Related Experiment Videos
Reimagining coastal water quality forecasting with hydrodynamic simulation and advanced machine learning
Seyed Arman Hashemi Monfared1, Jafar Jafari-Asl2, Kourosh Behzadian3
1School of Engineering, University of Warwick, Coventry, CV4 7AL, UK; School of Engineering, Department of Civil Engineering, University of Memphis, Memphis, TN, 38125, USA; AtkinsRealis UK, Newcastle Upon Tyne, NE4 7YB, UK.
Journal of Environmental Management
|August 11, 2026
Summary
This study introduces a novel hybrid machine learning (ML) framework for accurate coastal water quality forecasting, enhancing nutrient management and ecosystem health. The model effectively predicts nitrate and phosphate levels using hydrodynamic data, supporting sustainable coastal management.
Area of Science:
- Environmental Science
- Coastal Ecology
- Water Quality Management
Background:
- Effective coastal ecosystem management requires accurate water quality monitoring and forecasting, especially in polluted areas.
- Predicting nutrient concentrations (nitrate and phosphate) is challenging due to complex environmental interactions.
- Coastal water quality is vital for ecosystem health and managing pollutant loads.
Purpose of the Study:
- To introduce a novel hybrid machine learning (ML) framework for simulating and forecasting coastal water quality.
- To enhance the prediction accuracy of nitrate and phosphate concentrations in coastal waters.
- To provide a reliable tool for supporting sustainable coastal ecosystem management and nutrient control.
Main Methods:
- Developed a hybrid ML framework using least-squares support vector regression (LSSVR) optimized by the Improved Dragonfly Algorithm (IDA).
- Utilized physically grounded simulation data from the TELEMAC-WAQTEL numerical framework, incorporating hydrodynamic variables (free surface elevation, velocity, flow direction).
- Applied and validated the model on a dataset from Ha Long Bay, Vietnam, comparing its performance against other ML models.
Main Results:
- The IDA-LSSVR model demonstrated superior predictive capability compared to GPR, ε-LSSVR, ANN, Random Forest, and Model Tree.
- Achieved the lowest RMSE, MAE, and MAPE, with the highest R² for both nitrate and phosphate predictions on an independent test dataset.
- The model achieved high accuracy with RMSE values of 1.55E-02 mg/L for nitrate and 5.88E-04 mg/L for phosphate.
Conclusions:
- The proposed framework accurately and efficiently forecasts coastal water quality by integrating physics-based simulations with advanced ML and optimization.
- Offers valuable decision-support for nutrient management, eutrophication risk assessment, and environmental monitoring in dynamic coastal zones.
- Presents a promising approach for sustainable coastal ecosystem management and mitigating nutrient-driven degradation.
Related Concept Videos
01:12Modeling and Similitude
Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
01:30Typical Model Studies
Fluid mechanics model studies often utilize scaled-down systems to predict fluid behavior in full-scale environments, such as river flows, dam spillways, and structures interacting with open surfaces. Maintaining Froude number similarity in river models is crucial, as it replicates surface flow features like wave patterns and velocities.
01:24Rapidly Varying Flow
Rapidly varying flow (RVF) in open channels is characterized by abrupt changes in flow depth over a short distance, with the rate of depth change relative to distance often approaching unity. These flows are inherently complex due to their transient and multi-dimensional nature, making exact analysis difficult. However, approximate solutions using simplified models provide valuable insights into their behavior.Key Features of Rapidly Varying FlowRVF is commonly observed in scenarios involving...