Physical, Analytical and Numerical Study of Ship-Generated Waves-Induced Hydrodynamics and Associated Sediment Transport

Abstract

The Upper St. Lawrence River, a heavily regulated inland waterway connecting the Great Lakes to the Atlantic Ocean, has experienced intensified commercial navigation over the past several decades. This increase in vessel traffic has raised growing concern about the cumulative impact of ship-generated hydrodynamic disturbances—particularly vessel ship-generated waves—on nearshore environments. These disturbances are capable of resuspending bottom sediments, increasing turbidity, and accelerating the degradation of unprotected shorelines. Primary wave components, such as drawdown, and secondary components such as divergent or Kelvin ship-generated waves, can generate high-energy, short-duration events in shallow, sediment-rich areas. Although the ecological and morphological implications of ship-generated waves are widely acknowledged, their specific role in sediment resuspension—especially when isolated from wind-induced forces—remains insufficiently explored in confined, regulated waterways such as the Upper St. Lawrence River. This research addresses these knowledge gaps through an integrated framework that combines long-term field observations, data-driven modeling, and high-resolution numerical simulations. Two representative nearshore sites—Mariatown and Jacobs Island—were selected based on their contrasting geomorphic features, sediment characteristics, and proximity to the main shipping channel. The sites were monitored for over 300 days using six RBR Duo3 D.TU loggers deployed along shore-normal transects, collecting wave and turbidity data at a 2 Hz sampling frequency. A novel wave separation technique was developed to isolate ship-generated wave signals from overlapping natural waveforms. This approach combined spectral filtering, synthetic Gaussian noise reconstruction of wind and seiche activity, and wavelet-based verification to reliably differentiate anthropogenic from natural forcing events. Field results showed that secondary ship-generated waves, particularly secondary waves, were responsible for short-lived turbidity spikes up to five times greater than those caused by local wind events. These effects were most pronounced in shallow nearshore zones with fine-grained cohesive sediments and minimal shoreline protection. Spatial variability in turbidity response highlighted the importance of local depth, substrate composition, and logger distance from the navigation channel. By contrast, wind-induced waves produced sustained but lower-intensity turbidity increases, suggesting fundamentally different sediment mobilization mechanisms. To assess the influence of vessel parameters on wave dynamics and sediment transport, time-delayed multivariate regression and Long Short-Term Memory (LSTM) neural networks were applied. These models incorporated ship-specific inputs—such as speed, draft, length, and hull type—alongside wave and turbidity measurements. Both modeling approaches identified vessel speed and draft as the dominant predictors of wave energy and near-bed shear stress. The results confirmed that fast-moving tankers and cargo ships with deep drafts consistently exceeded shear stress thresholds for erosion, particularly during sequences of closely spaced transits. In the final phase, a numerical modeling framework was developed by coupling the U.S. Army Corps of Engineers’ Vessel Wake Prediction Tool (VWPT) with the DHI MIKE 21/3 Flow Model (FM) modeling suite. A nested large-domain hydrodynamic model was first created to generate boundary conditions based on river-scale flow, wind, and water level data. This was followed by high-resolution spectral wave and coupled wave-sediment transport simulations at both study sites. The model simulated the propagation of ship-generated waves into the nearshore zone, computed wave orbital velocities and shear stress, and predicted suspended sediment concentration (SSC) and bed-level changes over a synthetic annual time series representing approximately 2,500 ship passages. Model validation against field data demonstrated strong agreement in spatial wave energy distribution and SSC patterns. Results revealed that energy focusing occurred along concave shoreline segments, resulting in localized erosion and sediment mobilization. However, under current vessel traffic conditions, long-term morphological changes were predicted to be minimal, indicating that ship-generated wave-induced erosion is primarily episodic and site-specific rather than widespread or cumulative at the reach scale. This novel study contributes a validated, multi-scale framework for quantifying ship wave-induced sediment resuspension in confined inland waterways. The novel wave separation method enhances the classification of hydrodynamic drivers in mixed environments, while the integration of data-driven and numerical modeling provides mechanistic insights into the relationship between vessel traffic and shoreline processes. The findings support the development of evidence-based shoreline protection policies, including speed restrictions, hull design considerations, and localized structural interventions to ensure the sustainable management of navigation corridors subject to increasing anthropogenic pressure.