References

hydrophysics

Фундаментальная и прикладная гидрофизика

Fundamental and Applied Hydrophysics

2073-66732782-5221

St. Petersburg Research Center of the Russian Academy of Sciences

10.59887/2073-6673.2026.19(2)-9

tsnwkg

hydrophysics-1557

Research Article

ТЕХНИЧЕСКАЯ ГИДРОФИЗИКА

TECHNICAL HYDROPHYSICS

Оценка эффективности моделей машинного обучения для прогнозирования выбросов морского мусора на берег: использование автономной видеокамеры и данных реанализа

Evaluating the performance of machine learning models for prediction of marine litter beaching: using autonomous video camera and reanalysis data

https://orcid.org/0000-0002-6471-3646

Фетисов

С. В.

Fetisov

S. V.

117997, Москва, Нахимовский проспект, д. 36

Sergei V. Fetisov, Researcher, The Atlantic Branch of the IO RAS

36, Nakhimovsky Prosp., Moscow, 117997

fetisov.sv@atlantic.ocean.ru

https://orcid.org/0000-0003-3876-3022

Чубаренко

И. П.

Chubarenko

I. P.

117997, Москва, Нахимовский проспект, д. 36

Irina P. Chubarenko , Head Researcher, The Atlantic Branch of the IO RAS

36, Nakhimovsky Prosp., Moscow, 117997

irina_chubarenko@mail.ru

Институт океанологии им. П.П. Ширшова РАНРоссияShirshov Institute of Oceanology, Russian Academy of SciencesRussian Federation

2026

10082026

192102116

2026

Фетисов С.В., Чубаренко И.П.

Fetisov S.V., Chubarenko I.P.

Данная работа распространяется под лицензией Creative Commons Attribution 4.0.

This work is licensed under a Creative Commons Attribution 4.0 License.

https://hydrophysics.spbrc.ru/jour/article/view/1557

Выбросы на берег крупных пятен органического и антропогенного мусора после штормовых событий — явление, которое часто наблюдается вдоль морской береговой линии. В таких выбросах все больше наблюдается пластиковый мусор, что представляет значительную угрозу для прибрежных экосистем. Непрерывная видеосъемка длительностью 22 месяца, сделанная с помощью автономной стационарной камеры, позволила наблюдать выбросы на северном побережье Самбийского полуострова (Балтийское море). Анализировались гидрофизические и метеорологические параметры из данных реанализа для прогнозирования выбросов морского мусора с использованием моделей машинного обучения. Была оценена производительность нескольких моделей машинного обучения для прогнозирования времени, в которое морской мусор будет выброшен на берег. Сравнивалась точность моделей искусственной нейронной сети, классификатора случайного леса и классификатора градиентного усиления. Модель классификатора случайного леса (83,4 ± 7,6 % для метрики F1) и модель искусственной нейронной сети (81,7 ± 3,9 % для метрики F1) представляются наиболее эффективными моделями для прогнозирования выброса морского мусора на берег. Уровень моря, направление и крутизна волны оказались наиболее значимыми параметрами в рассматриваемых моделях машинного обучения.

The beaching of large patches of natural and anthropogenic debris following storm events is a phenomenon observed along numerous marine shorelines. Today, such wash-outs are becoming increasingly contaminated with plastic litter, which poses a significant threat to coastal ecosystems. A 22-month continuous video sequence, recorded by an autonomous stationary camera, allowed the observation the wash-outs on the northern shore of the Sambian Peninsula (the Baltic Sea). Hydrophysical and meteorological variables from reanalysis data are analyzed to predict the timing of marine litter beaching using machine learning models. The performance of multiple machine learning models was evaluated to assess their ability to predict the time at which marine litter would be beached on the shore. The accuracy of artificial neural network, random forest classifier and gradient boosting classifier of machine learning models are compared. The random forest classifier model (83.4 ± 7.6 % for F1-score) and the artificial neural network model (81.7 ± 3.9 % for F1-score) appear to be the most efficient models for predicting the wash-out formation and beaching. Sea level, wave direction and steepness are the most significant parameters in training models.

