Effect of prescribed fires on the export of dissolved organic matter, precursors of disinfection by-products, and water treatability.

In this study, we report for the first time the effect of prescribed fires on the export of dissolved organic matter (DOM) and precursors of disinfectant by-products (DBPs) from periodically (every 2-3 years) and seasonally (i.e., dormant and growing) burned forest fuel materials (i.e., live vegetation, woody debris, and detritus [litter and duff]) and treatability of its rainwater leachate. Periodically applied (every 2-3 years for 40 years) prescribed fires decreased total fuel load (62±10%), primarily detrital mass (75±2%). However, functional groups (i.e., phenolic compounds, proteins, carbohydrates, aromatic [1-ring], polycyclic aromatic hydrocarbons [PAHs], and lipids) attached to DOM of ground solid materials did not change significantly. Outside rainwater leaching (from forest fuel materials) experiments showed that the leaching capacity of dissolved organic carbon (DOC) from burned litter samples decreased by 40±20% regardless of burning season when compared to unburned litter samples. The leaching of total dissolved nitrogen (TDN), dissolved organic nitrogen (DON), ammonium (NH4+), and reactive phosphorus (PO43-) from burned materials decreased between 40 and 70% when compared to unburned materials. Also, DOM composition was affected by prescribed fire, which partially consumed humic-like substances based on fluorescence analyses. Thus, periodically applied prescribed fires also resulted in a reduction of trihalomethane (THM) (42±23%) and haloacetic acid (HAA) (42±20%) formation potentials (FPs), while DOC normalized reactivity of THM and HAA FPs did not change significantly. Additionally, the leaching of N-nitrosodimethylamine (NDMA) precursors, bromide ion (Br-), and selected elements (K, Ca, Mg, Mn, Fe, S, Na, B, and Al) were not significantly affected by prescribed fires. Finally, coagulant (i.e., alum and ferric) dose requirements and coagulation efficiencies were similar (i.e., removal of DOC, precursors of THMs and HAAs were 52-56%, 69-70%, 78-79%, respectively) in unburned and pre-burned leachate samples.

Low water treatability efficiency of wildfire-induced dissolved organic matter and disinfection by-product precursors.

Wildfire could alter both the quantity and composition of terrestrial organic matter exported into source water, and water treatability of fire-impacted dissolved organic matter (DOM) could be different from its unburned counterpart. Currently, there is no standard protocol to treat wildfire-impacted source water. To identify the best treatment practices in handling post-fire runoffs, we conducted a systematic controlled study using leachates of unburned white fir (Abies concolor) and Ponderosa pine (Pinus ponderosa) and black and white ashes (collected immediately and one year after the 2013 Rim Fire, California) to evaluate coagulation and oxidation strategies for controlling disinfection byproducts (DBPs) formation. Results showed that the efficiency (%) of alum coagulation in removing dissolved organic carbon and nitrogen followed the order of litter > ash immediately after the fire > ash one year after the fire. Alum coagulation was less effectiveness in removing DOM and DBP precursors in ash leachates, compared to litter leachates. This may be attributed to the loss of side chains and the decrease of DOM molecular weight during the wildfire, thus inducing lower removal efficiency of the DOM and DBP precursors during the alum coagulation. Considering use of brominated flame retardants by firefighters, the addition of bromide (Br-) (100 µg/L) greatly increased the formation of haloacetonitriles by chlorine, and this increase was relatively lower in ash leachates. The influence of reaction time and pH on DOM reactivity was similar among the leachates of litter and ash samples. Our results show that alum coagulation followed by chloramination at alkaline pH is an effective strategy for reducing post-fire DBP formation in drinking water.

Two years of post-wildfire impacts on dissolved organic matter, nitrogen, and precursors of disinfection by-products in California stream waters.

