Resolving Problematic Biofilms in Buildings and Compounds

Abstract

Biofilm problems persist in water distribution systems and, as a result, have been the focus of remediation strategy development. However, there have been limited reports of investigations of problematic biofilms in individual buildings and groups of buildings where biofilms may grow and affect a defined population. The authors have investigated three sites in subtropical areas—a school, a resort, and a condominium development. Each situation created different effects that were detected by the occupants and resulted in a water quality assessment. The common connection was problematic biofilm growth in the potable water distribution system resulting from site-specific incidents. In one case, the biofilm was formed as the system was not used for a long period. Another was a result of operational issues with the on-site water treatment process. The third case was caused by the water quality issues in the water delivered to the site. In each situation, solutions were sought to address the biofilm problem while keeping the buildings in service and reducing the potential for future recurrence. Applied control strategies included superchlorination and flushing, and were demonstrated to be effective. Operational strategies, including periodic pipe flushing and installation of a chlorine booster station, were recommended on a site-specific basis to prevent the return of problematic biofilms.

Introduction

Microorganisms are the most widely distributed life forms on the planet (Chapelle, 1993) and are known to inhabit and thrive in the presence of moisture and nutrients. Both moisture and nutrients exist in plentiful supplies in water distribution networks, where microorganisms can grow in the form of biofilms (Videla, 1996). As early as the 1980s, it had been firmly established that biofilms would inevitably be present in drinking water distribution systems; however, when conditions develop such that biofilms are not kept in check is when they become problematic (U.S. EPA, 1992). From the utility perspective, the undesirable accumulation of biofilms with actively growing slime layers can lead to biofouling, which in turn can degrade delivered water quality.

Numerous studies conducted since then have demonstrated that pathogens, opportunistic pathogens, and indicator organisms, including Clostridium sp., Escherichia coli, Enterobacter, Legionella, Pseudomonas, Vibrio, and Staphylococcus, among others, have been reported in biofilms collected from water distribution networks (LeChevallier et al., 1987; Emde et al., 1992; Geldreich, 1996; Sartory and Holmes, 1997; Norton and Le Chevallier, 2000; Murga et al., 2001; Lee and Kim, 2003; Hall-Stoodley and Stoodley, 2005; September et al., 2007). In addition to bacteria, other pathogenic microorganisms, including viruses, fungi, yeast, protozoa such as amoebae and ciliates, diatoms and other algae, and invertebrates, have also been isolated from biofilms (Bachmann and Edyvean, 2006 and references therein; Eboigbodin et al., 2008). Some of these biofilm organisms, if present in significant densities, can release toxins into the distribution system. If left uncontrolled, biofilms can become a considerable issue for water distribution systems, particularly in regard to hygienic, operational, and economic consequences.

In addition to health effects resulting from pathogens, biofilms can also contribute to taste, odor, and color issues, which may require operational changes at the treatment plant or in the transmission network to be resolved. Problematic biofilms may also compromise the proper enumeration of indicator organisms and weaken pipe integrity by microbially influenced corrosion, which permits the onset and acceleration of corrosion on a variety of pipe materials (Videla, 1996; Cantor et al., 2006; Zhang et al., 2010). For water distribution systems constructed with metallic materials, corrosion and corrosion control are on-going issues that, if not addressed, can result in pipe damage/failure, premature aging/replacement, clogging, and increased maintenance requirements (Dillon, 1995; Mains, 2008).

The economic significance of distribution system biofouling as a potential operational issue for potable water service utilities can be illustrated by noting the magnitude of piping currently in service in the United States and the rate at which distribution pipes are replaced and installed. Roughly 4,400 miles (7,080 km) of the estimated 880,000 miles (1.4 million km) of distribution system pipes are replaced each year (U.S. EPA, 2002a). The U.S. EPA estimates that transmission and distribution network maintenance represents about $4.2–$6.3 billion per year, including $200 million per year in pipe material costs alone (U.S. EPA, 2002b). It is also estimated that there will be nearly $80 billion in distribution piping that will be in need of replacement in the next 20 years (U.S. EPA, 2002b).

Biofilms are inevitable, and therefore they must be controlled and monitored. Many different methods have been used to control biofilms; however, in most circumstances, biofilm control requires the use of a variety of tools, and the relative effectiveness is typically site specific. In general, biofilms can be managed by removing organic matter and nutrients during water treatment (Batte et al., 2003), inactivation of microorganisms via sensible use of disinfectants and residuals (LeChevallier et al., 1990; Fass et al., 2003; Bachmann and Edyvean, 2006), and proper distribution system maintenance practices (i.e., flushing, avoiding stagnant conditions, minimizing the corrosion of iron pipe surfaces, maintaining disinfectant residuals, and managing contamination from external sources) (Walker and Morales, 1997; Crozes and Cushing, 2000; Kirmeyer et al., 2001; Eboigbodin et al., 2008). Niquette et al. (2000) demonstrated that proper pipe material selection is also important in biofilm control (Niquette et al., 2000). Polyvinyl chloride (PVC) and polyethylene had the least potential to accumulate biofilms, although polyethylene may promote the growth of Legionella more than PVC (Van der Kooij and Veenendaal, 2001). Ferrous materials were more likely to form biofilms than any other material (Bagh et al., 2002). If ferrous materials must be used, coatings (e.g., cement lining) can be used to discourage fouling and biofilm corrosion. It is difficult to make comparisons between biofilm formation on plastics and other materials since the biofilms slough off the smoother surfaces more readily but they still start to form readily.

