Search Results (31)

The chlorophyll-a (chl-a) concentration is a major indicator of marine ecosystem status, harmful algal blooms, and marine primary productivity. In coastal waters, however, complex hydrodynamic and ecological conditions lead to highly variable chl-a dynamics, driven by diverse and interacting mechanisms, posing substantial challenges for chl-a forecasts. To assess the applicability of machine learning approaches in predicting chl-a under complex coastal environments, we present a case study in the Taiwan Strait, where harmful algal blooms occur a few times every year. Based on satellite remote sensing data, a spatiotemporal imputation and prediction framework (STIMP), temporal models (Transformer, CrossFormer, Tsmixer), and spatiotemporal models (MTGNN and PredRNN) were applied to simulate chl-a spatiotemporal variability. A hydrodynamic–biogeochemical model was compared with these machine learning approaches to assess the model skills in coastal chl-a simulations. Results indicate that machine learning models trained with satellite data exhibit reasonable predictive skill offshore with pronounced seasonal variability and low data missing ratio, while their performance weakens in regions where seasonal signals are masked by short-term chl-a fluctuations with more missing data. In contrast, the hydrodynamic–biogeochemical model represents short-term variations in chl-a in nearshore regions with higher temporal resolution and accounts for the underlying mechanisms of phytoplankton biomass accumulation and die-off. When trained with model output, the machine learning approach shows improved performance in coastal chl-a forecasts, with much higher computational efficiency compared to the hydrodynamic–biogeochemical model. This study highlights the advantage of mechanistic and machine learning models in deciphering the spatiotemporal scales and governing mechanisms of chl-a variability in coastal regions and extracting spatiotemporal variability with computational efficiency, respectively. With input data of sufficient temporal resolution (e.g., daily to 3 days) and duration (5–10 years), a combination of the machine learning and mechanistic modeling approaches is recommended for operational coastal phytoplankton bloom forecasting. Full article

Harmful algal blooms (HABs) caused by Karenia brevis (K. brevis) present a persistent ecological and public health challenge across coastal Florida. Reliable bloom forecasting is critical for protecting public health, supporting coastal economies, and enabling timely management responses. This study develops a regionally integrated machine learning framework to predict weekly K. brevis bloom occurrence using environmental data from both the Peace and Caloosahatchee Rivers, combined with coastal bloom records from Southwest Florida and Tampa Bay to enhance the spatial and temporal continuity of the response record. A Random Forest classifier was trained on a multi-decadal dataset incorporating river discharge, nutrient concentrations (total nitrogen and total phosphorus), wind forcing, sea surface temperature, salinity, and sea surface height anomalies as a proxy for Loop Current variability. The model achieved strong predictive performance on a chronologically withheld test set, with an overall accuracy of ~90%, balanced accuracy of 87.6%, and ROC–AUC of 0.972, indicating strong discrimination between bloom and non-bloom conditions with high precision and recall for bloom events. Bloom timing and persistence were captured with strong agreement during ongoing bloom periods, while non-bloom conditions were identified with low false-positive rates. Feature-response analyses indicated that bloom probability increased most sharply under moderate discharge and nutrient conditions, with diminished sensitivity at higher extremes. Learning curve analysis demonstrated robust training performance and stable generalization, with validation accuracy plateauing near 84%, suggesting a data-limited ceiling on forecast skill. By aggregating nutrient inputs across multiple watersheds and integrating spatially aligned bloom observations, this study demonstrates the utility of multi-source machine learning frameworks for regional-scale HAB prediction. The results support the development of early warning tools and provide a reproducible foundation for evaluating how combined watershed loading and physical forcing are associated with K. brevis bloom occurrence in complex estuary systems with watershed and coastal coupling. Full article

