Impact of chloramination on the development of laboratory-grown biofilms fed with filter-pretreated groundwater

A review of research advances on disinfection strategies for biofilm control in drinking water distribution systems

The presence of biofilms in drinking water distribution systems (DWDS) is responsible for water quality deterioration and a possible source of public health risks. Different factors impact the biological stability of drinking water (DW) in the distribution networks, such as the presence and concentration of nutrients, water temperature, pipe material composition, hydrodynamic conditions, and levels of disinfectant residual. This review aimed to evaluate the current state of knowledge on strategies for DW biofilm disinfection through a qualitative and quantitative analysis of the literature published over the last decade. A systematic review method was performed on the 562 journal articles identified through database searching on Web of Science and Scopus, with 85 studies selected for detailed analysis. A variety of disinfectants were identified for DW biofilm control such as chlorine, chloramine, UV irradiation, hydrogen peroxide, chlorine dioxide, ozone, and others at a lower frequency, namely, electrolyzed water, bacteriophages, silver ions, and nanoparticles. The disinfectants can impact the microbial communities within biofilms, reduce the number of culturable cells and biofilm biomass, as well as interfere with the biofilm matrix components. The maintenance of an effective residual concentration in the water guarantees long-term prevention of biofilm formation and improves the inactivation of detached biofilm-associated opportunistic pathogens. Additionally, strategies based on multi-barrier processes by optimization of primary and secondary disinfection combined with other water treatment methods improve the control of opportunistic pathogens, reduce the chlorine-tolerance of biofilm-embedded cells, as well as decrease the corrosion rate in metal-based pipelines. Most of the studies used benchtop laboratory devices for biofilm research. Even though these devices mimic the conditions found in real DWDS, future investigations on strategies for DW biofilm control should include the validity of the promising strategies against biofilms formed in real DW networks.

The Impact of Pipe Material on the Diversity of Microbial Communities in Drinking Water Distribution Systems

As many cities around the world face the prospect of replacing aging drinking water distribution systems (DWDS), water utilities must make careful decisions on new pipe material (e.g., cement-lined or PVC) for these systems. These decisions are informed by cost, physical integrity, and impact on microbiological and physicochemical water quality. Indeed, pipe material can impact the development of biofilm in DWDS that can harbor pathogens and impact drinking water quality. Annular reactors (ARs) with cast iron and cement coupons fed with chloraminated water from a municipal DWDS were used to investigate the impact of pipe material on biofilm development and composition over 16 months. The ARs were plumbed as closely as possible to the water main in the basement of an academic building to simulate distribution system conditions. Biofilm communities on coupons were characterized using 16S rRNA sequencing. In the cast iron reactors, β-proteobacteria, Actinobacteria, and α-proteobacteria were similarly relatively abundant (24.1, 22.5, and 22.4%, respectively) while in the cement reactors, α-proteobacteria and Actinobacteria were more relatively abundant (36.3 and 35.2%, respectively) compared to β-proteobacteria (12.8%). Mean alpha diversity (estimated with Shannon H and Faith's Phylogenetic Difference indices) was greater in cast iron reactors (Shannon: 5.00 ± 0.41; Faith's PD: 15.40 ± 2.88) than in cement reactors (Shannon: 4.16 ± 0.78; Faith's PD: 13.00 ± 2.01). PCoA of Bray-Curtis dissimilarities indicated that communities in cast iron ARs, cement ARs, bulk distribution system water, and distribution system pipe biofilm were distinct. The mean relative abundance of Mycobacterium spp. was greater in the cement reactors (34.8 ± 18.6%) than in the cast iron reactors (21.7 ± 11.9%). In contrast, the mean relative abundance of Legionella spp. trended higher in biofilm from cast iron reactors (0.5 ± 0.7%) than biofilm in cement reactors (0.01 ± 0.01%). These results suggest that pipe material is associated with differences in the diversity, bacterial composition, and opportunistic pathogen prevalence in biofilm of DWDS.

