Automated under-ice monitoring system for fish detection in Central Arctic Ocean

Abstract

The Arctic Ocean ecosystem is undergoing dramatic changes as the sea ice cover retreats, underscoring the need for technologies that can monitor biological responses in the upper ocean, particularly beneath sea ice, where traditional ship-based investigations are severely constrained. Fish such as polar cod, which depend strongly on under-ice habitats, are widely regarded as key indicator species of Arctic warming, making their observation crucial for understanding ecosystem changes. In this study, we developed and deployed an integrated, fully automated observation system for long-term fish monitoring under the Arctic sea ice. The system combines a custom-designed underwater multi-focal automatic camera system (UMACS), a robust computing platform, and a satellite communication module to realize an autonomous ‘detection-to-transmission’ workflow under extreme polar conditions. A deep learning-based detection model trained on a multi-source dataset was implemented to address challenges such as low illumination, turbidity, and blurred backgrounds. A three-month continuous deployment in the Central Arctic Ocean demonstrated the robust engineering performance of the system under realistic field conditions. Although no fish were unambiguously confirmed, highlighting the intrinsic difficulty of discriminating small biological targets against a pure water background in this region, the system successfully achieved persistent, unattended, under-ice visual observation with data return. Therefore, this study provides a practical and transferable engineering framework for the scalable, technology-driven ecological monitoring of one of the planet’s most remote and fragile marine environments.

Keywords

Arctic / Under-ice monitoring / Fish detection / YOLOv5 / Automated system / Unmanned underwater camera

References

[1] Andriashev AP, Mukhomediyarov BF, Pavshtiks EA (1980) On mass congregations of the cryopelagic cod fishes (Boreogadus saida and Arctogadus glacialis) in circumpolar Arctic Basins. In: Vinogradov ME, Melnikov IA (eds) Biology of the central Arctic Basin. Nauka Science Press, pp 196–211

[2] Boldt JL, Williams K, Rooper CN, Towler RH, Gauthier S. Development of stereo camera methodologies to improve pelagic fish biomass estimates and inform ecosystem management in marine waters. Fish Res. 2018, 198: 66-77.

[3] Chen HN, Liu HY, Sun T, Lou HT, Duan XH, Bi LYet al. . MC-YOLOv5: a multi-class small object detection algorithm. Biomimetics. 2023, 8(4): 342.

[4] Cheng T, Wang W. MF-ID: a benchmark and approach for multi-category fine-grained intrusion detection. IEEE Trans Autom Sci Eng. 2025, 22: 3582-3597.

[5] Choi JG, Matsuoka A, Manizza M, Carroll D, Dutkiewicz S, Lippmann T. A new ecosystem model for Arctic phytoplankton phenology from ice-covered to open-water periods: implications for future sea ice retreat scenarios. Geophys Res Lett. 2024, 5119e2024GL110155.

[6] Deng LJ, Jin BW, Quan MY, Wang AM, Fan WJ, Wang H. Analysis of changes in Arctic sea ice and its influencing factors during 1979–2022. Mar Sci Bull. 2024, 2621-17

[7] Deng LX, Bi LY, Li HQ, Chen HN, Duan XH, Lou HTet al. . Lightweight aerial image object detection algorithm based on improved YOLOv5s. Sci Rep. 2023, 13: 7817.

[8] Douglas RH, Hawryshyn CW. Douglas RH, Djamgoz M. Behavioural studies of fish vision: an analysis of visual capabilities. The visual system of fish. 1990, London, Chapman and Hall373-418.

[9] Durant JM, Dupont N, Ono K, Langangen Ø. Interaction between three key species in the sea ice-reduced Arctic Barents Sea system. Proc R Soc Lond B Biol Sci. 2024, 291203220241408.

[10] Durve M, Orsini S, Tiribocchi A, Montessori A, Tucny JM, Lauricella M et al (2023) Benchmarking YOLOv5 and YOLOv7 models with DeepSORT for droplet tracking applications. Preprint at arXiv:2301.08189. https://doi.org/10.48550/arXiv.2301.08189

[11] Edwards JE, Hedges KJ, Kessel ST, Hussey NE. Multi-year acoustic tracking reveals transient movements, recurring hotspots, and apparent seasonality in the coastal-offshore presence of Greenland sharks (Somniosus microcephalus). Front Mar Sci. 2022, 9: 902854.