Environmental monitoring systems require robust uncertainty quantification for effective decision-making in complex ecological processes. Harmful algal blooms represent a critical challenge where prediction uncertainty directly impacts resource allocation and response timing, yet current remote sensing-based prediction systems provide only deterministic classifications without confidence measures. This gap between algorithmic predictions and actionable risk assessment limits operational utility for stakeholders managing water quality under varying risk tolerances. This study developed a transferable probabilistic forecasting framework integrating Sentinel-2 multispectral imagery with quantile regression and ensemble machine learning to generate continuous confidence indicators for cyanobacteria density prediction, demonstrated through its application to Lake Okeechobee, Florida. The methodology combines spectral indices extracted from Sentinel-2 data with XGBoost for quantile regression at 0.05, 0.50, and 0.95 probability levels, and LightGBM for multi-horizon temporal forecasting. Sentinel-2’s 13 spectral bands spanning visible to shortwave infrared wavelengths, combined with its 5-day revisit frequency provide a spectrally rich and temporally dense input space that is well-suited to gradient boosting methods such as XGBoost, which can exploit complex nonlinear interactions among spectral features to distinguish cyanobacterial signatures from background water constituents. LightGBM achieved mean absolute percentage errors of 2.9% for 10-day forecasts and 5.7% for 20-day forecasts, outperforming conventional regression models. The framework generates 90% prediction intervals that enable reliable risk classifications for operational bloom management. This approach bridges the gap between satellite-based algal bloom detection and actionable decision-making by quantifying predictive uncertainty, representing a shift from binary classifications to probability-based environmental monitoring systems that accommodate varying stakeholder risk tolerances in water quality management applications. Full article

Coastal and marine systems are governed by fragile water-quality dynamics, where disturbances can trigger harmful algal blooms with significant ecological and societal consequences. These pressures have intensified interest in forecasting systems that can anticipate bloom development and support environmental management. This study presents a systematic review of simulation-based and predictive environmental modeling approaches used for marine forecasting of water quality and harmful algal bloom phenomena. Following PRISMA guidelines, 11,185 records were identified, 127 articles were screened in full text for eligibility, and 40 peer-reviewed studies published between 2015 and 2025 were included and synthesized using a structured extraction framework capturing modeling paradigms, forecast targets, data inputs, spatial and temporal scope, validation practices, operational context, and reported limitations. The reviewed literature indicates the dominance of predictive and hybrid modeling approaches, with forecasting efforts primarily focused on coastal systems and short-term applications. Harmful algal blooms and chlorophyll-a emerge as dominant forecast targets, commonly supported by satellite observations, in situ measurements, and environmental forcing variables. Despite substantial methodological advances, persistent challenges related to data availability and quality, validation rigor, system integration, and operational deployment remain evident across modeling paradigms. Overall, the findings suggest that while marine forecasting models have become increasingly sophisticated, their translation into reliable and operational systems remains uneven, highlighting the need for closer alignment. Full article

Environmental filtering studies have revealed immense oceanic microbial diversity, yet the Southeast Pacific remains comparatively undersampled. We characterize the molecular diversity of phytoplankton across two biogeographic domains with contrasting oceanography—fjords and channels (41–53° S) versus the open Pacific (36–42° S)—where the frequency and intensity of harmful algal blooms (HABs) have increased. Using SSU rRNA metabarcoding, we retrieved community composition and biogeographic patterns for micro-phytoplankton. Diversity signals indicated broadly overlapping communities between domains with subtle shifts along hydrographic and nutrient gradients rather than sharp breaks. Phylogenetic resolution within bloom-forming genera recovered well-supported clades, and multiple ASVs matched historically relevant HAB taxa, including representatives of the Alexandrium complex, Dinophysis, Pseudo-nitzschia, and Karenia. Together, these results suggest that regional environmental filtering acts modestly at the community level while preserving clear signals of taxa of management concern. By providing a regionally resolved, DNA-based baseline for southern Chile’s fjords and adjacent open coast, this study helps fill the molecular diversity gap for the Southeast Pacific and supports improved HAB surveillance and ecosystem forecasting in a climate-sensitive seascape. Full article

Dinoflagellate resting cysts are critical to dinoflagellate ecology, acting as a key seed source for initiating harmful algal blooms (HABs) through their germination. However, the in situ germination dynamics of these cysts remain poorly understood due to technical challenges. To overcome this, we utilized the Germlings Harvester (GEHA), an in situ germination device we designed, to collect water samples containing dinoflagellate cysts germinated from marine sediments in Jiaozhou Bay, China, after 5 and 20 days of incubation. By combining the GEHA with metabarcoding analysis targeting 28S rDNA-specific primers for dinoflagellates, we identified 44 dinoflagellate species spanning 31 genera, 18 families, and 7 orders. Of these, 12 species were linked to HABs or recognized as toxic, including Azadinium poporum, Alexandrium leei, Alexandrium pacificum, Akashiwo sanguinea, Karlodinium veneficum, Stoeckeria algicida, and Luciella masanensis. Additionally, five species were newly identified as cyst producers, and one symbiotic dinoflagellate, Effrenium voratum, was detected. Our results also found that germinated dinoflagellate species increased from 23 to 34 with extended incubation, and the ratio of mixotrophic to heterotrophic species was approximately 2:1 in the samples of in situ sediments and seawater outside GEHA, as well as across germination durations (Sg-5 d vs. Sg-20 d). These findings provide essential field evidence for the role of resting cysts in driving HAB formation in this region and highlight the efficacy of the GEHA-based approach for studying in situ cyst germination dynamics, offering a robust tool for monitoring, early warning, prevention, and forecasting of HABs. Full article

