Using a random displacement model to simulate turbulent particle motion in a baroclinic frontal zone: A new implementation scheme and model performance tests

Numerical artifacts can limit accurate simulation of turbulent particle motion when Lagrangian particle-tracking models are implemented in hydrodynamic models with stratified conditions like fronts. Yet, modeling of individual particle motion in frontal regions is critical for understanding sediment dynamics as well as the transport and retention of planktonic organisms. The objective of this research was to develop a numerical technique to accurately simulate turbulent particle motions in a particle-tracking model embedded within a hydrodynamic model of a frontal zone. A new interpolation scheme, the 'water column profile' scheme, was developed and used to implement a random displacement model for turbulent particle motions. A new interpolation scheme was necessary because linear interpolation schemes caused artificial aggregation of particles where abrupt changes in vertical diffusivity occurred. The new 'water column profile' scheme was used to fit a continuous function (a tension spline) to a smoothed profile of vertical diffusivities at the x-y particle location. The new implementation scheme was checked for artifacts and compared with a standard random walk model using (1) Well Mixed Condition tests, and (2) dye-release experiments. The Well Mixed Condition tests confirmed that the use of the 'water column profile' interpolation scheme for implementing the random displacement model significantly reduced numerical artifacts. In dye-release experiments, high concentrations of Eulerian tracer and Lagrangian particles were released at the same location up-estuary of the salt front and tracked for 4days. After small differences in initial dispersal rates, tracer and particle distributions remained highly correlated (r=0.84 to 0.99) when a random displacement model was implemented in the particle-tracking model. In contrast, correlation coefficients were substantially lower (r=0.07 to 0.58) when a random walk model was implemented. In general, model performance tests indicated that the 'water column interpolation' scheme was an effective technique for implementing a random displacement model within a hydrodynamic model, and both could be used to accurately simulate diffusion in a highly baroclinic frontal region. The new implementation scheme has the potential to be a useful tool for investigating the influence of hydrodynamic variability on the transport of sediment particles and planktonic organisms in frontal zones. (c) 2006 Elsevier B.V All rights reserved.