Federal Register, Volume 62 Issue 128 (Thursday, July 3, 1997)

[Federal Register Volume 62, Number 128 (Thursday, July 3, 1997)]
[Proposed Rules]
[Pages 36100-36136]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 97-17210]

[[Page 36099]]

_______________________________________________________________________

Part II

Environmental Protection Agency

_______________________________________________________________________

40 CFR Parts 141 and 142

Drinking Water Monitoring Requirements for Certain Chemical
Contaminants--Chemical Monitoring Reform (CMR) and Permanent Monitoring
Relief (PMR); Proposed Rule

Federal Register / Vol. 62, No. 128 / Thursday, July 3, 1997 /
Proposed Rules

[[Page 36100]]

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 141 and 142

[FRL-5851-6]
RIN 2040-AC73

Drinking Water Monitoring Requirements for Certain Chemical
Contaminants--Chemical Monitoring Reform (CMR) and Permanent Monitoring
Relief (PMR)

AGENCY: Environmental Protection Agency (EPA).

ACTION: Advance notice of proposed rulemaking.

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

SUMMARY: EPA is providing advance notice that it is planning to propose
revising the drinking water monitoring requirements for sixty four
chemical contaminants. These chemicals may occur in the source water of
public drinking water systems, and are regulated on the basis of
chronic health effects over a seventy year period. The purpose of the
proposal would be to base the monitoring requirements for each water
system on its risk of contamination, and to establish a uniform and
simple sampling schedule for those systems without an apparent or
significant risk of contamination.
EPA is also soliciting comments on draft Permanent Monitoring
Relief (PMR) Guidelines under section 1418(b) of the Safe Drinking
Water Act (the Act), as amended August 6, 1996. The Act requires EPA to
issue guidelines, by August 6, 1997, for States to use in adopting
monitoring relief under Sections 1418 and 1453.
EPA is also soliciting comments on certain other changes under
consideration: the deadlines for decisions regarding ground water under
the direct influence of surface water and associated filtration
determinations; and reporting requirements for both public water
systems and State regulatory agencies. These potential changes were
raised by ``stakeholders'' in the drinking water community, through a
number of public meetings convened to explore ways of reducing the
burden created by the National Primary Drinking Water Regulations.
Today's action requests comments on the ``stakeholder'' suggestions,
which are described below under Suggestions for Regulatory Burden
Reduction Other Than Chemical.

DATES: Written comments must be postmarked or delivered by hand by
August 4, 1997. The public hearing dates are:

1. July 8, 1997, 9:00 a.m. to 5 p.m., Denver, Colorado
2. July 9, 1997, 9:00 a.m. to 5 p.m., Chicago, Illinois
3. July 22-23, 1997, 9:00 a.m. to 5 p.m., Washington, DC.

ADDRESSES: Send all written comments on this notice to the ``Chemical
Monitoring Reform Comment Clerk; Water Docket MC-4101 (Docket # W-97-
03); Environmental Protection Agency; 401 M Street, SW., Washington, DC
20460.'' Supporting documents for this proposed rulemaking are
available for review at EPA's Water Docket; 401 M Street, SW.,
Washington, DC 20460. For access to the Docket materials, call (202)
260-3027 between 9 a.m. and 3:30 p.m. for an appointment, and reference
``Docket #W-97-03''.
The public hearings will be held in the following locations:

1. EPA, Region VIII, Rocky Mountain Room in the 2nd floor Conference
Center, 999 18th Street, Denver, Colorado 80202
2. EPA, Region V, Lake Michigan Room (12th Floor), 77 West Jackson
Blvd., Chicago, Illinois 60604
3. Wyndham Bristol Hotel, Room Potomac 3, 2430 Pennsylvania Ave. NW.,
Washington, DC 20037.

FOR FURTHER INFORMATION CONTACT: The Safe Drinking Water Hotline, toll
free (800) 426-4791 for general information about, and copies of, this
document. To speak to the rule manager about today's proposal, contact
Mike Muse; Implementation & Assistance Division; Office of Ground Water
and Drinking Water; EPA (4604), 401 M Street SW., Washington, DC 20460;
telephone (202) 260-3874.

SUPPLEMENTARY INFORMATION: The Chemical Monitoring Reform portion of
this document presents many possible changes to the current
requirements in a detailed format, so that commenters can better assess
how the concepts in this document might work in the real world. In
addition, this document contains preliminary rule language so that
commenters may begin to address the details of regulatory
implementation. EPA is very open to suggestions for different and/or
additional changes to the current requirements, and to suggestions for
new or revised rule language for Chemical Monitoring Reform. After
considering and incorporating the public comments, the proposed changes
to the current regulations may be quite different from this document.
Concerning the Permanent Monitoring Relief Guidelines, EPA will
consider the comments received in response to this notice, and will
issue final guidelines by the August 6, 1997 statutory deadline. As
discussed in Section I.B below, EPA anticipates that regulations may be
needed in order to implement fully the Permanent Monitoring Relief
guidelines. Accordingly, EPA may propose such regulations at the same
time that the CMR regulations are proposed.
These changes would affect community water systems (CWSs) and non-
transient, non-community water systems (NTNCWSs). Community water
systems are those which serve at least 15 service connections used by
year round residents, or regularly serve at least 25 year round
residents e.g., cities, townships, district water authorities, private
water companies serving such communities. Non-transient, non-community
water systems are those which are not community water systems and which
serve at least 25 of the same persons over six months of the year e.g.,
schools, factories or other facilities with their own separate water
supply. The following table identifies the SIC code affected by this
action.

------------------------------------------------------------------------
SIC
Standard industrial classification description code
------------------------------------------------------------------------
Water Supply................................................... 4941
------------------------------------------------------------------------

If your comments pertain only to Chemical Monitoring Reform, only
to the Permanent Monitoring Relief Guidelines, or only to the other
ideas for burden reduction (e.g., deadlines for decisions regarding
ground water under the direct influence of surface water), please
indicate that in the first paragraph of your comments. Commenters are
requested to submit any references cited in their comments. Commenters
also are requested to submit an original and 3 copies of their written
comments and enclosures.
Commenters who want receipt of their comments acknowledged should
include a self-addressed, stamped envelope. No facsimiles (faxes) will
be accepted. The Agency would prefer for commenters to type or print
comments in ink. Commenters should subtitle each issue, including the
citation of the rule paragraph to which it pertains e.g.,
``Detection>MCL--Sec. 141.23(f):''.

Table of Contents

I. Summary of Today's Document
A. Chemical Monitoring Reform
B. Permanent Monitoring Relief (PMR) Guidelines
C. Suggestions for Regulatory Burden Reduction Other Than
Chemical Monitoring Reform
II. Background
A. Statutory Authority
B. Regulatory Background

[[Page 36101]]

C. Overview of Approach for Chemical Monitoring Reform &
Permanent Monitoring Relief
D. Anticipated Impact on Systems and States
III. Detailed Explanation of Draft Changes to Chemical Monitoring
Requirements
A. Affected Water Systems
B. Sampling Points
C. Time of Monitoring
D. Responsibility to Provide Information
E. Mandatory Monitoring
F. Detection\1/2\MCL
G. MCL Violation Determinations
H. Laboratory Certification Criteria
I. New Systems & New Sources
J. Sample Compositing
K. Records Kept by States
L. Special State Primacy Requirements
M. Safe Drinking Water Act Amendments
N. Permanent Monitoring Relief Guidelines
O. Suggestions for Regulatory Burden Reduction Other Than
Chemical Monitoring Reform
Appendix A to Preamble: EPA Technical Criteria Document for The
Analysis of Selected Chemicals in Drinking Water

Abbreviations Used in This Document

BAT: Best Available Technology
CWS: Community Water System
EPA: Environmental Protection Agency
FR: Federal Register
IMR: Interim Monitoring Relief
IOC: Inorganic Chemical
LFB: Laboratory Fortified Blank
MCL: Maximum Contaminant Level
MDL: Method Detection Level
NPDWR: National Primary Drinking Water Regulation
NTNCWS: Non-transient, Non-community Water System
PE: Performance Evaluation
PMR: Permanent Monitoring Relief
PQL: Practical Quantitation Level
PWS: Public Water System
RDL: Reliable Detection Level
RIA: Regulatory Impact Analysis
SDWA: Safe Drinking Water Act
SMF: Standard Monitoring Framework
SWAP: Source Water Assessment Program
SWRA: Source Water Review Area
SOC: Synthetic Organic Chemical
VOC: Volatile Organic Chemical
WHP: Wellhead Protection

List of Tables

Table A: Contaminants Affected by Chemical Monitoring Reform
Table B: Phase I Sampling Results of Organic Chemicals in Surface
Water
Table C: Phase I Sampling Results of Organic Chemicals in Ground
Water
Table D: Phase I Sampling Results for Ethylene Dibromide in Ground
Water
Table E: Phase I Sampling Results for Ethylbenzene in Ground Water
Table F: Aggregated VOC Compliance Sampling Data from Selected
States
Table G: Aggregated SOC Compliance Sampling Data from Selected
States
Table H: Chemical Monitoring Reform, Sampling Frequency Decision
Diagram
Table I: Standard Monitoring Framework, Repeat Sampling Frequency
Decision Diagram
Table J: Sampling Frequency Decision Diagram

I. Summary of Today's Document

A. Chemical Monitoring Reform

The purpose of this document is to suggest regulatory changes to
strengthen public health protection by reducing the chance of drinking
water contamination going undetected and unaddressed, and to reduce
unnecessary monitoring and reporting requirements. The reduction of
unnecessary monitoring will release public resources to focus on those
systems at risk of contamination, and on the contaminants posing such
risk.
The current monitoring requirements, specifically those under
Secs. 141.23 (a) through (c) and 141.24 (f) through (k), would be
replaced with a new approach that would (1) Consolidate the monitoring
requirements into a sampling frequency of once every five years for
those systems that States determine have very low risk of
contamination, (2) require States to target the `at risk' systems to
sample at a greater frequency based on the degree of each system's
vulnerability, and (3) provide for sampling during the periods of
greatest vulnerability. Further, this approach would promote the
implementation of source water protection to reduce systems'
vulnerability.
In addition, the quality control criteria for chemical analyses
would be consolidated into a separate technical criteria document that
would be incorporated by reference into a final Chemical Monitoring
Reform rule, as would the analytical methods and acceptance criteria
for these chemicals.

B. Permanent Monitoring Relief (PMR) Guidelines

Section 1418(b) of the Safe Drinking Water Act, as amended,
requires EPA to issue guidelines by August 6, 1997 for States to use in
adopting Permanent Monitoring Relief. Section 1418(b) authorizes a
State to offer a water system relief from the Federal monitoring
requirements, in accordance with the EPA guidelines, after the State's
Source Water Assessment Program has been approved by EPA and the local
source water assessment has been completed.
A draft of the Permanent Monitoring Relief Guidelines is presented
in this document under Section III.N. The key features are (1) Sampling
waivers under which systems could receive a waiver from sampling for a
five year period, if there is no risk to public health, (2) the
designation of surrogate sampling points under which systems could use
the results from some of their sampling points for other sampling
points, and (3) relaxed monitoring for nitrate under limited
conditions.
The final PMR guidelines will provide sufficient information about
monitoring provisions of the PMR for a State to ensure that its Source
Water Assessment Program will provide the data needed for PMR if the
State intends to avail itself of the alternative monitoring program
available under the PMR. However, EPA believes that to allow States to
implement the final guidelines, it may be necessary to revise the
monitoring requirements in 40 CFR Parts 141 and 142. EPA may need to
provide in the regulations that monitoring under PMR assures compliance
with applicable national primary drinking water regulations, thereby
allowing States to implement a monitoring plan that differs from the
current requirements. Second, certain provisions of the proposed
guidelines (Section III.N of this notice) would include specific forms
of monitoring flexibility and minimum elements for approvable State PMR
requirements that, if such provisions are to be included in the final
guidelines and be binding on States, may need conforming regulations.
The Agency solicits comments on what conforming changes, if any, might
be needed.
EPA may propose regulatory language to support the PMR in the
Federal Register notice proposing the CMR regulations. The Agency
expects to issue final regulations for the CMR, and if necessary the
PMR, by August 1998. This time-frame for regulatory support for PMR
should not pose a hardship for the States or PWSs. It will take some
time for many States to comply with the statutory pre-requisites for
granting PMR to its public water systems (i.e., approval of a Source
Water Assessment Program, completion of the relevant source water
assessments, and approval of a PMR program). The Agency would expect
necessary federal and State regulations to be in place well in advance
of PMR implementation.

C. Suggestions for Regulatory Burden Reduction Other Than Chemical
Monitoring Reform

As part of the President's initiative to ``Reinvent Environmental
Regulation'', EPA has been reviewing the National Primary Drinking
Water Regulations (NPDWRs) to find opportunities for reducing the
paperwork burden on public water systems and State drinking water
agencies. Through public meetings, EPA has solicited input from States,
water utilities, and environmental groups regarding ways to reduce this
paperwork burden. That

[[Page 36102]]

process looked at all of EPA's NPDWRs and yielded a number of
suggestions. Many of the suggestions made by these ``stakeholders'' are
incorporated in the Chemical Monitoring Reform approach in this
document. Some of the suggestions, however, were to make changes to
other parts of the NPDWRs.
EPA believes certain other suggestions deserve further
consideration, and is presenting these suggestions for comment, so the
Agency can more fully evaluate their merits for possible inclusion in
subsequent proposed rulemaking. The suggestions contained in this
document involve deadlines for decisions regarding ground water under
the direct influence of surface water and associated filtration
determinations, and requirements for water system and State reporting.
They can be found in Section III.Q. of this document. Stakeholder
suggestions pertaining to lead and copper requirements were presented
in the preamble for the proposal entitled, Maximum Contaminant Level
Goals and National Primary Drinking Water Regulations for Lead and
Copper, 60FR16348, April 12, 1996.

II. Background

A. Statutory Authority

The approach outlined in this document would amend the monitoring
requirements associated with certain National Primary Drinking Water
Regulations (NPDWRs) established pursuant to Section 1445 of the Safe
Drinking Water Act, as amended August 6, 1996 (the ``Act''). Section
1445 of the Act provides EPA with general information collection
authority. Namely, ``every person who is subject to any requirement of
this title ..., shall establish and maintain such records, make such
reports, conduct such monitoring, and provide such information as the
Administrator may reasonably require by regulation to assist the
Administrator in ... determining whether such person has acted or is
acting in compliance with this title.''

B. Regulatory Background

EPA first regulated chemicals in drinking water by establishing
maximum contaminant levels (MCLs) and sampling requirements for nine
inorganic chemicals (IOCs), and six synthetic organic chemicals (SOCs)
in the Interim Primary Drinking Water Regulations of 1975. In
accordance with the Safe Drinking Water Act Amendments of 1986, EPA
began adding to its list of regulated chemicals. In 1987, EPA adopted
standards for eight volatile organic chemicals (VOCs) in the Phase I
Rule. From that point on, regulations for contaminants in drinking
water have been referred to as National Primary Drinking Water
Regulations (NPDWRs).
EPA has since revised the standards for some chemicals, and
established new standards for other chemicals, in three separate
actions: Phase II Rule--January, 1991; Phase IIB Rule--July, 1991; and
Phase V Rule--July, 1992. These changes would affect sixty four (64) of
the chemicals for which NPDWRs have been established (13 IOCs, 30 SOCs
and 21 VOCs) as listed below in Table A.

