Odor Events in Surface and Treated Water: The Case of 1,3-Dioxane Related Compounds

A study has been carried out to identify the origin of the odorous compounds at trace levels detected in surface waters and in Barcelona's tap water (NE Spain) which caused consumer complaints. The malodorous compounds were 2,5,5-trimethyl-1,3-dioxane (TMD) and 2-ethyl-5,5-dimethyl-1,3-dioxane (2EDD) which impart a distinctive sickening or olive-oil odor to drinking water at low ng/L levels. Flavor profile analysis (FPA) or threshold odor number (TON) were used for organoleptic purposes. Levels up to 749 ng/L for TMD and 658 ng/L for 2EDD were measured at the entrance of the drinking water treatment plant. Three wastewater treatment plants (WWTPs) using industrial byproducts coming from resin manufacturing plants to promote codigestion were found to be the origin of the event. Corrective measures were applied, including the prohibition to use these byproducts for codigestion in the WWTPs involved. A similar event was already recorded in the same area 20 years ago.

Spatio-temporal variability of tastes and odors of drinking water within a distribution system

The threshold of human perception in the detection of tastes and odors (T&O) relating to compounds in drinking water is variable. For example, chlorine can be detected at the ppm level and geosmin can be perceived at the ppt level. In this paper, sensory tests (using a human panel), physicochemical analyses (total and free residual chlorine, temperature, metals, geosmin, and 2-methylisoborneol (2MIB)) and microbiological analyses (algae, Actinomycetes and heterotrophic plate count) were performed for water samples collected during a seventeen-month period at ten different locations of a municipal distribution network of Quebec City (Canada). The results showed that different flavors(1) assessed by a panel and aggregated into global flavor intensity (GFI) vary considerably spatially and seasonally. Multiple regression analysis showed that the factors best explaining the variability of GFI are (in order) the season, the location, the concentration of total residual chlorine and the presence of cyanobacteria. Results also demonstrate that chlorine has a masking effect on other T&O.