National Primary Drinking Water Regulations: Enhanced Surface Water Treatment Requirements; Proposed Rule ENVIRONMENTAL PROTECTION AGENCY

SUMMARY: EPA is proposing to amend the Surface Water Treatment Rule to

provide additional protection against disease-causing organisms

(pathogens) in drinking water. This action would primarily focus on

treatment requirements for the waterborne pathogens Giardia,

Cryptosporidium, and viruses. With the exception of one requirement

(sanitary surveys), this action would apply to all public water systems

that use surface water or ground water under the influence of surface

water, and serve 10,000 people or more. Among the features of the rule

would be a stricter watershed control requirement for systems using

surface water that wish to avoid filtration; a change in the definition

of ground water under the influence of surface water to include the

presence of Cryptosporidium; a periodic sanitary survey requirement for

all systems using surface water or ground water under the influence of

surface water, including those that serve fewer than 10,000 people; a

health goal (maximum contaminant level goal) of zero for

Cryptosporidium; and several alternative requirements for augmenting

treatment control of Giardia, Cryptosporidium, and viruses.

DATES: Comments should be postmarked or delivered by hand on or before

May 30, 1996. Comments received after this date may not be considered.

Public hearings will be held at the addresses indicated below under

ADDRESSES on August 30, 1996 (and 31, if necessary) in Denver, CO and

on September 13, 1994 (and 14, if necessary) in Washington, DC.

ADDRESSES: Send written comments to ESWTR Docket Clerk, Water Docket

(MC-4101); U.S. Environmental Protection Agency; 401 M Street, SW;

Washington, DC 20460. Please submit any references cited in your

comments. EPA would appreciate an original and three copies of your

comments and enclosures (including references). Commenters who want EPA

to acknowledge receipt of their comments should include a self-

addressed, stamped envelope. No facsimiles (faxes) will be accepted

because EPA cannot ensure that they will be submitted to the Water

Docket. The Agency requests commenters to follow the instructions

regarding format provided in Section IX of the Preamble, immediately

before the list of references.

The Agency will hold public hearings on the proposal at two

different locations indicated below:

1. Denver Federal Center, 6th and Kipling Streets, Building 25, Lecture

Halls A and B (3d Street), Denver, CO 80225 on August 30 (and 31, if

necessary), 1994.

2. EPA Education Center Auditorium, 401 M Street SW., Washington, D.C.

20460, on September 13 (and 14, if necessary), 1994.

The hearings will begin at 1:00 p.m., with registration at 12:30

p.m., on the first day. The hearings will begin at 9:30 a.m., with

registration at 9:00 a.m., on the second day. The Hearings will end at

4:00 p.m., unless concluded earlier. Anyone planning to attend the

public hearings (especially those who plan to make statements) may

register in advance by writing the ESWTR Public Hearing Officer, Office

of Ground Water and Drinking Water (4603), USEPA, 401 M Street, S.W.,

Washington, D.C. 20460; or by calling Tina Mazzocchetti, (703) 931-

4600. Oral and written comments may be submitted at the public hearing.

Persons who wish to make oral presentations are encouraged to have

written copies (preferably three) of their complete comments for

inclusion in the official record.

The proposed rule with supporting documents and all comments

received are available for review at the Water Docket at the address

above. For access to Docket materials, call (202) 260-3027 between 9 am

and 3:30 pm for an appointment.

FOR FURTHER INFORMATION CONTACT: The Safe Drinking Water Hotline,

Telephone (800) 426-4791. The Safe Drinking Water Hotline is open

Monday through Friday, excluding Federal holidays, from 9 a.m. to 5:30

p.m. Eastern Time. For technical inquiries, contact Stig Regli or Paul

S. Berger, Ph.D., Office of Ground Water and Drinking Water (MC 4603),

U.S. Environmental Protection Agency, 401 M Street SW., Washington DC

20460; telephone (202) 260-7379 (Regli) or (202) 260-3039 (Berger); or

Bruce A. Macler, Ph.D., Water Management Division, Region 9, U.S.

Environmental Protection Agency, 75 Hawthorne Street (W-6-1), San

Francisco, CA 94105-3901; telephone (415) 744-1884.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Statutory Authority

II. Regulatory Background

III. Discussion of Proposed Rule

A. Basis for Amending Existing SWTR: Limitations of SWTR

B. General Approach for Revising SWTR

C. Proposed Maximum Contaminant Level Goal and Treatment

Technique for Cryptosporidium

D. Proposed Revisions to SWTR under All Treatment Alternatives

1. Inclusion of Cryptosporidium in definition of ``groundwater

under the direct influence of surface water''

2. Inclusion of Cryptosporidium in watershed control

requirements

3. Sanitary surveys for all surface water systems

4. Possible supplemental requirements

E. Alternative Treatment Requirements

1. Options for defining pathogen densities in source waters

2. Treatment alternatives for controlling pathogens

IV. State Implementation

A. Special State Primacy Requirements

B. State Recordkeeping Requirements

C. State Reporting Requirements

V. Public Notification Language

VI. Economic Analysis

A. Cost of Proposed Rule

B. Benefits of Proposed Rule

VII. Other Statutory Requirements

A. Executive Order 12866

B. Regulatory Flexibility Act

C. Paperwork Reduction Act

D. Science Advisory Board, National Drinking Water Advisory

Council, and Secretary of Health and Human Services

E. Consultation with State, Local, and Tribal Governments

VIII. Request for Public Comment

IX. Instructions to Commenters

X. References

I. Statutory Authority

The Safe Drinking Water Act (SDWA or the Act), as amended in 1986,

requires EPA to publish a ``maximum contaminant level goal'' (MCLG) for

each contaminant which, in the judgment of the EPA Administrator, ``may

have any adverse effect on the health of persons and which are known or

anticipated to occur in public water systems'' (Section 1412(b)(3)(A)).

MCLGs are to be set at a level at which ``no known or anticipated

adverse effects on the health of persons occur and which allows an

adequate margin of safety'' (Section 1412(b)(4)).

At the same time EPA publishes an MCLG, which is a non-enforceable

health goal, it also must publish a National Primary Drinking Water

Regulation (NPDWR) that specifies either a maximum contaminant level

(MCL) or treatment technique (Section 1401(1), 1412(a)(3), and

1412(b)(7)(A)). A treatment technique may be set in lieu of an MCL if

it is not ``economically or technologically feasible'' to determine the

level of a contaminant.

Section 1412(b)(3) of the Act requires EPA to publish regulations

for at least 25 contaminants at three year intervals. Section

1412(b)(9) requires EPA to review each NPDWR every three years and

revise it if appropriate.

Section 1412(b)(7)(C) requires the EPA Administrator to publish a

NPDWR ``specifying criteria under which filtration (including

coagulation and sedimentation, as appropriate) is required as a

treatment technique for public water systems supplied by surface water

sources''. In establishing these criteria, EPA is required to consider

``the quality of source waters, protection afforded by watershed

management, treatment practices (such as disinfection and length of

water storage) and other factors relevant to protection of health''.

This section of the Act also requires EPA to promulgate a NPDWR

requiring disinfection as a treatment technique for all public water

systems and a rule specifying criteria by which variances to this

requirement may be granted.

Section 1445(a)(1) of the Act requires a public water system to

``establish and maintain such records, make such reports, conduct such

monitoring, and provide such information as the Administrator may

reasonably require by regulation . . .''.

Section 1414(c) requires each owner or operator of a public water

system to give notice to persons served by it of any failure to comply

with an MCL, treatment technique, or testing procedure required by a

NPDWR and any failure to comply with any monitoring required pursuant

to section 1445 of the Act.

II. Regulatory Background

Two NPDWRs control disease-causing microorganisms (pathogens) in

public water supplies--the Total Coliform Rule (TCR)(54 FR 27544; June

29, 1989) and the Surface Water Treatment Rule (SWTR)(54 FR 27486; June

29, 1989). A third regulation, the Groundwater Disinfection Rule

(GWDR), which is currently under development, will add further

protection for systems using ground water.

The SWTR met the requirements of Section 1412(b)(7)(C) and, for

surface waters, Section 1412(b)(8) of the SDWA, as amended in 1986. The

SWTR requires all systems using surface water, or ground water under

the direct influence of surface water, to disinfect. In addition, all

such systems are required to filter their water unless they demonstrate

that they have an effective watershed protection program and meet other

EPA-specified requirements (Sec. 141.71). The watershed control program

must minimize the potential for source water contamination by Giardia

cysts and viruses, and typically includes characterization of watershed

hydrology, land ownership by the system, and activities on the

watershed that might have an adverse effect on source water quality.

The rule also requires an annual on-site inspection of all systems that

wish to avoid filtration. This inspection must demonstrate that the

required watershed control program and disinfection treatment processes

are adequately designed and maintained. The SWTR also established MCLGs

of zero for Giardia lamblia, viruses and Legionella.

The SWTR requires all systems to achieve at least 99.9% (3-log)

removal/inactivation of Giardia lamblia cysts, and 99.99% (4-log)

removal/inactivation of enteric viruses. The intention of these

provisions was to provide appropriate multiple barriers of treatment to

control pathogen occurrence in finished drinking water. This rule was

promulgated as a treatment technique rather than an MCL, because EPA

believed that routine monitoring for the pathogens was not economically

or technologically feasible. Another pathogen, Cryptosporidium, was

considered for regulation under the SWTR, but was not addressed,

because EPA lacked sufficient health, occurrence, and water treatment

control data regarding this organism at that time.

The TCR established MCLGs of zero for total coliforms, which

includes fecal coliforms and E. coli. MCLs, monitoring requirements,

and analytical requirements were promulgated for these organisms. The

TCR requires all public water systems that collect fewer than five

samples per month to have an on-site sanitary survey every five years

(ten years for some systems). The purpose of this requirement is to

help ensure the long-term quality and safety of drinking water in small

systems that cannot be accomplished by infrequent coliform monitoring.

The TCR and SWTR were promulgated to minimize both epidemic and

endemic waterborne microbial illness. The public health goal, as

described in the preamble to the SWTR, was to provide treatment to

ensure an acceptable risk of less than one waterborne microbial illness

per year per 10,000 people.

In addition to the SWTR, TCR, and GWDR, EPA is also developing a

rule that would limit concentration levels of disinfectants and the

chemical disinfection byproducts (DBPs) resulting from their use. The

use of chemical disinfectants in water treatment results in a

substantial decrease in waterborne microbial illness and is an integral

part of a multiple-barrier removal/inactivation approach. However,

disinfectants and DBPs may present potential health risks themselves.

DBPs form when disinfectants used for microbial control in drinking

water react with various organic chemicals in the source water. Some of

these are known to be toxic to humans or are considered to be probable

human carcinogens. As such, a number of disinfectants and DBPs were

included on the 1991 Drinking Water Priority List (56 FR 1470, January

14, 1991) as candidates for future regulations. To address these health

issues, EPA is proposing elsewhere in today's Federal Register the

disinfectants/disinfection byproducts (D/DBP) rule, which includes

NPDWRs for several disinfectants and disinfectant byproducts.

To develop the D/DBP Rule, EPA instituted a formal regulation

negotiation process in 1992 with potentially affected parties (57 FR

53866; Nov. 13, 1992). The committee established to negotiate the

regulation included representatives from water utilities, State and

local health and regulatory agencies, environmental groups, consumer

groups, and EPA (hereafter the Negotiating Committee or the Committee).

One of the major goals addressed by the Committee was to develop an

approach that would reduce the level of exposure from disinfectants and

DBPs without undermining the control of pathogens. The intention was to

ensure that drinking water is microbiologically safe at the limits set

for disinfectants and DBPs and that these chemicals do not pose an

unacceptable risk at these limits. The approach in developing this rule

considered the constraints of simultaneously treating water for these

different concerns. As part of this effort, the Negotiating Committee

decided that the SWTR may need to be revised to address health risk

from high densities of pathogens in poor quality source waters and from

the protozoan, Cryptosporidium. If such requirements were deemed

necessary, and could be promulgated concurrently with new D/DBP

regulations (the regulations proposed elsewhere in today's Federal

Register and termed ``Stage 1 D/DBPR''), a system could comply with

both regulations and meet the intended public health goals.

