Fluorescence Excitation–Emission Matrix Regional Transformation and Chlorine Consumption to Predict Trihalomethane and Haloacetic Acid Formation

Abstract

Multifactor linear regression was used to correlate water quality characteristics of Iowa River water and four of its isolates to chlorinated disinfection byproduct (DBP) formation. Fluorescence regional integration (FRI) was used to measure changes in cumulative regional intensity of five operationally defined regions from excitation–emission matrices (EEM) subsequent to chlorination. The changes within the five regions alone correlated to the formation of Cl₃AA (R² = 0.75), Cl₂AA (R² = 0.38), CHCl₃ (R² = 0.33), and Cl₃AA + CHCl₃ (R² = 0.59). The combination of FRI and chlorine consumption increased the predictive capabilities for Cl₃AA (R² = 0.90), Cl₂AA (R² = 0.86), CHCl₃ (R² = 0.82), and Cl₃AA + CHCl₃ (R² = 0.92). Stepwise regression of fluorescence regions yielded similar results by utilizing only one region coupled with chlorine consumption; however, three different regions were required for the formation of each individual compound assessed. These results suggest that regional transformation coupled with chlorine consumption can be an effective tool for the prediction of DBP and provide insight to locations within the EEM for individual DBP compound precursor.