This study presents the results of three case studies addressing problematic biofilms in three different locations. At these sites, taste, odor, and discolored water events were indicative of potential occurrence of problematic biofilms. Microbial testing was conducted to confirm that problematic biofilms were indeed contributing to the cause of water quality deterioration. Existing pipelines could not be easily replaced or lined; therefore, treatment in situ was selected and applied as the best mitigation option.

Objectives

To more effectively select biofilm control strategies, a better understanding of warning signs of biofilms and associated water quality issues is necessary. The objective of this study was to identify the site-specific water quality factors and microbial monitoring indices that were associated with observed problematic biofilm formation, to identify engineering alternatives to control or eliminate the episodic occurrence of these biofilms, and to present the results of different control strategies applied on a site-specific basis.

Experimental Protocol

Three sites, located in tropical or subtropical climates, were investigated. Each site represented a different land use (a condominium development, a resort, and a school). All three sites shared similarities with regard to water quality concerns in the potable water distribution network. These case studies illustrate three monitoring strategies that lead to the conclusion that biofilms were problematic. For each study site, one or more of the following issues were observed to indicate the presence of problematic biofilms:

Hydrogen sulfide odors;

Elevated heterotrophic plate counts (HPC >500 colony-forming unit [CFU]/100 mL);

Detection of indicator organisms and target pathogens;

Poor esthetic water quality (color, taste, turbidity, etc.); and/or

Indications of biofilm formation in the pipe through visual observation during construction/installation of new connections.

Water sample collection

For each site, the sampling points included one or more of the following:

The point of entrance to the public water system to the property (n = 1 per site);

Any extremities within the property boundaries, particularly at dead ends, where stagnant water was most likely; and

Hot and cold water fixtures (whenever applicable, as it was not available in all locations).

Water samples were collected using 1.0-L or 500-mL sterile Whirl-Pak bags, using aseptic practices. Before sampling, any part of the fixture that was plastic or rubber (i.e., for showers, the shower head) was removed. The fixture was flamed to ensure that exterior surface bacteria would not be introduced and that no significant external biofilm material would slough into the sample container. The water was allowed to run continuously for at least 2 min to flush the line before the final sample was collected. For hot water sampling, the 2-min flushing time was started after it became hot to the touch, which was approximately between 40°C and 50°C (104°F–122°F). All water samples were immediately stored at 4°C and then delivered for analysis within 4 h of sampling except those from Site B, which were delivered within 24 h (the site was located outside the United States).

Biofilm sample collection

Biofilms were also sampled directly with sterile swabs using aseptic practices. Swabs were collected from inside wells; distribution piping; and distribution system taps and fixtures, such as showers, sinks, backflow devices, and inside faucets. Before sampling, any part of the fixture that was plastic or rubber (i.e., for showers, the shower head) was removed (see above). Then using a sterile swab, a sample was collected by gently swabbing the inside of the pipe surface. The collected specimen was then inserted into a presterilized Collection-Eze^® transport tube containing 2 mL of Amies Transport Medium. All swab samples were immediately stored at 4°C and then delivered for analysis within 48 h of sampling. For each site, distribution system sample taps were also tested.

Microbial characterization

The AWWA C651 Standard on disinfecting water mains (AWWA, 1992) suggests the use of the presence/absence test for coliform bacteria as an indicator of a sanitary hazard. Additional tests were performed to define specific organisms, which was the desired information in this project.

Water samples from sites A and B were analyzed for HPCs using the membrane filter method (SM 9215D) (APHA et al., 2005). For each sample, an appropriate volume of sample water (100 mL) was filtered through a sterile 47 mm × 0.45 μm gridded membrane filter under vacuum. Filters were washed with three 25-mL portions of sterile dilution water. Filters were then placed on an m-HPC agar Petri dish and incubated at 35°C for 48 h. Similarly for Site C, water samples were analyzed for HPCs using the spread plate method (SM 9215C) (APHA et al., 2005). For this method, water samples of 0.1 and 1.0 mL were pipetted onto the surface of a prepoured R2A or m-HPC agar Petri dish. The inoculum was distributed using sterile L-spreaders and then incubated at 35°C for 48 h. After the incubation period, the number of CFUs was counted manually and results were reported as CFU/mL. Appropriate controls and blanks were also analyzed with each batch of samples.

