%
GUIDANCE MANUAL
FOR
COMPLIANCE WITH THE
FILTRATION AND DISINFECTION REQUIREMENTS
FOR
PUBLIC WATER SYSTEMS
USING
SURFACE WATER SOURCES
MARCH 1991 EDITION
*
s
USE
PRC**'0
T>
2
ui
O
/
SCIENCE ANO TECHNOLOGY BRANCH
CRITERIA AND STANOAROS OIVISION
OFFICE OF DRINKING WATER
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON. D.C.
-------
GUIDANCE MANUAL
FOR
COMPLIANCE WITH THE
FILTRATION AND DISINFECTION REQUIREMENTS
FOR
PUBLIC HATER SYSTEMS
US INS
SURFACE HATER SOURCES
for
Science and Technlogy Branch
Criteria and Standards Division
Office of Drinking Hater
U.S. Environmental Protection Agency
Hashington, D.C.
Contract No. 68-01-6989
Malcolm Pirnie, Inc.
100 Elsenhower Drive
Paramus, New Jersey 07653
HDR Engineering, Inc.
5175 Hillsdale Circle
Eldorado Hills, CA 95630.
October, 1990
-------
Acknowledgments
Preparation of this document involved important contributions from
many people in two consulting engineering firms, several private
consultants, and the United States Environmental Protection Agency
(USEPA). Malcolm Pirnie, Inc., with technical contributions
provided by HDR Engineering Inc., conducted the day-to-day work
under contract with the USEPA.
Principal authors from Malcolm Pirnie were David J. Hiltebrand and
Linda Averell Wancho. Personnel from HDR involved in this work
were Jerry Troyan and Perri P. Garfinkel. Additional personnel
from Malcolm Pirnie who contributed to either the technical content
or the preparation of the manual included: John E. Dyksen, James K.
Schaefer, Scott L. Phillips and Peter B. Galant.
Private consultants who contributed to the document included:
Dr. Appiah Amirtharajah, Georgia Institute of Technology, Atlanta,
GA; Dr. Ovadia Lev, Hebrew University, Jerusalem, Israel;
Dr. Vincent Oliveri, formerly John Hopkins University, Baltimore,
MD; Dr. Phillip C. Singer, University of North Carolina, Chapel
Hill, NC; Dr. Mark wiesner, Rice University, Houston, TX.
Preparation of the document was overseen by Stig Regli, the USEPA
project officer. Valuable technical review and major contributions
to the text were provided by Thomas Grubbs, Office of Drinking
Water, USEPA, and Leigh Woodruff, Region X, USEPA.
Special thanks are given to the following individuals working for
USEPA whose review and comment on numerous drafts greatly
contributed to the evolution of this document: John Davidson,
Office of Policy, Planning and Evaluation; Edwin Geldreich,
Drinking Water Research Division, Office of Research and
Development; John Hoff, formerly with Drinking Water Research
Division, Office of Research and Development; Walt Jakubowski,
Environmental Monitoring and Support Laboratory, Office of Research
and Development; Dr. Gary Logsdon, formerly with Drinking Water
Research Division, Office of Research and Development; Kevin
Reilly, Region I; Margaret Silver, Office of General Council; and
Jim Westrick, Technical Support Division, Office of Drinking Water.
Appreciation is also expressed to state public health officials,
representatives of the drinking water industry, academicians, and
the American public for their participation in submitting timely
and insightful comments without which this document would not have
been possible.
Some of the appendices have primary authors which are noted on the
corresponding cover pages.
fli Apet'
-------
TABLE OF CONTENTS
1. INTRODUCTION
2. GENERAL REQUIREMENTS
2.1 Application
2.1.1 Types of Water Supplies
2.1.2 Determination of Applicable Sources
2.2 Treatment Requirements
2.3 Operator Personnel Qualifications
3. COMPLIANCE FOR SYSTEMS NOT FILTERING
3.1 Source Water Quality Criteria
3.1.1 Coliform Concentrations
3.1.2 Turbidity Levels
3.2 Disinfection Criteria
3.2.1 Inactivation Requirements
3.2.2 Determination of Overall Inactivation for Residual Profile, Multiple Disinfectants and Multiple
Sources and Multiple Sources
3.2.3 Demonstration of Maintaining a Residual
3.2.4 Disinfection System Redundancy
3.3 Site-Specific Conditions
3.3.1 Watershed Control Program
3.3.2 On-site Inspection
3.3.3 No Disease Outbreaks
3.3.4 Monthly Coliform MCL
3.3.5 Total Trihalomethane (TTHM) Regulations
-------
DESIGN AND OPERATING CRITERIA FOR FILTRATION AND
DISINFECTION TECHNOLOGY
4.1
Introduction
4.2
Selection of Appropriate Filtration Technology
4.2.1
General Descriptions
4.2.2
Capabilities
4.2.3
Selection
4.3
Available Filtration Technologies
4.3.1
Introduction
4.3.2
General
4.3.3
Conventional Treatment
4.3.4
Direct Filtration
4.3.5
Slow Sand Filtration
4.3.6
Diatomaceous Earth Filtration
4.3.7
Alternate Technologies
4.3.8
Nontreatment Alternatives
4.4
Disinfection
4.4.1
General
4.4.2
Recommended Removal/lnactivation
4.4.3
Total Trihalome.thane (TTHM) Regulations
CRITERIA FOR DETERMINING IF FILTRATION AND DISINFECTION
ARE
SATISFACTORILY PRACTICED
5.1
Introduction
5.2
Turbidity Monitoring Requirements
5.2.1
Sampling Location
5.2.2
Sampling Frequency
5.2.3
Additional Monitoring
5.3
Turbidity Performance Criteria
5.3.1
Conventional Treatment or Direct Filtration
5.3.2
Slow Sand Filtration
5.3.3
Diatomaceous Earth Filtration
5.3.4
Other Filtration Technologies
5.4
Disinfection Monitoring Requirements
5.5
Disinfection Performance Criteria
5.5.1
Minimum Performance Criteria Required Under the SWTR
5.5.2
Recommended Performance Criteria
5.5.3
Disinfection By-Product Considerations
5.5.4
Recommended Disinfection System Redundancy
5.5.5
Determination of Inactivation by Disinfection
5.6
Other Considerations
-------
REPORTING
6.1 Reporting Requirements for Public Water Systems Not Providing Filtration
6.2 Reporting Requirements for Public Water Systems Using Filtration
COMPLIANCE
7.1 Introduction
7.2 Systems Using a Surface Water Source Not Ground Water Under the Direct nfluence of
Surface Water)
7.3 Compliance Transition with Current NPDWR Turbidity Requirements
7.4 Systems Using a Ground Water Source Under the Direct Influence of a Surface Water
7.5 Responses for Systems not Meeting the SWTR Criteria
7.5.1 Introduction
7.5.2 Systems Not Filtering
7.5.3 Systems Currently Filtering
PUBLIC NOTIFICATION
EXEMPTIONS
9.1 Overview of Requirements
9.2 Recommended Criteria
9.3 Compelling Factors
9.4 Evaluation of Alternate Water Supply Sources
9.5 Protection of Public Health
9.6 Notification to EPA
-------
List of Tables
Table Description
2-1 Survey Form for the Classification of Drinking Water Sources
4-1 Removal Capabilities of Filtration Processes
4-2 Generalized Capability of Filtration Systems to Accommodate Raw Water Quality Conditions
6-1 Source Water Quality Conditions for Unfiltered Systems
6-2 Long Term Source Water Quality Conditions for Unfiltered Systems
6-3 CT Determination for Unfiltered Systems Monthly Report to Primacy Agency
6-4 Disinfection Information for Unfiltered Systems Monthly Report to Primacy Agency
6-5 Distribution System Disinfectant Residual Data for Unfiltered and Filtered Systems - Monthly
Report to Primacy Agency
6-6 Monthly Report to Primacy Agency for Compliance Determination - Unfiltered Systems
6-7 Daily Data Sheet for Filtered Systems
6-8 Monthly Report to Primacy Agency for Compliance Deterimination-Filtered Systems
7-1 Requirements for Unfiltered Systems
7-2 Requirements for Filtered Systems
2-1 Steps to Source Classification
3-1 Determination of Inactivation for Multiple Disinfectant Application to a Surface Water Source
3-2 Individually Disinfected Surface Sources Combined at a Single Point
3-3 Multiple Combination Points for Individually Disinfected Surface Sources
4-1 Flow Sheet for a Typical Conventional Water Treatment Plant
4-2 Flow Sheet for Typical Softening Treatment Plants
4-3 Flow Sheet for a Typical Direct Filtration Plant
4-4 Flow Sheet for a Typical Direct Filtration Plant with Flocculation
LIST OF APPENDICES
Appendix Description
A Use of Particulate Analysis for Source and Water Treatment Evaluation
B Institutional Control of Legionella
C Determination Of Disinfectant Contact Time D Analytical Requirements of the SWTR and a Survey of
the Current Status of Residual Disinfectant Measurement Methods for all Chlorine Species and Ozone
E Inactivation Achieved by Various Disinfectants
F Basis for CT Values
G Protocol for Demonstrating Effective Disinfection
H Sampling Frequency for Total Coliforms in the Distribution System
I Maintaining Redundant Disinfection Capability
J Watershed Control Program
K Sanitary Survey
L Small System Considerations
M Protocol for Demonstration of Effective Treatment
N Protocols for Point-of-Use Treatment Devices
0 Guidelines to Evaluate Ozone Disinfection
-------
1. INTRODUCTION
This Guidance Manual complements the filtration and disinfection
treatment requirements for public Mater systems using surface water
sources or ground water under the direct Influence of surface water
promulgated In 40 CFR Part 141, Subpart H. In this manual, these
requirements are referred to as In the Surface Water Treatment Rule
(SWTR).
The purpose of this manual 1s to provide guidance to United States
Environmental Protection Agency (USEPA) Regional Offices, Primacy Agencies
and affected utilities 1n the Implementation of the SWTR, and to help
assure that Implementation 1s consistent. For example, the SWTR sets
treatment requirements which apply to a large range of source water
conditions. The guidance manual suggests design, operating and perform-
ance criteria for specific surface water quality conditions to provide the
optimum protection from microbiological contaminants. These recommenda-
tions are presented as advisory guidelines only; unlike the provisions of
the SWTR, these recommendations are not mandatory requirements. In many
cases, it will be appropriate to tailor requirements to specific
circumstances; the guidance manual is designed to give the Primacy Agency
flexibility 1n establishing the most appropriate treatment requirements
for the systems within their Jurisdiction.
Throughout this document, the term "Primacy Agency" refers to a
State with primary enforcement responsibility for public water systems or
"primacy," or to mean EPA 1n the case of a State that has not obtained
primacy.
In order to facilitate the use of this manual, 1t has been
structured to follow the framework of the SWTR as closely as possible.
Brief descriptions of the contents of each section of this manual are
presented 1n the following paragraphs.
Section 2
This section provides guidance for determining whether a water
supply source 1s subject to the requirements of the SWTR including the
determination of whether a ground water source 1s under the direct
1-1
-------
Influence of surface water,I.e. at risk for the presence of 6lard1a cysts
•or other large microorganisms. The overall treatment requirements of the
SWTR are.also presented, along with recommendations for the qualifications
of operator personnel.
Section 3
For systems which are subject to the requirements of the SWTR and
which do not currently provide filtration, this section provides guidance
to the Primacy Agency for determining 1f a given system:
Meets the source water quality criteria
Meets the disinfection requirements Including:
99.9 and 99.99 percent 1nact1vat1on of Glardla cysts and
viruses and application of the CT (disinfectant residual
concentration x contact time) concept
Point of entry to distribution system requirements
Distribution system requirements
Provision for disinfection system redundancy
Maintains an adequate watershed control program
Meets the on-site Inspection requirements
Has not had an Identified waterborne disease outbreak
Complies with the requirements of the revised Total CoHform
Rule
Complies with Total Tribalomethane (TTHM) Rule
Section 4
This section pertains to systems which do not meet the requirements
to avoid filtration outlined 1n Section 3 and therefore are required to
Install filtration. Guidance 1s given for the selection of an appropriate
filtration technology based on the source water quality and the capabili-
ties of various technologies to achieve the required performance criteria.
In addition, recommended design and operating criteria are provided for
different filtration technologies.
1-2
-------
Section 5
Section 5 presents guidance to the Primacy Agency for determining
compliance with the turbidity and disinfection performance requirements,
and 1n turn, whether filtration and disinfection are satisfactorily
practiced. Recommendations are made for the level of disinfection to be
provided in order to meet the overall treatment requirements of the SWTR.
This section describes how to evaluate the adequacy of disinfection using
CT or other methods.
Section 6
Section 6 provides guidelines to the Primacy Agency for establishing
the reporting requirements associated with the SWTR. The requirements
include report content and frequency, and are applicable to both filtering
and nonfllterlng systems.
Section 7
This section provides an overview of the schedule for Primacy
Agencies and utilities to meet the requirements of the SWTR. Examples are
presented to provide guidance for corrective measures which can be taken
by systems which are not in compliance with the treatment requirements.
This section presents guidance on public notification. Included are
examples of events which would require notification, language for the
notices and the methods of notification.
Section 9
Section 9 provides guidance to the Primacy Agency for determining
whether a system is eligible for an exemption. The criteria for
eligibility for an exemption include:
Compelling factors (economic or resource limitations)
No available alternate source
Protection of public health
This section also provides guidance for evaluating the financial
capabilities of a water system, reviewing the availability of alternate
sources and suggests interim measures for protecting public health.
1-3
-------
Appendices
The manual also contains appendices which provide more detailed
guidance 1n specific areas. These Include:
Appendix A - EPA Consensus
Method for Glardla cvst Analysis
Several procedures are available for 61ard1a cyst analysis In water.
In 1983 the USEPA held a conference to establish a consensus on the
procedure to be used 1n the future. This consensus method would promote
uniformity 1n testing and provide a basis for future conparlsons. The
consensus method and the background data used to develop 1t are presented
In this appendix.
Appendix B - Institutional
Control of Legionella
Filtration and/or disinfection provides protection from Legionella.
However, 1t does not assure that recontamlnation or regrowth will not
occur 1n the hot water or cooling systems of buildings within the
distribution system. This appendix provides guidance for monitoring and
treatment which can be used by Institutional systems for the control of
Legionella.
Appendix C - Determination of Disinfectant
Contact Time
In many cases, the determination of disinfectant contact times
needed to evaluate the CT of a water system will necessitate the use of
tracer studies. This appendix provides guidance for .conducting these
studies. In some cases It may not be practical to conduct a tracer study.
For such cases guidance 1s given for estimating the detention time based
on the physical configuration of the system.
1-4
-------
Appendix D - Analytical Requirements
of the SWTR and A Survey of the Current
Status of Residual Disinfectant
Measurement Methods for all Chlorine
Spades and Ozone ; ;
This appendix includes a listing of the analytical methods required
under the SWTR, An executive stannary of a report on the analytical
methods used to measure the residual concentrations of the various
disinfectants is included. The reliability and limitations of each of the
methods are presented.
Appendix E - Inactivatlons Achieved
bv the Various Disinfectants
This appendix presents the log inactivatlons of Giardia cysts and
viruses which are achieved at various CT levels by chlorine, chlorine
dioxide, chloramlnes and ozone. Inactivatlons of viruses achieved by UV
absorbance are also included.
Aooendix F - Basis for CT Values
This appendix provides the background and rationale utilized In
developing the CT values for the various disinfectants. Included 1s a
paper by Clark and Regli, 1990, in which a mathematical model was used in .
the determination of CT values for free chlorine.
Appendix 6 - Protocol for Demonstrating
Effective Disinfection
This appendix provides the recommended protocols for demonstrating
the effectiveness of chloramlnes, chlorine dioxide and ozone as primary
disinfectants.
Appendix H - Sampling Frequency for
Total Conforms in the Distribution System
The sampling frequency required by the revised Total Col 1 form Rule
54 FR 27544 (June 29, 1989) is presented in this appendix.
1-5
-------
Appendix I - Maintaining
Redundant Disinfection Capability
This appendix details the conditions and equipment which should be
maintained by a system using chlorine, chlorine dioxide, ozone or
chloramines to assure that compliance with the SWTR requirement for
redundant disinfection Is met.
Appendix 4 - Watershed Control Program
This appendix provides a detailed outline of • watershed program.
This program may be adjusted by the Primacy Agency to serve the specific
needs of a particular water system.
Appendix K - Sanitary Survey
This appendix provides guidance for conducting a comprehensive
sanitary survey of a supply source and Its treatment and delivery to the
consumer. Suggested elements of an annual on-site Inspection are Included
1n Section 3.
Appendix L - Small System Considerations
This appendix describes difficulties which say be faced by small
systems in complying with the SWTR along with guidelines for overcoming
these difficulties.
Appendix M - Protocol for the
Demonstration of Effective Treatment
This appendix presents pilot study protocols to evaluate the
effectiveness of an alternate filtration technology in meeting the
performance requirements of the SWTR. It presents the use of particle
size analysis for demonstrating the actual removal of 61ard1a cyst
achieved by a treatment train. Guidance for conventional and direct
filtration plants to demonstrate that adequate filtration 1s being*
maintained at effluent turbidities between 0.5 and 1 Nephelometric
Turbidity Unit (NTU) Is also Included.
1-6
-------
Appendix N - Protocol for
Point-of-Use Treatment Devices
In some United cases, 1t nay be appropriate to Install po1nt-of-use
(POU) or polnt-of-entry (POE) treatnent devices as an Interim measure to
provide protection to the public health. This appendix provides a
protocol for evaluating and determining the efficacy of POU/POE treatment
devices.
Appendix 0 - Guidelines to
Evaluate Ozone Disinfection
The CT evaluation used for other disinfectants Is Inappropriate for
ozone. This appendix presents alternative methods for evaluating the
disinfection effectiveness of ozone systems.
1-7
-------
2. general requirements
2.1 Abb]Icfltlon
The SWTR pertains to all public water systems which utilize a surface
water source or ground water source under the direct Influence of,surface
water. The SWTR defines a surface water as all waters which are open to
the atmosphere and subject to surface runoff. Ground water under the
direct influence of surface water is defined as; any water beneath the
surface of the ground with (i) significant occurrence of insects or other
macroorganisms, algae, organic debris, or large-diameter pathogens such as
Giardia Iambi 1a, or (11) significant and relatively rapid shifts In water
characteristics such as turbidity, temperature, conductivity, or pH which
closely correlate to climatological or surface water conditions. Direct
influence must be determined for each individual source in accordance with
criteria established by the Primacy Agency. The Primacy Agency criteria
may provide for documentation of well construction and geology, with field
evaluation, or site-specific measurements of water quality as explained in
Section 2.1.2.
Saline water sources such as the ocean are not generally considered
to be subject to the requirements of the SWTR because of the low survival
time of pathogens in a saline environment (Geldreich, 1989). Pathogens
generally can only survive a few hours in saline water and any remaining
pathogens should be removed or inactivated during desalination. However,
it is up to the Primacy Agency's discretion to determine which systems
must meet the SWTR requirements. In cases where there is a sewage
discharge located near the water intake, it may be appropriate for the
Primacy Agency to require the system to comply with the SWTR.
