Regional Studies in Marine Science最新文献

{"title":"Elasmobranch fisheries in Tunisia: Characterization, fishermen’s insights and conservation options","authors":"Rihab Louhichi ,&nbsp;Samira Enajjar ,&nbsp;Mohamed Nejmeddine Bradai ,&nbsp;Bechir Saidi","doi":"10.1016/j.rsma.2025.104465","DOIUrl":"10.1016/j.rsma.2025.104465","url":null,"abstract":"<div><div>Elasmobranchs in Tunisia are facing an increased fishing pressure due to the decline of traditional fishery resources. This study aims to characterize the artisanal elasmobranch fisheries in Tunisia, assess their impact on species populations and recommend future management strategies. The method was based on inspections and 161 interviews with local fishermen targeting sharks and batoids during 2022 and 2023. Results showed that elasmobranch fisheries are practiced only in the Gulf of Gabès. Specific gillnets are used to target <em>Carcharhinus plumbeus</em>, <em>Mustelus mustelus</em>, <em>Mustelus punctulatus</em>, <em>Glaucostegus cemiculus</em> and <em>Aetomylaeus bovinus</em>, and the former species is also targeted with pelagic longlines, primarily during spring and summer. Furthermore, fishermen adapt their fishing gear and grounds in response to the appearance of species in the area when they move to shallow waters for reproduction. More than 90 % of fishermen stated that catch composition varies by month and target species, with most captures consisting of adult specimens of both sexes, particularly pregnant females, suggesting that these fisheries exploit species in their essential habitats. Most respondents reported declines in both catches and sizes of target species over the past 10 years, mainly due to overfishing. This exploitation, combined with the absence of management strategies, could jeopardize elasmobranch populations in the area. Conservation measures were deemed urgent to ensure the sustainable management of elasmobranch fisheries in Tunisia. Actions such as controlling fishing effort and closing critical habitats could yield significant benefits for the long-term sustainability of these fisheries.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104465"},"PeriodicalIF":2.4,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced settling velocity and floc formation in response to dredging-induced plumes: In-situ observations from a coastal sand mining site","authors":"Seong Woon Jeong ,&nbsp;Hun Jun Ha ,&nbsp;Ho Kyung Ha","doi":"10.1016/j.rsma.2025.104463","DOIUrl":"10.1016/j.rsma.2025.104463","url":null,"abstract":"<div><div><em>In-situ</em> observations were conducted to investigate suspended sediment dynamics within dredging-induced plumes at a sand mining site in Gyeonggi Bay, Korea. Combined acoustic and optical measurements captured the spatiotemporal variability in suspended sediment concentration (SSC) and settling velocity (<em>W</em>_<em>s</em>) during and after the dredging. Overflow discharge generated near-surface plumes with SSCs up to an order of magnitude higher than background levels, which were rapidly dispersed within one hour of dredging cessation. Particle size distributions revealed the formation of large flocs (&gt;100 μm) during active dredging, accompanied by increased <em>W</em>_<em>s</em> of 0.46–0.63 mm s⁻¹ . Post-dredging conditions showed reduced floc sizes and <em>W</em>_<em>s</em> (&lt;0.1 mm _<em>s</em>⁻¹), indicating rapid floc disaggregation. Flocculation was enhanced by high cohesive sediment supply and organic matter under calm hydrodynamic conditions. The findings from this study provide essential insights for assessing the environmental impacts of dredging on marine systems and serve as a valuable foundation for developing sustainable sediment management strategies.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104463"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144996813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Heavy metal accumulation and benthic foraminiferal response following the Peruvian Ventanilla crude oil spill (January 2022)","authors":"Leonela Rojas-Portilla ,&nbsp;Diana Ochoa ,&nbsp;Jorge Cardich ,&nbsp;Elizabeth Silva-Berrospi ,&nbsp;Sara Cárdenas-Farfán ,&nbsp;Arturo Aguirre-Velarde ,&nbsp;Matthieu Carré ,&nbsp;Irina Djouraev ,&nbsp;Samia Hidalgo ,&nbsp;Dimitri Gutiérrez ,&nbsp;Raul Loayza-Muro ,&nbsp;María Rivera-Ch ,&nbsp;Pedro E. Romero ,&nbsp;Jesús Barreto-Meza ,&nbsp;Edgar Cruz-Acevedo","doi":"10.1016/j.rsma.2025.104459","DOIUrl":"10.1016/j.rsma.2025.104459","url":null,"abstract":"<div><div>The accidental release of around 11,000 barrels of crude oil off the central Peruvian coast (11.92°S, 77.18°W) on January 15, 2022, affected ∼70 km of shoreline. This study analyzes the spatiotemporal variation of metal concentrations in intertidal and subtidal zones from October 2022 to September 2023, and their effects on benthic foraminiferal communities. Metal concentrations were found to exceed background levels established by the Peruvian Environmental Inspection Agency. Although no temporal trends were detected, spatial differences were recorded. Intertidal sites showed enhanced concentrations of V, Fe, Cr, and Mn. Notably, the site closest to the spill (Ventanilla Beach; 11.87ºS), recorded the highest concentrations of Pb, Cu, Zn, and As. In the subtidal zone, concentrations