Litter Pollution Research Articles

Detecting floating litter in freshwater bodies with semi-supervised deep learning

Researchers and practitioners have extensively utilized supervised Deep Learning methods to quantify floating litter in rivers and canals. These methods require the availability of large amount of labeled data for training. The labeling work is expensive and laborious, resulting in small open datasets available in the field compared to the comprehensive datasets for computer vision, e.g., ImageNet. Fine-tuning models pre-trained on these larger datasets helps improve litter detection performances and reduces data requirements. Yet, the effectiveness of using features learned from generic datasets is limited in large-scale monitoring, where automated detection must adapt across different locations, environmental conditions, and sensor settings. To address this issue, we propose a two-stage semi-supervised learning method to detect floating litter based on the Swapping Assignments between multiple Views of the same image (SwAV). SwAV is a self-supervised learning approach that learns the underlying feature representation from unlabeled data. In the first stage, we used SwAV to pre-train a ResNet50 backbone architecture on about 100k unlabeled images. In the second stage, we added new layers to the pre-trained ResNet50 to create a Faster R-CNN architecture, and fine-tuned it with a limited number of labeled images (≈1.8k images with 2.6k annotated litter items). We developed and validated our semi-supervised floating litter detection methodology for images collected in canals and waterways of Delft (the Netherlands) and Jakarta (Indonesia). We tested for out-of-domain generalization performances in a zero-shot fashion using additional data from Ho Chi Minh City (Vietnam), Amsterdam and Groningen (the Netherlands). We benchmarked our results against the same Faster R-CNN architecture trained via supervised learning alone by fine-tuning ImageNet pre-trained weights. The findings indicate that the semi-supervised learning method matches or surpasses the supervised learning benchmark when tested on new images from the same training locations. We measured better performances when little data (≈200 images with about 300 annotated litter items) is available for fine-tuning and with respect to reducing false positive predictions. More importantly, the proposed approach demonstrates clear superiority for generalization on the unseen locations, with improvements in average precision of up to 12.7%. We attribute this superior performance to the more effective high-level feature extraction from SwAV pre-training from relevant unlabeled images. Our findings highlight a promising direction to leverage semi-supervised learning for developing foundational models, which have revolutionized artificial intelligence applications in most fields. By scaling our proposed approach with more data and compute, we can make significant strides in monitoring to address the global challenge of litter pollution in water bodies.

Coastal current convergence structures in the Bay of Biscay from optimized high-frequency radar and satellite data

The southeastern Bay of Biscay has been described as a “dead end” for floating marine litter, often accumulating along small-scale linear streaks. Coastal Current Convergence Structures (CCS), often associated with vertical motions at river plume edges, estuarine fronts, or other physical processes, can be at the origin of the accumulation. Understanding the formation of CCS and their role in the transport of marine litter is essential to better quantify and to help mitigate marine litter pollution. The Lagrangian framework, used to estimate the absolute dispersion, and the finite-size Lyapunov exponents (FSLE), have proved very effective for identifying CCS in the current velocity field. However, the quality of CCS identification depends strongly on the Eulerian fields. Two surface current velocity data sets were used in the analysis: the remotely sensed velocities from the EuskOOS High-Frequency Radar (HFR) network and velocities from three-dimensional model outputs. They were complemented by drifting buoy velocity measurements. An optimization method, involving the fusion of drifting buoys and HFR velocities is proposed to better reconstruct the fine-scale structure of the current velocity field. Merging these two sources of velocity data reduced the mean Lagrangian error and the Root Mean Square Error (RMSE) by 50 % and 30 % respectively, significantly improving velocity reconstruction. FSLE ridgelines obtained from the Lagrangian analysis of optimized velocities were compared with remotely sensed concentrations of Chlorophyll-a. It was shown that ridgelines control the spatial distribution of phytoplankton. They fundamentally represent the CCS which can potentially affect marine litter aggregation. Analysis of the absolute dispersion revealed large stirring in the alongshore direction which was also confirmed by spatial distribution of FSLE ridgelines. The alignment between FSLE ridgelines and patterns of high Chlorophyll-a concentration was observed, often determining the limits of river plume expansion in the study area.