We investigated the effects of two California wildfires (Rocky and Wragg Fires, 2015) compared to an unburned reference watershed on water quality, dissolved organic matter (DOM), and precursors of disinfection by-products (DBPs) for two years' post-fire. The two burned watersheds both experienced wildfires but differed in the proportion of burned watershed areas. Burned watersheds showed rapid water quality degradation from elevated levels of turbidity, color, and suspended solids, with greater degradation in the more extensively burned watershed. During the first year's initial flushes, concentrations of dissolved organic carbon (DOC), dissolved organic nitrogen (DON), ammonium (NH4+/NH3), and specific ultraviolet absorbance (SUVA254) were significantly higher (67 ± 40%, 418 ± 125%, 192 ± 120%, and 31 ± 17%, respectively) in the more extensively burned watershed compared to the reference watershed. These elevated values gradually declined and finally returned to levels like the reference watershed in the second year. Nitrate concentrations were near detection limits (0.01 mg-N/L) in the first year but showed a large increase in fire-impacted streams during the second rainy season, possibly due to delayed nitrification. Changes in DOM composition, especially during the initial storm events, indicated that fires can attenuate humic-like and soluble microbial by-product-like (SMP) DOM while increasing the proportion of fulvic-like, tryptophan-like, and tyrosine-like compounds. Elevated bromide (Br-) concentrations (up to 8.7 µM]) caused a shift in speciation of trihalomethanes (THMs) and haloacetic acids (HAAs) to brominated species for extended periods (up to 2 years). Wildfire also resulted in elevated concentrations of N-nitrosodimethylamine (NDMA) precursors. Such changes in THM, HAA, and NDMA precursors following wildfires pose a potential treatability challenge for drinking water treatment, but the effects are relatively short-term (≤1 year).

Hurricane resulted in releasing more nitrogenous than carbonaceous disinfection byproduct precursors in coastal watersheds.

The frequency of Atlantic hurricanes has been predicted to increase significantly by the end of this century. Watershed disturbance initiated by hurricanes can alter dissolved organic matter (DOM) quantity and quality in source water dramatically. DOM is an important disinfection by-product (DBP) precursor, and thus hurricanes can have a significant impact on water treatability and drinking water safety. The interactions between land use and land cover (LULC) of a watershed and DBP formation potential (FP) in source water under hurricane events have rarely been evaluated. Here, we quantified the FPs of two carbonaceous (trihalomethanes [THMs] and haloacetic acids [HAA]) and two nitrogenous (haloacetonitrile [HAN] and N-nitrosodimethylamine [NDMA]) DBPs at eighteen sub-watersheds with varying LULC along the Yadkin-Pee Dee River basin across North and South Carolina during and after the flooding condition caused by the 2016 Hurricane Matthew. Using chlorine as a disinfectant, THM FP was 238% (±117%) higher (p < .001) under the flooding condition than baseflow condition, while HAA FP did not change significantly as a result of the flooding. DOM composition under the flooding condition changed in favor of the formation of THMs rather than HAAs by a decrease of fulvic acid-like compounds and an increase in DOM aromaticity (SUVA). The FPs of studied DBPs under the flooding condition compared with the baseflow, followed the order of HAN (356.5%) > NDMA (246.4%) > THM (115.2%) using chloramine as a disinfectant. Higher HAN FP and NDMA FP compared to THM FP suggested that more nitrogenous than carbonaceous DBPs precursors were released during this hurricane event. LULC analysis revealed that forested wetlands were the major contributor of THM, HAA, and HAN precursors, whereas NDMA precursor was derived from developed areas. This unique study highlights the dynamic interplay between LULC and exports of carbonaceous and nitrogenous DBPs precursors during and after hurricanes.

Control wildfire-induced Microcystis aeruginosa blooms by copper sulfate: Trade-offs between reducing algal organic matter and promoting disinfection byproduct formation.

Elevated levels of nutrients due to wildfire ash input into stream waters will likely cause algal blooms. When source water is impeded by algae and requires immediate restoration, copper algaecides are usually applied. Previous studies indicate that Cu2+ can promote reactivity of dissolved organic matter in forming disinfection byproducts (DBPs). However, it is unclear that how DBP formation is changed after the treatment of post-fire algal bloom by copper algaecide. In this study Microcystis aeruginosa was cultured in the medium containing black and white ash water extracts (BE and WE) to study DBP concentrations before and after 4-days exposures to low and high copper sulfate (0.5 and 1.0 mg-Cu/L). Dissolved organic matter (DOM) was characterized by UV-VIS absorption and fluorescence spectroscopy and chlorination/chloramination-based DBP formation potential (FP) experiments. DOM concentrations and algal population in the treatments were lower than that in control, regardless of types of water extract. N-nitrosodimethylamine FP in the treatments were 4-6 times higher than the control (0.23-0.34 vs. 0.05-0.06 µg/L), while haloacetonitrile FP revealed no significant difference (132-191 vs. 167-185 µg/L). Trade-offs between reducing algal population and promoting DBP-FP were more pronounced for the solutions containing BE than WE. Low copper concentration was as effective as high concentration in inhibiting algal growth while minimizing promotion of DBP formation. The results can serve to support risk evaluations of algal population and DBP concentration when wildfire-induced algal bloom is left untreated and treated by copper algaecides.

Removal of wastewater and polymer derived N-nitrosodimethylamine precursors with integrated use of chlorine and chlorine dioxide.

