Chlorine decay models provide efficient ways to develop disinfection strategies for water distribution systems, provided they account separately for bulk and wall decay, and accurately describe decay with a single set of coefficients. The augmented two-reactant (2RA) model is shown to be the simplest model to accurately describe effects of rechlorination dose/timing on bulk chlorine decay, in combination with effects of initial concentration and temperature over long periods. The two-reactant (2R) and variable reaction-coefficient (VRC) models provided predictions of comparable accuracy under higher and successive rechlorination doses at constant temperature. However, the 2RA model provides a more general basis for strategy development, as the VRC model cannot describe the effect of temperature variation. The minimal data-set required for 2RA calibration was similar for all cases considered. The 2RA model is readily applied by incorporation into system modelling software such as the multi-species extension (MSX) to EPANET software.

Management goals for many distribution systems include maintaining a minimum level of free chlorine and limiting disinfection by-products. As water resources become scarce and quality deteriorates, blends are often required, so achieving these goals becomes increasingly difficult. The augmented two-reactant (2RA) model describes chlorine decay in a single water, for various dosing levels and water temperatures. However, it is not known whether there is any effect on chlorine decay (either synergistic or antagonistic) arising from blending waters of distinctly different quality. Linked 2RA models of source waters were used to determine whether bulk decay in blends of various source waters could be accurately and generally modeled. Results showed that chlorine decay in blends of various waters could be described accurately without synergistic or antagonistic effects, implying that each water’s reactants reduced chlorine independently. This held for pairs of surface waters, groundwaters, and raw or t...

Using a laboratory-scale system, consisting of a primary disinfection tank (PDT) and three intermittently mixed reactors (R1–R3) in series, bulk water and biofilm contributions to chlorine decay were quantified. The reactors (surface-to-volume ratio: 23.7 m⁻¹; retention time in each reactor: 42.6 ± 1.18 h) were fed with plant-filtered water (PFW). Secondary disinfection was carried out in R1. Free chlorine concentration decreased with travel time (R1: 1.2 mg/L; R2: 0.6 mg/L; and R3: 0.12 mg/L). The bacterial number (ATP) decreased from 67 pg/mL in PFW and remained at ~2–3 pg/mL in R1 and R2 but increased back to 68 pg/mL in R3. First-order chlorine decay rate coefficients decreased from R1 to R2, as expected, but increased by five-fold from R2 to R3. The increased bacterial number (ATP) in R3 and batch chlorine decay tests confirmed that bulk water (soluble compounds, microbes, and sediments) contributed approximately 40% of the decay, and the biofilm contributed 60% in R3. When ATP levels in the reactors were combined with literature data, the bacterial number increased significantly when free chlorine decreased below 0.2 mg/L, but data between 0.2 and 0.5 mg/L are limited. More investigation is needed in the future for chlorine < 0.5 mg/L regarding bacterial regrowth and its effect on bulk water chlorine decay.