Total coliform bacteria were analyzed using either the membrane filtration (SM 9222B) or the enzyme substrate for presence/absence (SM 9223) (APHA et al., 2005). Briefly, the method uses lauryl tryptose broth pH 6.8 ± 0.2 (after autoclaving) for enrichment with incubation for 24 h at 35°C. If tubes were negative, they were incubated for an additional 24 h with monitoring for turbidity, fermentation tube (gas production), and yellow color for acid reaction. Analyses of 100-mL samples were conducted.

Samples were analyzed for Legionella from Site B only. SM 9260J was used for analysis of Legionella in water and swab samples (APHA et al., 2005). Briefly, samples were concentrated using membrane filtration; the entrapped cells were resuspended and plated on selective media containing antibiotics. After up to 10 days of incubation at 35°C in a humidified incubator, plates were examined for CFUs.

To further examine microbial composition of biofilms, isolates from the plates inoculated with the swabs were subjected to a battery of biochemical tests to speciate the organisms. This included colony morphology, catalase tests, coagulase tests, the RapID system (Remel, Inc.), and API Systems (20C and 20E; bioMerieux). Briefly, for bacteria, colonies from HPC testing were isolated on tryptone glucose yeast agar and then transferred to 0.85% saline for analysis by API 20E test strip manufactured by bioMerieux. The API 20E system consists of a plastic strip of 20 individual cupules, each containing a different reagent used to identify enteric bacteria in the family Enterobacteraceae by their metabolic capabilities. Each cupule was inoculated with a saline suspension from each isolated colony and incubated at 37°C for 24 h. After incubation, each tube was assessed for specific color changes and converted to a seven-digit code, which in turn was interpreted using an API code book software package. Whenever possible, identification was confirmed by visual observation under a microscope to confirm morphological characteristics. All tests were performed in accordance with manufacturer's instructions. It must be noted that only biofilm organisms that could be cultured were characterized in this study. Culturable organisms only represent a fraction of the total microbial population. Full analysis of biofilm composition using metagenomics was beyond the scope of this project.

Chlorine residual analyses

Total residual chlorine was analyzed using the N,N-diethyl-p-phenylenediamine (DPD) colorimetric method (SM 4500-Cl G) (APHA et al., 2005). A Hach DR4000 UV/Vis spectrophotometer was used at 530 nm to record the absorbance values. Ten milliliters of the sample was added to a prepared vial containing the reagents (potassium iodide, phosphate buffer, and DPD indicator solution). The cap was securely fastened and the tube inverted 20 times. After a reaction period of 2 min, the vial was placed in the chamber (530 nm), and results were recorded in absorbance values and converted to mg/L chlorine using a 5-point standard curve.

Biofilm disinfection protocol

For the problematic biofilm case studies, chemical disinfection (chlorine) was typically one of the control measures. Chemical disinfection was conducted consistent with AWWA standards (AWWA, 1992). For recalcitrant biofilms, a superchlorination protocol was used as follows (AWWA, 1992):

Open all faucets in the building. Equalize the pressure and flow by opening the first-floor fixtures just a little and the top-floor fixtures all the way open.

Cut the flow back in the water line coming into the building to a trickle on the top floors. Should be the same on the first floor if the faucet is closed enough.

The target chlorine concentration in the distribution system was 100 mg/L. To this end, 1,000 mg/L of chlorine from calcium hypochlorite tablets is prepared by dissolving 1.28 pounds of calcium hypochlorite per 100 gallons of water and bled into the system with water from a service line.

From a 100-gallon tank, pump the chlorine solution into the service line/building pipe. When the water appearing in the fixtures shows a white, milky water appearance with a strong chlorine odor, turn off the faucet.

Once the 100 gallons has been pumped in, close the valve on the water line and make another 100 gallons. Once the new solution is made, begin pumping and open the service line.

Repeat steps 4 and 5 until chlorine was detected at all of the faucets.

Close all valves.

Provide a residence time of 24 h (take adequate precautions to ensure that people know that there is a problem at the faucets or they will burn their skin).

Flush the system completely until you can no longer detect the added chlorine. Flushing toilets accelerate this process.

Results

Three case studies of problematic biofilm detection and control are the subject of this study. Table 1 provides a comparison of the three sites with respect to infrastructure characteristics. Each site is detailed briefly below.

Table 1.

Comparison of Case Study Site Attributes

	Site A	Site B	Site C
Development type	Condominium	Resort	School
	6 buildings	1 building	30 buildings
Piping material	PVC, copper	Copper	Ductile iron, copper
Disinfection residual type	Chloramines	None	Chloramines
Typical TRC	0.4 mg/L	0	<0.2 mg/L
Backflow present?	Yes	Yes	Yes
Age of site	<2 years	15 years	>30 years

PVC, polyvinyl chloride; TRC, total residual chlorine.