In this study, the effects of five different pre-oxidation scenarios (i.e., individual, simultaneous, and sequential applications of chlorine dioxide [ClO2] and chlorine [Cl2]) on the removal of N-nitrosodimethylamine (NDMA) formation potential (FP) from different water matrices (i.e., non-impacted natural waters, wastewater [WW]-impacted, and polymer-impacted waters) with subsequent chloramination were investigated. Practically relevant doses of ClO2 and Cl2 were applied for all scenarios to avoid the formation of disinfection by-products (DBPs) at regulatory levels. The removal efficiency of NDMA FP for all the oxidation scenarios (individual or simultaneous) was <20% in non-impacted natural water samples. In 20% WW-impacted waters, pre-oxidation with ClO2 at pH 7.8 resulted in a significant reduction in NDMA FP (56-73%), whereas pre-oxidation with Cl2 showed less removals (40-50%). For the integrated oxidation scenarios (i.e., simultaneous or sequential application), NDMA FP removals further increased (20-45%), especially, at pH 6.0 compared to individual application of oxidants in WW-impacted waters. The formation of NDMA in pre-oxidized water samples also decreased significantly under uniform formation condition (UFC). In polymer-impacted waters, integrated applications of Cl2 and ClO2 significantly improved the deactivation of polymer-derived NDMA precursors independent of oxidation time (10 vs. 60 min) and pH (6.0 vs. 7.8) compared to individual application of these oxidants. In addition, chlorite (ClO2-) formation was low and maintained well below 1 mg/L for integrated applications of Cl2 and ClO2, while chlorate (ClO3-) formation increased significantly as compared to application of ClO2 only.

Deactivation of wastewater-derived N-nitrosodimethylamine precursors with chlorine dioxide oxidation and the effect of pH.

In this study, the effect of chlorine dioxide (ClO2) oxidation on the deactivation of wastewater (WW)-derived N-nitrosodimethylamine (NDMA) precursors was investigated under various conditions (i.e., ClO2 application pH, dose and contact time). At pH 6.0, decreases in NDMA formation potentials (FPs) or occurrences (under uniform formation conditions [UFC]) were relatively low (<25%) with ClO2 oxidation regardless of WW-impact. A negative removal was also observed after ClO2 oxidation in some of the non-impacted waters. However, NDMA FP removals were significant (up to ~85%) under the same oxidation conditions in WW-impacted waters at pH 7.8. This indicates that the majority of WW-derived NDMA precursors can be deactivated with ClO2 oxidation above neutral pH. This was attributed to the better oxidative reaction of ClO2 with amines that have lone pair electrons to be shared at higher oxidation pH conditions. In addition, relatively short oxidation periods with ClO2 (i.e., ≤10 min) or low Ct (concentration × time, ~10 mg ∗ min/L) values were sufficient for the deactivation of WW-derived NDMA precursors. ClO2 oxidation was effective in freshly WW-impacted waters. Natural attenuation processes (e.g., sorption, biodegradation, etc.) can change the reactivity of WW-derived NDMA precursors for oxidation with ClO2. The effect of ClO2 on the removal of THM precursors was low (<25%) and independent of oxidation conditions. Given the low formation of regulated THMs and HAAs, ClO2 oxidation presents a viable option for the simultaneous control of NDMA and regulated DBP formation during water treatment, especially for utilities treating WW-impacted water sources.

Dynamic Changes of Disinfection Byproduct Precursors following Exposures of Microcystis aeruginosa to Wildfire Ash Solutions.

Wildfires can elevate dissolved organic matter (DOM) levels due to ash input and algal growth in source waters, and consequently impacting disinfection byproduct (DBP) formation in finished water; however, it remains unclear how quality and quantity of overall allochthonous and autochthonous DOM as well as associated DBP formation are changed during an entire algal life cycle. Microcystis aeruginosa was cultured in the medium containing low and high concentrations [10% and 65% (v/v)] of black and white ash water extracts (BE and WE) to study dynamic changes of carbonaceous, nitrogenous, and oxygenated DBP precursors during algal growth. DOM was characterized by absorption and fluorescence spectroscopy and chlorination/chloramination-based DBP formation experiments. Throughout the entire experiment, C-DBP precursors in the control ranged from 2.41 to 3.09 mmol/mol-C. In the treatment with 10% BE, the amount of C-DBP precursors decreased from 6.8 to 3.0 mmol/mol-C at initial-exponential phase then increased to 4.2 mmol/mol-C at death phase. The same trend was observed for O-DBP precursors. However, these dynamic changes of C- and O-DBP precursors exhibited opposite patterns in 65% extracts. Similar patterns were also observed in the WE treatments. On the other hand, N-DBP precursors continuously declined in all treatments. These results indicate that postfire ash loading and algal bloom stage may significantly affect DBP formation in source water.