Quantification of litter in cities using a smartphone application and citizen science in conjunction with deep learning-based image processing

Cities are a major source of litter pollution. Determination of the abundance and composition of plastic litter in cities is imperative for effective pollution management, environmental protection, and sustainable urban development. Therefore, here, a multidisciplinary approach to quantify and classify the abundance of litter in urban environments is proposed. In the present study, litter data collection was integrated via the Pirika smartphone application and conducted image analysis based on deep learning. Pirika was launched in May 2018 and, to date, has collected approximately one million images. Visual classification revealed that the most common types of litter were cans, plastic bags, plastic bottles, cigarette butts, cigarette boxes, and sanitary masks, in that order. The top six categories accounted for approximately 80 % of the total, whereas the top three categories accounted for more than 60 % of the total imaged litter. A deep-learning image processing algorithm was developed to automatically identify the top six litter categories. Both precision and recall derived from the model were higher than 75 %, enabling proper litter categorization. The quantity of litter derived from automated image processing was also plotted on a map using location data acquired concurrently with the images by the smartphone application. Conclusively, this study demonstrates that citizen science supported by smartphone applications and deep learning-based image processing can enable the visualization, quantification, and characterization of street litter in cities.

Spatiotemporal variation in marine litter distribution along the Bulgarian Black Sea sandy beaches: amount, composition, plastic pollution, and cleanliness evaluation

The threat of anthropogenic marine litter, particularly plastic pollution, to marine ecosystems and human health, has spurred mitigation initiatives and global scientific research. Following the Marine Strategy Framework Directive guidelines, this study evaluated marine litter distribution, cleanliness, and plastic pollution indices along Bulgarian Black Sea beaches in 2023. The survey integrates visual assessment, manual sampling, and drone mapping, distributing the beaches along the coastline to encompass a broader range, totaling 45, including 28 remote/natural, 10 semi-urban, and 7 urban beaches. Results indicate a 48% decrease in marine litter distribution on beaches from 1462 ± 147 items/100 m in 2021 to 753 ± 97 items/100 m in 2023, with Artificial polymer materials/plastic materials constituting 88.62% of the total litter amount. A comprehensive plastic macro litter pollution assessment was carried out along Bulgarian beaches using PAI for the first time. In 2023, the average cleanliness status of Bulgarian beaches was classified as “moderate” (CCI: 7.61 ± 1.00), with clean northern and central beaches contrasting with dirty southern beaches. Urbanized beaches were assessed with the highest level of pollution (PAIAV,23: 5.51; CCIAV,23: 18.16). In the long term, cleanliness and plastic pollution maintain “moderate” values with CCIAV,18-23: 8.81 ± 0.89, and PAIAV,18-23: 2.35 ± 0.32, persisting throughout the period, necessitating ongoing monitoring and intervention strategies. Despite identifying a significant number of clean beaches, none meet the EU threshold value of 20 litter items/100 m. This study highlights the urgent need for effective interventions to combat litter accumulation and plastic pollution, particularly in urban or semi-urban beaches, emphasizing multi-stakeholder collaboration for sustainable solutions and coastal ecosystem preservation.

Synthetic polymers: A global threat to aquatic benthic environments

Finding time-efficient and cost-effective data collection methods is a challenge when addressing aquatic litter pollution on a global scale. In this study, we analysed data on aquatic benthic debris collected worldwide by volunteer scuba divers through the Dive Against Debris® citizen science initiative, examining its relationship with spatial and socio-economic factors. Plastic-dominated litter was found in both marine (64 %) and freshwater (48 %) environments, followed by metal and glass. Lower litter abundances have been recorded in high income countries such as in Europe, Central Asia and North America. Plastic fragments and fishing lines were the most abundant seafloor litter items, while takeaway containers (aluminium cans, glass bottles) were dominant in freshwater environments. Single-use plastics, including objects for food and beverage consumption, accounted for about 1/3 of the total benthic aquatic debris. Our findings highlight the need to prioritise the fishing industry and change our fast-paced modern lifestyle. Citizen science initiatives, once data cleanup is conducted to overcome any bias, can provide valuable tools for better understanding and quantifying marine litter pollution. The outcomes gained can be leveraged to improve consumer awareness and inform environmental policies aimed at addressing aquatic litter pollution more effectively.

