A toolbox approach for meeting proposed changes to the U.S. EPA's MDBP rules

Resulting recommendations

The work group provided a report in November of 2023⁵ that provided 13 recommendations and received varying levels of support. The first and the thirteenth of those recommendations are the focus of this article:

1. Recommendation 1: Disinfectant Residual: Address the potential for no or low disinfectant residual in surface water Public Water System (PWS) distribution systems (DS). This recommendation has two parts: 1.) Raising the national minimum disinfectant residual requirement from the current value of “detectable” and include a range for consideration of setting a minimum specific value of up to 0.5 mg/L for free chlorine and 0.7 mg/L for total chlorine for chloraminating systems. In addition, EPA should consider removing alternative compliance criteria (e.g., HPC); 2.) Establish and require adoption of a disinfectant residual sampling and monitoring plan that will provide an accurate understanding of areas within the distribution system that have low or no disinfectant residual.
2. Recommendation 13: Ground Water Under the Direct Influence of Surface Water (GWUDI) – EPA should revisit the definition, determination methods, and guidance for GWUDI to ascertain what changes should be made to improve the protection of public health.

It is noted that many municipal groundwater wells in New England, from the authors’ experience, tend to be gravel pack wells, generally 60 feet or less in depth, and are often located near water bodies, generally wetlands. As testing for indicators for GWUDI improves, it is likely that more wells will be required to be treated like surface water. Also, the bulk of the municipal wells were constructed in the period from 1950 to 1975. Thus, many wells are reaching upwards of 75 years old, the timeframe that many would expect would be at the end of the useful life of a production well. Several methods exist to rehabilitate wells and extend their life which are outside the scope of this article, and likely many communities will invest in these technologies to extend the life of the wells as long as possible. However, as these wells begin to fail, they will also be more likely to experience surface water intrusion. 

As the above suggests, there may come a time in the near future of unity in the governing rules for disinfection for groundwater and surface water. It is quite possible, as detection methods and our understanding of groundwater improve, coupled with development pressures (e.g., septic systems located in well Zone II), that many groundwater wells will no longer be considered protective by providing the natural filtering ability to remove pathogen contaminants (or other contaminants for that matter). A large segment of these wells could be considered groundwater under the influence of surface water and beholden to those rules governing surface water, for which a numerical chlorine limit could be imposed for the distribution system of not only systems receiving surface water but groundwater as well. 

However, is a numerical chlorine residual limit required? Is it perhaps too blunt of a hammer to use to ensure water is safe for consumption as a one-size-fits-all approach? The purpose of this discussion is to present a case study using various common parameters and one novel parameter, adenosine triphosphate (ATP), for monitoring the biostability of a distribution system and guiding the use of targeted chlorination. 

ATP is considered the “energy currency” that all life uses. One of the potential benefits of ATP for this purpose is its ability to quickly measure all the biological activity from field samples within the distribution system instead of a smaller subset of select bacteria groups and provide a potential early warning system of changes in biostability. 

The results of this study could be considered an example ‘toolbox’ approach which would be used, if allowed by EPA and state regulatory agencies, to address the updates to the MDBP rules. 

Methods

The Wareham, Massachusetts Fire District (District) partnered with a local Massachusetts Department of Environmental Protection (MassDEP) certified laboratory for:

1. Coliform testing using EPA 1604.
2. Heterotrophic Plate Count using R2A agar - SM 9215 D. Results are after five days of incubation, estimated to be roughly system resident time at the extents of the distribution system without exceeding the ability to count the number of bacteria.

The remaining tests were performed in house as follows:

1. Cellular ATP was performed using LuminUltra’s Quench Gone Aqueous Test Kit and Lumitest C-1110 and equipment set (EQP-PAC-C110). The District purchased sterilized, 500 mL, wide-mouth, polypropylene bottles from VWR which were used to take field samples which were then analyzed back at the District office.
2. Free chlorine was measured using Hach’s Method 8021 using DPD and a DR900 colorimeter.
3. pH was measured using a Hach portable pH probe with temperature correction. The probe was calibrated with a pH 7 buffer and pH 10.01 buffer.

Weather data was taken from a National Oceanic and Atmospheric Administration (NOAA) station located in East Wareham. (USC00192451). 

Sources and treatment

The District has eight water supply sources. Three are listed on the Seawood Springs side of the distribution system and five are located on the Maple Springs side. Treatment is basic and consists of pH adjustment with lime for corrosion control and chlorination with sodium hypochlorite to maintain a detectable residual in the distribution system. There are two corrosion control facilities which add hydrated lime – one services the water from the Seawood Springs Wells and one from the Maple Spring Wells prior to the water entering the distribution system.

The well water raw quality is generally characterized as low alkalinity (generally less than 10 mg/L as calcium carbonate (CaCO₃₎), low total organic carbon (less than 2.0 mg carbon/liter (C/L)), low ultraviolet (UV) 254 (<0.015/centimeter), and low to moderate levels of iron and manganese, ranging up to 1.0 mg/L for each metal in select wells.