The Negotiating Committee also decided that to develop a reasonable

set of rules, including today's proposed rule, and to understand more

fully the limitations of the current SWTR, additional field data were

critical. Thus, the Committee agreed to the development of an

information collection rule (ICR) that would require, in part, systems

serving a population of 10,000 or greater to determine the density of

specific pathogens in their source water and to characterize their

treatment processes. Under the ICR, systems serving populations greater

than 100,000 would also be required to monitor for pathogens in their

finished water (depending upon the pathogen density in the source

water), and the concentration of DBPs and parameters related to their

formation at various steps in the treatment process. To this end, EPA

proposed the ICR on February 10, 1994 (59 FR 6332) that would require

this additional information. The Committee agreed to the requirements

in the proposed ICR as necessary and reasonable.

According to the regulatory strategy developed by the Committee,

systems serving a population of 10,000 or greater would use the

monitoring and treatment data collected under the ICR to decide what

additional treatment measures, if any, would be necessary to protect

the public from pathogens while controlling for DBPs. This decision

would be based on criteria specified either in amendments to the SWTR

or by guidance. Today's proposed rule includes a variety of regulatory

options, from requiring systems to provide minimum levels of treatment

based upon the density of pathogens in the source water to maintaining

the existing requirements of the SWTR.

According to the Committee's strategy, amendments to the SWTR would

be developed under two rules. The first of these rules, which is

today's proposed rule, would be an interim enhanced SWTR (ESWTR) that

would only pertain to systems serving 10,000 people or greater. Data

collected under the ICR would be used to determine the appropriate

regulatory option(s) under this rule, and then to implement it at the

time systems are required to comply with the Stage 1 D/DBP regulations.

Following the full compilation and analysis of all data collected under

the ICR rule and from other research findings, EPA would propose a

long-term ESWTR with which systems serving fewer than 10,000 people

would comply while also complying with the Stage 1 D/DBP rule. The

long-term ESWTR might also include additional refinements for systems

serving 10,000 people or greater. Today's proposal also satisfies the

provision in section 1412(b)(9) of the SDWA for review of NPDWRs every

three years; the SWTR was promulgated on June 29, 1989, and became

effective in stages, beginning December 31, 1990.

III. Discussion of Proposed Rule

A. Basis for Amending Existing SWTR: Limitations of SWTR

As discussed above, the SWTR requires all systems using surface

water, or ground water under the direct influence of surface water, to

disinfect. It also requires all such systems to filter their water

unless they can demonstrate that they have an effective watershed

control program, meet source water quality criteria, achieve minimum

disinfection requirements, and have no evidence of reported waterborne

disease among the served population. The SWTR also specifies that

systems using surface water must treat water to remove/inactivate at

least 99.9% of the Giardia lamblia cysts and at least 99.99% of the

enteric viruses, regardless of their densities in the source waters.

The SWTR does not require a system to monitor its source water or

drinking water for these pathogens. At the time of promulgation, EPA

recognized a variety of uncertainties and unknowns regarding potential

health risks, but these were not possible to address at that time.

Subsequently, additional information has become available that

indicated possible deficiencies in the SWTR. Some of these deficiencies

are described below.

SWTR Did Not Address Cryptosporidium

During the development of the SWTR, the United States experienced

its first recognized waterborne disease outbreak of cryptosporidiosis,

caused by the protozoan, Cryptosporidium (D'Antonio et al., 1985).

Other outbreaks caused by this pathogen have since been reported both

in the United States and other countries (Smith et al., 1988; Hayes et

al., 1989; Levine and Craun, 1990; Moore et al., 1993; Craun, 1993).

Because of the lack of data before 1989 on Cryptosporidium oocyst

occurrence and removal by treatment, EPA decided to regulate this

pathogen in a future rulemaking, rather than to delay publication of

the SWTR until this data was available. Thus, the SWTR does not now

specifically address Cryptosporidium treatment removal/inactivation

requirements, watershed control requirements for Cryptosporidium for

systems that wish to avoid filtration, or a definition of ground water

under the influence of surface water that includes Cryptosporidium.

Moreover, the assumptions about Giardia reduction under the SWTR may

not be applicable to Cryptosporidium, which, based on laboratory

studies, is much more resistant to common disinfection practices than

is Giardia (Korich et al., 1990; Korich et al., 1992). Since

publication of the SWTR in 1989, some information on Cryptosporidium

occurrence and control measures has been published. EPA will have new

data available shortly from systems that monitor for this organism

under the ICR and from research currently being carried out by the

Agency and the water industry. As a result, EPA believes that it will

soon be in a better position to develop a suitable regulation for

Cryptosporidium.

Specified Pathogen Reductions May Be Inadequate

The 3-log removal/inactivation of Giardia and 4-log removal/

inactivation of enteric viruses required by the SWTR were developed to

provide adequate protection from pathogens in average quality source

water, and thus may be inadequate when a system is supplied by poorer

quality source water with high levels of these or other pathogens. In

developing the SWTR, EPA assumed, on the basis of data available at

that time, that this level of treatment was adequate for the vast

majority of systems.

Additionally, risk assessments for the pathogens of concern had

high degrees of uncertainty, such that the risks associated with a

given level of pathogen contamination were unclear. Moreover, methods

for quantifying these organisms were not generally available.

Therefore, the Agency believed that a simple, yet conservative

treatment requirement was most appropriate. However, it was apparent

during the development of the SWTR that the level of treatment being

specified might not always be adequate. Therefore, the Agency published

associated guidance recommending greater treatment for systems supplied

by poorer quality source waters (EPA, 1991a).

Subsequent data on Giardia and virus densities in source water and

drinking water, however, bring into question the assumption that the

treatment specified in the SWTR was adequate for most systems. These

new data suggest that the concentrations of Giardia cysts and viruses

in the source waters of many systems may be too great for the specified

level of treatment to adequately control waterborne pathogens. For

example, LeChevallier et al. (1991a,b) examined Giardia and

Cryptosporidium levels in the source waters and filtered drinking

waters of 66 surface water systems in 14 States and one Canadian

province. They detected at least one of these two organisms in 97% of

the raw water samples. Giardia densities ranged from 0.04 to 66 cysts/L

(geometric mean of 2.77 cysts/L), while Cryptosporidium densities

ranged from 0.07 to 484 oocysts/L (geometric mean of 2.70 oocysts/L).

These investigators also detected at least one of the two organisms

in the drinking water of 39% of the systems. For those drinking waters

that were positive, Giardia densities ranged from 0.29-64 cysts/100L

(geometric mean 4.45 cysts/100L), while Cryptosporidium densities

ranged from 0.13 to 48 oocysts/100L (geometric mean 1.52 oocysts/100L).

According to the investigators, 78% of the systems that were positive

for Giardia or Cryptosporidium met the turbidity standards specified by

the SWTR. Based on a risk assessment model developed for Giardia, 24%

of the 66 systems might not meet the health goal in the SWTR of no more

than the one Giardia infection annually per 10,000 people per year

(LeChavallier et al., 1991b). (This incidence of infection is a

conservative estimate of illness, since not all infected people become

ill.) This study suggests that the SWTR may need to be revised if an

annual 10-4 risk level, or some other desired risk level, is to be

achieved by all systems in the United States.

EPA used the data in LeChevallier et al. (1991a,b) to calculate the

percentage of systems that use source waters containing various

densities of Giardia cysts. The Agency calculated that about 85% of the

source waters in the study contained 10 cysts/100L or more, while about

45% contained 100 cysts/100L or more. Many of these systems currently

provide four, five, or even six or more logs of removal/inactivation

and therefore are able to achieve EPA's 10-4 annual risk goal.

However, if such systems were to reduce existing levels of disinfection

to more easily meet new D/DBP regulations, and only marginally meet the

three-log removal/inactivation requirement for Giardia specified in the

current SWTR, they could experience significant increases in microbial

risk (Regli et al., 1993; Grubbs et al., 1992; EPA, 1994)).

An epidemiology study by Payment et al. (1991) also suggests that

the pathogen density reductions specified by the SWTR may not be

sufficient for adequate protection. The goal of this study was to

determine the extent to which drinking water caused gastrointestinal

disease in a community served by poor quality source water that was

subjected to full conventional treatment. In this study, the

investigators carried out a trial where 299 households in a community

drank water from a reverse-osmosis water filter, while 307 households

used the usual tapwater. According to the data, 35% of the reported

gastrointestinal illness was associated with the drinking water. The

etiologic agent(s) were not identified, but a plausible explanation is

that pathogens were in the finished water. A recent analysis by Haas et

al. (1993) also suggests that high levels of microbial risk far above

the health goal of the existing SWTR may be occurring in systems with

highly contaminated source waters that may only minimally comply with

the SWTR.

Several other recent studies have shown that Giardia and

Cryptosporidium cysts/oocysts can be found in filtered drinking waters

in systems served by highly contaminated source waters (Clancy, 1993;

EPA, 1993). If treatment is inadequate in reducing pathogens to an

acceptable level, EPA must consider revising and strengthening

treatment requirements. A mitigating factor is that, based upon a

microscopic examination of the cysts/oocysts detected by LeChevallier

et al. (1991b), most of the cysts/oocysts in the drinking water may not

be viable. This observation, however, has not yet been confirmed.

Virus CT Values May Be Greater Than Assumed by SWTR

The SWTR assumes that disinfection more readily controls viruses

than may actually be the case. The Guidance Manual to the SWTR (EPA,

1991a) identifies the disinfection CT values (disinfectant

concentration times the contact time) for viruses. These data are based

on laboratory studies in which a dispersed suspension (i.e., non-cell

associated, non-aggregated) of hepatitis A virus was used. These CT

values relative to the much higher CT values needed for Giardia

inactivation for systems to comply with the SWTR have led to the

assumption by some that systems which satisfactorily control for

Giardia cysts will adequately control for pathogenic viruses. (The CT

values to comply with the level of disinfection inactivation

requirements for viruses in the Guidance Manual to the SWTR are one to

two orders of magnitude below the CT values necessary to achieve the

inactivation requirements for Giardia.)

However this assumption may not always be valid. In environmental

waters, viruses are usually aggregated or associated with cell debris,

some of which may not be removed entirely by filtration processes. Such

cell-associated aggregates are considerably more resistant to

disinfection than free viruses (Williams, 1985; Sobsey, 1991).

Moreover, some pathogenic enteric viruses may be substantially more

resistant to disinfection than hepatitis A (Keswick et al., 1985). In

addition, laboratory studies to determine CT values for viruses, even

with applied uncertainty factors, may underestimate the actual CT

values necessary to achieve the desired level of inactivation, since

viruses in the environment may be hardier and less susceptible to

disinfection.

The detection of viruses in fully treated waters (i.e., after

coagulation, sedimentation, filtration, and disinfection) (Gerba and

Rose, 1990; Payment, 1985; Hurst, 1991) also suggests that viruses in

environmental sources have greater CT values than those published in

the Guidance Manual. Hurst (1991), for example, summarized the

published data on viruses in drinking water, and found that the

percentage of samples positive for viruses ranged from 0 to 100%. In

one study, Payment et al. (1985) detected enteroviruses in 7% of

finished water samples (1,000 L samples from 7 systems), with an

average density of 0.0006 most probable number of cytopathogenic units.

In another study, Payment (1981) detected 1-10 enteroviruses/100L in

most drinking water samples in a system using poor quality source

water.

The above data also suggests that EPA needs to reassess the 4-log

level of treatment required for viruses under the SWTR. Under this

requirement, a system may only provide a 2-log inactivation of viruses

by disinfection and still meet the 4-log overall treatment requirement

(under current EPA guidance, systems using conventional treatment are

assumed to achieve a 2- log removal of viruses by clarification

processes alone). For some systems, virus densities in surface waters

may be sufficiently high to warrant at least a 4-log or greater level

of inactivation by disinfection alone (and a 6-log or greater removal

of viruses with clarification and disinfection) to achieve desired risk

levels (Regli et al., 1991). The Agency would like to determine what

minimum level of disinfection inactivation is necessary for surface

water supplies to ensure adequate virus control, regardless of Giardia

densities. EPA intends to use data from the ICR to: (1) Help clarify

the adequacy of using Giardia as a target organism to control for

viruses in systems with different source water qualities, (2) determine

what assumptions can be made regarding quantification of virus removal

for different treatment processes and disinfection conditions, and (3)

determine what, if any, changes to the SWTR are needed to control

pathogenic viruses.

