environmental

Environmental studies 2026

Authors: Samah Dib, Arafa Mohamed, Fatima A. Al- Khayat, S. Veerasingam, V. M. Aboobacker, Jassim A. Al-Khayat & P. Vethamony This study examined microplastic (MP) accumulation in the gastrointestinal tracts of 170 fish from four commercially important species in Qatar. MP-like particles were extracted through chemical digestion and density separation, then analyzed via stereomicroscopy, revealing fibers as the dominant shape and blue as the most common color. Spectroscopy confirmed only 7 particles (4.12% of samples) as true MPs, matching >70% with polymer databases and identifying polyethylene (PE) and polypropylene (PP). The mean MP abundance was 0.070 ± 0.090 MPs/g of fish gut, relatively low compared to other Gulf and global regions.

01 - Distinguishing microplastics from microplastic-like particles in the marine fish from Qatar

02 - A machine learning-driven framework for enhancing underwater visual signal processing in marine ecosystem economic monitoring and anthropogenic impact assessment

Authors: Wang Minglong, Zhu Feng, and Jian Hu Underwater monitoring faces challenges from optical distortions, ecological variability, and dynamic behaviors, which conventional image processing methods struggle to address due to issues like spectral attenuation and scattering. Traditional terrestrial-based approaches are ineffective for tasks such as species tracking and habitat mapping. To resolve this, a machine learning framework combining physics-aware modeling and ecological adaptivity is proposed, featuring the BOANE module for correcting radiance distortions and encoding biological information, and the CAMSE module for real-time enhancement via ecological priors, feature alignment, and semantic filtering.

03 - Spatio-temporal dynamics of fish assemblages at an offshore wind farm and a comparison with mature artificial reefs

Authors: Kwang Tsao Shao, Ching Han Chang, Ching Yi Chen, Lin Tai Ho, Yi Ta Shao, Hsin-Chieh Chiang, Wei-Ning Lee, Hsien Ju Tsai, Tien-Yu Huang, and Chun-Yu Chang Offshore wind farms (OWFs) create new habitats for fish via their underwater foundations, but their ecological role compared to artificial reefs (ARs) remains uncertain. A multi-year study at Taiwan’s Formosa Wind Farm found 86 reef-associated fish species exclusively within 50 m of turbine foundations, absent in surrounding areas or prior records. Fish assemblages changed over time, and while species composition differed from established ARs in 2025, key ecological metrics (richness, diversity, trophic structure) showed no significant differences.

04 - Innovation below the surface: development of a canine underwater search training device for submerged scent detection

Authors: Paul Bunker, and Christina Brewster This paper introduces the Canine Underwater Search Training Device (CUSTD), a novel remotely operated system that safely delivers target odors from the front of a boat, conditioning canines to search and indicate detection from that position without contaminating water sources. This innovative methodology offers broad applications in environmental monitoring, oil spill response, search and recovery, and marine biology research, providing a versatile and ethical platform for advancing both scientific study and operational canine deployment in underwater detection.

05 - A global dataset of impact forces from submarine landslides on pipelines and cables

Authors: Xiaolei Liu, Shuzhou Wei, Xiangshuai Meng, Botao Xie, Xuejian Chen & Xingsen Guo Submarine pipelines and cables, vital for offshore energy and global communications, face growing threats from submarine landslides that can damage infrastructure. Existing research on landslide-pipeline interactions is fragmented and lacks standardized data, hindering comparative analysis. To address this, the authors compiled a dataset of 864 entries from 24 studies, capturing key parameters like impact velocity, flow type, and drag/lift forces. Standardizing force definitions and categorizing conditions by Reynolds number regimes enhances comparability. This dataset serves as a critical resource for risk assessment, offshore design, and modeling pipeline-landslide interactions.

