FLOCCULATION PROCESSES IN RIVER MOUTH FLUVIAL TO MARINE TRANSITIONS
Keyvani, Ali 1982-
MetadataShow full item record
Suspended mud delivered to coastal zones typically deposit as flocculated aggre- gates. This study aims to better understand how the flocculation process alters the size and properties of flocs in the unsteady turbulent shear conditions found in fluvial to marine transitions using a combination of laboratory and numerical experiments. The first set of experiments examines the impact of repeat exposure to multiple cycles of high and low shear rate on the process of flocculation (similar to cyclic tidal zones). To do this, a suspension of kaolinite and montmorillonite is introduced to a mixing chamber and allowed to flocculate under relatively low turbulent shear. Flocs are allowed to grow until an equilibrium size is reached. Following this, the suspension is deflocculated with vigorous mixing, and then reflocculated under the lower shear condition. This cycle is repeated seven times. The primary findings are that the equilibrium floc size is not dependent on the number of deflocculation and reflocculation cycles, whereas the floc growth rate up to equilibrium is. The second set of experiments examines continual shear-rate reduction on the evolution of flocs prior to reaching equilibrium (similar to river mouth jets and plumes). Flocs in these experiments do not form from primary particles under steady- state shear. Instead, at each successive shear rate, they grow from where they left off under the previous condition. Following this decay and growth of flocs, the shear rate is increased in a stepwise manner to examine breakup behavior. Results highlight how significantly floc settling velocity can change in such transitional conditions, sug- gesting that a static settling velocity might not be appropriate in fluvial to marine transitions. Data from these experiments are also useful for testing the ability of floc models developed from steady-state experiments to handle more natural unsteady conditions. The time-dependent mean floc size model of Winterwerp (1998) is used to model the data from both sets of experiments. An improvement to the Winterwerp model is proposed to better predict the reduction in growth rate that occurs as flocs ap- proach the equilibrium condition. The improvement comes through size-dependent aggregation and breakup coefficients.