Modelling the Giant Sandwaves of San Francisco Bay

A study on process based sandwave modelling, with an exploration on the mechanisms that results in the formation of these bedforms

More Info
expand_more

Abstract

Connected to the Pacific Ocean by the Golden Gate channel, the San Francisco Bay accommodates a widely spread field of sandwaves that show large spatial variations which are not yet fully understood. Insight into the formation of sandwaves is necessary to understand and predict the processes that control sand transport and sedimentation patterns in San Francisco Bay. When aiming at predicting large scale morphodynamic processes within a time span of years, a numerical model is imperative. The aim of this research is to study the capability of DELFT3D to model the sandwaves using a 2Dv-model, with emphasis on assessing the formation mechanisms. The model is built using a sinusoidal shaped bed level and flow velocities with large temporal variations (max. of 2.00 ms-1). Analysis of results focuses on two main aspects: sandwave equilibrium and residual flow patterns for one tidal cycle. Model results show that equilibrium in sandwave height is found for relatively large scale lengths (order of 100 m) and that modelled sandwaves show different growth and decay rates for steeper slopes and higher velocities. Stability in sandwave development is contributed mainly to a term which hinders the formation of steep slopes resulting in a so called equilibrium slope angle, and a growing term based on the residual near-bed current velocity. For the studied range of sandwave lengths, heights were found to be independent of the local water depth. And increasing flow velocities only allow the largest dimensions to reach equilibrium. Growth in height is found to be mainly caused by the hindering crest velocity term, being not large enough to compensate for the high near-bed upslope velocities. Decay in height is mainly attributed to asymmetric residual current flow, due to an initial bed level steepness that is too gentle.

Files