increased with water depth and reduced hydrodynamic conditions. The more sheltered site (Punta Mulatos; 11.76ºS) had higher Ni, Pb, Cu, Cd, Zn, and As levels compared to the more dynamic site (Bahía Blanca; 11.83ºS). Furthermore, a contribution associated with natural rock leaching was detected. These findings suggest that the spill's influence on metal concentration is amplified by low hydrodynamics, low oxygen, high productivity and riverine inputs. Dominant benthic foraminiferal taxa, typical of oxygen-poor environments, were associated with enhanced Cd, Cu, Ni, Cr, and Pb. However, isolating spill effects is challenging due to natural species tolerance and concurrent El Niño conditions affecting oxygen levels and organic matter fluxes. Finally, 18 months after the spill, heavy metals persist in both environments, underscoring the need for long-term monitoring and mitigation.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104459"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144996810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Environmental typology of Colombian Caribbean beaches based on sedimentological and colorimetric signatures","authors":"Nelson Rangel-Buitrago ,&nbsp;Alex Paternina-Ramos ,&nbsp;William J. Neal","doi":"10.1016/j.rsma.2025.104462","DOIUrl":"10.1016/j.rsma.2025.104462","url":null,"abstract":"<div><div>Understanding the spatial variability and depositional dynamics of beach sediments is essential for assessing coastal morphodynamics, sedimentary processes, and vulnerability in tropical high-energy environments. This study presents for first time, a comprehensive environmental typology of 23 beaches located along the central Caribbean coast of Colombia (Atlántico Department), based on the integration of granulometric parameters, CIELAB colorimetric data, and multivariate statistical analysis. Sediment samples were analyzed for grain size distribution (mean, sorting, skewness, kurtosis), and chromatic properties (L*, a*, b*), enabling the classification of sedimentary facies through discriminant indices (Y1, Y2, Y3), hierarchical clustering, and principal component analysis. Results reveal a predominance of well-sorted, quartz-rich sands indicative of swash-dominated, high-energy systems, punctuated by localized coarse storm berms and finer drift-line zones. Colorimetric parameters, particularly L* and b*, proved to be reliable proxies of sediment maturity, oxidation, and depositional environment. The integration of textural and optical descriptors facilitated the identification of four distinct facies types, offering new insights into sediment provenance, morphodynamic sensitivity, and anthropogenic modification. This typological framework enhances coastal sediment assessment and provides a transferable methodology for sediment monitoring, beach management, and resilience evaluation in sediment-limited tropical coasts.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104462"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144996779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Human impact on the frozen continent: Anthropogenic debris in a coastal Antarctic ecosystem","authors":"Fernando Augusto Bertazzo-Silva ,&nbsp;Flavia Helena Aires Sousa ,&nbsp;Jorge Renato Pinheiro Velloso ,&nbsp;Kamille Rodrigues Ferraz ,&nbsp;Alice Lemos Costa ,&nbsp;Carlos Ernesto Gonçalves Reynaud Schaefer ,&nbsp;Jair Putzke","doi":"10.1016/j.rsma.2025.104458","DOIUrl":"10.1016/j.rsma.2025.104458","url":null,"abstract":"<div><div>Coastal regions are subject to intense anthropogenic pressures, and Antarctica, despite its low human occupancy, is not exempt from these impacts. This study examines the occurrence and distribution of solid waste at Punta Elefante, Livingston Island, in Maritime Antarctica during the 43rd Brazilian Antarctic Operation. A total of 189 waste items were identified, with plastics being the most prevalent (96.3%). The waste characterization revealed that most items measured between 10–50<!--> <!-->cm (36.5%) and less than 5<!--> <!-->cm (30.7%), indicating ongoing fragmentation. Spatial dispersion patterns suggest the influence of ocean currents and winds. Beyond environmental contamination, interactions with biodiversity were documented, including fungi and plants colonizing plastic waste and seabirds ingesting plastic fragments. These findings underscore the need for mitigation strategies such as continuous monitoring, waste removal, and stricter regulation of human activities. The persistence of plastic debris in Antarctica poses a long-term ecological threat, necessitating collaborative efforts to mitigate its impact on these fragile ecosystems.