морской мусорискусственный интеллектнейронные сетимашинное обучениеБалтийское мореавтономная камераметеорологические и гидрофизические параметры

marine litterartificial intelligenceneural networksmachine learningBaltic Seaautonomous camerameteorological and hydrophysical conditions

The research was funded by the Russian Science Foundation (grant No. 24-17-00099, https://rscf.ru/project/24-17-00099/).

References1

Najafzadeh M, Basirian S, Li Z. Vulnerability of the rip current phenomenon in marine environments using machine learning models. Results in Engineering. 2024;21:101704. https://doi.org/10.1016/j.rineng.2023.101704

Paul A, Maheswaran PA, Satheesan K, Kottayil A, Harikrishnan M. A comparative assessment of different machine learning algorithms for estimating near realistic salinity in the North Indian Ocean. Ocean Modelling. 2025;197:102561. https://doi.org/10.1016/j.ocemod.2025.102561

Solomatine DP, Siek MB. Modular learning models in forecasting natural phenomena. Neural Networks. 2006;19(2):215– 224. https://doi.org/10.1016/j.neunet.2006.01.008

Xu G, Xie W, Lin X, et al. Detection of three-dimensional structures of oceanic eddies using artificial intelligence. Ocean Modelling. 2024;190:102385. https://doi.org/10.1016/j.ocemod.2024.102385

Berenjkoub M, Chen G, Günther T. Vortex boundary identification using convolutional neural network. In: 2020 IEEE Visualization Conference (VIS). 2020 Oct 25–30; Salt Lake City, UT, USA. p. 261–265. https://doi.org/10.1109/VIS47514.2020.00059

Krinitskiy M, Verezemskaya P, Elizarov S, Gulev S. Machine learning methods for the detection of polar lows in satellite mosaics: major issues and their solutions. arXiv. Preprint posted online 2020 Nov 9. arXiv:2011.04811. http://arxiv.org/abs/2011.04811 (Accessed 08.07.2023).

Kuma P, Bender FAM, Schuddeboom A, McDonald AJ, Seland Ø. Machine learning of cloud types in satellite observations and climate models. Atmospheric Chemistry and Physics. 2023;23(1):523–549. https://doi.org/10.5194/acp-23-523-2023

Liu H, Sun W, Yan S. Cloud type recognition using artificial intelligence of things. In: AGU Fall Meeting Abstracts. 2019. GC43D-1356.

Mommert M. Cloud identification from all-sky camera data with machine learning. The Astronomical Journal. 2020;159(4):178. https://doi.org/10.3847/1538-3881/ab744f

Bouzegag C, Bouzid-Lagha S, Djelal N. Forecasting the upwelling phenomenon using an artificial neural network. Polish Journal of Soil Science. 2020;53(2):245–259. https://doi.org/10.17951/pjss.2020.53.2.245-259

Litta AJ, Mary Idicula S, Mohanty UC. Artificial neural network model in prediction of meteorological parameters during premonsoon thunderstorms. International Journal of Atmospheric Sciences. 2013;2013:1–14. https://doi.org/10.1155/2013/525383

United Nations Environment Programme (UNEP). Comprehensive assessment on marine litter and plastic pollution confirms need for urgent global action. Published online 2021.

Onink V, Jongedijk CE, Hoffman MJ, Van Sebille E, Laufkötter C. Global simulations of marine plastic transport show plastic trapping in coastal zones. Environmental Research Letters. 2021;16(6):064053. https://doi.org/10.1088/1748-9326/abecbd

Rosas E, Martins F, Janeiro J. Marine litter on the coast of the Algarve: main sources and distribution using a modeling approach. Journal of Marine Science and Engineering. 2021;9(4):412. https://doi.org/10.3390/jmse9040412

Granado I, Basurko OC, Rubio A, et al. Beach litter forecasting on the south-eastern coast of the Bay of Biscay: a Bayesian networks approach. Continental Shelf Research. 2019;180:14–23. https://doi.org/10.1016/j.csr.2019.04.016

Haseler M, Balciunas A, Hauk R, et al. Marine litter pollution in Baltic Sea beaches — application of the sand rake method. Frontiers in Environmental Science. 2020;8:238. https://doi.org/10.3389/fenvs.2020.599978