Removal of N-nitrosodimethylamine precursors with powdered activated carbon adsorption.

The main objective of this study was to examine the roles of powdered activated carbon (PAC) characteristics (i.e., surface chemistry, pore size distribution, and surface area) in the removal of N-nitrosodimethylamine (NDMA) formation potential (FP) in surface and wastewater-impacted waters. Also, the effects of natural attenuation of NDMA precursors in surface waters, NDMA FP concentration, and carbon dose on the removal of NDMA FP by PAC were evaluated. Finally, the removal of NDMA FP by PAC at two full-scale DWTPs was monitored. Wastewater-impacted and surface water samples were collected to conduct adsorption experiments using different PACs and activated carbon fibers (ACFs) with a wide range of physicochemical characteristics. The removal efficiency of NDMA FP by PAC was significantly higher in wastewater-impacted than surface waters. Adsorbable NDMA precursors showed a size distribution in the waters tested; the adsorbable fraction included precursors accessing the pore size regions of 10-20 Å and <10 Å. Basic carbons showed higher removal of NDMA FP than acidic carbons on a surface area basis. The overall removal of NDMA FP by PAC on a mass basis depended on the surface area, pore size distribution and pHPZC. Thus, PACs with hybrid characteristics (micro and mesoporous), higher surface areas, and basic surface chemistry are more likely to be effective for NDMA precursor control by PAC adsorption. The application of PAC in DWTPs for taste and odor control resulted in an additional 20% removal of NDMA FP for the PAC doses of 7-10 mg/L. The natural attenuation of NDMA precursors through a combination of processes (biodegradation, photolysis and adsorption) decreased their adsorbability and removal by PAC adsorption.

Seasonal and temporal patterns of NDMA formation potentials in surface waters.

The seasonal and temporal patterns of N-nitrosodimethylamine (NDMA) formation potentials (FPs) were examined with water samples collected monthly for 21 month period in 12 surface waters. This long term study allowed monitoring the patterns of NDMA FPs under dynamic weather conditions (e.g., rainy and dry periods) covering several seasons. Anthropogenically impacted waters which were determined by high sucralose levels (>100 ng/L) had higher NDMA FPs than limited impacted sources (<100 ng/L). In most sources, NDMA FP showed more variability in spring months, while seasonal mean values remained relatively consistent. The study also showed that watershed characteristics played an important role in the seasonal and temporal patterns. In the two dam-controlled river systems (SW A and G), the NDMA FP levels at the downstream sampling locations were controlled by the NDMA levels in the dams independent of either the increases in discharge rates due to water releases from the dams prior to or during the heavy rain events or intermittent high NDMA FP levels observed at the upstream of dams. The large reservoirs and impoundments on rivers examined in this study appeared serving as an equalization basin for NDMA precursors. On the other hand, in a river without an upstream reservoir (SW E), the NDMA levels were influenced by the ratio of an upstream wastewater treatment plant (WWTP) effluent discharge to the river discharge rate. The impact of WWTP effluent decreased during the high river flow periods due to rain events. Linear regression with independent variables DOC, DON, and sucralose yielded poor correlations with NDMA FP (R(2) < 0.27). Multiple linear regression analysis using DOC and log [sucralose] yielded a better correlation with NDMA FP (R(2) = 0.53).

Assessing trihalomethanes (THMs) and N-nitrosodimethylamine (NDMA) formation potentials in drinking water treatment plants using fluorescence spectroscopy and parallel factor analysis.

The formation of disinfection byproducts (DBPs) is a major challenge in drinking water treatments. This study explored the applicability of fluorescence excitation-emission matrices and parallel factor analysis (EEM-PARAFAC) for assessing the formation potentials (FPs) of trihalomethanes (THMs) and N-nitrosodimethylamine (NDMA), and the treatability of THM and NDMA precursors in nine drinking water treatment plants. Two humic-like and one tryptophan-like components were identified for the samples using PARAFAC. The total THM FP (TTHM FP) correlated strongly with humic-like component C2 (r=0.874), while NDMA FP showed a moderate and significant correlation with the tryptophan-like component C3 (r=0.628). The reduction by conventional treatment was more effective for C2 than C3, and for TTHM FP than NDMA FP. The treatability of DOM and TTHM FP correlated negatively with the absorption spectral slope (S275-295) and biological index (BIX) of the raw water, but it correlated positively with humification index (HIX). Our results demonstrated that PARAFAC components were valuable for assessing DBPs FP in drinking water treatments, and also that the raw water quality could affect the treatment efficiency.