Influence of pipe material on biofilm microbial communities found in drinking water supply system

The biofilms and water samples from a model installation built of PVC-U, PE-HD and cast iron pipes were investigated using standard heterotrophic plate count and 16S rRNA Next Generation Sequencing. The results of the high throughput identification imply that the construction material strongly influences the microbiome composition. PVC-U and PE-HD pipes were dominated with Proteobacteria (54-60%) while the cast pipe was overgrown by Nitrospirae (64%). It was deduced that the plastic pipes create a more convenient environment for the potentially pathogenic taxa than the cast iron. The 7-year old biofilms were described as complex habitats with sharp oxidation-reduction gradients, where co-existence of methanogenic and methanotrophic microbiota takes place. Furthermore, it was found that the drinking water distribution systems (DWDS) are a useful tool for studying the ecology of rare bacterial phyla. New ecophysiological aspects were described for Aquihabitans, Thermogutta and Vampirovibrio. The discrepancy between identity of HPC-derived bacteria and NGS-revealed composition of biofilm and water microbiomes point to the need of introducing new diagnostical protocols to enable proper assessment of the drinking water safety, especially in DWDSs operating without disinfection.

Evaluating method and potential risks of chlorine-resistant bacteria (CRB): A review

Chlorine-resistant bacteria (CRB) are commonly defined as bacteria with high resistance to chlorine disinfection or bacteria which can survive or even regrow in the residual chlorine. Chlorine disinfection cannot completely control the risks of CRB, such as risks of pathogenicity, antibiotic resistance and microbial growth. Currently, researchers pay more attention to CRB with pathogenicity or antibiotic resistance. The microbial growth risks of non-pathogenic CRB in water treatment and reclamation systems have been neglected to some extent. In this review, these three kinds of risks are all analyzed, and the last one is also highlighted. In order to study CRB, various methods are used to evaluate chlorine resistance. This review summarizes the evaluating methods for chlorine resistance reported in the literatures, and collects the important information about the typical isolated CRB strains including their genera, sources and levels of chlorine resistance. To our knowledge, few review papers have provided such systematic information about CRB. Among 44 typical CRB strains from 17 genera isolated by researchers, Mycobacterium, Bacillus, Legionella, Pseudomonas and Sphingomonas were the five genera with the highest frequency of occurrence in literatures. They are all pathogenic or opportunistic pathogenic bacteria. In addition, although there are many studies on CRB, information about chlorine resistance level is still limited to specie level or strain level. The difference in chlorine resistance level among different bacterial genera is less well understood. An inconvenient truth is that there is still no widely-accepted method to evaluate chlorine resistance and to identify CRB. Due to the lack of a unified method, it is difficult to compare the results about chlorine resistance level of bacterial strains in different literatures. A recommended evaluating method using logarithmic removal rate as an index and E. coli as a reference strain is proposed in this review based on the summary of the current evaluating methods. This method can provide common range of chlorine resistance of each genus and it is conducive to analyzing the distribution and abundance of CRB in the environment.

Selective antibiotic resistance genes in multiphase samples during biofilm growth in a simulated drinking water distribution system: Occurrence, correlation and low-pressure ultraviolet removal

The aim of this study was to gain comprehensive insights into the characteristics of antibiotic resistance genes (ARGs) in multiphase samples from drinking water distribution pipelines using a simulated biofilm reactor. During 120 d of continuous operation, common parameters and six ARGs (ermA, ermB, aphA2, ampC, sulII, and tetO) in samples of three phases (water, particle, and biofilm) from the reactor were investigated, which demonstrated secondary contamination by ARGs. Abundances of the six ARGs in the reactor effluent increased gradually, and in the 120 d effluent, the relative abundances of aphA2 and sulII were the highest, at 9.9 × 10^-4 and 1.3 × 10^-3, respectively, with a 1.5-fold and 2.8-fold increase, compared with those in the influent. The relative abundances of the six ARGs in the biofilm phase increased significantly (P < 0.05) at 120 d, which was caused by robust bacteria in biofilm that was newly exposed following the detachment of a large piece of aging biofilm. In the particle phase, four of the ARGs did not change significantly during the 120 d period. The six ARGs in the samples of three phases showed a negative correlation with residual chlorine in the pipe water, which demonstrated that low abundance of ARGs in the samples of three phases was related to the improvement of residual chlorine. The proportion of cultivable bacteria illustrated that the robust and active bacteria were negatively correlated with the six ARGs in the biofilm. Total organic carbon (TOC) in the pipeline showed a positive correlation with the proportion of cultivable bacteria in both the water and biofilm phases, which indicated that a TOC reduction in the pipeline contributed to low abundance of ARGs. With low-pressure ultraviolet (LP-UV) irradiation of 20 mJ/cm², ARGs in the samples of three phases were efficiently controlled, which showed that LP-UV can be used for ARG removal in terminal water for supplemental bactericidal treatment of pipeline effluent.

