The hydrology of the upper Swan River Estuary with focus on an artificial destratification trial

A trial to artificially destratify part of the water column of the Swan-Can ning Estuary took place over four weeks in 1997. Destratification was attem pted with bubble plumes created by pumping compressed air through a porous pipe near the bed at a location in the upper estuary. The purpose of the tr ial was to determine whether the resultant bubble 'curtain' could mix the w ater column, thereby altering the extent of upstream and downstream mixing and the vertical structures of salinity, temperature, dissolved oxygen and turbidity. The destratification trial was monitored using three techniques: vertical profiles taken on several days at selected stations along the est uary, fine scale profiles in the vicinity of the bubble curtain on one day, and continuous water column profiles taken I km upstream of the curtain. T he seasonal upstream propagation of brackish water underneath a residual fr eshwater discharge was affected by the magnitude of freshwater discharge an d tidal elevations. Superimposed upon diurnal and semi-diurnal tides were n on-tidal water level changes induced mostly by variations in barometric pre ssure. These caused large changes in the salt wedge position. Destratificat ion by the bubble curtain was compromised by these large oscillations, whic h limited the exposure of the salt wedge region to the mixing action of the bubble curtain. Complete vertical mixing of the water column was observed up to 30 m either side of the curtain on 30 October. Disruptions to the den sity stratification were not evident beyond 350 m of the curtain. In the im mediate vicinity (similar to 30 m) of the curtain, deficits of dissolved ox ygen in bottom waters were generally reduced or obliterated and a naturally turbid plume of water near the bottom of the estuary was mixed through the water column. The effects of the bubble curtain on dissolved oxygen, turbi dity and temperature were, however, similar to those for salinity, with mix ing confined to around 30 m from the curtain and no effect observed further than 350 m away. The upper Swan River estuary, although relatively narrow and strongly vertically stratified, is unsuited to destratification using b ubble plumes. Cycling of spring and neap tides and non-tidal water level ch anges, together with freshwater inflows to the estuary, strongly limit the longitudinal extent of mixing by bubble curtains. Copyright (C) 2001 John W iley & Sons, Ltd.