The traditional concept that all water in subsurface aquifers 1s free
from pathogenic organisms is based upon soil being an effective filter
that removes microorganisms and other relatively large particles by
straining and antagonistic effects (Bouwer, 1978). In most cases
pathogenic bacteria retained 1n the soil find themselves in a hostile
environment, are not able to multiply and eventually die. However, some
underground sources of drinking water may be subject to contamination by
pathogenic organisms from the direct influence of nearby surface waters.
Only those subsurface sources which are at risk to contamination from
Giardia cysts will be subject to the requirements of the SWTR. Giardia
2-1
-------
cysts generally range 1n size from 7 to 12 um. Subsurface sources which
nay be at risk to contamination from bacteria and enteric viruses, but
which are not at risk from Giardia cysts will be regulated either under
the Total Collforra Rule or forthcoming disinfection treatment requirements
for ground waters. EPA Intends to promulgate disinfection requirements
for ground water systems in conjunction with regulations for disinfection
by-products by 1992.
2.1.1 Types of Water Supplies
Surface Haters
Surface water supplies that are often used as sources of drinking
water include two major classifications, running and quiescent waters.
Streams, rivers and brooks are examples of running water, while lakes,
reservoirs, impoundments and ponds are examples of quiescent waters. The
exposure of surface waters to the atmosphere results in exposure to
precipitation events, surface water runoff and contamination with micro
and macroorganlsms resulting from activities in their surrounding areas.
These sources are subject to the requirements of the SWTR.
Systems with rain water catchments not subject to surface runoff
(e.g. roof catchment areas) are not considered vulnerable to contamination
from animal populations which carry protozoan cysts pathogenic to humans
and are thus not subject to the SWTR requirements. However, such systems
should at least provide disinfection to treat for potential bacterial and
viral contamination coming from bird populations.1
Ground Waters under Direct Influence of Surface Water
Ground water sources which may be subject to contamination with
pathogenic organisms from surface waters include, springs, infiltration
galleries, wells or other collectors in subsurface aquifers. The
following section presents a recommended procedure for determining whether
a source will be subject to the requirements of the SWTR. These
determinations are to be made for each individual source. If the
determination will Involve an evaluation of water quality, eg. particulate
analysis, it 1s important that these analyses be made on water taken
One study (Markwell and Shortridge, 1981) indicates that a
cycle of waterbome transmission and maintenance of influenza
virus may exist within duck communities, and that it is
conceivable for virus transmission to occur in this manner to
other susceptible animals, including humans.
2-2
-------
directly from the source and not on blended water or water from the
distribution system.
2.1.2 Determination of Applicable Sources
The Primacy Agency has the responsibility for determining which water
supplies must meet the requirements of the SWTR. However, it is the
responsibility of the water purveyors to provide the Primacy Agency with
the information needed to make this determination. This section provides
guidance to the Primacy Agency for determining which water supplies are
surface waters or ground waters directly Influenced by a surface water and
are thereby subject to the requirements of the SWTR. Following the
determination that the source is subject to the SWTR, the requirements
enumerated In Sections 2.2 and 2.3 must be met.
The Primacy Agency must develop a program for evaluating ground water
sources for direct Influence by December 30, 1990. All community ground
water systems must be evaluated by June 29, 1994, while all non-community
systems must be evaluated by June 29, 1999. Primacy Agencies with an
approved Wellhead Protection (HHP) Program, may be able to use the WHP
program's requirements which include delineation of wellhead protection
areas, assessment of sources of contamination and Implementation of
management control measures. These same requirements can be used for
meeting the requirements of the watershed control program for ground water
under the direct Influence of a surface water.
A multiple step approach has been developed as the recommended method
of determining whether a ground water source is under direct influence of
a surface water. This approach includes the review of information
gathered during sanitary surveys. As defined by the USEPA, a sanitary
survey is an on-site review of the water source, facilities, equipment
operation and maintenance of a public water system for the purpose of
evaluating the adequacy of such source, facilities, equipment, operation
and maintenance for producing and distributing safe drinking water.
Sanitary surveys are required under the Total Coliform Rule and may be
required under the forthcoming disinfection requirements for ground water
systems as a condition for obtaining a variance or for determining the
level of disinfection required. Therefore, It is recommended that the
determination of direct Influence be correlated with the sanitary surveys
conducted under these other requirements.
2-3
-------
A. Source Evaluation Pr^g^l
As Illustrated on Figure 2-1, the determination of whether a source
is subject to the requirements of the SWTR nay involve one or more of the
following steps:
1. A review of the records of the system's source(s) to determine
whether the source is obviously a surface water, I.e. pond,
lake, streams, etc.
2. If the source 1s a well, determination of whether 1t Is clearly
a ground water source, or whether further analysis 1s needed
to determine possible direct surface water Influence.
3. A complete review of the system's files followed by a field
sanitary survey. Pertinent Information to gather 1n the file
review and field survey Includes: source design and construc-
tion; evidence of direct surface water contamination; water
quality analysis; Indications of waterborne disease outbreaks;
operational procedures (i.e. pumping rates, etc.); and customer
complaints regarding water quality or water related Infectious
Illness.
4. Conducting particulate analyses and other water quality
sampling and analyses.
Step K Records Review •
A review of information pertaining to each source should be carried
out to identify those sources which are obvious surface waters. These
would include ponds, lakes, streams, rivers, reservoirs, etc. If the
source is a surface water, then the SWTR would apply, and criteria in the
rule would need to be applied to determine if filtration is necessary. If
the source 1s not an obvious surface water, then further analyses, as
presented in Steps 2, 3, or 4, are needed to determine if the SWTR will
apply. If the source is a well (vertical or horizontal), go to Step 2.
If the source is a spring, infiltration gallery, or any other subsurface
source, proceed to Step 3 for a more detailed analysis.
Step 2.
-------
Obvious Surface
Sources:
Lakes, Reservoirs,
Streams, Creeks,
Rivers, etc.
SWTR Applies
All Public Water
Systems
Identify Source Type
Source Is Spring or
Infiltration Gallery
Review System Hie
and Conduct
Sanitary Survey
I
Source Directly
Influenced by
Surface Water?
C Undecided J)
i
Conduct Particulate
Analysis, Monitor
Changes In Water
Quality, Temperature,
etc.
Summary of Findings
Indicate Source is
Influenced by Surface
Water and Could
Contain Glardla?
Source Is Vertical or
Horizontal Well
Well Is Protected from
Direct Surface Influence
Based on Slate
Criteria
Mo #—~¦
<5
SWTR Does Not Apply
.0.
FIGURE 2-1 - STEPS TO SOURCE CLASSIFICATION
-------
consuming and labor Intensive. In an attempt to reduce the effort needed
to evaluate well sources, a set of criteria has been developed to Identify
wells In deep, well protected aquifers which are not subject to contamina-
tion from surface water. While these criteria are not as definitive as
water quality analysis, it 1s believed that they provide a reasonable
degree of accuracy, and allow for a relatively rapid determination for a
large number of well sources 1n the U.S.
Wells with perforations or a well screen less than or equal to 50
feet in depth are considered to be shallow wells, and should be evaluated
for direct surface Influence according to steps 3 and/or 4. For wells
greater than 50 feet 1n depth, State or system files should be reviewed
for the criteria listed below:
1. The well construction should include:
A surface sanitary seal using bentonite clay, concrete
or other acceptable material.
A well casing that penetrates a confining bed.
A well casing or collector laterals that are only
perforated or screened below a confining bed.
The importance of evaluating the hydrogeology of wells or
collectors, even those more than 200 feet from a surface water,
cannot be overstated. The porosity and transmlsslvlty of
aquifer materials, hydrologic gradients, and continuity of
confining layers above screens or perforations may need to be
considered 1n detail for some sources. Porous aquifer material
is more likely to allow surface water to directly Influence
ground water than finer grained materials. In addition, high
well pumping rates may alter the existing hydrologic gradient.
Ground water flow direction may change such that surface water
is drawn Into a collector, whereas under low pumping rates It
may not. Evaluating pumping rate effects and other hydrogeolo-
gic information must be done on a site specific basis.
If Information on well construction or hydrogeology are
incomplete or raise questions regarding potential surface water
influence, a more detailed analysis in steps 3 and 4 should be
considered.
2. The casing or nearest collector lateral should be located at
least 200 feet from any surface water.
3. The water quality records should indicate:
2-5
-------
No record of total coliform or fecal coll form contamina-
tion in untreated samples collected over the past three
y«ars%
- No history of turbidity problems associated with the
source. .
No history of known or suspected outbreak of Glardia. or
other pathogenic organism associated with surface water
(e.g. CrvgtQSPQritiiyffl). which has been attributed to that
source%
4. If data 1s available for particulate matter in the well there
should be:
No evidence of particulate matter associated with
surface water.
If data is available for turbidity or temperature from the well
and a nearby surface water there should be:
No turbidity or temperature data which correlates
to that of a nearby surface water.
Wells that meet all of the criteria listed above are not subject to
the requirements of the SWTR, and no additional evaluation is needed.
Wells that do not meet all the requirements listed require further
evaluation 1n accordance with Steps 3 and/or 4 to determine whether or not
they are directly influenced by surface water.
Step 3. On-site Inspection
For sources other than a well source, the State or system files
should be reviewed for the source construction and water quality
conditions as listed in Step 2. Reviewing historical records 1n State or
system files is a valuable Information gathering tool for any source.
However, the results may be Inconclusive. A sanitary survey In the field
may be helpful in establishing a more definite determination of whether
the water source 1s at risk to pathogens from direct surface water
influence.
Information to obtain during an on-site inspection include:
Evidence that surface water enters the source through defects
in the source such as lack of a surface seal on wells, infil-
2-6
-------
tration gallery laterals exposed to surface water, springs open
to the atmosphere, surface runoff entering a spring or other
collector, etc.
Distances to obvious surface water sources.
If the survey indicates that the well is subject to direct surface
water Influence, the source must either be reconstructed as explained
later 1n this section or 1t must be treated in accordance with the
requirements for the SWTR. If the survey does not show conclusive
evidence of direct surface water Influence, the analysis outlined 1n Step
4 should be conducted.
The Washington State Department of Social and Health Services has
developed a form to guide them and provide consistency in their evaluation
of sources for surface water influence (Notestine & Hudson, 1988). Table
2-1 provides a copy of this form as a guide for evaluating sources.
Step 4. Particulate Analysis and Other Indicators
a. Surface Water Indicators
Particulate analysis is intended to Identify organisms which only
occur in surface waters as opposed to ground waters, and whose presence in
a ground water would clearly indicate that at least some surface water has
been mixed with 1t. The EPA Consensus Method in Appendix A can be used
for Giardia cyst analysis.
In 1986 Hoffbuhr et. al. listed six parameters Identifiable in a
particulate analysis which were believed to be valid indicators of surface
contamination of ground water. These were: diatoms, rotifers, coccldia,
plant debris, insect parts, and Giardia cysts. Later work by Notestine
and Hudson (1988) found that microbiologists did not all define plant
debris in the same way, and that deep wells known to be free of direct
surface water influence were shown by particulate analysis to contain
"plant debris" but none of the other five indicators. Their work suggests
that "plant debris" may not currently be a useful tool in determining
direct surface water influence, but may be 1n the future when a standard
definition of "plant debris" is developed. Therefore, it is recommended
that only the presence of the other five parameters; diatoms and certain
other algae, rotifers, coccidia, insect parts, and Giardia. be used as
-------
TABLE 2-1
SURVEY FORM FOR THE CLASSIFICATION OF DRINKING WATER SOURCES
general
1. Utility Name (ID#)
2. Utility Person(s) Contacted
3. Source Type (As shown on state Inventory)
Spring Horizontal Well Vertical
Infiltration System Shallow Well Well
4. Source Name Year constructed
5. Is this source used seasonally or intermittently? No Yes
If yes, are water quality problems the reason? No ; Yes
6. Has there ever been a waterbome disease outbreak associated with
this source? Yes No If yes, explain
7. Have there been turbidity or bacteriological MCL violations within
the last five years associated with this source? No Yes
If yes, describe frequency, cause, remedial action (s) taken
8. Have there been consumer complaints within the past five years
associated with this source? No Yes If yes, discuss
nature, frequency, remedial action taken
9. Is there any evidence of surface water intrusion (pH, temperature,
conductivity, etc. changes) during the year? Yes No
Ifyes, describe
If not, submit supporting data.
10. Sketch of source in plan view (on an additional sheet)
-1-
-------
Shallow Wells
1. Does the well meet good sanitary practices regarding location, con-
struction, seal etc. to prevent the entrance of surface water?
Yes No If no, describe the deficiencies
2. What Is the depth of the well? (ft)
Elevation of top of casing? (ft msl)
Elevation of land surface? (ft msl)
3. Hydrogeology (Attach copy of well log or summarize it on reverse)
a. Depth to static water level? (Feet) "
b. Drawdown? (Feet)
c. What 1s the depth to the highest screen or perforation?
(Feet) .
d. Are there Impervious layers above the highest screen of
perforation?
Yes No Unknown
If yes, please describe ________________________
4. Is there a permanent or Intermittent surface water within 200 feet
of the well? Yes No If yes, describe (type, distance
etc.) and submit location map
What is the elvation of normal pool? (ft msl)
Elevation of 100 year flood level? (ft msl)
Elevation of bottom of lake or river? (ft msl)
5. Additional comments:
-------
Springs
1. a. What is the size of the catchment area (acres)?
b. Give a general description of the area (terrain; vegetation;
soil etc.)
2. What is the vertical distance between the ground surface and the
nearest point of entry to the spring collector(s) (feet)?
3. How rapidly does rainfall percolate into the ground around the
spring?
Percolates readily; seldom if ever any runoff.
Percolates readily but there is sane runoff in heavy rain.
Percolates slowly. Host local rainfall ponds or runs off.
Other
4. Does an impervious layer prevent direct percolation of surface water
to the collector(s)? Yes No Unknown
5. Is the spring properly constructed to prevent entry of surface
water? Yes No
6. Sediment
a. Is the spring box free of debris and sediment? Yes No ;
b. When was it last cleaned (Date)
c. .How often does it need to be cleaned? (month)
d. How much sediment accumulates between cleaning? (estimate in
inches)
7. Additionalcomments:
-3-
-------
Infiltration Systems
1. What are the shortest distances (vertical and horizontal separating
the collector from the nearest surface water? (Feet)
2. Does turbidity of the source vary 0.2 NTU or more throughout the
year? Yes No Not measured
If yes, describe how often and how much (pH, temperature,
conductivity, etc.)
3. AdditionalComments
Survey Conducted By: Date:
Decision? Surface Impacted Source Yes No If no,
further evaluation needed (particulate analysis, etc.)
-4-
-------
indicators of direct surface contamination. In addition, if other large
diameter (> 7 urn) organisms which are clearly of surface water origin such
as Diohilobothrium are present, these should also be considered as
indicators of direct surface water influence.
b.- interpretation
Since standard methods have not been developed specifically for
particulate analysis, there has not been consistency in the way samples
have been collected and analyzed. Differences in the degree of training
and experience of the microbiologists has added further to the difficulty
in comparing results from sample to sample, and system to system. The
current limitations in sample collection and analytical procedures must be
considered when interpreting the results. Until standardized methods are
developed, the EPA Consensus Method included in Appendix A is recommended
as the analytical method for particulate analysis. The following 1$ a
discussion of the significance of finding the six indicators identified
above.
Identification of a Giardia cyst in any source water should be
considered conclusive evidence of direct surface water influence. The
repeated presence of diatoms in source water should be considered as
conclusive evidence of direct surface water influence. However, it is
important that this determination be based on live diatoms, and not empty
silica skeletons which may only indicate the historical presence of.
surface water.
Bluegreen, green, or other chloroplast containing algae require
sunlight for their metabolism as do diatoms. For that reason their
repeated presence In source water should also be considered as conclusive
evidence of direct surface water influence.
Hoffbuhr (1986) indicates that rotifers and insect parts are
indicators of surface water. Others have pointed out though that rotifers
do not require sunlight, and not all rotifers require a food source such
as algae which originates in surface water. Their nutritional require-
ments may be satisfied by organic matter such as bacteria, or decomposing
soil organic material, not necessarily associated with surface water.
More precise identification of rotifers, i.e. to the species level, is
necessary to determine the specific nutritional requirements of the
rotifer(s) present. Further information on identifying rotifer species
and on which species require food sources originating in surface water,
2-8
-------
would be valuable, but 1s not readily available at this time. Without
knowledge of which species Is present, the finding of rotifers Indicates
that the source Is either a) directly Influenced by surface water, or b)
1t contains organic matter sufficient to support the growth of rotifers.
It could be conservatively assumed based on this evidence alone that such
a source 1s directly Influenced by surface water. However, It 1s
recommended that this determination be supported by other evidence, eg.
the source 1s near a surface water, turbidity fluctuations are signifi-
cant, etc.
Insects or insect parts likewise may originate 1n surface water, from
the soil, or they may be airborne in uncovered sources. If insects are
observed 1n a particulate analysis sample, 1t should be confirmed 1f
possible that there is no other route by which Insects could contaminate
the source other than surface water. For example, 1f a spring is sampled,
and the cover Is not well constructed, it is possible that Insects found
in a sample were airborne rather than waterborne. Insects which spend a
portion of their Hfecycle 1n water are the best indicators of direct
surface water influence, for example, larvae of mayflies, stoneflies,
damsel flies, and dragonflies. Terrestrial insects should not be ruled out
as surface water indicators though, since their accidental presence in
surface water is common.
Howell, (1989) has Indicated that some Insects may burrow and the
finding of eggs or burrowing larvae (eg. chironomids) may not be good
indicators of direct surface water influence. For some insects this may
be true, but the distance which insects burrow 1n subsurface sediments 1s
expected to be small, and insect larvae are generally large in comparison
to Giardia cysts. Until further research suggests otherwise, It 1s
recommended that Insects or insect parts be considered strong evidence of
surface water influence if not direct evidence in and of themselves. The
strength of this evidence would be increased if the source in question is
near a surface water, and particulate analysis of the surface water found
similar insects.
Coccidia are intracellular parasites which occur primarily in verte-
brates, eg. animals and fish, and live in various tissues and organs
including the intestinal tract (eg. Cryptosporidium). Though not
frequently identified by normal particulate analysis techniques, coccidia
are good indicators of direct surface water contamination since they
2-9
-------
require a vertebrate host or hosts and are generally large 1n size (10 -
20 um or greater). Cryptosporidium 1s commonly found 1n surface water,
but due to Its small size (4-6 um) 1t 1s not normally Identified without
specific antibody staining techniques.
Other macroorganlsms (>7 um) which are parasitic to animals and fish
may be found and are good Indicators of surface water Influence. Examples
Include, but are not limited to, helminths (e.g., tape worm cysts),
ascarls, and Dlphyllobothrium.
c. <«BpHna Method
A suggested protocol for collecting samples 1s listed below.
- Sampling Procedure
Samples should be collected using the equipment outlined 1n the
EPA Consensus Method Included In Appendix A.
Location
Samples should always be collected as close to the source as
possible, and prior to any treatment. If samples must be taken
after disinfection, samples should be noted and analyzed as
soon as possible.
- Number
A minimum of two samples should be collected during the period
the source 1s most susceptible to surface water Influence.
