Marine plastic pollution is a pressing global issue nowadays. To address this problem, automated image analysis techniques that can identify plastic litter are necessary for scientific research and coastal management purposes. The Beach Plastic Litter Dataset version 1 (BePLi Dataset v1) comprises 3709 original images taken in various coastal environments, along with instance-based and pixel-level annotations for all plastic litter objects visible in the images. The annotations were compiled in the Microsoft Common Objects in Context (MS COCO) format, which was partially modified from the original format. The dataset enables the development of machine-learning models for instance-level and/or pixel-wise identification of beach plastic litter. All original images in the dataset were extracted from beach litter monitoring records operated by the local government of Yamagata Prefecture in Japan. Litter images were taken in different backgrounds, such as sand beaches, rocky beaches, and tetrapods. The annotations for instance segmentation of beach plastic litter were made manually, and were given for all plastics objects, including PET bottles, containers, fishing gear, and styrene foams,all of which were categorized in a single class \"plastic litter\". Technologies developed using this dataset have the potential to enable further scalability for the estimation of plastic litter volume. This would help researchers, including individuals, and the the government to monitor or analyze beach litter and the corresponding pollution levels.

We evaluated the potential impacts from using a rapid same-day quantitative polymerase chain reaction (qPCR) monitoring method for beach posting outcomes at two Toronto beaches.
In total, 228 water samples were collected at Marie Curtis Park East and Sunnyside Beaches over the 2021 summer season. Water samples were processed using the USEPA 1609.1 Enterococcus qPCR-based method. Escherichia coli (E. coli) culture data and daily beach posting decisions were obtained from Toronto Public Health.
No significant correlation was observed between previous-day and same-day (retrospective) E. coli enumeration results at any Sunnyside Beach transect, and only relatively low (R = 0.41-0.56) or no significant correlation was observed at sampling transects for Marie Curtis Park East Beach. Comparing our same-day Enterococcus qPCR data to Toronto\'s 2-day E. coli geometric mean beach posting decisions, we noted the need for additional postings for 1 (2%) and 3 (8%) missed health-risk days at Sunnyside and Marie Curtis Park East Beaches, respectively. The qPCR data also pointed to incorrect postings for 12 (31%) and 6 (16%) lost beach days at Sunnyside and Marie Curtis Park East Beaches, respectively.
Application of a rapid Enterococcus qPCR method at two Toronto beaches revealed 5% of beach posting decisions were false negatives that missed health-risk days, while 23% of decisions were false positives resulting in lost beach days. Deployment of the rapid same-day qPCR method offers the potential to reduce both health risks and unnecessary beach postings.
RéSUMé: OBJECTIFS: Nous avons évalué, à deux plages de Toronto, l’effet possible de l’utilisation d’une méthode de surveillance rapide par PCR quantitative (qPCR) le même jour sur les avis de fermeture ou d’ouverture des plages. MéTHODE: En tout, 228 échantillons d’eau ont été prélevés aux plages Marie Curtis Park East et Sunnyside au cours de la saison estivale 2021. La présence d’Enterococcus dans les échantillons a été détectée par la méthode USEPA 1609.1, utilisant la qPCR. Les données sur les cultures d’Escherichia coli (E. coli) et les avis quotidiens de fermeture ou d’ouverture des plages ont été obtenus auprès du Bureau de santé de Toronto. RéSULTATS: Aucune corrélation significative n’a été observée entre les résultats (rétrospectifs) du dénombrement de E. coli obtenus la veille et le même jour dans les transects de la plage Sunnyside, et une corrélation significative faible (R = 0,41–0,56) ou nulle a été observée dans les transects d’échantillonnage de la plage Marie Curtis Park East. En comparant nos données sur Enterococcus obtenues le même jour par qPCR à la moyenne géométrique des avis de fermeture ou d’ouverture des plages sur deux jours liés à E. coli émis par le Bureau de santé de Toronto, nous avons remarqué qu’il aurait fallu émettre des avis de fermeture pour 1 jour de risques pour la santé manqué (2 %) à la plage Sunnyside et pour 3 jours de risques pour la santé manqués (8 %) à la plage Marie Curtis Park East. Les données de la qPCR ont aussi fait état d’avis de fermeture incorrects ayant entraîné la perte de 12 jours de plage (31 %) à Sunnyside et de 6 jours de plage (16 %) à Marie Curtis Park East. CONCLUSION: L’application d’une méthode de surveillance rapide d’Enterococcus par qPCR à deux plages de Toronto a montré que 5 % des avis étaient des faux négatifs qui n’ont pas détecté des jours de risques pour la santé, et que 23 % étaient des faux positifs qui ont entraîné des jours de plage perdus. Le déploiement de la méthode rapide par qPCR le même jour offre la possibilité de réduire à la fois les risques pour la santé et les avis de fermeture de plages inutiles.

