Abstract

Marine fog is defined as a turbulent air layer contiguous the ocean surface, laden with ~ 1-30 microns sized water droplets, characterized by the Meteorological Optical Range (i.e., visibility) less than 1 km. Fog disrupts transportation, poses security threats, disorients human perception and impacts communications and ecosystems. Net deposition of water vapor on hygroscopic aerosols in near-saturated marine environments leads to marine fog through collusion of dynamic, thermodynamic and physicochemical processes. On larger scales, temperature inhomogeneities of synoptic [low-pressure, colder] weather systems break down to the dissipation (Obukhov-Corrsin) scales, providing an entrée for marine-fog genesis. Evolving fog droplets and their aerosol nuclei are embedded in the smallest (Kolmogorov) eddies of atmospheric turbulence, and a host of two-phase microphysical process involving deposition/evaporation on/from the droplets, droplet surface tension, and eddy straining motions affect the growth, maturation and dissipation (i.e., lifecycle) of fog. This presentation will describe some major findings of a five-year (2021-26) multidisciplinary, multi-investigator, integrative project dubbed Fatima (Fog and turbulence interactions in the marine atmosphere) on marine fog. Ship and land/platform-based field observations in Grand Banks, Sable Island (an islet in the region where warm Gulf Stream and cold Labrador waters mix) and Hibernia Oil Platform in 2022 as well as multi-ship and aircraft observations in the Yellow Sea (off-coast of the Republic of Korea) in 2023, all accompanied by high-resolution and numerical weather prediction (NWP) model simulations, elicited new meteorological and [bio]physicochemical processes associated with fog lifecycle. The results elicited new physical processes, and indicated some commonly used concepts on fog dynamics need revisiting. This work was funded by the Grant N00014-21-1-2296 of the US Office of Naval Research, administered by the Marine Meteorology and Space Weather Program.

Abstract

Large-eddy simulation (LES) has been increasingly used to predict turbulent flows in navy hydrodynamics, such as flow-structure interaction and hydro-acoustics. These tasks require that LES should be time-accurate: it can correctly predict wavenumber-frequency spectra of velocity and pressure fields and their equivalent space-time correlations. The conventional turbulence models based on one single flow process suffers from their capability of representing the competitive balances of multiple flow processes, such as energy dissipation and random backscatter, attached and separated flows, and the numerical issues, such as stochastic and realization differentials. The machine learning method is potential to become the workhorse for turbulence modelling and numerical issues. In this talk, we present our recent work. (1) Data-driven turbulence models with random forcing: this class of models can be used to correctly predict wavenumber and frequency energy spectra and thus turbulence-generated noise; (2) Knowledge integrated additive (KIA) wall model: this model overcome the issue of “catastrophic forgetting” in machines learning and can be used to numerically simulate attached and separated flows. (3) LES-based shape optimization: the regularized ensemble Kalman method is introduced to overcome the blow-up of model gradients due to the chaotic nature of turbulence and the LES used for reduction of turbulence-generated noise. The application of LES to the noise radiated from turbulent flows around underwater vehicles is also presented.

Abstract

Resilience demands risk-coupled action. Yet risk remains poorly characterized where it matters most — at the scales, extremes, and compound interactions that drive impact. In coastal environments, this challenge is amplified by cyclone-driven hazards, including wind, rain, surge, and their cascading interactions, which remain difficult to resolve even with state-of-the-art climate models and their emulators. Resolving extremes is fundamentally a representation problem. Neither parameterization nor emulation alone is sufficient to capture the mechanisms that generate tail risk. This talk presents an approach to adaptive coastal digital twins grounded in high-resolution, physically consistent risk quantification. We focus on cyclone-driven hazards and demonstrate three complementary pathways that combine physics and machine learning to resolve extremes beyond conventional models: (1) tropical cyclones, using physics-constrained equation discovery to recover parsimonious dynamical representations; (2) extratropical cyclones, using stochastic transport formulations to capture structure and evolution; and (3) extreme precipitation, using statistical–physical adversarial downscaling to represent tail behavior. We further couple wind and rainfall to inundation, actively sampling the tails to identify and characterize the most damaging events, and extend these capabilities to forecasting—enabling, for example, prediction of peak inundation for an oncoming storm. We conclude with implications for adaptive coastal digital twins in Macao and South China, illustrating how these methods support decision-relevant, high-resolution representations of compound and cascading risks.

Abstract

Fish exhibit exceptional hydrodynamic performance, combining energy-efficient propulsion, agile manoeuvring, and adaptive environmental sensing. This keynote presents a multidisciplinary research effort, with a primary focus on marine hydrodynamics, aimed at exploring and characterizing biological swimming mechanisms to inform the design of next￾generation underwater vehicles and robots. Application areas include marine aquaculture, underwater monitoring, and autonomous exploration. The work integrates experimental, theoretical, and numerical approaches. Controlled experiments on live fish were conducted in swim tunnels using species representative of two different swimming modes. Configurations such as solitary and schooling arrangements were tested to analyse behavioural and hydrodynamic influences on swimming efficiency. These experiments also addressed boundary effects, body size, and critical swimming speed, factors relevant for both biological insight and aquaculture system optimisation. Complementing the experimental work, advanced numerical simulations were carried out using two-dimensional, self-propelled fish-like foils. These investigated the effects of body shape, motion strategies (e.g., prescribed undulation, rigid flapping, morphing bodies), and flow regimes on propulsive performance. Parametric analyses examined thrust generation, input power, recoil effects, and flow confinement, offering insights into how different features can influence locomotion and energy efficiency in engineered systems. Two pillars of ongoing research are also discussed. The first involves theoretical and numerical investigations using simplified hydrodynamic and structural models to advance the understanding of stability and manoeuvrability as functions of body morphology, fin positioning, and material flexibility. These studies support design strategies for robotic platforms capable of tuning stiffness or adapting morphology to improve directional control and passive stability. The second pillar concerns bio-inspired flow sensing, modelled after the lateral-line system. A digital twin and signal-processing framework is introduced to investigate how distributed passive sensors mounted on a vehicle can interpret wake dynamics and detect upstream obstacles. Both simulations and experimental validations support the feasibility of such sensing strategies to enhance environmental awareness and obstacle avoidance in cluttered or low-visibility environments. This keynote will also present relevant findings from the state-of-the-art in the field, such as the role of added-mass in fish-like manoeuvring, and will outline potential future research directions. While significant progress has been made, further work is required to fully uncover the key hydrodynamic principles of biological swimming and effectively translate them into adaptable, efficient, and sustainable technologies for underwater applications.