Such critical periods will vary from system to system and will
need to be determined case by case. For some systems, 1t may
be one or more days following a significant rainfall (eg. 2"
in 24 hours). For other systems 1t may be a period of maximum
flows and stream turbidities following spring snowmelt, or
during the summer months when water tables are elevated as a
result of Irrigation. In each case, particulate samples should
be collected when the source 1n question 1s most effected. A
surrogate measure such as source turbidity or depth to water
table may be useful 1n making the decision to monitor. If
there 1s any ambiguity 1n the particulate analysis results,
additional samples should be collected when there 1s the
greatest likelihood that the source will be contaminated by
surface water.
Sample volume should be between 5Q0 and 1000 gallons, and
should be collected over a 4 to 8 hour time period. It is
preferable to analyze a similar (+/- 10%) volume of water for
all sources, preferably a large volume, although this may not
always be possible due to elevated turbidity or sartplina
logistics. The volume filtered should be recorded for all
samples.
2-10
-------
d. Other Indicators
A number of other indicators could be used to provide supportive
evidence of surface influence. While particulate analysis probably
provides the most direct evidence that pathogens from surface water could
be migrating into a ground water source, other parameters such as
turbidity, temperature, pH arid conductivity could provide supportive, but
less direct, evidence.
Turbidity fluctuations of greater than 0.5 - 1 NTU over the course
of a year may be indicative of surface water influence. Considerable
caution should be used when evaluating turbidity changes though, since the
turbidity could be caused by very snail particles (< lum) not originating
in a surface water or it could be that larger particles are being filtered
out and only the very smallest particles migrate Into the water source.
Only ground water sources at risk to contamination from Giardia or other
large pathogens (> 7 um) are subject to the SWTR requirements.
Temperature fluctuations may also indicate surface water influence.
Fortunately these are easy to obtain and If there is a surface water
within 500 feet of the water source, measurements of both should be
recorded for comparison. Large changes in surface water temperature
closely followed by similar changes in source temperature would be
indicative of surface water influence. Also,, temperature changes (in
degrees F) of greater than 15 to 20% over the course of a year appear to
be a characteristic of some sources influenced by surface water (Randall.
1970). Changes in other chemical parameters such as pH, conductivity,
hardness,etc. could also be monitored. Again, these would not give a
direct Indication of whether pathogens originating in surface water were
present, but could Indicate whether the water chemistry was or was not
similar to a nearby surface water and/or whether source water chemistry
changed 1n a similar pattern to surface water chemistry. At this time no
numerical guidelines are available to differentiate what is or is not
similar, so these comparisons are more qualitative than quantitative.
b. Swwal Ssurcts
Some sources may only be used for part of the year, for example
during the summer months when water usage is high. These sources should
not be excluded from evaluation and, like other sources, should be
evaluated during their perlod(s) of highest susceptibility. Particular
2-11
-------
attention should be given to those sources which appear to be directly
influenced by surface water during part of the year. There may be times
during which these subsurface water sources are not Influenced by surface
water and other times when they are part or all surface water. If that is
the case, then 1t is critical that careful testing be done prior to,
during and at the end of the use of the source. This should be done over
several seasons to account for seasonal variation. In practice, 1t is
preferable to use sources which are less vulnerable to contamination since
susceptible sources will necessitate ongoing monitoring and close
attention to operation.
C. Modification of Sources
Sources directly influenced by surface water may be altered in some
cases to eliminate the surface water contamination. Primacy Agencies may
elect to allow systems with such sources to modify the construction of the
source and/or the area surrounding the source in an effort to eliminate
surface water contamination. Since this could be expensive and take
considerable time to evaluate for effectiveness, careful consideration
should be given to the decision to modify a source. In deciding whether
source modification 1s appropriate, systems and Primacy Agencies should
consider the following points:
Is the cause of the surface water contamination known? If the
specific cause or point of surface water contamination is not
known, it will not be possible to determine an effective
control strategy. Further, there may be several reasons why
the source is susceptible to direct surface water influence.
For example, an infiltration gallery may receive surface water
because some of its laterals are exposed in the bed of a nearby
stream, and also because laterals distant from the stream are
shallow and are affected by surface runoff. Simply modifying
or eliminating one or the other set of laterals 1n this case
would not entirely eliminate surface water Influence.
- What 1s the likelihood that modification of the source will be
effective? Assuming that the source of contamination has been
identified, the expected effectiveness of control measures
should be evaluated. If the cause is relatively evident, a
crack in a well casing or an uncovered spring box for example;
then there 1s a high degree of confidence that an effective
solution could be developed. Should the nature of the contami-
nation be more diffuse, or widespread, then the merits of
spending time and money to modify the source should be careful-
ly considered. In the case of the example above, eliminating
the use of the laterals under the stream will solve part of the
2-12
-------
problem. However, without considerably more hydrogeologic
Information about the aquifer and the placement of the other
laterals, 1t 1s not clear what, 1f any, control measures would
effectively eliminate direct surface water Influence 1n those
laterals distant from the stream.
If a source is Identified as being directly Influenced by surface
water, and 1t 1s decided to attempt to modify 1t, Interim disinfection
practices which will ensure at least 99.9% 1nact1vat1on of G1ard1a should
be considered. Methods and levels of disinfection which can.be used to
achieve such removals can be found 1n S141.72 (a) of the SWTR and 1n
Section 3.2 of this manual.
A partial listing of types of modifications which could be undertaken
Includes:
Diverting surface runoff from springs by trenching, etc.
Redeveloping springs to capture them below a confining layer.
Covering open spring collectors.
Reconstructing wells to Install sanitary seals, and/or to
screen them 1n a confined (protected) aquifer.
Repairing cracks or breaks In any type of source collector that
allows the entry of surface contaminants.
Discontinue the use of Infiltration laterals which Intercept
surface water.
An extended period of monitoring should follow reconstruction (eg.
through at least two years or critical periods) to evaluate whether the
source Is still directly Influenced by surface water. Preferably
particulate analysis would be used to make such evaluations, but It may be
helpful to use simpler measures, such as temperature and turbidity, as
screening tools. Longer term monitoring at critical times may also be an
appropriate agreement between the system and the Primacy Agency 1f there
1s still doubt about the long term effectiveness of the solution.
If modification Is not feasible, another alternative to avoid having
to comply with the SWTR may be to develop a new well either deeper or at
a different location,
2-13
-------
2.2 Treatment Requirements
According to the SWTR, all community and noncoaanunlty public water
systems which use a surface water source or a ground water under the
direct Influence of a surface water must achieve a minimum of 99.9 percent
(3-log) removal and/or 1nact1vat1on of Glardia cysts, and a minimum of
99.99 percent (4-1og) removal and/or inactivation of viruses. In the SWTR
and this manual, "viruses" means viruses of fecal origin which are
infectious to humans by waterbome transmission. Filtration plus
disinfection or disinfection alone may be utilized to achieve these
performance levels, depending on the source water quality and site
specific conditions. The SWTR establishes these removal and/or Inactiva-
tion requirements based on Giardia and viruses because this level of
treatment will also provide protection from heterotrophic plate count
(HPC) bacteria and Legionella2 as required 1n the SDWA amendments.
Guidelines for meeting the requirements of the SWTR are provided 1n
the remainder of this manual as outlined 1n Section 1. All systems must
meet the operator qualifications presented in Section 2.3.
2.3 Operator Personnel Qualifications
The SWTR requires that all systems must be operated by qualified
personnel. It 1s recommended that the Primacy Agency set standards for
operator qualifications, 1n accordance with the system type and size. In
order to accomplish this, the Primacy Agency should develop a method of
evaluating an operator's competence 1n operating a water treatment system.
Primacy Agencies which do not currently have a certification program are
thereby encouraged to implement such a program. An operator certification
program provides a uniform base for operator qualifications and an
organized system for evaluating these qualifications.
It Is recommended that plant operators have a basic knowledge of
science, mathematics and chemistry Involved with water treatment and
supply. The minimum requirements for at least one key staff member should
Include an understanding of:
In the SWTR and this manual "Legionella" means a genus of
bacteria, some species of which have caused a type of pneumonia
called Legionnaires Disease; the etiologic agent of most cases
of Legionnaires Disease examined has been pneumophila.
2-14
-------
The principles of water treatment and distribution and their
characteristics
The uses of potable water and variations 1n Its demand
The Importance of water quality to public health
The equipment, operation and maintenance of the distribution
system
The treatment process equipment utilized, Us operational
parameters and maintenance
The principles of each process unit (Including the scientific
basis and purpose of the operation and the mechanical compo-
nents of the unit)
Performance criteria such as turbidity, total collfonn, fecal
collform, disinfectant residual, pH, etc. to determine opera-
tlonal adjustments
Common operating problems encountered In the system and actions
to correct them
The current National Primary Drinking Water Regulations, the
Secondary Drinking Water Regulations and monitoring and
reporting requirements
Methods of sample collection and sample preservation
Laboratory equipment and tests used.to analyze samples (where
appropriate)
The use of laboratory results to analyze plant efficiency
Record keeping
Customer relations
Budgeting and supervision (where appropriate)
Training in the areas listed above and others is available through
the American Water Works Association (AWWA) training course series for
water supply operations. The course series Includes a set of four
training manuals and one reference book as follows:
Introduction to Water Sources and Transmission (Volume 1)
Introduction to Water Treatment (Volume 2)
Introduction to Water Distribution (Volume 3)
Introduction to Water Quality Analyses (Volume 4)
2-15
-------
Reference Handbook: Basic Science Concepts and Applications
Instructor Guide and Solutions Manual for Volumes 1, 2, 3 and
4
These manuals are available through the American Water Works Associa-
tion, 6666 West Qulncy Avenue, Denver, Colorado 80235 USA, (303) 794-7711.
The State of California also offers a series of training manuals for
water treatment plant operators prepared by the California State
University School of Engineering In Sacramento. The manuals Include:
1. Water Supply System Operation. (1 Volume)
2. Water Treatment Plant Operation. (2 Volumes)
These operator training manuals are available from California State
University, Sacramento, 6000 J Street, Sacramento, California 95819, phone
(916) 454-6142.
Completion of an established training and certification program will
provide the means of assuring that the operators have received training 1n
their respective area, and are qualified for their position. The
education and experience requirements for certification should be
commensurate with the size and the complexity of the treatment system. At
the present time, some states have Instituted a certification program
while others have not. Following is a summary of the basic contents of a
certification program, which can serve as a guide to the Primacy Agency 1n
developing a complete program.
Board of examiners for the development and Implementation of
the program.
Classification of treatment facilities by grade according to
the size and technology of the facilities.
Educational and experience requirements for operators of the
various treatment facilities according to grade.
A written/oral examination to determine the knowledge, ability
and judgement of the applicants with certification obtained
upon receiving a passing grade.
Renewal program for the license of certification, Including the
requirement of additional coursework or participation 1n
workshops.
The certification program should provide technically qualified
personnel for the operation of the plant.
2-15
-------
The extensive responsibility which 1s placed on the operating
personnel warrants the development of an outline of the responsibilities
and authority of the personnel members to aid them In the efficient
operation of the plant. The major responsibilities which should be
delegated In the outline of responsibilities Include: the. normal
day-to-day operations, preventive maintenance, field engineering, water
quality monitoring, troubleshooting, emergency response, cross-connection
control, Implementation of Improvements, budget formulation, response to
Complaints and public/press contact. A reference which the Primacy Agency
may utilize 1n developing the outline 1s "Water Utility Management
Practices" published by AWWA.
2-17
-------
3* CRITERIA FOR SYSTEMS NOT FILTERING
Th« provisions of the Surface Water Treatment Rule (SWTR) require
that filtration oust be included in the treatment train unless certain
criteria are Ret. These criteria are described in this chapter.' They
include*.
Source water Quality Conditions
1. Colifortn concentrations (total or fecal).
2. Turbidity levels.
Disinfection Criteria
1. Level of disinfection.
2. Point of entry disinfection.
3. Distribution system disinfection.
4. Oisinfection redundancy or automatic shutoff.
Site-Specific Criteria
1. Watershed control program.
2. On-site Inspections.
3. No waterborne disease outbreaks.
4. Complies with the total col 1 form MCI.
5. Complies with the Total Trihalomethane (TTHM) regulation.
Currently this only applies to systems serving more than
10,000 people.
The purpose of this section is to provide guidance to the Primacy
Agency for determining compliance with these provisions^
3.1 Source Water Quality Criteria
The first step in determining if filtration is required for a given
surface water supply is to determine whether the supply meets the source'
water quality criteria as specified in the SWTR. If the supply does not
meet the source water quality criteria, changes in operation to meet the
site-specific criteria may improve the water quality so that the source
3-1
-------
criteria will be act. However, If the Primacy Agency believes that the
source water quality criteria and/or the site-specific criteria cannot be
net, or that filtration Is appropriate regardless, the Primacy Agency may
require the Installation of filtration without a complete evaluation to
determine whether the system meets all the criteria required to avoid
filtration.
Sampling location
The SWTR requires that source water samples be collected at a loca-
tion just prior to the "point of disinfectant application,1* i.e., where
the water Is disinfected and no longer subject to surface runoff. For
example, a system which has multiple reservoirs In series, where each of
the reservoirs has previously been disinfected and receives surface
runoff, must take the raw water sample(s) just prior to the point of
disinfection or disinfection sequences used for calculating the CT
[disinfectant residual (mg/L) x contact time (min.)]. Disinfected water
in reservoirs receiving surface runoff cannot be counted toward CT credit.
It is also not appropriate for systems to monitor the source water after
the "point of disinfectant application" even if disinfection from this
point is not used for calculating CT credit.
3.1.1 Conform Concentrations; The SWTR states that, to avoid
filtration, a system must demonstrate that either the fecal col 1 form
concentration is less than 20/100 ml at the total coliform concentration
is less than 100/100 ml in the water prior to the point of disinfectant
application In 90 percent of the samples taken during the six previous
months. Where monitoring for both parameters has been or 1s conducted,
the rule requires that only the fecal coliform limit be met. However, EPA
recommends that the analytical results for both total coliforms and fecal
conforms be reported. In addition, If the turbidity of a surface water
source Is greater than 5 NTU and the surface source is blended with a
ground water source to reduce the turbidity, EPA recommends that the high
turbidity water prior to blending meet the fecal coliform source water
quality criteria.
Elevated coliform levels in surface water indicate higher probabili-
ties of fecal contamination, some of which could be protected from
exposure to disinfection by embodiment in particulate matter. Blending of
-------
- Ht CUSTOMER
INDIVIDUALLY DISINFECTED
SURFACE SOURCES COMBINED
AT A SINGLE POINT
FIGURE 3*2
C
A
1 at CUSTOMER
- OlStNPSBT&NT
application
constat a TION POINT
- • •Hf.lINO POINTS
FIGURE 3-3 MULTIPLE COMBINATION POINTS
FOR INDIVIDUALLY DISINFECTED
SURFACE SOURCES
-------
the surface water with ground water to reduce coHform levels nay obscure
the Indication of such possible effects. Thus, EPA does not recomend
blending to reduce collform levels In the source water. Furthermore, EPA
does not recoonend blending to reduce turbidity levels In cases where
elevated fecal contamination may be masked.
Ongoing monitoring Is required to ensure that these requirements are
continually net. The samples may be analyzed using either the multiple
tube fermentation method or the membrane filter test (MF) as described 1n
the 16th Edition of Standard Methods.
Sampling Frequency
Minimum sampling frequencies are as follows:
Grab samples must be taken on different days. In addition, one
sample must be taken every day during which the turbidity exceeds 1 NTU,
unless the Primacy Agency determines that the system, for logistical
reasons outside the system's control, cannot have the sample analyzed
within 30 hours of collection. If taken, these samples count towards the
weekly sampling requirement. Also, under the Total Collform Rule, systems
must take one collform sample In the distribution system near the first
service connection within 24 hours after a source water turbidity
measurement exceeds 1 NTU. This measurement must be Included 1n the total
collform compliance determination. The purpose of these requirements 1s
to ensure that the monitoring occurs during worst case conditions.
The Initial evaluation of the source water quality Is based on the
data from the previous 6 months. After the Initial evaluation, systems
must continue to conduct sampling each month to demonstrate compliance
with the source water quality criteria on an ongoing basis. If the
Population Served
Collform Samples/Week
£500
501-3,300
3,301-10,000
10,001-25,000
>25,000
1
2
3
4
5
criterion has not been met, the system must filter.
Use of Historical Data Base
3-3
-------
Some systems may already monitor their source water for total and/or
fecal conform concentration. The resulting historical data base may be
sufficient for the Primacy Agency to make the initial determination of
whether the system meets the source water quality criteria. The
historical data base Is considered sufficient for making this determina-
tion if:
The raw water sampling location is upstream of the point of
disinfectant application as previously defined.
• The monthly samples represent at least the minimum sampling
frequency previously mentioned.
The sampling period covers at least the previous six months.
3.1.2 Turbidity Levels; To avoid filtration, the turbidity of the
water prior to disinfection cannot exceed 5 NTU, on an ongoing basis,
based on grab samples collected every four hours (or more frequently) that
the system 1s 1n operation. A system may substitute continuous turbidity
monitoring for grab sample monitoring 1f It validates such measurements
for accuracy with grab sample measurements on a regular basis, as
specified by the Primacy Agency.1 If a public water system uses continu-
ous monitoring, 1t must use turbidity values recorded every four hours (or
some shorter regular time Interval) to determine whether it meets the
turbidity limit for raw water. A system occasionally may exceed the 5 NTU
limit and still avoid filtration as long as (a) the Primacy Agency
determines that each event occurred because of unusual or unpredictable
circumstances and (b) as a result of this event, there have not been more
than two such events in the past twelve months the system served water to
the public or more than five such events 1n the past 120 months the system
1 Validation should be performed at least twice a week based on the
procedure outlined 1n Part 214A in the 16th Edition of Standard
Methods. Although the 17th Edition 1s available, the 16th Edition is
that which is referred to in the rule. Improper installation of
continuous monitors may allow for air bubbles to enter the monitor
resulting 1n false turbidity spikes. To avoid air bubbles reaching the
turbidimeter, the sample tap should be installed below the center line
of the pipe and an air release valve may be included on the sample
line.
-------
served Mater to the public. An "event" 1s defined as a series of
consecutive days 1n which at least one turbidity Measurement each day
exceeds 5 NTU.
It is important to note that every event. I.e., exceedance of the 5
NTU limit, regardless of whether the system must filter as a consequence,
constitutes a violation of a treatment technique requirement. For
example, 1f the turbidity exceeded 5 NTU 1n at least one measurement each
day for three consecutive days, this would constitute one event and one
treatment technique violation. If this was the third event In the past 12
months the system served water to the public, or the sixth event 1n the
past 120 months the system had served water to the public, the system
would also be required to install filtration. In all cases, the system
must inform the Primacy Agency when the turbidity exceeds 5 NTU as soon as
possible, but no later than the end of the next business day.
The Primacy Agency should evaluate additional data from the utility
to determine the significance of the event with respect to the potential
health risk to the comounity and determine whether a boll water notice 1s
necessary. The additional data may Include raw water fecal coliform
levels, duration and magnitude of the turbidity excursion, nature of the
turbidity (organic or Inorganic), disinfectant residual entering the
system during the excursion and/or conform levels 1n the distribution
system following the excursion. Boil water notices are not required under
the SWTR, they may be Issued at the discretion of the Primacy Agency.