The wells are all located in the Plymouth-Carver aquifer and are gravel packed. The older wells tend to be 45 to 55 feet in depth and the two newer wells are upwards of 85 feet in depth. The Zone II is generally protected by a large portion of land located in the Miles Standish State Forest, and cranberry farming takes place upgradient of several of the wells.

The distribution system consists of 200 miles of pipe with roughly 30% being unlined asbestos cement pipe and the remaining a mixture of cast iron and ductile iron pipe. The distribution system also has three tanks: two steel standpipes (Bourne Hill and West Wareham) and one elevated glass-lined tank (Glen Charlie). Figure 1 provides a general layout of the facilities.

Specific notes on District operations are as follows:

1. The District implements a full annual unidirectional flush program roughly from April 15 to the first week of July. The District typically boosts the chlorine coming from the two corrosion control treatment facilities to address any bacterial issues associated with flushing the pipe system.
2. Up until 2009, the District did not implement any secondary chlorination. In 2009, they went to seasonal chlorination in the summer which quickly moved to continuous chlorination.
3. In October of 2018, the District revised their corrosion control program to a target pH of 9.5-10 to maintain better control limits on pH out in the distribution system.
The target chlorine level coming out of each corrosion control facility was 0.5 mg chlorine/liter (CL₂/L) which was increased during the flushing period. The District would boost the chlorine manually in the distribution tanks when chlorine residual was lost if it occurred in the summer.

Results

The District embarked on roughly a two-year study to observe the biological trends within the distribution system. The goal of the testing was to better understand when biostability within the system was changing. As such, the District sampled for ATP, pH, and temperature at the Total Coliform sample sites and for HPC-R2A at the tanks. Precipitation and air temperature readings were taken from a nearby NOAA weather station in East Wareham. The sites were sampled roughly weekly in the summer and every two to four weeks in the winter. The results are shown in Figures 2 through 13, and a three-moving-period average was used in Microsoft Excel to help smooth out the relative data trends as shown in select figures.

Discussion and system action targets

The District has historically struggled with Coliform detections due to a variety of causes. They were initially associated with the lack of chlorine and subsequently believed to be associated with the age of the supply and distribution system assets as well as water quality originating from the wells. The District implemented disinfection to meet the 4-log disinfection requirements of the Groundwater Rule for viruses. After this study, a new pressure filtration plant treating all the well water was constructed, as well as 4-log disinfection using UV reactors instead of chlorine. The District received a grant from the MassDEP Gap Energy Grant Program which allowed photovoltaic solar panels to be installed on the roof of the water treatment plant which offset the electrical demand from the UV reactors.

The District also wished to establish precursor parameters that would provide an indication that they may be at higher risk of a coliform detection and that biostability had shifted and changed. In this effort, they set out on a roughly two-year internal study to quantify potential trigger or alarm parameters which would indicate the need to address biostability of the distribution system and direct the chlorination practices.

To this end, the District began measuring the following on a regular basis within the system at the Total Coliform sampling sites, as well as at the three distribution tanks:

Over the course of sampling, the District observed the following:

1. ATP tended to be relatively good at providing an early indication of biological activity in the distribution system. Once ATP exceeded 2 picograms/milliliter (pq/ml) in the distribution tank, the District typically boosted the chlorine in the tanks by adding enough sodium hypochlorite to reach a 1 mg/L of chlorine in the tank (Figures 2 and 3). Similarly, 2 pq/ml in the distribution system would dictate an increase in the chlorine added at the corrosion control treatment stations. These events often occurred at a similar period.
   • ATP, while being a potentially good indicator for changes in biostability in the distribution system, did not necessarily guarantee that low levels of it equated to no detections of coliform. For example, as shown in Figures 2, 3 and 11, in July of 2019 when ATP values were less than 1 pq/ml, Total Coliform was detected in one location of the system. However, no E. coli were detected. In discussions with Dan Mahoney at the Sandwich, Massachusetts Water District, they also experienced similar difficulties in getting low ATP values to correlate with non-detections of Total Coliform (personal communication).
2. Free chlorine proved to be a good indicator of biological activity, as well as other chlorine demands which were occurring in the distribution system. Once the chlorine residual dropped to 0.1 mg Cl₂/L or below, the District would consider boosting the chlorine in the distribution tanks or increasing the chlorine be added at each corrosion control facility (Figures 4 and 5).
   • It is noted that Total Coliform was detected in both November of 2018 and 2019, after chlorine was boosted in the distribution tanks. It is possible that the District would benefit from a continual dosing of chlorine at the distribution tanks during the periods of October and November of each year since the water is warmer and residence times are higher (i.e., less demand). More work is required to best determine the amount, length, and frequency of dosing either at the tanks or alternative boosting of the chlorination locations (i.e., corrosion control facilities) to further tease this option out for successfully controlling Total Coliforms.
3. pH did not prove to be a useful indicator for biological stability by itself (Figures 6 and 7). However, as the distribution pH stabilized at a pH of 9.5, the residual chlorine appeared to be on an upward trend (Figures 4 and 5), perhaps due to some passivation of iron corrosion at the time.
4. Both air and water temperature seemed to be favorably associated with increases in biological activity, as would be expected (Figures 8, 9, and 11). Biological activity increased above temperatures of 50^o Fahrenheit (F) (10^o Celsius (C)) and tended to peak in the end of summer or early fall at around 70^oF (21.1^oC). As the District approached the 70^oF mark, the District would consider dosing the tanks with chlorine and/or boost the chlorine added at each corrosion control facility.
   • It should be noted here that the District experiences maximum water demands typically in the period between May and August each year. September brings an end to the tourist season and the water demands significantly decrease (Figures 12 and 13). Hence, the water age would increase overall in the distribution system. This, coupled with warmer temperatures still present in the tanks, was when the tanks were most at risk of developing a Total Coliform detection as illustrated by the data. Figure 11 shows that Total Coliform detections occurred both in November of 2018 and 2019, suggesting that warmer temperature with less water consumption (i.e., longer water age) was considered a main factor in the detection of coliforms. This suggests that practices like bleeding the tanks or selecting areas of the distribution system to add automatic flushing units or something similar may aid in lowering the risk of a Total Coliform detection.
5. HPC- R2A 5-day results showed a moderate increase in activity with warmer temperature and less chlorine. However, there tended to be a significant amount of noise in the data, so the testing for this parameter was stopped in 2019 (Figure 10). The District did not use this parameter for distribution system booster chlorine and found ATP measurements to be better suited for this purpose.
Precipitation had a limited association with detections of Coliform, since generally, a series of increasing monthly totals of rain of two or more months tended to be when Coliform was detected (Figure 11). However, temperature appears to be a stronger indicator for biological stability and potential coliform detection in the distribution system over precipitation - more work is needed to establish if there truly is a relationship. The District did not use this parameter to boost chlorine in the system.