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104458"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Water level predictions with machine learning in enclosed, semi-enclosed, inland and open (marginal) seas","authors":"Mehmet Ozturk,&nbsp;Furkan Altas","doi":"10.1016/j.rsma.2025.104448","DOIUrl":"10.1016/j.rsma.2025.104448","url":null,"abstract":"<div><div>In this study, we present a novel methodology to predict water levels (WLs) from its components (meteorological, hydrological, tidal, seawater, and air parameters) by using a machine learning (ML) approach. The novelty of the approach is 1) random selection of training data, 2) to determine significant predictors and eliminate the others by applying the t-test statistic, and 3) to check linear (via Variance Inflation Factor (VIF)) and nonlinear (via Distance Correlation (DisCor)) dependencies between the significant predictors to avoid using similar parameters. Various linear and non-linear ML approaches are employed to obtain the optimal regression model. We tested the prediction capacities of the model in four coastal seas: the enclosed Black Sea, the inland Sea of Marmara, the semi-enclosed Aegean Sea, and the open (marginal) Levantine Sea. We achieved the most significant correlations for the Ensemble of Trees (ET) approach in the Black Sea and Levantine Sea WL predictions, with correlation coefficients of R≅ 0.93–0.95 in the former and R≅ 0.89–0.93 in the latter.</div><div>On the other hand, the Gaussian Process Regression (GPR) yielded the most significant correlations in the Sea of Marmara (R≅0.93) and the Aegean Sea (R≅0.82). Water level (WL) variability was primarily driven by Danube River discharge and air pressure in enclosed and inland areas, the air pressure and tidal forces in semi-enclosed seas, and predominantly by tidal influences in open sea regions. Two parameters (sea surface temperature (SST) and 2-m air temperature (2m-T)) were relevant in all locations considered after applying the VIF test, and the lower-correlated predictor was excluded from predictions. Overall, the proposed method notably increased the predicted WLs' accuracy compared to previous ML studies.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104448"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144996713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"18S rRNA gene sequences of intestinal contents of Sebastiscus marmoratus in the coastal waters of the Ma’an Archipelago in the East China Sea","authors":"Kai Wang ,&nbsp;Jianing Sun ,&nbsp;Yuqing Wang ,&nbsp;Shouyu Zhang ,&nbsp;Yuheng Zhou ,&nbsp;Meiping Feng","doi":"10.1016/j.rsma.2025.104421","DOIUrl":"10.1016/j.rsma.2025.104421","url":null,"abstract":"<div><div><em>Sebastiscus marmoratus</em> is one of the key commercial species inhabiting the rocky reef habitats of the Ma’an Archipelago in the East China Sea, a national marine ranching area in China. Understanding its intestinal contents and feeding ecology is crucial for ecological functioning evaluation, yet 18S rRNA sequencing data remain limited. Here, we explored the feeding habits of wild <em>S. marmoratus</em> across different body length (90–166 mm) and genders using 18S rRNA sequencing. A total of 71 species from 55 genera and15 phyla were identified in intestinal contents. Arthropods dominated the diet (39.48 %), with fish, crustaceans, and cnidarians comprising the primary prey groups. Ontogenetic shifts were evident, with the proportion of crabs and cnidarians initially decreasing, then increasing with body length. Sexual dimorphism in feeding patterns was observed, with males exhibiting higher biodiversity than females. Significant differences were observed in intestinal community composition between genders (<em>P</em> &lt; 0.05). Potential prey items were also observed, including Neoloricata and gelatinous organisms (e.g. Actiniaria and Leptothecata). These findings reveal size- and gender-specific feeding strategies of <em>S. marmoratus</em>, which contribute to the understanding of its ecological role in the Ma’an Archipelago's marine ranching area.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104421"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144922875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling spatial and monsoonal dynamics of gelatinous zooplankton in a tropical estuary","authors":"Hattaya Jitrapat ,&nbsp;Pramot Sojisuporn ,&nbsp;Porntep Punnarak ,&nbsp;Ajcharaporn Piumsomboon","doi":"10.1016/j.rsma.2025.104461","DOIUrl":"10.1016/j.rsma.2025.104461","url":null,"abstract":"<div><div>Gelatinous zooplankton (GZ) frequently exhibit episodic blooms, intensifying their role in biogeochemical cycling within marine ecosystems. Despite growing concerns about fluctuations in GZ abundance, data on their spatiotemporal variations and species distributions remain limited. To address this gap, a study on the spatial and monsoonal variations of GZ species was conducted in a tropical estuary to identify distribution patterns and associated environmental factors. The study took place in the Inner Gulf of Thailand, with sampling sites categorized into upper (nearest to the river mouths), middle, and lower (farthest from the river mouths) areas over three sampling months in 2017 and 2018. Results revealed that the distribution patterns of GZ varied with distance from the river mouths, aligning with freshwater intrusion. The upper area supported higher GZ abundance and diversity than the middle and lower areas. Temporal aggregation of GZ assemblages was more pronounced in the middle to lower areas, where physicochemical conditions were relatively more stable than at the river mouth stations. Key contributors to the GZ assemblages included the chaetognath <em>Flaccisagitta enflata</em>, the doliolid <em>Doliolum nationalis</em>, and the hydromedusa <em>Liriope tetraphylla</em>. Additionally, certain species such as the siphonophore <em>Diphyes chamissonis</em>, the salp <em>Thalia democratica</em>, and the shelled pteropod <em>Creseis acicula</em> emerged occasionally. Among the measured variables, salinity, turbidity, and chlorophyll <em>a</em> concentration showed the strongest correlations with GZ abundance. Notably, <em>L. tetraphylla</em> dominated the deoxygenated water at the river mouth, comprising up to 70 % of the total GZ abundance. These findings provide insight into GZ distribution in tropical estuaries, highlighting the close linkage to land-based runoff and monsoonal forces. Monitoring GZ populations may, therefore, serve as a valuable tool for detecting environmental changes.