In order to determine if the periods with turbidity greater than
5 NTU are unusual or unpredictable, It is recommended that 1n addition to
the historical turbidity data, the water purveyor should collect and
provide to the Primacy Agency current and historical Information on flows,
reservoir water levels, cllmatological conditions, and any other informa-
tion that the Primacy Agency deems relevant. The Primacy Agency will then
evaluate this Information to determine if the event was unusual or
unpredictable. Examples of unusual or unpredictable events Include
hurricanes, floods and earthquakes. High turbidity events may be avoided
by:
Use of an alternate source which 1s not a surface water and
does not have to meet the requirements of the SWTR.
3-5
-------
Use of an alternate source which 1s not a surface water and
does not have to meet the requirements of the SWTR.
Use of an alternate source which 1s a .surface water and which
does meet the requirements of the SWTR.
Utilization of stored water to supply the community until the
source water quality meets the criteria.
3.2 Disinfection Criteria
3.2.1 Inactlvation Requirements
To avoid filtration, a system must demonstrate that 1t maintains
disinfection conditions which Inactivate 99.9 percent of Siardia cysts and
99.99 percent of viruses every day of operation except any one day each
month. If the disinfection conditions provide less than these 1nact1va-
tlons during more than one day of the month, the system is in violation of
a treatment technique requirement. If the system Incurs such a violation
during any two months in the previous 12 months, the system must Install
filtration, unless one of the violations was caused by unusual and
unpredictable circumstances as determined by the Primacy Agency. Systems
with three or more violations 1n the previous 12 months oust Install
filtration regardless of the cause of the violation. To demonstrate
adequate inactivatlons, the system must monitor and record the disinfec-
tant^) used, disinfectant residual(s), disinfectant contact time(s), pH.
(for chlorine), and water temperature, and use these data to determine if
1t is meeting the minimum total inactlvation requirements in the rule.
A number of disinfectants are available, Including ozone, chlorine,
chlorine dioxide and chloramlnes. The SWTR prescribes CT [C, residual
disinfectant concentration (mg/L) x T, contact time (m1n)] levels for
these disinfectants which will achieve different levels of Inactlvation
under various conditions. The dlslnfectant(s) used to meet the Inactlva-
tion requirements 1s Identified as the primary disinfectant throughout the
remainder of this document.
To determine compliance with the inactlvation requirements, a system
must calculate the CT value(s) for Its disinfection conditions during peak
hourly flow once each day that 1t is delivering water to its customers.
For the purpose of calculating CT value, T is the time (in minutes) it
3-6
-------
takes th« water, during peak hourly flow, to aove between the point of
disinfectant application and a point where, C, residual disinfectant
concentration is measured prior to the first customer. Residual
disinfectant concentration is the concentration of the disinfectant (in
mg/L) at a point before or at the first customer. Contact tine in
pipelines must be calculated based on plug flow (i.e., where all water
moves homogeneously in tine between two points) by dividing the Internal
volume of the pipeline by the peak hourly flow rate through that pipeline.
Contact tine within mixing basins, settling basins storage reservoirs, and
any other tankage nust be determined by tracer studies or an equivalent
method as determined by the Primacy Agency. The contact time determined
from tracer studies to be used for calculating CT 1s T10. Tl0 is the
detention time corresponding to the tine for which 90 percent of the water
has been in contact with at least the residual concentration, C. Guidance
for determining contact times for basins is provided in Appendix C.
The first customer is the point at which finished water is first
consumed. In many cases this will include the treatment plant Itself.
This definition of first customer pertaining to the point of first
consumption assures that the water has received the required disinfection
to provide protection from microorganisms for all consumers. Peak hourly
flow should be considered as the greatest volume'of water passing through
the system during any one hour in a consecutive 24 hour period. Thus, it
is not meant to be the absolute peak flow occurring at any Instant during
the day.
Systems with only one point of disinfectant application may
determine the total Inactivation based on one point of residual measure-
ment prior to the first customer, or on a profile of the residual
concentration after the point of disinfectant application. Methods of
disinfection measurement are presented in Appendix 0. The residual
profile and the total inactivation Is calculated as follows:
Measure the disinfectant residual, C, at any number of points
within the treatment train.
Determine the travel time, T, between the point of disinfec-
tant application and the point where C 1s measured for the
first section. For subsequent measurements of "C," T is the
3-7
-------
time It takes for water to move froa the previous "C" measure-
ment point to this point of Measurement.
Calculate CT for each point of residual measurement (CT{t({).
Determine the 1nact1vat1on ratio (CTtJ)t/CT„ ,) for each sec-
tion.
Sum the Inactlvatlon ratios for each section, I.e. C.T./CT.,,
~ C.T2/CT„ 9 + CJ#/CT„ , to determine the total Inactivat'on
ratio.
If the total inactlvatlon ratio (sum (01,,,,/CT,,,)) Is equal to or greater
than 1.0, the system provides greater than 99.9 percent Inactlvatlon of
Giardia cysts), and the system meets the disinfection performance re-
quirement. Further explanation of CT calculations 1s presented 1n Section
3.2.2.
Systems need only calculate one CT (CTtil{) each day, for a point at
or prior to the first customer; alternatively they have the option of cal-
culating numerous CTs after the point of disinfectant application but
prior to the first customer to determine the Inactlvatlon ratio. Profil-
ing the residual gives credit for the higher residuals which exist after
the disinfectant Is applied but before the first customer. Profiling the
residual may not be necessary If one CT 1s calculated (CTcllc), and this
exceeds the applicable CT„ ,. In this case, the system Is meeting the
disinfection performance requirement. For systems with a very low oxidant
demand In the water and long contact times, this approach may be the most
practical to use.
For systems with multiple points of disinfectant application, such
as ozone followed by chlorine, or chlorine applied at two different points
In the treatment train, the Inactlvatlon ratio of each disinfectant
section prior to the first customer Is used to determine the .total
Inactlvatlon- ratio. The disinfectant residual of each disinfection
2 CT99. 1s the CT value required to achieve 99.9 percent or 3-log Siardifl
cyst inactlvatlon for the conditions of pH, temperature and residual
concentration for each section. A section 1s the portion of the system
with a measurable contact time between two points of disinfection
application or residual monitoring.
3-8
-------
section «nd the corresponding contact time oust be measured at some point
prior to the subsequent disinfection application point(s) to determine the
inactivation ratio for each section, and whether the total Inactlvation
ratio is 1.0 or more. For example, if the first disinfection section
provided an inactivation ratio of 2/3 (or 99 percent Inactivation) and the
second disinfection section provided an inactivation ratio of 1/3 (or 90
percent inactivation), the total Inactivation ratio would equal 1.0 (2/3
4- 1/3 ¦ 1) Indicating that 99.9% Inactivation was provided and the
disinfection requirements are net. Further explanation of the-deternina-
tion of total inactivation provided Is contained in Section 3.2.2.
Hitntfllnfnfl Inactlration level
The SHTR establishes CTs for chlorine, chlorine dioxide, ozone and
chloramines which will achieve 3-log inactivations of Siardia cysts and at
least 4-log inactivation of viruses. Appendix E presents CTs for these
and other log Inactivations. A system must demonstrate compliance with
the Inactivation requirements based on conditions occurring during peak
hourly flow. Since a system generally can only Identify peak hourly flow
after it has occurred, hourly residual measurements during the day are
suggested. If the sampling points are remote, or manpower is limited and
collection of hourly grab samples is impractical, continuous monitors may
be Installed. In cases where continuous monitors are impractical, the
Primacy Agency may establish an acceptable monitoring program on a
case-by-case basis,* where possible this should be based on historical flow
patterns. Measurements for. the hour of peak flow can then be used In
calculating CT. The pH (for systems using chlorine) and temperature must
be determined dally for each disinfection sequence prior to the first
customer.
Since the system's Inactivation is determined during peak hourly
flow, the disinfectant dosage applied to meet CT requirements may not be
necessary during lower flow conditions. Continuing to apply a disinfec-
tant dosage based on the peak hourly flow could possibly result in
increased levels of disinfectant by-products, including TTHMs and
increased costs. Under lower flow conditions, a higher contact time is
available and a lower residual may provide the CT needed to meet the
Inactivation requirements. The system may therefore choose to adjust the
3-9
-------
disinfectant dost with changes In flow. The system should, however,
maintain a disinfectant residual which will still provide a 3-log
1nact1vat1on of Siardia cysts and a 4-log inactivation of viruses at
non-peak hourly flows. The system should therefore evaluate the residual
needed to provide the required Inactivation under different flow
conditions and set the dosage accordingly. The following provides an
example of maintaining the required inactivation.
Example
A 5 agd non-filtering system disinfecting with free chlorine at one
point of application, has a contact time of 165 minutes during'a peak flow
of 5 MGO. The flow varies from 1 to 5 MGO. The pH and temperatures of
the water are 7 and 5 C, respectively. At a residual of 0.9 mg/L, a CT of
148 mg/L-min is required to meet the disinfection requirements. The CT
for 0.9 mg/L residual Is determined by straight line interpolation between
. 0.8 mg/L and 1.0 mg/l residuals. Under lower flow conditions, the
available contact time is longer and a lower residual would provide the
required disinfection. Based on existing contact time and using the
appropriate CT tables (in this case, Table E-2) in Appendix E for a 3-log
siardia cyst inactivation, the required disinfection would be provided by
maintaining the following chlorine residuals for the indicated flow:
Contact CT (mg/L-min) Free Chlorine
now tMGQ^ Iliw (min) Rwirtti— Rcsldml f,wg/U
5 161 148 0.9
4 206 145 0.?
3 275 143 0.6
2 412 139 0.4
1 825 139 0,2
This table Indicates the variation of residuals needed for the
system to provide the required inactivation. For chlorine, the disinfec-
tant residual cannot be adjusted in direct proportion to the flow because
the CT needed for disinfection is dependent upon the residual. Since it
is not practical to continuously adjust the residual and, since a
disinfection level for a 3-log Siardia cyst inactivation must be
maintained under all flow conditions, it 1s suggested that the flow
variation at the utility be divided into ranges and the residual needed at
3-10
-------
the higher flow rate of each range be Maintained for all flows within the
range to ensure the required disinfection. The following flow ranges and
residuals are suggested for the system:
Free Chlorine
Flow Ranoe fMGDl Residual fmo/n
1-1.9 0.4
2 - 3.9 0.6
4-5 0.9
By maintaining these residuals, the utility 1s ensuring the provision of
the required disinfection while minimizing the disinfectant application,
which should result in lower disinfection by-products and costs.
Although these residuals will meet the inactivatlon requirements,
maintaining a residual 1n the distribution system must also be considered.
If no other point of disinfection exists prior to the distribution system,
the residual for disinfection must be maintained at a level which will
also provide a residual throughout the distribution system. The complete
range of flows occurring at the plant should be evaluated for determining
the required residual. A utility may establish the residual requirements
for as many flow ranges as 1s practical.
The CTs determined from the daily system data should be compared to
the values 1n the table for the pH and temperature of the water, to
determine if the required CT has been achieved. Only the analytical
methods prescribed in the SWTR, or otherwise approved by EPA, may be used
for measuring disinfectant residuals. Methods prescribed in the SWTR are
listed 1n Appendix 0. The Appendix also contains a paper which describes
monitoring methods for various disinfectants and conditions.
The Primacy Agency should make periodic checks on its utilities to
assure that they are maintaining adequate disinfection at non-peak flow
conditions.
Meeting the Inactivation Requirement Usino Free Chlorine
When free chlorine Is used as a disinfectant, the efficiency of
Inactivation Is influenced by.the temperature and pH of the water. Thus,
the measurement of the temperature and pH for the determination of the CT
is required. The SWTR provides the CT requirements for free chlorine at
3-11
-------
various temperatures and pHs which may occur In a source water. These
values are presented In Table E-l through Table E-7 In Appendix E. The
basis for these values Is discussed In Appendix F. For free chlorine, a
3-log 1nact1vat1on of Giardia cysts will provide greater than a 4-log
1nact1vat1on of viruses, thus meeting the SWTR Inactlvatlon requirements.
As Indicated 1n Table E-2, a raw water temperature of 5 C, a pH of
7.0, and a residual chlorine concentration of 1.4 mg/L require a CT of 155
mg/L-a1n to provide a 3-log inactlvatlon of Giardia cysts. Therefore, to
meet the Inactlvatlon requirement under these conditions with one point of
residual measurement, a contact time of 111 minutes [(155 mg/L-m1n)/ (1.4
mg/L)] prior to the first customer would be required.
Meeting the Inactivation Requirement Using Ch1oram1n«
Chloramlnes are a much weaker oxidant than free chlorine, chlorine
dioxide and ozone. The CT values for chloramlnes presented 1n Table E-12
are based on disinfection studies using preformed chloramlnes and In vitro
excystatlon of Giardia muris cysts (Rubin, 1988). No safety factor was
applied to the laboratory data on which the CT values were based since EPA
believes that chloramlnation, conducted in the field, 1s more effective
than using preformed chloramlnes.
In the laboratory testing using preformed chloramlnes, ammonia and
chlorine were reacted to form chloramlnes before the addition of the
microorganisms. Under field conditions, chlorine 1s usually added first
followed by amsonla addition further downstream. Also, even after the
addition of ammonia, some free chlorine residual may persist for a period
of time. Therefore, free chlorine is present for a period of time prior
to the formation of chloramlnes. Since this free chlorine contact time 1s
not duplicated 1n the laboratory when testing with preformed chloramlnes,
the CT values obtained by such tests may provide conservative values when
compared to those CTs actually obtained in the field with chlorine applied
before ammonia. Also, other factors such as mixing 1n the field (versus
no mixing 1n the laboratory) may contribute to disinfection effectiveness.
For these reasons, systems using chloramlnes for disinfection may
demonstrate effective disinfection 1n accordance with the procedure 1n
Appendix G in lieu of meeting the CT values 1n Appendix E.
3-12
-------
If a system uses chloramlnes and Is Able to achieve the CT values
for 99.9 percent Inactlvatlon of filardla cysts, It Is not always
appropriate to assume that 99.99 percent or greater Inactlvatlon of
viruses was also achieved, New data Indicate that Hepatitis A virus Is
more sensitive than Giardla cysts to Inactlvatlon by preformed chloranlnes
(Sobsey, 1988). The CT values required to achieve 99.99 percent
Inactlvatlon of Hepatitis A with preformed chloranlnes are lower than
those needed to achieve 99.9 percent Inactlvatlon of filardia cysts. These
data contrast with other data which Indicate that rotavirus . Is more
resistant than G1ard1a cysts to preformed chloranlnes (Hoff. 1986).1
However, rotavirus is very sensitive to Inactlvatlon by free chlorine,
much more so than Hepatitis A (Hoff, 1986;4 Sobsey, 1988). If chlorine
1s applied prior to amnonla, the short term presence of free chlorine
would be expected to provide at least 99.99 percent Inactlvatlon of
rotavirus prior to the addition of ammonia and subsequent formation of
chloramlnes. Thus, EPA believes 1t Is appropriate to use Hepatitis A
data, In lieu of rotavirus data, as a surrogate for defining minimum CT
values for Inactlvatlon of viruses by chloramlnes, under the condition
that chlorine Is added to the water prior to the addition of amnonla.
A system which achieves a 99.9 percent or greater Inactlvatlon of
61ard1a cysts with chloramlnes can be considered to achieve at least 99.99
percent Inactlvatlon of viruses, provided that chlorine Is added to the
water prior to the addition of ammonia, Table E-13 provides CT values for
achieving different levels of virus Inactlvatlon. However, If anmonla Is
added first, the CT values 1n the SWTR for achieving 99.9 percent
inactlvatlon of Giardia cysts cannot be considered adequate for achieving
99.99 percent inactlvatlon of viruses.
Under such cases of chloramine production, the SWTR requires systems
to demonstrate through on-site challenge studies, that the system Is
1 CT values 1n excess of 5,000 are required for a 4-log Inactlvatlon of
rotavirus by preformed chloramlnes but no minimum CT values have been
determined.
4 CT values ranging from 0.025 to 2.2 achieve 99 percent inactlvation of
rotavirus by free chlorine at pH ¦ 6 -10 and 4 - 5BC (Hoff, 1986).
3-13
-------
achieving at least a 4-log 1nact1vat1on of viruses. Guidance for
conducting such studies Is given In Appendix G. Once conditions for
achieving a 4-log 1nact1vat1on of viruses has been established, the
Primacy Agency should require systems to report their disinfection
operating conditions on an ongoing basis. These conditions should verify
that the system 1s operating at CT values 1n excess of that needed to
achieve a 4-log virus Inactivation or 3-log Giardia cyst 1nact1vat1on,
whichever 1s higher.
Meeting the Inactivation Requirement Using Chlorine Dioritte
Under the SWTR, the CT values for the inactivation of Giardia cysts
using chlorine dioxide are independent of pH. Under the SWTR the only
parameter affecting the CT requirements associated with the use of
chlorine dioxide 1s temperature. Table E-8 in Appendix E presents the
chlorine dioxide CT values required for the Inactivation of Giardia cysts
at different temperatures. The basis for these CT values is discussed 1n
Appendix F. Systems which use chlorine dioxide are not required to
measure the pH of the disinfected water for the calculation of CT. For
chlorine dioxide, a 3-log Inactivation of Giardia cysts•will generally
result 1n greater than a 4-log virus Inactivation, and assure meeting the
SWTR Inactivation requirements. However, for chlorine dioxide, unlike
chlorine where this relationship always holds true, at certain tempera-
tures, the 4-log virus CTs may be higher than the 3-log Giardia cyst CTs.
The Primacy Agency may allow lower CT values than those specified 1n
the SWTR for individual systems based on information provided by the
system. Protocols for demonstrating effective disinfection at lower CT
values Is provided In Appendix G.
As Indicated In tables E-8 and E-9, the CT requirements for chlorine
dioxide are substantially lower than those required for free chlorine.
However, chlorine dioxide is not as stable as free chlorine or chloramines
In a water system and may not be capable of providing the required
disinfectant residual throughout the distribution system. In addition,
out of concern for toxicologlcal effects, EPA's current guideline 1s that
the sum of the chlorine dioxide, chlorate and chlorite residuals, be less
than 1.0 mg/L at all consumer taps. This guideline may be lowered as more
health effects data become available. These concerns further reduce the
3-14
-------
feasibility of using chlorine dioxide as a secondary disinfectant for
distribution systems. Therefore, the use of chlorine dioxide as a primary
disinfectant may result In the need for the application of a secondary
disinfectant, such as chlorine or chloraalnes, that *111 persist In the
distribution system and provide the required residual protection.
Iteetlnn the Inactlvatlon Requirement Using Oione
Another disinfectant to Inactivate Giardia cysts and viruses 1s
ozone. As with chlorine dioxide, under the SWTR, the CT values for ozone
are Independent of pH. Tables £-10 and E-ll present the CT requirements
for ozone at different source water temperatures. The basis for the CT
values for ozone Is given In Appendix F. As for free chlorine, a 3-log
Giardia cyst Inactlvatlon with ozone will result In greater than a 4-log
virus Inactlvatlon. Unlike chlorine, for cases where only a 1-log or
lower Giardia Inactlvatlon Is needed with ozone, the CT values for virus
Inactlvatlon may be higher than the CT for 61ard1a. The Primacy Agency
¦ay allow lower CT values for Individual systeas based on information
provided by the system that demonstrates that CT values lower than those
specified In the rule achieve the same Inactlvatlon efficiencies (see
Appendix 6).