Effects of chlorination/chlorine dioxide disinfection on biofilm bacterial community and corrosion process in a reclaimed water distribution system

In this work, reclaimed water treated with sodium hypochlorite (NaClO) or chlorine dioxide (ClO₂) at 1, 2, and 4 mg/L was operated successively for 30 days respectively, in annular reactors with new cast iron coupons, corresponding to stages I (days 0-30), II (days 31-60), and III (days 61-90). The Illumina HiSeq 2500 sequencing platform was used to analyze the bacterial community composition, scanning electron microscopy and X-ray diffraction analyses were conducted to characterize corrosion scales, and the weight loss method was served to determine the general corrosion rate. Results reveal the precise disinfection effect on biofilm bacteria to be dose dependent and species specific. In stage I, disinfection caused a reduction in the number of operational taxonomic units, but, had little effect on biofilm composition. In stage II, NaClO and ClO₂ induced a reduction of Proteobacteria proportion, but increased the dominance of Firmicutes; the diminished Proteobacteria in NaClO test mainly included Gammaproteobacteria, while, that in ClO₂ test mainly included the Gammaproteobacteria and Betaproteobacteria. In stage III, Firmicutes presented a certain resistance to NaClO and ClO₂ as the accumulation of corrosion scales. Results also indicated that disinfection enhanced the corrosion process, and the promoting effect of ClO₂ was more pronounced than that of NaClO. Moreover, this promoting effect was more obvious in stage I than that in the latter two stages. The strong oxidization effect associated with disinfection in stage I was the dominant factor promoting corrosion, whereas, the bacterial community also played a crucial role in stages II and III.

Understanding, Monitoring, and Controlling Biofilm Growth in Drinking Water Distribution Systems

In drinking water distribution systems (DWDS), biofilms are the predominant mode of microbial growth, with the presence of extracellular polymeric substance (EPS) protecting the biomass from environmental and shear stresses. Biofilm formation poses a significant problem to the drinking water industry as a potential source of bacterial contamination, including pathogens, and, in many cases, also affecting the taste and odor of drinking water and promoting the corrosion of pipes. This article critically reviews important research findings on biofilm growth in DWDS, examining the factors affecting their formation and characteristics as well as the various technologies to characterize and monitor and, ultimately, to control their growth. Research indicates that temperature fluctuations potentially affect not only the initial bacteria-to-surface attachment but also the growth rates of biofilms. For the latter, the effect is unique for each type of biofilm-forming bacteria; ammonia-oxidizing bacteria, for example, grow more-developed biofilms at a typical summer temperature of 22 °C compared to 12 °C in fall, and the opposite occurs for the pathogenic Vibrio cholerae. Recent investigations have found the formation of thinner yet denser biofilms under high and turbulent flow regimes of drinking water, in comparison to the more porous and loosely attached biofilms at low flow rates. Furthermore, in addition to the rather well-known tendency of significant biofilm growth on corrosion-prone metal pipes, research efforts also found leaching of growth-promoting organic compounds from the increasingly popular use of polymer-based pipes. Knowledge of the unique microbial members of drinking water biofilms and, importantly, the influence of water characteristics and operational conditions on their growth can be applied to optimize various operational parameters to minimize biofilm accumulation. More-detailed characterizations of the biofilm population size and structure are now feasible with fluorescence microscopy (epifluorescence and CLSM imaging with DNA, RNA, EPS, and protein and lipid stains) and electron microscopy imaging (ESEM). Importantly, thorough identification of microbial fingerprints in drinking water biofilms is achievable with DNA sequencing techniques (the 16S rRNA gene-based identification), which have revealed a prevalence of previously undetected bacterial members. Technologies are now moving toward in situ monitoring of biomass growth in distribution networks, including the development of optical fibers capable of differentiating biomass from chemical deposits. Taken together, management of biofilm growth in water distribution systems requires an integrated approach, starting from the treatment of water prior to entering the networks to the potential implementation of "biofilm-limiting" operational conditions and, finally, ending with the careful selection of available technologies for biofilm monitoring and control. For the latter, conventional practices, including chlorine-chloramine disinfection, flushing of DWDS, nutrient removal, and emerging technologies are discussed with their associated challenges.