06 - Research on a Lightweight Detection Method for Underwater Diseased Corals

Authors: Mingqi Li and Ming Chen This study addresses challenges in underwater diseased coral detection, such as high variability, small pathological regions, and environmental interference. A novel lightweight method, CD-YOLO, is proposed. Key improvements include: 1) a lightweight CDShufflenet backbone to reduce model complexity while maintaining performance; 2) a SPDConv downsampling module to preserve fine-grained coral details; and 3) integrated attention mechanisms (EMA and SENetV2) to enhance pathological feature representation and suppress environmental noise.

07 - The Greater Agulhas Current System – Circulation, Variability, Long-Term Trends and Impacts on Weather, Climate and Ecosystems.

Authors: Issufo Halo et al. The Greater Agulhas Current System (GACS) is essential for the exchange between the Indian and Atlantic Oceans and influences weather, climate, and marine ecosystems in southern Africa. Since the last major publication nearly twenty years ago, advancements in technology and research have improved understanding of GACS from 2006 to 2025. Key developments include new observational tools and models, enhanced knowledge of transport and eddy dynamics, and links to Southern Hemisphere wind patterns. The review highlights knowledge gaps and suggests future research needs, focusing on continuous observations and better modeling of fine-scale processes.

08 - Using surface drifters to characterise near-surface ocean dynamics in the southern North Sea: a data-driven approach

Authors: Jimena Medina-Rubio, Madlene Nussbaum, Ton S. van den Bremer, and Erik van Sebille Traditional drifters struggle to replicate the transport of buoyant ocean-surface objects due to complex interactions with wind, currents, and waves. To address this, 12 ultra-thin drifters were deployed in the southern North Sea for 68 days, and their velocities were modeled using hydrodynamic and atmospheric data using linear and machine-learning (random forests, support vector regression) approaches. Analysis revealed tidal forcing as the dominant driver of zonal motion, while wind primarily influenced meridional movement, though its effect saturated at higher speeds.

09 - Research Progress and the Prospect of Artificial Reef Preparation and Its Impact on the Marine Ecological Environment

Authors: Hao-Tian Li, Ya-Jun Wang, Jian-Bao Zhang, Peng Yu, Yi-Tong Wang , Jun-Guo Li, Shu-Hao Zhang, Zi-Han Tang, and Jie Yang Artificial reefs are widely used globally for marine ecological restoration and fishery enhancement, with countries like Japan, the U.S., China, and others leading research and implementation. This paper reviews recent advancements in reef materials, including concrete alternatives such as steel slag, blast furnace slag, sulfoaluminate cement, and silica fume, which improve mechanical properties (e.g., up to 20% higher compressive strength) and neutralize surface pH, fostering marine organism adhesion.

10 - More eddying of subtropical western boundary currents boosts stratification and cools shelf seas

Authors: K. L. Gunn, & L. M. Beal Climate change is intensifying eddies in western boundary currents, driving cross-slope fluxes that alter shelf conditions. Agulhas Current mooring data show eddies converging heat and salt to the core, cooling shelves while broadening and stratifying the current. Inshore 10-km instabilities pump nutrients up; offshore 100-km meanders transport heat, accelerating surface warming but cooling at depth. Similar trends imply more extremes in subtropical shelf seas despite steady currents.

11 - Mangrove Ecosystems: Importance, Threats and Opportunities for Restoration

Authors: Elijah I. Ohimain, Robert Eugene Turner, and Beth A. Middleton Mangroves are vital for biodiversity, coastal protection, local livelihoods, and climate mitigation. They help shield coasts from storms and rising sea levels. However, their health is threatened by infrastructural development, urban expansion, aquaculture, farming, and oil and gas exploration. This review focuses on the threats and restoration opportunities for African mangroves, which suffer significantly from oil spills. A key challenge in restoration is reestablishing suitable hydrologic and salinity conditions before natural regeneration or planting propagules can occur.