Site A

Site A is a complex of high-density residential buildings (six in total) in south Florida. Four of the buildings consist of town homes, whereas the other two buildings are five-story, high-rise condominiums. Water service is provided by the local government-owned utility. The original developers installed a new pipeline, which was dedicated to the utility, and services for the six buildings. At the time when the first samples were collected in the study, the pipelines were <2 years old and were made of PVC and copper. Backflow prevention devices were provided on each of the connections. A review of the water quality files submitted to the Health Department did not indicate the presence of high HPCs or microbial concerns for the finished water entering the distribution network. Chloramines are used for disinfection. Total chlorine residuals were above 0.2 mg/L on all public taps as measured in the field. Internal samples had low or no detectable chlorine residuals.

Before sampling at Site A, the developer noticed hydrogen sulfide odors escaping from water fixtures, during construction. This phenomenon appeared to be random. Although half of the units had never been occupied, water and power had been supplied to all units for over 18 months. When certain pipelines were flushed, turbid and colored water was noted, and corrosion was present around plumbing fixtures. Although not unusual in potable water systems, the amount and persistence of the material being discharged and the pungent hydrogen sulfide odors raised concerns as discussed by Dillon (1995) and Videla (1996). Thus, the combination of sulfide odors and deteriorated water quality indicated that problematic biofilm growth may have been present.

At Site A, water quality sampling was conducted in the two structures located at the extremities (Buildings 1 and 6) of the property as well as the unoccupied five-story tower in the middle of the complex (Building 5). For each building, a representative set of water fixtures was sampled. No units were ever occupied in Building 1, so Units A through F in Building 1 were used for test purposes. Buildings 2 and 5 could not be tested in as systematic manner as the other property buildings because all units are interconnected and some units had periodic occupancy. Table 2 summarizes the results obtained from the initial sampling event indicated HPCs as high as 2,400 CFU/100 mL was measured in the interior of Building 1 (which has six units, A to F). Biofilm organisms were identified in samples from all buildings tested. The presence of typical biofilm organisms, including various Pseudomonas species, was found in unflushed (stagnant) units. No exterior test sites showed any HPCs.

Table 2.

Summary of Microbial Analyses for Site A

	No action		Before treatment				Filter ^a	Superchlorination
	Unit A Inside	Unit F Inside	Unit C Inside	Unit C Outside	Unit D Outside	Unit E Outside	Unit B Inside	Unit C Inside	Unit D Inside	Unit D Outside	Unit E Inside	Unit E Outside
HPC (CFU/100 mL)	3,200	240	2,400	<1	<1	<1	1,200	<1	<1	<1	<1	<1
Corynebacterium sp.			×	×
Pseudomonas aeruginosa	×						×
Pseudomonas alcaligenes	×	×	×	×			×
Pseudomonas stutzeri	×
Bacillus sp.					×	×			×	×		×
Pantoea agglomerans			×
Stenotrophomonas maltophilia			×
Staph-coag negative		×		×				×			×
Alcaligenes xylosoxidans		×

“ × ” indicates species detected.

Filter—filter change.

CFU, colony-forming unit; HPC, heterotrophic plate count.

On the basis of the water quality problems observed and microbial testing, four actions were recommended to the developer (1) do nothing (Units A and F), (2) change the 5-μm sediment prefilter unit at hot water heater location (Unit B), (3) perform flushing (Unit C), and (4) superchlorinate (Units D and E). Table 2 also summarizes the results after implementing the recommended actions.

The HPC values remained high in unoccupied Units A and F, where the developer did nothing. In Unit B, where a 5-μm sediment prefilter was changed at the hot water location, Pseudomonas spp. remained detectable. Flushing of Unit C with chlorinated water from the public water supplier for 15–30 min was effective at reducing the level of bacteria at the inside tap to below detection limits. Superchlorination in Units D and E resulted in nondetects from the inside taps.

Factors contributing to problematic biofilms and delivered water quality degradation may be a result of a number of factors at Site A. The presence of stagnant water in unoccupied buildings may be hypothesized to contribute to detectable HPC populations, as evidenced in Units A and F. In this underutilized distribution system, a combination of limited flushing of accumulated biomass material by flowing water as well as depletion of chlorine residual because of long detention time allows for microbial regrowth and thus problematic biofilm accumulation (LeChevallier et al., 1990). A similar observation was reported in a Massachusetts distribution system where increased biofilm and microbial water quality degradation was reported for an encrusted distribution system with depleted chlorine residual (Reilly and Kippin, 1983). As a long-term solution, flushing regularly with incoming chlorinated public water supply water would improve the situation (Walker and Morales, 1997). Periodic superchlorination, proven to be effective in Units D and E, may still be needed if consistent flushing is not implemented.

Site B

Site B is a Caribbean resort property consisting of an eight-story building with ancillary structures that include lodging space (400 + rooms), restaurants, a casino, and other amenities. At the time of the study, the building was <20 years old. Water supply to the resort is provided by a publicly owned desalination facility. The water is supplied to the resort by an 8-inch water line, and service lines within the resort branch off from this main line. Routine sampling and analysis is conducted using European Union standards (i.e., Legionella). Chlorine is not used as a primary disinfectant by the local public utility, but the resort has a protocol and chlorine disinfection system once water reaches the property. The purpose of the chlorine disinfection system is to provide a residual to prevent the growth of biofilms and eliminate pathogens in the pipeline. The pipe materials are primarily copper.

