Search for: All records

Abstract The salinity distribution of an estuary depends on the balance between the river outflow, which is seaward, and a dispersive salt flux, which is landward. The dispersive salt flux at a fixed cross‐section can be divided into shear dispersion, which is caused by spatial correlations of the cross‐sectionally varying velocity and salinity, and the tidal oscillatory salt flux, which results from the tidal correlation between the cross‐section averaged, tidally varying components of velocity and salinity. The theoretical moving plane analysis of Dronkers and van de Kreeke (1986) indicates that the oscillatory salt flux is exactly equal to the difference between the “local” shear dispersion at a fixed location and the shear dispersion which occurred elsewhere within a tidal excursion; therefore, they refer to the oscillatory salt flux as “nonlocal” dispersion. We apply their moving plane analysis to a numerical model of a short, tidally dominated estuary and provide the first quantitative confirmation of the theoretical result that the spatiotemporal variability of shear dispersion accounts for the oscillatory salt flux. Shear dispersion is localized in space and time due to the tidal variation of currents and the position of the along‐channel salinity distribution with respect to topographic features. We find that dispersion near the mouth contributes strongly to the salt balance, especially under strong river and tidal forcing. Additionally, while vertical shear dispersion produces the majority of dispersive salt flux during neap tide and high flow, lateral mechanisms provide the dominant mode of dispersion during spring tide and low flow. 

Abstract Flow separation has been observed and studied in sinuous laboratory channels and natural meanders, but the effects of flow separation on along‐channel drag are not well understood. Motivated by observations of large drag coefficients from a shallow, sinuous estuary, we built idealized numerical models representative of that system. We found that flow separation in tidal channels with curvature can create form drag that increases the total drag to more than twice that from bottom friction alone. In the momentum budget, the pressure gradient is balanced by the combined effects of bottom friction and form drag, which is calculated directly. The effective increase in total drag coefficient depends on two geometric parameters: dimensionless water depth and bend sharpness, quantified as the bend radius of curvature to channel width ratio. We introduce a theoretical boundary layer separation model to explain this parameter dependence and to predict flow separation and the increased drag. The drag coefficient can increase by a factor of 2–7 in “sharp” and “deep” sinuous channels where flow separation is most likely. Flow separation also enhances energy dissipation due to increased velocities in bends, resulting in greater loss of tidal energy and weakened stratification. Flow separation and the associated drag increase are expected to be more common in meanders of tidal channels than rivers where point bars that inhibit flow separation are more commonly found. The increased drag due to flow separation reduces tidal amplitude and affects velocity phasing along the estuary and could result in morphological feedbacks.