12 - Wind as an Influential Factor in the Transport and Destination of Oil from Spills Along the Brazilian Semiarid Coast (Ceará State, Northeast Brazil)

Authors: Alexandre Medeiros de Carvalho et al. (10 scientists) Oil spills along Brazil's northeast coast pose severe risks to coastal environments and ecosystems. This study assessed wind-driven transport of oil fragments inland using field observations and GIS tools, finding that wind can carry oil over hundreds of meters. Key factors like exposed beachrock, low dunes, and high wind potential trap oil on beaches, fragmenting it and enabling deeper inland contamination. The findings highlight the need to evaluate risks as oil integrates into aeolian and inland systems.

13 - Efficiency Assessment of Crude Oil Contamination Remediation Using Green Surfactants and Biofoam Material: A Case Study of the Bodo Region, Nigeria

Authors: Kabari Visigah, Dongmei Wang, Jin Zhang, and Surojit Gupta This study evaluates the effectiveness of green surfactants combined with a lignin-based biofoam to remediate crude oil contamination in a simulated mangrove environment in Bodo, Niger Delta, Nigeria. Four soil types (sand, mud, peat, and peat–mud) were tested. Key findings include: (1) the best-performing surfactants reduced interfacial tension to 10-¹ mN/m, significantly improving contaminant mobilization and extraction, particularly in sand and mud samples, and (2) the biofoam achieved a 40% oil absorption rate, with only brine water remaining in the contaminated oil. The results highlight a promising, sustainable approach for cleaning oil-polluted muddy soils.

14 - Near-Bottom ROV-Borne Self-Potential Exploration of Seafloor Massive Sulfi de Deposits on the Southwest Indian Ridge

Authors: Zuofu Nie, Chunhui Tao, Zhongmin Zhu, and Jianping Zhou ROV-borne SP surveys at ~10 m altitude over the Yuhuang field on the Southwest Indian Ridge mapped SMS mineralization via adaptive octree meshing and 3D inversion of current density. Results identified a main anomaly zone matching known deposits plus an additional potential zone, consistent with independent observations and drilling. The method provides a reliable approach for imaging deep-sea SMS mineralization.

15 - Specific Oil Detection by Canines: Discrimination of Fresh Spill Hydrocarbons from Weathered Background Oil in Coastal Environments

Authors: Paul Bunker, and Ed Owens This study found that specially trained canines can distinguish between unweathered oil and naturally weathered oil (e.g., tarballs) in shoreline environments. Two canines, trained solely on unweathered oil samples, consistently detected fresh oil while ignoring naturally weathered oil. This targeted detection method could improve spill response efficiency by rapidly identifying new contamination, thereby reducing costs, response time, and environmental damage.

16 - MorphoCal: a multi-stage deep learning framework for fish length estimation in challenging underwater pond environments

Authors: Vijayalakshmi M., Sasithradevi A., Sabari Nathan, and P. Prakash Underwater imaging is crucial for monitoring aquatic ecosystems, but challenges like turbidity and uneven lighting hinder accurate fish pose and length estimation. To address this, the study introduces MorphoCal, a deep learning framework combining AquaYOLO-PoseCA—a YOLO-based keypoint detection model enhanced with coordinate attention—for improved fish anatomy localization. The framework integrates single-shot calibration and ray-plane projection for centimetre-level length reconstruction. On the DePondFi'24 dataset, AquaYOLO-PoseCA achieves high accuracy with efficient computation, and its length reconstructions closely match ruler-based measurements.

17 - Marine Geographic Information Systems, Spatial Analysis Tools in the Management Process of Spanish Marine Protected Areas

Authors: Dulce Mata, Paula Gil, Ángela Bellido, and Olvido Tello Spain’s vast marine jurisdiction requires robust geospatial frameworks for ecosystem assessment and policy. This study presents IEO-CSIC GIS methodologies developed for LIFE IP INTEMARES and MSFD implementation. Workflows integrate diverse datasets (bathymetry, ROV transects, sediments, species records) using terrain modeling, geomorphometric indices, classification, and spatial statistics to quantify habitats and pressures. Integrated analysis supports MPA zoning, while WebGIS platforms enhance data access and engagement. Outputs deliver high-resolution MSFD-aligned maps and indicators, demonstrating GIScience’s value for reproducible marine monitoring and management.

