This study is the first to experimentally show two wave mechanisms regarding density wave amplification with long horizontal slurry transport. There is bed-driven and suspended-driven density wave amplification, in which the grain size determines which mechanism is dominant.

Density wave amplification in hydraulic pipeline transport causes significant risk during operation with the consequences of blockage. Current design methodology for pipeline transport considers mixture velocity and density constant over space and time. However, these conditions are only possible in laboratory circuits where conditions can be controlled carefully. In real-world conditions, concentration varies significantly over time due to the natural dredging process in which a dredging vessel takes slurry from the seabed. Density wave amplification can be differentiated into two different flow categories. Both long horizontal transport and a combination of vertical and horizontal transport. With the first category, there are two main theories that explain the amplification of density waves: 'erosion and sedimentation imbalance' and 'the unstable slip point of the bed'. Here, density wave amplification only occurs in the presence of a bed. 
In the second category, there is one theory called the: 'transient accumulation theory' which is applicable to a combination of horizontal and vertical transport. With this density wave, amplification can occur far above the deposit limit velocity. Mixture velocities change when density waves travel from horizontal to vertical orientation and vice versa. When mixture velocity changes density will change. The influence of grain size, concentration and the centrifugal pump on density wave amplification has not been researched yet.

A test loop has been built with an inner diameter of 46mm and a length of 46 meters. The goal of this laboratory circuit is to investigate the mechanisms that result in the amplification of density waves. Two types of density waves were measured: bed-driven density waves occurring with coarse sediments (Dorsilit 7: d₅₀=1040 μm and Dorsilit 8: d₅₀=619 μm) and suspended-driven density waves occurring with fine sediments (Dorsilit 9: d50=316 μm and Zilverzand: d₅₀=240 μm). With bed-driven density waves, there is fast amplification and multiple sharp waves which can result in areas where no concentration is left. With suspended-driven density waves, there is one smooth wave, and amplification takes multiple loop lengths.

Density waves inhibit the stability of (long) pipelines, which cannot be predicted using current day design tools. Density waves are formed due to an inverse relationship between sand eroding and settling in a pipeline. By designing and building a dedicated flow loop, an experimental study on how the erosion and sedimentation of sand in a pipe is influenced by particle size and concentration is conducted. Moreover, it is determined whether the erosion and sedimentation process can be modeled numerically using existing analytical pick-up functions. This could be applied for predicting density waves in horizontal pipelines.
From the experimental study it is found that at higher concentrations, the caterpillar-like movements of the bed, associated with the sedimentation and erosion unbalance at high concentrations, were observed more and more frequently. The influence of the mean grain size diameter can be characterised as the ability of the mixture to trigger amplifying density waves in the horizontal measurement section. Only the coarser sands used in the experiments were found to experience the sliding-stopping behaviour. Implementing different analytical pick-up functions to simulate the experimental study resulted in sufficiently accurate results. Depending on the pick-up function on how much calibration was required the measurements could be simulated adequately.