DBP Rule May Undermine Pathogen Control

Some systems currently exceed the required Giardia and virus

reductions specified by the SWTR. The DBP Rule may potentially

undermine pathogen control in these systems by prompting them to reduce

DBP concentrations at the expense of pathogen control (e.g., by

shifting point of disinfection to later in the treatment chain,

reducing disinfection dose, or switching to a weaker disinfectant such

as chloramines). These systems would still have to comply with the

removal/inactivation criteria in the current SWTR, but would do so with

a lower margin of safety. This situation might result in a substantial

increase in waterborne illness for systems using a poor quality source

water. For example, according to a model developed by EPA (Regli et

al., 1993), a reduction of the MCL for total trihalomethanes (TTHMs)

(one of the toxic byproducts) from 100g/L to 75g/L

could increase the incidence of waterborne giardiasis in some systems

by as many as 10,000 per million people per year, if the existing SWTR

is not amended to require higher levels of treatment for poor quality

source waters.

This situation is further evidence that EPA may need to revise the

SWTR to ensure that measures taken by systems to comply with the

forthcoming DBP Rule do not increase the health risk from pathogens.

B. General Approach for Revising SWTR

Under the negotiated rulemaking, the Negotiating Committee agreed

to propose three rules: (1) ICR, proposed on February 10, 1994 (59 FR

6332), (2) D/DBP regulations (proposed in today's Federal Register),

and (3) ESWTR. EPA is planning to remedy the shortcomings of the SWTR

indicated above through two sequential stages--an interim ESWTR and a

long-term ESWTR. Today's rule proposes the interim ESWTR. The Agency

and the Negotiating Committee decided that this phased approach was

appropriate because of the uncertainties associated with lack of data.

EPA needs much more data on the concentrations of Giardia,

Cryptosporidium, and enteric viruses for various qualities of source

waters, with variations over time and season, to determine the need for

additional treatment. Some members of the Negotiating Committee

believed that health effects information, especially dose response data

for pathogens of concern, is also important to ensure that EPA selects

the most appropriate control option. In addition, EPA needs more field

data on the effectiveness of different types of water treatment for

controlling these pathogens. Data from the ICR and various research

studies would provide much of this information, sufficient in EPA's

view to refine the present proposed interim rule. Additional work would

culminate in a long-term ESWTR that would be protective for all surface

water system sizes (including those that serve fewer than 10,000

people) and would also include possible refinements to any interim

requirements for larger systems. EPA believes that the interim and

long-term ESWTR rules are essential for providing adequate human health

protection; however, some members of the Negotiating Committee believed

that the most appropriate regulatory criteria to provide this

protection are not yet apparent.

Schedule of Regulations

Table I-1 indicates the schedule agreed to by the Negotiating

Committee for proposing, promulgating, and implementing these rules.

Implementation dates for the ICR are indicated under the columns of the

Stage 2 D/DBP rule and ESWTR to reflect the relationship between these

rules. Although the schedule for proposing these rules has slipped

slightly, EPA believes the scheduled promulgation dates for the ESWTR

and D/DBP Rule can still be met.

The Negotiating Committee believes that the December 1996 scheduled

date for promulgating the Stage 1 D/DBP Rule reflects the shortest time

possible by which the interim ESWTR, if necessary, could also be

promulgated. EPA is proposing that the Stage 1 D/DBP regulations and

the interim ESWTR become effective on the same date of June 30, 1998,

for those surface water systems, or ground water systems under the

direct influence of surface water, serving 10,000 people or more. This

strategy is necessary so that systems do not degrade pathogen control

in attempting to comply with the Stage 1 D/DBP regulations.

Table I-1.--Proposed D/DBP, ESWTR, ICR Rule Development Schedule

------------------------------------------------------------------------

Time

line State 1 DBP rule Stage 2 DBP rule ESWTR

------------------------------------------------------------------------

12/93.. .................... Propose information Propose information

collection collection

requirements for requirements for

systems>100k. systems >10k.

3/94... Propose required Propose Stage 2. Propose interim

enhanced MCLs for TTHMs (40 ESWTR for

coagulation for g/l), HAA5 systems>10k.

systems with (30 g/l),

conventional BAT is precursor

treatment. MCLs- removal with

TTHMs (80g/ chlorination.

l), HAA5 (60ug/l),

bromate, chlorite.

Disinfectant limits.

6/94... .................... Promulgate ICR...... Promulgate ICR.

8/94... Close of public .................... Public comment

comment period. period for

proposed ESWTR

closes.

10/94.. .................... Systems>100k begin Systems>100k begin

ICR monitoring. ICR monitoring.

1/95... .................... .................... Systems 10-100k

begin source water

monitoring.

10/95.. .................... SW systems>100k, GW

systems>50k begin

bench/pilot studies

unless source water

quality criteria

met.

11/95.. .................... .................... NOA for monitoring

data, direction of

interim ESWTR.

1/96... .................... .................... Systems>10k

complete ICR

monitoring. End

NOA public comment

period.

3/96... .................... Systems complete ICR Systems>100k

monitoring. complete ICR

monitoring.

12/96.. Promulgate Stage 1.. .................... Promulgate interim

ESWTR for

systems>10k.

6/97... .................... Notice of Propose long-term

availability for ESWTR for

Stage 2 reproposal. systems10k.

10/97.. .................... Complete and submit

results of bench/

pilot studies.

12/97.. .................... Initiate reproposal--

begin with 3/94

proposal.

6/98... Effective. Effective Close of public Interim ESWTR

for SW systems comment period. effective for

serving systems>10k. 1994-

greater>10k, 6 monitoring data

extended compliance used to determine

date for GAC or treatment level.

membrane technology.

12/98.. .................... Propose for CWSs, Publish long-term

NTNCWSs. ESWTR.

6/00... Stage 1 limits Promulgate Stage 2 Long-term ESWTR

effective for for all CWSs, effective for all

surface water NTNCWSs. system sizes.

systems10k.

1/02... Stage 1 limits Stage 2 effective,

effective for GW compliance for GAC/

systemsm=log-1 (1/n x [log X1+log X2+...log Xn]),

Where n = number of samples and Xi is the measured density for

each sample. For example, the geometric mean of the values 1, 10, and

100 would be 10. The geometric mean is more appropriate than the

arithmetic mean for representing the central tendency for data that

have a skewed distribution. However, the geometric mean is less

conservative, i.e., it would generally estimate a lower mean density

and therefore lower risk for pathogens than the arithmetic mean (for

example, the arithmetic mean of 1, 10, and 100 is 37, versus the

geometric mean of 10). Nevertheless, depending upon the assumptions

made in the risk assessment calculation (e.g., percentage of cysts/

oocysts viable), use of the geometric mean may be adequately

conservative for estimating exposures and consequently appropriate

levels of treatment (Regli et al., 1991).

c. Use of the 90th percentile value. Another alternative for

defining pathogen density is to base this value on the 90th percentile

of all data for a particular pathogen. This is the value below which

fall 90% of the data points and above which fall 10% of the data

points. This approach is more conservative in terms of risk than the

arithmetic mean and geometric mean, because for sources where pathogen

density varies significantly throughout the year, use of this value

will be more representative of the elevated risk associated with peak

contamination periods.

Use of the 90th percentile measured value, however, has the obvious

drawback that it requires a sufficient number of samples to provide a

good 90th percentile estimate without interpolation. In many cases,

particularly for small water systems, cost considerations will prevent

extensive sampling. For example, the proposed ICR would require only

six raw water samples over the period of a year for systems from 10,000

to 100,000 people served. The 90th percentile value could be

interpolated from the two highest values.

d. Use of the maximum measured value. This approach would dictate

the use of the highest density measured under the ICR for raw water.

Since few systems have the resources for routine, frequent, and long-

term sampling for pathogens such as Giardia, Cryptosporidium and

viruses, it is clear that episodic periods of microbial contamination

may escape detection. EPA is particularly concerned with the risks from

unusually high level contamination events that might exceed the

removal/inactivation capacity of a treatment system. While the maximum

measured value may not be representative of the normal pathogen density

in a source water, it would be more indicative of potential short-term

risks.

Additionally, since the published dose-response values for Giardia

and other pathogens were developed in healthy adult populations and

therefore are not conservative with respect to protecting public

health, EPA might select use of the maximum value, which is the most

conservative statistical option, to offset this problem. Alternatively,

it may be more appropriate to use a less conservative method for

estimating microbial densities but to use more conservative criteria

for deriving the actual level of treatment requirements as they relate

to pathogen densities. For example, if EPA assumes that all Giardia

cysts detected are viable and infectious to humans in specifying the

level of treatment needed, this approach may be sufficiently

conservative to warrant the density calculation by one of the other

above described methods.

A major problem with basing the density calculation on the maximum

value is that if a utility collects more than the minimum number of

samples required in the interest of better defining potential

exposures, it has a greater likelihood of collecting a sample with a

higher pathogen density than would occur with the minimum required

number of samples. In this case, the system may face a more stringent

(and thus more expensive) standard.

Use of the maximum measured density may be more appropriate than

other statistical methods for systems that have not collected

sufficient data to allow calculation of an adequately representative

mean value or 90th percentile value. With such limited data, the

maximum value might be suitable for determining level of treatment.

EPA is soliciting comment on which approach is most appropriate for

defining pathogen density. The Agency is also requesting comment on

whether the approach used should be based on the number of pathogen

samples collected, i.e., the maximum measured value would be required

for systems taking only six samples under the ICR (systems serving

between 10,000 and 100,000 people) and 90th percentile value for

systems that collect at least 10 samples.

2. Treatment alternatives for controlling pathogens. To determine

what regulatory controls are most appropriate for controlling pathogens

in drinking water, EPA must decide what constitutes acceptably safe

drinking water. The SDWA frames this discussion in determining MCLGs

and MCLs. MCLG levels, which are not legally enforceable, are based

solely on health concerns. As required by the SDWA, they are set ``at

the level at which no known or anticipated adverse effects on the

health of persons occur and which allows an adequate margin of

safety''. The corresponding enforceable regulation consists of either

an MCL set as close to the MCLG as feasible, taking cost and

availability of treatment into account; or, when it is not

technologically or economically feasible to monitor for the

contaminants, a treatment technique(s) to achieve an acceptable risk.

The SWTR promulgated an MCLG for Giardia of zero, i.e., no Giardia

cysts should be allowed in drinking water. A system using surface water

cannot usually attain this goal in any practical sense. Therefore, the

SWTR preamble suggested a more practical health goal for Giardia:

drinking water should not cause more than one Giardia lamblia infection

annually per 10,000 exposed persons (10-4 risk). In contrast to

this goal, EPA policy for specific chemical carcinogens is for

theoretical lifetime upper bound risks to be no greater than within a

range of 10-4 to 10-6. For non-carcinogenic chemical

contaminants, EPA policy is to base the MCLG on the reference dose

(RfD) for the given chemical. The RfD is calculated to be below any

known level of exposure resulting in adverse health effects, so that

drinking water at the resulting MCLG over a lifetime should be without

known risk.

In developing the D/DBP rule (proposed elsewhere in today's Federal

Register), EPA is attempting to ensure that drinking water remains

microbiologically safe at the limits set for disinfectants and

byproducts, and that the disinfectants and byproducts themselves do not

pose an unacceptable risk at these limits.

Based on data on microbial illness and death in the U.S. compiled

by Bennett et al. (1987), the estimated annual risk of waterborne

illness during 1985 was about 4 x 10-3 and the estimated lifetime

risk of death was about 3 x 10-4. As stated above, the goal of the

SWTR was for systems to achieve a risk of less than 10-4

infections per year for Giardia. Because Giardia is relatively

difficult to inactivate compared to virus and bacterial pathogens, the

SWTR assumed that water treatment adequate to achieve a 10-4 risk

for Giardia would provide an even higher level of protection against

pathogenic viruses and bacteria in untreated surface waters. Applying

the Bennett et al. (1987) data regarding the ratio of mortality to

waterborne illness (0.1 percent = 10-3), if a system achieves an

incidence of 10-4 waterborne infections per year or less, the

associated lifetime risk of death would be less than 7 x 10-6.

This is a 40-fold decrease in risk relative to those estimated by

Bennett et al. (1987), who used 1985 data.

