RSRC), College of Information Science and Engineering, Ningbo University, Ningbo 315211, China, Ningbo Institute of Industrial Technology, Chinese Academy of Sciences, Ningbo 315201, China, Department of Ecology & Evolutionary Biology, University of California, Santa Cruz, CA 95060, USA, Centre d’Études Biologiques de Chizé, UMR 7372 CNRS-Université de La Rochelle, 79360 Villiers-en-Bois, France, Department of Biological Sciences, Macquarie University, Sydney, New South Wales 2109, Australia, Institute for Marine and Antarctic Studies, University of Tasmania, Private Bag 05, Tasmania 7001, Australia, Sydney Institute of Marine Science, 19 Chowder Bay Road, Mosman, New South Wales 2088, Australia, Instituto de Oceanografia, Caixa Postal 474, Rio Grande 96201-900, Brazil, Australian Institute of Marine Science, Indian Ocean Marine Research Centre, University of Western Australia (M096), 35 Stirling Highway, Crawley, Western Australia 6009, Australia, Department of Computer Science, City University of Hong Kong, Hong Kong, China
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: MDPI AG 2020
• Subjects: Kerguelen; Kerguelen Island; Elephant Seal; Macquarie Island; Southern Elephant Seal
• Online Access: http://hdl.handle.net/10754/661631; https://doi.org/10.3390/app9142935

The analysis of animal movement from telemetry data provides insights into how and why animals move. While traditional approaches to such analysis mostly focus on predicting animal states during movement, we describe an approach that allows us to identify representative movement patterns of different animal groups. To do this, we propose a carefully designed recurrent neural network and combine it with telemetry data for automatic feature extraction and identification of non-predefined representative patterns. In the experiment, we consider a particular marine predator species, the southern elephant seal, as an example. With our approach, we identify that the male seals in our data set share similar movement patterns when they are close to land. We identify this pattern recurring in a number of distant locations, consistent with alternative approaches from previous research. Seal data from Macquarie Island, Davis and Casey Stations were sourced from the Integrated Marine Observing System (IMOS). IMOS is a national collaborative research infrastructure, supported by the Australian Government. It is operated by a consortium of institutions as an unincorporated joint venture, with the University of Tasmania as Lead Agent. M.M. acknowledges support from CNPq. The Kerguelen Island work was supported by the CNES-TOSCA and the French Polar Institute as part of the SNO-MEMO. This research was funded by King Abdullah University of Science and Technology’s (KAUST) Sensor Innovation Initiative, the National Natural Science Foundation of China (NO. 61802372), the Qianjiang Talent Plan (NO. QJD1702031), and Natural Science Foundation of Ningbo, China (NO. 2018A610050).