Results of a large number of erosion tests on artificially generated and relatively dense sand-mud mixtures are presented. Soil sample compositions are varied concerning clay-silt and sand-silt ratio, and clay mineralogy. The experimental setup consists of a re-circulating small-scale rectangular erosion flume with unidirectional flow conditions. The erosion threshold and erosion rate are studied through step by step increasing the flow rate during a test. Results clearly indicate time-decreasing erosion during which individual flocs are randomly eroded, and time-independent (steady) erosion during which both sand and mud particles are continuously and uniformly eroded. These two erosion types appear to be floc and surface erosion, respectively (Winterwerp and Van Kesteren, 2004). Floc erosion relates to the stochastic character of both the flow conditions and (surficial) sediment strength, whereas surface erosion relates to the plasticity index, which is a bulk soil mechanical parameter characterizing cohesiveness. The surface erosion threshold is discussed following a geotechnical approach, which argues that surface erosion is a drained process. This implies that cohesiveness rather than packing density is important for the erosion threshold, which is confirmed by the experimental data. Simultaneously with the erosion tests, also the undrained shear strength of the applied soil samples was determined. A model is proposed and validated to predict the undrained shear strength as function of the granular porosity in combination with the plasticity index. The comparison of the undrained shear strength with the surface erosion threshold further confirms the applicability of a geotechnical approach to understand the erosion of mixed sediments. Finally, the study provides a valuable data set that can be used as a reference for future research on erosion behavior of (natural) sediment mixtures.

[1] This paper proposes a conceptual framework for erosion of cohesive sediment beds. We focus on cohesive beds, distinguishing between floc erosion, surface erosion, and mass erosion. By (our) definition, surface erosion is a drained soil mechanical process, whereas mass erosion occurs under undrained conditions. The eroding shear stress is modeled through a probability density function. This yields a continuous description of floc erosion and surface erosion as a function of mean bed shear stress. Furthermore, we assume a distribution for the bed strength. The mean values of the bed strength are derived from soil mechanical theory, assuming that the surface erosion rate is limited by the swelling rate from the undrained shear strength in the bed to its drained value at its surface. The rate of erosion then relates to the undrained shear strength of the soil, and its consolidation (swelling) coefficient. The critical shear stress for erosion is slightly larger than the true cohesion of the bed, i.e., the drained strength, and follows a power law relation with the plasticity index. The conceptual framework proposed herein has been validated against a limited number of experimental data, and has a series of advantages above other methods of direct measuring erodibility, as it is inexpensive and can be used to attain space-covering information on the sediment bed. Moreover, the use of bulk soil mechanical parameters accounts implicitly for the effects of organic material, though the role of, e.g., macrophytobenthos mats and/or bioturbation is difficult to capture a priori.

No abstract

Information on the sedimentological composition of sediment beds in marine wetlands is important for the study of the complicated interactions between physical, biological and chemical processes. In situ soil sample collection and subsequent laboratory analyses using traditional methods is rather time consuming. The present paper presents the Medusa (Multi Detector system for Underwater Sediment Activity) RhoC system. ‘Rho’ refers to density and ‘C’ to the activity concentration of the decaying isotopes adhered to the sediments. The new instrument directly translates (the attenuation of) natural radioactivity to sedimentological data concerning the depth-averaged sediment composition and vertical density profiles of the upper 15 cm of the sediment bed. The accuracy and applicability of the instrument were assessed to illustrate its potential and limitations. Results from a field campaign on several intertidal flats and from similar measurements in the laboratory for controlled circumstances were compared with data obtained by traditional analyses. The instrument generates accurate results for the depth-averaged sediment composition. Vertical density profiles are also well represented by the RhoC after smoothing and correcting the data for partly saturated soils. Thus, Medusa RhoC is a useful and practical tool to provide accurate sedimentological data in a fast and cost-effective way. The combination of sedimentological relations with the data obtained by RhoC further increases the applicability of the new instrument.