Search Results (143)

Saxitoxin (STX) is one of the most potent marine neurotoxins, produced by several species of freshwater cyanobacteria and marine dinoflagellates. Although omics-based approaches have advanced our understanding of STX biosynthesis in recent decades, the origin, regulation, and ecological drivers of STX in dinoflagellates remain poorly resolved. Specifically, dinoflagellate STX biosynthetic genes (sxt) are extremely fragmented, inconsistently expressed, and unevenly distributed between toxic and non-toxic taxa. Environmental studies further report inconsistent relationships between abiotic factors and STX production, suggesting regulation across multiple genomic, transcriptional, post-transcriptional, and epigenetic levels. These gaps prevent a comprehensive understanding of STX biosynthesis in dinoflagellates and limit the development of accurate predictive models for harmful algal blooms (HABs) and paralytic shellfish poisoning (PSP). Artificial intelligence (AI), including machine learning and deep learning, offers new opportunities in ecological pattern recognition, molecular annotation, and data-driven prediction. This review explores the current state of knowledge and persistent knowledge gaps in dinoflagellate STX research and proposes an AI-integrated multi-omics framework highlighting recommended models for sxt gene identification (e.g., DeepFRI, ProtTrans, ESM-2), evolutionary reconstruction (e.g., PhyloGAN, GNN, PhyloVAE, NeuralNJ), molecular regulation (e.g., MOFA+, LSTM, GRU, DeepMF), and toxin prediction (e.g., XGBoost, LightGBM, LSTM, ConvLSTM). By integrating AI with diverse biological datasets, this novel framework outlines how AI can advance fundamental understanding of STX biosynthesis and inform future applications in HAB monitoring, seafood safety, and PSP risk management in aquaculture and fisheries. Full article

Harmful algal blooms (HABs) represent an escalating threat in marine and coastal ecosystems, posing increasing risks to ecological balance, public health, and blue economy industries including fisheries, aquaculture, and tourism. This review examines the impact of climate change and anthropogenic pressures on the escalation of HAB occurrences, focusing especially on vulnerable regions in Mexico, which are the primary case study for this investigation. The methodological framework integrates HAB risk assessment (RA) methods found in the literature. Progress in detection and monitoring technologies—such as sensing, in situ sensor networks, and prediction tools based on machine learning—are reviewed for their roles in enhancing early-warning systems and aiding decision support. The key findings emphasize four linked aspects: (i) patterns of HAB risk in coastal zones, (ii) deficiencies and prospects in HAB-related policy development, (iii) how governance structures facilitate or hinder effective actions, and (iv) the growing usefulness of online monitoring and evaluation tools for real-time environmental observation. The results emphasize the need for coupled technological and governance solutions to reduce HAB impacts, protect marine biodiversity, and enhance the resilience of coastal communities confronting increasingly frequent and severe bloom events. Full article

