Comparing RBF with Bench‐Scale Conventional Treatment for precursor reduction

Reduction of disinfection by-product (DBP) precursors upon riverbank filtration (RBF) at three drinking water utilities in the midwestern United States was compared with reductions obtained using a bench-scale conventional treatment train on the corresponding river waters. The river waters were subjected to a treatment train consisting of coagulation, flocculation, sedimentation, filtration, and ozonation. RBF performed as well as or better than the bench-scale conventional treatment with respect to DBP precursor removal. Total and dissolved organic carbon concentrations were reduced by 20 to 50% after bench-scale treatment, compared with reductions between 30 and 70% after subsurface travel to the closer wells at the three sites. Reductions in precursor material for a variety of DBPs (trihalomethanes, haloacetic acids, haloacetonitriles, haloketones, chloral hydrate, and chloropicrin) after bench-scale treatment were generally in the range of 40 to 80%, whereas reductions after RBF ranged from 50 to 100%. After RBF and bench-scale treatment, a shift was observed from the chlorinated to the more-brominated DBP species, with the shift more pronounced for the bank-filtered waters. This shift was likely attributable to the increase in the ratio of bromide to dissolved organic carbon.