The above calculations refer to the average individual and an

average pathogenic organism. Available dose-response data show that the

risk of infection for a given pathogen density in the consumed water

ranges over several orders of magnitude for different organisms (Regli

et al., 1991). Setting a generic microbiological drinking water

standard based on one dose-response curve will either overestimate or

underestimate risks from other organisms. The risk of death from

infection likewise varies widely with the organism (Bennett et al.,

1987). Therefore, the severity of the illness associated with a given

organism must be considered.

Additional considerations in assessing acceptable waterborne

microbial risks involve the human subpopulations sensitive to

infection, illness and death. Infection does not always result in

illness; many infections are asymptomatic (Rendtorff, 1954; Lopez et

al., 1980). The progression and severity of illness following microbial

infection are more a function of individual physiology than the

magnitude of dose, as is true for many toxic chemicals. Acute

gastrointestinal illness, the most common microbial illness, is

generally considered non-life threatening in normally healthy adults.

However, this is not necessarily true for those subpopulations that are

more sensitive to microbial infection or illness. Some studies (Glass

et al., 1991; Lew et al., 1991) indicate that infants and those over 70

years old have mortalities of 3-5 percent from diarrhea requiring

hospitalization. As discussed above, Cryptosporidium infections, mild

in healthy persons, are sometimes fatal to the immuno-compromised.

Other identified sensitive human subpopulations include pregnant women

and those with cardiovascular disease. EPA estimates that about 15% of

the U.S. population is in these higher risk groups.

Prudent health policy would be to protect these groups from their

higher risks of waterborne microbial infection. Use of a one percent

mortality-to-illness rate (instead of a 0.1 percent) to represent more

deadly organisms and a 10-fold uncertainty factor (as used in EPA's RfD

calculations) to account for these sensitive subpopulations may be

appropriate for estimating potential risks resulting from systems

achieving regulatory goals. A risk calculation based on this approach,

assuming that the system achieves the risk goal of 10-4 annual

infections for the average population, might result in a 7 x 10-5

(i.e., 10-4 x 10-2 x 70 years) lifetime risk of death for

certain subpopulations. The 7 x 10-5 lifetime risk of death (which

is a more severe endpoint than cancer) is barely within the 10-4-

10-6 guideline for excess lifetime cancer risk that EPA uses for

regulating chemical carcinogens in drinking water.

These calculations, while based on estimates and approximations and

having large uncertainties, suggest that the risk level of 10-4

annual infections may be acceptable, albeit barely so. If EPA were to

accept a more stringent annual risk level of 10-5 or 10-6

infections to achieve a greater consistency between lifetime mortality

risks from waterborne pathogens and most regulated drinking water

chemicals, substantial increases in treatment might be required. EPA

solicits comment on the appropriateness and magnitude of specific

acceptable risk levels for microbial infection and illness.

To counter waterborne illness, EPA is proposing five treatment

alternatives for controlling Giardia, Cryptosporidium, and/or viruses.

Within each alternative, several options are addressed. The Agency may

promulgate one or more of these alternatives. Alternative A addresses

enhanced treatment for Giardia only. Alternatives B and C address

treatment for Cryptosporidium only. Alternative D addresses enhanced

treatment for viruses only. Alternative E maintains existing level of

treatment requirements for Giardia and viruses. EPA requests comment on

what alternative(s) is most suitable.

a. Alternative A. Enhanced treatment for Giardia. This alternative

bases the extent of treatment required on the Giardia density in the

source water. The SWTR currently requires a 99.9 percent (3-log)

removal/inactivation of Giardia for all surface waters, regardless of

Giardia cyst concentration in the source water. As discussed above and

in the SWTR, EPA believes that for source waters of high quality (low

pathogen densities), this level of treatment should result in less than

one case of giardiasis (and most other waterborne disease) per 10,000

people per year. This risk level for Giardia is associated with an

infectious Giardia cyst density in the source water of less than one

cyst/100 liters and assumes that 3 logs of removal/inactivation is

consistently achieved on such source water. For more information about

Giardia risk calculations and associated uncertainties and assumptions,

refer to Rose et al. (1991), Regli et al. (1991), and Macler and Regli

(1993).

Under Alternative A, systems using source waters with higher

Giardia densities would be required to meet higher levels of treatment

to satisfy the desired acceptable risk level, e.g., the annual

10-4 risk or perhaps a more stringent goal. Specifically, under

one option of alternative A, EPA is proposing that systems meet the

level of treatment for Giardia associated with the following Giardia

concentrations in the source water to achieve a 10-4 annual risk

level:

------------------------------------------------------------------------

Required treatment

No. of giardia/100L level (percent)

------------------------------------------------------------------------

99.............................................. 99.9999 (6-log).

------------------------------------------------------------------------

The determination by utilities and States of removal and

inactivation efficiencies for specific treatment strategies would be

based on EPA guidance and information, as is currently done under the

existing SWTR. EPA would revise the existing SWTR Guidance Manual based

on data collected under the ICR and research to complement any criteria

promulgated under the ESWTR. The Agency expects that data collected

under the ICR will be used by States and utilities to define the source

water concentration and consequently the appropriate level of treatment

for individual systems. If a utility has not collected data on pathogen

densities in source water under the ICR, it would be required to do so

to define the appropriate level of treatment.

Depending on the method used for calculating pathogen density,

assumptions used for estimating risk (e.g., whether to assume that all

or only a portion of the detected cysts in the source water are viable

and infectious to humans), the desired acceptable risk level, concern

about DBP risk, and the technical and economic feasibility of achieving

different levels of treatment, it may be appropriate to specify

treatment requirements that address higher source water pathogen

concentrations than described above. EPA is not aware to what extent

physical removal greater than 2.5 logs can reasonably be achieved by

systems using conventional water treatment approaches commonly

practiced in the United States. The Agency believes that systems that

optimize their treatment may be able to achieve substantially higher

levels of removal. Membrane filtration techniques, although promising

for much higher levels of removal, may not yet be technologically or

economically feasible for large numbers of systems. The balance of the

removal/inactivation requirement may have to rely on the use of

chemical disinfectants. However, while EPA has confidence in the use of

disinfectants to achieve current SWTR requirements, it has not been

demonstrated that CT values extrapolated from tables in the SWTR or

other sources are valid for higher levels of disinfection (e.g., to

achieve a 4- or 5-log reduction of Giardia). EPA requests comment on

approaches to achieve higher levels of treatment by physical means and

on the use of existing CT values in the EPA guidance (EPA, 1991) to

predict, by extrapolation, higher levels of inactivation that could be

achieved by disinfection.

EPA is considering an alternative version of the above described

treatment requirements that would instead require greater Giardia

reductions for source waters beginning with Giardia concentrations of

10 or more cysts/100 liters, as indicated below:

------------------------------------------------------------------------

Required treatment

No. giardia/100L level (percent)

------------------------------------------------------------------------

10-99............................................ 99.99 (4-log).

100-999.......................................... 99.999 (5-log).

>1000............................................ 99.9999 (6-log).

------------------------------------------------------------------------

EPA solicits comment on the two treatment options described above

and on associated variations.

b. Alternative B. Specific treatment for Cryptosporidium. EPA is

proposing a treatment technique rather than an MCL for Cryptosporidium,

because EPA believes that it is not currently economically or

technologically feasible for a system to monitor for this organism in

the finished water to determine whether it meets an acceptable risk

level. The Agency bases its belief on three factors: (1) The

variability of Cryptosporidium spatially and temporally may be

considerable, and consequently systems would have to collect frequent

samples and inordinately large sample volumes to properly characterize

the density of this organism, (2) current methods for Cryptosporidium

analysis are difficult and expensive, (3) it is not yet possible to

predict the risk resulting from a specific level of exposure to

Cryptosporidium, and (4) even if Cryptosporidium could be detected at

the lowest concentrations of concern in the finished water, the

exposure and associated risk would have already occurred, thereby

reducing the significance of monitoring non-compliance.

Under this rule, all community and non-community public water

systems using any surface water source, or groundwater under the direct

influence of surface water, would be required to treat these sources as

described below. EPA anticipates that human dose-response data for

Cryptosporidium will be available within the next year and will include

these data in a Notice of Availability, probably in March 1996. Because

they are not yet available, basing the treatment level on a specific

acceptable risk level, as proposed by EPA for Giardia, cannot be used

for Cryptosporidium in the present notice. Data collected to date

suggest that the dose-response for Cryptosporidium may be similar to

that for Giardia. If this is true, then the required reduction level

for Cryptosporidium may be the same as for Giardia to achieve an

equivalent risk level for similar source water densities.

In the absence of dose-response data, EPA is proposing a wide

variety of options. One sub-alternative would be to base the level of

treatment on the Cryptosporidium densities found in the source water,

as presented in the Table below.

------------------------------------------------------------------------

Required treatment

No. cryptosporidium/100L level (percent)

------------------------------------------------------------------------

99.............................................. 99.9999 (6-log).

------------------------------------------------------------------------

EPA is concerned, however, that it may not be technologically or

economically feasible to achieve the treatment levels above, given that

Cryptosporidium is much more resistant to disinfection than is Giardia.

Conventional treatment of coagulation, sedimentation and filtration may

not reliably achieve more than 2.5-log or 3-log Cryptosporidium oocyst

reduction under typical operating conditions. While membrane filtration

technologies (ultrafiltration, nanofiltration), possibly following

conventional treatment processes, appear to promise considerably

greater reductions in Cryptosporidium densities, their widespread use

for this purpose raises other concerns such as waste disposal of the

concentrate, water wastage, potential failure of the membrane, and

significant costs. Unless systems can feasibly achieve higher removal

levels for Cryptosporidium by physical means, they would have to

achieve this additional reduction by the use of disinfectants. However,

uncertainties exist with respect to disinfection of Cryptosporidium.

Current data suggests that chlorine and chlorine-based disinfectants

are relatively ineffective in inactivating Cryptosporidium, and the

Agency is not certain if alternative disinfectants, such as ozone, are

more effective than chlorine to allow systems to comply with the

removal/inactivation levels above.

With this in mind, EPA is also considering two other treatment sub-

alternatives for Cryptosporidium, as follows:

------------------------------------------------------------------------

Required treatment

No. cryptosporidium/100L level (percent)

------------------------------------------------------------------------

99.............................................. 99.999 (5-log).

------------------------------------------------------------------------

------------------------------------------------------------------------

Required treatment

No. cryptosporidium/100L level (percent)

------------------------------------------------------------------------

99.............................................. 99.99 (4-log).

------------------------------------------------------------------------

EPA requests comment on all treatment alternatives discussed above

for Cryptosporidium.

c. Alternative C. 99% (2-log) removal of Cryptosporidium. Under

this alternative, EPA would require systems to achieve at least 99% (2-

log) removal of Cryptosporidium by filtration (with pretreatment)

alone. This alternative is based on the premise that the 3-log removal/

inactivation requirement specified for Giardia is not economically or

technologically feasible for Cryptosporidium, since laboratory data

suggests that Cryptosporidium is considerably more resistant to

disinfection than is Giardia. In addition, it may not be practical to

remove more than two logs of Cryptosporidium consistently by

clarification and filtration processes. EPA believes, however, that a

two-log removal of Cryptosporidium is feasible using current

conventional treatment methods of coagulation, sedimentation and

filtration, as specified under the SWTR.

Under this treatment option, EPA would continue to assess new field

and laboratory data to control Cryptosporidium by physical removal and

disinfection. If these data indicate that proportionally higher levels

of Cryptosporidium removal/inactivation can be achieved at a reasonable

cost, then EPA would revise the ESWTR accordingly as part of the long-

term ESWTR regulatory development. The Agency would also revise the

SWTR Guidance Manual to suggest approaches for improving system design

and operations for controlling Cryptosporidium. When sufficient human

dose-response information for Cryptosporidium becomes available to

allow calculation of drinking water health risks from this organism,

EPA will consider a risk-based approach to establishing adequate

treatment levels.

EPA solicits comment on whether a higher minimum removal

requirement than two logs should be specified for Cryptosporidium under

this alternative. The Agency also requests comment on whether the

removal requirement should be increased if treatment were to include

disinfection.

d. Alternative D. Specific disinfection treatment for viruses. The

SWTR required systems to achieve a four-log reduction/ inactivation of

viruses. This is to be achieved through a combination of filtration and

disinfection or, for systems not required to filter their source

waters, by disinfection alone. Viruses are of particular concern, given

that one or several virus particles may be infectious (Regli et al.,

1991) and that several enteric viruses are associated with relatively

high mortality rates (Bennett et al., 1987). Failure or impairment of

filtration performance could allow substantial pathogen contamination

of drinking water, particularly if the disinfection barrier following

filtration is minimal.

