AWWA JAW49837

Utility experiences were examined to assess the reliability of solubility models and decision trees in forecasting control of lead (Pb) and copper (Cu) corrosion byproducts. Such approaches cannot quantitatively predict effectiveness of corrosion control. For example, even the best-fit solubility models explain only about 35 and 18 percent of the quantitative variation in 90th percentile Cu and Pb release, respectively. However, the decision trees and solubility models are qualitatively consistent with utility experience and are thus useful when promising water quality changes are identified to mitigate corrosion byproduct release. Utility experiences confirm that an optimal alkalinity range (20-40 mfg/L as Ca co3) exists for Pb corrosion control above pH 8.5, a higher mass ratio of chloride to sulfate tends to worsen Pb byproduct release, and detectable color can signal a natural organic material content that may adversely affect compliance with the Cu action limit. Includes 35 References, figures.