Chapter 14 The development of algorithms to parameterise the mass settling flux of flocculated estuarine sediments

Abstract

This paper describes a new empirical flocculation model, referred to as the Manning floc settling velocity model, which is based entirely on an existing data set of 157 individually observed floc populations. The in situ data was acquired from a wide range of estuarine water column conditions in the Tamar, Gironde and Dollard estuaries. The model was developed during the recent Estuary Process Research project (EstProc). A parametric multiple regression statistical technique was chosen to analyse the empirical data matrix and produce the model algorithms. This study has identified three key component algorithms which best parameterise a floc population in turbulent water: the changes in the macrofloc (>160 μm) and microfloc (<160 μm) settling velocities, together with how the suspended matter is distributed across each floc sub-population. When all three algorithms are combined as a single equation, they can be used to describe the mass settling flux of flocculated estuarine cohesive sediment at a specified point, both spatially and temporally, in the water column. The derived Manning floc algorithms are valid for suspended particulate matter concentration and turbulent shear stress values ranging between 10 and 8600 mg/L and between 0.04 and 2.13 Pa (with extrapolation extending this up to 5–10 Pa), respectively. This range allows the empirical model to be more readily implemented into numerical simulation models of sediment transport. Initial 1-D Eulerian testing of the Manning floc model, using an independent data set, indicated that it could reproduce 93% of the total mass settling flux observed throughtout a complete tidal cycle. Accuracy was further improved within the turbidity maximum zone. In contrast, a constant 0.5 mm/s settling rate only estimated 22% of the tidal cycle flux, whereas a fixed 5 mm/s settling rate over-predicted the mass flux by 116%.