Conclusions, Implications for Drinking-Water Treatment and River Management, and Possible Future Studies


While some of the initial results of this study were used to inform recent upgrades to the water-treatment plant at the LO DWTP (Kari Duncan, City of Lake Oswego, oral commun., 2010), it is hoped that future studies build on this research and identify specific areas or activities that produce DPB precursors so that watershed-management efforts can be targeted and wisely prioritized.


The DOM precursor pool in the Clackamas River basin was strongly influenced by season, streamflow, and perhaps most notably, the effects of storms. However, not all storms were alike in terms of the amount or character of carbon exported, which reveals that factors such as time of year, antecedent flow, snowpack conditions, and patterns in rainfall and runoff affect carbon export. This study identified the primary sources of organic matter that contributed to DBP precursors in raw source-water supplies in the lower Clackamas River, which turn out to be primarily dissolved organic compounds that are terrestrial in nature. Multiple lines of evidence also supported the hypothesis that algae may, at times, be contributors to the DBP precursor pool, especially THMs. 


Higher DBPFPs in some unfiltered samples compared to filtered samples suggested that, at times, a considerable amount of the total DBP precursor pool was composed of filterable particles such as detritus, soil particles, and algae. This suggests that filtration prior to chlorination could reduce finished-water DBP concentrations. Differences between watershed THM and HAA precursor sources, as well as their treatability, were evident, suggesting different actions may be necessary to manage for these types of DBP precursors. 


Although terrestrial sources of carbon dominated the Clackamas River in 2010–11, some of the results obtained here suggest that algae also contributed some carbon to the river. While benthic algae reached nuisance levels in the mainstem and caused large and synchronized swings in dissolved oxygen concentrations and pH—key photosynthesis indicators—conditions did not lead to clogged intakes or substantial sloughing as has occurred years past. Such material also contributes carbon that, to some degree, contains DBP precursors. Future in-situ, high-frequency monitoring of FDOM may present opportunities not offered during this study to evaluate the degree to which algae may affect source-water quality and DBP concentrations in treated water by providing an early warning that allows for timely sampling of such conditions. Although algal blooms appear to be a regular phenomenon, weather conditions and growth–senescence processes are dynamic in the Clackamas River as algae accumulates on the riverbed or as blooms develop in the reservoirs. Sampling finished-water DBPs during periods when water-column chlorophyll-a is high at the Oregon City (or Estacada) monitors, or during periods of obvious sloughing of benthic algae or after particularly large blooms of blue‑green algae in the reservoirs, could provide data to further evaluate this hypothesis. The occasional taste and odor event also provides opportunities to identify where in the system these compounds are coming from, and may be ideal times to screen for blue-green algae toxins (Graham and others, 2010). The high specific THMFP value from North Fork Reservoir suggests this source is worthy of future monitoring if THM concentrations continue to be a concern. Studies that examine the temporal changes in organic matter reactivity and potential production of DBP precursors, taste and odor compounds, and algal toxins over the course of a bloom could be helpful for understanding the effect that algae may have on downstream water quality. In addition to providing a better understanding of watershed processes, this knowledge might help guide future DBP-treatability approaches that remove DOC, TPC, and DBP precursors. This could yield great benefits, especially if algal-derived organic matter is less amenable to removal by coagulation methods employed at these direct-filtration plants.


High frequency, in-situ measurements of FDOM proved to be an excellent proxy for DOC concentration in the Clackamas River, suggesting further development and refinement of these sensors have the potential to provide information that can inform DWTP operations and upgrades. The in-situ FDOM measurements revealed short-term, rapid changes in DOC concentrations in the river in response to storms. For this system, the close association between source‑water in-situ FDOM and finished-water HAA5 concentrations demonstrates the utility of these instruments in providing a robust proxy for DBPs continuously, in real‑time. This technology represents a new tool that can be used to optimize treatment-plant operations by adjusting water‑intake rates or modifying coagulant doses during critical time periods, for example, to minimize the DOC and DBP-precursor content of treated water (Kraus and others, 2010). The link between DOM fluorescence and the presence of DBP precursors for other classes of DBPs other than THMs and HAAs, including for example nitrogenous DBPs, which are of emerging concern, clearly warrants future study (Hua and others, 2007; Henderson and others, 2009; Chen and Westerhoff, 2010). Research into the use of fluorescence to monitor and predict treatability, estimate biological oxygen demand, identify and quantify wastewater inputs, and act as an early-warning system for contaminants also shows promise (Bieroza and others, 2008; 2009a, b; Hudson and others, 2008; Henderson and others, 2009; Baghoth and others, 2011; Bridgeman and others, 2011; Goldman and others, 2012). 


Other studies could sample forest soils and streams to refine the current understanding of what processes lead to DOM leaching, where DBP precursors are most prevalent, and how these processes are trending over time. Information on how carbon is stored and released would help modeling and prediction of DBP precursors and be especially helpful for understanding the effects of potential changes in land management, fire, climate, precipitation patterns, and other factors. Bacterial and fungal activity within soils and decomposing wood on the forest floor likely play important roles in mediating carbon sequestration and losses over time, and more information on their status would be useful. Future studies might also examine how forest management influences forest-soil carbon dynamics and the types of organic matter (or “components” as described here) that are most prone to leaching, and which ones contribute DBP precursors. Additional measures of bromide and chloride concentrations from Austin Hot Springs in the upper basin would be helpful for detecting possible trends over time, and might provide insight into the seasonality of DBP speciation in finished drinking water.


Monitoring of in-situ DOM using high-frequency continuous FDOM sensors can provide information regarding trends in the amount and composition, and thus origin and reactivity, of organic matter present in the river and in source water. These data have not only the potential to provide real-time information that can be used to manage DWTP operations, better understand DOM treatability, and predict finished-water DBP concentrations, but also provide information about watershed hydrology and processes that affect carbon dynamics in both terrestrial and aquatic ecosystems.


First posted February 11, 2013