Ozone Is extremely reactive and dissipates quickly after applica-
tion. Therefore, a residual5 can only be expected to persist a short time
5 The residual must be measured using the Indlao .Trlsulfonate Method
(Bader I Holgne, 1981) or automated methods which are calibrated 1n
reference to the results obtained by the Indigo Trlsulfonate method, on
a regular basis as determined by the Primacy Agency. The Indigo
Trlsulfonate method 1s included In the 17th Edition of Standard
Methods. This method Is preferable to current standard methods because
of the selectivity of the Indigo Trlsulfonate Indicaor In the presence
of most Interferences found 1n ozonated waters. The ozone degrades an
acidic solution of Indigo trlsulfonate In a 1:1 proportion. The
decrease In absorbance Is linear with Increasing ozone concentrations
over a wide range. Mai on 1c acid can be added to block Interference
from chlorine. Interference from permanganate, produced by the
ozonation of manganese, 1s corrected by running a blank 1n which ozone
1s destroyed prior to addition of the indigo reagent. The samples can
be analyzed using a spectrophotometer at a 600 nm wavelength which can
detect residuals as low as 2 ug/L or a visual color comparison method
which can measure down to 10 ug/L ozone. Although currently available
monitoring probes do not use the Indigo Trlsulfonate Method, they can
3-15
-------
after application. In addition, the application of ozone to water Is
dependant on mass transfer. For these reasons, the method of CT
determination used for the other disinfectants 1s Impractical for ozone*
The CTtllc Bust be determined for the ozone contactor alone. The contactor
will have some portions where the ozone Is applied and other portions of
the contactor where ozone Is no longer applied, which are referred to as
the reactive flow chanters.
For many ozone contactors, the residual 1n the contactor will vary
1n accordance with the method and rate of application, the residual will
be nonuniform and Is likely to be zero In a portion of the contactor. As
previously Indicated, the CT value Is based on the presence of a known
residual during a specific contact time. Thus disinfection credit Is only
provided for the time when a residual Is present. Besides the nonuniform-
Ity of the residual, monitoring the residual will be difficult because of
the ozone's high reactivity and the closed design of the contactors.
In addition to the difficulty In determining the ozone residual for
the CT calculation, the contact time will vary between basins depending
on their flow configuration. Several types of devices are available for
adding ozone to water Including porous diffusers, submerged turbines,
Injector, packed towers and static mixers. Each type of device can be
used in either single or multiple chamber contactors. The flow through a
single chamber turbine contactor will approximate a completely mixed unit,
while flow through a single chamber diffused contactor, or a multiple
chamber diffused contactor, will more closely represent plug flow. This
variation In flow In different contactors makes the use of Tl# inappropri-
ate for some contactors.
The differences between ozone contactors and other disinfection
systems resulted In the development of several approaches for determining
the inactlvatlon provided by ozone, Including:
Evaluation of C and T
Segregated Flow Analysis (SFA)
- Continuously Stirred Tank Reactor (CSTR)
Site Specific Evaluation
be calibrated via this method.
3-16
-------
The method which Is appropriate for a parlcular system will depend on
system configuration and the required level of Inactlvatlon. Another
significant difference 1s that ozone nay be applied to provide only a
portion of the overall 3-log Giardia cyst and 4-log virus Inactlvatlon"
with the remainder of the Inactivatlon provided by another disinfectant.
Appendix 0 provides details for selecting the appropriate method of
evaluation for specific cpnditlons.
The evaluation of C and T involves separate determination of the
ozone residual concentration, C, and the contact time, T, In the
contactor. C can be determined for individual chambers of -a' contactor
based on the residual measured at several points throughout the chamber,
or at the exit of the chamber. The T value can be determined through a
tracer study or an equivalent method as approved by the Primacy Agency
with air or oxygen applied during testing, using the same feed gas rate as
used during operation. Appendix 0 provides details for the CT approach.
SFA Is based on the results of a tracer study used in conjunction
with the measured ozone residual to determine the survival of microorgan-
isms exiting the contactor. The survival corresponds to a certain
inactlvatlon. Guidelines for this approach are included in Appendix 0.
The CSTR approach Is applicable for contactors which have a high
degree of mixing. Experience has shown that for contactors such as
turbine units, the ozone residual is generally uniform throughout the
contactor. The ozone residual measured at the exit 1f the contactor is
used in an equation for CSTRs to determine the inactlvatlon provided.
Appendix 0 provides details for conducting CSTR analysis.
Site specific evaluations may include:
Measurement of an observable parameter to correlate with C
Mathematical model for disinfection efficiency
Microbial Indicator studies for disinfection efficiency
to more closely determine the inactlvatlon provided in a particular
system. Appendix 0 provides details for applying site specific evalua-
tions.
3-17
-------
Suaaarv
Hany systeas which do not provide filtration will have difficulty In
. providing the contact tlae necessary to satisfy the inactlvatlon
requlresents prior to the first custoiMr. For exaaple, a systea using
free chlorine at a Mater temperature of 5 C, a pH of 7.0 and a chlorine
residual of 1.4 ag/L would require 111 alnutes of contact tlae to Beet the
Inactlvatlon requirement. Potential options for these systeas Include:
Installation of storage facilities to provide the required
contact tlae under aaxlaua flow conditions.
- Use of an alternate prlaary disinfectant such as ozone or
chlorine dioxide which has Cf values lower than those required
for free chlorine for the required Inactlvatlon.
For soae systeas, the difficulty 1n obtaining the required
Inactlvatlon aay only be a seasonal problea. A systea that has raw water
temperatures which reach 20 C during the suaner aonths at a pH of 7.0, aay
have sufficient contact tlae to aeet the CT of 56 ag/L-aln (Table E-5) at
a chlorine concentration of 1 ag/L. However, assuring the saae pH and
chlorine concentration, It aay not have sufficient contact tlae to aeet
the CT requlrenent at 5 C, 149 ag/L-a1n (Table E-2), or at 0.5 C,
210 ag/L-aln (Table E-l). Under those conditions, a systea could choose
to use ozone or chlorine dioxide on a seasonal basis, since they are
stronger disinfectants requiring a shorter contact tlae.
As Indicated 1n Table E-12, the CT values for chloraaines aay be
lapractlcal to attain for aost systeas. Systeas which currently utilize
chloraaines as a prlaary disinfectant aay need to use either free chlor-
ine, chlorine dioxide or ozone 1n order to provide the required disin-
fection. However, systeas using chloraaines as a prlaary disinfectant aay
chose to demonstrate the adequacy of the disinfection. Appendix G
presents • method for Baking this demonstration.
M-tine the inactlvatlon Requirement Uslno Alternate Disinfectants
For systems using disinfectants other than chlorine, chloraaines,
chlorine dioxide, or ozone, the effectiveness of the disinfectant can be
demonstrated using the protocol contained In Appendix G. The protocol in
Appendix G.3 for batch testing should be followed for any disinfectant
3-18
-------
which can be prepared in an aqueous solution and trill be stable throughout
the testing. For disinfectants which are not stable, the pilot study
protocol outlined in Appendix G.4 should be followed.
3.2.2 Determination of Overall Inactivation for Residual Profile,
Multiple Disinfectants and MultlaU Sn»rr»<
For systems which apply dislnfectant(s) at more than one point, or
choose to profile the residual from one point of application, the total
inactivation is the sun of the inactivation ratios between each of the
points of disinfection or between each of the residual aonitorlng points,
respectively. The portion of the system with a measurable contact time
between two points of disinfection application or residual aonitorlng will
be referred to as a section. The calculated CT (CTtllt) for each section
1s determined dally.
. The CT needed to fulfill the disinfection requirements is CTlf,,
corresponding to a 3-log inactivation of Siardia cysts and greater than or
equal to a 4-log inactivation of viruses (except for chloramlnes and
sometimes chlorine dioxide as explained In Section 3.2.1). The Inactiva-
tion ratio for each section Is represented by CTtI(c/CTf# #> as explained in
Section 3.2.1, and Indicates the portion of the required inactivation
provided by the section. The sum of the inactivation ratios from each
section can be used to determine the overall level of disinfection
provided. Assuming inactivation 1s a first order reaction, the inac-
tivation ratio corresponds to log and percent inactlvations as follows;
£Ic,u£It,.« Log Inactivation Ptrccnt Inactivation
0.17 - 0.5 log 68 %
0.33 - I log 90%
0.50 - 1.5 log » 96.8%
0.67 ¦ 2 log 99%
0.83 - 2.5 log « 99.7%
1.00 - 3 log 99.9%
1.33 - 4 log ¦ 99.99%
3-19
-------
CTn , «an be determined for each section by referring to Tables I-J
through E-13 in Appendix E, using the pH (when chlorine is the disinfec-
tant) and temperatures of the water for the respective sections. These
tables present the log inactivation of Siardia cysts and viruses achieved
by CTs at various water temperatures and pHs.
Log inactivations are additive, so:
O.S Log + 1.0 Log ¦ 1.5 Log or
0.17CT,,, ~ 0.33CT„, % 0.5CTff f
If the sua of the Inactivation ratios is greater than or equal to
one, the required 3-1og inactivation of fiiardia cysts has been achieved.
An inactivation ratio of at least 1.0 is needed to demonstrate compliance
with the fiiardia cyst Inactivation requirements for unfiltered systems.
The total log inactivation can be determined by multiplying the sun
of the inactivation ratios (sum (CTl4ll/CTlf,)), by three. The total log
Inactivation can be determined in this way because CT„ , 1s equivalent to
a 3-log inactivation. The total percent inactivation can be determined as
follows:
y ¦ JM - Ififi Equation ' (1)
10*
where: y •% inactivation
x ¦ log inactivation
For example:
x - 3.0 log Inactivation
y - 100 - IQfw * 99.9 % inactivation
As explained In Section 3.2.1, the CTMlt determined for each disin-
fection section 1s the product of the disinfectant residual in mg/L and
the detention time 1n minutes through the section at peak hourly flow.
However, for many water systems, peak hourly flow will not necessarily
occur simultaneously In all sections. The extent to which the occurrence
of peak hourly flow will vary between sections of the system depends on
3-20
r
-------
the characteristics of an Individual system Including Its size, storage
capacity within the distribution system, the number of sources, and
hydraulic capacities between different sections. In order to simplify the'
determination of peak hourly flow for the system, 1t should be taken as
peak hourly flow 1n the last section of the system prior to the first
customer.
The CT values for all the sections should be calculated for the flow
and the residuals occurring during the hour of peak flow in the last
section. The most accurate way to determine the flow in a particular
section 1s through the use of a flow meter. However, some sections of the
system may not have a flow meter. The following guidelines can be used to
determine the flow to be used In calculating CT:
For sections which do not have meters, the flow should be
assumed to be the higher of the two flows occurring 1n the
closest upstream and downstream sections with meters.
In cases where a section contains a pipeline and a basin with
the flow meter located prior to the basin, the metered flow
does not represent the discharge rate of the basin. The
difference in Inlet and discharge rates from a basin will
impact the water level in the basin. As explained 1n Appendix
C, falling water levels will result 1n lower T,# values.
To assure that the detention time of a basin is not
overestimated, the discharge flow from a basin should be
used 1n lieu of the Influent flow, unless the Influent,
flow 1s higher.
To estimate the discharge flow from a basin the closest
flow meter downstream of the basin should be used.
The following example presents the determination of the total
percent Inactivation for multiple points of disinfection, with variation
in flow between sections.
fiffflnlf
A coMunity of 6,000 people obtains Its water supply from a lake
which Is 10 miles from the city limits. Two 0.2 MG storage tanks are
located along the 12-inch transmission line to the dty. The water is
disinfected with chlorine dioxide at the exit from the lake and with
chlorine at the discharge from the first and second storage tanks. The
3-21
-------
average water demand of the coominlty 1s 1 HGO with a peak hourly demand
of approximately 2 M60. For the calculations of the overall percent
1nact1vat1on, the supply system 1s divided Into three sections as shown on
Figure 3-1..
'Section 1 - from the lake to the discharge from the first storage
tank,
Section 2 - from the discharge from the first storage tank to the .
discharge from the second tank
Section 3 - from the discharge of the second storage tank to the
first customer
The overall 1nact1vat1on 1s computed dally for the peak hourly flow condi-
tions. Sections 1 and 3 contain flow meters to monitor the water being
withdrawn from the lake and the water being delivered to the distribution
system as shown on Figure 3-1. On the day of this example calculation,
the peak hourly flow In section 3 was 2 HGO. During this hour, water was
being withdrawn from the lake at a rate of 1.5 mgd. Considering the
placement of flow meters, the flow of 2 mgd measured 1n section 3 should
be used for calculating CT for that section. Since section 2 does not
have a flow meter, the meter In section 3 serves as a measure of the
discharge from storage tank 2 and should be the flow used in the
calculation of CT for section 2. The flow meter 1n section 1 records the
flow through the transmission main which should be used 1n the calculation
of CT for the pipeline. However, this meter does not represent the
discharge from storage tank 1. Since the water 1s being pumped to the
distribution system at a higher rate than the flow entering storage tank
1, the flow of 2 mgd measured in section 3 should be used for calculating
the CT for storage tank 1.
The pH, temperature and disinfectant residual of the water were
measured at the end of each section just prior to the next point of
disinfection and the first customer during the hour of peak flow. The
water travels through the 12-inch transmission main at 177 ft/nin at
3-22
-------
1.5 MGD.1 The detention times of the storage tanks were read off the T.B
vs. Q plots generated from tracer studies conducted on the storage tanks
(see Appendix C). The data for the inactivation calculation are as
follows.*
length of pipe (ft)
flow (mgd)
pipe
tank
contact time (nin)
pipe
tank
total
disinfectant
residual (mg/L
temperature (C
pH
Section 1
15,840
1.5
2.0
89
116
205
chlorine
dioxide
0.1
5
8
Section 2
26,400
2.0
2.0
111
114
225
chlorine
0.2
5
8
Section 3
10,560
2.0
45
0
45
chlorine
0.4
5
8
This information is then used in conjunction with the CT„ , values in
Appendix I to determine the (CTMU/CTH ,) in each section as follows:
Section 1 - Chlorine dioxide
CTcllc - 0.1 ng/L x 105 minutes • 20.5 ng/L-nin
From Table E-8 at a temperature of 5 C and pH • 8,
-
, is 26 mg/L-min
CTftU/CT,
HI
26 mg/L-mln
0.7*
SttCtlQn 2 - Chlorine
CT
call
> 0.2 mg/L x 225 minutes ¦ 45 mg/L-min
From Table 1-2 at a temperature of 5 C and pH • 8,
CT,, , Is 198 mg/L-min
CTmU/CTh.« * IS W/Lrlln ¦ 0-23
198 mg/L-min
fl - 1.5 X 10' qalrtav X 1 ft' t X day, - 177 ft/«in
A (1 ft7 /4) 7.48 gal 1440 min
3-23
-------
5*gt1on 3 - Chlorine
CTtn« % 0,4 "9/L-«fn x 45 ain ¦ 18 ag/L-a1n
Froa Table E-2 at a teaperature of 5 C and pH ¦ 8,
CTf| | Is 198 ag/L-ain
CTe,,(/CT|j } ¦ 18 wo/L-ain ¦ 0>09
198 ag/L-a1n
The sua of CTtIlt/CT„, 1s equal to 1.11, Mhfch 1s greater than 1,
therefore, the systea aeets the requireaents of providing a 3-log
1nact1vat1on of 61ard1a cysts. The log 1nact1vat1on provided 1s:
* % 3 x ' £ICI„ - 3 x 1.11 . 3.33
CT
Uliu
The percent 1nact1vat1on can be determined using equation 1.
y ¦ 100 - lflO ¦ 100 - Iflfl ¦ 100 - 0.05 ¦ 99.95% 1nact1vat1on
10* " 2,138
The system oeets the requirement of providing a 99.9 percent Inactlvation
of 61ard1a cysts.
The SHTR also requires that the public be provided with protection
froa Legionella as well as filardla cysts and viruses. Inactlvation levels
have not been set for Legionella because the required Inactlvation of
filardla cysts will provide protection froa Legionella.7 However, this
level of disinfection cannot assure that all Legionella will be Inacti-
vated and that no recontaaination or regrowth In recirculating hot water
systeas of buildings or cooling systeas will occur. Appendix B provides
Kuchta et al. (1983)-reported a aaxiaua CT requlreaent of 22.5 for a
99 percent Inactlvation of Legionella In • 21 C tap water at.a pH of
7.6-6.0 when usino free chlorine.Using first order kinetics, a 99.9
percent inactlvation requires a CT of 33.8. Table A-5 presents the CTs
needed for free chlorine to achieve a 99.9 percent Inactlvation of
Slardia cysts at 20 C. This table Indicates that the CT required for
a 3-log inactlvation of 61ardia at the teaperature and pH of the
Legionella test ranges froa 67 to 108 depending on chlorine residual.
These CT's are two to three tlaes higher than that which is needed to
achieve a 3 log inactlvation of Legionella.
3-24
-------
guidance for monitoring and treatment to control u^ianglla in institu-
tional systems.
The above discussion pertains to a system with one source with-
sequential disinfection. Another system say blend More than one source,
and disinfect one or sore of the sources independently prior to blending.
System conditions which nay exist include:
- All the sources are combined at one point prior to supplying
the coamunity but one or sore of the sources are disinfected
prior to being combined, as shown on Figure 3-2.
- Each source Is disinfected individually and enters the
distribution system at a different point, as shown on Fig-
ure 3-3.
For all systems combining sources, the first step in determining the
CT should be to determine the CTtllt provided from the point of blending
closest to the first customer using the contact time and reslduaT at peak
hourly flow for that portion of the distribution system. This corresponds
to section 0 on Figure 3-2 and section E on Figure 3-3. If the CTlllt for
section 0 or I provides the required inactlvation, no additional CT credit
is needed and no further evaluation is required. However, if the CT for
section D or E 1s not sufficient to achieve the required inactlvation,
then the inactlvation ratio (CTclle)/(CT,# ,) should be determined for each
section to determine the overall Inactlvation provided for each source.
The total inactlvation must be greater than or equal to one for all
sources in order to comply with the requirements for 3-log inactlvation of
SlittLii cysts.
On Figure 3-2, sections A, B, C and 0 contain sampling points a, b,
c and d, respectively. The sum of the Inactlvation ratios for sections
A+D, B+0 and C+0 must each be greater than or equal to one for the
disinfection requirements to be met.
The total Inactlvation for each source on Figure 3-2 should be
determined as follows:
Source I
Oetermine CT(|I for sections A and 0 based on the residual
measurements at sample points a and d, and the travel time
3-25
-------
through each section under peak hourly flow conditions for the
.respective section.
Determine CT„ . for the pH and temperature conditions In eacK
section using the tables in Appendix E
Calculate the inactlvatlon ratios (CT(ilt/CT„ ,) for sections
A and D.
Calculate the sun of the inactivation ratios for sections A.
and 0 to determine the total inactlvatlon for source I.
If the sua of the inactlvatlon ratios is greater than or equal
to 1.0, the system has provided the required 3-1 og fiiardia
cyst inactlvatlon.
Determine CTt|lt for section I based on the residual measured
at sample point b and the travel time through the section
under peak hourly flow conditions.
Determine CT„. for section 8 for the pH and temperature
conditions in the section using the appropriate tables 1ft
Appendix I.
Calculate the inactlvatlon ratio (CT^/CT,,,) for section B.