Summary

Distribution biostability is a relatively new concept in the drinking water industry. Historical operation strategy holds that if the water is flowing, everything is good. In recent years though, corrosion control has become a prominent theme, along with maintaining water quality such that there are no pathogens detected in the distribution system.

The recent NDWAC working group on the MDBP rules revision has recommended that a minimum disinfect residual be maintained (i.e., Recommendation 1) at the end of the distribution system for surface water suppliers, as well as enhanced GWUI testing to occur more frequently for groundwater suppliers (i.e., Recommendation 13). It is not a stretch of the imagination that a mandatory chlorine residual in the distribution system could be extended to groundwater systems in the future as well, based on these recommendations.

Increasing the amount of chlorine to minimum levels at the ends of the distribution system (0.5 mg/L for chlorine or 0.7 mg/L for chloramines) is likely a much higher level of chlorination than many systems implement now. It will likely shift many systems to move away from chlorine to chloramines and push many systems to remove more total organic carbon to maintain compliance with the Stage 2 Disinfection Byproducts Rules DBPR or its future revisions. By doing this, what other unintended consequences would they face? Some examples are simultaneous compliance issues (e.g., higher corrosion rates and higher DBPs at higher residual chlorine levels), emerging contaminants (e.g., nitrogenous DBPs), as well as public outreach as water quality changes with higher chlorine values or a change in disinfectant (i.e., chlorine to chloramine or taste and odor issues). 

The authors here suggest that, should the EPA working group’s recommendations be adopted as proposed, both EPA and state regulatory agencies allow the flexibility of public water systems to implement a toolbox of approaches to address the rule revisions, rather than on a one-size-fits-all minimum chlorine residual in the distribution system. This article provides some suggestions for parameters that public water systems could use to meet the intent of the coming rule updates.

In this article, we provided some of the parameters tested by the District which allowed the detection of a biostability issue in the distribution system. Using both nontraditional methods, like ATP, along with traditional methods, like chlorine residual, the District was able to be proactive in implementing its chlorine dosing. Coupled with this, the water quality monitoring allowed the District to use varying levels of residual chlorine as required by the distribution system instead of targeting a flat, set numerical value. 

Acknowledgements

The authors gratefully acknowledge the Wareham Fire District Water Commissioners Edward J. Tamagini, III, John B. English, III, and Richard England for their foresight and support for the improvements to the system and operations to provide better water quality. 

Special thanks also go to Dan Mahoney at the Sandwich Water District for reviewing the draft version of this article and providing critical feedback.

References

1. US Environmental Protection Agency, Stage 1 and Stage 2 Disinfectants and Disinfection Byproducts Rules
2. “Proposed Settlement Agreements, Safe Drinking Water Act Claims,” April 16, 2020, https://www.federalregister.gov/documents/2020/04/16/2020-07980/proposed-settlement-agreements-safe-drinking-water-act-claims
3. Settlement Agreement, Jun 1, 2020. https://waterkeeper.org/wp-content/uploads/2020/06/Settlement-Agreement.pdf
4. Crystal Rodgers-Jenkins, “Potential Revisions to Microbial & Disinfection Byproducts (MDBP) Rules: Background & Charge to NDWAC,” Oct. 11, 2023. https://www.epa.gov/system/files/documents/2023-10/presentation-mdbp-rules-background-and-charge-to-the-ndwac-october-11-2023_3.pdf
5. US Environmental Protection Agency, Stage 1 and Stage 2 Disinfectants and Disinfection Byproducts Rules