Table A.--Contaminants Affected by Chemical Monitoring Reform
------------------------------------------------------------------------

-------------------------------------------------------------------------
Inorganic Chemicals (IOCs):
[1] Antimony, [2] Arsenic, [3] Asbestos, [4] Barium, [5] Beryllium,
[6] Cadmium, [7] Chromium, [8] Cyanide, [9] Fluoride, [10] Mercury,
[11] Nickel, ^{1 [12] Selenium, [13] Thallium.
Synthetic Organic Chemicals (SOCs):
[1] 2,4-D (Formula 40 Weeder 64); [2] 2,3,7,8-TCDD (Dioxin); [3]
2,4,5-TP (Silvex); [4] Alachlor (Lasso); [5] Atrazine; [6]
Benzo[a]pyrene; [7] Carbofuran; [8] Chlordane; [9] Dalapon; [10]
Di(2-ethylhexyl)adipate; [11] Di(2-ethylhexyl)phthalate; [12]
Dibromochloropropane (DBCP); [13] Dinoseb; [14] Diquat; [15]
Endothall; [16] Endrin; [17] Ethylene dibromide (EDB); [18]
Glyphosate; [19] Heptachlor epoxide; [20] Heptachlor; [21]
Hexachloro-cyclopentadiene; [22] Hexachlorobenzene; [23] Lindane;
[24] Methoxychlor; [25] Oxamyl (Vydate); [26] Pentachlorophenol;
[27] Picloram; [28] Polychlorinated Biphenyls (PCBs); [29]
Simazine; [30] Toxaphene.
Volatile Organic Chemicals ( VOCs ):
[1] 1,1-Dichloroethylene; [2] 1,1,2-Trichloroethane; [3] 1,1,1-
Trichloroethane; [4] 1,2,4-Trichlorobenzene; [5] 1,2-
Dichloropropane; [6] 1,2-Dichloroethane; [7] Benzene; [8] Carbon
tetrachloride; [9] cis-1,2-Dichloroethylene; [10] Dichloromethane;
[11] Ethylbenzene; [12] Monochlorobenzene; [13] o-Dichlorobenzene;
[14] p-Dichlorobenzene; [15] Styrene; [16] Tetrachloroethylene;
[17] Toluene; [18] trans-1,2-Dichloroethylene; [19]
Trichloroethylene; [20] Vinyl Chloride; [21] Xylenes.
------------------------------------------------------------------------

When EPA published the Phase II rule in January, 1991, it
established the Standard Monitoring Framework. This framework is in
effect today, and applies to all chemicals regulated under the Phase I,
II, IIB and V rules, including those regulated under previous IPDWRs--
except arsenic. ² The Standard Monitoring Framework was
intended to provide a uniform monitoring structure for all current and
subsequent NPDWRs involving chemical contaminants. However, it soon
became apparent that the Standard Monitoring Framework (a) could be
redesigned to identify contaminated drinking water more quickly and
effectively, (b) is too prescriptive in several areas, and (c) is
complex and difficult to implement efficiently.
---------------------------------------------------------------------------

\1\ Although the MCL for Nickel has been stayed by a Federal
Court, the monitoring requirements remain in force.
\2\ Arsenic was excluded from the Standard Monitoring Framework
at the time Phase II was promulgated, because revision of the
arsenic MCL was thought to be imminent at that time. As indicated by
Table A, these changes include arsenic.
---------------------------------------------------------------------------

It also appears that the high rates of water supply contamination
anticipated in the late 1980s and early 1990's, upon which the Standard
Monitoring Framework is largely based (e.g., EPA cited VOC
contamination of about 20% of the water systems), have not been borne
out by the sampling results since then. According to the data in EPA's
national data base for tracking violations (the Safe Drinking Water
Information System--SDWIS), an average of about \1/2\% or less of the
systems that sample for the sixty four chemicals, had MCL violations
for any one of those chemicals during 1993-1995.³ Although
the data available to EPA are not definitive, they are significant
because they represent thousands of systems. EPA invites the submittal
of sampling data to support or refute the preliminary findings upon
which these changes are based.
---------------------------------------------------------------------------

\3\ Based on 28 States reporting.
---------------------------------------------------------------------------

(1) Monitoring Results from Phase I Unregulated Contaminants in 1988-
1991
The following discussion presents chemical occurrence data that EPA
States gathered from public drinking water systems. The sampling
results from thirty three States \4\ were compiled

[[Page 36103]]

for fourteen organic chemicals. These chemicals were sampled as
unregulated contaminants under the Phase I rule in 1988 through 1991 (A
Statistical Survey of the Unregulated Contaminant Data, prepared by
Computer Sciences Corporation). Twelve VOCs have since been regulated,
and EDB and DBCP have since been regulated as SOCs, under the Phase II,
IIB or V rules.
---------------------------------------------------------------------------

\4\ The following States, Territories and home rule
jurisdictions contributed data: Alabama, Arkansas, Colorado,
Delaware, District of Columbia, Florida, Georgia, Hawaii, Illinois,
Indiana, Louisiana, Maryland, Massachusetts, Missouri, Nebraska,
Nevada, New Jersey, New York, North Carolina, North Dakota, Ohio,
Oklahoma, Pennsylvania, Rhode Island, South Carolina, Tennessee,
Texas, Vermont, Virginia, Virgin Islands, Washington, West Virginia
and Wyoming. The underlined States reported only results showing
detection. They are included here because the data were taken from a
table in which the sampling results were consolidated for all the
States and it was impossible to separate these States out.
---------------------------------------------------------------------------

For systems served by surface water, these data show that thirteen
of the fourteen contaminants were detected at less than 3% of the
facilities tested, and that the fourteenth contaminant
(dichloromethane) was detected at slightly more than 5% of the
facilities (see Table B). In ground water, the data show that only one
contaminant (tetrachloroethylene) was detected at more than 3% of the
facilities sampled (see Table C). In summary, only a small percentage
of the facilities sampled has detected any of these contaminants.

Table B.--Phase I Sampling Results of Organic Compounds in Surface Water
(1988-1991)
------------------------------------------------------------------------
Percent
Chemical name and (phase) No. sites No. sites w/ sites w/
sampled \5\ detects detects
------------------------------------------------------------------------
cis/trans-1,2-Dichloroethylene
(2)............................. 1,670 15 0.90%
Dichloromethane (5).............. 1,588 81 5.10
1,2-Dichloropropane (2).......... 1,581 5 0.32
Ethylbenzene (2)................. 1,526 15 0.98
Ethylene Dibromide [EDB] (2)..... 1,180 34 2.88
Dibromochloropropane [DBCP] (2).. 1,204 28 2.33
Monochlorobenzene (2)............ 1,531 5 0.33
o-Dichlorobenzene (2)............ 1,504 3 0.20
Styrene (2)...................... 1,496 4 0.27
Tetrachloroethylene (2).......... 1,579 32 2.03
Toluene (2)...................... 1,529 37 2.42
1,2,4-Trichlorobenzene (5)....... 1,119 0 0.00
1,1,2-Trichloroethane (5)........ 1,523 20 1.31
Xylenes (2)...................... 1,606 23 1.43
------------------------------------------------------------------------

Table C.--Phase I Sampling Results of Organic Compounds in Ground Water
(1988-1991)
------------------------------------------------------------------------
Percent
Chemical name and (phase) No. sites No. sites w/ sites w/
sampled detects detects
------------------------------------------------------------------------
cis/trans-1,2-Dichloroethylene
(2)............................. 12,798 205 1.60
Dichloromethane (5).............. 12,263 294 2.40
1,2-Dichloropropane (2).......... 12,213 42 0.34
Ethylbenzene (2)................. 12,219 107 0.88
Ethylene Dibromide [EDB] (2)..... 9,339 61 0.65
Dibromochloropropane [DBCP] (2).. 9,293 40 0.43
Monochlorobenzene (2)............ 12,215 14 0.11
o-Dichlorobenzene (2)............ 12,162 8 0.07
Styrene (2)...................... 12,092 29 0.24
Tetrachloroethylene (2).......... 12,349 447 3.62
Toluene (2)...................... 12,218 222 1.82
1,2,4-Trichlorobenzene (5)....... 11,535 16 0.14
1,1,2-Trichloroethane (5)........ 12,211 11 0.09
Xylenes (2)...................... 12,743 150 1.18
------------------------------------------------------------------------

As shown in Tables D and E, the rates of detection also vary from
State to State.\6\ In Table D, the detection of ethylene dibromide
(EDB) in ground water ranges from < 1% of the facilities sampled in 13
of 17 States to 3.4% of the facilities in North Carolina and 12.5% of
the facilities in Alabama. In Table E, the variation of ethylbenzene
detections in ground water ranges from less than 1% of the facilities
sampled in 12 of 20 States to 5%--5.5% in Alabama, Missouri and North
Carolina.

\5\ The report from which this data is taken describes a point
as the ``number of unique sample sites and collection points'' for
each water system.
\6\ States reporting only results showing detections have been
excluded from Tables D and E, because the data presented to EPA
allowed us to identify and delete these States. Otherwise, these
tables include data from the States in which laboratories reported
the results of analyzing one or more samples for these specific
analytes.

[[Page 36104]]

Table D.--Phase I Sampling Results for Ethylene Dibromide [EDB] in
Ground Water (1988-1991)
------------------------------------------------------------------------
Number of Number of Percent of
State name sites sites w/ sites w/
Sampled detects detects
------------------------------------------------------------------------
Alabama.......................... 160 20 12.50
Colorado......................... 18 0 0.00
Delaware......................... 132 0 0.00
Minnesota........................ 119 0 0.00
Missouri......................... 130 0 0.00
North Carolina................... 383 13 3.39
North Dakota..................... 374 0 0.00
Nevada........................... 33 0 0.00
New Mexico....................... 968 0 0.00
New York......................... 378 1 0.26
Ohio............................. 5,747 3 0.05
Pennsylvania..................... 359 6 1.67
Rhode Island..................... 159 0 0.00
South Dakota..................... 17 0 0.00
West Virginia.................... 97 0 0.00
Wyoming.......................... 247 0 0.00
------------------------------------------------------------------------

Table E.--Phase I Sampling Results For Ethylbenzene in Ground Water 1988-
1991
------------------------------------------------------------------------
Percent
State name No. sites No. sites w/ sites w/
sampled detects detects
------------------------------------------------------------------------
Alabama.......................... 160 8 5.00
Colorado......................... 30 0 0.00
Delaware......................... 130 0 0.00
Hawaii........................... 28 0 0.00
Maryland......................... 131 2 1.53
Minnesota........................ 117 1 0.85
Missouri......................... 264 14 5.30
North Carolina................... 384 21 5.47
North Dakota..................... 414 2 0.48
Nevada........................... 58 0 0.00
New Mexico....................... 1,217 10 0.82
New York......................... 519 0 0.00
Ohio............................. 5,747 22 0.38
Pennsylvania..................... 371 1 0.27
Rhode Island..................... 166 2 1.20
South Dakota..................... 17 0 0.00
Washington....................... 2,112 4 0.19
West Virginia.................... 97 1 1.03
Wyoming.......................... 247 9 3.64
------------------------------------------------------------------------

The data above have several shortcomings, including the fact that
they are not nationally representative. The reasons for this include
(1) five States reported only positive results, which are included in
Tables B and C,⁷ and (2) the laboratory sensitivity in
detecting each contaminant is unknown, but can be assumed to vary from
one State to the next. The first factor tends to skew the data in
Tables B and C to an uncertain extent in favor of higher detection
rates than are likely to be found in data representing a cross section
of systems. The effect of the second factor is unknown. Further, the
samples were probably not collected during the periods of greatest
vulnerability, and many VOCs may evaporate from surface water, which
may skew the results in favor of lower detection rates. Nevertheless,
this is one of the largest collections of data available today, and
provides substantial support for the initial conclusion that relatively
few systems are contaminated.
---------------------------------------------------------------------------

\7\ Florida, Indiana, Massachusetts, Michigan and Nebraska.
---------------------------------------------------------------------------

(2) Sampling Results for Organic Compounds From 1992-1994
Several States have volunteered compilations of their sampling
results for organic chemicals.⁸ A detailed presentation of
this data is available in the docket under, Sampling Results for
Organic Compounds from 1992-1994. These results indicate VOC
contamination rates that are significantly lower than those reported
from the Phase I data. This difference may be due to improved waste
solvent management practices mandated under the Resource Conservation &
Recovery Act (RCRA), and to the closure of many of the contaminated
wells identified by the Phase I monitoring.
---------------------------------------------------------------------------

\8\ Although States have been sampling for most of the IOCs for
20 years, few provided useful compilations. Most IOC occurrence is
geologically based, and therefore not subject to rapid change.
Today's notice would represent the first set of national drinking
water monitoring requirements to recognize and account for the
potential of IOCs to occur as a result of human activity.
---------------------------------------------------------------------------

An aggregation of these data for eleven States, ⁹ Table
F, shows that, for

[[Page 36105]]

a very high percentage of the several thousand sites sampled, none of
the organic chemicals affected by these changes was detected. Only
three VOCs were detected at more than 2% of the sites sampled (`boxed'
numbers in right column). Exceedance of the MCL averaged less than 1%
of the sampling points for each VOC (bottom row, second column from the
right).
---------------------------------------------------------------------------

\9\ Alabama, Alaska, Arkansas, California, Georgia, Kansas,
Massachusetts, Mississippi, New Jersey, New Mexico and Oregon are
the States that have volunteered data to the Association of State
Drinking Water Administrators (ASDWA). They do not necessarily
represent a valid cross section of all States, and the data for any
one State may not represent a valid cross section for that State,
but these data do represent the most complete and the most current
information that EPA has received.

Table F.--Aggregated VOC Compliance Sampling Data From Selected States
----------------------------------------------------------------------------------------------------------------
Percent of Percent of Total
Number of sites w/ sites w/ percent of
Chemical name and (phase) sites detects < detects > sites w/
sampled MCL MCL detects
----------------------------------------------------------------------------------------------------------------
Benzene (1)................................................. 41,742 0.52 0.11 0.63
Carbon Tetrachloride (1).................................... 41,531 0.45 0.16 0.61
cis-1,2-Dichloroethylene (2)................................ 38,404 1.11 0.02 1.13
1,2-Dichloroethane (1)...................................... 41,501 0.58 0.10 0.68
1,1-Dichloroethylene (1).................................... 41,514 0.78 0.08 0.85
Dichloromethane (5)......................................... 41,506 1.13 0.13 1.26
1,2-Dichloropropane (2)..................................... 40,778 0.29 0.01 0.30
Ethylbenzene (2)............................................ 41,240 0.54 0.00 0.55
Monochlorobenzene (2)....................................... 41,713 0.12 0.00 0.12
o-Dichlorobenzene (2)....................................... 41,313 0.10 0.00 0.10
p-dichlorobenzene (1)....................................... 41,326 0.39 0.00 0.39
Styrene (2)................................................. 36,455 0.13 0.00 0.13
trans-1,2-Dichloroethylene (2).............................. 41,453 0.14 0.00 0.14
Tetrachloroethylene (2)..................................... 41,789 3.34 0.61 3.96
Toluene (2)................................................. 41,233 1.05 0.01 1.06
1,2,4-Trichlorobenzene (5).................................. 36,388 0.09 0.00 0.09
1,1,1-Trichloroethane (5)................................... 41,523 2.61 0.00 2.62
1,1,2-Trichloroethane (5)................................... 40,990 0.09 0.02 0.11
Trichloroethylene [TCE] (1)................................. 41,803 3.11 0.59 3.70
Vinyl Chloride (1).......................................... 41,471 0.06 0.05 0.11
Xylenes (2)................................................. 41,059 0.97 0.00 0.97
----------------------------------------------------------------------------------------------------------------

In the 1991 regulation, EPA offered no estimate of drinking water
contamination by synthetic organic chemicals (SOCs), such as
pesticides. Table G presents data gathered from ten
States.¹⁰ Only three SOCs were detected at more than 2% of
the sites--Dibromochloropropane (DBCP), Di(2-ethylhexyl)-phthalate, and
Di(2-ethylhexyl)-adipate. Only DBCP and phthalate exceeded the MCL at
more than \1/2\% of the sites sampled. Virtually all the DBCP
detections were in California, where the product was produced and
heavily used into the 1970s. Many of the phthalate and adipate
detections are thought to be due to plasticizers leaching from plastic
laboratory equipment containers and tubing, rather than from source
water contamination.
---------------------------------------------------------------------------

\10\ Alabama, Alaska, Arkansas, California, Georgia, Kansas,
Mississippi, Oregon, New Jersey and Wisconsin are the States that
have volunteered data to the Association of State Drinking Water
Administrators (ASDWA). They do not necessarily represent a valid
cross section of all States, and the data for any one State may not
represent a valid cross section for that State, but these data do
represent the most complete and the most current information that
EPA has received.
---------------------------------------------------------------------------

As before, these data have several shortcomings, which include the
fact that they may not be representative of the nation. The reasons for
this are that (1) the data were volunteered by only a few States, and
(2) the detection levels vary among these States. The effect of these
factors is unknown. Also, it is unknown whether the systems for which
sampling results were reported are representative of those in each
State, or whether the sampling was targeted to the periods of greatest
vulnerability. Based on this information, EPA believes that relatively
few systems are contaminated with SOCs.

Table G.--Aggregated SOC Compliance Sampling Data From Selected States
----------------------------------------------------------------------------------------------------------------
Percent of Percent of Total
Number of sites w/ sites w/ percent of
Chemical name and (phase) sites detects < detects > sites w/
sampled MCL MCL detects
----------------------------------------------------------------------------------------------------------------
Alachlor (2)................................................ 8,798 0.13 0.00 0.13
Atrazine (2)................................................ 9,596 0.85 0.00 0.85
Benzo[a]pyrene (5).......................................... 6,074 0.26 0.00 0.26
Carbofuran (2).............................................. 8,214 0.28 0.00 0.28
Chlordane (2)............................................... 9,324 0.02 0.00 0.02
Dalapon (5)................................................. 7,161 0.47 0.00 0.47
Dibromochloropropane [DBCP] (2)............................. 10,187 2.95 1.36 4.32
Di(2-ethylhexyl)-adipate (5)................................ 4,573 2.01 0.00 2.01
Di(2-ethylhexyl)-phthalate (5).............................. 6,556 2.81 0.78 3.58
Dinoseb (5)................................................. 7,242 0.33 0.00 0.33
Dioxin [2,3,7,8-TCDD] (5)................................... 1,165 0.00 0.00 0.00
Diquat (5).................................................. 5,592 1.07 0.02 1.09
Endothall (5)............................................... 5,424 0.04 0.00 0.04
Endrin (5).................................................. 9,229 0.26 0.00 0.26
Ethylene Dibromide [EDB] (2)................................ 10,184 0.16 0.36 0.52

[[Page 36106]]

Glyphosate (5).............................................. 6,796 0.06 0.00 0.06
Heptachlor (2).............................................. 8,770 0.06 0.01 0.07
Heptachlor Epoxide (2)...................................... 8,773 0.13 0.02 0.15
Hexachlorobenzene (5)....................................... 7,651 0.01 0.00 0.01
Hexachlorocyclopentadiene (5)............................... 7,340 0.07 0.00 0.07
Lindane (2)................................................. 7,369 0.20 0.00 0.20
Methoxychlor (2)............................................ 9,224 0.09 0.00 0.09
Oxamyl (5).................................................. 7,626 0.01 0.00 0.01
Picloram (5)................................................ 4,602 0.02 0.00 0.02
Pentachlorophenol (2)....................................... 6,428 0.06 0.00 0.06
Polychlorinated Biphenyls [PCBs] (2)........................ 7,945 0.04 0.01 0.05
Silvex [2,4,5-TP] (2)....................................... 8,522 0.55 0.00 0.55
Simazine (5)................................................ 9,608 0.23 0.01 0.24
Toxaphene (2)............................................... 7,373 0.04 0.00 0.04
2,4-D (2)................................................... 8,739 0.47 0.00 0.47
Avg. % Detections........................................... ........... 0.46 0.09 0.54
----------------------------------------------------------------------------------------------------------------

In summary, EPA and the States have been discussing ways to reduce
unnecessary monitoring requirements and to use chemical monitoring
resources more efficiently since late 1992. EPA also sought input from
outside organizations through public forums. The sampling results
summarized above indicate that few systems are contaminated and that
contamination levels vary widely among States. EPA believes that public
resources can be used more efficiently by allowing States to focus on
contaminated systems and systems at relatively high risk of
contamination.