Prevenios M, Zeri C, Tsangaris C, Liubartseva S, Fakiris E, Papatheodorou G. Beach litter dynamics on Mediterranean coasts: distinguishing sources and pathways. Marine Pollution Bulletin. 2018;129(2):448–457. https://doi.org/10.1016/j.marpolbul.2017.10.0013

Escobar-Sánchez G, Haseler M, Oppelt N, Schernewski G. Efficiency of aerial drones for macrolitter monitoring on Baltic Sea beaches. Frontiers in Environmental Science. 2021;8:560237. https://doi.org/10.3389/fenvs.2020.560237

Franceschini S, Mattei F, D’Andrea L, et al. Rummaging through the bin: modelling marine litter distribution using artificial neural networks. Marine Pollution Bulletin. 2019;149:110580. https://doi.org/10.1016/j.marpolbul.2019.110580

Marin I, Mladenović S, Gotovac S, Zaharija G. Deep-feature-based approach to marine debris classification. Applied Sciences. 2021;11(12):5644. https://doi.org/10.3390/app11125644

Scarrica VM, Aucelli PPC, Cagnazzo C, et al. A novel beach litter analysis system based on UAV images and convolutional neural networks. Ecological Informatics. 2022;72:101875. https://doi.org/10.1016/j.ecoinf.2022.101875

Kaandorp MLA, Ypma SL, Boonstra M, Dijkstra HA, Van Sebille E. Using machine learning and beach cleanup data to explain litter quantities along the Dutch North Sea coast. Ocean Science. 2022;18(1):269–293. https://doi.org/10.5194/os-18-269-2022

Corbau C, Buoninsegni J, Olivo E, Vaccaro C, Nardin W, Simeoni U. Understanding through drone image analysis the interactions between geomorphology, vegetation and marine debris along a sandy spit. Marine Pollution Bulletin. 2023;187:114515. https://doi.org/10.1016/j.marpolbul.2022.114515

Garcia-Garin O, Monleón-Getino T, López-Brosa P, et al. Automatic detection and quantification of floating marine macro-litter in aerial images: introducing a novel deep learning approach connected to a web application in R. Environ mental Pollution. 2021;273:116490. https://doi.org/10.1016/j.envpol.2021.116490

Zaaboub N, Guebsi R, Chaouachi RS, et al. Using unmanned aerial vehicles (UAVs) and machine learning techniques for the assessment of Posidonia debris and marine (plastic) litter on coastal ecosystems. Regional Studies in Marine Science. 2023;67:103185. https://doi.org/10.1016/j.rsma.2023.103185

Domnin D, Chubarenko B, Grave A. Baseline assessment of beach cast appearance in the South-Eastern Baltic by video monitoring at a pilot site in the Kaliningrad Oblast (Russia). Marine Pollution Bulletin. 2021;173:112994. https://doi.org/10.1016/j.marpolbul.2021.112994

Franzén D, Infantes E, Gröndahl F. Beach-cast as biofertiliser in the Baltic Sea region: potential limitations due to cadmium content. Ocean & Coastal Management. 2019;169:20–26. https://doi.org/10.1016/j.ocecoaman.2018.11.015

Isachenko I, Esiukova E, Chubarenko I. Beach-cast appearance on the tide-less sea shore: parameters of favoring surface waves. Estuarine, Coastal and Shelf Science. 2023;281:108219. https://doi.org/10.1016/j.ecss.2023.108219

Ruju A, Buosi C, Coco G, et al. Ecosystem services of reed and seagrass debris on an urban Mediterranean beach (Poet to, Italy). Estuarine, Coastal and Shelf Science. 2022;271:107862. https://doi.org/10.1016/j.ecss.2022.107862

Gómez-Pujol L, Orfila A, Álvarez-Ellacuría A, Terrados J, Tintoré J. Posidonia oceanica beach-cast litter in Mediterranean beaches: a coastal videomonitoring study. Journal of Coastal Research. 2013;65(sp2):1768–1773. https://doi.org/10.2112/SI65-299.1

