Historically, a repeated concentration of discoloration events has been used to justify rehabilitation in the form of pipeline replacement or relining. This is no longer an acceptable management strategy, serviceability should be maintained to customers at least cost. Pipe renewal is unlikely to be justified by a sole driver, for example the potential for discoloration events. The development and realisation of proactive management and operation practices is therefore imperative. However, understanding of the detailed mechanisms that lead to discoloration events is limited and hence control is difficult.

Flushing programs have been identified as a means by which discoloration events may be managed. Hydrants are opened to create the hydraulic forces required to remove material that contributes to discoloration events from the localised distribution network.

Flushing is predominately carried out as a reactive measure in response to customer complaints. Accurate predictive modeling would allow the formulation of proactive flushing strategies within network operational management schemes, improving water quality and reducing customer complaints.




It has can be shown that once in suspension the materials collected during flushing activities are unlikely to deposit in water distribution networks. Therefore, the modelling situation is not one of erosion, transport and deposition, as for larger cohesionless sediments. The scenario that requires modeling in distribution systems is the generation, development and erosion of cohesive layers and subsequent material transport as a permanently suspended wash load or solute. These model concepts can be accommodated by consideration in conjunction with calibrated network hydraulic models.





Specific project objectives are to;

  • develop a model to predict discoloration events. The model will operate in conjunction with the EPANET distribution network simulation freeware;
  • collect hydraulic, temperature, continuously monitored turbidity and colour field data and concurrent water samples for the analysis for colour, turbidity, iron content and other water quality parameters in the laboratory;
  • analyse the field and water sample data with the aid of the model in order to further understand the chemical and physical mechanisms associated with discoloration in response to network operational events;
  • utilise the model in a sensitivity analysis to investigate the influence on discoloration of;
    • i) network make up (e.g. pipe material, diameter)
    • ii) hydraulic regime
    • iii) water quality
    • iv) temperature
  • undertake ongoing refinement of the model based on the findings of the fieldwork and analysis;
  • investigate the transferability of the model between areas of different water quality and between seasons;
  • derive an end of pipe model to describe the release of sediment build up from dead end mains upon hydrant operation during flushing;
  • investigate hydrant behaviour during flushing operation;
  • derive a basic computational routine/model to identify areas of most likely deposition of larger bed load type particles.
Password required