The SWTR considered Giardia to be a surrogate for viruses, and

assumed that if viruses were present in the source water, treatment

requirements adequate to reduce Giardia by three logs would also reduce

viruses to safe levels. This assumption may not be appropriate if a

system were to achieve a 3-log removal of Giardia by physical means and

provide little disinfection inactivation. Viruses may be present in

substantial numbers even in the absence of detectable Giardia cysts.

Treatment designed to minimize Giardia may not be optimal for

viruses. Viruses are substantially smaller than Giardia cysts or

Cryptosporidium oocysts and may pass through certain filter media that

will remove the larger protozoa. Therefore, use of data on Giardia,

Cryptosporidium, or even coliform bacteria (intermediate in size

between viruses and protozoa) in assessing treatment efficacy may not

be adequate for virus control. Studies by Payment et al. (1991) showed

that conventionally treated water meeting current Canadian microbial

drinking water advisory levels still led to substantial illness in the

studied population. These authors suggested that much of this illness

could have resulted from viral infection.

For the above reasons, particularly for strengthening the treatment

barrier by disinfection, EPA is proposing to require that systems

provide sufficient disinfection such that by disinfection alone it

would achieve at least a 0.5-log inactivation of Giardia or,

alternatively, a 4-log inactivation of viruses. This requirement would

be independent of the level of physical removal, e.g., if filtration

was able to remove three logs of Giardia, the system would still have

to provide at least an additional 0.5-log inactivation of Giardia or 4-

log inactivation of viruses by disinfection. Therefore, this would mean

that the system would provide 6 logs of virus removal/inactivation,

assuming it is removing 2-logs of viruses by filtration alone. EPA

would provide guidance to indicate the appropriate CT values to use

with these two alternatives.

The SWTR assumed that a 0.5-log inactivation of Giardia would

result in a 4-log inactivation of viruses. This assumption was based on

a study where the effect of free chlorine on the hepatitis A virus was

examined (Sobsey, 1991). Subsequent investigations, however, have

suggested that some viruses, such as the Norwalk agent, are

substantially more resistant to disinfection by chlorine than is the

hepatitis A agent. Additionally, use of disinfectants other than free

chlorine to achieve the 0.5-log inactivation of Giardia may not yield a

4-log inactivation of viruses. Therefore, a requirement to provide

sufficient disinfection to inactivate 4 logs of viruses may be more

conservative than the alternative requirement of providing sufficient

disinfection to inactivate 0.5 logs of Giardia.

Either of these two approaches could result in several additional

logs of pathogen removal/inactivation for systems that practice

conventional treatment. For example, where the system can remove by

physical means at least 2-logs of viruses, the disinfection requirement

would yield a total 6-log removal/inactivation of viruses (i.e, 2 logs

by physical means and 4 logs by disinfection).

e. Alternative E. No change to existing SWTR treatment requirements

for Giardia and viruses. Under this alternative, the existing SWTR

requirements for treatment for Giardia and viruses would not change.

For Cryptosporidium control, EPA could either regulate this organism

directly (e.g., Alternative C above) or make a finding that

Cryptosporidium is adequately controlled by filtration and disinfection

requirements in the existing SWTR. The Agency may choose this

alternative to allow the Agency time to fully develop analyses of the

ICR data and accumulate additional data on pathogen occurrence,

treatment performance, and health effects, given the view that the

current SWTR has not been in effect long enough to evaluate the

projected improvements in drinking water quality and resulting public

health benefits. EPA would consider additional regulatory alternatives

while developing the long-term ESWTR, based on this new data. The

Agency requests comment on this alternative, as well.

IV. State Implementation

This section describes the regulations and other procedures and

policies States would have to adopt, or have in place, to implement the

rule proposed today. States must continue to meet all other conditions

of primacy in 40 CFR Part 142.

Section 1413 of the SDWA establishes requirements that a State must

meet to maintain primary enforcement responsibility (primacy) for its

public water systems. These include (1) adopting drinking water

regulations that are no less stringent than Federal NPDWRs in effect

under sections 1412(a) and 1412(b) of the Act, (2) adopting and

implementing adequate procedures for enforcement, (3) keeping records

and making reports available on its activities that EPA requires by

regulation, (4) issuing variances and exemptions (if allowed by the

State) under conditions no less stringent than allowed by sections 1415

and 1416, and (5) adopting and being capable of implementing an

adequate plan for the provision of safe drinking water under emergency

situations.

40 CFR Part 142 sets out the specific program implementation

requirements for States to obtain primacy for the public water supply

supervision (PWSS) program, as authorized under section 1413 of the

Act. In addition to adopting the basic primacy requirements, States may

be required to adopt special primacy provisions pertaining to a

specific regulation. These regulation-specific provisions may be

necessary where implementation of the NPDWR involves activities beyond

those in the generic rule. States are required by 40 CFR 142.12 to

include these regulation-specific provisions in an application for

approval of their program revisions. These State primacy requirements

would apply to the rule proposed today, along with the special primacy

requirements discussed below.

To implement today's proposed rule, States would be required to

adopt revisions to Sec. 141.2, Definitions; Sec. 141.52, Maximum

contaminant level goals for microbiological contaminants; Sec. 141.70,

General requirements; Sec. 141.71, Criteria for avoiding filtration;

and Sec. 141.74, Analytical and monitoring requirements.

A. Special State Primacy Requirements

In addition to adopting drinking water regulations at least as

stringent as the Federal regulations listed above, EPA would require

that States adopt certain additional provisions related to this

regulation to have their program revision application approved by EPA.

Because this rule would provide considerable State latitude on

implementation, today's rule would require a State to include, as part

of its State program submission, its implementation policies and

procedures. This information would advise the regulated community of

State requirements and help EPA in its oversight of State programs. In

concert with promulgating the interim ESWTR, EPA would revise the SWTR

Guidance Manual (EPA, 1991). This guidance would assist States in

developing appropriate criteria in the regulations they adopted.

To ensure that the State program includes all the elements

necessary for a complete enforcement program, today's notice proposes

that to obtain EPA approval for implementing this rule, the State's

application would be required to include the following:

(1) Adoption of the promulgated ESWTR.

(2) Description of the protocol the State will use to judge the

adequacy of watershed protection programs for minimizing the potential

for contamination by Giardia cysts, Cryptosporidium oocysts, and

viruses in the source water. The SWTR required States to specify the

methodology they would use to judge the adequacy of a watershed control

program to control the presence of waterborne Giardia. This rule would

add Cryptosporidium. The addition of Cryptosporidium is significant

because it may prohibit or substantially limit certain watershed uses

such as cattle farming and feedlots. The location of cattle feedlots on

a watershed would require additional control measures.

(3) Description of the criteria and methods the State will use for

the conduct and review of sanitary surveys. If the State elects to

allow non-State personnel to conduct the surveys, the State must

specify the criteria for approval and oversight of these personnel and

of the surveys.

(4) Description of the procedures for determining the level of

treatment required of systems to meet removal and/or inactivation

requirements under the rule. If Alternative A described in Section IIIE

above is promulgated, demonstration by the State that it has in place

enforceable design and operating criteria for achieving the levels of

Giardia removal and/or inactivation required. If either Alternative B

or C described in Section IIIE above is promulgated, demonstration by

the State that it has in place enforceable design and operating

criteria for achieving the levels of Cryptosporidium removal and/or

inactivation required. Compliance with the design and operating

criteria would be judged on a system-by-system basis.

B. State Recordkeeping Requirements

Changes to the existing recordkeeping requirements to implement the

provisions proposed in this notice would require, under general

recordkeeping requirements, States to maintain records on the level of

treatment necessary to achieve the required levels of removal and/or

inactivation of Giardia, Cryptosporidium and/or viruses. States would

also be required to maintain a record of any decisions made as a result

of sanitary surveys. These records must be kept for 40 years (as

currently required by Sec. 142.14 for other State decision records) or

until a subsequent determination is made, whichever is shorter. If the

final rule requires systems to base level of treatment on source water

pathogen densities, then the State must maintain record of these

densities.

C. State Reporting Requirements

Currently States must report to EPA information under 40 CFR 142.15

regarding violations, variances and exemptions, enforcement actions and

general operations of State public water supply programs. Today's rule

would require States to provide additional information to EPA within

the context of the existing special report requirements for the SWTR

(Sec. 142.15(c)(1)) on microbial densities in the source water and the

resulting required levels of treatment for each public water system

supplied by a surface water source or by ground water under the direct

influence of surface water.

V. Public Notice Language

The SDWA (section 1414(c)) requires that notices of violation of

the MCL or treatment requirement for a specific contaminant include

EPA-specified language on the adverse health effects of that

contaminant. Requirements for public notification are found in 40 CFR

141.32. In this notice, EPA is proposing that the existing language for

violating the treatment technique requirements in Subpart H of the

SWTR, found in 40 CFR 141.32(e)(10), not be changed. This decision is

based on EPA's belief that language is sufficiently broad to include

the adverse health effects from Cryptosporidium exposure.

VI. Economic Analysis

A. Cost of Proposed Rule

This proposed rule would result in treatment costs, monitoring

costs, and State implementation costs. These costs are difficult to

estimate because of uncertainty in the number of systems that would

have to improve treatment and the extent of that improved treatment.

This information would depend primarily on the results of future

monitoring under the ICR. Under the ICR, systems using surface water

and serving 10,000 people or more would determine raw water pathogen

densities and, in some cases, the efficiency of treatment for reducing

pathogen concentrations. Given the above uncertainties, the cost

estimates can now only be addressed in the most general way, across a

wide range of possibilities.

With regard to treatment costs, if ICR results indicate that the

existing SWTR ensures adequate levels of treatment for most systems,

then minimal additional treatment costs would be necessary. Regardless

of whether the SWTR is amended to require higher levels of treatment,

at least some systems would be expected to upgrade existing levels of

treatment based on EPA guidance and ICR monitoring results. Similarly,

some systems might reduce existing levels of disinfection upon a

finding that their source water is of better quality than expected.

Also, some costs will be incurred by systems correcting for

deficiencies identified through the sanitary survey requirement.

If ICR monitoring indicates that many source waters contain

considerably higher pathogen concentrations than anticipated under the

SWTR, then substantial national treatment costs would result in

mitigating the associated health risk. These costs could involve

increasing disinfection contact time or dosage, switching to stronger

disinfectants, or improving filtration efficiencies through upgrades or

installation of new technologies.

In estimating possible costs resulting from an ESWTR EPA assumed

that (1) national Giardia density in source waters are represented by

the survey results of LeChevallier et al. (1991a), (2) all systems are

at least meeting the treatment requirements of the existing SWTR, (3)

some systems, as indicated by the survey results of LeChevallier et al

(1991), are providing higher levels of treatment than required by the

SWTR, (4) systems would be required to provide sufficient treatment of

their source water to achieve no greater than a 10-4 annual risk

level for Giardia, based on the dose-response data and risk assessment

methodology developed by Rose et al. (1991) and Grubbs et al. (1992),

(5) additional Giardia reduction beyond the requirements of the SWTR to

achieve the 10-4 risk level would be achieved solely by using

chlorine as the disinfectant and providing additional disinfectant

contact time (i.e., increasing the CT value by increasing the contact

basin size), (6) when all ancillary construction costs including site-

specific factors are taken into account, the average total capital cost

per system is twice the capital cost of increasing the size of the

contact basin alone. Based on the assumptions of Rose et al. (1991) and

Grubbs et al. (1992), EPA calculates that systems would need a Giardia

removal/inactivation level of 3, 4, 5, or 6 logs for Giardia

concentrations in the source water of -4 annual risk level target was used as an example;

costs for achieving different acceptable risk levels, of course, will

differ considerably. Although other treatment measures could be used to

reduce Giardia levels, EPA believed that the national cost based on

providing additional disinfectant contact time is probably

representative, on average, of other modifications that systems might

implement. The Agency chose this methodology for estimating costs

because it was the most simple. Moreover, insufficient data prevents

the Agency from predicting with any reliability the mix of different

technologies systems would use to comply. EPA recognizes that in lieu

of expanding contact basin size, some systems may achieve the required

Giardia reductions through increased disinfectant dosages, improved

sedimentation and filtration efficiencies, or use of a stronger

disinfectant such as ozone. Smaller systems, especially those serving

fewer than 1,000 people, might use cartridge filters, or membrane

technology rather than additional contact time to achieve compliance

with the long-term ESWTR and other drinking water regulations at lower

cost. In addition, the costs for utilities to meet the D/DBP

regulations (proposed elsewhere in today's Federal Register) are not

necessarily additive with the costs for utilities to meet either the

interim or long-term ESWTR. For example, systems installing membrane

technology to comply with the D/DBP rule would also be expected through

use of this technology to comply with the ESWTR. Use of technologies

other than increasing contact basin time might be more feasible and

less expensive for some systems, depending on site-specific factors

(e.g., limited availability of land) and overall treatment objectives

(e.g., meeting other regulatory requirements such as the D/DBR rule).