Reliable forecasts of large-scale chlorophyll-a (Chl-a) levels one week ahead in the Murray–Darling Basin are essential for water resources management, as increasing Chl-a levels in water bodies indicate possible harmful algal blooms, a serious threat for freshwater security. A lack of high-resolution data in space and time is a major constraint for delivering early warnings. To address data scarcity, we developed a forecasting model integrating remote sensing data and time-series modelling. Using in situ Chl-a measurements from Murray–Darling Basin water bodies, we locally recalibrated a two-band ratio algorithm, namely the Normalized Difference Chlorophyll Index (NDCI), from Sentinel-2 data to derive Chl-a levels. The recalibrated model significantly improved the accuracy of high Chl-a estimates in our dataset after mitigating data heteroscedasticity. Building on these improved satellite-derived Chl-a estimates, we developed a time-series model for forecasting weekly Chl-a levels including quantification of forecast uncertainty through prediction intervals. The developed model, validated at eight sites for 2021–2022 data, performed well at shorter lead times, showing R² = 0.41 and RMSE = 8.1 μg/L for overall performance at a one-week lead time. The prediction intervals generally aligned well with nominal levels, demonstrating their reliability. This study provides a valuable tool for the water managers/decision-makers to issue early warnings of algal blooms in the Murray–Darling Basin. Full article

Cyanobacteria harmful blooms (Cyano-HABs) have become a globally critical environmental issue, threatening freshwater ecosystems by degrading water quality and posing risks to human and aquatic life. Chlorophyll-a (Chl-a), a key biomarker of bloom intensity, offers crucial insights into algal bloom dynamics. However, predicting Chl-a concentrations remains challenging due to the complex interactions between various environmental factors. This study utilizes machine learning (ML) models to predict Chl-a concentrations, focusing on Lake Taihu in China, a large eutrophic lake that serves as an example of numerous freshwater lakes suffering from Cyano-HABs. The research leverages nine critical water quality parameters—water temperature, pH, dissolved oxygen, turbidity, electrical conductivity permanganate index, ammonia nitrogen, total phosphorus, and total nitrogen—to develop an ensemble ML model using XGBoost, known for its ability to handle nonlinear relationships and integrate multiple variables. The XGBoost model achieved superior predictive accuracy with an R² value of 0.78 and RMSE of 8.97 mg/m³ on the test set, outperforming traditional models like linear regression, decision trees, multi-layer perceptrons, support vector regression, and random forests. Feature importance analysis identified electrical conductivity, turbidity, and water temperature as the most significant predictors of Chl-a levels. This study further enhances model interpretability through Pearson correlation analysis, which quantifies the relationships between Chl-a concentrations and other water quality factors. Additionally, we employed principal component analysis (PCA), mutual information, Spearman rank correlation coefficients, and SHAP models to analyze feature importance and model interpretability in ML. The model’s robustness was tested across multiple monitoring sites in Lake Taihu, demonstrating its potential for broader application in other eutrophic lakes facing similar environmental challenges. By providing a reliable tool for forecasting Chl-a concentrations, this research contributes to the development of early warning systems that can help mitigate the impacts of Cyano-HABs, aiding in more effective water resource management. Full article

Biotoxins produced by harmful algal blooms (HABs) are a substantial global threat to ocean and human health. Domoic acid (DA) is one such biotoxin whose negative impacts are forecasted to increase with climate change and coastal development. This manuscript serves as a review of DA toxicosis after environmental exposure in humans and wildlife, including an introduction to HAB toxins, the history of DA toxicosis, DA production, toxicokinetic properties of DA, susceptibility, clinical signs, DA detection methods and other diagnostic tests, time course of toxicosis, treatment, prognostics, and recommendations for future research. Additionally, we highlight the utility of California sea lions (CSLs; Zalophus californianus) as a model and sentinel of environmental DA exposure. Full article

