To meet energy demand towards a low-carbon future, the global market demand is growing for metals such as cobalt and nickel which are major elements in batteries. Polymetallic nodules, which are formed on abyssal plains at depths ranging from 4 to 6 kilometres and are distributed in high abundance on the top of the seabed, contains several times more cobalt and nickel than the entire global terrestrial reserves. This has raised the interest to exploit these resources from the deep ocean. The seafloor mining tool (SMT) can move along the soft sea bottom and can collect polymetallic nodules. While doing so, it will also entrain sediments and water. The excess of water and sediment entrained is discharged at the back of the SMT, forming a sediment plume. The sediment plume dispersion has strong adverse impacts on deep-sea environment.Thus, it is essential to study the sediment plume behavior in order to limit plume dispersion and thus to reduce its environmental impact. Experimental research is a powerful technique to study the plume behaviour. However, experiments sometimes take a long time due to complex set-up. In comparision, numerical analysis can save time and costs when solving complex problems. Furthermore, numerical modelling can provide deeper understanding and flexibility for boundary conditions and sediment types, which is applicable on both model and prototype scale. Previous numerical studies have noted the significant role of flocculation in limiting plume dispersion, but flocculation process has not been modelled explicitly. This study aims to establish a numerical model to study flocculation process and its effect on sediment transport. Previous flocculation-fluid dynamics modelling has applied a Euler-Euler method with additional population balance equations. The disadvantage is that many equations need to be solved. To avoid excessive computational costs, the sediment transport is described by a multiphase drift-flux model in this study. The flocculation process is modelled by a discretized form of population balance equations. The author has found that, by multiplying the particle volume, the population balance can be efficiently incorporated in the phase continuity equations in the drift-flux model. The population dynamics of particle aggregation and breakup can thus be characterized by the phase transition terms in the phase continuity equations. Hence, no additional equation needs to be introduced and solved. Verification is carried out to check conservation relationships and iterative convergence of numerical results. Then, an initial numerical investigation has shown the results can qualitatively show the three settling stages (i.e., flocculent settling, hindered settling and compression settling) found in the experimental studies. Afterwards, the collision efficiency is calibrated using the settling column tests conducted by Enthoven (2021). The results of calibration show a good fit to the experimental data. Another advantage is that numerical simulations can provide the particle size distribution over time, which is not measured in the experiments. The major novelty of this study is the coupling of the drift-flux model and the population balance equations, which inherits both the characteristics of population balance and the merits of drift-flux model in reducing computational costs. The flocculation modeling technique as proposed in this study can be incorporated as a module into an extended drift-flux model to predict the dispersion of deep-sea mining plumes.

Turbidity currents are common in the waters around the world. They can be caused by earthquakes, collapsing slopes or other geological disturbances. However, turbidity currents can also be caused by deep sea mining activities. Deep sea mining is done with special mining vehicles. There are three main operations within this mining vehicle: collecting nodules, separating the nodules from the mixture and discharging the mixture of sediment and water towards the environment. After impingement of the discharged flow with the seabed, it is expected that the flow will take the form of a turbidity current. Turbidity currents belong to a larger class of flows called gravity currents. Furthermore, turbidity currents are typically defined as dilute flows in which particles are dominantly supported by fluid turbulence. These currents have an interstitial fluid that is a liquid, generally water. The objective of this research is to increase knowledge of the behavior of these currents and to obtain experimental results that can be used for validation of CFD models. In this research the influence of the initial concentration on the dispersion, deposition and entrainment of the current is investigated. In addition, the spread of particles within the current is researched. Furthermore, different types of sediment are used with various particle size ranges and particle properties. The experimental methodology can be divided in two parts. The first part are experiments that involve a full-depth lock release of a fixed volume suspension of a sediment with different particle size ranges and particle properties into water. The initial concentration is varied and the obtained current is recorded with a high-speed camera. The second part consists of multiple calibration procedures. During these procedures, different concentrations of sediment are mixed with water and the resulting solution is recorded. This data is used to create a calibration function that in turn can be used to quantify the concentrations of sediment in the previous recorded currents. The experimental methodology differs from previous work due to this calibration procedure. Afterwards, video processing is used to perform an analysis of the current. The resulting turbidity currents go through three phases as observed in the experiments and previous work. At the start, there is an initial phase where the front is formed and the current accelerates. During this phase a limited amount of particles settle out of the current. Afterwards, the current transitions into a second phase when the velocity of the front starts to decrease due to inertial forces and due to the start of particle settling at the rear of the current. The last phase is when the backflowing bore created by the ambient fluid at the release of the gate reaches the front and decelerates the current until viscous forces start to dominate until the current vanishes. The experiments shows that currents transporting fine particles reach higher velocities and may travel for longer distances in comparison to currents composed of larger particles. Furthermore, the combination of fine and large particles has a substantial effect on the currents dynamics. The currents composed of a mixture of these two sediment types travel longer distances than currents with only large particles. The performed experiments also focus on varying the initial concentration and sediment type. These show that larger particles tent to entrain more with the ambient fluid in comparison to smaller particles and thus they have a larger vertical dispersion. Furthermore, currents composed of sediment with only fine particles will have a more rapid transition into a dense front and a less dense middle and rear pat of the current. Currents composed of fine sediment and a combined sediment with small and large particles have a clear division between a dense basal layer and a less dense top layer within the current. At last, deposition patterns are different per sediment type. Sediment composed of larger particles will have a larger deposition rate than sediment composed of smaller particles. Furthermore combining both sediment types will increase the deposition rate at the beginning of the current.