Introduction: Traditional approaches to discover teleconnections and quantify uncertainty, such as global sensitivity analysis, Monte Carlo experiments, decomposition analysis, etc., are computationally intractable for large-scale process-based Coupled Human and Natural Systems (CHANS) models. This study hypothesizes that machine-learned emulator models provide “computationally efficient” algorithms for discovering teleconnections and quantifying uncertainty within and across dynamically evolving human and natural systems. Objectives: This study aims to harness machine-learned emulator models to discover the relative contributions of internal- versus external-to-the-lake teleconnected processes driving the emergence of Harmful Algal Blooms (HABs) and trophic regime shifts. Three objectives are pursued: (1) build emulators; (2); quantify uncertainty and (3) identify teleconnections. Methods: Six machine-learned emulator models are trained on ~3.8 million observations for ~52 features derived from 332 scenarios simulated in an integrated process-based CHANS model that predicts water quality in Missisquoi Bay of Lake Champlain under alternate hydro-climatic and nutrient management scenarios for the 2001–2047 timeframe. The regression random forest (RRF), regression LightGBM (RLGBM) and regression XGBoost (RXGB) models predict the average surface mean of ChlA. Further, the classifier random forest (CRF), classifier LightGBM (CLGBM) and classifier XGBoost (CXGB) predict four trophic states of Missisquoi Bay. Relative importance and partial dependence plots are derived from all six emulator models to quantify relative uncertainty and importance of external-to-the-lake (climatic, hydrological, nutrient management) and internal-to-the-lake (P and N sediment release) drivers of HABs. Results: RXGB (R² = 96%, 48 features) outperforms RLGBM (R² = 95%, 37 features) and RRF (R² = 93%, 20 features) in predicting the average surface mean of ChlA. CLGBM (F1 = 96.15, 4 features) outperforms CXGB (F1 = 95.66, 48 features) and CRF (F1 = 93.06, 23 features) in predicting four trophic states. We discovered that predictor variables representing snow, evaporation and transpiration dynamics teleconnect hydro-climatic processes occurring in terrestrial watersheds with the biogeochemical processes occurring in the freshwater lakes. Conclusions: The proposed approach to discover teleconnections and quantify uncertainty through machine-learned emulator models can be scaled up in different watersheds and lakes for informing integrated water governance processes. Full article

The impact of nitrogen on harmful algal blooms (HABs) and functions of biota in marine ecosystems under eutrophication is a topical issue of growing importance. The article aimed at describing the diversity of planktonic bloom-forming dinoflagellates in the SW Baltic Sea coastal waters under variable eutrophication. The analysis of 44 year-long database revealed 82 dinoflagellate species and demonstrated diversity patterns of ten common bloom-forming species, including seven mixotrophs from the genera Prorocentrum, Dinophysis, and Ceratium, under variable eutrophication evaluated using total nitrogen (TN) content in water. Based on the Intermediate Disturbance Hypothesis (IDH), we presumed those coastal waters with total nitrogen concentrations that are optimal to dinoflagellates to host greater taxonomic diversity compared to areas with non-optimum TN content. The results showed that the highest dinoflagellate species richness was associated with much lower TN concentrations than the optimum values for these species. Thus, our findings disagreed with the IDH. We suggested and discussed possible reasons of this inconsistency, including algal growth rates and disturbance frequency. We also updated the classification of eutrophication levels in the Baltic Sea based on the distribution of TN content and diversity of HAB-forming dinoflagellates. The results can contribute to predictive assessment of HABs under growing eutrophication. Full article

Accurate monitoring of chlorophyll-a (Chl-a) is essential for managing coastal aquaculture, as Chl-a indicates phytoplankton biomass and water quality. This study developed a hybrid deep learning model integrating convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and an attention mechanism (Attention) to classify Chl-a using hourly, water quality datasets collected from the GOT001 station in Si Racha Bay, Eastern Gulf of Thailand (2020–2024). A random forest (RF) identified sea surface temperature (SEATEMP), dew point temperature (DEWPOINT), and turbidity (TURB) as the most influential variables, accounting for over 90% of the accuracy. Chl-a concentrations were categorized into ecological groups (low, medium, and high) using quantile-based binning and K-means clustering to support operational classification. Model performance comparison showed that the CNN–BiLSTM model achieved the highest classification accuracy (81.3%), outperforming the CNN–LSTM model (59.7%). However, the addition of the Attention did not enhance predictive performance, likely due to the limited number of key predictive variables and their already high explanatory power. This study highlights the potential of CNN–BiLSTM as a near-real-time classification tool for Chl-a levels in highly variable coastal ecosystems, supporting aquaculture management, early warning of algal blooms or red tides, and water quality risk assessment in the Gulf of Thailand and comparable coastal regions. Full article

