Projects

Salt dispersion in channels of complex geometry

The main objectives of this subproject are:

1. to establish the degree to which salinity variations in time-series are a lagged response to the forces imposed by the river discharge, tidal dynamics, storm surges and seasonal salinity variations near the estuary mouth.

For this purpose, we aim to develop a harmonic regression model applicable for (non-stationary) estuarine salinity time series.  The external forces will be embedded in a formulation describing the river to sea salinity curve (an s-curve). Subsequently, through a site-specific optimalization procedure, the response and temporal salinity response to forcing are determined.  We prove the predictive ability of this new method in the Ems-river, which is a single channel estuary with several well-positioned salinity sensors over the salt intrusion length.  We will extent this methodology to more complex channel-networks of the Rhine-Meuse estuary (RME) with salinity stations located predominantly upstream of the average salt intrusion limit.

2. to develop a new generic methodology to infer and parameterize 1D dispersion coefficients from 3D hydrodynamic models.

Generally, 1D hydrodynamics models fail to accurately predict salinity variations. This can largely be attributed to an oversimplification of the channel geometry and formulations used to parameterize the dispersive mechanism. In this new methodology, we will tackle this problem by directly inferring dispersion from 3D models, since these models implicitly solve the large-scale local dispersive mechanisms. Inferred dispersion and gained system understanding will be used to investigate the dependency between dispersion and channel irregularities, such as harbors basins. We will adjust existing formulations describing local dispersive mechanisms. These formulations will be made partially empirical such that they can be fed with geometric properties or bathymetric details from the model, to better represent the resulting local dispersion. The aim is to embed this in a framework that can be applied to obtain local dispersions coefficients from 3D models and improve 1D hydrodynamic simulations of salt.

3. To improve understanding of lateral salinity variation and salt mixing occurring in bank boundary regions.

1D models represent salinity as a constant for an entire cross-section, whereas salinity varies in both the vertical and/or lateral direction. Remarkably, lateral salinity variations occurring in the near bank region have received little attention. This while it is known that bank boundary layer processes contribute to mixing. We will investigate mixing in near-bank regions under two opposed situations (also see figure); mixing along prismatic channel banks where the flow structure is highly influenced by a rigid boundary, and mixing along natural embanked channels where the boundary region provides significant storage space.  To investigate this, two unique datasets will be collected of the near bank salinity- and flow field. These datasets will used to unravel the mixing mechanisms and estimate the contributions to 1D salt dispersion. The gained new insights will contribute to improving dispersion formulations, but also provide a knowledge basis for good water management.