Michigan\'s water-quality standards specify that E. coli concentrations at bathing beaches must not exceed 300 E. coli per 100 mL, as determined by the geometric mean of culture-based concentrations in three or more representative samples from a given beach on a given day. Culture-based analysis requires 18⁠-⁠24 h to complete, so results are not available on the day of sampling. This one-day delay is problematic because results cannot be used to prevent recreation at beaches that are unsafe on the sampling day, nor do they reliably indicate whether recreation should be prevented the next day, due to high between-day variability in E. coli concentrations demonstrated by previous studies. By contrast, qPCR-based E. coli concentrations can be obtained in 3-4 h, making same-day beach notification decisions possible. Michigan has proposed a qPCR threshold value (qTV) for E. coli of 1.863 log10 gene copies per reaction as a potential equivalent value to the state standard, based on statistical analysis of a set of state-wide training data from 2016 to 2018. The main purpose of the present study is to assess the validity of the proposed qTV by determining whether the implied qPCR-based beach notification decisions agree well with culture-based decisions on two sets of test data from 2016⁠-⁠2018 (6,564 samples) and 2019-2020 (3,205 samples), and whether performance of the proposed qTV is similar on the test and training data. The results show that performance of Michigan\'s proposed qTV on both sets of test data was consistently good (e.g., 95% agreement with culture-based beach notification decisions during 2019⁠-⁠2020) and was as good as or better than its performance on the training data set. The false-negative rate for the proposed qTV was 25-29%, meaning that beach notification decisions based on the qTV would be expected to permit recreation on the day of sampling in 25-29% of cases where the beach exceeds the state standard for FIB contamination. This false-negative rate is higher than one would hope to see but is well below the corresponding error rate for culture-based decisions, which permit recreation at beaches that exceed the state standard on the day of sampling in 100% of cases because of the one-day delay in obtaining results. The key advantage of qPCR-based analysis is that it permits a large percentage (71-75%) of unsafe beaches to be identified in time to prevent recreation on the day of sampling.

The unusual arrival of Sargassum on Caribbean beaches is an emerging problem that has generated numerous challenges. The monitoring, visualization, and estimation of Sargassum coverage on the beaches remain a constant complication. This study proposes a new mapping methodology to estimate Sargassum coverage on the beaches. Semantic segmentation of geotagged photographs allows the generation of accurate maps showing the percent coverage of Sargassum. The first dataset of segmented Sargassum images was built for this study and used to train the proposed model. The results demonstrate that the currently proposed method has an accuracy of 91%, improving on the results reported in the state-of-the-art method where data was also collected through a crowdsourcing scheme, in which only information on the presence and absence of Sargassum is displayed.

This dataset consists of 3500 images of beach litter and 3500 corresponding pixel-wise labelled images. Although performing such pixel-by-pixel semantic masking is expensive, it allows us to build machine-learning models that can perform more sophisticated automated visual processing. We believe this dataset may be of significance to the scientific communities concerned with marine pollution and computer vision, as this dataset can be used for benchmarking in the tasks involving the evaluation of marine pollution with various machine learning models. The beach litter images were obtained from coastal environment surveys conducted between 2011 and 2019 by the Yamagata Prefectural Government, Japan. These images were originally obtained owing to the reporting guidelines concerning regular coastal-environmental-cleanup and beach-litter-monitoring surveys. Based on these images, the Japan Agency for Marine-Earth Science and Technology created 3500 images comprising eight classes of semantic masks for beach litter detection [1].

Mitigating and preventing beach litter from entering the ocean is urgently required. Monitoring beach litter solely through human effort is cumbersome, with respect to both time and cost. To address this problem, an artificial intelligence technique that can automatically identify different-sized beach litter is proposed. The technique was established by training a deep learning model that enables pixel-wise classification (semantic segmentation) using beach images taken by an observer on the beach. Eight segmentation classes that include two beach litter classes were defined, and the results were qualitatively and quantitatively verified. Segmentation performance was adequately high based on three metrics: Intersection over Union (IoU), precision, and recall, although there is room for further improvement. The potency of the method was demonstrated when it was applied to images taken in different places from training data images, and the coverage of artificial litter calculated and discussed using drone images provided ground truth.

