SIMULATION OF WIND-BLOWN SAND FOR GEOMORPHOLOGICAL APPLICATIONS: A SMOOTHED PARTICLE HYDRODYNAMICS APPROACH

Sandy coastal landscapes can be relentlessly dynamic owing to waves, tides, extreme weather events, and perpetual winds. In these settings, the ever-changing landscape can necessitate costly risk mitigation measures and landscape maintenance. Currently, the research on coastal landscape change is focused on waves and storm surge, whereas the role of wind in reshaping coastal landscapes is still poorly understood.

Aeolian transport models that are intended for management applications should be physically based, computationally efficient, and capable of representing complex terrain scenarios. However, this is difficult to achieve owing to the complexity of aeolian transport; landscape-scale models often rely heavily on empirics, whereas physics-based grain-scale models can be computationally intractable. Sauermann et al. (2001) used a mean field approach to develop a physics-based continuum model of saltation. In this model, the saltation layer is modeled as a fluid, and thus is described by its density and velocity fields. However, the nonlinearity of the model can make it difficult to solve for complex landscape scenarios.

We present a Smoothed Particle Hydrodynamics (SPH) implementation of the bivariate Sauermann model for geomorphological applications. In the SPH framework, the saltation layer is represented by an ensemble of sampling particles, which flow with the saltation layer and embody its velocity and density. The velocity and density at any location is equal to the weighted sum of the velocities and densities that are carried by neighboring particles. The contribution of each particle is weighted by its mass and density, and a spatial weighting function, which is a function of the distance to the particle. The SPH implementation offers a high level of computational efficiency and robustness to complex landscapes that are typical of anthropogenic influence. Furthermore, the particle nature of SPH naturally lends itself well to automatic adaptive resolution, which can be a considerable advantage for geomorphological applications.

The model application to a small section of the Outer Banks, NC, will be presented. The wind field, which drives saltation, will be simulated using computational fluid dynamics software over Light Detection and Ranging (LiDAR)-derived terrain.