Particle-based methods in the modelling of turbidity currents and turbidites
|Location||International Geological Congress,oslo 2008|
|Author||Al-Khayat, Omar۱; Loseth, Tore۲; Bruaset, Are Magnus۱; Langtangen, Hans Petter۱|
|Holding Date||11 October 2008|
Turbidity currents constitute the most common type of underwater density flows. The flow moves down-slope due to en excess density relative to the ambient water, owing to sediment particles held in suspension by turbulent vortices. They commonly develop from slides or slumps that are initiated when unstable slopes disintegrate and fail (surge-type turbidity currents). They can also be developed when a turbulent, sediment-laden flow enters a basin, plunges on its marginal slope and continues to spread in the basin (sustained or hyperpycnal flows). Turbidity currents can flow for large distances (several hundreds of kilometres) and over gentle slopes (much less than one degree), but eventually they decelerate and loose turbulence as the slope and the speed of the current decreases. This causes a reduction in flow momentum and deposition of the transported sediments.
The deposits resulting from turbidity currents are called turbidites. Thick successions of stacked turbidites are common in the stratigraphic record and many of them are important petroleum reservoirs. The modelling and prediction of such deposits are therefore of prime interest both in the academic community and in the industry.
In recent years, significant improvement has been made for modelling and simulation of turbidity currents. Numerous models now exist, and they are almost without exception based on traditional computational fluid dynamics (CFD). These are based on the assumption that the fluid and other quantities are in principle a continuum. Although improvement has been made, such models are associated with challenges. It is for instance difficult phenomenologically to describe and model the dynamics, deposition and erosion of sand particles in complex fluid flow from a traditional CFD approach.
Recently, an attempt to model turbidity currents using a particle-based method called Cellular Automata was reported by IFP. Particle-based models are founded on the idea that a macroscopic system is composed of interacting discrete building blocks. These particles are conjoined through a predefined interaction rule which serves to model the molecular dynamics numerically. These are then summed up in various manners to describe the macroscopic system.
The aim of this study is to investigate whether, from a mathematical point of view, some particle-based models can have larger potential for mathematical simulation of turbidity currents and turbidite formations than the CFD methods. In particular, the study focuses on a particle-based method called the Lattice-Boltzmann Method (LBM), which shows good potential in describing complex flow. Several examples of models using LMB for fluid modelling will be shown. Finally, we will present a turbidity current application based on the LBM paradigm. We will report current status and outlook on the development process.