Water quality modelling of eutrophied reservoirs in South Africa
Doctoral Thesis
1996
Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
License
Series
Abstract
Governmental agencies in South Africa became concerned about the increase in eutrophication-related water quality problems during the early 1970's. The first step taken to control eutrophication was introduction of an effluent phosphate standard that limited the phosphorus concentration in effluents being discharged in certain sensitive catchments to a maximum of 1 mg P04-P 1-1. This standard applied only to point sources, because of an initial belief that the contribution from non-point sources was relatively minor, and the absence of practical economic measures to control phosphorus discharges from non-point sources. Subsequent to introduction of the 1 mg P Standard several modelling studies were undertaken, as there was a need to describe the response of eutrophic reservoirs to altered phosphate inputs. Most of the work was done on the hypertrophic Hartbeespoort Dam reservoir. The models utilised were empirical, zero-dimensional models that treated the reservoir as a completely mixed reactor. Usually these models considered only the steady state, or at most, annual changes. The models simulated annual mean phosphate-P concentrations with varying degrees of success, but a significant relationship between observed and simulated chlorophylla concentrations could not be obtained, i.e. these models could not be used to predict the response of eutrophic reservoirs to different management strategies aimed at alleviating eutrophication-related water quality problems. Consequently, a further study was initiated by the Water Research Commission to test the applicability of more sophisticated hydrodynamic and water quality models, developed in the USA and Australia, to stratified reservoirs under South African climatic conditions. Several models of varying complexity were available. From these four models were selected for study. The models that were tested were the one-dimensional models DYRESM (developed in Australia), and MINLAKE (developed in the USA) , and the two-dimensional models CE-QUAL and WASP (developed in the USA). The DYRESM model can simulate hydrodynamic behaviour only, whereas MINLAKE, CE-QUAL and WASP can simulate both hydrodynamic and water quality behaviour. This report covers the extensive study that conducted on the MINLAKE model. The study is justified in view of the potential of the MINLAKE model to evaluate different treatment options: of the four models selected, it is the only model that can simulate more than one algal class. Thus it is an ideal tool to assess the effect of a chosen treatment option on, for example, algal succession.
Description
Includes bibliography.
Keywords
Reference:
Venter, A. 1996. Water quality modelling of eutrophied reservoirs in South Africa. University of Cape Town.