This study presents an explainable artificial intelligence (XAI)-based explainable planning management (EPM) framework designed to provide interpretable prediction-driven insights for water quality management. Although deep learning models such as the multi-layer perceptron (MLP) effectively predict water quality indicators, they have limited interpretability and practical use. To address this limitation, Shapley additive explanations (SHAP) were applied to quantify each input feature’s contribution to model-predicted chlorophyll-a (Chl-a) values and to support the construction of scenario-based analyses. The proposed framework was applied at the Dasan water quality observation station in the Nakdong river basin, Republic of Korea. Daily water quality data from 2014 to 2023 were used for model training, and 2024 data were used for prediction. The model excluding turbidity achieved the lowest root mean squared error (RMSE) of 7.3922. Scenario analyses were performed by varying Chl-a(t−1) and major variables in 10% increments, guided by influence identified through SHAP analysis. Results indicated that pH, which had the highest Shapley value excluding Chl-a(t−1), was the most influential variable, reducing algal bloom warning occurrences by up to 34%. These results demonstrate that the proposed EPM framework enhances interpretability and supports the exploration of prediction-based planning strategies, without implying causal or mechanistic relationships among water quality variables. Full article

Chlorophyll-a (Chl-a) concentration is a key parameter for assessing the degree of eutrophication and the algal bloom risk in water bodies. Accurate and robust monitoring of Chl-a is crucial for effective water quality management of inland and coastal optically complex Case-II waters. This study proposes a stratified integrated framework that combines optical water type (OWT) classification and multi-index blending models and evaluates the capability of ZY-1 02E hyperspectral imagery in the retrieval of Chl-a concentration in Case-II waters. This research is based on ZY-1 02E hyperspectral remote sensing images and ground synchronous measurement data from four typical water bodies in China (Dongpu Reservoir, Nanyi Lake, Tangdao Bay, and Moon-lake Reservoir). Using Fuzzy C-Means (FCM) clustering combined with spectral feature analysis, three different OWTs were identified, and the bands sensitive to Chl-a for each water type were recognized. Subsequently, the most suitable semi-empirical indices (BR, TBI) were selected, and a new suspended matter correction index (SMCI) was constructed by integrating spectral bands and TSM data specifically for high-turbidity waters to facilitate the retrieval of Chl-a concentration. The RMSE and MAPE of the model constructed based on the unclassified dataset were 3.1586 μg·L⁻¹ and 30.82%, respectively. When the stratified ensemble method based on optical water type classification was employed, the overall RMSE and MAPE were reduced to 1.5832 μg·L⁻¹ and 16.36%. The results demonstrate that this hierarchical ensemble framework significantly improved the retrieval accuracy of Chl-a concentration. An uncertainty assessment of the Chl-a retrieval model for highly turbid waters incorporating SMCI was conducted using the Monte Carlo method, revealing a mean coefficient of variation of 0.0567 and a coverage rate of 95.65% for the 95% confidence interval, indicating high predictive stability and reliability of the model. This study emphasizes the importance of the integrated framework strategy that combines OWTs classification and multi-index blending models for accurate and robust remote sensing estimation of Chl-a concentration under optically complex environmental conditions. It confirms the application potential of ZY-1 02E hyperspectral data in monitoring Chl-a in inland and near-coastal waters at medium and small scales. Full article

Harmful algal blooms, particularly those caused by cyanobacteria (blue-green algae) and green algae, pose an increasing risk to aquatic ecosystems and public health. This risk is intensified by climate change and nutrient pollution. This study presents a methodology for in situ monitoring and assessment of algal contamination in surface waters, combining UV/Vis and fluorescence spectroscopy with a fuzzy pattern classifier for consideration of uncertainties. The system incorporates detailed data pre-processing to minimise measurement uncertainty and uses full-spectrum feature extraction to enhance classification accuracy. To assess the methodology under both controlled and real-world conditions, a mobile submersible probe was tested alongside a laboratory setup. The results demonstrate a high degree of agreement between the two systems, showing particular sensitivity to biological signals, such as the presence of algae. The assessment method successfully identified cyanobacterial and green algal contamination, and its predictions aligned with external observations, such as official warnings and environmental changes. By explicitly accounting for measurement uncertainty and employing a comprehensive spectral analysis approach, the system offers robust and adaptable monitoring capabilities. These findings highlight the potential for scalable, field-deployable solutions for the early detection of harmful algal blooms. Full article