Before conducting a microbial water quality sampling survey at Site B, concerns about Legionella in the potable water system were raised. In addition, report patron complaints of color, odor, and biomass in the showers were documented. Sampling indicated Legionella densities in the water as high as 7,400 CFU/100 mL, levels that imply potential human health effects (Fever ID₅₀ = 20–129 organisms using guinea pigs) (U.S. EPA, 1999). These factors and Legionella testing indicated that problematic biofilms were likely a cause of deteriorated delivered water quality.

Water samples and swabs were collected from a number of individual units distributed around the resort, including showerheads and shower walls, because of concerns raised by resort patrons. Table 3 summarizes the results of fungal and bacterial incubation and speciation. Except for the gray water system sample and the HVAC system, the water and swab samples are virtually negative for common biofilm builders. Quantities of the fungus Acremonium and the bacterium Pantoea agglomerans were isolated from the incoming water. However, the output from the freshwater storage tank after chlorination was completely negative for bacteria and fungi. Only seven of the remaining 17 samples developed visible bacterial colonies upon culturing. Pseudomonas fluorescens was isolated in four of the samples. This bacterium is a ubiquitous slime former, and it would not be surprising to find this microorganism in any water system, especially at the end of the distribution network (Schmeisser et al., 2003). Likewise, the finding of coagulase-negative Staphylococcus sp. may indicate biofilm potential (Geldreich, 1996; Lee and Kim, 2003), but neither the Pseudomonas sp. nor coagulase-negative Staphylococcus sp. represent a public health concern, and both are controllable with chlorine.

Table 3.

Summary of Microbial Analyses for Water and Swab Samples at Site B

	ND	L	Pf	Pa	Ps	B	Ah	Fo	F	Cp	Ab	Pag	S−	Ac
Freshwater input												×		×
Freshwater tank output	×
Freshwater tank discharge^a
Gray water output				×		×	×	×		×
HVAC condenser pump		30		×	×	×			×		×	×
Hot water tank			×			×
Beach shower	×
Room 127 hot	×
Room 127 cold	×
Room 127 shower head^a
Room 349 hot			×
Room 501 hot	×
Room 501 cold	×
Room 527 hot													×
Room 527 cold	×
Room 533 hot	×
Room 533 cold			×
Room 533 shower wall^a
Room 533 shower head^a
Room 867 hot
Room 867 cold	×												×
Room 811 hot	×
Room 811 cold			×
Room 849 cold													×
Ice machine #27	×
Engineering sink	×

“ × ” indicates species detected; bold-faced “ × ” indicates heavy concentrations.

Swab sample.

Ps, Pseudomonas stutzeri; Ab, Acinetobacter baumannii; ND, no microbial species detected; B, Bacillus sp.; Pag, Pantoea agglomerans; Ah, Aeromonas hydrophila; S−, coagulase-negative Staphylococcus; L, Legionella; Fo; Flavobacterium odoratum; Ac, Acremonium (fungus); Pf, Pseudomonas fluorescens; F, Flavobacterium sp.; Pa, Pseudomonas aeruginosa; Cp, Clostridium perfringens.

In general, the diversity of organisms identified and number of locations containing biofilm organisms coupled with color and odor complaints indicate that problematic biofilm is a concern at this site. The water samples swabbed from the shower head in one room was negative. Therefore, the amount of biofilm activity existing in the pipeline was likely limited. Further investigation revealed that on-site operators found that the resort guests objected to the use of chlorine, and despite a directive by the management to operate the chlorination system provided, this was only done intermittently. Application of a chlorine residual continuously has been proven to keep help reduce accumulated biomass and keep future microbial biofilms in check (LeChevallier et al., 1990; Fass et al., 2003). A 0.3-mg/L chlorine residual protocol was implemented, which eliminated the problem (note resampling for chlorine residual was not possible given the distance to the site).

Since the nature of the incoming water was demonstrated to support the development of problematic biofilm when chlorination is only intermittently applied, the recommended corrective actions included flushing the system and regular use of the existing chlorination system as designed. With on-going use of the on-site chlorinator, no further biofilm-related problems or resort patron water-related complaints were noted.

Site C

Site C is a community college campus with 30 buildings in the southeastern United States that includes a library, a day-care center, offices, classrooms, and a central chiller building. Water supply to the campus is provided by a local governmentally owned utility. Of note is that the campus is located at the farthest extremity of the municipal water distribution network. The utility did not indicate the presence of biofilm organisms in the main system, but total chlorine residuals lower than 0.2 mg/L were recorded. A series of internal water supply pipelines provide service to the different campus buildings. The network consists of a variety of pipe materials (including some unlined ductile iron pipe that is on the order of 30–40 years old) and ages, since the lines were installed during construction of each separate building. The following improvements were made to the system before the study taking place:

Installation of backflow prevention devices on all buildings and fire lines;

Installation of new piping to provide looping for all dead-end water mains to limit areas of stagnant water;

Replacement of several small, older pipelines with new PVC pipelines; and

Replacement of copper pipelines in the chiller building and certain other buildings.

