N.A. Voennyy1, A.B. Domracheva2, K.V. Domrachev3, D.A. Okutin4
1–2, 4 Bauman Moscow State Technical University (Moscow, Russia)
3 Lomonosov Moscow State University (Moscow, Russia)
1 voennyyyna@student.bmstu.ru, 2 domracheva@bmstu.ru, 3 domrachevkv@my.msu.ru, 4 okutinda@student.bmstu.ru

Current global maps of whale migration routes contain no data on whale movements in Russian coastal waters. This information is particularly important for Russia, as it relates to two key objectives: the conservation of humpback whales and other whale species and maintaining aboriginal subsistence whaling practices vital for several Russian ethnic groups. The aim of this study was to develop an information system for studying humpback whale migration routes, with a focus on Russian coastal waters.

Data on humpback whale sightings over the past decade were gathered from two open sources: the global HappyWhale photo-identification database, which holds over 230,000 records, and a printed catalogue compiled by researchers from Lomonosov Moscow State University covering Russian coastal waters between 2017 and 2023. All data were stored in a purpose-built SQLite database. The records include the geographic coordinates of each sighting and the date and time of observation.

An application was developed to reconstruct the migration routes of individual humpback whales, whole populations, and researcher-defined groups. Individual whales were assigned to groups using cluster analysis (k-means). The application was built in Python, Go, and TypeScript. Four trajectory interpolation methods were compared for both individual animals and whole populations or selected groups: quadratic Bezier curve, cubic Bezier curve, B-spline, and Catmull–Rom spline.

A machine learning microservice based on the EfficientNet-B4 architecture was developed to detect humpback whales in photographs, achieving an accuracy of 0.95, an F1-score of 0.95, and a ROC AUC of 0.97.

Several technical challenges were also addressed, including the antimeridian problem and the automated detection of migration route intersections with land areas.

The system can be extended to cover the entire World Ocean and adapted to include other whale species as data become available. This will enable accurate mapping of migration corridors for these remarkable animals and help maintain the aboriginal subsistence whaling practices of several Russian ethnic groups.

Voennyy N.A., Domracheva A.B., Domrachev K.V., Okutin D.A. Development of an information system for humpback whale migration analysis // Neurocomputers. 2026. V. 28. № 3. P. 65–75. DOI: https://doi.org/10.18127/j19998554-202603-08.

1. Cooke J.G. Megaptera novaeangliae. The IUCN Red List of Threatened Species 2018. DOI: 10.2305/IUCN.UK.2018-2.RLTS.T13006
   A50362794.en.
2. Johnson C.M., Reisinger R.R., Palacios D.M. et al. Protecting Blue Corridors: Challenges and Solutions for Migratory Whales Navigating National and International Seas. WWF International. 2022. DOI: 10.5281/zenodo.6196131.
3. Palacios D.M., Irvine L.M., Lagerquist B.A. et al. A satellite-linked tag for the long-term monitoring of diving behavior in large whales. Animal Biotelemetry. 2022. V. 10. № 26. DOI: 10.1186/s40317-022-00297-9.
4. Titova O. Gorbaty`e kity` Chukotki: fotokatalog. Tula: Borus-Print. 2024. 112 s. (in Russian).
5. Baza danny`x OBIS-SEAMAP / provajder danny`x: HappyWhale. URL: https://seamap.env.duke.edu/dataset/list.
6. Tremblay Y., Shaffer S.A., Fowler S.L. et al. Interpolation of animal tracking data in a fluid environment. Journal of Experimental Biology. 2006. V. 209. № 1. P. 128–140. DOI: 10.1242/jeb.01970.
7. Natural Earth. URL: https://www.naturalearthdata.com/downloads.
8. Ikotun A.M., Ezugwu A.E., Abualigah L., Abuhaija B., Heming J. K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Information Sciences. 2023. V. 622. P. 178–210. DOI: 10.1016/j.ins.2022.11.139.
9. Tan M., Le Q.V. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning. PMLR. 2019. V. 97. P. 6105–6114.
10. Martin R.S. Chistaya arxitektura: iskusstvo razrabotki programmnogo obespecheniya. Per. s angl. SPb.: Piter. 2018. 352 s. (in Russian).
11. Voenny`j N.A., Domracheva A.B. Postroenie cifrovoj karty` migracionny`x putej gorbaty`x kitov. Sb. tezisov III Vserossijskogo kongressa s mezhdunarodny`m uchastiem. Russkij inzhener. 2025. S. 200–201. (in Russian).