Alexandrium spp., globally recognized as harmful algal bloom (HAB) species, pose severe threats to marine ecosystems, fisheries, and public health. Based on 469 occurrence records and 24 marine environmental variables, this study employed the Biomod2 ensemble modeling framework to predict the potential distribution of Alexandrium spp. under current and future climate scenarios, and to assess the role of key environmental factors and the spatiotemporal dynamics of habitat centroid shifts. The results revealed that (1) the ensemble model outperformed single models (AUC = 0.998, TSS = 0.977, Kappa = 0.978), providing higher robustness and reliability in prediction; (2) salinity range (bio18, 19.1%) and mean salinity (bio16, 5.8%) were the dominant factors, while minimum temperature (bio23) also showed strong constraints, indicating that salinity determines “whether persistence is possible,” while temperature influences “whether blooms occur”; (3) under present conditions, high-suitability habitats are concentrated in Bohai Bay, the Yangtze River estuary to the Fujian coast, and parts of Guangdong; (4) climate change is predicted to drive a southward shift of suitable habitats, with the most pronounced expansion under the high-emission scenario (RCP8.5), leading to the emergence of new high-risk areas in the South China coast and adjacent South China Sea; (5) centroid analysis further indicated a pronounced southward migration under RCP8.5 by 2100, highlighting a regional reconfiguration of ecological risks. Collectively, salinity and temperature are identified as the core drivers shaping the ecological niche of Alexandrium spp., and future warming is likely to exacerbate HAB risks in southern China. This study delineates key prevention regions and proposes a shift from reactive to proactive management strategies, providing scientific support for HAB monitoring and marine ecological security in China’s coastal waters. Full article

Cyanobacterial blooms have been increasingly detected in estuaries and freshwater tidal rivers. To enhance detailed monitoring, an efficient approach to detecting algal blooms through remote sensing is needed to focus more detailed monitoring focused on cyanobacteria. In this study, we compared different remote sensing processing methods to determine an efficient approach to mapping chlorophyll-a. Using a subset of paired chlorophyll-a observations with Sentinel-2 imagery (2015–2022), with sites located in the Chesapeake Bay and Indian River selected along gradients of salinity, turbidity, and trophic status, we compared the combined performance of two different atmospheric processing methods (Acolite, Polymer) and a suite of empirical (band ratio, spectral shape indices) and machine learning algorithms for chlorophyll-a prediction. Acolite outperformed Polymer, resulting in 176 observation points, compared to 106 observation points from Polymer, and a greater range in chlorophyll-a values (0–74 μg/L from Acolite compared to 0–36 μg/L from Polymer), although Polymer showed more responsiveness at lower chlorophyll-a levels. Two algorithms performed best in predicting chlorophyll-a, as well as trophic state and HABs risk classes: the machine learning mixture density network (MDN) approach and the one band-ratio approach (Mishra). Full article