Not only in the crowd: Benthic litter characterization in a low population density area still reveals widespread pollution and local malpractices

Various anthropogenic activities affect marine coastal habitats, leading to heavy litter pollution. However, whilst high litter concentrations are nowadays common in the proximity of metropolises, few studies investigated the magnitude of this phenomenon around coastal villages and small towns. We hereby characterized the benthic litter occurring in the trawlable grounds of the Gulf of Policastro (Tyrrhenian Sea, central-western Mediterranean), a low population density area that becomes a popular tourist destination during summer. We furthermore tested differences between two depths (∼100–200 and ∼500–600 m) and the impact of tourism on the shallower waters. The area was characterized by a litter abundance of 651.12 ± 130.61 item/km2, with plastic being almost totalitarian (93%). The shallower waters hosted two-thirds of the litter found. Almost all (∼95%) the litter items had a land-based origin, while the sea-based litter was mostly found at higher depths. About 14% of the litter was found to be fouled, with the development of litter-associated communities that somehow mimic the natural ones living on hard substrates. The higher litter presence noticed during the touristic peak (July–August) suggests that tourism is an important source of local litter, although it contributed to the local accumulation in a synergic way with other factors. The majority of the litter items presumably originated from the nearby coastline, while the deeper waters were or are used as a dumping site by the local trawling fleet. The discovery of such a critical waste accumulation and management in a somehow remote area contributes to widen the perspectives on the presence of benthic litter mostly in territories characterized by wide anthropization. Moreover, it confirms that appropriate local policies and communication plans are urged even at a regional level to stimulate citizen consciousness and mitigate the ever growing litter pollution.

Assessing the relationship between urbanization and plastic litter on sandy beaches in California, USA

Urbanization can increase the threat plastic litter poses to beaches around the world. However, many oceanic beaches lack baseline data on how urbanization may be altering plastic litter abundance. Here, I documented the abundance and type of plastic litter at 23 oceanic beaches that spanned an urbanization gradient (as measured by population density) along the California, USA coast in October 2023. Across all 23 sites, I found that urbanization significantly increased abundance for all types of plastic litter including (large microplastic litter 2–5 mm; mesoplastic 5–25 mm; macroplastic >25 mm). I also found that urban beaches had significantly more plastic litter (132.7 pieces/m2 ± 84.2 S.D.) than non-urban beaches (67.4 pieces/m2 ± 30.9 S.D.). In addition, I found that urban beaches had significantly higher abundances of cigarette butts, hard plastic, and foamed plastic than non-urban beaches. Collectively, my findings suggest that plastic litter pollution poses an important threat to California beaches, that oceanic beaches in more urban areas may be more impacted by this threat over time, and that plastic waste from recreation (both on and off-shore) could be an important source of pollution.

Ingestion of fishing gear and Anisakis sp. infection in a beached Indo-Pacific finless porpoise (Neophocaena phocaenoides) in the Jeju Island, Republic of Korea: findings from post-mortem computed tomography and necropsy

BackgroundHuman fishing activities have significantly affect environmental concern for marine ecosystems, conservation of marine mammals, and human health. Coastal cetaceans are highly vulnerable to ingestion of fishing gear, bycatching, or entanglement, all of which can be fatal for these animals. In particular, certain coastal dolphins and porpoises are heavily impacted by fishing gear such as angling gear or stownet, as their food often overlap with the target fish species of human fisheries.Case presentationThis study presents a case of an Indo-Pacific finless porpoise (Neophocaena phocaenoides) beached on the coast of Jeju Island, Republic of Korea, with ingestion of fishing gear and severe Anisakis infection. Although this species inhabits waters ranging from the Persian Gulf to Taiwan, several stranded carcasses have been reported on Jeju Island in recent years. Post-mortem computed tomography revealed a bundle of four fishing hooks in the forestomach, along with nylon lines and steel lines with connectors, which were assumed to be angling gear for Jeju hairtail (Trichiurus lepturus). Further necroscopic investigation revealed that the forestomach contained a large number of Anisakis spp. (Nematoda: Anisakidae). Histological examination revealed a thickened forestomach wall with pinpoint and volcanic ulcerations, a thickened layer of stratified squamous epithelium, and infiltrated stroma in the squamous epithelium.conclusionsThis study emphasizes the urgent need to address the impact of fishing activities on marine mammals, marine litter pollution, and the bycatch problem in Korean seawater. In addition, the occurrence of N. phocaenoides in seawater around Jeju Island should be raised in future geographical ecology or veterinary pathology studies and when its distribution is updated.