18 - Wave Climate Trends and Teleconnections in the Gulf of Mexico and the Caribbean Sea

Authors: Miqueas Diaz-Maya, Marco Ulloa, and Rodolfo Silva This study analyzes wave climate trends (1981–2022) in the Gulf of Mexico and Caribbean Sea via WAVEWATCH III simulations validated against buoys. Using Mann–Kendall and Theil–Sen tests, it finds basin-wide rises in significant wave height (Hs) and wave power (P) with strong spatial/seasonal variation—most robust in winter/spring, weaker or negative in summer/fall. ENSO is the main driver of Caribbean interannual variability via remote swell, NAO exerts regional storm-track influence, and AMO plays a secondary role; the Gulf shows limited climate-mode sensitivity, dominated by local wind-sea processes. Results inform coastal hazards, shipping, and wave energy planning.

19 - Assessing the Impact of Soil Hydrocarbon Properties on Plant Functional Types Using Hyperspectral Data in the Niger Delta

Authors: Abdullahi A. Kuta, Stephen Grebby, Doreen S. Boyd, and Christopher H. Vane This study examined how soil hydrocarbon parameters (TPHs, TOC, and soil toxicity) affect vegetation in the Niger Delta using hyperspectral leaf-scale data and vegetation indices (mND705, PRI, NDVVI844,447, and MDATT). Analysis of five vegetation types revealed species-specific responses to contamination: mangrove vegetation was most affected by total petroleum hydrocarbons (R = −0.683), while mango vegetation showed greatest sensitivity to organic carbon (R = −0.725) and soil toxicity (rs = 0.870). The findings demonstrate that red-edge- based hyperspectral techniques effectively assess vegetation stress in contaminated coastal ecosystems.

20 - YOLOv8m-CGSE: An Improved Lightweight YOLOv8m for Marine Oil Spill Detection

Authors: Qingyang Wang, Junjie Lu, Bin Yang, Chen Jiao, Tao Yue, Bo Song, Jianwu Jiang, Guoqing Zhou, and Jingwen Li This study proposes YOLOv8m-CGSE, an improved YOLOv8m model for UAV oil spill detection. It replaces convolutions with GSConv, substitutes C2f with SENetV2, and adds lightweight CCFM for multi-scale features. With Mosaic augmentation, it achieves mAP50 of 91.2% and mAP50-95 of 73.3%, reducing parameters by 16.1% and computation by 12.6%. The model also cuts false positives from 21 to 15 on deceptive sea surfaces, balancing precision, robustness, and efficiency for real-time monitoring.

21 - Evolution and Challenges of Marine Oil Spill Governance in Taiwan over Two Decades

Authors: Chih-Wei Chang, Shiau-Yun Lu, Chun-Pei Liao, Wen-Yan Chiau, and Yi-Che Shih This study analyzes two decades of marine oil spill governance in Taiwan via four key cases, comparative analysis, stakeholder interviews, policy review and international comparisons. As a non-UN member, Taiwan encounters unique barriers engaging conventions like MARPOL 73/78. Findings reveal gaps in decision-support tools, fragmented agency coordination and inadequate compensation laws. The paper also proposes a phased unified task force, data-driven contingency plans and alignment of the Marine Pollution Control Act with global standards through sub-national partnerships.

22 - Oil Spill Segmentation in Marine Radar Imager via an Enhanced GA-RBF-MBO Hybrid Approach

Authors: Jin Xu, Bo Xu, Jin Yan, Lihui Qian, Boxi Yao, Zekun Guo, Minghao Yan, and Peng Liu Global maritime trade growth has increased oil spill risks, threatening ecosystems. An improved Monarch Butterfly Optimization (MBO) algorithm was proposed for accurate oil spill segmentation in radar images. Preprocessing enhanced contrast via grayscale conversion and background removal; a genetic algorithm optimized an RBF network to extract ROIs in 3D feature space. The MBO employed a multi-objective fitness function with dynamic parameter adjustment, reverse learning, and elite reproduction for threshold optimization.