Poyang Lake, China’s largest freshwater lake, is an ecologically significant but increasingly vulnerable system threatened by eutrophication and harmful algal blooms driven by human activities. Phytoplankton organisms, as primary producers and sensitive bioindicators, provide critical insights into these ecological changes; however, comprehensive seasonal assessments remain scarce. This study examined intra-annual phytoplankton dynamics at 15 representative sites, with the objectives of quantifying seasonal and spatial variations in community composition, density, biomass, and diversity, and identifying key environmental drivers. Surface water samples were collected during four seasons. Phytoplankton were identified microscopically, and diversity was quantified using Shannon–Wiener, Pielou’s evenness, and Margalef’s richness indices. Concurrent measurements included water temperature (WT), dissolved oxygen (DO), nutrients (TN, TP, NO₃⁻-N, NO₂⁻-N, NH₄⁺-N), chemical oxygen demand (COD), pH, and transparency. Pearson correlation and redundancy analysis (RDA) were applied to evaluate phytoplankton–environment relationships. A total of 118 phytoplankton species belonging to 7 phyla were identified. Chlorophyta, Cyanobacteria, and Bacillariophyta exhibited the highest species richness. The highest seasonal abundances were observed for Microcystis wesenbergii (0.998) in winter, Aulacoseira granulata var. angustissima (0.780) in spring, and Snowella lacustris (0.520) in autumn, indicating pronounced seasonal shifts in dominant taxa across Poyang Lake. Phytoplankton density and biomass peaked in summer, while diversity indices significantly declined with increasing WT. RDA revealed that WT, DO, TP, and transparency collectively explained 45.7% of the community variation, with DO emerging as the most influential factor. These findings demonstrate that physical drivers, particularly thermal conditions and oxygen availability, exert stronger influences on phytoplankton diversity than nutrients alone, challenging nutrient-centric paradigms. Management should integrate hydrological and oxygen regulation with nutrient control, while long-term monitoring, depth-stratified sampling, and trait-based approaches are recommended to improve predictive models under climate variability. Full article

Satellite remote sensing has emerged as a reliable and cost-effective approach for monitoring inland water quality, offering spatial and temporal advantages over traditional in situ methods. Lake Victoria, the largest tropical lake and a critical freshwater resource for East Africa, faces increasing eutrophication driven by nutrient inflows from agriculture, urbanization, and industrial activities. This study assessed the spatiotemporal dynamics of water quality along Uganda’s Lake Victoria coast by integrating field measurements (2014–2024) with Landsat 8/9 imagery. Chlorophyll-a, a proxy for algal blooms, and Secchi disk depth, an indicator of water clarity, were selected as key parameters. Cloud-free satellite images were processed using the Dark Object Subtraction method, and spectral reflectance values were correlated with field data. Linear regression models from single bands and band ratios showed strong performance, with adjusted R² values of up to 0.88. When tested on unseen data, the models achieved R² values above 0.70, confirming robust predictive ability. Results revealed high algal concentrations for nearshore and clearer offshore waters. These models provide an efficient framework for monitoring eutrophication, guiding restoration priorities, and supporting sustainable water management in Lake Victoria. Full article

A novel Gram-negative bacterium, designated strain SN16^T, was isolated from a harmful algal bloom (HAB). Strain SN16^T exhibited potent, broad-spectrum algicidal activity against the colony-forming alga Phaeocystis globosa and eight other HAB-causing species, highlighting its potential as a promising candidate for the biological control of HABs. A phylogenetic analysis of 16S rRNA gene sequences placed strain SN16^T within the genus Flagellimonas. The average nucleotide identity (ANI) and digital DNA–DNA hybridization (dDDH) values between strain SN16^T and its relatives were 75.4–91.4% and 19.3–44.0%, respectively. These values fall below the established thresholds for species delineation, confirming that SN16^T represents a novel species. A chemotaxonomic analysis revealed its dominant cellular fatty acids to be iso-C_15:0 and iso-C_15:1 G. The major polar lipid was phosphatidylethanolamine, and the primary respiratory quinone was menaquinone-6. Genome mining identified 11 biosynthetic gene clusters (BGCs), including those encoding for terpenes, ribosomal peptide synthetases, and non-ribosomal peptide synthetases. By integrating BGC analysis with the observed algicidal activities, we predicted that pentalenolactone and xiamycin analogues are the likely causative compounds. Based on this polyphasic evidence, strain SN16^T is proposed as a novel species of the genus Flagellimonas, named Flagellimonas algicida sp. nov. This is the first report of Flagellimonas species exhibiting broad-spectrum algicidal activity, including activity against the colonial form of P. globosa—a key ecological challenge in HAB mitigation. Full article