The diurnal presence of the culturable bacterial indicators of fecal contamination in the water environment has been shown to be highly variable over time due to natural die-off and injury from effects of sunlight and other environmental stressors. Molecular analytes of a quantitative polymerase chain reaction (qPCR) method for measuring fecal contamination degrade considerably slower than the alternative of culturable fecal indicator bacteria. The rapid qPCR method holds the promise of more timely notification decisions with respect to postings or closure being made on the basis of microbial water quality samples collected earlier on the same day. In the case of culture-based methods requiring a 24 h or longer incubation period, decisions must be based on samples collected no sooner than the previous day. To examine the effect of this lag in assay results, temporal stability of a molecular Enterococci target analyte with that of traditional culture-based cells is compared using data from USEPA studies conducted between 2003 and 2007 on seven freshwater and marine beaches that were impacted by publicly-owned treatment works. Generally, levels of the molecular indicator were more consistent throughout the day between 8:00 am and 3:00 pm. The difference in temporal consistency is even more pronounced when the 24-h lag in culture-based results is taken into account.

The atypical arrival of pelagic Sargassum to the Mexican Caribbean beaches has caused considerable economic and ecological damage. Furthermore, it has raised new challenges for monitoring the coastlines. Historically, satellite remote-sensing has been used for Sargassum monitoring in the ocean; nonetheless, limitations in the temporal and spatial resolution of available satellite platforms do not allow for near real-time monitoring of this macro-algae on beaches. This study proposes an innovative approach for monitoring Sargassum on beaches using Crowdsourcing for imagery collection, deep learning for automatic classification, and geographic information systems for visualizing the results. We have coined this collaborative process \"Collective View\". It offers a geotagged dataset of images illustrating the presence or absence of Sargassum on beaches located along the northern and eastern regions in the Yucatan Peninsula, in Mexico. This new dataset is the largest of its kind in surrounding areas. As part of the design process for Collective View, three convolutional neural networks (LeNet-5, AlexNet and VGG16) were modified and retrained to classify images, according to the presence or absence of Sargassum. Findings from this study revealed that AlexNet demonstrated the best performance, achieving a maximum recall of 94%. These results are good considering that the training was carried out using a relatively small set of unbalanced images. Finally, this study provides a first approach to mapping the Sargassum distribution along the beaches using the classified geotagged images and offers novel insight into how we can accurately map the arrival of algal blooms along the coastline.

Most Great Lakes communities rely on culture-based E. coli methods for monitoring fecal indicator bacteria (FIB) at recreational beaches. These cultivation methods require 18 or more hours to generate results. As a consequence, public notifications about beach action value (BAV) exceedance are based on prior-day water quality. Rapid qPCR monitoring of bacteria in beach water solves the 24-h delay problem, though the USEPA-approved qPCR method targets enterococci bacteria, while Great Lakes communities are familiar with E. coli monitoring. For an E. coli qPCR method to be useful for water quality management, it is important to systematically characterize method performance, and establish BAVs for public notification purposes. In this study, we 1) evaluated a draft USEPA E. coli qPCR method, 2) compared E. coli qPCR measurements with two established FIB (E. coli culture and enterococci qPCR) results, and explored potential strategies to establish E. coli qPCR BAV criteria in the absence of an epidemiological study. Based on analyses of 288 water samples collected from eight of Chicago\'s Lake Michigan beaches, the E. coli qPCR method demonstrates acceptable performance characteristics. The method is prone to low level DNA contamination, possibly originating from assay reagents derived from E. coli bacteria. Both E. coli and enterococci BAVs were exceeded in approximately 18% of the samples. E. coli qPCR values were correlated with both E. coli culture (r = 0.83; p < 0.0001) and enterococci qPCR (r = 0.67; p < 0.0001) values. The approach recommended by the USEPA in its Technical Support Material (TSM) was used to generate candidate E. coli qPCR BAVs, as was receiver operating characteristic (ROC) analysis. Potential BAV thresholds differed substantially, ranging from 200.9 calibrator cell equivalents (CCE)/100 mL (ROC analysis, enterococci qPCR BAV as the reference) to 1000 CCE/100 mL (TSM analysis, enterococci qPCR BAV as the reference). Because we found that different approaches to establishing potential BAVs generate quite different values, guidance from USEPA about approaches to defining comparable BAVs would be useful.

This study assessed the levels of marine debris pollution and identified its main sources in Korea. The surveys were bimonthly conducted by NGO leaders and volunteers on 20 beaches from March 2008 to November 2009. The quantities of marine debris were estimated at 480.9 (±267.7) count⋅100 m(-1) for number, 86.5 (±78.6) kg⋅100 m(-1) for weight, and 0.48 (±0.38) m(3)⋅100 m(-1) for volume. The level of marine debris pollution on the Korean beaches was comparable to that in the coastal areas of the North Atlantic ocean and South Africa. Plastics and styrofoam occupied the majority of debris composition in terms of number (66.7%) and volume (62.3%). The main sources of debris were fishing activities including commercial fisheries and marine aquaculture (51.3%). Especially styrofoam buoy from aquaculture was the biggest contributor to marine debris pollution on these beaches.