Wave Effects on Negatively Buoyant Jets: CFD Analysis

Computational Fluid Dynamics (CFD) uses computers to simulate fluid flow, addressing limitations of traditional analytical and experimental methods. Fluid mechanics is governed by three conservation principles: mass (neither created nor destroyed), momentum (Newton's second law expressing force equals rate of change of momentum), and energy (including temperature variations). Together, mass and momentum equations form the Navier-Stokes equations describing all fluid flows. These equations are partial differential equations with nonlinear terms and are highly coupled—requiring simultaneous solution, making analytical solutions for three-dimensional problems impossible. CFD involves three main steps: discretization (breaking domains into elements—the most critical step as poor meshing produces unreliable results), solving (converting continuous equations to linear algebraic systems using finite difference, finite volume, or finite element methods), and postprocessing (validating results through visual inspection, mass conservation checks, and force verification).

Navier-Stokes equations govern fluid motion through conservation laws: mass conservation (∂ρ/∂t + ∇·(ρ**u) = 0) and momentum conservation (∂(ρ**u)/∂t + ∇·(ρ**u**u) = -∇p + ∇·[μ(∇**u** + (∇**u**)T)] + **f**). Three spatial directions require three momentum equations. Solution methods include direct (Gauss elimination, LU, TDMA) and iterative (SOR, ILU, ADI) approaches. Turbulent flow is chaotic, three-dimensional, and dissipative with enhanced mixing through energy cascade. Reynolds number (Re = UL/ν) distinguishes flow regimes, transitioning at Re = 2304 for pipe flow. Reynolds averaging creates closure problem requiring turbulence models: RANS (time-averaged), DES (RANS near walls, LES elsewhere), LES (resolves large scales), and DNS (resolves all scales without modeling).

Computational Fluid Dynamics (CFD) is a powerful simulation tool that solves the Navier-Stokes equations to analyze fluid systems, integrating into the mechanical design process for aerodynamic and fluid dynamics applications; the CFD workflow involves creating geometry, generating a mesh to discretize the fluid domain, setting up the solver with appropriate turbulence models (such as RANS equations for turbulent flows), running the simulation, and post-processing results, with mesh quality, solver settings, and convergence monitoring being critical factors that determine simulation accuracy and computational efficiency.

CFD solves Navier-Stokes equations using numerical methods. The process involves dividing an object into millions of tiny boxes called cells, creating a mesh. Each cell interacts with its neighbors. Engineers set known boundary values along the edges of the mesh, and computers perform brute-force iteration to balance all cells and their interactions. This is a generalized algorithm that can be applied to any geometry.

Mesh generation is crucial for CFD accuracy. Structured meshes (quadrilateral/hexahedral) are orthogonal and better for viscous flow near surfaces, requiring fewer cells. Unstructured meshes (triangular/tetrahedral) handle complex geometries better but may require more cells. Mesh quality is measured by orthogonality (ideally 90° for structured meshes) and skewness. Discretization methods include Finite Difference (solves differential equations directly), Finite Volume (solves integral forms), and Finite Element (uses basis functions). Time integration uses explicit and implicit schemes, with explicit schemes limited by CFL number (CFL = αΔt/Δx² < 0.5 for stability). Turbulence modeling approaches include DNS (solving all terms directly), LES (modeling small eddies), and RANS (modeling all turbulent scales). Each approach has trade-offs between accuracy and computational cost.

Linear wave theory is a simplified approach to analyzing surface waves that assumes small-amplitude sinusoidal waves with constant water depth, periodic motion, and negligible viscous effects; it classifies waves into shallow water (depth/wavelength < 1/20), intermediate water (1/20 ≤ depth/wavelength ≤ 1/2), and deep water (depth/wavelength > 1/2) regimes, where wave celerity follows different equations (celerity = √(g×depth) for shallow water, celerity = √(g×λ/(2π)) for deep water), and predicts that water particles follow orbital motions that decrease in size with depth.

The dispersive wave equation relates wavelength (L), period (T), and water depth (h). Deep water waves (h > L/2) have celerity depending only on period. Shallow water waves (h < L/20) have celerity depending only on depth. Orbital velocities are derived by differentiating the velocity potential. In deep water, both horizontal and vertical velocities reach maximum values and vary exponentially with depth. In shallow water, horizontal velocity remains constant from surface to bottom, while vertical velocity varies linearly and becomes zero at the bottom. The bottom boundary condition requires vertical velocity to be zero at z = -h.