The coastal waters of Qinhuangdao are a major hotspot for harmful algal blooms (HABs) in the Bohai Sea, with Noctiluca scintillans being one of the primary algal species responsible for these events. A comprehensive understanding of the microbial community structure and functional responses to N. scintillans bloom events is crucial for elucidating their underlying mechanisms and ecological impacts. This study investigated the microbial community dynamics, metabolic shifts, and the environmental drivers associated with a N. scintillans bloom in the coastal waters of Qinhuangdao, China, using high-throughput sequencing of 16S and 18S rRNA genes, co-occurrence network analysis, and metabolic pathway prediction. The results revealed that the proliferation of autotrophic phytoplankton, such as Minutocellus spp., likely provided a nutritional foundation and favorable conditions for the N. scintillans bloom. The bloom significantly altered the community structures of prokaryotes and microeukaryotes, resulting in significantly lower α-diversity indices in the blooming region (BR) compared to the non-blooming region (NR). Co-occurrence network analyses demonstrated reduced network complexity and stability in the BR, with keystone taxa primarily belonging to Flavobacteriaceae and Rhodobacteraceae. Furthermore, the community structures of both prokaryotes and microeukaryotes correlated with multiple environmental factors, particularly elevated levels of NH₄⁺-N and PO₄³⁻-P. Metabolic predictions indicated enhanced anaerobic respiration, fatty acid degradation, and nitrogen assimilation pathways, suggesting microbial adaptation to bloom-induced localized hypoxia and high organic matter. Notably, ammonia assimilation was upregulated, likely as a detoxification strategy. Additionally, carbon flux was redirected through the methylmalonyl-CoA pathway and pyruvate-malate shuttle to compensate for partial TCA cycle downregulation, maintaining energy balance under oxygen-limited conditions. This study elucidates the interplay between N. scintillans blooms, microbial interactions, and functional adaptations, providing insights for HAB prediction and management in coastal ecosystems. Full article

Algal blooms pose a serious threat not only to the lake ecosystem of Lake Bosten but also by negatively impacting its rapidly developing fisheries and tourism industries. This study focuses on Lake Bosten as the research area and utilizes multi-source remote sensing imagery from Landsat TM/ETM+/OLI and Sentinel-2 MSI. The Adjusted Floating Algae Index (AFAI) was employed to extract algal blooms in Lake Bosten from 2004 to 2023, analyze their spatiotemporal evolution characteristics and driving factors, and construct a Long Short Term Memory (LSTM) network model to predict the spatial distribution of algal-bloom frequency. The stability of the model was assessed through temporal segmentation of historical data combined with temporal cross-validation. The results indicate that (1) during the study period, algal blooms in Lake Bosten were predominantly of low-risk level, with low-risk bloom coverage accounting for over 8% in both 2004 and 2005. The intensity of algal blooms in summer and autumn was significantly higher than in spring. The coverage of medium- and high-risk blooms reached 2.74% in the summer of 2004 and 3.03% in the autumn of 2005, while remaining below 1% in spring. (2) High-frequency algal bloom areas were mainly located in the western and northwestern parts of the lake, and the central region experienced significantly more frequent blooms during 2004–2013 compared to 2014–2023, particularly in spring and summer. (3) The LSTM model achieved an R² of 0.86, indicating relatively stable performance. The prediction results suggest a continued low frequency of algal blooms in the future, reflecting certain achievements in sustainable water-resource management. (4) The interactions among meteorological factors exhibited significant influence on bloom formation, with the q values of temperature and precipitation interactions both exceeding 0.5, making them the most prominent meteorological driving factors. Monitoring of sewage discharge and analysis of agricultural and industrial expansion revealed that human activities have a more direct impact on the water quality of Lake Bosten. In addition, changes in lake area and water environment were mainly influenced by anthropogenic factors, ultimately making human activities the primary driving force behind the spatiotemporal variations of algal blooms. This study improved the timeliness of algal-bloom monitoring through the integration of multi-source remote sensing and successfully predicted the future spatial distribution of bloom frequency, providing a scientific basis and decision-making support for the sustainable management of water resources in Lake Bosten. Full article