C. Overview of Approach for Chemical Monitoring Reform, Permanent
Monitoring Relief and Anticipated Impact on Systems and States

The approach outlined in this document would result in new
monitoring requirements, and refine the required laboratory practices
for the contaminants listed in Table A, above. ¹¹ These new
requirements would replace the current requirements for Inorganic
Chemicals (IOCs), Synthetic Organic Chemicals (SOCs) and Volatile
Organic Chemicals (VOCs). ¹² The new monitoring requirements
would be consolidated under one section (Sec. 141.23). The current
monitoring requirements for nitrate and nitrite ¹³ would
remain unchanged, but would be moved to Sec. 141.24(a). The maximum
contaminant levels (MCLs) and designations of best available technology
(BAT) would remain unaffected, as would the monitoring requirements for
unregulated contaminants. All the provisions for radionuclides would
remain unaffected, as would the requirements for lead and copper, for
total trihalomethanes, and for microbial contaminants. The quality
control criteria for chemical analyses would be consolidated in a
separate technical criteria document incorporated by reference into the
rule, as would the analytical methods and acceptance criteria for these
chemicals. Note that EPA may, in a separate action, reformat all of the
drinking water regulations in Part 141 and that would require the
citations to change accordingly.
---------------------------------------------------------------------------

\11\ The MCLs for these contaminants are listed under 40 CFR
141.11(b), 141.61(a), 141.61(c), 141.62(b)(1)-(6) and 141.62(b)(10)-
(15).
\12\ These requirements currently appear under Secs. 141.23(a)-
(c) and 141.24(f)-(k).
\13\ Currently under Secs. 141.23(d) and 141.23(e)
---------------------------------------------------------------------------

Chemical Monitoring Reform is based on six concepts. (1) Some
systems are not sampling at the appropriate time of year or with
sufficient frequency to detect significant levels of contamination.
Several reports, including a U.S.G.S. study of the Mississippi River
Basin, entitled Contaminants in the Mississippi River, 1987-1992,
U.S.G.S. Circular 1133, 1995, and an Environmental Working Group (EWG)
study entitled Weed Killers by the Glass, document springtime peaks in
pesticide contamination of surface water supplies. (2) The percentage
of systems that are contaminated is very low. The sampling data that
support this view are summarized under Regulatory Background and
included in the record for this document. (3) Public resources should
be focused more on the systems that are contaminated or at risk of
contamination, and less on systems that have low risk of contamination.
(4) Because of their first hand knowledge of each system's operating
environment and vulnerability, States are better able than EPA to
determine which systems are at risk of contamination and which are not.
For the same reason, States are also better able to determine the time
of year and frequency of sampling that are most likely to detect
contamination at its highest levels. (5) Source water protection
measures should be expanded to minimize the number of systems
contaminated in the future. (6) The current requirements are complex,
and should be streamlined.
EPA is considering addressing these concepts by revising the
Federal monitoring framework, within which States operate, to provide
States the flexibility to focus their resources on systems at risk of
contamination. This would be accomplished by consolidating the baseline
sampling requirements for all contaminants and all classes of systems
into a single five year frequency, except for the `at risk' systems.
States would be assigned the responsibility to review the vulnerability
of all their systems, and to schedule the `at risk' systems to sample
more frequently than once every five years based on the degree of each
system's vulnerability. Further, all systems would generally be
required to sample during the periods of greatest vulnerability as
directed by the State. This would reduce the chance of contamination
going undetected, and hence unaddressed.
The development of these system-specific sampling schedules will
typically involve the identification of potential contamination
source(s) and an assessment of contaminant use patterns and the
resulting periods of greatest vulnerability based on the management of
those sources and intervening hydrogeologic or climatic features. This
targeting, assessment and

[[Page 36107]]

scheduling activity closely parallels the efforts required under State
Source Water Assessment Programs,¹⁴ and can be accomplished
most efficiently by conducting a single assessment under both programs.
---------------------------------------------------------------------------

\14\ State Source Water Assessment Programs are mandated under
section 1453 of the Act.
---------------------------------------------------------------------------

If the monitoring results for the systems sampling every five years
are below \1/2\ of the MCL, the systems would continue sampling every
five years. If their sampling results are equal to or above \1/2\ of
the MCL, the systems would sample more frequently as directed by the
State.
In their primacy applications to adopt Chemical Monitoring Reform,
States would describe their programs to screen all systems and identify
and schedule ``at risk'' systems for increased sampling, to determine
the periods of greatest vulnerability, and to determine whether and how
to schedule increased sampling for systems exceeding the trigger level.
These State program descriptions would then undergo public review and
comment, before their submittal to EPA for approval. EPA's review of
the primacy applications would assure that each State has an effective
plan, and the legal authority, to implement these provisions. As a last
resort, EPA may intervene to schedule increased sampling for individual
systems at risk of contamination, if the State fails to act. Table H
highlights the main features of the Chemical Monitoring Reform approach
in a flow chart.
BILLING CODE 6560-50-P

[[Page 36108]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.000

BILLING CODE 6560-50-C

[[Page 36109]]

EPA expects that States will take advantage of the simplicity of
today's approach. Table I illustrates the current sampling requirements
starting in 1996 for most systems.¹⁶ There are different
sampling frequencies for IOCs, SOCs and VOCs. For IOCs and VOCs, the
requirements vary by type of source water i.e., surface water or ground
water. For SOCs, the requirements vary by size of system i.e., larger
or smaller than 3,300. As with Chemical Monitoring Reform, these
requirements apply to each sampling point, and many small systems have
three or four sampling points.

\16\ Under the Phase V rule, systems serving < 150 service
connections are not required to begin sampling for the Phase V
contaminants until 1996. However, almost all States incorporated the
sampling schedules of these systems into the sampling schedules
under Phase II, which began in 1993, in the interests of
administrative simplicity.
---------------------------------------------------------------------------

BILLING CODE 6560-50-P
[GRAPHIC] [TIFF OMITTED] TP03JY97.001

BILLING CODE 6560-50-C

[[Page 36110]]

Before EPA finished developing these changes for Chemical
Monitoring Reform, Congress enacted the 1996 Amendments to the Safe
Drinking Water Act. The Amendments that are functionally related to
Chemical Monitoring Reform are discussed in Section III.M. Through one
of these amendments,¹⁷ Congress authorized States that have
received EPA approval of their Source Water Assessment Programs to
offer Permanent Monitoring Relief to public water systems. The systems
must have completed their source water assessments under the State
program to be eligible for Permanent Monitoring Relief. ¹⁸
Congress also directed EPA to publish guidelines by August 6, 1997, for
States to follow in developing their PMR requirements.
---------------------------------------------------------------------------

\17\ See section1418(b).
\18\ See section 1453(a)(3)
---------------------------------------------------------------------------

The new requirements of Chemical Monitoring Reform would be
complemented by the draft Permanent Monitoring Relief (PMR) Guidelines
in Section III.N., which will allow States to offer additional
monitoring relief under specific conditions. Under the draft PMR
guidelines in this document, States could allow systems to forgo
monitoring of individual chemicals at specified sampling points during
a five year period, either by granting a waiver, or by allowing the use
of surrogate sampling results from other points. Systems could also be
allowed to reduce the sampling frequency for nitrate under limited
circumstances. In all cases, the State would make system-specific
determinations in accordance with the PMR Guidelines.
The draft PMR guidelines provide, generally, that systems that
qualify for waivers will be those with long records of no detection,
and for which a vulnerability assessment unambiguously shows that the
system is not at risk of contamination. Monitoring results from a
sampling point(s), or from a group of points, that are used as
surrogates for the results from other sampling points, will be from
samples of the most vulnerable portion of the same source water serving
all of the sampling points. Reduced nitrate sampling will be allowed
only where the sampling results over a long period are very low and the
State determines that the prognosis is for more of the same.

D. Anticipated Impact on Systems and States

EPA expects that States will support this approach because it
provides flexibility to allocate more of their resources to
contaminated systems and systems at risk of contamination, by one or
more chemicals, and to reduce the monitoring burden for those systems
where specific chemicals do not pose a risk to public health. For
example, the same system may be at risk of contamination by certain
pesticides, but have a very low risk of contamination by the other
chemicals. By reducing the sampling burden at that system to one sample
every five years for the low risk chemicals, the State can often `buy
enough economic elbow room' to increase the sampling frequency for the
high risk pesticides without imposing a significant net increase in
monitoring burden. In many cases, even where the sampling for one or
more contaminants under a single laboratory method is increased, the
net effect for the system may be a decrease in overall sampling costs.
EPA believes that most systems, including very small systems, would
have a net decrease in sampling burden and cost and that only a small
percentage of systems would have a net increase in sampling burden.
Further, that net increase would occur only where the State assessment
of public health risk indicates that the increase is warranted as an
appropriate response to identified risk. For States, EPA believes that
the net program burden would also be reduced, because the aggregate
reduction in sampling frequencies would reduce the burden of tracking
compliance with the sampling requirements, even though States would be
required to develop plans for identifying at risk systems. This net
reduction in sampling cost for the 64 chronic contaminants may provide
further ``elbow room'' for systems and States to concentrate on higher
priority contaminants. EPA seeks comment on this summary of the net
effect of today's approach on system and State program burden.

III. Detailed Explanation of Draft Changes to Chemical Monitoring
Requirements

A. Affected Water Systems

Under Sec. 141.23 of these changes, the chemical monitoring
requirements would apply to all community water systems (CWSs) and non-
transient, non-community water systems (NTNCWSs); this is the same as
the current rule. Community water systems are those which serve at
least 15 service connections used by year round residents, or regularly
serve at least 25 year round residents e.g., cities, townships,
district water authorities, and private water companies serving such
communities. Non-transient, non-community water systems are those which
are not community water systems, and which serve at least 25 of the
same persons over six months of the year e.g., schools, factories or
other facilities with their own separate water supply. Henceforth in
this discussion, CWSs and NTNCWSs will be referred to collectively as
water systems or systems.

B. Sampling Points

Under Sec. 141.23(a) of these changes, all water systems would
sample at each entry point to the distribution system after treatment.
Under the PMR guidelines, exceptions to this may be allowed. However,
some States may require sampling at each source water withdrawal point
in order to quickly identify contaminated sources and initiate remedial
action. States could establish alternative or additional sampling
points, as long as the water delivered to the consumer is tested; this
is the same as the current rule.
In addition, systems would sample at any sampling point the State
designates in addition to the entry point to the distribution system.
For example, systems may be vulnerable to contamination from the
asbestos cement pipes in the distribution system, or to infiltration
where leaking solvents have dissolved portions of polyvinyl piping.
States could address these situations by determining where systems must
sample in addition to the entry point to the distribution system.

C. Time of Monitoring

Under Sec. 141.23(b) of these changes, sampling would generally be
conducted during the periods of greatest vulnerability, according to a
schedule specified by the State. Periods of greatest vulnerability mean
the periods during which contamination is most likely to occur at the
highest concentration at a particular sampling point, based on the
history of relevant factors for that sampling point e.g., U.S. Weather
Bureau rainfall averages, local pesticide application practices.
Under the current requirements, systems must sample according to
nationally uniform schedules, based on prior sampling results and other
factors (see 56 FR 3600-3612, January 30, 1991). The most frequent
sampling is quarterly, which is designed to account for the seasonal
variation in contaminant concentrations. Systems may satisfy this
requirement by sampling at any time during each quarter. If systems are
not sampling quarterly, they are sampling annually, triennially or less
frequently, depending

[[Page 36111]]

on the type of system and the contaminant.
Because the current requirements do not specify the time of
sampling more precisely, contamination may go undetected; this is
especially true for systems served by surface water (particularly river
systems), or by ground water under the direct influence of surface
water. For example, pesticides are typically applied during the Spring
and Summer months and a high frequency series of sampling results from
surface water systems during this period may show frequent spikes of
contamination from runoff. However, a system sampling in early April
may miss the contamination, and have a false sense of security about
the safety of its drinking water.
Today's approach would remedy this potential problem by assigning
States the responsibility to schedule sampling during the periods of
greatest vulnerability. This responsibility would require States to use
sound science in assessing local patterns of contaminant use, where
there are systems susceptible to significant seasonal variation in
contaminant levels. The State set asides that are available from the
new State Revolving Fund established under the 1996 Amendments to the
Safe Drinking Water Act, for conducting source water assessments, could
be used to assist States in making these determinations. EPA expects
that the State schedules would evolve toward greater precision based on
State experience and the growing knowledge of local industrial and
agricultural practices.
There has been some concern expressed about the workload impact on
the capacity of laboratories to handle a large number of samples in a
short period of time. Two factors mitigate this issue. First, the
number of systems that are scheduled to sample more frequently than
once every five years should not be great, and sampling for the other
systems, which constitute the great majority of systems, can be divided
over a five year period i.e., only a fifth of the systems under the
five year sampling frequency would sample each year. Second, many
systems (about 80%) are served by water supplies that are not subject
to significant fluctuation over time e.g., deep ground water systems in
geological settings other than fractured bedrock. States could schedule
these systems to sample at different periods than the surface water
systems (i.e., Autumn and Winter) to further balance the work load.
EPA intends to prepare technical guidance, in consultation with the
States, to assist them in scheduling sampling during the periods of
greatest vulnerability, if this approach is promulgated.

D. Responsibility to Provide Information

Under Sec. 141.23(c) of these changes, systems would be required to
provide any information requested by the State. States may need
information they do not have in their files to decide whether a system
should sample more frequently than every five years. Failure by a
system to provide this information would be cause for the State to
schedule the system for increased sampling.
The requirement to report all sampling results, including
detections and non-detections, would be continued. EPA would clarify
this provision by specifying that detections equal to or greater than
the laboratory's MDL ²³ must be reported as detections. The
reporting of detections is necessary because, if contamination is
detected, it means the sampling point is vulnerable to contamination.
States need this information for determining which systems may need to
sample more frequently than every five years.
---------------------------------------------------------------------------

\23\ Method Detection Limit (MDLs) are defined under 40 CFR Part
136, Appendix B.
---------------------------------------------------------------------------

EPA recognizes that some detections at the MDL may be incorrectly
identified as to the chemical involved, owing to the difficulty of
qualitatively characterizing contamination at that level. This is a
general problem that can occur at any level, and that gets worse as the
level of contamination gets lower i.e., closer to the MDL. But, it is
also true that detection at the MDL means there is chemical
contamination in the sample. States could recommend that systems direct
their laboratories to use qualitative confirmation techniques to verify
or invalidate all detections (see Methods Development and
Implementation for the National Pesticides Survey, Munch, D.J., Graves,
R.L., Maxey, R.A., and Engel, T.M., Environmental Science Technology,
Vol. 24, No. 10, 1990. pp.1450-1451).

E. Mandatory Monitoring

Under Sec. 141.23(d)(1) of these changes, as Chemical Monitoring
Reform is implemented, systems would sample according to schedules
specified by the State. If the State has made a screening decision and
informed the system that the State will not specify a schedule for
increased monitoring, the system would sample at least once every five
years at each sampling point and this sampling would be conducted
during the periods of greatest vulnerability as determined by the
State. If the State does not specify a period of greatest
vulnerability, the system is responsible for doing so, and must
describe to the State its risk-based reasons for the period it
specified. For example, a system might sample at an appropriate time in
May because it knows that is the peak period of pesticide application.
Sampling during the period of greatest vulnerability may require
some systems to perform the same test more than once. This is because
contaminants may have different periods of vulnerability and if they
are covered by the same analytical method, the same test would have to
be repeated. EPA seeks comment on whether this multi-period sampling
might impose a significant burden, and if it would, specific examples
of the burden and concrete proposals as to what might be done to reduce
the burden while maintaining the capacity to monitor during vulnerable
periods.
This approach is different than the current requirements, under
which the systems must sample according to a nationally uniform
schedule (see 40 CFR, Secs. 141.23 through 24). There are four reasons
why EPA is considering moving from current monitoring requirements to
relying on States to schedule system specific monitoring requirements.
First, States have gained a far more complete understanding of
drinking water quality as it is affected by these chemicals. Today,
most systems have completed several rounds of sampling or they have
been granted sampling waivers based on the State's assessment of their
vulnerability to contamination. States have established a base of
information and experience related to the local conditions of
individual water systems within each State that did not exist in 1991.
Therefore, the level of detail in the current Federal monitoring
requirements may no longer be necessary.
Second, the compliance sampling results available today indicate
that the number of drinking water sources contaminated with one of the
chemicals affected by these changes is very low. As noted in the
background discussion, the contamination of public water systems by any
of the regulated organic chemicals in the systems for which sampling
data was provided ranges from 5% to less than 0.5 %, and averages less
than 1%.
Third, the current monitoring requirements are complex, as
illustrated in the monitoring decision diagram in Table I, above. This
complexity is the result of establishing nationally uniform monitoring
requirements that account for the differences among types and sizes of
systems and contaminants.

