Alexandra Ershova
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alexandra Ershova has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Aleksandr Sergeevich Danilov
Department of Geoecology, Saint Petersburg Mining University, Saint Petersburg 199106, Russia
2 shared publications - Elizaveta Serdiukova
Department of Geoecology, Saint Petersburg Mining University, Saint Petersburg 199106, Russia
2 shared publications Anna Trofimova
Northern Arctic Fed Univ MV Lomonosov, Arkhangelsk, Russia
1 shared publication- Danil S. Vorobiev
Biological Institute, Tomsk State University, 36 Lenin Ave., 634050 Tomsk, Russia
1 shared publication - Egor D. Vorobiev
Biological Institute, Tomsk State University, Tomsk, Russia
1 shared publication Ekaterina Vorotnichenko
Russian State Hydrometeorol Univ, PlasticLab, St Petersburg, Russia
1 shared publication
Latest Publications
- Open Access
Detection of Marine Litter on Arctic Beaches Based on Aerial Photography Using the YOLOv8s CNN
Marine litter beach accumulation is a pressing issue not only for urban coastal areas but also for remote and uninhabited regions such as the Arctic. Existing manual litter collection and identification methods remain labor-intensive. An assessment of the volume and composition of marine litter was conducted for the Russian Arctic coastlines using high-resolution airphotos. The method is based on training the You Only Look Once (YOLOv8, 2023) convolutional neural network (CNN) in the Oriented Bounding Box (OBB) configuration to recognize plastic, metal, fishing gear, wood, and other types of marine litter on images. The best detection performance was obtained for metal and plastic—large, visually contrasting litter items. The worst performance was obtained for wood and other debris, primarily due to their similarity to the natural background. Machine learning (ML) methods combined with remote sensing data represent a new high-precision tool for large-scale environmental monitoring in the Arctic.
Open AccessDetection of Marine Litter on Arctic Beaches Based on Aerial Photography Using the YOLOv8s CNN
Marine litter beach accumulation is a pressing issue not only for urban coastal areas but also for remote and uninhabited regions such as the Arctic. Existing manual litter collection and identification methods remain labor-intensive. An assessment of the volume and composition of marine litter was conducted for the Russian Arctic coastlines using high-resolution airphotos. The method is based on training the You Only Look Once (YOLOv8, 2023) convolutional neural network (CNN) in the Oriented Bounding Box (OBB) configuration to recognize plastic, metal, fishing gear, wood, and other types of marine litter on images. The best detection performance was obtained for metal and plastic—large, visually contrasting litter items. The worst performance was obtained for wood and other debris, primarily due to their similarity to the natural background. Machine learning (ML) methods combined with remote sensing data represent a new high-precision tool for large-scale environmental monitoring in the Arctic.
- Open Access
- 14 August 2026
- Published by MDPI AG in Preprints.org
Pollution Assessment of the Arctic Coastlines Based on Aerial Photography Using the YOLOv8 CNN
Pollution of coastal territories by marine litter is a pressing issue not only for large cities but also for remote and uninhabited regions such as the Arctic. At the same time, the scale of pollution is growing annually, while existing environmental monitoring methods remain labour-intensive. For the first time, an assessment of the volume and composition of marine litter was conducted for Arctic coastlines using high-resolution airphotos. The method is based on training the YOLOv8 (2023) convolutional neural network in the OBB configuration to recognise plastic, metal, fishing nets, wood, and other types of marine litter in images. Based on a comparative analysis, it was established that machine learning methods combined with remote sensing data represent a new high-precision tool for environmental monitoring, which can significantly contribute to reducing the anthropogenic pressure on the ecosystems of Arctic territories.