Due to their immense economic and recreational value, the monitoring of Great Lakes water quality is of utmost importance to the region. Historically, this has taken place through a combination of ship-based sampling, buoy measurements, and physical models. However, these approaches have spatial and temporal deficiencies which can be improved upon through satellite remote sensing. This study details a new approach for using long time series of satellite remote sensing data to identify historical and near real-time anomalies across a range of data products. Anomalies are traditionally detected as deviations from historical climatologies, typically assuming that there are no long-term trends in the historical data. However, if present, such trends could result in misclassifying ordinary events as anomalous or missing actual anomalies. The new anomaly detection method explicitly accounts for long-term trends and seasonal variability by first decomposing a 10-plus year data record of satellite remote sensing-derived Great Lakes water quality parameters into seasonal, trend, and remainder components. Anomalies were identified as differences between the observed water quality parameter from the model-derived expected value. Normalizing the anomalies to the mean and standard deviation of the full model remainders, the relative anomaly product can be used to compare deviations across parameters and regions. This approach can also be used to forecast the model into the future, allowing for the identification of anomalies in near real time. Multiple case studies are detailed, including examples of a harmful algal bloom in Lake Erie, a sediment plume in Saginaw Bay (Lake Huron), and a phytoplankton bloom in Lake Superior. This new approach would be best suited for use in a water quality dashboard, allowing users (e.g., water quality managers, the research community, and the public) to observe historical and near real-time anomalies. Full article

Harmful algal blooms (HABs) pose significant threats to coastal ecosystems and public health. Accurately predicting the chlorophyll-a (Chl) concentration, a key indicator of algal biomass, is crucial for mitigating the impact of algal blooms. Long short-term memory (LSTM) networks, as deep learning tools, have demonstrated significant potential in time series forecasting. However, missing data, a common occurrence in environmental monitoring systems, can significantly degrade model performance. This study examines the impact of missing input parameters, particularly the absence of Chl data, on the predictive performance of LSTM models. To evaluate the model’s performance and the effectiveness of different imputation techniques under various missing data scenarios, we used data collected from 2008 to 2018 for training and data from 2020 and 2021 for testing. The results indicated that missing Chl data can significantly reduce predictive accuracy compared to other parameters such as temperature or dissolved oxygen. Edge-missing data had a more pronounced negative effect on the model than non-edge missing data, and the model’s performance declined more steeply with longer periods of missing data. The prediction of high Chl concentrations was relatively more negatively affected by missing data than by low Chl concentrations. Although LSTM imputation methods help mitigate the impact of missing data, ensuring data completeness remains critical. This study underscores the importance of reliable data collection and improved imputation strategies for accurate forecasting of algal blooms. Full article

Marine eutrophication, primarily driven by nutrient over input from agricultural runoff, wastewater discharge, and atmospheric deposition, leads to harmful algal blooms (HABs) that pose a severe threat to marine ecosystems. This review explores the causes, monitoring methods, and control strategies for eutrophication in marine environments. Monitoring techniques include remote sensing, automated in situ sensors, modeling, forecasting, and metagenomics. Remote sensing provides large-scale temporal and spatial data, while automated sensors offer real-time, high-resolution monitoring. Modeling and forecasting use historical data and environmental variables to predict blooms, and metagenomics provides insights into microbial community dynamics. Control treatments encompass physical, chemical, and biological treatments, as well as advanced technologies like nanotechnology, electrocoagulation, and ultrasonic treatment. Physical treatments, such as aeration and mixing, are effective but costly and energy-intensive. Chemical treatments, including phosphorus precipitation, quickly reduce nutrient levels but may have ecological side effects. Biological treatments, like biomanipulation and bioaugmentation, are sustainable but require careful management of ecological interactions. Advanced technologies offer innovative solutions with varying costs and sustainability profiles. Comparing these methods highlights the trade-offs between efficacy, cost, and environmental impact, emphasizing the need for integrated approaches tailored to specific conditions. This review underscores the importance of combining monitoring and control strategies to mitigate the adverse effects of eutrophication on marine ecosystems. Full article

