부산대학교 도서관

Wave-induced ripples are significant tools in reconstructing a range of paleo-environments and in establishing predictors of sediment transport in modern estuarine, marine and coastal environments. Previous studies, however, have focused on the dynamics of bedforms composed primarily of well-sorted, non-cohesive, sand. This restriction to one sediment type is far from being a realistic representation of natural sediment size distributions. This thesis reports on a flume experiments and aligned field investigations of rippled formed in mixed (cohesive and non-cohesive) sediments substrates under combined flows. The flume experiments were conducted in the Total Environment Simulator at the University of Hull and comprised of 6 separate runs, in which 5 runs were conducted under identical sets of regular (monochromatic) waves with period of 2.48 s. A control experiment, with a pure non-cohesive sand bed, proceeded runs where the bed clay content was systematically varied in its composition ranging from a bed comprising 4.2% clay through to 7.4%. A series of state-of-the-art measurements were employed to quantify interactions of near-bed hydrodynamics, sediment transport, and turbulence over rippled beds formed by wave action. The experimental results demonstrate the significant influence of cohesive clay materials in the substrate on ripple evolution under waves. Most importantly, addition of clay in the bed remarkably slows the rate of ripple development and ripple evolution. Both equilibrium time of wave ripple length and height exponentially increased with increases in clay content, from 40 minutes to 200 minutes for ripple wavelength and from 30 minutes to 120 minutes for ripple height. The slower ripple growth rates with higher cohesive substrate fractions is shown to alter the critical shear stress and reduce clay winnowing rates. However, the results also highlight that the equilibrium size of ripples is independent of increasing substrate clay fraction, with equilibrium length and height standing around 140 mm and 20 mm, respectively. Laser granulometry of cores from the final substrates verified that ripple crests were composed of pure sand layers and that substrates within the ripple troughs remained mixtures, reflecting a relatively higher winnowing efficiency at wave ripple crest. The winnowing process, and its efficiency, is shown to be inexorably linked to wave ripple development and evolution. Furthermore, the remaining clay was able to stabilize the rippled bed, ripple migration rate decreasing from 0.06 mms−1 to 0.02 mms−1 and bed sediment flux falling from 0.43 mm2s−1 to 0.22 mm2s−1 from the non-cohesive run to the run with the highest initial concentrations of clay. Finally, the results also highlight the influence of clay concentrations on damping of turbulence, which is related to the stratification of suspended sediments. In concert with the flume experiments, a series of field surveys were conducted at Red Cliff sand bar in the upper Humber estuary NE England. The fieldwork aims were to study bedform dynamics under combined current-wave flows. Field surveys were conducted in a matrix of periods covering spring and neap tides and during periods of high and quiescent wind conditions. Field surveys were also conducted across both winter and summer periods to investigate the influence of biological cohesive Extracellular polymeric substances (EPS) on ripple size. The results identified that two-dimensional current-induced ripples dominated under low wave forcing calm conditions. While the bed morphology under high wind conditions with significant wave forcing on the bed resulted in a morphology dominated by plane bed and 2D wash-out ripples. Finally, although the EPS content increased from 0.003% in winter to 0.013% in summer survey periods, its influence on the size of the ripple size was very negligible. However, the ripple symmetry index (RSI) does show a distinctive seasonal trend, with ripples forming in winter being more asymmetrical than those in the summer. Further work is needed to tease out the complex interactions of biological factors and combined flows in estuarine settings. However, these results here highlight a complex interplay, which has significant implications for paleoenvironmental reconstructions.