Distribution and typologies of anthropogenic seafloor litter in the Pearl River Estuary and adjacent coastal waters, China

Marine litter pollution poses a significant threat to offshore ecosystems, eliciting widespread concern. We investigated seafloor litter patterns in the Pearl River Estuary and adjacent coastal waters of China in 2023 via bottom trawl survey. Average number and weight densities were found to be 154.34 ± 30.95 n/km2 and 2384.63 ± 923.98 g/km2, respectively. Plastic was the most abundant material by number density (79.07 %), and rubber the highest by weight density (22.93 %). Overall number density varied from 40.50 ± 22.50 to 221.13 ± 52.44 n/km2, with the highest in Daya Bay and the lowest in Guanghai Bay. Weight density varied from 189.93 ± 71.94 to 5386.70 ± 3050.30 g/km2, with the highest in Qiao Island and the lowest in Honghai Bay. The main source was plastic bags and wrappers. The Pearl River Delta and Daya Bay were identified as seafloor litter distribution hotspots. Controlling plastic waste input is crucial for reducing seafloor litter in the Pearl River Estuary.

Revision of China’s marine environmental protection law: history, background and improvement

The law on marine environmental protection is a branch of the law on environmental protection and is the basis for the conduct of marine environmental protection. Since its promulgation in 1982, China’s Marine Environmental Protection Law has undergone two revisions and three amendments, forming a legal system based on the Constitution, with the Environmental Protection Law as the overarching principle, the Marine Environmental Protection Law as the core, and various laws on the protection of marine elements as the gripping force. The revision of the Marine Environmental Protection Law in 2023 was based on the 2018 supervision and inspection of the enforcement inspection team of the Standing Committee of the National People’s Congress, the implementation of the construction of a strong maritime country, and the adaptation to the reform of the ocean governance system. Compared with the previous Marine Environmental Protection Law, the improvements of the newly revised law are mainly reflected in the following: optimizing the institutional mechanism, reflecting the Land and Sea Coordination, strengthening supervision and management, strengthening marine litter pollution prevention and control, enhancing the protection of marine biodiversity, strengthening the regulation of estuaries, and strict liability.

Assessing beach litter trapping efficiency in Mediterranean sandy coasts: A comparative study between typical and invaded embryonic dunes

Beach litter is one of the most evident indicators of marine litter pollution, an anthropogenic component that can affect and interact with the habitat of coastal dunes. In this study, we aim to assess the role of the Mediterranean embryonic dunes in trapping beach litter. Moreover, we investigate if dunes with native vegetation and those invaded by the alien plant C. acinaciformis differ in the trapping of beach litter. To this end, two samplings were carried out in the Tavolara-Punta Coda Cavallo Marine Protected Area, considering four beaches with different morphologies, using a paired sampling method that considers plots in the embryonic dunes and in the same habitat with C. acinaciformis. Our results indicate that plastic is the primary type of beach litter and that its distribution varied across the different beaches; especially, the greatest amount was found on pocket beaches. Based on our results, we can conclude that there are no differences between embryonic dunes with native vegetation and their invaded form, but the different beach morphologies may play a role in the distribution of beach litter. These findings may support habitat conservation initiatives such as the eradication of C. acinaciformis since it has no additional role in trapping beach litter.

Classification, composition, and sources of marine litter on beach sediment of Kondang Merak Coast, Malang, Indonesia

Marine litter pollution has been examined extensively as a significant environmental concern. Beaches are particularly vulnerable to this form of pollution since it is located between land and sea. Malang is one of the cities on Java Island, with a long coastline and numerous beautiful beaches facing the Indian Ocean. Kondang Merak Beach is a popular tourist destination among locals. However, it is under constant threat from marine litter pollution. The beach is divided into two distinct areas: the foreshore, which is influenced by sea waves and currents, and the backshore, which is significantly impacted by tourist activity. Therefore, this study aimed to examine marine litter on beach sediment and to distinguish the composition of marine litter between foreshore and backshore areas. Quadrat transects of 5 x 5 m and 50 x 50 cm were used to collect bigger and micro-size marine litter, respectively. Of the 895 litter items, 61 % were collected from the backshore and 39 % from the foreshore areas. Bigger-sized litter found in the backshore area was the result of tourism activities. On the other hand, micro-size litter in the foreshore area resulted from fragmentation from bigger-size litter.