Seafood, especially from the ocean, is now seen as a greener and more sustainable source of protein, causing an increase in its demand. This has also led to people making choices towards seafood as a replacement for carbon-intensive protein sources. As a result, the demand for seafood is growing, and as the aquaculture industry looks to increase production, keeping products safe and sustainable is imperative. There are many challenges faced by the aquaculture industry in meeting these increased demands. One such challenge is the presence of harmful algal blooms (HABs) in the ocean, which can have a major impact on aquatic life. In this paper, we look at the impact of this challenge on aquaculture and monitoring strategies whilst illustrating the potential for technological interventions to help mitigate the impact of an HAB. We will focus on Abagold Limited, a land-based marine aquaculture business that specialises in the large-scale production of abalone (Haliotis midae) based in Hermanus, South Africa. HABs are considered a threat to commercial-scale abalone farming along the South African coastline and require continuous monitoring. The most recent HAB was in February–April 2019, when the area experienced a severe red-tide event with blooms of predominantly Lingulodinium polyedrum. We present some of the monitoring strategies employing digital technologies to future-proof the industry. This article presents the development of a novel hybrid water quality forecasting model based on a TriLux multi-parameter sensor to monitor key water quality parameters. The actual experimental real water quality data from Abagold Limited show a good correlation as a basis for a forecasting model which would be a useful tool for the management of HABs in the aquaculture industry. Full article

Harmful algal blooms (HABs) have been deteriorating global water bodies, and the accurate prediction of algal dynamics using the modelling method is a challenging research area. High-frequency monitoring and deep learning technology have opened up new horizons for HAB forecasting. However, the non-stationary and stochastic process behind algal dynamics monitoring largely limits the prediction performance and the early warning of algal booms. Through an analysis of the published literature, we found that decomposition methods are widely used in time-series analysis for hydrological processes. Predictions of ecological indicators have received less attention due to their inherent fluctuations. This study explores and demonstrates the predictive enhancement for chlorophyll fluorescence data based on the coupling of three decomposition algorithms with conventional deep learning models: the convolutional neural network (CNN) and long short-term memory (LSTM). We found that the decomposition algorithms can successfully capture the time-series patterns of chlorophyll fluorescence concentrations. The results indicate that decomposition-based models can enhance the accuracy of single models in predicting chlorophyll concentrations in terms of the improvement percentages in RMSE (with increases ranging from 25.7% to 71.3%), MAE (ranging from 28.3% to 75.7%), and R² values (increasing ranging from 14.8% to 34.8%). In addition, the comparison experiment for different decomposition methods might suggest the superiority of singular spectral analysis in hourly predictive tasks of chlorophyll fluorescence over the wavelet transform and empirical mode decomposition models. Overall, while decomposition methods come with their respective strengths and weaknesses, they are undeniably efficient in combination with deep learning models in dealing with the high-frequency monitoring of chlorophyll fluorescence data. We also suggest that model developers pay more attention to online data preprocessing and conduct comparative analyses to determine the best model combinations for forecasting algal blooms and water management. Full article

Harmful algal blooms (HABs) adversely impact aquatic organisms, human health, and the marine economy. The need to understand the origins and mechanisms of HAB occurrence and development determines the relevance of the study of these phenomena, including using remote sensing methods and assets. Here we present the results of a comprehensive study of conditions and precursors of some intense HABs detected in the water areas near the island of Chiloe (Chile, 2016), near the Kamchatka Peninsula (Russia, 2020), near the island of Hokkaido (Japan, 2021), among others. The study involves statistical analysis of long-term satellite and model data arrays on significant parameters of the marine environment and near-surface atmosphere, as well as empirical modeling of HAB risks. Information products on the following environmental parameters were used: sea surface temperature (SST, NOAA OISST, since 1981), the level of photosynthetically active radiation (PAR) and chlorophyll-a concentration (MODIS Ocean Color SMI, since 2000), sea surface salinity and height (HYCOM, since 1993), and near-surface wind speed and direction (NCEP CFSv2, since 1979). Quantitative assessments of the dynamics of informative criteria were applied. The key criterion is the ratio (${\mathsf{\Delta }}_{\mathsf{\sigma }}$) of the absolute deviation of the studied parameter from the expected norm to the RMS deviation of its values. Intense HABs were often preceded by excessive SST (up to ${\mathsf{\Delta }}_{\mathsf{\sigma }}$ ~1.99) and PAR (up to ${\mathsf{\Delta }}_{\mathsf{\sigma }}$ ~2.25) values, as well as low near-surface wind speed (up to ${\mathsf{\Delta }}_{\mathsf{\sigma }}$ ~−1.83). These environmental parameters considerably contribute to HAB generation and intensification. An approach and empirical function were proposed that allow us to assess the risk of HAB phenomena and reveal their precursors. Using the proposed approach and empirical function, the precursors of ten HABs were identified, nine of which were confirmed by in situ data. The results can be used as a tool for forecasting and studying the conditions for the occurrence of HABs, representing one of the promising directions for monitoring these dangerous phenomena. Full article