- 14 August 2026
- Published by MDPI AG in Preprints.org
Open AccessPollution Assessment of the Arctic Coastlines Based on Aerial Photography Using the YOLOv8 CNN
Pollution of coastal territories by marine litter is a pressing issue not only for large cities but also for remote and uninhabited regions such as the Arctic. At the same time, the scale of pollution is growing annually, while existing environmental monitoring methods remain labour-intensive. For the first time, an assessment of the volume and composition of marine litter was conducted for Arctic coastlines using high-resolution airphotos. The method is based on training the YOLOv8 (2023) convolutional neural network in the OBB configuration to recognise plastic, metal, fishing nets, wood, and other types of marine litter in images. Based on a comparative analysis, it was established that machine learning methods combined with remote sensing data represent a new high-precision tool for environmental monitoring, which can significantly contribute to reducing the anthropogenic pressure on the ecosystems of Arctic territories.
- 8 June 2024
- Published by Elsevier in Marine Pollution Bulletin
- Vol. 204, 116517
Beach litter composition, distribution patterns and annual budgets on Novaya Zemlya archipelago, Russian Arctic
- 8 June 2024
- Published by Elsevier in Marine Pollution Bulletin
- Vol. 204, 116517
Beach litter composition, distribution patterns and annual budgets on Novaya Zemlya archipelago, Russian Arctic
- Open Access
- 1 January 2024
- Published by Tomsk State University in Vestnik Тomskogo gosudarstvennogo universiteta. Khimiya
- No. 34,p. 89-105
Comparability of riverine microplastic sampling and processing techniques: intercalibration experiment for the Yenisei River
Current methodologies for microplastic detection in the environment, including surface freshwaters, are imperfect; no standardized methods for sampling and sample preparation are available. The paper discusses the issues of comparability of microplastic surveys conducted by different methods. An intercalibration experiment carried out on the Yenisei River by two research labs – Research Center ‘Microplastic Siberia’, Tomsk State University, and PlasticLab, Russian State Hydrometeorological University – is described in detail, including the laboratory protocols and QA/QC issues. Two different sampling techniques – Manta trawl and filtration pump – showed that the total microplastic content in trawl samples was significantly lower (p < 0.01) than that in pump samples (on average 30-fold lower): 2.04–4.85 and 93.0–107 items/m3, correspondingly. The problem of incomparability of the quantitative estimates obtained by different sampling methods was confirmed, suggesting their complementarity. At the same time, differences in the morphology of the detected particles sampled by different instruments suggest that river surface and subsurface layers transport different microplastics. The study showed that sampling methods but not different laboratory protocols for sample processing are primarily important for the consistency of the results of quantitative analysis of riverine microplastics, which suggests relevance of harmonization of sampling methods.
- 1 January 2024
- Published by Tomsk State University in Vestnik Тomskogo gosudarstvennogo universiteta. Khimiya
- No. 34,p. 89-105
Open AccessComparability of riverine microplastic sampling and processing techniques: intercalibration experiment for the Yenisei River
Current methodologies for microplastic detection in the environment, including surface freshwaters, are imperfect; no standardized methods for sampling and sample preparation are available. The paper discusses the issues of comparability of microplastic surveys conducted by different methods. An intercalibration experiment carried out on the Yenisei River by two research labs – Research Center ‘Microplastic Siberia’, Tomsk State University, and PlasticLab, Russian State Hydrometeorological University – is described in detail, including the laboratory protocols and QA/QC issues. Two different sampling techniques – Manta trawl and filtration pump – showed that the total microplastic content in trawl samples was significantly lower (p < 0.01) than that in pump samples (on average 30-fold lower): 2.04–4.85 and 93.0–107 items/m3, correspondingly. The problem of incomparability of the quantitative estimates obtained by different sampling methods was confirmed, suggesting their complementarity. At the same time, differences in the morphology of the detected particles sampled by different instruments suggest that river surface and subsurface layers transport different microplastics. The study showed that sampling methods but not different laboratory protocols for sample processing are primarily important for the consistency of the results of quantitative analysis of riverine microplastics, which suggests relevance of harmonization of sampling methods.