Add the inactlvatlon ratios for sections B and D to determine
the total inactlvatlon for source II.
If the sum of the inactlvatlon ratios is greater than or equal
to 1.0, the system has provided the required 3-log Giardia
cyst inactlvatlon for the source.
Determine CT(ll. for section C based on the residual measured
at sample point c and the travel time through the section
under peak hourly flow conditions.
Determine CT„. for section C for the pH and tepperature
conditions in the section using the appropriate tables in
Appendix £.
Calculate the inactlvatlon ratio (CT«4lt/CT„ ,) for section C.
Add the inactlvatlon ratios for sections C and D to determine
the total inactlvatlon for Source III.
3-26
-------
If the sum of the inaetlvatlon ratios Is greater than or equal
to 1.0, the system has provided the required 3-log Giardia
cyst Inaetlvatlon for the source.
The determination of the total Inaetlvatlon for each source nay
require more calculations for systems such as that on Figure 3-3 than on
Figure 3-2. On Figure 3-3 sections A, B, C, 0, and E contain sampling
points a, b, c, d, and e respectively. In order to minimize the
calculations needed, the determination of the total Inaetlvatlon should
begin with the source closest to the first customer.
The total Inaetlvatlon for each source on Figure 3-3 should be
determined as follows:
Source III
Determine CT ,{ for sections C and E based on the residual
measurement at sample points c and e and the detention time In
each section under peak hourly flow conditions for the
respective section.
Determine CT99 . for the pH and temperature conditions in each
section using the tables in Appendix E.
Calculate the inaetlvatlon ratios (CTtllt/CTfl,) for sections
C and E.
Calculate the sum of the Inactivatfon ratios for sections C
and E to determine the total inaetlvatlon for source III.
If the sum of the inaetlvatlon ratios is greater than or equal
to 1.0, the system has provided the required 3-1og Giardia
cyst inaetlvatlon for source III.
Scarce II
Determine CT , for section D based on the residual measured
at sample point d and the detention time through the section
under peak hourly flow conditions.
Determine CTI9. for section D for the pH and temperature
conditions in the section using the appropriate tables 1n
Appendix E.
- Calculate the inaetlvatlon ratio (CTtllt/CT„ ,) for section D.
Add the inaetlvatlon ratios for sections D and E to determine
the overall inaetlvatlon.
3-27
-------
- 'If the sun of the inactlvatlon ratios 1s greater than or equal
I?..1'?:.'.I*I,M ,"'ov1,l3*d 3-log Et.rdl.
cyst Inactlvatlon for source II, as well as source I since the
water from each of these sources are combined prior to
sections 0 and E.
If the total Inactlvatlon ratio for sections 0 and E 1s less
than 1.0, additional calculations are needed. Proceed as
follows for source II.
Determine CT , for section B based on the residual measured
at sample point b and the detention time through the section
under peak hourly flow conditions.
Determine CT„. for section B for the pH and temperature
conditions In the section using the appropriate tables in
Appendix E.
Calculate the Inactlvatlon ratio (CTt|U/CTl# ,) for section B.
Add the Inactlvatlon ratios for sections B, D and E to
determine the total Inactlvatlon for source II.
If the sum of the Inactlvatlon ratios 1s greater than or equal
to 1.0, the system has provided the required 3-log GlardU
cyst Inactlvatlon for the source.
Source I
As noted In the determination of the Inactlvatlon provided for
source II, if tht sum of the Inactlvatlon ratios for sections D and E Is
greater than or equal to 1.0, the system has provided the required 3-log
Gfardla cyst Inactlvatlon. However, If this sum Is less than 1.0
additional calculations will be needed to determine the overall Inactlva-
tlon provided for source I. The calculations are as follows:
Source I
Determine CT , for section A based on the residual measured
at sample point a and the detention time In the section under
peak hourly flow conditions.
- Determine CT.ri for section A for the pH and temperature
conditions In'the section using the appropriate tables In
Appendix E.
- Calculate the Inactlvatlon ratio (CT„lt/CT„ ,) for section A.
- Add the Inactlvatlon ratios for sections A, D, and E to
determine the total Inactlvatlon for source I.
3-28
-------
If the sua of the inactivation ratios is greater than or equal
to 1.0, the systea has provided the required 3-1og Siardla
cyst Inactivatlon for the source.
3.2.3 Dcinqnstrfltlon of Maintaining a Residual
The SWTR establishes two requirements concerning the Maintenance of
a residual. The first requirement is to maintain a minimus residual of
0.2 mg/l entering the distribution systea. The second is to aaintain a
detectable residual throughout the distribution systea. The disinfectant
used to aeet these requirements is Identified as the secondary disinfec-
tant throughout the remainder of this document. These requirements are
further explained in the following sections.
Maintaining a Residual Entering the Distribution SvtW
To avoid filtration, the disinfectant residual in water entering the
distribution system cannot be less than 0.2 ag/1 for more than four hours,
with one exception noted below. Systeas serving aore than 3,300 persons
aust monitor continuously. If there is a failure in the continuous
aonitoring equipment, the systea aay substitute grab sampling every four
hours for up to five working days following the failure of the equipment.
Systems serving 3,300 or fewer people aay aonitor continuously or take
grab samples at the frequencies prescribed below:
System Size by Population Samto/day*
*Samples cannot be taken at the same tiae.
The sampling intervals are subject to Primacy Agency review and
approval.
If at My tiae the residual disinfectant concentration falls below 0.2
ag/1 in a systea using grab saaple monitoring, the systea aust continue to
take a grab saaple every four hours until the residual disinfectant
concentration is equal to or greater than 0.2 ag/1. For all systeas, if
the residual concentration 1s not restored to at least 0.2 ag/1 within
four hours after a value of less than 0.2 ag/1 is observed, the system is
$500
501-1,000
1,001-2,500
2,501-3,300
1
2
3
4
3-29
-------
in violation of a treatment technique requirement, and Bust Install
filtration. However, 1f the Privacy Agency finds that the exceedance Mas
caused by an unusual and unpredictable circumstance, the Primacy Agency,
¦ay choose not to require filtration. EPA expects Primacy Agencies to use
this provision sparingly; it is intended to encompass catastrophic events,
not infrequent large stom events. In addition, any time the residual
concentration falls below 0.2 ng/1, the system wist notify the Primacy
Agency. Notification Bust occur as soon as possible, but no later than
the end of the next business day. The system also Bust notify the Primacy
Agency by the end of the next business day whether or not the residual was
restored within four hours.
Failure of a monitoring or reporting requirement does not trigger a
requirement to filter although they are violations.
Maintaining a Residual Within the SvitM
To avoid filtration, the disinfectant residual In the distribution
system cannot be undetectable in more than five percent of the samples in
a month, for any two consecutive months that the system serves water to
the public. Systems may measure HPC instead of disinfectant residual.
Sites with HPC concentrations of less than or equal to 500/ml are
considered equivalent to sites with detectable residuals for the purpose
of determining compliance. Public water systems must monitor for the
presence of a disinfectant residual (or HPC levels) at the same frequency
and locations as total coliform measurements taken pursuant to the Total
Coliform Rule. However, If the Primacy Agency determines, based on site-
specific considerations, that a system has no means for having a sample
transported and analyzed for HPC by a certified laboratory within the
requisite time and temperature conditions (Method 907, APHA, 1985), but
that the system Is providing adequate disinfection in.the distribution
system, this requirements does not apply to that system.
For systems which use both surface and ground water sources, the
Primacy Agency may allow the system to take disinfectant residual or HPC
samples at points other than the total coliform sampling locations if it
determines that such points are more representative of treated (disin-
fected) water quality within the distribution system.
3-30
-------
Disinfectant residual can be measured as total chlorine, free
chlorine, combined chlorine or chlorine dioxide (or HPC level). The SWTR
lists the approved analytical Methods for these Analyses. For example,'
several test methods can be used to test for chlorine residual in the
water, Including amperometric titration, DPD colorinetric, DPD ferrous
titrlmetric method and lodometrlc method, as described 1n the 16th Edition
of Standard Methods.1 Appendix 0 provides a review and summary of
available techniques to measure disinfectant residuals.
If a system fails to maintain a detectable disinfectant r-esldual or
an HPC level of less than or equal to 500/ml in more than 5 percent of the
samples during'a month, for any two consecutive months the system serves
water to the public, the system is in violation of a treatment technique
requirement. In addition, this system must Install filtration unless the
Primacy Agency determines that the violation was not due to a deficiency
1n treatment of the source water (e.g., the violation was due to a
deficiency in the distribution system, such as cross-connection contamina-
tion or failure 1n the pipeline).
The absence of a detectable disinfectant residual in the distribu-
tion system may be due to a number of factors, including:
Insufficient chlorine applied at the treatment plant
Interruption of chlorlnation
A chanae 1n chlorine demand In either the source water or the
distribution system
Long standing times and/or, long transmission distances
Available options to correct the problem of low disinfectant
residuals In distribution systems Include:
Routine flushing
1 Also, portable test kits are available which can be used 1n the field
to detect residual upon the approval of the Primacy Agency. These kits
may employ titration or colorimetrlc test methods. The colorimetric
kits employ either a visual detection of a residual through the use of
. a color wheel, or the detection of the residual through the use of a
hand held spectrophotometer.
3-31
-------
- -Increasing disinfectant doses at the plant
- Cleaning of the pipes (either mechanically by pigging or by
the addition of chemicals to dissolve the deposits) in the
distribution systea to remove accumulated debris which may be
exerting a disinfectant demand;
Flushing and disinfection of the portions of the distribution
systea in which a residual Is not Mlntalned; or
Installation of satellite disinfection feed facilities within
the distribution system.
For systems unable to Maintain a residual, the Primacy Agency say
determine that It Is not feasible for the system to monitor HPC and judge
that disinfection 1s adequate based on site-specific conditions.
Additional Information on maintaining a residual In the system Is
available In the AHHA Manual of Hater Supply Practices and Hater
Chlorlnatlon Principles and Practices.
3-2-4 Disinfection System Redundancy
Another requirement for unflltered water supply . systems Is
disinfection facility redundancy. A system providing disinfection as the
only treatment 1s required to assure that the water delivered to the
distribution system Is continuously disinfected. The SHTR requires either
redundant disinfection equipment with auxiliary power and automatic
start-up and alarm; or an automatic shutoff of delivery of water to the
distribution system when the disinfectant residual level drops below 0.2
mg/L. In order to fulfill the requirement of providing redundant
disinfection facilities, the following system Is recomended:
All components have backup units with capacities equal to or
greater than the largest unit on-line.
- A minimum of two storage units of disinfectant which can be
used alternately - e.g., two cylinders of chlorine gas, two
tanks of hypochlorite solution
- Where the disinfectant Is generated on-site, such as ozone,
backup units with a capacity equal to or greater than that of
the largest unit on-line.
- Automatic switchover equipment to change the feed from one
storage unit to the other before the first empties or becomes
Inoperable
3-32
-------
Feed systems with backup units with capacities equal to or
greater than the largest unit on-line.
An alternate power supply such as a standby generator with the
capability of running all the electrical equipment at the
disinfection .station. The generator should be on-site and
functional with the capability of automatic start-up on power
failure
Systems providing disinfection may have several different configura-
tions for type and location of disinfectant application. The following
guidelines are provided to assist Primacy Agencies and utilities in
determining the need for redundancy. Possible disinfection configurations
include:
one disinfectant used for primary and Secondary disinfection
- one point of application
- multiple points of application
two different disinfectants used for primary and secondary
disinfection
In many cases one disinfectant will be used to fulfill both the
total inactivation and residual requirements. One or sore application
points may be used to accomplish this. When one application point is used
to meet both the primary and secondary disinfection requirements, the
system 1s required to include redundant disinfection facilities.
When multiple points of application are used, redundancy 1s
recommended for the disinfection facilities at each point of application
which Is essential to meet the total Inactivation requirements. In
addition, to assure the maintenance of a residual entering and throughout
the distribution system, either:
the last point of application prior to the distribution system
should have redundancy, or
the point of application Immediately prior to this point
should have redundancy and sufficient capacity to assure a
residual entering the distribution system.
Systems may also use two different disinfectants, one to fulfill the
Inactivation requirements and the second to maintain a residual. An
3-33
-------
example of this would Include a system using ozone as a primary disinfec-
tant and chloramlnes as a secondary disinfectant. EPA recommends that:
the disinfection facilities at each point of disinfectant'
application In the primary system essential In providing the
overall Inactlvatlon requirements Include redundancy, and
the secondary disinfection facilities Include redundancy,
unless the disinfectant used for primary disinfection can
provide a residual for the distribution system as well. If
the primary disinfectant can be used for residual maintenance,
the last point of primary disinfectant application should
Include redundancy and sufficient capacity , to- assure a
residual entering the distribution system.
Appendix I contains more specific Information to assist the Primacy
Agency In establishing requirements for providing redundant disinfection
facilities.
Providing automatic shutoff of water delivery requires approval by
the Primacy Agency. The Primacy Agency must determine that this action
will not result 1n an unreasonable risk to health or Interfere with fire
protection. This determination should Include the evaluation of the
system configuration to protect against negative pressures In the system,
and providing for high demand periods Including fire flow requirements.
Automatic shutoff should be allowed only If systems have adequate
distribution system storage to maintain positive pressure for continued
water use.
3-34
-------
3.3 site-specific cmmmwK
In addition to meeting source water quality criteria and disinfec-
tion criteria, nonflltering systems using surface water supplies must meet .
the following criteria:
Maintain a watershed control program
Conduct a yearly on-site Inspection
Determine that no waterborne disease outbreaks have occurred
Comply with the revised annual total col 1 form MCI
Comply with TTHM regulations (currently applies to systems
serving >10,000 people)
Guidelines for meeting these other criteria are presented 1n the following
sections.
3.3.1 Watershed Control Program
A watershed control program is a surveillance and monitoring program
which is conducted to protect the quality of a surface water source. An
aggressive and detailed watershed control program is desirable to
effectively limit or eliminate potential contamination by human viruses.
A watershed program may impact parameters such as turbidity, certain
organic compounds, viruses, total and fecal coll forms, and areas of wild-
life habitation. However, the program is expected to have little or no
Impact on parameters such as naturally occurring Inorganic chemicals.
Limiting human activity In the watershed may reduce the likelihood of
animals becoming infected with pathogens and thereby reduce the transmis-
sion of pathogens by wildlife. Preventing animal activity near the source
water Intake prior to disinfection may also reduce the likelihood of
pathogen occurrence at the intake.
The effect of a watershed program 1s difficult to quantify since
many variables that Influence water quality are beyond the control or
knowledge of the water supplier. As a result, the benefit of a watershed
control program or specific control measures must 1n many cases be based
on accumulated cause and effect data and on the general knowledge of the
Impact of control measures rather than on actual quantification. The
effectiveness of a program to limit or eliminate potential contamination
by human viruses will be determined based on: the comprehensiveness of
the watershed review; the ability of the water system to effectively carry
3-35
-------
oyt and aonitor the aanagement decisions regarding control of detrimental
activities Occurring in the watershed; and the potential for the water
systea to aaxialze land ownership and/or control of land use within the
watershed. According to the SWTR, a watershed control prograa should
include as a minimum:
- A description of the watershed including its hydrology and
land ownership
- Identification, ¦onitoring and control of watershed character-
istics and activities in the watershed which a«y have an
adverse effect on the source water quality
- A prograa to gain ownership or control of the land within the
watershed through written agreements with land owners, for the
purpose of controlling activities which will adversely affect
the Microbiological quality of the water
An annual report which identifies special concerns in the
watershed and how they are being handled, identifies activi-
ties in the watershed, projects adverse activities expected to
occur in the future and how the utility expects to address
thea.
Appendix J contains a aore detailed guide to a comprehensive
watershed prograa.
In preparing a watershed control prograa, surface water systems
should draw upon the State watershed assessments and nonpoint source (HPS)
pollution aanageaent prograas required by S319 of the Clean Hater Act.
Inforaatlon on these prograas 1s available froa State water quality
agencies or EPA's regional offices. Assessaents Identify HPS pollutants
In water and assess the water quality. Utilities should use the
assessaents when evaluating pollutants 1n their watershed. Surface water
quality assessaents can also be obtained froa the lists of waters prepared
under S304(l) of the Clean Water Act, and State biennially prepared
5305(b) reports.
State HPS aanageaent prograas Identify best aanageaent practices
(BMPs) to be eaployed In reducing MPS pollution. These aanagement
prograas can be Incorporated in the watershed prograa to protect against
degradation o.f the source water quality.
3-36
-------
For systems using ground water sources under the Influence of
surface water, the control measures delineated In the Wellhead Protection
(WHP) program encompass the requirements of the watershed control program,
and can be used to fulfill the requirements of the watershed control
program. Guidance on the content of State Wellhead Protection Programs
and the delineation of wellhead protection areas Is given In: "Guidance
for Applicants for State Wellhead Protection Program Assistance Funds
Under the Safe Drinking Water Act," June, 1987, and "Guidelines for
Oellheatlon of Wellhead Protection Areas," June, 1987, available from the
EPA office of Ground-Water Protection (WH-550G).
As a minimum, the WHP program must:
Specify the duties of State agencies, local governmental
entitles and public water supply systems with respect to the
development and Implementation of Programs;
Determine the wellhead protection area (WHPA) for each
wellhead as defined 1n subsection 1428(e) based on all
reasonably available hydrogeologlc Information, ground-water
flow, recharge and discharge and other Information the State
deems necessary to adequately determine the WHPA;
Identify within each WHPA all potential anthropogenic sources
of contaminants which may have any adverse effect on the
health of persons;
• Describe a program that contains, as appropriate, technical
assistance, financial assistance, Implementation of control
measures, education, training and demonstration projects to
protect the water supply within WHPAs from such contaminants;
A
Present contingency plans for locating and providing alternate
drinking water supplies for each public water system In the
event of well or wellfleld contamination by such contaminants;
Consider all potential sources of such contaminants within the
expected wellhead area of a new water well which serves a
public water supply system; and
Provide for public participation.
3.3.2 On-site Inspection
v
The watershed control program and on-site inspection are Inter-
related preventive strategies. On-site Inspection Is actually a program
which Includes and surpasses the requirements of a watershed program.
3-37
-------
While the watershed program Is mainly concerned with the water source,
on-site Inspection Includes some additional requlreaents for source water
quality control and Is also concerned with the disinfection facilities.
As defined by the EPA, an.on«s1te Inspection Includes review of the water
source, disinfection facilities and operation and Maintenance of i public
water system for the purpose of evaluating the adequacy of such systems
for producing safe drinking water.
The SWTR requires an annual on-site Inspection to evaluate the
watershed control program and disinfection facilities. The-Inspection
must be performed by a party approved by the Privacy Agency. The Inspec-
tion should be conducted by competent Individuals such as sanitary and
civil engineers, sanitarians, and technicians who have experience and
knowledge In the operation, Maintenance, and design of water systems, and
who have a sound understanding of public health principles and waterbome
diseases. Guidance for the contents of an Inspection are Included In the
following paragraphs. Appendix K presents guidelines for a sanitary
survey which Includes and surpasses the requlreaents of an on-site
Inspection.