[[Page 36112]]

There are sixty four chemicals, thirty two trigger levels, two types of
source water (surface water, ground water), and two sizes of systems
(greater than 3,300, less than 3,300).
Fourth, the current monitoring requirements assume that all systems
are vulnerable to contamination, and require each system to sample at
relatively high frequencies, unless the State reduces the sampling
frequency by granting a sampling waiver. In order to provide relief to
systems that are not vulnerable, many States have invested resources to
design and implement sampling waiver programs. That investment will now
assist them to narrow the focus to those water systems that are already
contaminated or at risk of contamination.
Rather than initially presuming vulnerability of all systems,
States' screening review should be neutral, but looking to good
scientific data from State waiver programs, wellhead protection
programs, source water assessments, and the like for a reasonably
substantive basis to place systems in the ``at risk'' or ``not at
risk'' categories. Under today's approach, States would now review the
vulnerability of their systems to identify those with an apparent risk
of contamination. States would schedule these systems for increased
sampling according to the degree of their vulnerability. This would
relieve those systems that are not contaminated, and that have little
risk of contamination, of current burdens and complexity by
consolidating and reducing the standard sampling frequency for all
contaminants and all classes of systems to a minimum of one sample
every five years. This will reduce the State resource burden enough to
allow States to focus on systems that need to sample more frequently
than every five years.
EPA believes the five year sampling period is protective of public
health, because the sampling will be conducted during the periods of
greatest vulnerability, because the States will target those systems
that are contaminated or at risk of contamination to sample at a
greater frequency and because the MCLs of the contaminants affected by
these changes are based on chronic health effects, which for most of
the contaminants covers a seventy year period. The five year period has
the advantages of coinciding with several periodic, important bases for
developing data that will inform State determinations, including: (1)
the five year time of travel adopted by many State Wellhead Protection
Programs (WHPPs) for delineating Wellhead Protection Areas and Source
Water Protection Areas, (2) many State schedules for conducting
sanitary surveys at small water systems, and (3) the cycle for updating
section 305(b) reports which inventory the quality of the nation's
surface waters.
The chemical monitoring reform work group considered other time
periods for the frequency of the default sampling period and chose five
years for the reasons mentioned above. EPA seeks comment on whether the
Agency should propose a shorter or longer time period and, if so, why.
EPA is considering default sampling periods ranging from every three
years to six years.
Shorter periods, such as three years, may appear to provide
nominally more protection than the five year period, but would require
more State resources to administer compliance with the shorter time
frame and to respond to a higher demand for waivers than would be the
case under the five year period. In most cases, these additional
resources would be diverted from working on high priority water systems
i.e., those that are already contaminated or at risk of contamination.
Thus, it is not clear that a shorter time frame would automatically
result in greater protection.
Six years may appear to provide more relief than five years for
systems that have little risk of contamination. That would require
additional State resources to develop adequate information for the
``not at risk'' determinations because, as noted above, most State
Wellhead Protection Programs are referenced to a five year time of
travel.

F. Detection \1/2\ MCL

Under Sec. 141.23 (e) through (f) of these changes, if any
contaminant were detected at a level equal to or greater than \1/2\ of
the MCL, the system would sample according to a schedule specified by
the State. This trigger level was selected by considering the need to
provide an adequate margin of safety in identifying potential MCL
exceedances before they occur, the capability of laboratories across
the country to identify contamination below the MCL, and the need to
simplify the current requirements.
When contamination is detected \1/2\ of the MCL, States
would determine the level of additional monitoring required to fully
characterize the contamination. This deference to State discretion, in
scheduling the follow up sampling based on local circumstances, is more
effective than the current provisions at detecting MCL exceedances
because the sampling schedule most likely to accurately characterize
contamination depends on the history of sampling results at the
sampling point and neighboring points, the susceptibility of the water
supply to contamination, the most vulnerable periods of contamination,
and local commercial practices. Today's approach would require States
to consider those factors in establishing follow up sampling schedules.
Today's approach would also require systems that exceed the MCL to take
at least one sample during each of the following three quarters. And,
whenever the levels of contamination may vary significantly during a
quarter, the sampling schedule would have to account for the expected
frequency and amplitude of that variation.
Under the current monitoring requirements, any system that exceeds
the trigger level must sample every quarter. There is no requirement
for systems to follow up more quickly to characterize the contamination
and there is no requirement for systems to sample during the periods of
greatest vulnerability. Therefore, systems could mischaracterize the
extent of contamination under the current requirements.
The trigger level in these changes can be explained far more easily
than the trigger levels under the current monitoring requirements,
because the new trigger level would always be based on the potential
for exceeding the MCL. This will enhance the ability of States and
systems to assess MCL compliance, by focusing on the risk of MCL
exceedances, rather than trying to figure out which trigger level
applies to which contaminant.
Under the current requirements, the trigger level for organic
chemicals is detection. For VOCs the detection limit is 0.5g/
l, and for SOCs the EPA specified detection limit varies by
contaminant. Thirty nine percent of the trigger levels for all organic
chemicals are less than 1% of the MCL and fifty three percent of them
are less than 5% of the MCL. Because all sampling under today's
approach would be scheduled during the periods of greatest
vulnerability, the sampling results would reflect the worst case level
of contamination. Additionally, all detections must be reported to
States under today's approach. While it is true that detection
indicates a path of contamination, most water supplies are not subject
to dramatic fluctuations in contamination levels and such low level
detections rarely signal imminent exceedance of the MCL, at least in
monitoring samples taken during the time of greatest vulnerability.
Therefore, setting the trigger level at \1/2\MCL would be protective of
public health, and would minimize the chances of

[[Page 36113]]

undetected MCL exceedances during other times of the year.
Under the current requirements for inorganic chemicals (IOCs),
systems do not have to begin quarterly sampling until the contaminant
exceeds the MCL. This approach would be protective for naturally
occurring contaminants, because the natural levels of fluctuation are
usually slight and slow to change. However, when these chemicals
contaminate water supplies as a result of human activity, the levels of
fluctuation and time periods involved tend to mimic those of organic
chemicals. Since virtually all of the IOCs can occur as a result of
human activity, it would be more protective to establish a trigger
level below the MCL for these contaminants.
In summary, it is EPA's view that the trigger level in these
changes would: (1) establish a uniform, understandable and practical
criterion for increased sampling that is protective of public health;
and (2) strike a reasonable balance between responding to contamination
at very low levels, and taking no action until a contaminant has
exceeded the MCL.
EPA is, however, seeking comment on alternatives for proposing the
trigger levels, recognizing that there is no perfect level for any one
contaminant under all circumstances. Three of the possible alternatives
are: (1) \1/2\ of the MCL or the practical quantitation level
(PQL),²⁴ whichever is higher; (2) detection of the
contaminant; and (3) requiring use of the most sensitive methods.
---------------------------------------------------------------------------

\24\ The PQL is the lowest concentration at which a contaminant
can be reliably measured.
---------------------------------------------------------------------------

(1) Trigger=\1/2\MCL or the PQL, Whichever Is Higher
This option would have the benefit of not requiring State action
until the PQL has been exceeded. This means there would be a reasonable
degree of certainty that a quantifiable level of contamination has
actually occurred before the State would undertake its review to
establish a sampling frequency based on the specifics of the
contamination. For twenty five contaminants,²⁵ however, the
PQL equals the MCL. Therefore, this option has the potential problem of
inadequately characterizing, and failing to responding to,
contamination until it has exceeded the MCL.
---------------------------------------------------------------------------

\25\ Antimony, Thallium, Alachlor, Benzo[a]Pyrene, Chlordane,
Dibromochloropropane, Di(2-ethylhexyl)phthalate, Ethylene-dibromide,
Heptachlor, Heptachlor Epoxide, Hexachlorobenzene, Lindane,
Polychlorinated Biphenyls, Pentachlorophenol, Toxaphene, Dioxin,
Benzene, Carbon Tetrachloride, 1,2-Dichloroethane, Dichloromethane,
1,2-Dichloropropane, Tetrachloroethylene, 1,1,2-Trichloroethane,
Trichloroethylene, Vinyl Chloride.
---------------------------------------------------------------------------

(2) Trigger=Detection
This option would offer the benefit of providing earlier warning of
contamination than the options at higher levels. However, a trigger
lower than \1/2\ of the MCL may not provide a real benefit in
identifying potential MCL exceedances, because contaminant levels
generally take many months to change significantly. Because the time of
greatest vulnerability generally indicates the maximum level of
contamination, this option would have the drawback of triggering many
State reviews where MCL exceedances are unlikely, and would therefore
impose a burden on States that may be unwarranted.
This option also raises the issue of defining a detection.
Detection should be the lowest concentration at which a laboratory can
consistently detect, and correctly identify, individual contaminants in
a variety of drinking water samples. Detection is more difficult in
dirty water than in clean water. Detection is also determined by other
variables, including the sensitivity of the analytical method used for
measurement, the sophistication and age of the laboratory testing
equipment, and the training and expertise of the laboratory staff.
Therefore, detection will vary by laboratory and by system. EPA has not
established SOC detection criteria for laboratory certification. That
issue is being addressed under the new laboratory performance
requirements described below in section III.J.
(3) Require Analytical Methods With the Most Sensitive Detection Levels
Under this option, laboratories would be required to use the most
sensitive analytical laboratory method for each contaminant. This may
offer some assurance of early detection of low level contamination.
However, many labs would be required to purchase new equipment to run
these methods. This would raise the cost of the drinking water program
for all systems, and could create a lab capacity problem, if many labs
are unable to secure the necessary funding i.e., there would be fewer
certified laboratories (and possibly an inadequate number) to conduct
compliance analyses. As more contaminants become regulated, more new
equipment would have to be purchased. That would further raise the cost
of the program, and could make the lab capacity problem worse. Finally,
due to the variability of laboratory expertise, some laboratories using
the most sensitive methods may operate at higher (less sensitive)
detection levels than are routinely achieved by other laboratories with
more skillful personnel, who are using ostensibly less sensitive
analytical methods.
G. MCL Violation Determinations
Under Sec. 141.23(g) of these changes, all MCL violations would be
determined by the average annual concentration of the contaminant. This
is very similar to the current provisions for determining violations
when the system has been sampling at a quarterly frequency i.e., MCL
violations are based on the running annual average of the prior year's
sampling results. Under today's approach, all MCL violations would be
determined by the average of four consecutive quarterly values,
beginning with the quarter in which the initial MCL exceedance occurs.
²⁶ The States would schedule the sampling in each subsequent
quarter to include the periods of greatest vulnerability during that
quarter. Each quarterly value would be determined by the time balanced
average of all samples taken in that quarter i.e., the State would
divide each quarter into equal segments, and use the average of the
sampling results from each segment to calculate the quarterly value. By
limiting the annual calculation to four quarterly values, we would
avoid skewing the annual average to the periods of highest sampling
frequency.
---------------------------------------------------------------------------

\26\ Sometimes, the MCL exceedance may occur at the end of a
quarter, and therefore, may not be representative of a time balanced
average of multiple samples taken throughout the quarter. In this
case, the State should choose to begin calculating the annual
average concentration in the quarter following the quarter in which
the initial MCL exceedance occurred, so that the MCL compliance
determination is based on four consecutive quarterly values that are
representative of each quarter.
---------------------------------------------------------------------------

For example, a State might divide a quarter into one month
segments. The State might then schedule only one sample during each of
the two months considered low vulnerability segments, and ten samples
(three days apart) during the month it considers to be the high
vulnerability segment. The ten samples from the high vulnerability
month would be averaged to provide a single data point for that
segment. The quarterly value would be the average of the three monthly
data points. The State may require only one sample during those
quarters in which the contaminant concentration is not expected to vary
significantly.
This process of segmentation would accomplish three objectives. (1)
It would yield an annual value representative of the average annual
contaminant concentration that includes

[[Page 36114]]

representation from the periods of highest concentration. (2) As
mentioned above, it would avoid unduly skewing the annual average to
the sampling results showing the highest concentrations. (3) It would
prevent systems from using periods of low concentration to load up on
the sampling results that would cast a downward bias onto the annual
average.
If the average of one or more quarters would cause the average
annual concentration to exceed the MCL, the system would be in
violation of the MCL from the end of that quarter. This assures that
compliance determinations would be made as soon as the average annual
contaminant levels can be established as > MCL, but not until then.
EPA also seeks comment on whether systems failing to comply with a
State schedule to characterize contamination after an MCL exceedance
should be required to notify the public of a potential MCL violation.
Specifically, EPA is considering a provision that would require any
system that has exceeded the MCL, and subsequently failed to comply
with a State schedule to fully characterize the average annual
contamination levels, to issue a public notice under Sec. 141.32 within
30 days of its failure to comply with the State sampling schedule.
This notice would include the health effects language under
Sec. 141.32 for the contaminant exceeding the MCL, and would further
state (a) that the MCL has been exceeded, (b) that an MCL violation is
based on the average annual level of contamination, (c) that the
sampling schedule to effectively characterize the average annual level
of contamination is based on local circumstances of contaminant
fluctuation, and (d) that the system has failed to comply with the
State sampling schedule to determine whether the system is in violation
of the MCL. Failure to issue a public notice in accordance with these
requirements would be a violation of the Safe Drinking Water Act.

H. Laboratory Certification Criteria

The quality control provisions associated with measuring the
chemicals covered by these changes, the approved analytical methods for
measuring compliance with the MCL, and the Performance Evaluation (PE)
acceptance limits for those contaminants, would be consolidated in EPA
Technical Criteria Document for the Analysis of Selected Chemical
Contaminants in Drinking Water (i.e. the EPA Technical Criteria
Document) incorporated by reference under Sec. 141.23(j). A copy of
this document is attached to this discussion as Appendix A, so the
reader may review its provisions in conjunction with the other
provisions of this document. This subsection would specify that all
samples must be analyzed by laboratories certified by EPA or by the
State, and that the State or EPA may suspend or revoke a laboratory's
certification for failure to achieve the prescribed operating
requirements and standards. This provision would supersede Sec. 141.28
for lab certification under Sec. 141.23.
The incorporation by reference of the EPA Technical Criteria
Document into the Federal Regulations means that the requirements in
the technical document would be part of the regulations and would be
fully enforceable. The reason for moving the laboratory provisions into
a separate document is that the audience for these requirements is
different than the audience for the general program monitoring
requirements. State program managers, their staff and EPA Regional
Office program coordinators are interested in the program requirements
described in these draft changes. The State laboratory certification
officers, State lab directors, EPA Regional Office laboratory
certification officers and private lab personnel are mainly interested
in the highly technical requirements pertaining to laboratory
measurement of chemicals. A technical manual is a much better format
for system technicians and laboratory analysts who need an operational
reference document.
With the exception of four changes described below, and highlighted
in the text of the criteria document (Appendix A), the laboratory
requirements in this document are the same as the current laboratory
requirements (see 40 CFR Sections 141.23-24). Since those provisions
have already undergone notice and comment, EPA is not opening those
provisions for further public comment today. EPA is describing the
current requirements in this preamble (1) So the reader can better
understand how today's approach would fit into the total structure of
laboratory requirements; and (2) because these requirements are being
consolidated from several parts of the current rule into the technical
criteria document identified above.
In a concurrent effort to the development of today's approach, EPA
has been reviewing several inexpensive methods for detecting and
measuring drinking water contaminants. These are generally referred to
as immuno-assays, or immuno-assay kits. They cost about $15 to $30 a
test, which is much less than some of the methods currently approved,
which can cost up to several hundred dollars. EPA requests comment on
the following concepts.
(1) EPA has long required that laboratories pass performance
evaluation (PE) samples within prescribed acceptance limits, but has
not specified a frequency for these tests. All States require labs to
pass these PE tests at least every year, and EPA believes that is an
appropriate requirement. These changes would adopt the universal State
requirement for laboratories to successfully analyze PE samples at a
minimum of once each year as provided by EPA, the State, or other
parties that have been approved by the State or EPA.
(2) Under the current requirements of EPA's methods, laboratories
using a method for the first time must calculate their method detection
limits (MDL) for each contaminant covered by that method. However,
there are no parameters for the time frame over which the MDL samples
must be analyzed. Therefore, EPA is considering proposing that the
extraction and analysis of the MDL samples must be performed over a
period of at least three days. This same procedure was adopted under
the Information Collection Rule (61 FR 24354, May 14, 1996), because
EPA believes that this procedure results in a more realistic MDL
determination.
(3) Under the current requirements of EPA methods, laboratories
must analyze a laboratory fortified blank (LFB) with each batch of
samples. LFBs are quality control samples of purified water with known
concentrations of certain contaminants (i.e., the regulated
contaminants affected by these changes) that are subjected to
laboratory analysis, as a check on the reliability of the results
produced from real world samples of unknown contaminant concentrations.
The requirements for LFBs are specified in the individual EPA methods,
which labs must follow. Most EPA methods require laboratories to
analyze LFBs at a concentration equal to ten times the method detection
limit (MDL), ten times the estimated detection limit (EDL), or at a
mid-point of the measurement calibration curve.
Under these changes, laboratories would have to analyze a subset of
these LFBs at the trigger level of \1/2\ of the MCL or less, and at the
level used to calculate the laboratory MDL. A record of the results of
each LFB would have to be maintained until the next State certification
audit or for five years--whichever is longer, and would be available to
the State upon request. States would make these records

[[Page 36115]]

available to EPA upon request. Generally, the analyses of LFBs at
specified concentrations would not affect the regulatory burden under
the current requirements, because those analyses must be performed
anyway and the cost of performing an analysis at one contaminant level
is usually the same as performing it at another level. However, EPA
seeks comment on whether running LFBs at these levels, which may be
lower than the current customary levels, would result in a significant
increase in the incidence of recalibrating or fine tuning the
laboratory measuring equipment and whether that would result in a
significant increase in laboratory operating costs.
The record of each laboratory's operational sensitivity at the
trigger level, and the level used to calculate the MDL, would serve the
following objectives. One, the records would provide a means for States
to assure that laboratory performance is sufficiently reliable to
protect public health. Two, a statistical analysis of these records
would provide the basis for States or EPA to establish uniform
performance criteria at these levels.
These changes would require laboratories to analyze an LFB at \1/2\
of the MCL or less at least once per week during any week in which
drinking water compliance samples are analyzed. This provision would
provide an ongoing check on the reliability of each laboratory's
ability to identify contamination at the trigger level. These changes
would also require laboratories to analyze at least one LFB per month
at the concentration that was used to calculate the MDL, during any
month in which drinking water compliance samples are analyzed. The
purpose of this is to maintain an ongoing record of each laboratory's
ability to detect low level contamination.
It is important to characterize what ``no detection'' means for
each laboratory, because the systems that contract with each laboratory
will be reporting all detections to the State. The States will be
making system targeting decisions and sampling waiver determinations
based in part on whether or not contamination has been detected at the
sampling point. For this reason, today's approach is considering
requiring laboratories, as a condition of certification, to maintain
records of these analyses in the format in paragraph IV of the
technical criteria document, at least until the next State
certification audit report has been completed.
(4) These changes would set the trigger for polychlorinated
biphenyls (PCBs) at 0.00025 mg/L (i.e., \1/2\ of the MCL), measured as
decachlorobiphenyl. However, the approved PCB screening methods in the
technical criteria document that determine whether or not the trigger
level has been exceeded do not measure decachlorobiphenyl. They measure
Aroclors, the values for which can be converted to decachlorobiphenyl
using the conversion table under paragraph III.A. of the technical
criteria document. Laboratories must use one of the EPA approved
screening methods in analyzing LFBs at the trigger level for PCBs.

