Observations of Mud‐Induced Periodic Stratification in a Hyperturbid Estuary

Abstract

This study presents new observations of fluid mud transport and of the interaction between mud‐induced stratification and the flow. Data collected in a hyperturbid estuarine tidal channel reveal details of the intratidal entrainment asymmetry, characterized by quasi‐instantaneous entrainment and upstream pumping of mud during flood, and a gradual reduction of layer thickness by shear dispersion during ebb. Rapid restratification early during the flood phase restores the predominant two‐layer structure and delimits the transport period, which is then significantly shorter than the overall flood duration. The hydraulic cross section is reduced, causing an increase of the salinity intrusion into the estuary at the end of the flood. The fluid mud layer occupies on average 40% of the water depth, and stratification exceeds that of highly stratified salt wedge estuaries. These data show how mud‐induced periodic stratification influences flow structure and sediment transport and thereby contribute to the understanding of the dynamics of hyperturbid estuaries.

Modeling the Transition to High Sediment Concentrations as a Response to Channel Deepening in the Ems River Estuary

“…These results capture the qualitative characteristics of the observed ETM; observations made by Talke, De Swart, and Schuttelaars () during one cruise in 2006 show sediment concentrations near the bed of 10–30 kg/m 3 and near the surface of approximately 1 kg/m 3 in the entire zone between km 35 and 60. Similar or higher near‐bed concentrations and fluid mud have been observed by Wang (), Papenmeier et al (), Winterwerp et al (), and Becker et al (). De Jonge et al () reports observed surface concentrations locally exceeding 3 kg/m 3 .…”

“…Once the transition to high sediment concentrations has occurred, the model lacks several physical processes that are essential to describe the sediment dynamics. This shows in the 2005 model results by an overestimation of the M 4 tidal water level amplitude and insufficient vertical structure to capture the observed distinct fluid mud layers (e.g., Becker et al, ). Furthermore, while lower sediment concentrations are observed during periods of high discharges (Winterwerp et al, ), the strong flushing found in the model for discharges higher than 70 m 3 /s does not correspond to the observations.…”

“…The model therefore lacks mechanisms that retain the fluid mud during periods of high discharges. Possible mechanisms that need to be included for a better description of the 2005 state are a critical shear stress for erosion, multiple sediment fractions, internal dynamics (i.e., the effect of a temporally variable density structure on the eddy viscosity and eddy diffusivity; Becker et al, ; Winterwerp et al, ), and dynamic effects that resolve the time scale for depleting a bottom pool of sediment (e.g. Schoellhamer, ).…”

“…After dredging operations that occurred between 1960 and 1994, the ETM had moved upstream and had become a wide turbidity zone between approximately Pogum/Terborg (km 15–25) and the tidal weir at Hebrum (km 64). Typical sediment concentrations in this ETM are 1–4 kg/m 3 at the surface (De Jonge et al, ) and 10–100 kg/m 3 near the bed (Becker et al, ; Papenmeier et al, ; Talke, De Swart, & Schuttelaars, ; Wang, ).…”

Acoustic Observations of Kelvin‐Helmholtz Billows on an Estuarine Lutocline

“…The velocities were measured with a Flowquest 600 kHz acoustic Doppler current profiler (ADCP) sampling at 0.5 Hz and recording an average over 60 s to remove high‐frequency variability, with a vertical bin size of 0.5 m. Velocity data within the near‐bed region (~0.07 H , where H is the water depth, Nystrom et al, 2007) are affected by sidelobe interference and are not used in further analysis. Although 600 kHz ADCP has been confirmed valid in current measurements in high SSC environments (Becker et al, 2018), the ADCP velocity records can be biased by spikes due to high SSC (Cao et al, 2012), which is the case in this study. Hence, a 11‐min moving average filter was applied to the ADCP velocity profiles to remove the spikes, similar to that of Becker et al (2018).…”