Characterization of dissolved organic matter in drinking water sources impacted by multiple tributaries

The characterization of dissolved organic matter (DOM) in drinking water sources is important as this material contributes to the formation of disinfection by-products (DBPs) and affects how water treatment unit operations are optimized. Drinking water utilities often draw water from sources impacted by multiple tributaries, with possible shifts in DOM concentrations and reactivity over time, depending on specific environmental conditions. In this study, results are presented on the characterization of DOM under varying ambient conditions from the four main tributaries of Lake Mead, a large reservoir in the southwest United States. The tributaries include the Las Vegas Wash (LVW), Muddy River (MR), Virgin River (VR) and the upper Colorado River (UCR). One additional sample was collected at the outflow of the reservoir (lower Colorado River (LCR)). The DOM was characterized by both bulk parameters (specific ultraviolet absorbance (SUVA)) and specific physicochemical properties, i.e. size, polarity and fluorescence. The analyses were performed emphasizing limited changes in its natural configuration by eliminating analytical preparation steps, excluding sample filtration (0.45 microm filter). Results indicate that each tributary had a different molecular weight distribution, as well as fluorescence properties, which helped in the identification of the relative source of DOM (allochthonous versus autochthonous). The largest apparent molecular weight distribution was observed for DOM samples collected at the MR site, which is fed mostly by groundwater seepage. The smallest apparent molecular weight was observed for DOM collected at the LCR site, suggesting that retention in the reservoir resulted in a decrease in molecular weight as a probable result of photo oxidation and microbial processes. Fluorescence analysis aided the differentiation of DOM by clearly identifying waters that were affected by microbial activity (LVW, UCR, and LCR), either by wastewater influence or by autochthonous processes, versus limited microbial influence (MR and VR). Polarity analysis revealed clear differences in the hydrophobic/hydrophilic nature between waters, including temporal differences within individual waters at a particular site. The DOM from the LVW and VR sites had higher hydrophobic character, as measured by retention onto non-polar sorbents. Additionally, the DOM collected at the LCR had the least hydrophobic character. This type of analysis would be beneficial to utilities who want to better understand and manage their source waters, especially in the evaluation of temporal variation within a watershed.