Abstract

In the late 1970s and early 1980s, several papers dealing with instabilities of water waves were published. In 1978, Longuet-Higgins coined the terminology “superharmonic instability”. The distinguishing feature of that instability is that it is co-propagating with the wave. In other words, it has the same wavelength and the same speed as the wave. This feature is to be contrasted with subharmonic instabilities and the modulational (Benjamin-Feir) instability where the perturbation has a different wavelength and a different speed. The link between wave breaking and instabilities has been made several times. But which instabilities are the most relevant for wave breaking? The superharmonic instability develops into crest instability. For the case of unstable periodic Stokes waves, the wave-breaking scenario was observed in the numerical simulations of Jillians (1989). Jillians took the eigenfunction as a small perturbation to the Stokes wave and integrated in time. A microbreaker emerged from the superposition of a periodic travelling wave and a superharmonic unstable eigenfunction. Mansar et al. (2025) checked the robustness of the appearance of this crest instability leading to breaking. They added a perturbation to a large-amplitude unstable Stokes wave, which was then taken as initial data in a direct numerical solution of the Navier-Stokes equations, using the Basilisk numerical software package. The talk will be devoted to the 3D instability of 2D waves as well as the instability of 3D waves. We will present a theory for breaking of 3D water waves, based on instability of short-crested Stokes waves travelling in deep water. We find that these waves are susceptible to a crest instability at large amplitude, with a dipole structure of the eigenfunctions that varies periodically along the crest. Studying the full 3D problem numerically, with the unstable eigenfunctions as initial data, leads to overturning crests that generate the structure of a microbreaker, similar to what has been observed in the open ocean. This talk will be in the spirit of Prof. T. Wu, who deeply engaged in the science of wave phenomena using mathematics and physics to understand complex behaviors in waves.

Abstract

Waves have distinctive properties that enable intriguing applications, such as cloaking, super-resolution imaging, defect detection, and noise and vibration control. In this lecture, the speaker will focus on two key properties: time reversibility and subwavelength control. He will use experimental and numerical examples to demonstrate and explain these properties across various types of waves. Following this, he will present findings from the application of these techniques in over 15 real-world projects related to water supply and drainage systems. Additionally, the speaker will discuss recent research on subwavelength control of coastal gravity waves using Helmholtz resonators and highlight their significant potential for coastal engineering applications.

Abstract

Nature-based solutions are increasingly integrated into coastal and river engineering, from vegetated shorelines to restored wetlands and eco-engineered channels. While vegetation is widely recognized for reducing wave energy, its influence extends far beyond wave attenuation. It fundamentally alters currents, turbulent mixing, and the spreading of jets and plumes, with important implications for coastal protection, sediment stability, and water quality. Over the past decade, eco-hydraulics has moved from qualitative descriptions to predictive, physics-based understanding. Research has clarified how vegetated canopies dissipate wave energy, modify flow structure, and reshape turbulence. These changes affect both advective transport and diffusive mixing, controlling how sediments, nutrients, and pollutants are redistributed in natural and engineered environments. Recent advances include improved modelling of finite-amplitude wave attenuation, better characterization of vegetation drag and canopy geometry, and new insights into how jets spread and dilute when interacting with vegetated currents. Scaling relationships now link plant properties to measurable hydrodynamic effects, providing guidance for numerical modelling and engineering design. By connecting hydrodynamic mechanisms to practical decision-making, this lecture shows that eco-hydraulics is not only an ecological perspective, but a necessary component of resilient coastal design and effective environmental management.

Abstract

Large-scale floating solar farms (FSFs) are rapidly advancing worldwide, with generation capacity approaching ~1 GW in calm-reservoir settings. Modern FSFs are typically modular, consisting of hundreds of thousands of interconnected floats supporting PV panels and covering extensive water surfaces. Accurate assessment of their impacts on water quality therefore requires environmental-hydraulic modelling that captures post-construction changes. Improved physical understanding is also needed to represent how periodic surface coverage modifies surface hydrodynamics. The next frontier for modular FSFs is sheltered coastal waters, where available surface area can far exceed that of reservoirs but where stronger currents and waves pose greater engineering and environmental challenges. Pilot coastal projects—such as the 6 MW installation in Singapore completed in 2022—and additional developments at the scale of O(10 MW) are now under way. In coastal environments, tidal currents strongly influence processes of interest, so models must account for how FSFs alter surface drag, redistribute turbulence in the surface boundary layer, and affect sediment transport. In this talk, I shall present recent research addressing these issues, with the goal of informing future coastal FSF designs while supporting sustainable integration into urban coastal systems.