EPA solicits comment on how many systems might use these

alternative approaches for meeting ESWTR requirements and whether the

use of such technologies would lead to substantially different cost

estimates. EPA does not believe there are sufficient data to predict

the costs for reducing Cryptosporidium to a desired risk level as has

been done for Giardia. EPA solicits comment on what approaches might be

taken for estimating national treatment costs for systems to provide

different levels of Cryptosporidium removal depending on

Cryptosporidium densities in the source water. Also, EPA requests

comment on whether it is reasonable to assume that any treatment

changes that are made to remove Cryptosporidium would also remove

Giardia, thereby not duplicating costs for compliance.

Table VI-2 indicates a range of estimated increases in household

costs by system size category for systems needing to achieve an

additional 0.5 log to 3 log reduction of Giardia to comply with the

ESWTR option described above. By this analysis 35 percent of the

systems would not be required to make any changes in treatment and

would incur no costs. For the interim ESWTR, estimated increases in

household costs for systems required to make changes in treatment would

range from $11 to $49 per household per year in the smallest size

category (serving a median population of 15,000 people) to $3.1 to $24

per household per year in the largest size category (serving a median

population of 1,550,000 people).

If the analyzed criteria for the interim ESWTR were extended to

smaller systems under the long-term ESWTR, and systems used additional

disinfectant contact time to meet such criteria, increases in

annualized household costs would range from $360 to $1100 per household

per year in the smallest size category (serving a median population of

57 people) to $27 to $85 per household per year in systems with a

median population of 5,500. As stated above, EPA believes that smaller

systems should be able to use a more economic treatment alternative

than additional disinfectant contact time.

EPA solicits comment on whether the system level costs to achieve

the different log reductions indicated in Table VI-2 by disinfection,

or other means, are reasonable and accurate.

Table VI-3 indicates the estimated labor effort by the number of

full time employees (FTEs), hours, and dollar costs for States to

implement the interim ESWTR. If systems, rather than the State, were to

fund some or all sanitary surveys, then State costs would be reduced

accordingly. Further details of this analysis are available in the

Regulatory Impact Analysis (EPA, 1994).

Table VI-1--Estimated National Contact Basin Costs for Enhanced SWTR

--------------------------------------------------------------------------------------------------------------------------------------------------------

Cost estimates based on additional disinfectant contact basin size\1\

-------------------------------------------------------------------------------------------

Number of affected systems Total capital cost (M$) Total annualized cost (M$)

Population per -------------------------------------------------------------------------------------------

System size category system Filtering systems Filtering systems Filtering systems

-------------------------------------------------------------------------------------------

W/out W/out W/out

soft W/soft Total soft W/soft Total soft W/soft Total

--------------------------------------------------------------------------------------------------------------------------------------------------------

1......................................... 25-100 493 8 501 22 0.5 23 4 0.09 4

2......................................... 101-500 454 13 467 42 1.4 43 8 0.3 8

3......................................... 501-1K 415 47 462 86 12 98 16 2 18

4......................................... 1K-3.3K 586 61 647 200 24 224 25 3 28

5......................................... 3.3K-10K 627 104 731 360 72 432 45 9 54

6......................................... 10K-25K 291 67 358 330 94 424 34 10 44

7......................................... 25K-50K 167 42 209 340 110 450 36 12 47

8......................................... 50K-75K 77 24 101 230 92 322 24 10 34

9......................................... 75K-100K 63 7 70 250 37 287 26 4 30

10........................................ 100K-500K 88 24 112 620 230 850 67 25 92

11........................................ 500K-1M 22 5 27 600 170 770 64 18 83

12........................................ 1M+ 9 1 10 460 97 557 50 11 60

-------------------------------------------------------------------------------------------------------------

Totals:

Interim Rule (Systems > 10K)........ ................ 716 170 887 2830 831 3661 302 88 391

Long-Term Rule (All Systems)........ ................ 3,290 404 3,694 3,540 941 4,481 401 103 504

--------------------------------------------------------------------------------------------------------------------------------------------------------

Notes:

\1\Cost estimates were developed on the basis of the following assumptions: 1) 35 percent of surface water systems currently meet ESWTR inactivation

requirements (based on LeChevallier et al., 1991); 2) the amount of additional inactivation required by systems that do not currently meet ESWTR

requirements is based on the distribution of source water Giardia concentrations in the LeChevallier data; and 3) the additional basin volume is based

upon CT requirements of the SWTR guidance document with: pH = 8 (non-softening) or 9 (softening), t10 : ttheoretical = 0.7, temperature = 5 deg.C,

and C12 residual = 1 mg/l.

Table VI-2--Estimated Increases in Annual Houshold Costs for Systems Expanding Contact Basin Size to Meet an

Enhanced SWTR\2\

----------------------------------------------------------------------------------------------------------------

Total household costs, $/hh/yr\3\

-----------------------------------------------------------

Cat. # Median pop. Filter\4\ Filter and Soften\4\

-----------------------------------------------------------

Min Avg\1\ Max Min Avg\1\ Max

----------------------------------------------------------------------------------------------------------------

1................................... 57 360 420 960 430 500 1,100

2................................... 225 170 200 450 200 240 540

3................................... 750 88 120 310 110 150 350

4................................... 1,910 36 52 110 43 59 130

5................................... 5,500 27 28 67 30 34 85

6................................... 15,000 11 15 40 13 19 49

7................................... 35,000 7.9 12 30 10 15 39

8................................... 60,000 6.6 10 26 8.8 13 34

9................................... 88,100 6.1 8.8 24 7.5 12 32

10.................................. 175,000 5.1 8.5 24 6.6 11 33

11.................................. 730,000 3.5 6.7 20 4.7 9.1 27

12.................................. 1,550,000 3.1 6.0 18 4.2 8.2 24

----------------------------------------------------------------------------------------------------------------

Notes:

\1\Costs assume that 35 percent of systems currently meet ESWTR requirements (LeChevallier et al., 1991) and

therefore do not require contact basin modifications.

\2\Assumes Giardia and level of treatment distributions per LeChevallier et al, 1991 are nationally

representative of arithmetic averages, and that systems under ESWTR are required to provide additional

disinfection inactivation to meet a less than 1/10,000 annual infection rate at the first customer calculated

according to the Giardia infectivity dose response curve of Rose et al. (1991).

\3\Household costs represent costs for affected systems. Minimum costs are based on costs for systems requiring

additional 0.5 log inactivation while maximum costs are based on requiring an additional 3-log inactivation.

Average costs are based on distribution of costs for achieving different inactivations based on data by

LeChevallier et al. (1991).

\4\Contact basin size dependent upon chlorine residual, pH and temperature. Contact basin costs will increase if

chlorine residual or temperature decrease or if pH increases. For non-softening systems (``Filter'' and

``Unfilt'') pH=8, temperature=5 deg.C, and chlorine residual=1 mg/L. For softening systems (``Soften''),

pH=9, temperature=5 deg.C, and chlorine residual=1 mg/L.

Table VI-3.--Interim Enhanced Surface Water Treatment Rule State Program Costs Model

----------------------------------------------------------------------------------------------------------------

National burden

Variable Default assumptions ------------------------- Cost

In FTEs In hours

----------------------------------------------------------------------------------------------------------------

Regulation Adoption and 0.5 FTE per State............. 1 28.0 47,040 $1,540,000

Program Development.

Review Plans and Specs........ 10 days/large system.......... 112 \1\N/A \1\N/A \1\N/A

Log Removal Determination..... 3 days per surface water 19 19.0 31,958 1,046,250

system.

---------------------------------------------------------------------------------

Subtotal (x)............ .............................. ....... 47 78,998 2,586,250

Staff Training (Rule Specific) 10 days per technical FTE=f(x) 0 2.1 3,591 117,557

Sanitary Surveys\3\........... 5 days/lg. sys./survey; 2 days/ 79 78.7 132,147 4,326,250

sm. sys./survey.

=================================================================================

Total................... .............................. ....... 128 214,736 7,030,057

Average Annual Cost over 3.5 .............................. ....... 37 61,353 2,008,588

years\2\.

----------------------------------------------------------------------------------------------------------------

\1\Costs for reviewing plans and specifications for the ESWTR are counted as a joint activity undertaken with

the same step of implementing the Stage 1 DBP Rule and are included in the DBP Rule RIA.

\2\Total cost and burden are divided by 3.5 years, the time between promulgation of this rule and the final

ESWTR.

\3\Recordkeeping burden is assumed to equal approximately 2% of the burden shown (i.e. approximately 1.6 FTEs,

2,640 hours, $86,000).

B. Benefits of Proposed Rule

The level of reduction of waterborne illness resulting from

implementation of this rule will largely depend on the particular

option(s) promulgated. Even if EPA could predict the most suitable

option(s), the Agency cannot yet predict the number of illnesses

avoided until more data become available. EPA anticipates that much of

such data, particularly on national pathogen occurrence and existing

treatment levels, will become available under the forthcoming ICR.

With the limited available data, EPA has used a disinfection

byproducts risk assessment model (DBPRAM) to estimate potential risks

from Giardia that might result from systems complying with different

DBP standards both with the existing SWTR and an ESWTR (Grubbs et al

1992; Regli et al 1993; Cromwell et al 1992). In this analysis, EPA

assumed that the ESWTR would require systems to remove/inactivate

Giardia by 3, 4, 5, or 6 logs if the Giardia concentrations in the

source water were (III.C) Rationale for setting MCLG of zero and a treatment

technique for Cryptosporidium

(III.D.1) Ground water under the direct influence of

surface water

--Including Cryptosporidium in rule language in definition of ground

water under the direct influence of surface water

--Revising guidance defining ground water under the influence of

surface water

--Most appropriate procedure for determining credits for removal/

inactivation and treatment requirements for systems using ground waters

under the direct influence of surface water

(III.D.3) Sanitary surveys

--Prerequisites for individuals performing sanitary survey (academic

degree, etc.)

--Revisions needed in SWTR Guidance Manual for conducting sanitary

survey for filtered systems and for evaluating vulnerability to

Cryptosporidium

--Frequency of sanitary surveys (three vs. five years)

(III.D.4) Possible supplemental requirements

--Whether to publish national rule to require systems to cover finished

water reservoirs and storage tanks, or whether this should be left to

State discretion. What would the cost be for such a rule, and what

waiver provisions would be appropriate.

--Whether EPA should require States and/or systems to have a cross-

connection control program; what specific criteria, if any, should be

included therein; how often such a program should be evaluated; under

what conditions a waiver could be granted; and whether only those

connections identified as a cross-connection by the public water system

or the State should be subject to a cross-connection program.

--Identification of measures other than cross-connection control

program to prevent the contamination of drinking water already in the

distribution system (e.g., minimum pressure requirements in the

distribution system).

--Whether to require systems to notify the State for persistent

turbidity levels above the performance standard (but not in violation

of this standard)

(III.E) Alternative Treatment Requirements

1. Options for defining pathogen densities

--Appropriateness of requiring systems whose population served grows to

over 10,000 to perform ICR monitoring.