At each location investigated for Site C, water quality was reviewed by taking the following action steps. First, a thorough review of existing laboratory analyses was conducted. Results showed consistently high HPCs in samples associated with the plumbing of the buildings. Next, a survey of the water distribution system was conducted via field site visits and plan reviews. Then, a timeline of corrective and preventative maintenance actions was assembled. Finally, water samples for microbiological analysis were collected and analyzed, before forming a recommended implementation plan for corrective measures.

At Site C, biofilms were discovered during construction renovations to the pipeline of the chiller building. Subsequent targeted sampling enumerated HPCs in excess of 177,000 CFU/100 mL in two nearby buildings. The highest HPCs were found in fixtures within buildings that were in proximity to the chiller line. Although coliforms were not detected, the persistence of this particular biofilm was a concern and, as a result, water was shutdown to the affected buildings, and waterless foam soap was provided to the occupants to prevent potential sanitary and hygiene issues. During this time, turbid and colored water was also encountered at many water fixtures within the buildings.

Table 4 summarizes the results obtained from sampling at Site C. Before the work conducted as part of this study, Pseudomonas spp. (177,000 CFU/100 mL) and unidentifiable species (158,000 CFU/100 mL) were enumerated in the building water. Although no maximum contaminant level applies to HPCs, the levels recorded here indicate a potentially significant problem consistent with findings of Geldreich (1996). Knowledge of the piping system indicated that these high HPC levels were not related to cross connections or fecal contamination.

Table 4.

Summary of Microbial Analyses for Site C

	HPC (CFU/100 mL)
	Date
Location	9/8/2003	9/29/2003	8/10/2004	8/23/2004	8/26/2004	8/30/2004
Building 42 backflow	250					250
Building 47 backflow			470	4,400		220
3rd FL men room			920	1,160		25,000
3rd FL women's room			530	1,700		810
2nd FL men's room			480	1,310		650
2nd FL women's room			920	920		700
1st FL men's room			6,300	80		860
1st FL women's room			1,710	850		4
2nd kitchen				7,400		720
Room 308			2,430	183		780
Building 56 fire hydrant					22,000
Building 48 backflow	19,500			390		220
3rd FL men's room	3,400	550	71,000	1,010		400
3rd FL drinking fountain	14,200
3rd FL women's room		890	18,900	880		110
2nd FL men's room			85,000	1,410		55
2nd FL drinking fountain	14
2nd FL women's room	21,400	400	47,000	800		151
Room 206			28	0		2
1st FL men's room			21	1,190		98
1st FL drinking fountain	34,800
1st FL women's room	184	1,560	58,000	1,850		300
2nd kitchen
Room 308
City backflow S^a			4	<1	40,000	80
City backflow N^a			5	1	280 + TCpositive	69
Building 62 backflow						300
Building 41 backflow						420
Building 61 backflow						11,600
Building 57 backflow						39
Library					64

Initial two samples collected on 6/9 and 6/29.

TC, total coliforms.

The results of swab testing indicate high quantities of typical slime-forming bacteria that contribute to the establishment of biofilms (e.g., Pseudomonas, Bacillus, and Mycobacteria), evidence of iron bacteria (Sparganium natans and iron-encrusted filaments), and fungi (Aspergillus) (LeChevallier et al., 1987; Emde et al., 1992; Norton and LeChevallier, 2000; Lee and Kim, 2003). However, no coliforms were detected at any location except in one instance at the utility connection point.

While this site is serviced by a municipal water system that utilizes disinfection residual as a process to protect distribution system quality, this location is at the end of the system. To resolve the problem, it was recommended that the entire distribution system and all buildings be superchlorinated. This was scheduled in December after conclusion of the Fall semester, when the buildings are less occupied, and applied as described in the Methods section. Re-sampling after superchlorination showed no HPCs and no fecal coliforms.

To ascertain a longer-term solution for this site, an evaluation of disinfection results was undertaken. One of the problems is that the site is a large user of water, but has 2–3-week periods with virtually no flow. The site is also located at the end of the distribution system (although there are two feeds, they are both located at the end of the system). Customers at the ends of distribution networks are more likely to receive water with depleted chlorine residuals than other customers, and biofilms that detach in the pipe system are also more likely to flow toward the extremities, where they may find conditions suitable for re-colonization.

