Laboratory model studies of flushing of trapped salt water from a blocked tidal estuary

Abstract

Results are presented from a series of laboratory model studies of the flushing of saline water from a partially- or fully-closed estuary. Experiments have been carried out to determine quantitatively the response of the trapped saline volume to fresh water Hushing discharges Q for different values of the estuary bed slope a and the density difference (Δρ)0 between the saline and fresh water. The trapped saline water forms a wedge within the estuary and for maintained steady discharges, flow visualisation and density profile data confirm that its response to the imposition of the freshwater purging flow occurs in two stages, namely (i) an initial phase characterised by intense shear-induced mixing at the nose of the wedge and (ii) a relatively quiescent second phase where the mixing is significantly reduced and the wedge is forced relatively slowly down and along the bed slope. Scalings based upon simple energy balance considerations are shown to be successful in (i) describing the time-dependent wedge behaviour and (ii) quantifying the proportion of input kinetic energy converted into increasing the potential energy of the wedge/river system. Measurements show that the asymptotic value of the energy conversion factor increases with increasing value of the river Froude number Fr 0 at small values of Fr 0, thereafter reaching a maximum value and a gradual decrease at the highest values of Fr 0. Dimensional analysis considerations indicate that the normalised, time-dependent wedge position (x w)3 (g′)0/q 2 can be represented empirically by a power-law relationship of the form (x w)3 [(g′)0/q 2]⅓ = C {(t)[(g′)0 2/q]⅓} n where the proportionality coefficient C is a function of both Fr 0 and the slope angle a and the exponent n has a value of 0.24. Successful attempts are made to relate the model data to existing field observations from a microtidal estuary. Experiments with multiple, intermittent periodic flushing flows confirm the importance of the starting phase of each flushing event for the timedependent behaviour of the saline wedge after reaching equilibrium in the intervals between such events. For the parameter ranges investigated and for otherwise-identical external conditions, no significant differences are found in the position of the wedge between cases of sequential multiple flushing flows and steady single discharges of the same total duration.