--Appropriateness of the four approaches for calculating pathogen

densities, and whether approach selected should be based on number of

pathogen samples collected

2. Treatment alternative for controlling pathogens

--Appropriateness and magnitude of specific acceptable risk levels for

microbes

--Which treatment approach(es) is most appropriate

--Identification of approaches for achieving levels of pathogen removal

greater than 2.5 logs by physical means

--Utility of extrapolating CT values in SWTR Guidance Manual to predict

the effect on pathogens of higher levels of disinfection

--Appropriateness of two treatment alternatives, with possible

variations, for removal of Giardia

--Appropriateness of indicated treatment alternatives for

Cryptosporidium

--Feasibility of removing more than two logs of Cryptosporidium (with

and without disinfection being considered). -

Appropriateness of not

--Changing treatment specifications in SWTR

(VI) Economic Analysis

--What approaches are reasonable for estimating the national treatment

costs of requiring systems to remove a level of Cryptosporidium that

would depend on Cryptosporidium densities in the source water

--Are the system level costs in Table VI-2 for increasing the

disinfectant contact time reasonable and accurate

--The number of systems that might use control measures other than

increasing contact basin time requirements and whether the use of such

technologies would lead to substantially different cost estimates

--Assumption in estimating economic impact that treatment changes to

control Cryptosporidium will also control Giardia

--Soundness of assumptions made in disinfection byproducts risk

assessment model (DBPRAM) used to estimate potential risks from Giardia

that might result from systems complying with different DBP standards

both with the existing SWTR and an ESWTR, and how these assumptions

could be improved

Other Issues

How should EPA decide, in developing a forthcoming Notice

of Availability, what treatment approach(es) is most suitable for

additional public comment?

What criteria, if any, should the ESWTR include to ensure

that systems optimize treatment plant performance?

Should any turbidity performance criteria in the SWTR be

modified? For example, should the ESWTR require systems to base

compliance with the turbidity standards on monitoring the turbidity at

the effluent of each filter separately, in lieu of (or in addition to)

the confluence of all filters? Should any performance standard value be

changed?

To what extent should the ESWTR address the issue of

recycling filter backwash, given its potential for increasing the

densities of Giardia and Cryptosporidium on the filter?

Should the ESWTR define minimum certification criteria for

surface water treatment plant operators? Currently, the SWTR

(Sec. 141.70) requires such systems to be operated by ``qualified

personnel who meet the requirements specified by the State.''

Should the ESWTR include a performance standard(s) for

particle removal?

Under what conditions could systems be allowed different

log removal credits than is currently recommended in the SWTR Guidance

Manual?

IX. Instructions to Commenters

To ensure that EPA can read, understand and therefore properly

respond to comments, the Agency would prefer that commenters type or

print comments in ink, and cite, where possible, the paragraph(s) in

this proposed regulation (e.g. 141.76(b)) to which each comment refers.

Commenters should use a separate paragraph for each method or issue

discussed.

X. References

Addiss, DG, and EJ Lengerich (1994). Hypokalemic myopathy induced by

Giardia lamblia. N. Engl. J. Med. 330(1):66.

AWWA. American Water Works Association (1992). Jackson County Oregon

cryptosporidiosis outbreaks: January-June 1992. Summary, Expert

Meeting, August 3, 4, 1992.

AWWA. American Water Works Association (1993). Officers & Committee

Directory. AWWA, Denver.

Bailey SW, and EC Lippy (1978). Should all finished water reservoirs

be covered. Public Works for April 1978. P. 66-70.

Bennett JV, SD Holmberg, MF Rogers and SL Solomon (1987). Infectious

and parasitic diseases. Am J Prev Med 3: 102-114. In: RW Amler and

HB Dull (eds), Closing the gap: the burden of unnecessary illness.

Oxford University Press, Oxford.

Berger PS, S Regli and L Almodovar (1992). Cryptosporidium control

in drinking water. Am Water Works Assoc Proceedings, 1992 Annual

Conf, Mgmt and Regs, pp 845-864. American Water Works Association,

Denver.

Casemore DP and FB Jackson (1984). Hypothesis: cryptosporidiosis in

human beings is not primarily a zoonosis. J Infect 9: 153-156.

Casemore DP (1990). Epidemiological aspects of human

cryptosporidiosis. Epid Infec 104: 1-28.

CDC. Centers for Disease Control (1982). Cryptosporidiosis:

assessment of chemotherapy of males with acquired immune deficiency

syndrome (AIDS). Morbidity and Mortality Weekly Report 31: 589-592.

Clancy JL (1993). Interpretation of microscopic particulate analysis

data--a water quality approach. Water Quality Technology Conference

Proceedings, 1992, II: 1831-1844. American Water Works Association,

Denver.

Craun G (1991). Causes of waterborne outbreaks in the United States.

Wat Sci Technol 24:17-20.

Craun GF (1993). Conference conclusions. In: (GF Craun, ed.) Safety

of Water Disinfection: Balancing Chemical and Microbial Risks.

International Life Sciences Inst. Press, Washington.

Craun G (1994). Memorandum from G. Craun to U.S. Environmental

Protection Agency (P. Berger), dated 1/19/94. Waterborne outbreak

data 1981-90, community water systems.

Cromwell JE, Zhang X, Letkiewicz FJ, et al. Analysis of potential

tradeoffs in regulation of disinfection by-products. Office of Water

Resource Center. Washington D.C. EPA-811-R-92-008. 1992.

Current WL, NC Reese, JV Ernst, WS Bailey, MB Heyman and WM

Weinstein (1983). Human cryptosporidiosis in immunocompetent and

immunodeficient persons. New Eng J Med 308(21): 1252-1257.

D'Antonio RG, RE Winn, JP Taylor, et al. (1985). A waterborne

outbreak of cryptosporidiosis in normal hosts. Ann. Intern. Med.

103:886-888.

De Mol P, S Mukashema, J Bogaerts, W Hemelhof and J-P Butzler

(1984). Cryptosporidium related to measles diarrhoea in Rwanda.

Lancet 2(8393): 42-43.

EPA. Environmental Protection Agency (1983). Assessment of

Microbiology and Turbidity Standards for Drinking Water. EPA 570-9-

83-001. Washington, DC.

EPA. Environmental Protection Agency (1989a). Regulatory Impact

Analysis: Benefits and Costs of Final Surface Water Treatment Rule,

prepared by Wade Miller Associates, Inc., February 17, 1989,

Washington, DC.

EPA. Environmental Protection Agency (1989b). Cross-Connection

Control Manual. EPA 570/9-89-007. Washington, DC.

EPA. Environmental Protection Agency (1991a). Guidance Manual for

Compliance With the Filtration and Disinfection Requirements for

Public Water Systems Using Surface Water Sources. Washington, DC.

EPA. Environmental Protection Agency (1991b). Manual of Small Public

Water Supply Systems. EPA 570/9-91-003. Washington, DC.

EPA. Environmental Protection Agency (1991c). Manual of Individual

and Non-Public Water Supply Systems. EPA 570/9-91-004. Washington,

DC.

EPA. Environmental Protection Agency (1992). Consensus method for

determining groundwaters under the direct influence of surface water

using microscopic particulate analysis (MPA). EPA 910/9-92-029.

EPA. Environmental Protection Agency (1993). Drinking Water Criteria

Document for Cryptosporidium (Draft). Office of Science and

Technology (Office of Water), EPA, Washington, DC.

EPA. Environmental Protection Agency (1994). Regulatory Impact

Analysis of Proposed Interim Enhanced Surface Water Treatment Rule

(Draft). Office of Ground Water and Drinking Water.

Ernest J, B Blagburn, D Lindsay and W Current (1986). Infection

dynamics of Cryptosporidium parvum (Apicomplexa: Cryptosporidiidae)

in neonatal mice (Mus musculus). J Parasitology 72(5):796-798.

Fayer R, and BLP Ungar (1986). Cryptosporidium spp. and

cryptosporidiosis. Microbiol. Rev. 50:458-483.

Gerba, C, and J. Rose. (1990). Viruses in source and drinking water.

Chap.18, pp 380-396. In: G. McFeters (ed), Drinking Water

Microbiology. Springer-Verlag, New York.

Glass RI, JF Lew, RE Gangarosa, CW LeBaron and M-S Ho (1991).

Estimates of morbidity and mortality rates for diarrheal diseases in

American children. J Pediatrics 118: 27-33.

Grubbs WD, Macler B. Regli S. (1992). Modelling Giardia occurrence

and risk. EPA-811-B-92-005. Office of Water Resource Center.

Washington D.C.

Haas CN, JB Rose, C Gerba, S Regli (1993). Risk assessment of virus

in drinking water. Risk Analysis 13: 545-552.

Harrington, W.A., Krupnick, A.J., and Spofford, W.O.,Jr.. The

Benefits of Preventing an Outbreak of Giardiasis Due to Drinking

Water Contamination. Resources for the Future, 1616P Street, NW.,

Washington, DC. September, 1985.

Hayes EB, TD Matte, TR O'Brien, TW McKinley, GS Logsdon, JB Rose,

BLP Ungar, DM Word, PF Pinsky, ML Cummings, MA Wilson, EG Long, ES

Hurwitz and DD Juranek (1989). Large community outbreak of

cryptosporidiosis due to contamination of a filtered public water

supply. New Eng J Med 320: 1372-1376.

Herwaldt BL, GF Craun, SL Stokes and DD Juranek (1991). Waterborne

disease outbreaks, 1989-1990. In: CDC Surveillance Summaries,

Morbidity and Mortality Weekly Report: 40(SS-3): 1-21. Hurst C

(1991). Presence of enteric viruses in freshwater and their removal

by conventional drinking water treatment process. Bull World Health

Org. 69 (1): 113-119.

Keswick, BH, et al (1985). Inactivation of Norwalk virus in drinking

water by chlorine. Appl. Environ. Microbiol. 50: 261-264.

Korich DG, JR Mead, MS Madore, NA Sinclair and CR Sterling (1990).

Effects of ozone, chlorine dioxide and monochloramine on

Cryptosporidium parvum oocyst viability. Appl Environ Microbiol 56:

1423-1428.

Korich D, M Yozwiak, M Marshall, M Arrowood, N Sinclair, and C

Sterling (1992). Cryptosporidium viability: assessment and

correlation with infectivity. Water Quality Technology Conference

Proceedings, 1991, I: 65-74. American Water Works Assoc, Denver.

LeChevallier MW, DN Norton and RG Lee (1991a). Occurrence of Giardia

and Cryptosporidium spp. in surface water supplies. Appl Environ

Microbiol 57: 2610-2616.

LeChevallier MW, DN Norton and RG Lee (1991b). Giardia and

Cryptosporidium spp. in filtered drinking water supplies. Appl

Environ Microbiol 57:2617-2621.

Levine WC and GF Craun (1990). Waterborne disease outbreaks, 1986-

1988. In: CDC Surveillance Summaries, Morbidity and Mortality Weekly

Report 39(SS-1): 1-13.

Lew JF, RI Glass, RE Gangarosa, IP Cohen, C Bern and CL Moe (1991).

Diarrheal deaths in the United States, 1979 through 1987. A special

problem for the elderly. J Am Med Assoc 265: 3280-3284.

Lopez et al. (1980). Waterborne giardiasis: a communitywide outbreak

of disease and a high rate of asymptomatic infection. Am J Epid 112:

495-507.

Macler BA and S Regli (1993). Use of microbial risk assessment in

setting US drinking water standards. Int J Food Microbiol 18: 245-

256.

Miller RA, MA Bronsdon and WR Morton (1990). Experimental

cryptosporidiosis in a primate model. J Infect Dis 161: 312-315.

Moore AC, BL Herwaldt, GF Craun, RL Calderon, AK Highsmith, and DD

Juranek (1993). Surveillance for waterborne disease outbreaks--

United States, 1991-1992. Morbidity and Mortality Weekly Report:

42(SS-5): 1-22.

Payment, 1981. Isolation of viruses from drinking water at the

Point-Viau water treatment plant. Can. J. Microbiol. 27:417.

Payment P, M Trudel, et al., (1985). Elimination of viruses and

indicator bacteria at each step of treatment during preparation of

drinking water at seven water treatment plants. Appl. Environ.

Microbiol. 49:1418-1428.

Payment P, L Richardson, J Siemiatychi, R Dewar, M Edwardes and E

Franco (1991). A randomized trial to evaluate the risk of

gastrointestinal disease due to consumption of drinking water

meeting current microbiological standards. Am J Publ Health 81: 703-

708.

Regli S, JB Rose, CN Haas and CP Gerba (1991). Modeling the risk

from Giardia and viruses in drinking water. J Am Water Works Assoc

83 (11): 76-84.

