You are here:
Modeling and testing of reactive contaminant transport in drinking water pipes: Chlorine response and implications for online contaminant detection
Citation:
YANG, Y. J., J. A. GOODRICH, R. M. CLARK, AND S. Y. LI. Modeling and testing of reactive contaminant transport in drinking water pipes: Chlorine response and implications for online contaminant detection . doi:10.1016/j.watres, M. Henze (ed.), WATER RESEARCH. Elsevier Science Ltd, New York, NY, 42(6-7):1397-1412, (2008).
Impact/Purpose:
to provide information
Description:
Reactive contaminants introduced to chlorinated drinking water can cause water quality change directly related to their reactivity and other physiochemical properties. This general principle is further developed and utilized in a proposed real-time event adaptive detection, identification and warning (READiw) method for paired sensor monitoring stations in a water distribution pipe. The READiw process begins with adaptive treatment of sensor outputs to enhance contaminant signals, followed by anomaly detection using a statistical coarse filter, detection verification and classification in multiple criteria that include the relationships among measured relative changes of multiple parameters, inferred contaminant properties in chemical discrimination diagrams, and the reactive mass transport between the sensor stations. This paper describes principles and utilities of the READiw method, and examines its applicability for a composite dataset generated in a series of pilot-scale flow experiments for 11 contaminants at 3 initial concentrations. After adaptive treatment of the sensor outputs, reaction and kinetic differences become apparent between pesticide and herbicides (aldicarb, glyphosate and dicamba), alkaloids (nicotine and colchicine), E.coli and biological growth media (nutrient broth, terrific broth, and tryptic soy broth), and inorganic chemicals (mercuric chloride and potassium ferrocyanide). The differences are consistent with previously published contaminant properties, providing the basis for classification. In application to the composite dataset, the minimal sensor responses for accurate contaminant event detection are 0.1% for pH and ORP, 0.9% for free chlorine, 1.6% for total chlorine, and 0.9% for chloride, which are significantly smaller than the background variations or noise.
