Abstracts
Résumé
Cet article rapporte les résultats de campagnes de mesure de Carbone Organique Dissous Biodégradable sur 2 sites en exploitation. Les mesures de CODB ont été effectuées par la méthode rom et tom. La première partie de l'étude concerne le site de KERNE UHEL. Il s'agit d'une usine de traitement d'eau de surface comprenant une étape d'ozonation suivie d'une filtration sur CAG récemment mise en place. L'ozonation provoque une augmentation de CODB de 0,3 à 0,6 mg O. Le CAG élimine le CODB : on a 100 % d'élimination lorsqu'il est neuf, et jusqu'à 80 % d'élimination sur un pilote après 1 an de fonctionnement. Dans le réseau, an observe une diminution du CODB lorsque la concentration de celui-ci est de 0,6 - 0,7 mg 1-1 dans l'eau traitée, cela malgré la présence de chlore libre. Lorsque le CODB de l'eau traitée est de 0,14 mgl-1 (après mise en route de la filière CAG) l'évolution dans le réseau est beaucoup plus difficile à mesurer (0,14 < CODB < 0,27 mg l-1). Le deuxième site étudié est le réseau de VIENNE-BRIANCE. 4 campagnes de mesures effectuées entre Juin et Août 1991 sur 10 points de prélèvement, montrent qu'il y a consommation de CODB, surtout en période chaude et aux extrémités du réseau. On a également des relargages de CODB, un nombre significatif de valeurs se trouvant au-dessus de la valeur maximale observée à l'usine. On remarque une grande variabilité des valeurs sur le réseau (< 0,1 à 0,6 mg l-1) en comparaison avec la gamme de valeurs obtenues sur l'eau traitée (0,21- 0,30 mg l-1).
Mots-clés:
- Eau potable,
- matière organique biodégradable,
- réseaux de distribution
Abstract
This study reports results of BDOC measurements from two experimental sites. BDOC was measured by JORET and LEVI’S method, in which a sample is incubated at 20 °C with colonised sand. Dissolved Organic Carbon is evaluated every day, and BDOC calculated as Initial DOC minus the minimal DOC observed during incubation. Reliability of this method was verified by measuring BDOC of standard acetate solutions. We obtained coefficients warrying between 2 and 5 % for BDOC about 1,0 mg l-1.
The first part of the study concerns KERNE UHEL treatment plant and its distribution network. BDOC was measured on sand filtered water, ozonated water, GAC filtered water, treated water and on five reservoirs along the network. Results from the treatment plant show that ozonation generates BDOC (increase of 0.3 to 0.6 mg l-1) GAC filtration allows very low levels of BDOC to be obtained when filtration start up (< 0.1 mg l-1) On GAC pilot plant, BDOC removed was between 50 % and 80 % after one year. In distribution, BDOC concentration 'decreased tram 0,6 - 0,7 mg l-1 to 0,2 - 0,4 mgl-1, during an estimated residence lime of 190 h. When the entire GAC filters were commissioned on the treatment plant, treated water BDOC was equal to 0.14 mg l-1; and between 0.14 and 0.27 mg l-1 in the network. This shows that BDOC reduction in the treatment plant reduces its evolution in the distribution system.
The second part of the study was conducted on the distribution network of VIENNE BRIANCE, which is supplied with water from the river VIENNE. This water is treated by coagulation, settlement and sand filtration. This network is very extensive and provides water to more than 20 rural villages. Ten points were selected on the network, and were monitored for BDOC, chemical and bacteriological parameters from June to August 1991. Results of physico chemical analyses showed variation of temperature (-3 °C to + 3 °C in the network compared to the treated water at the station); pH (slight variations < 0,5 pH units) and turbidity (increase in the network with peak values as high as 3,5 NTU). Treatment plant's BDOC values were between 0.2 - 0.4 mg l-1. In the network, a great variation is observed : < 0.1 to 0.6 mg l-1. This shows that microbiological activity in the system causes BDOC consumption, but also BDOC is released kilo the water. Heterotrophic bacteria coutils increases during distribution and filamentous fungi have also been isolated in this network.
The results of this study pointed out the difficulty to interprete BROC evolution in the network, especially when the hydraulic of the system is not well understood.
Keywords:
- Drinking water,
- biodegradable Organic Matter,
- Distribution Network