Regli S, JE Cromwell, X Zhang, AB Gelderloos, WD Grubbs, F

Letkiewicz and BA Macler (1993). Framework for decision making: EPA

perspective. In: (GF Craun, ed.) Safety of water disinfection:

Balancing chemical and microbial risk. pp 487-538. International

Life Sciences Institute Press, Washington, DC.

Rendtorff RC (1954). The experimental transmission of human

intestinal protozoan parasites. II. Giardia lamblia cysts given in

capsules. Am J Hyg 59: 209-220.

Rose JB, CN Haas and S Regli (1991). Risk assessment and control of

waterborne giardiasis. Am J Publ Health 81: 709-713.

Smith HV, RWA Girdwood, WJ Patterson, et al. (1988). Waterborne

outbreak of cryptosporidiosis. Lancet 2: 1484.

Sobsey MD, F Takashi, and RM Hall (1991). Inactivation of cell-

associated and dispersed Hepatitis A virus in water. J Am Water

Works Assoc 83 (11): 64-67.

Williams, F (1985). Membrane-associated viral complexes observed in

stools and cell culture. Appl. Environ. Microbiol. 50:523-526.

Wittenberg DF, NM Miller and J van den Ende (1989). Spiramycin is

not effective in treating Cryptosporidium diarrhea in infants:

results of a double-blind randomized trial. J Infect Dis 159(1):

131-132.

Wolfe, MS (1990). Clinical symptoms and diagnosis by traditional

methods. Pp. 175-185, In: Giardiasis (EA Meyer, ed.), Elsevier, New

York.

List of Subjects

40 CFR Part 141

Intergovernmental relations, Reporting and recordkeeping

requirements, Water supply.

40 CFR Part 142

Administrative practice and procedure, Reporting and recordkeeping

requirements, Water supply

Dated: June 7, 1994.

Carol M. Browner,

Administrator.

For the reasons set forth in the preamble, Title 40 of the Code of

Federal Regulations is proposed to be amended as follows:

PART 141--NATIONAL PRIMARY DRINKING WATER REGULATIONS

1. The authority citation for part 141 continues to read as

follows:

Authority: 42 U.S.C. 300f, 300g-1, 300g-2, 300g-3, 300g-4, 300g-

5, 300g-6, 300j-4, 300j-9.

2. Section 141.2 is amended by revising the definition of ``Ground

water under the direct influence of surface water'' to read as follows:

Sec. 141.2 Definitions.

* * * * *

Ground water under the direct influence of surface water means any

water beneath the surface of the ground with:

(1) Significant occurrence of insects or other macroorganisms,

algae, or large-diameter pathogens such as Giardia lamblia or

Cryptosporidium, or

(2) Significant and relatively rapid shifts in water

characteristics such as turbidity, temperature, conductivity, or pH

which closely correlate to climatological or surface water conditions.

Direct influence must be determined for individual sources in

accordance with criteria established by the State. The State

determination of direct influence may be based on site- specific

measurements of water quality and/or documentation of well construction

characteristics and geology with field evaluation.

* * * * *

3. In Sec. 141.52, the Table is amended by adding a new entry, in

numerical order, to read as follows:

Sec. 141.52 Maximum contaminant level goals for microbiological

contaminants.

* * * * *

------------------------------------------------------------------------

Contaminant MCLG

------------------------------------------------------------------------

* * * * *

(5) Cryptosporidium............................................. zero

------------------------------------------------------------------------

4. Section 141.71 is amended by revising the first three sentences

of paragraph (b)(2) introductory text to read as follows:

Sec. 141.71 Criteria for avoiding filtration.

* * * * *

(b) * * *

(2) The public water system must maintain a watershed control

program which minimizes the potential for contamination by Giardia

lamblia cysts, Cryptosporidium oocysts, and viruses in the source

water. The State must determine whether the watershed control program

is adequate to meet this goal. The adequacy of a program to limit

potential contamination by Giardia lamblia cysts, Cryptosporidium

oocysts, and viruses must be based on: * * *

5. Section 141.73 is amended by revising paragraph (d) to read as

follows:

Sec. 141.73 Filtration.

* * * * *

(d) Other filtration technologies. A public water system may use a

filtration technology not listed in paragraphs (a) through (c) of this

section if it demonstrates to the State, using pilot plant studies or

other means, that the alternative filtration technology, in combination

with disinfection treatment that meets the requirements of

Sec. 141.72(b), consistently achieves 99.9 percent removal and/or

inactivation of Giardia lamblia cysts and 99.99 percent removal and/or

inactivation of viruses and 99 percent removal of Cryptosporidium

oocysts between the source water and the first customer. For a system

that makes this demonstration the requirements of paragraph (b) of this

section apply.

6. Section 141.74 is amended by adding a new paragraph (d) to read

as follows:

Sec. 141.74 Analytical and monitoring requirements.

* * * * *

(d) Sanitary surveys for all systems. (1) A public water system

that uses a surface water source or a ground water source under the

influence of surface water shall be subject to an initial sanitary

survey by [insert date 5 years after publication of the final rule] and

a subsequent sanitary survey every five years [ALTERNATIVE: every three

years] thereafter.

(2) The sanitary survey shall be performed by either the State, or

an agent approved by the State. An agent approved by the State shall be

paid by the system. In exceptional circumstances, the State may approve

the public water system to conduct its own sanitary survey. In this

case, the public water system shall certify that the system conducted

the sanitary survey is in accordance with Sec. 141.2 and that the

sanitary survey report is true and accurate.

(3) If the State or an agent approved by the State is not available

to conduct the sanitary survey within the time frame specified in this

section, the system must conduct the sanitary survey. If an agent

approved by the State or the system itself conducts the sanitary

survey, the system must submit the sanitary survey report to the State

within 90 days of completing the survey and before the end of the five

year period.

Alternative A

7. Section 141.70 is amended by revising paragraph (a)(1) to read

as follows:

Sec. 141.70 General requirements.

(a) * * *

(1)(i) At least 99.9 percent (3-log) removal and/or inactivation of

Giardia lamblia cysts for systems serving fewer than 10,000 people. A

system serving 10,000 people or more must achieve a Giardia removal/

inactivation level by [insert date 18 months after publication of the

final rule in the Federal Register that depends on the concentration of

Giardia in the source water(s), as follows:

(A) If the source water(s) contains less than 1 cyst/100 liters,

the system must achieve at least 99.9 percent (3-log) reduction;

(B) If the source water(s) contains 1 to 9 cysts/100 liters, the

system must achieve at least 99.99 percent (4-log) reduction [OPTION:

99.9 percent (3-log) reduction];

(C) If the source water(s) contains 10 to 99 cysts/100 liters, the

system must achieve at least 99.999 percent (5-log) reduction [OPTION:

99.99 percent (4-log) reduction];

(D) If the source water(s) contains more than 99 cysts/ 100 liters,

the system must achieve at least 99.9999 percent (6-log) reduction

[OPTION: 99.999 percent (5-log) reduction].

(ii) Systems must achieve the required Giardia removal/inactivation

level, as specified above, between the source water and the first

customer. To calculate the Giardia density in source water from

monitoring data obtained during the sampling period specified by

Sec. 141.140 of this part, use the:

Option 1: Arithmetic mean of measured values.

Option 2: Geometric mean of measured values.

Option 3: 90th percentile value of measured values.

Option 4: Highest measured value.

* * * * *

Alternative B

8. Section 141.70 is amended by adding new paragraph (a)(3) to read

as follows:

Sec. 141.70 General requirements.

(a) * * *

(3) Beginning 18 months after promulgation of this rule, a system

serving 10,000 people or more must achieve a Cryptosporidium removal/

inactivation level between the source water and first customer that

depends on the concentration of Cryptosporidium in the source water(s),

as follows:

(i) If the source water(s) contains less than 1 oocyst/100 liters,

the system must achieve at least 99.9 percent (3-log) reduction

[OPTION: 99 percent (2-log) reduction];

(ii) If the source water(s) contains 1 to 9 oocysts/100 liters, the

system must achieve at least 99.99 percent (4-log) reduction [OPTION 1:

99.9 percent (3-log) reduction; OPTION 2: 99 percent (2-log)

reduction];

(iii) If the source water(s) contains 10 to 99 oocysts/100 liters,

the system must achieve at least 99.999 percent (5-log) reduction

[OPTION 1: 99.99 percent (4-log) reduction; OPTION 2: 99.9 percent (3-

log) reduction];

(iv) If the source water(s) contains more than 99 oocysts/ 100

liters, the system must achieve at least 99.9999 percent (6-log)

reduction [OPTION 1: 99.999 percent (5-log) reduction; OPTION 2: 99.99

percent (4-log) reduction];

Systems must achieve the required Cryptosporidium removal/

inactivation level, as specified above, between the source water and

the first customer. To calculate the Cryptosporidium density in source

water from monitoring data obtained during the sampling period

specified by section 141.140 of this part, use the:

Option 1: Arithmetic mean of measured values.

Option 2: Geometric mean of measured values.

Option 3: 90th percentile value of measured values.

Option 4: Highest measured value.

* * * * *

Alternative C

9. Section 141.73 is amended by adding a new paragraph (e) to read

as follows:

Sec. 141.73 Filtration.

* * * * *

(e) Public water systems that filter their source water must

achieve at least 99 percent (2-log) removal of Cryptosporidium between

the source water and the first customer.

Alternative D

10. Section 141.72 is amended by adding a new paragraph (c) to read

as follows:

Sec. 141.72 Disinfection.

* * * * *

(c) Public water systems that serve 10,000 people or more and use

either surface water or ground water under the direct influence of

surface water must achieve, by disinfection alone, at least a 0.5-log

inactivation of Giardia [ALTERNATIVE 1: 4-log inactivation of viruses].

Alternative E

11. No change in existing SWTR regarding level of removal/

inactivation requirements.

PART 142--NATIONAL PRIMARY DRINKING WATER REGULATIONS

IMPLEMENTATION

1. The authority citation for part 142 continues to read as

follows:

Authority: 42 U.S.C. 300f, 300g-1, 300g-2 300g-3, 300g-4, 300g-

5, 300g-6, 300j-4, 300j-9.

2. Section 142.16 is amended by adding paragraph (g) to read as

follows:

Sec. 142.16 Special primacy requirements.

* * * * *

(g) An application for approval of a State program revision that

adopts the requirement specified below must contain the following:

(1) The State must designate the method it will use to judge the

adequacy of watershed protection programs in minimizing the potential

for contamination by Giardia lamblia cysts, Cryptosporidium oocysts,

and viruses in the source water.

(2) The State must describe its criteria for the conduct of

sanitary surveys and the method it will use to judge the adequacy of

each sanitary survey. If the State elects to allow non-State personnel

to conduct the surveys, the State must specify the criteria to be used

to approve the non-State personnel. If the State intends to allow

public water systems to conduct sanitary surveys, the State must

specify procedures it will use for oversight and review of the surveys.

Alternative A

3. The following special primacy requirements are associated with

Alternative A from item 7, above.

(3) Section 141.70(a)(1). The State must demonstrate that it has in

place enforceable design and operating criteria for achieving the

levels of Giardia lamblia removal/inactivation required. Alternatively,

the State may institute a procedure for establishing design and

operating conditions on a system-by-system basis (e.g., a permit

system).

Alternative B

4. The following special primacy requirements are associated with

Alternative B from item 8, above.

(3) Section 141.70(a)(3). The State must demonstrate that it has in

place enforceable design and operating criteria for achieving the

levels of Cryptosporidium removal/inactivation required. Alternatively,

the State may institute a procedure for establishing design and

operating conditions on a system-by-system basis (e.g., a permit

system).

Alternative C

5. The following special primacy requirements are associated with

Alternative C from item 9, above.

(3) Section 141.73(e). The State must demonstrate that it has in

place enforceable design and operating criteria for achieving 2-log

removal of Cryptosporidium between the source water and the first

customer. Alternatively, the State may institute a procedure for

establishing design and operating conditions on a system-by-system

basis. (e.g., a permit system)

Alternative D

6. The following special primacy requirements are associated with

Alternative D from item 10, above.

(4) Section 141.72(c). The State must demonstrate that it has in

place enforceable design and operating criteria for achieving the level

of Giardia (virus) inactivation required. Alternatively, the State may

institute a procedure for establishing design and operating conditions

on a system-by-system basis (e.g., a permit system).

[FR Doc. 94-17650 Filed 7-28-94; 8:45 am]

BILLING CODE 6560-50-P