Les systèmes de transport intelligents, École Polytechnique de Montréal

Articles du groupe MADITUC

An integrated approach to understand water quality failures in distribution system

Référence:

GAUTHIER, Vincent, BESNER, Marie-Claude, TRÉPANIER, Martin, PRÉVOST, Michèle (2001). An integrated approach to understand water quality failures in distribution system, American Waterworks Association Distribution System Symposium, San Diego

Type:
Conférence

Organisme:
Autres

Retour

Nouvelle recherche

Résumé

Maintaining treated water quality throughout the distribution system (DS) up to the consumer’s tap remains today a real challenge. Degradation of the microbiological quality usually results from treatment plant breakthrough, from local contamination (intrusion), or from the survival and regrowth of micro-organisms. In this latter case, micro-organisms may have penetrated the system upstream or in the past, or may be associated to the deposits (biofilms, sediments) present in the system. While treatment breakthrough may be quite readily identified, the localization of contaminations downstream in the DS may be much more difficult to pinpoint, since transport and survival capacities of bugs are often ill-defined. As a result, only few studies were able to determine the origin of DS contaminations (Gedreich et al., Wat. Res.1992, Clark et al., Aqua, 1996, MacMath et al., WQTC 2000). While some attempts were made to predict coliform occurrences (Volk & Joret, Sciences de l’Eau, 1994; LeChevallier et al., AEM, 1996; Gale et al., Aqua, 1997), applications were not very successful which may probably be explained by the fact that such models did not considered local phenomena such as the hydraulic or the operation of the DS, but mainly relied on water quality data. These last types of data - which may help to explain water quality variations in DS - are very difficult to compare since most of them vary in time and space.

In the last years, major developments were brought these data types: flexible and affordable hydraulic models were made available to describe transport of contaminants throughout DS, and to provide water quality related data, such as the age of water at each location. On the other side, GIS provide a spatial reference to compile and visualize all types of data, which may be attributed to a pipe, a hydrant ort a civic address. This provides a unique opportunity to understand water quality failures in distribution systems by considering not only water quality at treatment plant and DS locations, but also all types of DS and water treatment plant events (maintenance and accidents), the link between data being achieved through the hydraulic model.

This research project presents the development and application of such an “integrated approach” to understand water quality failures in DS. This includes a number of steps such as (a) the selection of relevant databases and configurations; (b) the collection of such databases and their assemblage using homogeneous formats; (c) the geocoding of data to be able to display them on maps; (d) the exploration of these databases to identify relationships between water quality failures and all kind of “events” (both planned and accidental on the DS and at the treatment plant). Hydraulic model outputs are used to define the upstream and downstream sections around a quality failures, and to examine the low flow or old water areas.

This approach is tested with five large municipalities in Europe and North-America. Examined water quality failures are coliform occurrences, turbidity peaks and consumer complaints. Obtained results provides new insights on the fraction of water quality failures which may be attributed to DS operation, and reinforce the need to operate and maintain DS by taking into account the possible interactions with drinking water quality.

gbisaillon@polymtl.ca 2024-11-05 16:19:12