As the first step In determining which SWTR requirements, If any, a
source Is subject to, EPA recommends that utilities conduct a detailed,
comprehensive sanitary survey. Appendix K presents a comprehensive list
of water system features that the person conducting the survey should be
aware of and review as appropriate. This Initial Investigation estab-
lishes the quality of the water source, Its treatment and delivery to the
consumer. EPA reco—ends that this comprehensive evaluation be repeated
every three years for systems serving 4,100 people or less and every five
years for systems serving more than 4,100 people. Also, under the Total
Collfor* Rule, ground water systems which take Itss than 5 coHform
samples per month must conduct such sanitary surveys within every 5 or 10
years dependiog upon whether the source is protected and disinfected.
The annual on-site Inspection to fulfill the SWTR requirements
should include as a minimum:
3-38
-------
1 at CUSTOM
PM
T
STORAGE
STORAOE
PM >
r-t
TANK 1
t
TANK 2
M
CHLORINE
OIOXIOC
CHLORINE
CHLORINE
SECTION
SECTION
I SECTION
FIGURE 3* 1-DETERMINATION OF IN ACTIVATION FOR
MULTIPLE DISINFECTANT APPLICATION
TO A 8URFACE WATER SOURCE
-------
1. Source Evaluation
a. Review the effectiveness of the watershed control
program (Appendix J).
b. Review -the physical condition and protection of the
source Intake.
c. Review the Maintenance program to Insure that all
disinfection equipment Is appropriate and has received
regular Maintenance and repair to assure a high operat-
ing reliability.
2. Treatment Evaluation
a. Review Improvements and/or additions made to disinfec-
tion processes during the previous year to correct
deficiencies detected in earlier surveys.
b. Review the condition of disinfection equipment.
c. Review operating procedures.
d. Review data records to assure that all required tests
are being conducted and recorded and disinfection 1s
effectively practiced (CT calculations should be spot
checked to ensure that they were done correctly)
e. Identify any needed Improvements 1n the equipment,
system maintenance and operation, or data collection.
In addition to these requirements, a periodic sanitary survey 1s
recommended for all systems, Including those with filtered and unflltered
supplies. The sanitary survey should Include the Items listed 1n 1 and 2
above as well as:
3. Distribution System Evaluation
a. Review the condition of storage facilities.
b. Determine that the system has sufficient pressure
throughout the year.
c. Verify that system equipment has received regular
aalntenance.
d. Review additions/improvements Incorporated during the
year to correct deficiencies detected 1n the initial
inspection.
3-39
-------
e. Review cross connection prevention program, includina
annual teiting of backflow prevention devices.
f. Review routine flushing program for effectiveness.
g. Evaluate the corrosion control program and its impact on
distribution water quality.
h. Review the adequacy of the program for periodic storage
reservoir flushing.
i. Review practices in repairing water sain breaks to
assure they Include disinfection.
4. Management/Operation Evaluation
a. Review the operations to assure that any difficulties
experienced during the year have been adequately
addressed. .
b. Review staffing to assure adequate numbers of properly
trained and/or certified personnel are available.
c. Verify that a regular Maintenance schedule is followed.
d. Audit systems records to verify that they are adequately
¦aintained.
e. Review bacteriological data from the distribution system
for colifora occurrence, repeat samples and action
. response.
3.3.3 No Disease Outbreaks
Under the provisions of the SWTR, a surface water system which does
not filter must not have been identified as a source of waterbome
disease, or if 1t has been so identified, the system must have been
modified sufficiently to prevent another such occurrence, as determined by
the Primacy Agency. If a waterborne disease outbreak has occurred and the
outbreak was or Is attributed to a treatment deficiency, then the system
must install filtration unless the system has upgraded Its treatment
system to remedy the deficiency which led to the outbreak and the Primacy
Agency has determined that the system is satisfying this requirement. If
the Primacy Agency has determined the disease outbreak was the result of
a distribution system problem rather than a source water treatment
deficiency, the system is not required to install filtration.
3-40
-------
In order to determine whether the above requirement Is being met,
the responsible federal, state and local health agencies should be
surveyed to obtain the current and historical information on waterborne-
disease outbreaks which nay have occurred within a given system. Whether
conducted by the Primacy Agency or submitted by the water purveyor, this
Information should include:
1. Source of the Information:
a. Name of agency
b. Name and phone number of person contacted
c. Oate of inquiry
2. Outbreak Data
a. Known or suspected incidents of waterborne disease
outbreaks
b. Oate(s) of occurrence(s)
c. Type or identity of illness
d. Number of cases
3. Status of Disease Reporting:
a. Changes In regulations; e.g., giardiasis was not a
reportable disease until 1985
4. If a Oisease Outbreak has Occurred:
a. Has the reason for the outbreak identified; e.g.,
inadequate disinfection?
b. Did the outbreak occur while the system was in its
current configuration?
c. Has remedial action taken?
d. Have there been any further outbreaks since the remedial
action was taken?
If • review of the available Information Indicates that the system
or network for disease reporting Is Inadequate within the Primacy Agency's
area of responsibility, efforts should be made to encourage the appropri-
ate agencies to upgrade the disease reporting capabilities within th?
area.
3-41
-------
3'3*4 Monthly CoHfara MTI
To avoid filtration, a system Bust comply with the MCI for total
coliforiM, established in the Total Coliform Rule, for at least 11 out of
the previous 12 aonths the system served water to the public on an ongoing
basis, unless the Primacy Agency determines that failure to meet this
requirement was not caused by a deficiency 1n treatment of the source
water. If the Primacy Agency makes such a determination, the system is .
not required to install filtration. The Total Coliform Rule requires
systems using surface water or ground water under the Influenced surface
water which do not filter to collect a sample at or near the first
customer each day that the turbidity level exceeds 1 NTU within 24 hours
of learning of the result and to analyze the sample for the presence of
total conforms. (If the Primacy Agency determines that It 1s not
possible for the system to have such a sample analyzed within 24 hours,
this time limit may be extended on a case-by-case basis.) This sample may
be used to fulfill the routine coo^Hance monitoring requirements of the
Total Coliform Rule. The results of the additional sample must be
Included in determining whether the system 1s In compllance with the
monthly MCI for total coll forms.
3.3.5 Total Trlhaloffiethm (TTHH) Regulations
For the system to continue . to use disinfection as the only
treatment, 1t must comply with the total trihalomethane (TTtM) MCL
regulation. The current regulation established an MCL for total TTHM of
0.10 mg/L for systems serving a population greater than 10,000. Both the
NCI and the system population covered may be reduced 1n the future, and
this should be considered when'planning disinfectant application.
One alternative to meet the CT requirements of the SWTR is to
Increase the disinfectant dose. For many systems, a higher chlorine dose
will refeliTt In increased formation of TTHMs. Changes 1n disinfection
practice should maintain TTHM levels of less than 0.10 mg/L. In lieu of
Increasing chlorine dose, use of an alternate disinfectant which produces
fewer TTHMs could be considered. Alternate disinfectants Include the use
of ozone or chlorine dioxide as primary disinfectants with chlorine or
chloramlnes as secondary (residual) disinfectants. It 1s important to
note that EPA also will promulgate regulations for disinfectants and
3-42
-------
disinfection by-products which My Halt application of som of these
disinfectants. EPA recommends that Primacy Agencies keep informed through
coonunlcation with EPA on interim guidance on how to avoid conflict for
systems to comply with both the SWTR and the forthcoming regulations on
disinfectants and disinfection by-products. Any changes which appear to
not meet the by-product regulations should not be Implemented.
3-43
-------
4. OESIGN ANO OPERATING CRITERIA FOR
FILTRATION AND DISINFECTION TECHNOLOGY
4.1 introduction
To comply with the SWTR, public water systems must Include filtra-
tion, or some other approved particulate removal technology, In their
treatment process unless they are able to satisfy certain conditions.
Those conditions Include compliance with source water quality criteria and
site-spedf 1c criteria. Guidance for determining whether these conditions
•re met 1s provided 1n Section 3 of this manual. Systems unable to
satisfy these conditions must provide particulate removal and meet
criteria pertaining to operation, design and performance. These criteria
are specified In part 1n the definitions of technologies 1n the SWTR and
more specifically as determined by the Primacy Agency.
This section provides guidance both for those water systems which
currently do not have filtration equipment and must add It, and for
systems which have existing filtration processes. Guidance on additional
alternatives for small systems 1s presented 1n Appendix L.
This section Includes guidance on the following topics:
a. Filtration Technology: Descriptions, capabilities, design
criteria and operating requirements for each technology, and
a listing of major factors to be considered In their
selection, Including raw water quality considerations.
b. 01s1nfect1on: Descriptions of the most applicable disin-
fection technologies used with filtration systems, and a
presentation of the relative effectiveness of these disinfec-
tion technologies with respect to 1nact1vat1on of bacteria,
cysts and viruses.
c. Alternate Technologies: Descriptions of some currently
available alternate filtration technologies.
d. Other Alternatives: Includes a description of some nontreat-
ment alternatives Including regionalIzatlon and use of an
alternate source.
4.2 Selection of Appropriate Filtration Technology
Filtration 1s generally provided by passing water through a bed of
sand, a layer of dlatomaceous earth or a combination bed of coarse anthra-
4-1
-------
cite coil overlaying finer sand. Filters are classified and named in a
nusber of ways. For example, based on application rate, sand filters can
be classified as either slow or rapid; yet these two types of filters
differ in aany sore characteristics than just application rate. They
differ in their reaoval process, bed material, Mthod of cleaning, and
operation. Based on the type of bed saterial, filters can be classified
as sand, dlatomaceous earth, dual-media (coal-sand) or even multi-media
in which a third layer of high density sand is used.
4.2.1 fianeral Descriptions
Current technologies specified by the SWTR are:
a. Conventional Treataent: A series of processes Including
coagulation, flocculatlon, sedimentation and filtration.
b. Direct Filtration; A series of processes including coagula-
tion (and perhaps flocculatlon) and filtration, but excluding
sedimentation.
c. Slow Sand Filtration; A process which involves passage of raw
water through a bed of sand at low velocity, generally less
than 0.4 meters/hour (1.2 ft/hr), resulting in substantial
particulate removal by physical and biological mechanisms.
d. Dlatomaceous Earth Filtration: A process that meets the
following conditions:
A precoat cake of dlatomaceous earth filter media is
deposited on a support membrane (septus)
- The water 1s filtered by passing it through the cake on
the septus; additional filter sedia, known as body feed,
is continuously added to the feed water in order to
maintain the perseability of the filter cake.
*
4-2
-------
«. Alternate Technologies: Any filtration process other than
those listed above. Available alternate filtration technolo-
gies include, but are not Halted to:
Package Plants1
Cartridge Filters
4.2.2 Capabilities
Filtration processes provide various levels of turbidity and
¦Icroblal contaminant removal. When properly designed and operated and
when treating source waters of suitable quality, the above filtration
processes are capable of achieving at Itut a 2-1 og (99 percent) renoval
of Glardla cysts and at least a l-1og (90 percent) removal of viruses
without disinfection (Logsdon, 1987b; USEPA, 1988b; Roebeck, 1962). The
exception 1s cartridge filters which nay not provide effective virus
renoval. A summary of the renoval capabilities of the various filtration
processes 1s presented In Table 4-1.
As Indicated 1n Table 4-1, conventional treatment without disinfec-
tion Is capable of achieving up to a 3-1og renoval of Glardla cysts and
up to a 3-log renoval of viruses. Direct filtration can achieve up to a
3-log renoval of Glardla cysts and up to a 2-1 og renoval of viruses.
Achieving the maxlnum renoval efficiencies with these treatnent processes
requires the raw water to be properly coagulated and filtered. Factors
which can adversely affecy renoval efficiencies Include:
Raw water turbidities less than 1 NTU
Cold water conditions
Non-optlnal or no coagulation
Inproptr filter operation Including:
1 Depending upon the type of treatnent units In place, historical
perfornance and/or pilot plant work, these plants could be categorized
as one of the technologies 1n a-d above at the discretion of the State.
Several studies have already Indicated that sone package plants
effectively renove Glardla cysts. If such plants provided adequate
disinfection so that the conplete treatnent train achieves at least a
3-log renoval/Inactlvatlon of Glardla cysts and a 4-1 og renoval/inacti-
vatlon of viruses, use of this technology would satisfy the minimum
treatment requirenents.
-------
No filter to waste
Interaittent operation
Sudden rate changes
Poor housekeeping
- Operating the filters after turbidity breakthrough
Studies of slow sand filtration have shown that this technology
(without disinfection) 1s capable of providing greater than a 3-1 og
removal of fiiardia cysts and greater than a 3-log reaoval of viruses.
Factors which can adversely affect reaoval efficiencies Include:
Poor source water quality
Cold water conditions
Increases 1n filtration rates
Decreases 1n bed depth
Iaproper sand sire
Inadequate ripening
Olatooaceous earth (DE) filtration can achieve greater than a 3-log
reaoval of filardla cysts when sufficient precoat and body feed are used.
However, turbidity and total collfora removals are strongly Influenced by
the grade of OE eaployed. Conversely, OE filtration Is not very effective
for removing viruses unless the surface properties of the dlatoaaceous
earth have been altered by pretreataent of the body feed with alua or a
suitable polyaer. In general, OE filtration Is assuaed to achieve only
a 1-log reaoval of viruses unless demonstrated otherwise. Factors which
can affect the reaoval of fiianlii cysts and viruses Include:
Precoat thickness
- Aaount of body feed
- Gride of OE
- • roper conditioning of septua
-- iaproper pretreataent of the body feed
Package plants can be used to treat water supplies for coaaunlties
as well is for recreational areas, parks, construction caaps, ski resorts,
allltary installations and other facility where potable water 1s not
available froa a aunlclpal supply. Operator requlreaents vary signifi-
cantly with specific situations. Under unfavorable raw water conditions.
4-4
-------
TABLE 4-1
REMOVAL CAPABILITIES QFFILTRATION PROCESSES^
Log Removali
Proem
Giardia{l)
SML.
Viruses
Total(I)
Col 1 form
Conventional Treataent
2*3
1 - 3(,)
>4
Direct Filtration
2-3
1 - 2l,)
1-3
Slow Sand Filtration
2 - 3(,)
% . 3<«>
1 -2
Olatoaaceous Earth
Filtration
2 - 3
1 - 3
Note:
1. Without disinfection.
2. logsdon, 1917b.
3. Roebeck fii iX 1962.
4.. Poynter tnd Slide, 1977.
5. These technologies generally achieve greater than a 3-log removal.
-------
package plants could dMand full-time attention. Package plants art most
widely used to trtat surface supplies for removal of turbidity, color and
conform organism prior to disinfection. They are currently available
1n capacities up to 6 mgd.
Colorado State University conducted a series of tests on one package
plant over a 5-month period during the winter of 1985-86 (Horn and
Hendricks, 1986). Existing Installations 1n Colorado had proven effective
for turbidity removal, and the tests at the university were designed to
evaluate the system's effectiveness 1n removing collform bacteria and
filardla cysts fro* low turbidity, low temperature source waters. The test
results showed that the filtration system could remove greater than
99 percent of 61ard1a cysts for waters which had less than 1 NTU turbidity
and less than 5 C temperatures, as long as proper chemical treatment was
applied, and the filter rate was 10 gpa/ft2 or less. In addition, an
alternate water source having a turbidity ranging from 3.9 to 4.5 NTU was
used In 12 test runs with coagulant doses ranging from 15 to 45 mg/L. The
effluent turbidities from these runs were consistently less than 0.5 NTU.
Surveys of existing facilities Indicate that while package plants
may be capable of achieving effective treatment, many have not consistent-
ly met the Interim MCL for turbidity, and In some cases, conforms were
detected In the filtered water (Morand et al., 1980; Morand and Young,
1983). The performance difficulties were primarily the result of the
short detention time Inherent 1n the design of the treatment units, the
lack of skilled operators with sufficient time to devote to operating the
treatment facilities, and the wide-ranging variability In quality of the
raw water source. For Instance, raw water turbidity was reported to often
exceed 100 NTU at one site. Improvements 1n operational techniques and
methods at this site resulted 1n a substantial Improvement In effluent
quality. After adjustments were made, the plant was capable of producing
a filtered water with turbidities less than 1 NTU, even when Influent
turbidities Increased from 17 to 100 NTU within a 2-hour period, as long
as proper coagulation was provided.
One of the major conclusions of these surveys was that package water
treatment plants manned by competent operators can consistently remove
4-5
-------
turbidity -and bacteria from surface waters of a fairly uniform quality.
Package plants Applied where raw water turbidities are variable require
a high degree of operational skill and nearly constant attention by the
operator. Regardless of the quality of the raw water source, all package
plants require at least a minimum level of Maintenance and operational
skill and proper cheaical treatment if they are to produce satisfactory
water quality.
Cartridge filters using mlcroporous filter elements (ceramic, paper
or fiber) with pore sizes as saall as 0.2 ua aay be suitable for producing
potable water from raw water supplies containing aoderate levels of
turbidity, tlgae and aicrobiological contaminants. The advantage to saall
systems of these cartridge filters is that, with the exception of
disinfectant, no other chenicals are required. The process is one of
strictly physical removal of saall particles by straining as the water
passes through a porous cartridge. Other than occasional cleaning or
cartridge replacement, operational requirements are not complex and do not
require skilled personnel. However, the SWTR does require each surface
water system to be operated by a qualified operator, as determined by the
Primacy Agency. Such a system may be suitable for some small systems
where, generally, only maintenance personnel are available for operating
water supply facilities. However, the use of cartridge filters should be
liaited to low turbidity source waters because of their susceptibility to
rapid headloss buildup. For exaaple, manufacturer's guidelines for
achieving reasonable filter run lengths with certain polypropylene filter
elements are that the raw water turbidity be 2 NTU or less (USEPA, 1988b).
Long (1983) analyzed the efficacy of a variety of cartridge filters
.using turbidity measurements, particle size analysis, and scanning
electron microscope analysis. The filters were challenged with a
suspension of microspheres averaging 5.7 ua in diameter which Is smaller
than a filitfli cyst. The microspheres were applied at a concentration of
40,000 to 65,000 spheres per ml. Ten of 17 cartridge filters removed over
99.9 percent of the microspheres.
In tests using live Infectious cysts from a human source, cartridge
filters were found to be highly efficient 1n removing GlflrtU cysts
4-6
-------
(Hibltr, 1986). Each test Involved challenging a filter with 300.000
cysts at a concentration of 10,000 cysts/ml. The average removal for five
tests was 99.86 percent, with removal efficiencies ranging from 99.5 per-
cent to 99.99 percent.
The application of cartridge filters to small water systems using
either cleanable ceramic or disposable polypropylene cartridges appears
to be a feasible method for removing turbidity and most microbiological
contaminants. However, data regarding the ability of cartridge filters
to remove viruses are not available. Since disinfection by Uself could
achieve a 4-log Inactlvatlon of viruses, If the cartridge filter removes
greater than or equal to 3 logs of fiiintti, then the filter plus
disinfection would achieve the overall minimum requirements, regardless
of whether only negligible Glardla Inactlvatlon Is achieved (e.g., less
than 0.5 log). However, consideration should be given to the feasibility
of providing multiple barriers of treatment for each target organism,
I.e., some 61ard1a and virus removal by each barrier (I.e., some removal
by filtration and some Inactlvatlon by disinfection) as protection In case
one of the barriers falls. The efficiency and economics of the process
must be closely evaluated for each situation. Pretreatment In the form
of roughing filters (rapid sand or multi-media) or fine mesh screens may
be needed to remove larger suspended solids which, If not removed, could
cause the rapid buildup of headloss across the cartridges (USEPA, 1988a).
