Modelling the physical dynamics of estuaries for management purposes.

Abstract

South African estuaries are characterised by highly variable inflows owing to the semi-arid nature of the land mass which they drain. The interaction of this variability with that of the marine environment (seasonality, high wave events, synoptic effects) gives rise to the distinctive character of South African estuaries. In general, they are small, micro-tidal, bar-built systems with strong flood tidal dominance. Approximately half of the 273 systems along the coast exhibit intermittent closure of the mouth, while a number can become hypersaline during dry periods. In view of the increasing development pressures on the rivers and estuaries of South Africa and their strong dependence on freshwater flow for the maintenance of their character and functioning, and the need for justifiable, scientifically-based decision making regarding the freshwater requirements of estuaries is evident. This study was initiated to address this issue by first developing a model to simulate the physical dynamics of South African estuaries over time scales from months to years, so enabling prediction of the medium to long term consequences of alterations in the freshwater inflow on the abiotic components of an estuary. Thereafter, the efficacy of management policies involving water releases and mouth breachings could be evaluated in terms of their success in maintaining the character and functioning of an estuary. A semi-empirical estuarine systems model incorporating seven state variables, namely water volume, salt content, stratification, circulation, tidal flushing, freshwater flushing and the height of the sill at the mouth, was formulated and implemented on two case studies. Estuarine physics concepts were incorporated dynamically in the model in a novel manner. For instance, the bulk densimetric Froude number and the Estuarine Richardson number are used in the simulation of the stratification-circulation states, while the Ackers and White sediment transport formula was modified to yield results which agreed with field observations of the closure and breaching of the mouth of the Great Brak Estuary. Additionally, tidal exchange through the mouth was modelled phenomenologically and successfully calibrated against observations for both case studies. Model results were found to be fairly robust to uncertainties in parameter values. However, most encouraging of all is that behaviour known to occur in shallow estuaries, such as modulation of the n11.:.m water level by low frequency forcing and the generation of overtides, was reproduced by the estuarine systems model although it was not specifically included in the model formulation. The model is thus considered to reliably predict the physical dynamics of South African estuaries over time scales of months to years. A number of management policies involving freshwater allocations, water releases and breachings of the mouth (where appropriate) were tested on the two case studies, namely the Great Brak Estuary, a small, temporarily open system, and the permanently open Kromme Estuary. The results indicate an increase in marine dominance as freshwater flow to the estuaries decreases. The variability in the estuarine environment declines and the systems become more inert to freshwater flooding and more sensitive to marine forcing. By applying the estuarine systems modelling approach, the performance of different management policies could be evaluated in comparison with reference policies. Accordingly, for both case studies, preferred management policies which utilize the present total annual allocations to the estuaries more beneficially could be indicated. Further management applications included the use of the estuarine systems model in a linked system of abiotic and biotic models to facilitate more comprehensive prediction of the consequences of freshwater abstraction and so more informed assessment of estuarine freshwater requirements. The estuarine systems model results were critical in enabling the prediction of the faunal and floral responses in the intermittently closed Great Brak Estuary as it is presently the only abiotic model capable of simulating the closure and breaching of the estuary mouth over a number of years. It is anticipated that further developments will occur in biological prediction in the near future and that this could require developments or adaptations to the estuarine systems model, particularly when details of the type of information required for biological prediction becomes known. Additionally, the use of the estuarine systems model in a strategic management sense is suggested. It could play a role as a screening tool for regional water resource planning, while the preliminary quantification of the extent of anthropogenic influence in expediting the movement of estuaries towards the later successionary stage of a coastal lagoon is a powerful indication of the level of prediction which could become possible in the future. Thus enhanced management decision making is now possible on a site specific basis and at a more strategic water resources planning level.