GenZs environmental attitudes and ecology behavior nexus: Urgent education message

Environmental education in schools is critical to help decrease plastic litter pollution because plastic pollution could be the most significant health problem of the 21st century. Our research examined the environmental behaviors and attitudes of Generation Z (Gen Z) high school and undergraduate students in South Texas, USA. The research was important because we need to understand what young people believe and value if we are going to see any change in environmental behavior. Results supported research on how teachers can influence student environmental attitudes, behaviors, and knowledge with education. Feedback from students revealed significant attitude change and a rich narrative of pollution and watershed ecology knowledge and behavior themes. These themes disclosed that a holistic environmental science curriculum is necessary to connect Earth systems with the plastic litter pollution cycle.

Anthropogenic debris in three sympatric seal species of the Western Antarctic Peninsula

Litter pollution is a growing concern, including for Antarctica and the species that inhabit this ecosystem. In this study, we investigated the microplastic contamination in three seal species that inhabit the Western Antarctic Peninsula: crabeater (Lobodon carcinophaga), leopard (Hydrurga leptonyx) and Weddell (Leptonychotes weddellii) seals. Given the worldwide ubiquity of this type of contaminant, including the Southern Ocean, we hypothesized that the three seal species would present anthropogenic debris in their feces. We examined 29 scat samples of crabeater (n = 5), leopard (n = 13) and Weddell (n = 11) seals. The chemical composition of the items found were identified using micro-Raman and micro-FTIR spectroscopies. All the samples of the three species presented anthropic particles (frequency of occurrence - %FO – 100 %). Fibers were the predominant debris, but fragments and filaments were also present. Particles smaller than 5 mm (micro debris) were predominant in all the samples. Leopard seals ingested significantly larger micro-debris in comparison with the other seal species. The dominant color was black followed by blue and white. Micro-Raman and micro-FTIR Spectroscopies revealed the presence of different anthropogenic pigments such as reactive blue 238, Indigo 3600 and copper phthalocyanine (blue and green). Carbon black was also detected in the samples, as well as plastic polymers such as polystyrene, polyester and polyethylene terephthalate (PET), polyamide, polypropylene and polyurethane These results confirm the presence of anthropogenic contamination in Antarctic seals and highlight the need for actions to mitigate the effects and reduce the contribution of debris in the Antarctic ecosystem.

Beach litter pollution along the odisha coast: Composition, abundance, biodiversity impact and management practices

Beach litter pollution at Puri, Gopalpur, and the world-famous turtle nesting beach at Rushikulya along the Odisha coast are studied during 2018, 2019, and 2021. The composition, abundance, colour and sources of the litters revealed five major categories; plastic, foam, metal, rubber and glass in all three beaches. Pollutant load shows an increasing trend with maximum at Gopalpur followed by Rushikulya and Puri. Litter of ten colours were observed with a maximum percentage of white and translucent litter. Plastics were the most abundant litter in Puri and Gopalpur while rubber was most abundant in Rushikulya. Tourism activities are identified as the major source of pollution at Puri and Gopalpur, while river discharge contributes the maximum to beach litter pollution at Rushikulya. The study discusses on the impacts of these litters on sea turtle nesting habitat and population. Clean Coast Index (CCI) for Puri, Rushikulya and Gopalpur indicates the beaches under the very clean category. Beach litter management practices along the Odisha coast are discussed and methods for improvement are suggested.

