Pierre Le Hir

Numerical modelling of mixed-sediment consolidation

Sediment transport modelling in estuarine environments, characterised by cohesive and non-cohesive sediment mixtures, has to consider a time variation of erodibility due to consolidation. Generally, validated by settling column experiments, mud consolidation is now fairly well simulated; however, numerical models still have difficulty to simulate accurately the sedimentation and consolidation of mixed sediments for a wide range of initial conditions. This is partly due to the difficulty to formulate the contribution of sand in the hindered settling regime when segregation does not clearly occur. Based on extensive settling experiments with mud-sand mixtures, the objective of this study was to improve the numerical modelling of mixed-sediment consolidation by focusing on segregation processes. We used constitutive relationships following the fractal theory associated with a new segregation formulation based on the relative mud concentration. Using specific sets of parameters calibrated for each test—with different initial sediment concentration and sand content—the model achieved excellent prediction skills for simulating sediment height evolutions and concentration vertical profiles. It highlighted the model capacity to simulate properly the segregation occurrence for mud-sand mixtures characterised by a wide range of initial conditions. Nevertheless, calibration parameters varied significantly, as the fractal number ranged from 2.64 to 2.77. This study investigated the relevance of using a common set of parameters, which is generally required for 3D sediment transport modelling. Simulations were less accurate but remained satisfactory in an operational approach. Finally, a specific formulation for natural estuarine environments was proposed, simulating correctly the sedimentation-consolidation processes of mud-sand mixtures through 3D sediment transport modelling.

Investigating suspended particulate matter in coastal waters using the fractal theory

Suspended particulate matters (SPM) in coastal waters were investigated with an approach combining suspended particulate matter concentrations (SPMCs) measured by an optical backscatter sensor (OBS), particle size distributions measured by a laser in situ scattering transmissometer (LISST), and the fractal theory. The aim was to investigate whether changes in the fractal dimension indicate variations in the SPM composition. The method relies on the fractal theory that relates the floc excess density ∆ρ to the ratio between floc size Df and primary particle size Dp as follows: Δρu2009=u2009(ρpu2009–u2009ρw)[Df/Dp]nfu2009−u20093. The best-fit fractal dimension nfxa0was determined by matching the OBS-derived SPMC and the SPMC calculated from the LISST assuming fixed primary particle size and density. This method was applied to measurements from four tidal cycles characterized by different organic matter (OM) contents. Spurious data caused by schlieren were identified and discarded. The fractal dimension was relatively homogeneous over each tidal cycle except one where a strong apparent decrease in the fractal dimension was related to an important episode of sediment resuspension. This apparent decrease could result from the limit of the LISST to measure valid data when SPMC is high enough to induce multiple scattering and/or a change in the SPM population in suspension. Results showed also strong differences between the tidal cycles such as an increase of the fractal dimension with increasing OM content. Sensitivity analyses of fractal dimension and settling velocity were performed on major parameters: primary particle size and density, slope of the OBS calibration relationship, and optical model of inversion of the LISST. Results showed that the assumed primary particle size and the OBS calibration relationship significantly affect fractal dimension and settling velocity calculations.