[12] Gallagher AJ, Alsudairy NA, Shea BD, Payne NL, Duarte CM. First application of 360-degree camera technology to marine predator bio-logging. Front Mar Sci. 2021, 8707376.

[13] Golikov AV, Stauffer JB, Schindler SV, Taylor J, Boehringer L, Purser A et al (2023) Video observations of a bathypelagic cirrate octopod Cirroteuthis muelleri. PANGAEA. https://doi.org/10.1594/PANGAEA.957196

[14] He ZJ, Wang K, Fang T, Su L, Chen R, Fei XH (2024) Comprehensive performance evaluation of YOLOv11, YOLOv10, YOLOv9, YOLOv8 and YOLOv5 on object detection of power equipment. In: 2025 37th Chinese Control and Decision Conference (CCDC). IEEE, pp 1281–1286. https://xplorestaging.ieee.org/document/11090973/citations#citations

[15] Henson HC, Puts IC, Sejr MK, Sørense JL, Holding JM. Glacial meltwater increases coastal carbon dioxide uptake and sensitivity to biogeochemical change. Commun Earth Environ. 2025, 6687.

[16] Hermann NT, Hammer LJ, Hussey NE, Marcoux M, Hedges KJ, Walter RPet al. . Year-round monitoring of Arctic species of sculpin to identify residency and seasonality of movement behavior. Can J Fish Aquat Sci. 2023, 80111798-1812.

[17] Ishihara M, Matsuno K, Tokuhiro K, Ando Y, Sato K, Yamaguchi A. Geographic variation in population structure and grazing features of Calanus glacialis/marshallae in the Pacific Arctic Ocean. Front Mar Sci. 2023, 10: 1168015.

[18] Jiang YQ, Xu JX, Ye L, Cao J, Li YM, Hou DD (2023) AUV applications and technologies of deep sea exploration in the Arctic: a review. In: Yang D (ed) METSD 2023. Lecture notes in civil engineering, vol 375. Springer, Singapore, pp 372–384. https://doi.org/10.1007/978-981-99-4291-6_28

[19] Kim J, Cho J. Exploring a multimodal mixture-of-YOLOs framework for advanced real-time object detection. Appl Sci-Basel. 2020, 102612.

[20] Li WT, Zhao DP, Yuan B, Gao Y, Shi ZW. PETDet: proposal enhancement for two-stage fine-grained object detection. IEEE Trans Geosci Remote Sens. 2024, 625602214.

[21] Lin YH, Zhang YY, Yang DS, Zhou YKP, Wang LT. DLFE-YOLO: an enhanced framework for underwater object detection based on YOLOv8. Intell Mar Technol Syst. 2025, 337.

[22] Liu XP, Zhao KK, Liu C, Chen L. Bi2F-YOLO: a novel framework for underwater object detection based on YOLOv7. Intelligent Marine Technology and Systems. 2025, 39.

[23] Mou JF, He XB, Liu K, Huang YQ, Zhang SY, Zu YCet al. . Benthic biodiversity by baited camera observations on the Cosmonaut Sea shelf of East Antarctica. Diversity-Basel. 2024, 165277.

[24] Snoeijs-Leijonmalm P, Flores H, Sakinan S, Hildebrandt N, Svenson A, Castellani Get al. . Unexpected fish and squid in the central Arctic deep scattering layer. Sci Adv. 2022, 87eabj7536.

[25] Walters V. Winter abundance of Arctogadus glacialis in the Polar Basin. Copeia. 1961, 2236-237.

[26] Wang S, Lu N, Zheng H, Li H, Peng JG, Zhang Tet al. . Wildlife video object detection based on deep learning. Forest Grassl Resour Res. 2024, 1: 34-40

[27] Wang ZY, Men SJ, Bai YT, Yuan YT, Wang JM, Wang KLet al. . Improved small object detection algorithm CRL-YOLOv5. Sensors. 2024, 24196437.

[28] Yang ZY, Lan X, Wang H. Comparative analysis of YOLO series algorithms for UAV-based highway distress inspection: performance and application insights. Sensors. 2025, 2551475.

Funding

Key Technologies Research and Development Program(2022YFC2807603)

National Natural Science Foundation of China(42576286)

Shanghai Jiao Tong University(SL2020ZD206)

RIGHTS & PERMISSIONS

The Author(s)