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104461"},"PeriodicalIF":2.4,"publicationDate":"2025-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144996809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Abundance and distribution of microplastics in sediments of the Eastern Aegean Sea (Izmir Bay, Türkiye)","authors":"Ebru Yeşim Özkan ,&nbsp;Ayşah Öztekin ,&nbsp;Levent Bat ,&nbsp;Nusret Kaya","doi":"10.1016/j.rsma.2025.104457","DOIUrl":"10.1016/j.rsma.2025.104457","url":null,"abstract":"<div><div>Microplastics (MPs) have accumulated extensively in the marine environment due to the increasing use of plastic materials and their inadequate disposal. Several investigations have been focused on the presence of MPs in the marine environment but a few of them investigate the presence of MPs in the sediment profiles and distribution dynamics. The aim of the study was to investigate the abundance of MPs in the seabed sediments in the İzmir Bay of the Aegean Sea, Türkiye and to determine the distribution dynamics of MPs and the factors governing them. The abundance of MPs was 2979 ± 1560 pieces/kg dry sediment. The dominant MP shape was fragment (59.54 %), followed by fiber (36.77 %), film (3.55 %), the remaining types were found to be less than 1 %. The majority of MPs are smaller than 1 mm. In the distribution of MPs by color, the dominant color was transparent (52 %) followed by blue with 11.07 % and other colors account for less than 10 %. It was determined from FT-IR and Raman analyses that the sediments of the Inner Gulf chemically contain 7 different MPs and polyethylene (PE) and polypropylene (PP), constituting the majority of identified particles. The distribution of MPs in the sediment may be explained by Fe, Mn, Calcite (CaCO₃), Biogenic Silica (BSi) and hydrodynamic features in the water column (cyclonic, anticyclonic eddies in the water column, diatom substrate selection, littering and sinking with density increase).</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104457"},"PeriodicalIF":2.4,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144925933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A coastline automatic extraction and classification method by using multi-source remote sensing and DEM data","authors":"Dongyue Chai ,&nbsp;Zhixiang Fang ,&nbsp;Fenfang Wu","doi":"10.1016/j.rsma.2025.104450","DOIUrl":"10.1016/j.rsma.2025.104450","url":null,"abstract":"<div><div>The efficient coastline extraction and classification are essential for environmental monitoring, disaster prevention, and coastal planning. However, the extraction and classification of coastlines from remote sensing imagery remains challenging, especially in complex environmental conditions with varying textures, waves, and tidal flats or man-made structures. This study aims to address these challenges by developing a high-precision framework for automated coastline extraction and classification. We developed an innovative approach that integrates Sentinel-1 SAR Ground Range Detected (GRD), Sentinel-2 Multispectral Instrument (MSI), and Copernicus Digital Elevation Model (DEM) data with a flexible region-growing algorithm, enabling effective adaptation to complex and diverse coastal environments. The interactive effect between spectral and shape features was used to enhance coastline classification. Three large-scale ports—Sanya Port in China, Los Angeles Port in the United States, and Botany Port in Australia were selected as experimental areas. Ground-truth coastline samples were manually annotated at these ports. The proposed method was subsequently compared with the SAET model, which employs multiple water indices and K-means clustering to extract sub-pixel-level coastlines, as well as the Segment Anything Model (SAM). The proposed method achieved over 91 % extraction accuracy with an average positional error under 6 m. Compared with traditional approaches, accuracy improved by 10.28 %, 29.29 %, and 1.2 %, respectively. For classification, Random Forest outperformed other models, and the inclusion of feature interaction and contextual updating improved overall classification accuracy to 82.2 %, representing a 4.5 % increase over baseline models. This study presents a robust and automated approach for coastline extraction and classification in complex environments, enhancing the applicability of remote sensing for coastal analysis and management.</div></div>","PeriodicalId":21070,"journal":{"name":"Regional Studies in Marine Science","volume":"90 ","pages":"Article 104450"},"PeriodicalIF":2.4,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144916489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}