I. New Systems and New Sources

Under Sec. 141.23(k) of these changes, any public water system or
source of water supplying a public water system that begins operation
after (the publication date of the final rule), would have to
demonstrate compliance with all applicable MCLs in this part within a
period of time specified by the State, unless the State waives testing
for certain contaminants in accordance with its approved waiver
process.

J. Sample Compositing

The current requirements allow systems to combine two to five
samples before they are analyzed for contamination. This feature allows
systems to reduce sampling costs by half or more, depending upon the
number of samples composited. However, this feature may allow
contamination to go undetected, where the contamination in one sample
is masked by dilution from the other samples. In an extreme case,
contamination at the MCL in one sample could be invisible to the
laboratory analysis, where it is masked by four clean samples and where
the laboratory detection sensitivity is hovering at or just above one
fifth of the MCL.
For this reason, EPA is considering whether to discontinue its use.
Some States, however, have expressed an interest in continuing
compositing under conditions that would assure the same levels of
detection sensitivity as those available for single sample analyses.
EPA is open to suggestions to allow sample compositing in the limited
cases where the criteria for single sample analysis would not be
sacrificed.
Commenters wishing to allow systems to use sample compositing under
Chemical Monitoring Reform should identify which contaminants would be
covered, the single sample detection criterion the State would
establish for each contaminant, and explain how the detection criteria
would be enforced for both single sample analyses and composited sample
analyses. The single sample detection criterion should be sufficiently
far below the trigger level of \1/2\ of the MCL as to assure that
quantitation at \1/2\ of the MCL will be within reasonable precision.
That requirement will probably eliminate many contaminants as
candidates for compositing, because the composite sample detection
criteria must be consistent with the single sample criterion i.e., if
the State sets the single sample detection criterion at one tenth of
the MCL (five times lower than the quantitation at \1/2\ of the MCL),
the detection criterion for a composite of two samples would be one
twentieth the MCL (i.e., \1/2\ the single sample detection criterion)
and it would be one fortieth of the MCL for a composite of four
samples, etc.

K. Records Kept by States

40 CFR 142.14(d) (4) through (5) requires States to keep records of
vulnerability and monitoring decisions. This document clarifies these
provisions by describing examples of the most recent vulnerability
decisions and monitoring frequency decisions. Under Sec. 142.15(d)(4),
the most recent State decisions include those related to targeting
systems for increased sampling and those involving sampling points that
have exceeded the trigger level. Under Sec. 142.15(d)(5), records of
the most recent monitoring frequency decisions include those based on
the targeting and vulnerability determinations identified above.
Included in the records would be the data that States used in making
these decisions.

L. Special State Primacy Requirements

Under Section 1413(c) of the Safe Drinking Water Act, as amended, a
State that has primary enforcement authority for all drinking water
regulations, would have interim primacy for Chemical Monitoring Reform
beginning on the date the State submits its regulations and a complete
primacy application to EPA, and ending when the Administrator makes a
determination of the primacy application.
State program revisions would include: (1) the State's regulations
or implementing provisions under Secs. 141.2 and 141.23; (2) the State
Targeting Plan described below; and (3) State's certification that its
program, including the targeting plan, is enforceable under State law.
Once adopted, the State program must operate in accordance with
Secs. 141.2 and 141.23, the approved State Targeting Plan, and the
provisions of Sec. 142.16(e)(3) for scheduling sampling when
contaminants are detected \1/2\ of the MCL.

[[Page 36116]]

1. Implementing Provisions
The implementing provisions under Part 141 are:

Sec. 141.2 Definitions
Sec. 141.23(a) General (types of systems affected)
Sec. 141.23(b) Sampling Points
Sec. 141.23(c) Responsibility to Provide Information
Sec. 141.23(d) Mandatory Monitoring
Sec. 141.23(e) Detection \1/2\ of the MCL
Sec. 141.23(f) Detection > MCL
Sec. 141.23(g) Violation Determinations
Sec. 141.23(h) Laboratory Certification Criteria
Sec. 141.23(i) New Systems & New Sources

Under Sec. 141.23 (e) through (f) of these changes, whenever a
system detects a contaminant at a concentration equal to or greater
than the draft trigger level of \1/2\ of the MCL, the system would be
required to sample at an increased frequency as directed by the State.
If a contaminant exceeds the MCL, the system must take at least one
sample per quarter for the following three quarters, in addition to any
additional samples required by the State to assure that the average
annual level of contamination is fully characterized. State decisions
must be documented in writing.
States would be required under Sec. 142.16(e)(3) of these changes
to include specific factors in their review of these detections,
including: (i) The history of sampling results for the sampling point
and for neighboring sampling points; (ii) The susceptibility of the
water supply to contamination; (iii) The periods most vulnerable to
contamination for the sampling point; (iv) The contaminant's solubility
and other characteristics; and (v) The agricultural and commercial
practices, and the efficacy of any source water protection measures
that have been enacted, within the source water review area. Further,
States would have to account for the estimated frequency and amplitude
of contaminant fluctuation in each sampling schedule.
2. State Targeting Plans
Under today's approach, States would identify those systems that
need to sample more frequently than every five years based on local
vulnerability, and every system scheduled by the State to sample more
frequently than every five years under Sec. 141.23(d), must do so.
Systems must also sample during the periods of greatest vulnerability
as designated by the State. Under Sec. 142.16(e)(2), States would be
required to describe their strategy for implementing this flexibility
in a State Targeting Plan.
Specifically, a State Targeting Plan would describe the State's
plans to screen all systems to identify vulnerable systems and the
sampling points that need to sample more frequently than once every
five years, for determining the frequency of sampling based on the
degree of vulnerability, and for updating the State's list of targeted
sampling points based on changing information. The targeting plan would
also describe the factors the State would consider in determining the
periods of greatest vulnerability and for scheduling the time of year
and frequency at which each system must sample.
A State targeting plan would also indicate that the State may
require a system to sample more frequently than every five years, at a
minimum, based on any one or a combination of the following factors:
(1) the fate and transport of a contaminant; (2) any agricultural,
commercial or industrial activity in the source water review area; (3)
the susceptibility of the source water to contamination; or (4) the
results of source water assessments conducted under section 1453 of the
Safe Drinking Water Act. States may list additional factors upon which
they would require a system to sample more frequently than every five
years, and States may subsequently require systems to sample more
frequently than every five years based on a factor not listed in its
targeting plan.
Finally, each State would provide the EPA Regional Administrator
with its initial list, or categorical description, of systems that it
has targeted to sample more frequently than every five years, within
one year after it has submitted a complete primacy revision application
to EPA. States would be required to update this list annually, and to
make it available to the public upon request. EPA seeks comment on
whether one year (which is in addition to the time prior to the
submission of the State's primacy revision application) is sufficient
time for the screening decisions, or whether a different period is
appropriate for States to inform all of their systems of their
individual sampling schedules. EPA also seeks comment on whether to
require systems to continue sampling in accordance with their current
schedules until the State has informed them of its screening and
monitoring decisions.
EPA is considering another option to the version described above.
The second version includes the approach above, and would also require
States to specifically target systems served by surface water, or by
ground water under the direct influence of surface water, to sample
more frequently than every five years, unless (or until) the State
determines that increased sampling is not required based on the degree
of an individual system's vulnerability to contamination (e.g., the
contaminant is not used in the source water review area), or based on a
finding that the risk posed by such levels of contamination is not
significant. This provision would establish a presumption of
vulnerability for surface water systems, and for ground water systems
under the direct influence of surface water, because of their inherent
susceptibility to contamination, and regardless of the presence or
absence of potential contamination sources in the Watershed & Recharge
Area.
EPA also seeks comment on whether the initial detection of a
contaminant within the source water review area should be an
alternative basis for the presumption of vulnerability. This criterion
would apply to any detection from the most recent round of sampling
that has not been discarded as a false detection in accordance with
State sampling confirmation procedures. The presumption would not apply
to detections for which the sources of contamination have been
identified, and the health risk posed by the contamination has been
described, to the satisfaction of the State.
3. State Certification
The requirement for States to certify that their program revisions
are fully enforceable under State law is not new, but the significance
of the certification under these changes would be greater than usual.
In reviewing State primacy programs and certifications, EPA would give
special attention to the State's authority to impose and enforce
requirements for individual systems to sample more frequently than
every five years, and to sample during the periods of greatest
vulnerability.
4. Oversight of State Decisions
There would be two avenues for EPA intervention into State chemical
monitoring decisions, short of initiating primacy withdrawal. The first
method is provided by section 1431(a) of the Safe Drinking Water Act,
which authorizes EPA to take such actions as necessary to protect
public health, whenever a contaminant may present an imminent and
substantial endangerment to public health. EPA may exercise this option
under the appropriate circumstances, without regard to any other
provision in these draft changes. For circumstances that do not warrant
a finding of imminent and substantial endangerment, EPA would rely on
40

[[Page 36117]]

CFR 142.18, as presented in this document.
Section 142.18 of the current regulations authorize an EPA Regional
Administrator to annul State sampling waiver determinations. This
section provides EPA with an alternative to primacy withdrawal, if EPA
should find a pattern of State decisions that are contrary to the
approved State program. In today's action, EPA is considering
increasing the list of State decisions in which a Regional
Administrator can intervene to include (1) the absence of State action
to require increased monitoring under Secs. 141.23 (c) through (g) and
(2) State decisions to grant monitoring relief under section 1418 of
the Safe Drinking Water Act. EPA could issue a monitoring order to:
annul a State waiver; annul a State surrogate sampling point
designation; annul a State monitoring relief decision made pursuant to
section 1418 of the SDWA; or make a determination to increase
monitoring in the absence of State action. EPA seeks comment on which
of these State decisions the Regional Administrators should be
authorized to annul in addition to waivers.
Neither the current provisions, nor the possible changes described
above are intended to authorize regular, random or arbitrary EPA
intervention on individual State monitoring decisions. They are
intended to authorize an appropriate EPA response to a pattern of State
decisions which conflict substantially with the bases in an approved
program on which the State has agreed to make those decisions. The EPA
monitoring order would be based on a failure by the State to implement
its approved program, and would take effect only after public notice
and comment. This provision is a safety valve that would provide for
EPA action, short of primacy withdrawal, in the face of a State's abuse
of its discretion.
Finally, as explained in the overview of this document, EPA expects
most States to support today's approach to reform the chemical
monitoring requirements. However, as shown in the table below, some
provisions in the current requirements are more stringent and some are
less stringent. EPA considers the current monitoring requirements, that
were published on January 30, 1991 and that have been adopted by all
States, to be as stringent, taken as a whole, as the provisions in this
document. Therefore, EPA is considering allowing States to continue
operating under the current requirements indefinitely. EPA seeks
comment on allowing States to continue under the current requirements,
if they prefer to do so.

M. Safe Drinking Water Act Amendments

Prior to the enactment of the Safe Drinking Water Act (SDWA)
Amendments of 1996, Chemical Monitoring Reform (CMR) was envisioned as
a free-standing initiative for monitoring revision and burden
reduction. During the development of CMR, Section 1418(b) of the SDWA
Amendments directed EPA to publish, by August 6, 1997, guidance for
``Permanent Monitoring Relief'' (PMR). This PMR would authorize States
to provide ``tailored alternative monitoring requirements'' for public
water systems upon completion of source water assessments in States
with approved programs under Section 1453 of the SDWA Amendments. This
notice describes in detail below the relationship between potential
characteristics of CMR, PMR, and the source water assessment activities
that are required by the SDWA Amendments.
As described below, Section 1418(b) authorizes PMR's features for
monitoring flexibility to be broader in coverage than CMR was framed to
be. If EPA develops two parallel programs for monitoring relief, there
could be substantial potential for confusion, overlap, conflict, and
unnecessary expenditure of scarce resources. EPA believes that Section
1418(b) directs EPA to frame PMR as a broad program for monitoring
relief. To implement the Amendments effectively and efficiently, EPA
must examine the actions it is required to take under the PMR
provisions of the Amendments, and ensure that its exercise of
discretion to frame CMR complements rather than complicates the
implementation of PMR by States and public water systems.
Today EPA provides (1) advance notice of its intent to revise
current monitoring regulations to provide for targeted, heightened
monitoring for systems at risk of contamination and a new, simplified
framework of reduced monitoring for systems not at risk (CMR), and (2)
draft guidelines for PMR which would include additional burden
reduction features. This advance notice is being provided in this form
for two reasons. First, while it might have been possible to frame this
entire monitoring initiative as PMR under SDWA Section 1418(b), EPA has
decided to issue these proposals as two joint elements--PMR and CMR.
EPA developed CMR in consultation with many members of the drinking
water community over a period of nearly two and a half years, most of
which pre-dated the enactment of the SDWA Amendments of 1996. Congress
was aware of the CMR process when it enacted additional relief in the
form of PMR. EPA believes separate approaches best meshes the
expectations for CMR, and its responsibilities under the 1996
Amendments for PMR.
Second, this notice contains what EPA believes to be a reasonable
and coherent alignment of the several components of a more flexible but
potentially more protective monitoring regime. Under the approach in
this notice, States can choose to retain their approved primacy
regulations for the current monitoring framework for Phase II and Phase
V chemical monitoring, and adopt (or not, if they choose) the burden
reduction features of PMR (additional waiver authority, surrogate
sampling, reduced nitrate sampling). Or, they can choose to adopt CMR
as their new primacy regulation for monitoring--which includes CMR's
basic, simplified monitoring framework and its provisions requiring
targeted monitoring for systems at risk of contamination--and adopt (or
not) the burden reduction features of PMR.
EPA recognizes that if a State adopts CMR before it obtains
approval of its Source Water Assessment Program and source water
assessments are completed for individual systems, the State would be
unable to grant monitoring waivers. This feature of the strategy for
integrating CMR and PMR may have the unintended consequence of
discouraging States from adopting CMR and retaining Phase II and V,
since Phase II and V provide for waivers. To address this, EPA is
considering allowing States that proceed with adopting CMR to retain
their existing approved waiver programs until the expiration of the
State's timetable for completing the assessments. States would not be
able to renew waivers after this date, unless it has met these
statutory requirements. EPA solicits comments on this issue.
EPA further seeks comment on whether or not to apply this same
approach to renewing waivers to States that choose to retain the Phase
II and V rules. This would preclude States from renewing waivers for
any public water system for which the State has failed to complete a
source water assessment after the expiration of the State's timetable
for completing all such assessments. The rationale for this approach
would be that it is important for States to apply any updated
information generated by the assessments to waiver decisions that would
be made after the assessments are completed. Although EPA is only

[[Page 36118]]

seeking comment on this approach, there are at least two reasons to
expect that it would not be burdensome for States or systems. First,
EPA is taking steps to provide States with the maximum amount of time
available under the law to begin and complete their assessment program
by the most cost effective and prioritized approaches possible, using
up to the full amount of the more than $120 million made available for
assessments by Congress. Second, any water system with an existing
waiver would already have a substantial and, in some cases, the full
amount of information needed for a source water assessment, meaning
these systems are among the likeliest candidates for expeditious
completion of assessments.
EPA believes this array of features would in general present a
reasonable, coherent and effective approach, but acknowledges that
alternative arrays of these features within CMR or PMR are possible.
Because alternative arrays could have significant implications for
coherence, operation, and (potentially) compliance with various
requirements of SDWA, EPA wants to present this notice for public
comment on its substance as well as on the operational implications of
this particular form in which the features of monitoring are arrayed.
The following are the various key components from which a State may
choose to frame its monitoring regime. EPA is requesting comment on
whether to delete or rearrange any elements of CMR or PMR. A complete
presentation of EPA's proposed guidelines for PMR can be found in
Section III.N, below.