Emission, Transport and Retention of Floating Marine Macro-Litter (Plastics): The Role of Baltic Harbor and Sailing Festivals

Every year, harbor and sailing festivals attract close to 20 million visitors in the Baltic Sea region, but their consequences on marine litter pollution are still unknown. We combine field studies with model simulations and literature reviews to quantify the annual emissions of floating macro-litter and to assess its retention in estuaries and role in Baltic Sea pollution. Results focusing on Hanse Sail in Rostock and Kiel Week are extrapolated to the entire Baltic Sea region. After the Hanse Sail 2018, the harbor pollution amounted to about 950 floating macro-litter particles/km²; 85–90% were plastics. We calculated an emission between 0.24 and 3 particles per 1000 visitors, depending on the year and the waste management system. About 0.02% of all waste generated during a festival ends up in the harbor water. The Hanse Sails contributes less than 1% to the total annual macro-litter emissions in the Warnow estuary. Model simulations indicate that over 99% of the emitted litter is trapped in the estuary. Therefore, Hanse Sails are not relevant to Baltic Sea pollution. The extrapolated Baltic-Sea-wide annual emissions are between 4466 and (more likely) 55,830 macro-litter particles. The over-30 harbor and sailing festivals contribute an estimated <0.05% to the total annual macro-litter emissions in the Baltic Sea region.

Evaluation of Marine Litter Pollution Level in a Urban Beach (Aegean Coast of Izmir, Turkey)

Evaluation of Marine Litter Pollution Level in a Urban Beach (Aegean Coast of Izmir, Turkey)

Assessment of marine litter on Morocco’s Mediterranean beaches

Marine litter pollution is one of the world’s major environmental challenges. In the Mediterranean, marine litter on beaches is mainly generated by tourism and recreational activities. The present study is part of an "Adopt-a-Beach" pilot project, initiated as part of the Marine Litter MED (ML MED) project, coordinated by the United Nations Environment Programme UNEP/MAP. The main objective of this work was to assess and quantify marine litter on four beaches along Morocco’s Mediterranean coast (Merkala Tanger, Amsa Tetouan, Sabadia Al Hoceima, and Miami Nador) over a seasonal cycle during 2018–2019, in line with UNEP/MAP guidelines. On average, 1768 items/100 m were collected from the four beaches, with a density of 0.34 items/m2. "Plastic/polystyrene" was the main component of the total debris collected, accounting for 74.1% in number of items, followed by "paper/cardboard" at 6.4%. The dominant categories of top 10 items were "plastic/polystyrene pieces < 50 cm" (G76), "cigarette butts and filters" (G27), "plastic caps and lids" (G21/24), and "crisps packets/sweets wrappers" (G30/31) that all comprised 49.9% of the total items. Tourism and recreational wastes were the most represented sources at 52%.

Progress in Research on Marine Litter-Related Monitoring Technologies

The problem of marine litter pollution is becoming increasingly serious and has become one of the important global environmental issues. From the perspective of source and sink, floating riverine litter (FRL) is the main source of floating marine litter (FML), while beach litter (BL) and FML are the source and sink of each other. It is necessary to research the transformation mechanism and collaborative governance policy among these three kinds of marine litter, based on accurate monitoring data. However, conventional monitoring methods have problems such as high labor costs, insufficient representativeness and poor comparability. Hence, developing efficient and intelligent monitoring technology is the focus of recent future. Since 2018, monitoring technique of marine litter relevant research literature has grown rapidly, showing this technique a hot spot. In this paper, the relevant research literature is divided into three litter types (BL, FML, and FRL) and two ways (fixed monitoring and mobile monitoring). Similarities and differences in research focus, platform selection, result accuracy and result application are summarized and reasoned, with a view to accelerating the research progress and better supporting BL, FML, and FRL control in China.

Assessment of riverbank macrolitter contamination in two mediterranean river catchments: Laou and Martil, Morocco

The Mediterranean Sea has been characterized in recent years by marine litter pollution. Our study aims to quantify waste flows in two watersheds in the northern Mediterranean region of Morocco. For Laou wadi and Martil wadi, floating and bank-accumulated waste was monitored during the autumn 2022 and spring 2023 seasons respectively. The results revealed that for Laou wadi, 384 objects/100 m were collected on the banks with a density of 0.20 objects/m2, and "plastic/polystyrene" constituted the main component of the total debris collected representing 57.19% in number of pieces, followed by "glass/ceramic" with 28.44%. For Martil wadi, 4715 objects/100 m were collected, with a density of 6.74 objects/m2, and "plastic/polystyrene" also constituted the main component of the total debris collected with 94.38%. The debris recorded are mainly associated with waste discharge along the river banks. It is therefore recommended to improve solid waste management and activate environmental control in the two Mediterranean basins studied in order to protect the Mediterranean Sea.