Measurements indicated that the chlorine residuals delivered to the campus were <0.2 mg/L periodically, which is a cause for concern. Analysis of water quality reports submitted to the Health Department by the utility demonstrated problems meeting the minimum chlorine residuals throughout their service area in 2006 and 2007, and particularly at the meters to Site C. This may have contributed to the problematic biofilm issue returning to noticeable levels again in 2008 (See Table 5). Recommendations to address the problem included

Routine flushing;

Monitoring of local chlorine residuals so that they do not approach zero; and

Installation of a liquid chlorine booster station to maintain the chlorine residual at acceptable levels, when necessary.

Table 5.

Results of Chlorine and Heterotrophic Plate Count Sampling for Site C

	May 2008			June 2008			July 2008
Sample	Total residual chlorine (mg/L)	Total residual chlorine (mg/L)	HPC ^a (CFU/mL)	Total residual chlorine (mg/L)	HPC ^a (CFU/mL)	HPC ^b (CFU/mL)	Total residual chlorine (mg/L)	HPC ^a (CFU/mL)	HPC ^b (CFU/mL)
Golf course check valve	BDL	0.10	96	0.05	60	60	BDL (0.03)	115	60
NW check valve	BDL	BDL	109	BDL (0.03)	>1,000		BDL (0.01)	>1,000	329.5
41—check valve	BDL	0.15	210	BDL	>1,000		BDL	110	47.5
46—1st floor men's bath	BDL	BDL					BDL	177.5	180
47—3rd floor men's bath	BDL	BDL	474	BDL	>1,000	85	BDL
48—3rd floor men's bath	BDL	BDL		BDL			BDL
56—1st floor sink	BDL	BDL	174	BDL	>1,000	170	BDL		175
56—1st floor sink DUP							BDL		267.5
Blank	BDL	0.05	0	BDL	0	0	BDL		0

HPC using R2A agar.

HPC using plate count agar.

BDL, below detectable level (0.00 unless otherwise indicated).

From the analysis of chlorine residual data, it was recommended that the site owner work with the utility to increase the chlorine residual to 0.6 mg/L on the campus. However, the utility has been unwilling to support an on-site system. Politically, for the college to manage a utility sanctioned on-site system, it would require the college to become a consecutive system and hire operators. To date, the problem has not been resolved, and the campus is considering becoming a consecutive system operation as a result.

Summary

Biofilms commonly occur in distribution systems. They have the tendency to develop in piping systems with the following characteristics:

Low flow/dead zones;

High nutrients;

Exposed ferrous materials; and

Low or no chlorine residual.

Three sites were investigated because of water quality issues that appeared to be associated with biofilms, including sulfur odors, elevated HPCs, and visual biofilms. The intent was to determine the types of microorganisms present and develop solutions to reduce potential impacts from them. Table 6 summarizes the actions taken at each site (or those recommended, such as periodic monitoring of HPCs).

Table 6.

Actions Taken at Each Study Site

Action taken	Site A	Site B	Site C
Flushing	×	×	×
Chlorine addition		×	×
Superchlorination	×		×
Plug-superchlorination			×
Bacteria monitoring	×	×	×
Chlorine residual monitoring	×	×	×
Pipe replacement			×

In reviewing the results of these actions, it was found that HPCs were largely eliminated for all sampling points for all three sites by maintaining chlorine residuals, although the biofilms were not completely eliminated. The units treated with superchlorination and flushed regularly at Site A had no further issues to date. At Site B, maintaining a chlorine residual (higher initially) eliminated the issue, indicating that the biofilm was just starting to establish. At Site C, superchlorination of buildings and plug chlorination of the on-site distribution system seemed to control the problem. However, elevated HPCs returned after an incident when the chlorine residual dropped below detection limits. It was recommended that the Site C consider re-chlorinating the system with sodium hypochlorite to reduce the potential for further problems.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

APHA, AWWA, and WEF. 2005. Standard Methods for the Examination of Water and Wastewater, 21st. Washington, DC: American Public Health Association.

AWWA. 1992. Standard for Disinfecting Water Mains(Standard C651-92)Denver: American Water Works Association.

Bachmann

R.T.

, Edyvean

R.G.J.

2006. Biofouling: an historic and contemporary review of its causes, consequences and control in drinking water distribution systems. Biofilms, 2:197.

Bagh

L.K.

, Albrechtsen

H.-J.

, Arvin

, Ovesen

2002. Biofilm formation in a hot water systems. Water Sci. Technol., 46:95.

Batte

, Mathieu

, Laurent

, Prevost

2003. Influence of phosphate and disinfection on the composition of biofilms produced from drinking water, as measured by fluorescence in situ hybridization. Can. J. Microbiol., 49:741.

Cantor

A.F.

, Bushman

J.B.

, Glodoski

M.S.

, Kiefer

, Bersch

, Wallencamp

2006. Copper pipe failure by microbiologically influenced corrosion. Mater. Perform., 45:38.

Chapelle

F.H.

1993. Groundwater Microbiology and Geochemistry. New York: John Wiley and Sons.

Crozes

G.F.

, Cushing

R.S.

2000. Evaluating Biological Regrowth in Distribution Systems. Denver: American Water Works Association Research Foundation.

Dillon

C.P.