BILLING CODE 6560-50-P

[[Page 36119]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.002

BILLING CODE 6560-50-C

[[Page 36120]]

For commenters who propose transferring to a CMR rule all or part
of the burden reduction features proposed today for PMR, EPA requests
that their comments also discuss what similar or different burden
reduction features should be included in PMR, for which EPA is required
to publish guidelines, and how they believe these two frameworks for
monitoring should be coordinated, operationally and structurally. In
light of Congress' enactment of Section 1418(b), if EPA is to place any
of PMR's burden reduction features within CMR, specific benefits and
functions that could only be achieved within CMR should be identified.
Comments proposing modifications to the monitoring requirements
under CMR for systems with little to no risk of contamination (one test
in five years) should address the expected public health implications
of the proposed modifications.
EPA also requests comments on the basis for the monitoring
requirements for systems at risk of contamination: the Phase II and
Phase V requirements currently in place; the targeted approach for
specifying heightened monitoring proposed as a part of CMR today; or
some other approach, such as a range of monitoring frequencies EPA
could specify to apply to different categories of contaminant, source
water or water system conditions that would trigger increased
monitoring. It is necessary to consider these requirements when
commenting on PMR because Section 1418(b)(3) specifies that public
water systems that are monitoring under PMR provisions and that detect
contaminants at levels that are not ``reliably or consistently below''
the MCL and that do not ``eliminate the contamination problem'' must
return to the monitoring frequencies specified in the applicable NPDWR
(Section 1418(b)(3)(B)). Currently, the monitoring frequencies under
the applicable NPDWR are those specified under the existing Standard
Monitoring Framework for Phase II and Phase V contaminants, which
requires quarterly monitoring for these systems. The monitoring
frequencies under CMR would be the heightened monitoring requirements
for systems at risk of contamination.
EPA cannot consider comments proposing the actual or effective
deletion of PMR, because it is required under Section 1418(b) to
publish guidelines for PMR.
Greater economic efficiency is an important value of the SDWA
Amendments of 1996 because it can enable the limited funds of public
water systems and States to be focused on the greatest risks to health.
Nonetheless, protection of public health itself remains the dominant
consideration under the SDWA. In monitoring as elsewhere in
implementation of the SDWA Amendments, EPA has a statutory obligation
to see that structures and decisions in PMR and CMR equally are based
on the adequate scientific information necessary to ensure that public
health is protected. To strengthen public confidence in drinking water
safety, consumers must know that a decision to reduce monitoring of
their water supplies is well-grounded in adequate scientific data and
analysis of their water system, that any waiver of monitoring is based
on a scientific judgement that the contaminant will not be present at
problematic levels during the waiver period, and that any detection at
problematic levels of a contaminant subject to reduced monitoring will
quickly lead to appropriately heightened monitoring. The following
discussion identifies the means to provide such scientific information.
In Section 1418, Congress expressly provided that completion for a
water system of a source water assessment, pursuant to an approved
State Source Water Assessment Program (SWAP) under Section 1453, was a
prerequisite to granting PMR to that system. Section 1453 requires
States to establish and implement SWAPs. To do this work, Section
1452(k)(1)(C) makes available to States, and allows them to obligate
over 4 fiscal years, up to 10 percent of the funds allotted to them for
State Revolving Funds in Fiscal Year 1997, a total of over $120 million
nationally. EPA is committed in Headquarters and in the Regions to
ensure successful assessments, and will as needed assist States on the
Drinking Water SRF set-asides, on stretching assessment dollars by
strong involvement of all capable participants in the assessments, and
by encouraging exchange of information about good models for
assessments and use of existing information to place within the
assessments. EPA believes that this funding and support will yield
useful assessments that can enable PMR to be provided where
appropriate, and will place source water protection on a firm base.
The Safe Drinking Water Act Amendments of 1996 also require EPA to
publish guidance for these two efforts--Source Water Assessments and
Permanent Monitoring Relief--at the same time, one year after enactment
(that is, on August 6, 1997). This timing ensures that, as States began
to develop their SWAPs under Section 1453 guidance, they will know what
information is needed to provide their systems with Permanent
Monitoring Relief, and can frame their assessment programs to generate
(among other things) the data necessary for PMR. EPA's draft Source
Water Assessment guidance of April 4, 1997, proposed that existing
delineations and source inventories done under approved State Wellhead
Protection Programs would be adequate to fulfill the delineation and
inventory requirements of Section 1453 for those ground water based
systems. However, States should examine whether these delineations and
inventories provide sufficient information to support all aspects of
PMR, and should consider modifying them under their SWAPs where
necessary to take full advantage of the regulatory flexibilities
offered in PMR.
Under CMR, the basic monitoring frequency in the proposed rule, of
1 sample every 5 years, is to be founded on the determination that the
system is not at risk. In deciding what information is necessary to
make determinations under CMR, today's proposal relies on a level of
information rather than the process to generate that information (that
is, the source water assessment process) specified for PMR under
Section 1418(b).
The kinds of data on source water, occurrence, susceptibility, use,
and the like that would be generated by a source water assessment
appear necessary to make adequately informed determinations for all
functions of CMR: on which systems are or are not at risk, to develop a
targeting plan for at-risk systems, and to specify sampling times of
greatest vulnerability for systems that are not. States are not
required to undertake a formal source water assessment process to
generate such data for CMR, but they are required to have and apply the
level of data that would be generated by an assessment to make CMR
determinations. This level of data will be consistent with the criteria
for completion of assessments each State has defined in their EPA-
approved assessment program, and likely will vary depending on the
nature and condition of a system (i.e., community or non-community, at
risk or not at risk, etc.). In other words, States can apply a screen
that is essentially equivalent to a source water assessment to ensure
they have adequate scientific data to make CMR determinations, but they
need not complete a formal assessment to do so. EPA may also include in
its final source water assessment guidance (to be published no later
than August 6, 1997) a provision in which States can use this
``assessment equivalence'' concept to allow the use of information
generated

[[Page 36121]]

by States for Chemical Monitoring Reform to be used to complete source
water assessments, at a level appropriate to the situation of the water
system.
This data requirement (Sec. 142.16(e)(2)(i)(A) of these changes)
should not slow CMR implementation. Many States have already gathered
considerable data on contamination sources, performed vulnerability
assessments, and analyzed monitoring data on these contaminants in
implementing the Phase II and V rules and in developing approved waiver
programs under those rules. Many States have also performed similar
work in developing wellhead protection programs. States are also
required to submit to EPA their source water assessment programs by
February, 1999. Because EPA does not expect to promulgate final CMR
regulations before August, 1998, States can thus incorporate the
characteristics of completed ``assessment equivalents''--waiver
programs, monitoring results, and wellhead protection programs--into
their overall CMR plan, for targeting at-risk systems and providing the
simplified monitoring framework for systems not at risk. States can put
their overall CMR plan into effect when EPA approves their primacy
regulations for CMR. They can determine which systems are or are not at
risk where they have this ``assessment equivalence'' level of
information.

N. Permanent Monitoring Relief Guidelines

Introduction
The Permanent Monitoring Relief provision of the Safe Drinking
Water Act (the Act) authorizes primacy States to adopt ``tailored
alternative monitoring requirements'' for most chemical contaminants.
Under that provision, State monitoring relief must comply with guidance
published by EPA, as well as ``assure compliance with, and enforcement
of, the applicable national primary drinking water regulations.''
Congress directed EPA to publish guidelines ``for States to follow
in proposing alternative monitoring requirements.'' These guidelines
must (1) assure that ``public health will be protected from drinking
water contamination,'' (2) require States to apply this monitoring
relief ``on a contaminant-by-contaminant basis,'' and (3) require that,
to be eligible for monitoring relief, a system must show that the
contaminant is not present in the water supply, or, if present, is
reliably and consistently below the MCL, or that ``action has been
taken to eliminate the contamination problem.''
Congress also specified that each State must develop, and secure
EPA approval of, a Source Water Assessment Program under section 1453
of the Act, and that a source water assessment must be complete for any
system to which such alternative monitoring requirements would be
available.²⁷ The guidance for approvable State Source Water
Assessment Programs must be published by August, 6, 1997.
---------------------------------------------------------------------------

\27\ See sections 1418(b) and 1453(a)(3)
---------------------------------------------------------------------------

Overview
States may offer Permanent Monitoring Relief for the sixty four
(64) contaminants listed in Table I, below, and for nitrate. Permanent
Monitoring Relief is not available for microbial contaminants, for
indicators thereof, or for contaminants formed within a distribution
system as a result of disinfection or corrosion.

Table I.--Contaminants Affected by Chemical Monitoring Reform

Inorganic Chemicals (IOCs):
[1] Antimony, [2] Arsenic, [3] Asbestos, [4] Barium, [5] Beryllium,
[6] Cadmium, [7] Chromium, [8] Cyanide, [9] Fluoride, [10] Mercury,
[11] Nickel, [12] Selenium, [13] Thallium.
Synthetic Organic Chemicals (SOCs):
[1] 2,4-D (Formula 40 Weeder 64); [2] 2,3,7,8-TCDD (Dioxin); [3]
2,4,5-TP (Silvex); [4] Alachlor (Lasso); [5] Atrazine; [6]
Benzo[a]pyrene; [7] Carbofuran; [8] Chlordane; [9] Dalapon; [10]
Di(2-ethylhexyl)adipate; [11] Di(2-ethylhexyl)phthalate; [12]
Dibromochloropropane (DBCP); [13] Dinoseb; [14] Diquat; [15]
Endothall; [16] Endrin; [17] Ethylene dibromide (EDB); [18]
Glyphosate; [19] Heptachlor epoxide; [20] Heptachlor; [21]
Hexachloro-cyclopentadiene; [22] Hexachlorobenzene; [23] Lindane;
[24] Methoxychlor; [25] Oxamyl (Vydate); [26] Pentachlorophenol;
[27] Picloram; [28] Polychlorinated Biphenyls (PCBs); [29]
Simazine; [30] Toxaphene.
Volatile Organic Chemicals (VOCs):
[1] 1,1-Dichloroethylene; [2] 1,1,2-Trichloroethane; [3] 1,1,1-
Trichloroethane; [4] 1,2,4-Trichlorobenzene; [5] 1,2-
Dichloropropane; [6] 1,2-Dichloroethane; [7] Benzene; [8] Carbon
tetrachloride; [9] cis-1,2-Dichloroethylene; [10] Dichloromethane;
[11] Ethylbenzene; [12] Monochlorobenzene; [13] o-Dichlorobenzene;
[14] p-Dichlorobenzene; [15] Styrene; [16] Tetrachloroethylene;
[17] Toluene; [18] trans-1,2-Dichloroethylene; [19]
Trichloroethylene; [20] Vinyl Chloride; [21] Xylenes.

For contaminants identified in Table I, States could, under PMR,
grant waivers to permit systems to forgo the sampling requirements of
one sample every five years, and can allow systems to conduct surrogate
sampling from sampling points within a system, or among two or more
systems, in lieu of sampling at every entry point to the distribution
system. These waiver and surrogate sampling provisions are presented in
greater detail in Sections A and B, respectively. For nitrate, States
could permit systems to reduce the sampling frequency from annual to
biennial under certain conditions. These provisions are described in
Section C. Section D explains the process for State adoption and EPA
approval of Permanent Monitoring Relief and Section E provides
definitions of key terms used in these guidelines.

Section A--Sampling Waivers for Chronic Contaminants

Under the Chemical Monitoring Reform approach, water systems would
sample at a minimum of once every five years during the time of
greatest vulnerability for each of the sixty four contaminants listed
in Table I, above. Under the PMR guidelines, a State could allow a
system to forgo monitoring at specified sampling points during a
monitoring period by granting a sampling waiver.
EPA seeks comment from States and systems on whether the relief
provided by five year waivers would be meaningful, in light of the cost
difference between sampling once every five years or updating a
vulnerability analysis to review a waiver every five years,
understanding that waivers could be granted on an area wide basis, and
do not have to be done on an individual system basis.
(1) State Findings Required for Waivers: Under PMR, a State could
grant a waiver allowing a system to forgo sampling during a five year

[[Page 36122]]

monitoring period, if the State, at a minimum, makes one of the
following determinations.
(a) The State may determine that the sampling point is free of
contamination and there is a high probability that it will remain so
during the term of the waiver. A State may not make this determination,
if the contaminant has been detected within the source water review
area of the sampling point within the last five years.
(b) The State may determine that the contaminant level will remain
reliably and consistently below the MCL during the sampling period
based on a finding that:
(i) the natural occurrence levels are stable and the contaminant
does not occur because of human activity; or
(ii) all the sources of potential contamination within the source
water review area: have been identified, brought under control, and
will pose no increased or additional risk of contamination to the
source water withdrawal point during the sampling period; and the
contaminant levels have peaked based on the history of sampling results
and the duration of the contaminant in the environment; or
(iii) the treatment at the sampling point is properly operated and
maintained, and is working reliably and effectively.
(c) A State may not make any of the three determinations under this
paragraph, if the contaminant was detected at a level \1/2\
of the MCL in the most recent sampling series for that sampling point.
(2) General Considerations: In making waiver decisions the State
shall, at a minimum, consider the following factors.
(a) the fate and transport of the contaminant;
(b) the patterns of contaminant use;
(c) the location of potential contamination sources within the
source water review area;
(d) the hydrogeologic features within the source water review area;
(e) the integrity of the structures delivering source water to the
sampling point;
(f) the results of all source water assessments that have been
completed within the source water review area;
(g) the efficacy of any source water protection measures that have
been enacted, and;
(h) for waivers based on the contaminant remaining reliably and
consistently below the MCL for the sampling period, the relationship of
the sampling results to the MCL, the variability of the sampling
results over time, and the trend of the sampling results.
(3) System Responsibility: Each water system granted a sampling
waiver under this paragraph shall notify the State within 30 days of
the time it first learns of any change in any of the conditions under
which a waiver was granted.
(4) State Review of Waiver Determinations: The State shall review
its decision to grant or renew a waiver, whenever it learns of a change
in the circumstances upon which the waiver was granted. The State may
amend the terms of a waiver, or revoke a waiver at any time.
(5) Waiver Renewals: A State may renew a sampling waiver by making
the same determination it made to initially grant the waiver, after
reviewing current assessments of the factors that are subject to change
during the term of the waiver, and that affect the finding(s) upon
which the waiver is based.
(6) Waivers for Cyanide: Before granting a waiver for cyanide, the
State shall determine whether cyanide is present in the system's source
water.

Section B--Surrogate Sampling Points

A State may allow a system, or several systems, to use the
monitoring results from the sampling point(s) designated by the State
as surrogate point(s), if the State determines that the source water
serving the surrogate sampling points is drawn from the most vulnerable
portion of the same contiguous source water.
(1) Intra-system Surrogate Sampling: For designating surrogate
sampling points within one system, the State shall consider a
sufficient record of the pertinent information below and the results of
the source water assessments that have been completed under section
1453 of the Safe Drinking Water Act
(a) Monitoring data demonstrating that the sampling results are <
\1/2\ MCL;
(b) Well log or surface water hydrology data demonstrating that the
points to be included in the surrogate sampling point program draw from
the same contiguous source water; and
(c) An inventory of the potential contamination sources within the
source water review area affecting all the sampling points to be
included in the surrogate sampling point program.
The State shall also require the system to validate the results of
the surrogate sampling points. For example, where one sampling point
among three in a small system has been designated as the surrogate
point, the State might require the other two points to rotate the
sample every five years. This would reduce the system sampling burden
by one third.
(2) Inter-system Surrogate Sampling: For designating surrogate
sampling points among systems, a State must first receive EPA approval
of its criteria and procedures for implementing an Inter-system
Surrogate Sampling Point Program, that meets the criteria of this
paragraph. Two or more systems may use the monitoring results from
surrogate sampling points designated by the State, based on a complete
assessment of the contiguous source water that has been approved by the
State and that describes:
(a) The requirements for validation sampling (For example, where
several sampling points among dozens in several systems have been
designated as the surrogate points, the State might require the next
most vulnerable tier of sampling points to ``round robin'' the sample
every five years. This could significantly reduce the overall sampling
burden.);
(b) The location of potential contamination sources that could
affect any of the Community Water Systems or Non-transient, Non-
community Water Systems drawing from the contiguous source water.
(c) The hydrogeologic features of the contiguous source water; and
(d) The relationships among potential contamination sources, the
hydrogeologic features and the source water withdrawal points, with
particular regard to their relative locations.
(3) Validation Sampling: Whenever the sampling results at a
surrogate point are \1/2\ of the MCL, the State shall
require the systems to conduct validation sampling at each of the
points represented by that surrogate point. Surrogate sampling shall be
discontinued for that sampling point, and for any sampling points that
it represents, if the contaminant is \1/2\ MCL. The State
shall then decide which sampling points to target for increased
sampling, which, if any, to default to once every five years, and
which, if any, may be appropriate for a smaller surrogate sampling
arrangement.
(4) System Responsibility: Each system shall notify the State
within 30 days of the time it first learns of any change in any of the
conditions under which any surrogate sampling point has been
designated.
(5) State Review of Surrogate Sampling Point Designations: The
State shall review its decision to designate any surrogate sampling
point, whenever it learns of a change in the circumstances upon which
the point was designated.
EPA seeks comment on its distinction between intra-system surrogate
sampling and inter-system surrogate sampling, and the requirements

[[Page 36123]]

associated with each. EPA made the distinction because it believes that
inter-system surrogate sampling is likely to be more complex and
require more sophisticated analyses that intra-system surrogate
sampling. There may be situations, however, where inter-system
surrogate sampling is simple or where intra-system surrogate sampling
is complex. EPA seeks comment on whether the distinction should be made
on the complexity of analyses as opposed to the intra-system and inter-
systems distinction. Commenters should provide specific suggestions for
making an alternative distinction.