Characterization of a Drinking Water Distribution Pipeline Terminally Colonized by Naegleria fowleri

Free-living amoebae, such as Naegleria fowleri, Acanthamoeba spp., and Vermamoeba spp., have been identified as organisms of concern due to their role as hosts for pathogenic bacteria and as agents of human disease. In particular, N. fowleri is known to cause the disease primary amoebic meningoencephalitis (PAM) and can be found in drinking water systems in many countries. Understanding the temporal dynamics in relation to environmental and biological factors is vital for developing management tools for mitigating the risks of PAM. Characterizing drinking water systems in Western Australia with a combination of physical, chemical and biological measurements over the course of a year showed a close association of N. fowleri with free chlorine and distance from treatment over the course of a year. This information can be used to help design optimal management strategies for the control of N. fowleri in drinking-water-distribution systems.

Response of Simulated Drinking Water Biofilm Mechanical and Structural Properties to Long-Term Disinfectant Exposure

Mechanical and structural properties of biofilms influence the accumulation and release of pathogens in drinking water distribution systems (DWDS). Thus, understanding how long-term residual disinfectants exposure affects biofilm mechanical and structural properties is a necessary aspect for pathogen risk assessment and control. In this study, elastic modulus and structure of groundwater biofilms was monitored by atomic force microscopy (AFM) and optical coherence tomography (OCT) during three months of exposure to monochloramine or free chlorine. After the first month of disinfectant exposure, the mean stiffness of monochloramine- or free-chlorine-treated biofilms was 4 to 9 times higher than those before treatment. Meanwhile, the biofilm thickness decreased from 120 ± 8 μm to 93 ± 6-107 ± 11 μm. The increased surface stiffness and decreased biofilm thickness within the first month of disinfectant exposure was presumably due to the consumption of biomass. However, by the second to third month during disinfectant exposure, the biofilm mean stiffness showed a 2- to 4-fold decrease, and the biofilm thickness increased to 110 ± 7-129 ± 8 μm, suggesting that the biofilms adapted to disinfectant exposure. After three months of the disinfectant exposure process, the disinfected biofilms showed 2-5 times higher mean stiffness (as determined by AFM) and 6-13-fold higher ratios of protein over polysaccharide, as determined by differential staining and confocal laser scanning microscopy (CLSM), than the nondisinfected groundwater biofilms. However, the disinfected biofilms and nondisinfected biofilms showed statistically similar thicknesses (t test, p > 0.05), suggesting that long-term disinfection may not significantly remove net biomass. This study showed how biofilm mechanical and structural properties vary in response to a complex DWDS environment, which will contribute to further research on the risk assessment and control of biofilm-associated-pathogens in DWDS.

Core-satellite populations and seasonality of water meter biofilms in a metropolitan drinking water distribution system

Drinking water distribution systems (DWDSs) harbor the microorganisms in biofilms and suspended communities, yet the diversity and spatiotemporal distribution have been studied mainly in the suspended communities. This study examined the diversity of biofilms in an urban DWDS, its relationship with suspended communities and its dynamics. The studied DWDS in Urbana, Illinois received conventionally treated and disinfected water sourced from the groundwater. Over a 2-year span, biomass were sampled from household water meters (n=213) and tap water (n=20) to represent biofilm and suspended communities, respectively. A positive correlation between operational taxonomic unit (OTU) abundance and occupancy was observed. Examined under a 'core-satellite' model, the biofilm community comprised 31 core populations that encompassed 76.7% of total 16 S rRNA gene pyrosequences. The biofilm communities shared with the suspended community highly abundant and prevalent OTUs, which related to methano-/methylotrophs (i.e., Methylophilaceae and Methylococcaceae) and aerobic heterotrophs (Sphingomonadaceae and Comamonadaceae), yet differed by specific core populations and lower diversity and evenness. Multivariate tests indicated seasonality as the main contributor to community structure variation. This pattern was resilient to annual change and correlated to the cyclic fluctuations of core populations. The findings of a distinctive biofilm community assemblage and methano-/methyltrophic primary production provide critical insights for developing more targeted water quality monitoring programs and treatment strategies for groundwater-sourced drinking water systems.