Pattiaratchi C, Wijeratne S. Seagrass wrack and coastal structures: lessons from South-Western Australia. Published online 2019. https://doi.org/10.18451/978-3-939230-64-9_062

Fetisov S, Esiukova E, Lobchuk O, Chubarenko I. Abundance and mass of plastic litter on sandy shore: contribution of stormy events. Marine Pollution Bulletin. 2024;207:116911. https://doi.org/10.1016/j.marpolbul.2024.116911

Von Thenen M, Von Loh M, Robbe E, et al. Beach wrack on Baltic and Mediterranean Sea beaches: litter pollution and management perspectives. Estuaries and Coasts. 2025;48(3):61. https://doi.org/10.1007/s12237-025-01497-1

Fetisov S, Chubarenko I. Marine litter stormy wash-outs: developing the neural network to predict them. Pollutants. 2021;1(3):156–168. https://doi.org/10.3390/pollutants1030013

Chubarenko B, Woelfel J, Hofmann J, et al. Converting beach wrack into a resource as a challenge for the Baltic Sea (an overview). Ocean & Coastal Management. 2021;200:105413. https://doi.org/10.1016/j.ocecoaman.2020.105413

Esiukova E. Plastic pollution on the Baltic beaches of Kaliningrad region, Russia. Marine Pollution Bulletin. 2017; 114(2):1072–1080. https://doi.org/10.1016/j.marpolbul.2016.10.001

Esiukova EE, Lobchuk OI, Volodina AA, Chubarenko IP. Marine macrophytes retain microplastics. Marine Pollution Bulletin. 2021;171:112738. https://doi.org/10.1016/j.marpolbul.2021.112738

Fetisov SV, Chubarenko IP, Esiukova EE. Analysis of the meteorological and hydrophysical factors of massive wash-outs of marine litter to the shore of Sambian Peninsula (The Baltic Sea). In: Chaplina T, editor. Processes in GeoMedia — Volume V. Cham: Springer International Publishing; 2022. p. 145–156. https://doi.org/10.1007/978-3-030-85851-3_17

Bučas M, Daunys D, Olenin S. Overgrowth patterns of the red algae Furcellaria lumbricalis at an exposed Baltic Sea coast: the results of a remote underwater video data analysis. Estuarine, Coastal and Shelf Science. 2007;75(3):308–316. https://doi.org/10.1016/j.ecss.2007.04.038

Alsina JM, Jongedijk CE, Van Sebille E. Laboratory measurements of the wave-induced motion of plastic particles: influence of wave period, plastic size and plastic density. Journal of Geophysical Research: Oceans. 2020;125(12): e2020JC016294. https://doi.org/10.1029/2020JC016294

Pärn O, Davulienė L, Macias Moy D, Vahter K, Stips A, Torsvik T. Effects of Eulerian current, Stokes drift and wind while simulating surface drifter trajectories in the Baltic Sea. Oceanologia. 2023;65(3):453–465. https://doi.org/10.1016/j.oceano.2023.02.001

Li H, Feng X, Ni X, Feng W. Beaching process of floating marine debris associated with the evolution of the nearshore wave. Marine Pollution Bulletin. 2023;197:115695. https://doi.org/10.1016/j.marpolbul.2023.115695

European Union — Copernicus Marine Service. Global ocean hourly sea surface wind and stress from scatterometer and model. Published online 2022. https://doi.org/10.48670/MOI-00305

European Union — Copernicus Marine Service. Baltic Sea wave analysis and forecast. Published online 2020. https://doi.org/10.48670/MOI-00011

European Union — Copernicus Marine Service. Baltic Sea physics analysis and forecast. Published online 2020. https://doi.org/10.48670/MOI-00010

Lee C. Calculating water wavelength using dispersion relation and approximation. Undergraduate Journal of Mathematical Modeling: One + Two. 2019;9(2). https://doi.org/10.5038/2326-3652.9.2.4902

Nguyen DT, Ngo TPN, Lou Z, Klar M, Beggel L, Brox T. Robust learning under label noise with iterative noise-filtering. arXiv. Preprint posted online 2019. https://doi.org/10.48550/ARXIV.1906.00216