Section C--Reduced Nitrate Sampling

States may reduce the nitrate monitoring frequency from annual to
biennial sampling for a sampling point served exclusively by ground
water under the following conditions:
(1) Maximum Allowed Concentration: Nitrate measured as N has not
exceeded a concentration equal to or greater than 2 milligrams per
liter at any time during the past ten years;
(2) Integrity of Structures and Equipment: The State has determined
that the design and construction of the structures and equipment
delivering water from the wellhead to the distribution system fully
comply with current State code for such structures and equipment;
(3) Freedom from Surface Water Intrusion: The State has determined
that the ground water serving the sampling point is not under the
direct influence of surface water, and is not susceptible to
significant changes in contamination levels during the period for which
the sampling would be reduced e.g., not a shallow well, not in
fractured bedrock;
(4) State Determination: The State has determined that (a) nitrate
sampling is not required as a precursor to microbial or viral
contamination, (b) land uses, or relevant land use based conditions
(such as the effective operation of septic systems) in the area
affecting the sampling point are unlikely to change in a way that would
increase the risk of nitrate contamination, and (c) any contamination
at the sampling point is very unlikely to exceed the 2 mg/l during the
reduced sampling period;
(5) Effect of Detection 2 mg/l: If nitrate is detected
at 2 mg/l, measured as N, the system shall return to an
annual sampling frequency under the State requirements adopted pursuant
to the national primary drinking water regulations; and
(6) System Responsibility and State Review: Each system shall
notify the State within 30 days of the time it learns of any change the
conditions under which the reduced sampling for nitrate has been
allowed, particularly of any change in land use practices. The State
shall review its decision to reduce the sampling frequency, whenever it
learns of a change in the circumstances upon which its decision was
based.
EPA also seeks comment on [a] whether the Agency should use a
threshold other than 2 mg/l as one of the bases for reduced monitoring,
[b] whether EPA should set a reduced frequency other than biennial
sampling, or [c] whether EPA should establish a sliding scale of longer
sampling frequencies e.g., three year frequency based on a threshold of
2 mg/l, and five year frequency based on a threshold of 1 mg/l.

Section D--State Adoption and EPA Approval of Permanent Monitoring
Relief

The Act specifies that State Permanent Monitoring Relief provisions
will be treated as ``applicable'' national primary drinking water
regulations, which means they must be enforceable under both State and
Federal law.²⁸ The Act defines an enforceable State
requirement as a ``State program approved pursuant to this part.''
²⁹ In order to assure that the State Permanent Monitoring
Relief provisions will be Federally enforceable, EPA must review and
approve the State program. Therefore, any State adoption of alternative
monitoring requirements to offer Permanent Monitoring Relief must be at
least as stringent as these requirements and adhere to each of the
following steps.
---------------------------------------------------------------------------

\28\ See Sec. 1418(c)
\29\ See Sec. 1414(i)(4).
---------------------------------------------------------------------------

(1) State Program Description: The State shall describe the
information it will review, and its procedures and decision criteria
for issuing waivers under Section A, designating surrogate sampling
points under Section B, or allowing systems to sample biennially for
nitrate under Section C. At a minimum, the State Program Description
shall include the criteria under Sections A-C (respectively) for each
form of monitoring relief that the State proposes to offer, and specify
that the State will retain a record of the most recent vulnerability
determination for each sampling point, including:
(a) Those resulting in a decision to grant a sampling waiver under
Section A;
(b) Those resulting in a decision to allow the use of intra-system
surrogate sampling points under Section B(1); and
(c) Those resulting in the approval of source water assessments and
the location of geographically targeted sampling points based on those
source water assessments under Section B(2).
(2) Notice and Comment: The State must provide notice and
opportunity for public comment on the requirements.
(3) Attorney General Certification: The Attorney General must
certify in writing that the alternative State monitoring requirements
were duly adopted under State law, are enforceable under State law, and
comply with EPA's Permanent Monitoring Relief Guidelines and with
Secs. 1418 (b) through (c) of the Safe Drinking Water Act, as amended
August 6, 1996.
(4) State Source Water Assessment Program: EPA must have approved
the State's Source Water Assessment Program.
(5) EPA Review and Decision: Unless EPA notifies the State of its
disapproval of the State requirements within 9 months of EPA's receipt
of a complete set of the proposed State requirements, the State
requirements will take effect on the date of the State's submittal of a
complete program, or the effective date of its regulations, whichever
occurs later.
(a) A notice of disapproval will include the identification of the
part(s) of the State requirements at issue and the remedies necessary
to render those parts approvable.
(b) The State requirements shall not take effect until the State
has corrected the problems identified by EPA, and resubmitted its
revised program for review.
(6) EPA Review of State Determinations: A Regional Administrator
may annul a State decision to grant a waiver, to designate a surrogate
sampling point, or to reduce nitrate sampling, under the procedures
specified in 40 CFR, Part 142.18. EPA is seeking comment on whether to
expand this authority to these and other State decisions.

Section E--Definitions

(1) Contiguous source water means, for the purposes of these
guidelines, a source or several inter-connected sources of public
drinking water:
(a) Comprised of surface water, or ground water, or ground water
under the direct influence of surface water, or any combination
thereof, that serves two or more source water withdrawal points; and
(b) From within which contamination that can reach any one of the
source water withdrawal points, can also reach any of the other source
water withdrawal points.

[[Page 36124]]

(2) Monitoring period means a five year period during which water
systems are required under 40 CFR 141.23 to take at least one sample
during the time of greatest vulnerability.
(3) Source Water Review Area (SWRA) means the surface and
subsurface area within which a contaminant can reach the source water
withdrawal point, or any point between it and the entry point to the
distribution system (e.g., an aqueduct), during the time between
regularly scheduled samples. The size and shape will vary depending
upon several factors, including the sampling period and the
hydrogeologic features within the area. Where systems use ground water,
the SWRA could be the Source Water Protection Area (SWPA) established
under the Safe Drinking Water Act, where the SWPA is based on a time of
travel delineation consistent with the sampling period i.e., 5 years.
For surface water, the SWRA is the watershed upstream of the source
water withdrawal point.
(4) Surrogate sampling points mean the sampling point(s) within a
group of sampling points: within one water system e.g., under a
Wellhead Protection Program, that meets the criteria for intra-system
surrogate sampling point designations; or within a group of water
systems, that are designated by the State as the most vulnerable to
contamination and, therefore, can be used to represent all the sampling
points within the group.
(5) Validation sampling means sampling at one or more points
represented by surrogate sampling points, in order to verify that the
surrogate points are representative of those sampling points.

O. Suggestions for Regulatory Burden Reduction Other Than Chemical
Monitoring Reform

As explained in the Summary of Draft Changes, as part of the
President's initiative to ``Reinvent Environmental Regulation'', EPA
has been reviewing the National Primary Drinking Water Regulations
(NPDWRs) to find opportunities for reducing the paperwork burden on
public water systems and State drinking water agencies, and has
solicited input from States, water utilities, and environmental groups.
That process yielded a number of suggestions, including many which have
been incorporated into the Chemical Monitoring Reform approach that is
presented today. ``Stakeholders'' did, however, make suggestions other
than those related to Chemical Monitoring Reform. EPA believes a few of
these suggestions deserve further consideration. Consequently, we are
presenting those suggestions below, and are requesting comment, data,
or other relevant information on each so that the Agency can more fully
evaluate their merits for possible subsequent rulemaking.
It should be noted that none of the following suggestions were
unanimously embraced by all stakeholders, and some received more
stakeholder support than others. The suggestions follow:
(1) Surface Water Treatment Requirements
Extend various deadlines associated with filtration of ground
waters under the direct influence of surface water.
Section 142.16(b)(2)(B) of the regulations require States to
determine which community water systems are served by ground water
under the direct influence (GWUDI) of surface waters by June 29, 1994,
and which noncommunity water systems are GWUDI by June 29, 1999.
Section 141.71 of the regulations then requires that, within 18 months
after a system has been designated as a GWUDI, the State must determine
whether the system has to install filtration treatment or is able to
avoid filtration.
It has been suggested that provisions be adopted which would allow
for extensions of these two requirements. Some stakeholders believed
that while many GWUDI determinations are relatively easy, others are
quite complex--requiring additional time to complete. Some States also
have many more such determinations to make. The suggestion was to
provide States with additional time to make the determinations for
these more complex cases or where an extremely large number of
determinations is required. It was suggested that States be allowed
additional time to make the filtration determinations where they are
particularly complex or there are an extremely large number of
determinations to make.
In both cases, the suggestion was to allow for such extensions on a
case by case basis, possibly through a formal request to EPA for an
extension for specific systems. The suggestions also envisioned that
the extensions would be for a finite time period (possibly 2 to 5
years), to be specified in the federal regulations.
(2) General Reporting Requirements
(a) Eliminate the requirement for water systems to report
monitoring violations to the State. Section 141.31(b) of the current
federal regulations requires public water systems to report a violation
of any regulatory requirement to the State. One such requirement is
that a system must notify the State any time it fails to conduct any
required monitoring. In practice, States do not typically rely on water
systems to inform them of such failures. A system which does not
perform some required monitoring is not likely to notify the State of
that failure. Rather, States normally treat failure to receive
laboratory analytical results as the indicator that monitoring did not
occur. As a result, it has been suggested that the federal
requirement--that systems report instances of failure to monitor--is
redundant, and is serving no useful purpose. The interpretation is that
since there is a Federal requirement for water systems to report
analytical results of all monitoring to the State, a requirement to
notify the State of a failure to monitor is, in effect, redundant, and
thus unnecessary.
The intended purpose behind the requirement was to ensure that
States knew where required monitoring was not occurring so that they
could take some type of action to correct that failure. Advocates of
this approach believe that experience suggests that purpose is being
served without needing the support of the federal requirement. It has
been suggested, therefore, that EPA eliminate the federal requirement
that water systems must report monitoring violations to the State.
Systems would still be required to report analytical results of all
required monitoring to the State. With the suggested change, however,
States would have the option of continuing to require systems to report
monitoring failures to the State (although this would now be through
State, rather than federal, regulations), or treating any failure to
provide the analytical results as a monitoring failure. In either case
the State would know that follow-up action was necessary--fulfilling
the intent of the original federal requirement. Further, a water system
would still be required to notify the public of its failure to conduct
the required monitoring [Sec. 141.32(b)]. The consumers would,
therefore, be aware that some required monitoring had not occurred and
could take citizen action to resolve that failure. In addition, States
would still be required to follow-up on, and resolve, such failures.
Finally, States would still be required to notify EPA of all water
system monitoring failures. Advocates believe that EPA would,
therefore, continue to have all the information that it currently has
about such failures and the Federal oversight and enforcement
capabilities

[[Page 36125]]

would not be diminished. The suggested change would, in this view, not
alter a State's knowledge about a water system's failure to monitor, a
State's obligation to correct that failure, a State's obligation to
report the failure to EPA, the system's obligation to inform the public
about the failure, or EPA authorities to take an enforcement action
against the system. The change would only give a State the flexibility
to decide how it wants to arrive at a determination that a system has
failed to conduct some required monitoring.
(b) Reduce the frequency of reporting violation information to EPA.
Section 142.15(a) of the current regulations requires States to submit
to the Agency, quarterly reports of; (a) new violations by public water
systems, (b) new enforcement actions taken by the State against public
water systems, and (c) new variances and exemptions granted by the
State during the previous quarter. The violations and enforcement data
include acute and chronic contaminants, violations of actual safety
standards (MCLs, treatment techniques, etc), and failures to sample or
report according to schedule. Some of these violations represent a
greater risk to public health than others and some are more time
sensitive than others. As an example, violations of acute contaminants
(such as e-Coli, or fecal coliforms) or violations associated with
acute contaminants (such as total coliforms), typically need to be
addressed sooner than do violations of chronic contaminants. As such,
the regulatory agency needs to be aware of a violation of an acute
contaminant sooner than it does a violation of a chronic contaminant.
Similarly, violations of maximum contaminant levels (indicating actual
contamination) typically require more immediate attention than do
violations of monitoring requirements. Even different types of
monitoring violations deserve different levels of attention. ``Major''
monitoring violations (those in which none of the required monitoring
was conducted) need to be addressed and resolved much sooner than do
``minor'' monitoring violations (those in which some, but not all of
the required monitoring was conducted).
There is also a distinction in the urgency for any violation
information among the different users of that information. States are
typically the primary enforcement authority for the drinking water
requirements, with EPA serving a secondary role. The primary
enforcement authority needs to make decisions about violation severity
and appropriate remedy, and therefore, typically needs information more
quickly than does the secondary overseer. In States where a State
agency has been delegated this primary enforcement authority, EPA
typically becomes involved only when a violation is considered
``significant'', or where it is clear that EPA involvement is necessary
to resolve the problem. Other than these special situations, EPA's role
is one of evaluating the success of the drinking water program through
the surrogate of compliance/violation statistics.
For these reasons, some stakeholders questioned EPA's need for all
of the above information, on a quarterly basis. It has been suggested
that EPA align the frequency of State reporting to the importance of
the information to the Agency. One suggestion was to continue to
require quarterly reporting of violations of all maximum contaminant
levels (MCLs), treatment techniques, and State enforcement actions
against those violations, but to reduce to annually all other State
reporting. Another suggestion was to require quarterly reporting of all
information (MCL, treatment technique, reporting, etc) related to acute
contaminants, but to reduce to annually the reporting of all
information related to chronic contaminants.
It should be noted that a few stakeholders believed that reducing
the reporting frequency would actually increase, rather than decrease,
the burden on States. Some stakeholders noted the problems and
obstacles faced by States in transmitting violation data to EPA (such
things as identifying why certain data is rejected by the automated
data system), and believed that ``saving'' resolution of all these
problems until the end of the year would actually take much more time
than would have been required if done on a quarterly basis.
EPA requests comment on these suggestions and solicits ideas for
other ways of reducing the frequency of reports from the State to EPA.

List of Subjects in 40 CFR Parts 141 and 142

Environmental protection, Administrative practices and procedures,
Intergovernmental relations, Reporting and recordkeeping requirements,
Water supply, Indians.

Dated: June 26, 1997.
Carol M. Browner,
Administrator.

Appendix A to Preamble: EPA Technical Criteria Document for the
Analysis of Selected Chemicals in Drinking Water

(The four suggestions for change described in the preamble and subject
to comment are highlighted in the following technical criteria document
with [brackets].)
Contaminant Performance Criteria: In order to receive and retain
certification for analyzing samples to determine compliance under 40
CFR 141.23 and Part 141, Subpart I, a laboratory would have to meet the
following requirements.
I. Laboratory Method Detection limits (MDLs): Before initially
using an EPA approved method to analyze compliance samples, each
laboratory would calculate the MDL for each regulated contaminant
covered by that method using at least seven replicates in accordance
with the procedure in 40 CFR, Part 136 Appendix B, [except that the
LFBs used to calculate the MDL must be extracted (if applicable), and
analyzed over a period of at least three days]. The requirement to
calculate the initial MDL over a three day or longer period does not
apply to MDL calculations conducted before October 1, 1997.
A. Each laboratory would achieve an MDL of 0.5 g/l for
each VOC listed under Sec. 141.61(a), an MDL of 1 g/l for
lead, and for copper--an MDL of 1 g/l or 200 g/l when
atomic absorption direct aspiration is used.
B. Each laboratory would achieve the detection limits specified by
the State for all other contaminants listed under Secs. 141.11(b),
141.61(c) and 141.62(b) (1) through (6), 141.62(b) (10) through (15)
and 141.82(c)(3).
II. Ongoing Quality Control: Each laboratory would analyze a
laboratory fortified blank (LFB) with each batch of samples. The spike
levels of each LFB would be as specified by the individual methods or
consistent with standard laboratory practices, except that:
A. [Trigger Level LFBs--(i) Each laboratory would extract (if
applicable) and analyze at least one LFB per week at a concentration
equal to or less than \1/2\ of the MCL in any week during which
drinking water compliance samples are either (1) analyzed directly
without the use of an extraction step; or (2) extracted for future
analysis.
(ii) For polychlorinated biphenyls, the LFBs would be analyzed
using an approved PCB screening method under paragraph V. of this
document. The

[[Page 36126]]

conversion table below would be used to determine if a laboratory can
detect Aroclors at \1/2\ of the MCL.