Role of biofilm roughness and hydrodynamic conditions in Legionella pneumophila adhesion to and detachment from simulated drinking water biofilms

Biofilms in drinking water distribution systems (DWDS) could exacerbate the persistence and associated risks of pathogenic Legionella pneumophila (L. pneumophila), thus raising human health concerns. However, mechanisms controlling adhesion and subsequent detachment of L. pneumophila associated with biofilms remain unclear. We determined the connection between L. pneumophila adhesion and subsequent detachment with biofilm physical structure characterization using optical coherence tomography (OCT) imaging technique. Analysis of the OCT images of multispecies biofilms grown under low nutrient condition up to 34 weeks revealed the lack of biofilm deformation even when these biofilms were exposed to flow velocity of 0.7 m/s, typical flow for DWDS. L. pneumophila adhesion on these biofilm under low flow velocity (0.007 m/s) positively correlated with biofilm roughness due to enlarged biofilm surface area and local flow conditions created by roughness asperities. The preadhered L. pneumophila on selected rough and smooth biofilms were found to detach when these biofilms were subjected to higher flow velocity. At the flow velocity of 0.1 and 0.3 m/s, the ratio of detached cell from the smooth biofilm surface was from 1.3 to 1.4 times higher than that from the rough biofilm surface, presumably because of the low shear stress zones near roughness asperities. This study determined that physical structure and local hydrodynamics control L. pneumophila adhesion to and detachment from simulated drinking water biofilm, thus it is the first step toward reducing the risk of L. pneumophila exposure and subsequent infections.

Bacterial community analysis of drinking water biofilms in southern Sweden

Next-generation sequencing of the V1-V2 and V3 variable regions of the 16S rRNA gene generated a total of 674,116 reads that described six distinct bacterial biofilm communities from both water meters and pipes. A high degree of reproducibility was demonstrated for the experimental and analytical work-flow by analyzing the communities present in parallel water meters, the rare occurrence of biological replicates within a working drinking water distribution system. The communities observed in water meters from households that did not complain about their drinking water were defined by sequences representing Proteobacteria (82-87%), with 22-40% of all sequences being classified as Sphingomonadaceae. However, a water meter biofilm community from a household with consumer reports of red water and flowing water containing elevated levels of iron and manganese had fewer sequences representing Proteobacteria (44%); only 0.6% of all sequences were classified as Sphingomonadaceae; and, in contrast to the other water meter communities, markedly more sequences represented Nitrospira and Pedomicrobium. The biofilm communities in pipes were distinct from those in water meters, and contained sequences that were identified as Mycobacterium, Nocardia, Desulfovibrio, and Sulfuricurvum. The approach employed in the present study resolved the bacterial diversity present in these biofilm communities as well as the differences that occurred in biofilms within a single distribution system, and suggests that next-generation sequencing of 16S rRNA amplicons can show changes in bacterial biofilm communities associated with different water qualities.

Biogas production using anaerobic groundwater containing a subterranean microbial community associated with the accretionary prism

In a deep aquifer associated with an accretionary prism, significant methane (CH₄) is produced by a subterranean microbial community. Here, we developed bioreactors for producing CH₄ and hydrogen (H₂) using anaerobic groundwater collected from the deep aquifer. To generate CH₄, the anaerobic groundwater amended with organic substrates was incubated in the bioreactor. At first, H₂ was detected and accumulated in the gas phase of the bioreactor. After the H₂ decreased, rapid CH₄ production was observed. Phylogenetic analysis targeting 16S rRNA genes revealed that the H₂-producing fermentative bacterium and hydrogenotrophic methanogen were predominant in the reactor. The results suggested that syntrophic biodegradation of organic substrates by the H₂-producing fermentative bacterium and the hydrogenotrophic methanogen contributed to the CH₄ production. For H₂ production, the anaerobic groundwater, amended with organic substrates and an inhibitor of methanogens (2-bromoethanesulfonate), was incubated in a bioreactor. After incubation for 24 h, H₂ was detected from the gas phase of the bioreactor and accumulated. Bacterial 16S rRNA gene analysis suggested the dominance of the H₂-producing fermentative bacterium in the reactor. Our study demonstrated a simple and rapid CH₄ and H₂ production utilizing anaerobic groundwater containing an active subterranean microbial community.