Wang Z, Huang Y, Zhang Y, Jia G, Chambers J. An improved Kalman filter with adaptive estimate of latency probability. IEEE Transactions on Circuits and Systems II: Express Briefs. 2020;67(10):2259–2263. https://doi.org/10.1109/TCSII.2019.2952090

Ali SA, Ahmad A. Forecasting MSW generation using artificial neural network time series model: a study from metropolitan city. SN Applied Sciences. 2019;1(11):1338. https://doi.org/10.1007/s42452-019-1382-7

Singh D, Satija A. Prediction of municipal solid waste generation for optimum planning and management with artificial neural network — case study: Faridabad City in Haryana State (India). International Journal of System Assurance Engineering and Management. 2018;9(1):91–97. https://doi.org/10.1007/s13198-016-0484-5

Biau G, Scornet E. A random forest guided tour. TEST. 2016;25(2):197–227. https://doi.org/10.1007/s11749-016-0481-7

Scornet E, Biau G, Vert JP. Consistency of random forests. The Annals of Statistics. 2015;43(4). https://doi.org/10.1214/15-AOS1321

Liu Y, Wu S, Wu Z, Zhou S. Application of gradient boosting machine in satellite-derived bathymetry using Sentinel-2 data for accurate water depth estimation in coastal environments. Journal of Sea Research. 2024;201:102538. https://doi.org/10.1016/j.seares.2024.102538

Chen Z, Wu Q, Han S, Zhang J, Yang P, Liu X. A study on geological structure prediction based on random forest method. Artificial Intelligence in Geosciences. 2022;3:226–236. https://doi.org/10.1016/j.aiig.2023.01.004

Theerthagiri P. Predictive analysis of cardiovascular disease using gradient boosting based learning and recursive feature elimination technique. Intelligent Systems with Applications. 2022;16:200121. https://doi.org/10.1016/j.iswa.2022.200121

Kingma DP, Ba J. Adam: a method for stochastic optimization. arXiv. Preprint posted online 2014. https://doi.org/10.48550/ARXIV.1412.6980

Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn: machine learning in Python. Journal of Machine Learning Research. 2011;12(85):2825–2830.

Haibo He, Yang Bai, Garcia EA, Shutao Li. ADASYN: adaptive synthetic sampling approach for imbalanced learning. In: 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). Piscataway, NJ: IEEE; 2008. p. 1322–1328. https://doi.org/10.1109/IJCNN.2008.4633969

Afshan N, Mushtaq Z, Alamri FS, Qureshi MF, Khan NA, Siddique I. Efficient thyroid disorder identification with weighted voting ensemble of super learners by using adaptive synthetic sampling technique. AIMS Mathematics. 2023;8(10):24274–24309. https://doi.org/10.3934/math.20231238

Walker S, Khan W, Katic K, Maassen W, Zeiler W. Accuracy of different machine learning algorithms and added-value of predicting aggregated-level energy performance of commercial buildings. Energy and Buildings. 2020;209:109705. https://doi.org/10.1016/j.enbuild.2019.109705

Haarr ML, Westerveld L, Fabres J, Iversen KR, Busch KET. A novel GIS-based tool for predicting coastal litter accumulation and optimising coastal cleanup actions. Marine Pollution Bulletin. 2019;139:117–126. https://doi.org/10.1016/j.marpolbul.2018.12.025

Soomere T. Climate change and coastal processes in the Baltic Sea. In: Oxford Research Encyclopedia of Climate Science. Oxford: Oxford University Press; 2024. https://doi.org/10.1093/acrefore/9780190228620.013.897

Eelsalu M, Soomere T, Parnell K, Viška M. Attribution of alterations in coastal processes in the southern and eastern Baltic Sea to climate change-driven modifications of coastal drivers. Oceanologia. 2025;67(1). https://doi.org/10.5697/LXTZ5389

Haarr ML, Pantalos M, Hartviksen MK, Gressetvold M. Citizen science data indicate a reduction in beach litter in the Lofoten archipelago in the Norwegian Sea. Marine Pollution Bulletin. 2020;153:111000. https://doi.org/10.1016/j.marpolbul.2020.111000

The authors declare that there are no conflicts of interest present.