----------------------------------------------------------------------------------------------------------------
Aroclor in Conversion Decachlorobiphenyl
Aroclor mg/L factor in mg/L
----------------------------------------------------------------------------------------------------------------
1016............................................................. 0.00013 1.92 0.00025
1221............................................................. 0.000095 2.63 0.00025
1232............................................................. 0.000115 2.17 0.00025
1242............................................................. 0.00013 1.92 0.00025
1248............................................................. 0.00015 1.67 0.00025
1254............................................................. 0.000165 1.52 0.00025
1260............................................................. 0.00018 1.39 0.00025
----------------------------------------------------------------------------------------------------------------

(iii) In any week during which a laboratory is using method 508A to
analyze drinking water compliance samples, it would extract and analyze
at least one LFB at a concentration equal to or less than \1/2\ MCL
using that method.
B. MDL LFBs--Each laboratory would extract, if applicable, and
analyze at least one LFB per month during any month in which drinking
water compliance samples are either (1) analyzed directly without the
use of an extraction step; or (2) extracted for future analysis. In
either case, the laboratory would spike each LFB at the same level as
that used to calculate the method detection limit in the initial
demonstration of capability.]
C. Each laboratory would reliably achieve the accuracy and
precision parameters, if any are specified by the State under paragraph
A above, and the detection sensitivity, if any are specified by the
State under paragraph B, in the analyses of these LFBs.
III. Approved Analytical Methods, PE Samples and Acceptance Limits:
All samples used to determine compliance with the maximum contaminant
levels under Secs. 141.11(b), 141.61(a), 141.61(c) and 141.62(b) (1)
through (6) and 141.62 (10) through (15) would be analyzed in
accordance with the methods, preservation techniques and holding times
specified under paragraph V. Approved Analytical Methods and Acceptance
Limits Under Chemical Monitoring Reform, of this document and in the
method descriptions.
A. [At a minimum, each laboratory must successfully analyze
Performance Evaluation (PE) samples every year as provided by EPA, the
State, or other parties that have been approved by the State or EPA.]
This series of PE samples must be tested for the contaminants, and
achieve the quantitative acceptance limits, under paragraph V. of this
document
B. Each laboratory must achieve the quantitative acceptance limits
under paragraph V. of this document for at least 80 percent of the
regulated organics listed in Sec. 141.61(a)(2) through (a)(21).
IV. Recording Results of Sampling Analyses and Laboratory Quality
Assurance Analyses:
A. Each laboratory would report the results of all sample analyses,
including all detections, in the manner and format specified by the
State. For the purposes of 40CFR141.23 only, ``detection'' means any
value observed in a drinking water sample that is equal to or greater
than the MDL as determined by the procedures in 40CFR136, Appendix B,
by paragraphs I and II. of this criteria document, and by criteria
established by the State.
B. [Each laboratory would report the results of analyzing the
Performance Evaluation (PE) Samples under paragraph III. to the State,
at a minimum frequency of once each year.
C. Each laboratory would maintain a record of each MDL analysis and
calculation under paragraph I, in the format specified by the State,
until the next State laboratory certification audit report has been
completed, or for five years, whichever period is longer.
D. Each laboratory would maintain a record of each LFB analysis
conducted under paragraph II., in the format specified below, until the
next State laboratory certification audit report has been completed, or
for five years, whichever period is longer.
E. The records under Paragraphs C. and D. (above) would be provided
to the State upon request, in the manner and format specified by the
State.]

Record of Analyzing Laboratory Fortified Blanks

Purpose of LFB (check one):
{time} Weekly Trigger Level Check
{time} Monthly MDL Level Check

Units of Measure (check one):
{time} Milligrams per Liter (mg/l)
{time} Micrograms per Liter (g/l)

Laboratory Name and Address

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

Lab Identification Number:---------------------------------------------

Contact Person:--------------------------------------------------------
Phone : ( ) - --------------------------------------------

Method Identification :------------------------------------------------

Description of deviations from published method, if any (e.g.,
columns, detectors, etc). Use reference to laboratory SOP or other
QA documentation when appropriate.

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

----------------------------------------------------------------------------------------------------------------
Date Analyte Fortified concentration Measured concentration
----------------------------------------------------------------------------------------------------------------
......................................... .....................
......................................... .....................
......................................... .....................
......................................... .....................
......................................... .....................
......................................... .....................
......................................... .....................
----------------------------------------------------------------------------------------------------------------

BILLING CODE 6567-50-P

[[Page 36127]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.003

[[Page 36128]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.004

[[Page 36129]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.005

[[Page 36130]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.006

[[Page 36131]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.007

[[Page 36132]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.008

[[Page 36133]]

[GRAPHIC] [TIFF OMITTED] TP03JY97.009

BILLING CODE 6560-50-C

[[Page 36134]]

For the reasons set out in the preamble, Chapter I of Title 40 of
the Code of Federal Regulations is proposed to be amended as follows:

PART 141--[AMENDED]

1. The authority 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, and 300j-9.

2. Section 141.2 is amended by adding the following definitions in
alphabetical order.

Sec. 141.2 Definitions.

* * * * *
Periods of greatest vulnerability means the periods during which
contamination is most likely to occur at the highest concentration at a
particular sampling point, based on the history of relevant factors for
that sampling point e.g., Weather Bureau precipitation averages, local
pesticide application practices.
* * * * *
Time balanced average means the average of values representing
equal segments of time, which are themselves the average of individual
data points within each segment of time. For example, the sampling
results throughout each quarter would be divided among the months of
the quarter and the individual sampling results within each month would
be averaged to determine the value for that month. The quarterly value
would be the average of the three monthly values.
* * * * *
3. Section 141.23 is revised to read as follows:

Sec. 141.23 General monitoring provisions.

(a) General: Each community water system (CWS) and each non-
transient, non-community water system (NTNCWS)--hereafter ``system'' in
Secs. 141.23 and 142.16(e)--shall monitor the contaminants under
Secs. 141.11(b), 141.61(a), 141.61(c), 141.62(b) (1) through (6) and
141.62(b) (10) through (15) in accordance with the requirements of this
section. Failure to sample, or to report to or notify the State, in
accordance with this section, or as directed by the State under this
section and Sec. 142.16(e), is a violation of the Safe Drinking Water
Act.
(b) Sampling Points:
(1) Each system shall monitor, at each entry point to the
distribution system, after treatment (if any).
(2) Systems shall sample at any sampling points the State may
designate in addition to the entry point to the distribution system.
(c) Responsibility to Provide Information:
(1) Each system shall report the results of all sampling conducted
under this section to the State, including detections the
Method Detection Limit (MDL), in accordance with Sec. 141.31 and in the
format prescribed by the State.
(2) Each system shall provide any information requested by the
State, within the time frame and in the format specified by the State.
A failure to provide this information is sufficient reason for the
State to require a system to sample more frequently than every five
years.
(d) Mandatory Monitoring:
(1) Each system shall sample at least once every five years at each
sampling point for the contaminants under Secs. 141.11(b), 141.61(a),
141.61(c), 141.62(b) (1) through (6) and 141.62(b) (10) through (15).
(2) If, for any reason, the State directs a system to sample more
frequently than once every five years, the system shall sample at the
frequency specified by the State.
(3) Each system shall sample during the periods of greatest
vulnerability designated by the State. If the State does not designate
the periods of greatest vulnerability, the system shall determine the
periods of greatest vulnerability, describe to the State the risk-based
reasons for the periods it specified, and sample at those times.
(4) If any of the following VOCs are detected at 0.5
g/l at any sampling point, the system shall monitor for vinyl
chloride at that sampling point within 30 days: trichloroethylene;
tetrachloroethylene; 1,2-dichloroethane; 1,1,1-trichloroethane; cis-
1,2-dichloroethylene; trans-1,2-dichloroethylene; or 1,1-
dichloroethyhlene.
(e) Detection \1/2\ of the MCL: If a contaminant is
detected \1/2\ of the MCL, including detections >MCL, the
system shall sample as scheduled by the State under Sec. 142.16(e)(3).
(f) Detection >MCL: If the results of a sample exceed the MCL, in
concert with the requirements of paragraph (e), the system shall sample
during each of the following three quarters. If the State schedules
multiple samples during any quarter, a time balanced average must be
used to determine the value for that quarter.
(1) Once an MCL violation has been established for a contaminant
under paragraph (g) of this section, the system shall sample every year
for that contaminant during the period of greatest vulnerability,
unless the State specifies a different sampling schedule.
(2) If an MCL violation is not established upon completion of the
monitoring required under this paragraph, the system shall continue
sampling as directed by the State.
(g) MCL Violations Determinations: A system is in violation of the
MCL if :
(1) The average of the four quarterly values exceed the MCL; or
(2) Any quarterly value, or any combination of less than four
quarterly values, would cause the average annual concentration to
exceed the MCL.
(h) Laboratory Certification Criteria:
(1) All samples to determine compliance with the MCLs in
Secs. 141.11(b), 141.61(a), 141.61(c), 141.62(b) (1) through (6) and
141.62(b) (10) through (15) must be analyzed by laboratories certified
by EPA, or by the State in accordance with, and meeting the
requirements described in, EPA Technical Criteria Document for Selected
Chemical Contaminants in Drinking Water.
(2) The State or EPA may suspend or revoke a laboratory's
certification for failure to consistently achieve the standards
established under this paragraph.
(i) New Systems & New Sources: All public water systems and sources
of water supplying a public water system that begin operations after
[insert publication date of the final rule], shall demonstrate
compliance with all applicable MCLs in this part within a period of
time specified by the State, unless the State waives testing for
certain contaminants in accordance with paragraph (h) of this section.
In a State where EPA has primary enforcement authority, a new system or
new source must demonstrate full compliance with the MCLs in
Secs. 141.11(b), 141.61(a), 141.61(c), 141.62(b) (1) through (6) and
141.62(b) (10) through (15), within the period of time specified by the
Regional Administrator.

PART 142--[AMENDED]

4. The authority citation for Part 142 continues to read as
follows:

Authority: 42 U.S.C. 300g-1, 300g-2, 300g-3, 300g-4, 300g-5,
300g-6, 300j-4, and 300j-9.

* * * * *
5. Section 142.14 is amended by revising the introductory text of
paragraph (d) and paragraphs (d)(4) and (d)(5) to read as follows:

Sec. 142.14 Records kept by States.

* * * * *
(d) Each State which has primary enforcement responsibility shall
retain, for not less than 12 years, files which shall include for each
public water system in each State:
* * * * *

[[Page 36135]]

(4) A record of the most recent targeting and vulnerability
determination for each sampling point, including the monitoring results
and other data supporting the determination, the State's findings based
on the supporting data and any additional bases for such determination;
except that it shall be kept in perpetuity or until a more current
vulnerability determination has been issued. These records shall
include State decision:
(i) Determinations related to targeting systems for increased
sampling;
(ii) Determinations involving sampling points that have exceeded
the trigger level;
(iii) Determinations related to the review of any such decisions
that has been undertaken because of a change in the circumstances upon
which the original decision was based.
(5) A record of all current monitoring requirements and frequencies
for each contaminant and each sampling point, including those based on
the targeting and vulnerability determinations identified under
paragraph (d)(4) of this section. These records shall be kept in
perpetuity, or until a more recent monitoring frequency decision has
been issued.
* * * * *
6. Section 142.16(e) is revised to read as follows:

Sec. 142.16 Special primacy requirements.

* * * * *
(e) Chemical Monitoring Reform.
(1) Prior to implementing the provisions of 40 CFR 141.23, a State
shall submit a primacy revision application that meets the requirements
specified below. Approved State programs must operate in accordance
with the provisions under Sec. 141.23 and paragraph (e)(3) of this
section and the approved State Targeting Plan.
(i) An application for approval of a State program revision to
adopt the requirements under Sec. 141.23, must include the State
regulations (or implementing provisions) adopting those requirements, a
description of the State Targeting Plan under paragraph (e)(2) of this
section and a certification from the Attorney General that each of the
provisions in its primacy revision application, and in any supplements
thereto, are enforceable under State law.
(ii) The State's primacy revision application must also include a
summary of public participation in the development of the State's
program. At a minimum, the State process shall include an opportunity
for public review of and comment upon the program elements identified
above.

Alternative I for Paragraph (e)(2)

(e)(2) Targeting Plans. The State shall identify, and prescribe a
sampling schedule for, each sampling point within each community water
system and within each non-transient, non-community water system that
may be vulnerable to contamination during the next five years. The
State shall transmit its list of these sampling points to the Regional
Administrator within one year after EPA has approved its primacy
revision application, and thereafter upon request of the Regional
Administrator. The State shall also update its list of targeted
sampling points annually, and shall make the list available to the
public upon request.
(i) The State shall develop a Targeting Plan describing:
(A) The State's procedures under Sec. 141.23(d)(2) to screen all
systems in order to identify vulnerable systems to sample more
frequently than once every five years, and for determining the
frequency of sampling based on the degree of vulnerability;
(B) The factors the State will consider in determining the periods
of greatest vulnerability; and
(C) The State plans for periodically updating its list of targeted
sampling points.
(ii) At a minimum, the targeting plan shall specify that a sampling
point may be targeted to sample more frequently than every five years
based on any one or a combination of the following factors:
(A) The fate and transport of a contaminant;
(B) The agricultural, commercial or industrial activities in the
source water review area; or
(C) The susceptibility of the source water withdrawal point to
contamination.
(iii) At a minimum, the State's factors for scheduling systems to
sample during the periods of greatest vulnerability shall include each
of the factors listed in paragraph (e)(2)(ii) of this section.
(iv) The State shall notify all systems of their sampling
requirements in writing.

Alternative II for Paragraph (e)(2)

(e)(2) Targeting Plans: The State shall identify and prescribe a
sampling schedule for each sampling point within each community water
system and within each non-transient, non-community water system that
must sample more frequently than once every five years, based on each
sampling point's vulnerability to contamination. The State shall
transmit its list of these sampling points to the Regional
Administrator within one year after EPA has approved its primacy
revision application.
(i) The State shall develop a plan describing
(A) The State's procedures under Sec. 141.23(d)(2) to screen all
systems in order to identify vulnerable systems to sample more
frequently than once every five years and for determining the frequency
of sampling based on the degree of vulnerability,
(B) The factors the State will consider in determining the periods
of greatest vulnerability, and
(C) The State plans for periodically updating its list of targeted
sampling points.
(ii) The State plan shall specifically target those sampling points
served by surface water, or by ground water under the direct influence
of surface water, to sample more frequently than every five years as
specified by the State, unless (or until) the State determines that
those points do not need to sample more frequently than every five
years based on the degree of their vulnerability, or on the risk that
such levels may pose to public health.
(iii) At a minimum, the targeting plan shall specify that a
sampling point may be targeted to sample more frequently than every
five years based on any one or a combination of the following factors:
(A) The fate and transport of a contaminant;
(B) The agricultural, commercial or industrial activities in the
source water review area; or
(C) The susceptibility of the source water withdrawal point to
contamination.
(iv) At a minimum, the State's factors for scheduling systems to
sample during the periods of greatest vulnerability shall include each
of the factors listed in paragraph (e)(2)(iii) of this section.
(v) The State shall notify all systems of their sampling
requirements in writing.
(e)(3) Detection \1/2\ of the MCL: Whenever the
sampling result for a contaminant is \1/2\ MCL, the State
shall require the system to sample according to a special monitoring
schedule, that has been designed to account for the estimated frequency
and amplitude of contaminant fluctuation.
(i) In establishing a special monitoring schedule for a sampling
point under this paragraph and Sec. 141.23(e), the State shall
consider:
(A) The history of sampling results for the sampling point and for
neighboring sampling points;

[[Page 36136]]

(B) The sources of contamination and the susceptibility of the
water supply to contamination;
(C) The periods of greatest vulnerability;
(D) The contaminant's solubility and other relevant
characteristics; and
(E) The agricultural and commercial practices, and the efficacy of
any source water protection measures that have been enacted, within the
source water review area.
(ii) A State may determine that detections \1/2\ of the
MCL, but less than the MCL, will remain reliably and consistently below
the MCL for five years, and may allow the system to sample at a minimum
of once every five years.
(iii) The State shall document each sampling schedule, or the basis
of its determination that the contaminant will remain reliably and
consistently below the MCL, in writing.
* * * * *
7. Section 142.18 is revised to read as follows:

Sec. 142.18 EPA Review of State Determinations.

(a) A Regional Administrator may:
(1) Annul a State decision to grant a waiver, to designate a
surrogate sampling point or to reduce nitrate monitoring under the
Permanent Monitoring Relief provisions of section 1418 of the Safe
Drinking Water Act; or
(2) Make a determination in the absence of State action under
Secs. 141.23(c) through (g)--in accordance with paragraph (b) of this
section.
(b) When information available to a Regional Administrator, such as
the results of an annual review, indicate that either a State
monitoring determination, or the absence of a State monitoring
determination, fails to apply the standards of the approved State
program or of the guidelines published under section 1418(b)(2) of the
Safe Drinking Water Act as amended, he may propose to annul the State
monitoring determination or initiate an EPA monitoring determination by
sending the State and the affected PWS a draft Monitoring Order. The
draft Monitoring Order shall:
(1) Identify the PWS, the State determination and the provisions at
issue;
(2) Explain why the State determination, or absence thereof, is not
in compliance with the State program and must be changed; and
(3) Describe the actions and terms of operation the PWS will be
required to implement.
(c) The State and PWS shall have 60 days to comment on the draft
Monitoring Order.
(d) The Regional Administrator may not issue a Monitoring Order to
impose conditions less stringent than those imposed by the State.
(e) The Regional Administrator shall also provide an opportunity
for comment upon the draft Monitoring Order, by
(1) Publishing a notice in a newspaper in general circulation in
the communities served by the affected system; and
(2) Providing 30 days for public comment on the draft order.
(f) The State shall demonstrate that its determination is
reasonable, based on its approved program.
(g) The Regional Administrator shall decide within 120 days after
issuance of the draft Monitoring Order to:
(1) Issue the Monitoring Order as drafted;
(2) Issue a modified Monitoring Order; or
(3) Cancel the Monitoring Order.
(h) The Regional Administrator shall set forth the reasons for his
decision, including a responsiveness summary addressing significant
comments from the State, the PWS and the public.
(i) The Regional Administrator shall send a notice of his final
decision to the State, the PWS and all parties who commented upon the
draft Monitoring Order.
(j) The Monitoring Order shall remain in effect until canceled by
the Regional Administrator. The Regional Administrator may cancel a
Monitoring Order at any time, so long as he notifies those who
commented on the draft order.
(k) The Regional Administrator may not delegate the signature
authority for a final Monitoring Order or the cancellation of an order.
(l) Violation of the actions, or terms of operation, required by a
Monitoring Order is a violation of the Safe Drinking Water Act.

[FR Doc. 97-17210 Filed 7-2-97; 8:45 am]
BILLING CODE 6560-50-P}