An overview on the reactors to study drinking water biofilms

The development of biofilms in drinking water distribution systems (DWDS) can cause pipe degradation, changes in the water organoleptic properties but the main problem is related to the public health. Biofilms are the main responsible for the microbial presence in drinking water (DW) and can be reservoirs for pathogens. Therefore, the understanding of the mechanisms underlying biofilm formation and behavior is of utmost importance in order to create effective control strategies. As the study of biofilms in real DWDS is difficult, several devices have been developed. These devices allow biofilm formation under controlled conditions of physical (flow velocity, shear stress, temperature, type of pipe material, etc), chemical (type and amount of nutrients, type of disinfectant and residuals, organic and inorganic particles, ions, etc) and biological (composition of microbial community - type of microorganism and characteristics) parameters, ensuring that the operational conditions are similar as possible to the DWDS conditions in order to achieve results that can be applied to the real scenarios. The devices used in DW biofilm studies can be divided essentially in two groups, those usually applied in situ and the bench top laboratorial reactors. The selection of a device should be obviously in accordance with the aim of the study and its advantages and limitations should be evaluated to obtain reproducible results that can be transposed into the reality of the DWDS. The aim of this review is to provide an overview on the main reactors used in DW biofilm studies, describing their characteristics and applications, taking into account their main advantages and limitations.

MS2 bacteriophage reduction and microbial communities in biosand filters

This study evaluated the role of physical and biological filter characteristics on the reduction of MS2 bacteriophage in biosand filters (BSFs). Three full-scale concrete Version 10 BSFs, each with a 55 cm sand media depth and a 12 L charge volume, reached 4 log10 reduction of MS2 within 43 days of operation. A consistently high reduction of MS2 between 4 log10 and 7 log10 was demonstrated for up to 294 days. Further examining one of the filters revealed that an average of 2.8 log10 reduction of MS2 was achieved within the first 5 cm of the filter, and cumulative virus reduction reached an average of 5.6 log10 after 240 days. Core sand samples from this filter were taken for protein, carbohydrate, and genomic extraction. Higher reduction of MS2 in the top 5 cm of the sand media (0.56 log10 reduction per cm vs 0.06 log10 reduction per cm for the rest of the filter depth) coincided with greater diversity of microbial communities and increased concentrations of carbohydrates. In the upper layers, "Candidatus Nitrosopumilus maritimus" and "Ca. Nitrospira defluvii" were found as dominant populations, while significant amounts of Thiobacillus-related OTUs were detected in the lower layers. Proteolytic bacterial populations such as the classes Sphingobacteria and Clostridia were observed over the entire filter depth. Thus, this study provides the first insight into microbial community structures that may play a role in MS2 reduction in BSF ecosystems. Overall, besides media ripening and physical reduction mechanisms such as filter depth and long residence time (45 min vs 24 ± 8.5 h), the establishment of chemolithotrophs and proteolytic bacteria could greatly enhance the reduction of MS2.

Differential resistance of drinking water bacterial populations to monochloramine disinfection

The impact of monochloramine disinfection on the complex bacterial community structure in drinking water systems was investigated using culture-dependent and culture-independent methods. Changes in viable bacterial diversity were monitored using culture-independent methods that distinguish between live and dead cells based on membrane integrity, providing a highly conservative measure of viability. Samples were collected from lab-scale and full-scale drinking water filters exposed to monochloramine for a range of contact times. Culture-independent detection of live cells was based on propidium monoazide (PMA) treatment to selectively remove DNA from membrane-compromised cells. Quantitative PCR (qPCR) and pyrosequencing of 16S rRNA genes was used to quantify the DNA of live bacteria and characterize the bacterial communities, respectively. The inactivation rate determined by the culture-independent PMA-qPCR method (1.5-log removal at 664 mg·min/L) was lower than the inactivation rate measured by the culture-based methods (4-log removal at 66 mg·min/L). Moreover, drastic changes in the live bacterial community structure were detected during monochloramine disinfection using PMA-pyrosequencing, while the community structure appeared to remain stable when pyrosequencing was performed on samples that were not subject to PMA treatment. Genera that increased in relative abundance during monochloramine treatment include Legionella, Escherichia, and Geobacter in the lab-scale system and Mycobacterium, Sphingomonas, and Coxiella in the full-scale system. These results demonstrate that bacterial populations in drinking water exhibit differential resistance to monochloramine, and that the disinfection process selects for resistant bacterial populations.