23 - Antibiotic residues in aquatic environment: ecological disruption, and biotechnological solutions for environmental safety

Authors: Mahipal Singh Sankhla, Negaa, Vaibhav Sharma, Rajeev Kumar, Garima Awasthi, Raj Shukla, Baljeet Yadav, and Kumud Kant Awasthi Antibiotic residues are emerging pharmaceutical contaminants in aquatic ecosystems, originating from human and veterinary medicine and agriculture. Inadequate wastewater treatment allows these residues to contaminate drinking and groundwater. Continuous exposure to trace antibiotic levels causes long-term harm to aquatic and terrestrial life, promoting the development and spread of antibiotic-resistant bacteria (ARB) and resistance genes (ARGs), which disrupts microbial communities. This study examines global antibiotic occurrence trends from 2004–2025 across various water bodies.

24 - Gauging the Effectiveness and Translatability of Oil Spill Response Technologies to Plastic Pellet Spills

Authors: Marko Jugo, Christopher M. Reddy, Bryan D. James, and Tarzan Legovic Plastic pellet spills are a growing environmental concern with limited response strategies. This study evaluates repurposing oil spill tools like booms and skimmers for pellet recovery at sea, noting how pellet properties influence dispersion and cleanup. It identifies limitations from oil-centric designs and proposes cross-sector strategies, validated by a successful field test in the Northern Adriatic Sea.

25 - Cellulosic Absorbent Materials for Oil Spill Response: A Review

Authors: Adilet Nogayev, Kenes Kudaibergenov, Aliya Togasheva, Akshyryn Zholbasarova, Ryskol Bayamirova, Bakytzhan Burkhanov, Ainagul Abdygaliyeva, and Nurzhamal Zhumagaliyeva Plastic pellet spills are a growing environmental concern with limited response strategies. This study evaluates repurposing oil spill tools like booms and skimmers for pellet recovery at sea, noting how pellet properties influence dispersion and cleanup. It identifies limitations from oil-centric designs and proposes cross-sector strategies, validated by a successful field test in the Northern Adriatic Sea.

26 - Unsupervised Oil Spill Detection in Shipborne Radar Imagery Using Autoencoder-Enhanced Q-Learning and Improved Bat Optimization

Authors: Jin Yan, Binghui Chen, Jin Xu, Zekun Guo, Minghao Yan, Mengxin Sun, and Lin Qiao Marine oil spills threaten environments, requiring efficient detection amid radar image issues like blurred boundaries, interference, and noise. This study proposes a method from Dalian Bay data integrating autoencoder feature extraction, PCA reduction, K-Means pseudo-labeling, Q- learning for ROI extraction, and an improved bat algorithm for segmentation. Experiments outperformed classic algorithms; ablation verified components, offering a new offshore monitoring approach.

27 - Two-Stage Oil Spill Detection in SAR Using a Domain-Adapted Segment Anything Model

Authors: George Giannopoulos, Maria Kremezi, Vasilia Karathanassi, Vassilis Andronis, Dimitris Bliziotis, Katerina Kikaki, Ana Sofia Oliveira, and Ariane Müting This study presents a two-stage deep learning method for mapping oil spills using Sentinel-1 SAR images. In the first stage, a ConvNeXt-T classifier identifies areas likely containing slicks. In the second stage, a modified Segment Anything Model (SAM) performs segmentation to find spill boundaries using Sentinel-1 VV backscatter and GLCM texture features. The adapted SAM achieves an F1-score of 0.86, surpassing traditional models. Wind speed impacts detection but does not solely determine segmentation quality.

Document published by

Frontiers

Document published by

Scientific reports (Nature)

Document published by

Frontiers

Document published by

Frontiers

Document published by

Frontiers

Document published by

Scientific data (Nature)

Document published by

MDPI

Document published by

Egusphere / Doi Foundation

Document published by

Ocean science - European Geosciences Union

Document published by

MDPI

Document published by

Nature Climate Change

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

Frontiers

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI

Document published by

MDPI