1995. Corrosion Resistance of Stainless Steels. New York: Marcel Dekker, Inc.

10.

Eboigbodin

K.E.

, Seth

, Biggs

C.A.

2008. A review of biofilms in domestic plumbing. J. AWWA, 100:131.

11.

Emde

K.M.E.

, Smith

D.W.

, Facey

1992. Initial investigation of microbially influenced corrosion (MIC) in low temperature water distribution systems. Water Res., 26:169.

12.

Fass

, Block

J.-C.

, Boualam

, Gauthier

, Gatel

, Cavard

, Benabdallah

, Lahoussine

2003. Release of organic matter in a discontinuously chlorinated drinking water network. Water Res., 37:493.

13.

Geldreich

E.E.

1996. Microbial Quality of Water Supply in Distribution Systems. Boca Raton: CRC Press.

14.

Hall-Stoodley

, Stoodley

2005. Biofilm formation and dispersal and the transmission of human pathogens. Trends Microbiol., 13:7.

15.

Kirmeyer

, Richards

, Smith

C.D.

1994. An Assessment of Water Distribution Systems and Associated Research Needs. Denver: American Water Works Association Research Foundation.

16.

Kirmeyer

G.J.

, Friedman

, Martel

K.D.

, Noran

P.F.

, Smith

2001. Practical guidelines for maintaining distribution system water quality. J. AWWA, 93:62.

17.

LeChevallier

M.W.

, Babcock

T.M.

, Lee

R.G.

1987. Examination and characterization of distribution-system biofilms. Appl. Environ. Microbiol., 53:2714.

18.

LeChevallier

M.W.

, Lowry

C.D.

, Lee

R.G.

1990. Disinfection of biofilms in a model distribution system. J. Am. Water Works Assoc., 82:87.

19.

Lee

D.-G.

, Kim

S.J.

2003. Bacterial species in biofilm cultivated from the end of the Seoul water distribution system. J. Appl. Microbiol., 95:317.

20.

Mains

2008. Biofilm control in distribution systems. Tech. Brief, 8:1.

21.

Murga

, Forster

T.S.

, Brown

, Pruckler

J.M.

, Fields

B.S.

, Donlan

R.M.

2001. Role of biofilms in the survival of Legionella pneumophila in a model potable-water system. Microbiology, 147:3121.

22.

Niquette

, Servais

, Savoir

2000. Impacts of pipe materials on densities of fixed bacterial biomass in drinking water distribution system. Water Res., 34:1952.

23.

Norton

C.D.

, LeChevallier

M.W.

2000. A pilot study of bacteriological population changes through potable water treatment and distribution. Appl. Environ. Microbiol., 66:268.

24.

Reilly

J.K.

, Kippin

J.S.

1983. Relationship of bacterial counts with turbidity and free chlorine in two distribution systems. J. AWWA, 75:309.

25.

Sartory

D.P.

, Holmes

1997. Chlorine sensitivity of environmental, distribution system, and biofilm coliforms. Water Sci. Technol., 35:289.

26.

Schmeisser

, Stockigt

, Raasch

, Wingender

, Timmis

K.N.

, Wenderoth

D.F.

, Flemming

H.-C.

, Liesegang

, Schmitz

R.A.

, Jaeger

K.-E.

, Streit

W.R.

2003. Metagenome Survey of Biofilms in Drinking-Water Networks. Appl. Environ. Microbiol., 69:7298.

27.

September

S.M.

, Els

F.A.

, Venter

S.N.

, Brözel

V.S.

2007. Prevalence of bacterial pathogens in biofilms of drinking water distribution systems. J. Water Health, 5:219.

28.

U.S. EPA. 1992. Control of Biofilm Growth in Drinking Water Distribution SystemsEPA/625/R-92/001Washington, DC: Office of Research and Development.

29.

U.S. EPA. 1999. LegionellaHuman Health Criteria Document, EPA-822-R-99-001Washington, DC: Office of Science and Technology, Office of Water.

30.

U.S. EPA. 2002a. Deteriorating Buried Infrastructure Management Challenges and Strategies. Office of Ground Water and Drinking Water Distribution System Issue Paper. Washington, DC: Office of Water.

31.

U.S. EPA. 2002b. The Clean Water and Drinking Water Infrastructure Gaps AnalysisEPA-816-R-02-020Washington, DC: Office of Water.

32.

Van der Kooij

, Veenendaal

H.R.

2001. Biomass production potential of materials in contact with drinking water: methods and practical importance. Water Sci. Technol. Water Supply, 1:39.

33.

Videla

H.A.

1996. Manual of Biocorrosion. Boca Raton: CRC Press.

34.

Walker

J.T.

, Morales

1997. Evaluation of chlorine dioxide (ClO₂) for the control of biofilms. Water Sci. Technol., 35:319.

35.

Zhang

, Griffin

, Edwards

2010. Effect of nitrification on corrosion of galvanized iron, copper and concrete. J. AWWA, 102:83.