This extensive section covers linear wave theory fundamentals and wave classification. It includes: (1) Linear wave theory (Airy wave theory) providing 95% accuracy for 95% of coastal engineering calculations under small amplitude assumptions; (2) Wave classification based on relative depth (D/L): deep water (D/L > 1/2, no seabed contact), transitional water (1/25 < D/L < 1/2, partial interaction), and shallow water (D/L < 1/25, seabed contact dominates); (3) Wave celerity (phase velocity) definition as C = L/T representing energy propagation, not material transport; (4) Sinusoidal wave profile representation η = H/2 cos(θ); (5) Water particle velocities: horizontal u = ωa cos(θ) cosh(k(z+D))/cosh(kD); vertical w = ωa sin(θ) sinh(k(z+D))/cosh(kD); (6) Acceleration relationships showing 90-degree phase shifts with velocity; (7) Wave number k = 2π/L and its role in dispersion relations.

In coastal engineering, linear wave theory (Airy theory) classifies water depths into shallow (h/L < 1/20), transitional (1/20 < h/L < 0.5), and deep water (h/L > 0.5) based on the ratio of depth to wavelength, where the wavelength depends on both depth and wave period. The maximum horizontal particle velocity at any depth is calculated using the formula u = (k * A * g / σ) * cosh(k(z+h)) / cosh(kh) * cos(θ), where k is the wave number, A is the wave amplitude, g is gravity, σ is the angular frequency, z is the depth coordinate, and θ is the phase angle. This velocity is critical for assessing erosion potential on coastal structures, as particle velocities exceeding material resistance thresholds indicate erosion risk.

In coastal waters, currents and waves interact through turbulence, where waves increase turbulence which reduces current velocity and shear stress, but the total shear stress can either increase or decrease depending on the relative magnitudes of the current and wave components; this coupling can be modeled by decomposing the governing equations into separate current and wave components while maintaining consistent turbulent viscosity, resulting in logarithmic velocity profiles with modified shear velocities.

Desalination produces concentrated brine as a waste product, which cannot be discharged directly into the ocean without environmental consequences. Modern desalination plants typically recover about half of their intake flow, meaning the brine stream becomes approximately twice the concentration of normal seawater. Direct discharge would harm marine life, particularly organisms near the sea floor, since the dense, super-saline solution tends to sink. To mitigate environmental impacts, most plants use diffusers to spread out the brine so it dilutes faster, or blend it with other water streams like power plant cooling lines or wastewater effluent before release. When neither option is feasible, some plants inject brine into the ground, though this expensive approach adds significantly to operational costs.

Desalination plants use diffusers at the end of discharge pipes to spread brine and dilute it, introducing air to raise saline levels so the brine spreads and drops rather than sinking. However, this technology has limitations. The diffusers only dilute the brine but cannot prevent the overall increase in salinity levels. Additionally, the diffusers aim to prevent concentrated dense masses of higher salinity water from sinking and spreading insidiously in local areas.

This webinar presents research on managing environmental impacts of seawater discharge (brine) from desalination plants, emphasizing that proper engineering solutions like diffusers and pre-dilution can limit salinity increases to 1-5% within 100-600 meters, preventing significant ecological harm when properly implemented; successful desalination projects require collaboration between academia, public authorities, and private industry to establish scientifically-based environmental standards and monitoring programs that ensure sustainable long-term operation.

Desalination produces brine at approximately twice the salinity of input seawater (around 60,000-70,000 mg/L compared to 35,000 mg/L for seawater). Proper disposal requires achieving initial dilution to below 2,000 mg/L before reaching sensitive benthic habitats. Research demonstrated that optimal diffuser angles maximize mixing efficiency through buoyancy effects, with upward discharge creating larger dilution zones. The Monterey project designed its diffuser to achieve 1,600 mg/L at the ocean floor, below regulatory thresholds. Continuous monitoring using salinity sensors ensures ongoing compliance, with provisions for adaptive modifications if conditions change over time.

Seawater contains 35,000-40,000 ppm TDS, requiring complex engineering for safe drinking water. The process begins with intake systems using coarse and fine screens to prevent marine life entry, followed by coagulation, flocculation, and air flotation to remove particles. Multi-media filtration with anthracite, sand, and gravel layers removes remaining impurities. Chlorine must be removed before membrane treatment using sodium metabisulfite. High-pressure pumps (55-80 bar) drive water through spiral-wound RO membranes with polyamide active layers. Membrane elements are arranged in series, with later elements facing higher salinity and reduced efficiency. RO produces clean permeate and concentrated brine (up to 50,000 ppm TDS). Energy recovery devices transfer pressure from brine to incoming seawater, reducing energy consumption. Brine is discharged through long pipes with diffuser nozzles to prevent environmental damage.