A 1989 survey provides insights on how utilities perceive taste and odor problems in their water supplies, what treatment processes they use, and which processes they rated effective. A taste and odor questionnaire was sent to 826 water utilities. A total of 377 usable responses were received from across the United States and Alberta, Canada. Sixteen percent of the utilities considered their taste and odor problems to be serious, and 43 percent had experienced a taste and odor event lasting longer than one week. The major odor problems reported by the utilities were characterized as chlorine, earthy, fishy, and medicinal; major taste problems were reported as sour, metallic, and chlorine. There were differences between the taste and odor problems encountered by utilities treating groundwater and those treating surface water. Most utilities attributed taste and odor problems to the distribution system. In addition, groundwater plants identified the disinfectant used and mineral residuals as other sources of taste and odor problems. Surface water systems blamed plankton and decaying vegetation in the influent water, the disinfectant used, and conditions at reservoirs.

Water treatment plants in the US may operate under the assumption that chlorine masks earthy and musty odors from geosmin and 2-methylisoborneol (MIB) in drinking water. To test this hypothesis, we evaluated the effects of chlorine and chloramines on geosmin and MIB by two sensory analysis approaches--a statistical Pairwise Comparison Test, and Flavor Profile Analysis (FPA). All Pairwise Ranking test statistics were significant (p<0.05); we conclude that panelists can differentiate minor differences in geosmin and MIB concentrations in a Pairwise Comparison Test even in the presence of chlorine. FPA appeared to be more challenging in discerning subtle differences in concentrations of geosmin or MIB than did the Pairwise Comparison Test, and the presence of chlorine (0.5-20 mg/L) and chloramines (3-24 mg/L) confused the panelists (i.e showed a larger error in the intensity of response reported by the panel), but did not necessarily mask geosmin or MIB.