In general, conventional treatment, direct filtration, slow sand
filtration and dlatomaceous earth filtration can be designed and operated
to achieve the maximum removal of the water quality parameters indicated
In Table 4-1. Howmver, for the purpose of i«1ect1no the appropriate
filtration and disinfection technologies and for determining design
criteria, these filtration processes should be assumed to achieve a 2-1 og
removal of-Siardla cysts and a 1-log removal of viruses. This conserva-
tive approach will assure that the treatment facility has adequate
capability to respond to non-optimum performance due to changes 1n raw
water quality, plant upsets, etc. The balance of the required removals
and/or Inactlvatlon of filardla cysts and viruses would be achieved through
the application of appropriate disinfection.
4-7
-------
Tht performance of alternate technologies such as cartridge filters,
and possibly package plants, depending upon the unit under consideration
cannot be stated with certainty at this tlae. Because of these perform-
ance uncertainties, pilot studies should be used to demonstrate their
efficacy for a given mater supply.
4.2.3 Selection
For any specific site and situation, a nuaber of factors will
determine which filtration technology is most appropriate. .'Among these
are: raw water quality conditions, space and personnel availability, and
economic constraints. A discussion of the iapact of raw water quality on
the technology selection Is presented here. The iapact of site-specific
factors and economic constraints 1s presented In the USEPA document
"Technologies and Costs for the Reaoval of Microbial Contestants froa
Potable Water Supplies- (USEPA, 1988b).
Raw Water Quality Conditions
Tht nuaber of treataent barriers provided should be coaaensurate
with the degree of contaalnatlon In the source water. The four technolo-
gies specified in the SWTR vary in their ability to aeet the performance
criteria when a wide range of raw water quality 1s considered. While the
numerical values of raw water quality that can be accoaaodated by each of
the four technologies will vary froa site to site, general guidance can
be provided. General guidelines for selecting filtration processes, based
on total colifora count, turbidity, and color are presented in Table 4-2.
It Is not rtcoaaendtd that filtration systeas other than those listed in
Table 4-2 be used whin the general raw water quality conditions exceed
the valuat listed, unless 1t has been demonstrated through pilot testing
that the technology can aeet the perforaance criteria under the raw water
quality conditions expected to occur at the site.
The filtration processes listed In Table 4-1 are capable of
achieving the required perforaance criteria when properly designed arrd
operated 1f they are treating a source water of suitable quality (i.e..
generally within the ranges indicated in Table 4-2). One of the causes
4-8
-------
TABLE 4-2
GENERALIZED CAPABILITY OF FILTRATION SYSTEMS
TO ACCflfttOATE RAW WATER OHALtTY CONDITIONS
general Restrictions ;
Color
(CT)
<75(I)
<75tJ)
<40(4)
<10t,»
<5<»
<5(1)
1. Depends on algae population, alua or cat ion 1c polywer
coagulation — (Clcasby «t al.t 1984.)
2. USEPA, 1971.
3. Letteraan, 1986.
4. Bishop it at., 1980.
5. Slezak and S1as, 1984.
Treatment
Conventional with
predlsInfectIon
Convantlonal without
predlslnfectlon
Direct filtration
with flocculation
In-line filtration
Slow land filtration
Olatooaceous earth
filtration
Total
Colifoms
(f/iQQ m
<2Q,mm
<5,000(,)
<500(,)
<500(,)
<800(,)
<50(1)
Turbidity
CHTU)
No restrictions111
No restrictions*1*
«7-l4(l)
<7-14(,)
<10(,)
<5d)
Notes;
-------
of filtration failures Is the use of Inappropriate technology for a given
raw water quality (Logsdon, 1987b). These criteria arc general guide-
lines. Periodic occurrences of raw water collfora, turbidity or color
levels In excess of the values presented In Table 4-2 should not preclude
the selection or use of a particular filtration technology. For example,
the following alternatives are available for responding to occasional raw
water turbidity spikes:
a. Direct Filtration
Continuous monitoring and coagulant dose adjustment
More frequent backwash of filters
Use of presedlmentatlon
b. Slow Sand Filtration
Use of a roughing filter
Use of an Infiltration gallery
c. Olatomaceous Earth Filtration
Use of a roughing filter
Use of excess body feed
For the above alternatives, EPA recoanends that pilot testing be
conducted to demonstrate the efficacy of the treatment alternative.
The characteristics of each filtration technology are • major factor
1n the selection process. Significant characteristics Include perfornance
capabilities (contaalnant removal efficiencies), design and construction
requirements, and operation and maintenance requirements. Details
regarding each of the four filtration technologies are presented In the
following section.
4.3 Available Filtration Technologies
4.3.x introduction
As Indicated In the preaable to the SWTR, the historical responsi-
bility of the States to establish design and operating criteria for public
drinking water plants will continue. The purpose of the following
sections Is to provide guidance on how the design and operating criteria
may need to be changed In order to assure that the performance criteria
In the SWTft are met.
4-9
-------
The design criteria for the various filtration technologies found
1n the 118? edition of Recommended Standard* far Water Work* (6feit Lakes,
1987) are the minimum design criteri* that a majority of states are
currently following.1 These standards are referred to as Ten States
Standards in the reminder of this Manual. The design criteria contained
in the Ten States Standards have not been duplicated here. Rather, the
reader is referred to the Ten States Standards directly. EPA recosmends
the following additions and/or changes to the Ten State Standards In order
to assure compliance with the performance criteria of the SWTft.
4.3.2 general
The following recommendations apply to all filtration plants:
a. All filtration plants should provide continuous turbidity
monitoring of the effluent turbidity from each individual
filter. * If continuous monitoring Is impractical, routine
monitoring of individual filters 1s recommended as a minimum.
b. All filtration systems should be concerned with the peak
turbidity levels in the filtered water after backwashing and
2 Based upon the results of a survey conducted for the American Hater
Works Association Research Foundation (AWWARF), some 38 states use the
Ten States Standards entirely or in modified form (AWWARF, 1986).
1 Although this is not a requirement of the SWTR, 1t is rec amended
because of the possibility that not all filters In a treatment plant
will produce the same effluent turbidity. This may be due to a variety
of conditions that include bed upsets, failure of media support or
underdrain systems, etc. Although the combined effluent from all the
filters may meet the turbidity requirements of the SHTR, the turbidity
1tv«1 from an individual filter may substantially exceed the limits.
TMs may result in the passage of Siardia cysts or ether pathogens.
1 Validation should be performed at least twice a week based on the
procedure outlined in Part 214A in the 16th Edition of Standard Methods.
It should be noted that Improper installation of continuous monitors
may allow for air bubbles to enter the monitor resulting in false
turbidity spikes. To avoid air bubbles reaching the turbidimeter the
sample tap should be installed below the center line of the pipe and
an air .release valve may be included on the sample line.
• 4-10
-------
make every attempt to operate tht filters to minimize the
.magnitude and duration of these turbidity spikes.4
Individual filters should be monitored as discussed In Section
4.3.2.1 end when excessive turbidity spikes ire found, corrective actions
taken. Ourlng these turbidity peaks, GlardU cysts and other pathogens
¦ay be passed Into the finished water. There 1s evidence that a 0.2 to
0.3 NTU Increase 1n the turbidity during the first period of the filter
run can be associated with rises In Glardla cyst concentrations by factors
of twenty to forty (Logsdon, 1985). Special studies should be conducted
to determine the extent of the turbidity spike problems.
There are basically four approaches available for correcting
probleas with turbidity spikes after backwashlng. These are as follows
(Bucklln, et al 1988):
Proper chemical conditioning of the Influent water to the
filter can minimize the magnitude and duration of these
turbidity spikes. This could Include proper control of the
prlnary coagulant chemicals such as alum or Iron compounds.
In some cases filter aids using polymers may be needed to
control the turbidity spikes.
Gradually increasing the filtration rate 1n Increments when
placing the filter 1n operation. Starting the filter at a low
flow rate and then Increasing the flow in small Increments
over 10 to 15 minutes has been shown to reduce the turbidity
spikes 1n some cases (Logsdon, 1987).
Addition of coagulants to the backwash water has also been
shown to reduce the extent of turbidity spikes after backwash.
Typically the same primary coagulant used 1n the plant Is
added to the backwash water. Polymers alone or In combination
with the primary coagulant may also be used.
Filter-to-waste may be practiced where a portion of the
filtered water Immediately after starting the filter is
. wasted. This Is only possible where the filter system has
s For most high rate granular bed filters, there Is a period of
conditioning, or break-In Immediately following backwashlng, during
which turbidity and particle removal Is at a minimum, referred to as
the break-In period. The turbidity peaks are thought to be caused by
remnants of backwash water within the pores of and above the media
passing through the filter, and/or floe breakup during the filter
ripening period before 1t can adequately remove Influent turbidity.
4-11
-------
provided the necessary valves and piping to allow this
procedure. There Is some concern whether or not this practice
is beneficial. The extra valve operations needed for fllter-
to-waste can disrupt the filter flow rate to the extent that
they create their own turbidity spikes. Some knowledge of the
time actually needed for filter-to-waste Is also needed before
It can be determined that this Is an effective procedure for
controlling turbidity spikes. If the length of tine the
f11ter-to-waste Is practiced Is less than that before the
turbidity spike passes, the disruption caused by the valve
operation My actually Increase the turbidity spike.
Different plants and the Individual filters within the plant say
have different turbidity spike characteristics. The four approaches
presented above, therefore, Bust be evaluated on a case-by-case basis.
Special studies will be required to Identify those filters with the
turbidity spike problems and assist In selecting which of the four
approaches 1s best for correcting the problem. It has been generally
found that turbidity spikes can be alnlalzed through one or a combination
of the first three approaches.
In order to establish filter-to-waste operating guidelines, the
following procedure 1s suggested:
Review the effluent turbidity data for each filter and deter-
mine which filter historically has the highest effluent tur-
bidity.
• Following backwasMng of the filter with the poorest perfor-
mance, place that filter Into service and collect orab samples
every 5 to 10 minutes for a period of at least 60 minutes.1
Analyze the grab samples for turbidity and determine how long
the filter must be 1n operation before the effluent turbidity
drops
to less than or equal to 0.5 NTU
- or 1 NTU In cases where a filtered water turbidity of
less than or equal to 1 NTU Is allowed.
1 Continuous turbidity monitoring can be used 1n place of grab sampling
4-12
-------
Halted Information exists on the typical aagnltude and duration of
peak turbidity levels after backwashlng and what levels are considered
acceptable to assure that these turbidity spikes are not associated with
passage of filardla cysts. Information froa plant scale tests, showing
the typical aagnltude and duration of these turbidity spikes 1s available
froa two plants (Bucklln et al;. 1988). Studies conducted at these plants
over a year showed that these peaks occurred within the first few alnutes
after the filter was placed back In operation, their effects lasted for
several hours, and varied In Magnitude froa 0.08 to 0.35 NTU.on average.
For existing plants without provisions for fllter-to-waste, the
decision to add the necessary piping to provide this capability should be
Bade only after carefully evaluating the other three approaches. If the
results of special studies show that the other three options are not
effective In alnlalzlng the turbidity spikes then the expense of adding
the filter-to-waste capabilities aay be justified.
For new plants the capability of f1 lter-to-waste say be required by
the Prlaacy Agency or should be considered. By haying this capability,
additional flexibility will be available for turbidity spike control.
This flexibility say also be useful for other filter Maintenance functions
such as after aedla replacitent or when heavy chlorlnatlon of the filter
1s needed after aalntenance.
4.3.3 Conventional Treatment
Conventional treatment 1s the aost widely used technology for
removing turbidity and Microbial contaalnants fro* surface water supplies.
Conventional treatment includes the pretreataent steps of chenlcal
coagulation, rapid alxlng, flocculatlon and sedlaentatlon followed by
filtration- These conventional treataent plants typically use alualnun
and Iron eoapounds In the coagulation processes. Polyaers aay also be
used to enhance the coagulation and filtration processes. A flow sheet
for a conventional treataent plant 1s presented on Figure 4-1.
Llae softening 1s a treataent process used to reaove hardness and
turbidity froa surface waters. Treataent 1s typically accoapllshed with
conventional process units. The llae softening process removes the
4-13
-------
calciua and aagnesiua froa the water by precipitating thea as calcium
carbonate and aagnesiua hydroxide. Turbidity levels 1n the water are also
reduced by this process. Liae and possibly soda ash is added to the raw
water to raise Its pH to a point at which these precipitates are foried
and then removed froa the'water during sedimentation end filtration. Line
softening My be used for the reaoval of carbonate hardness in the pH
range of 9 to 10 through a single stage process. Two-stage liae/soda ash
softening at a pH of 10 to 12 can be used for the reaoval of non-carbonate
hardness and aagnesiua. Two-stage softening includes recafbonation to
neutralize the caustic alkalinity, reducing the pH to the range of 8.S to
9.5. A flow- sheet for typical one- and two-stage softening plants Is
presented on Figure 4-2.
Each of these three conventional treatment processes uses filtration
following sedimentation. Three different types of filters are used. Sand
filters, normally found in older plants, use a single aedia of sand to
fora a filter bed, and are generally designed with a filtration rate of
2 gpa/ft1. Newer plants noraally use dual-aedia or aixed Media filters.
Dual aedia filters use a coabination of anthracite coal along with a sand
to fora the filter bed. Mixed aedia filters use coal, sand, and a third
Material to fora the filter bed. Dual and aixed aedia filters can be
designed to operate at higher filtration rates than sand filters, i.e.,
4 to 6 gpa/ftz.
Ptsjgn Cftttrli
The ainiaua design criteria in the Ten State Standards for
conventional treatment are considered sufficient for the purposes of
coaplying with the SWTR with the following addition:
• The criteria for sediaentatlon should be expanded to Include
other aethods of solids reaoval including dissolved air
flotation. Plate separation and upflow-solids contact
clarifiers included in the 1987 Ten State Standards should
also be considered.
Operating Requirement!
In addition to the operating requireaents in the Ten State
Standards, a coagulant should be used at all tiaes the treatment plant is
4-14
-------
In operation.7 Conventional and direct filtration plants must be monitored
carefully because failure to maintain optimum coagulation can result 1n
poor filter performance and breakthrough of cysts and viruses.1 Although
the detention time provided by the settling basins results In some Margin
of safety, the loss of coagulation control at the chemical feed or rapid
mix points nay not be noticed until the poorly coagulated water reaches
the filters, after the process has failed. Failure to effectively Monitor
and control filter operation can result In undetected poor filter
performance (Logsdon, 1987a; Logsdon, 1987b).
Effective operation of a conventional treatment plant requires
careful aonltorlng and control of:
Chemical Feed
- Rapid Mix
Flocculatlon
i - Sedimentation
Filtration
For the purposes of the SWTR, the requirements for effective
operation of a conventional water treatment plant can be summarized as
follows:
a. The application of a coagulant and the maintenance of
effective coagulation and flocculatlon at all times when a
treatment plant Is In operation.' Proper process control
7 Dependable removal of fiUrdla cysts can not be guaranteed 1f a water
Is filtered without being properly coagulated (Logsdon, 1987b; A1-Ani
et •!., 1985). This 1s true even 1f the raw water turbidity is less
than 1 NTU.
1 As Indicated In the preamble to the proposed SWTR, 33 percent of the
reported cases of giardiasis In waterbome disease outbreaks were
attributed to Improperly operated filtration plants.
' Some conventional water treatment plants which treat low turbidity
source waters (<1 NTU) reportedly discontinue the application of
coagulant(s) during periods of low turbidity since the raw water already
meets the turbidity MCL. However, studies have shown that cyst removal
for low turbidity waters Is the most difficult to achieve and requires
optimum pretreatment Including coagulation to achieve effective removals
4-15
-------
procedures should be used at the plant to assure that chemical
feeds are adjusted and Maintained 1n response to variations
1n raw water temperature and turbidity.
b. Maintenance of effective filtration will require proper
operation procedures to meet the turbidity requirements of the
SHTR. Proper operation should Include:
Proper chealcal conditioning of the water ahead of the
filter to assure adequate turbidity removal through the
filter.
- Control of the flow rates and elimination of rapid
changes 1n flow rate applied to the filter.
Backwashlng of filters before the filtered water quality
Is degraded to the point that the plant falls to Met
the turbidity requirements of the SHTR. The criteria
on which to base Initiation of backwash will have to be
determined for each plant. Experience with operation
cycles Including run times and headloss data may serve
as the basis for this site specific criteria.
After backwash bringing the clean filters, back on line
so that excessive turbidity spikes 1n the filtered water
are not created. Section 4.3.2.B of this manual
discusses these turbidity spikes and approaches
available to minimize them.
c. Filters removed from service generally should be backwashed
upon start-up. However, In some cases, it may be Impractical
to backwash filters each time they are removed from service.
Accordingly, the Primacy Agency may choose to allow start-up
without backwashlng under certain conditions on a s1te-by-site
basis. In making this decision, the following should be
considered:
. _ • the length of time the filter was off-line
performance of the filter while being put on-line
The filter should be brought back on-line 1n such a way that
no turbidity spikes that could be associated with passage of
(Al-Anl et al.. 1985).
4-16
-------
gfardfa cysts and other pathogens occur. If probleas with
turbidity spikes art found when starting up dirty filters,
special studies should be used to evaluate 1f any of the
approaches discussed 1n Section 4.3.2.B of this aanual are
effective In a1n1a1z1ng the turbidity spikes.
4.3.4 Direct nitration
A direct filtration plant can include several different pretreataent
unit processes depending upon the application. In Its simplest fora, the
process includes only In-Hne filters preceded by chealcal coagulant
application and alxlng. The alxlng step, particularly in pressure
filters, can be satisfied by influent pipeline turbulence. In larger
plants with gravity filters, an open rapld-aix basin with aechanlcal
aixers typically Is used. Figure 4-3 illustrates the unit processes of
a typical direct filtration plant.
Another variation of the direct filtration process consists of the
addition of a coagulant to the raw water followed by rapid alxlng and
flocculatlon, as Illustrated on Figure 4-4. The chealcally conditioned
and flocculated water 1s then applied directly to a dual- or aulti-nedia
filter (USEPA, 1988b).
Pesion Criteria
The 1987 edition of the Ten State Standards recoaaends pilot studies
to deteralne aost design criteria. For the purposes of iapleaentation of
the SVTR this requlreaent is considered sufficient with the following
exception:
a. A coagulant aust be used at all tiaes when the treataent plant
Is in operation.10
10 Optlaua coagulation is critical for effective turbidity and aicroDiolog-
ical removals with direct filtration (Al-Ani et al., 1985).
4-17
-------
flntritlng Riaulrtmenta
Operating considerations for dlrtct filtration plants art assent 1 al-
ly Identical to those for conventional treatMnt plants. The aajor
difference 1s that a direct filtration plant will not have a clarifier,
and aay or My not have a flocculatlon or contact basin. In addition, IPA
recoenends that all direct filtration plants, both new and existing, be
required to Mice provisions to alnlalie the break-In tlM of a filter
being put on-line.11
As with conventional treatMnt, the Initiation of backwashlng a
filter should first be based on filter effluent turbidity values, then by
headloss and run tlM. Effluent turbidity son 1 toring equipMnt should be
set to Initiate filter backwash at an effluent value of 0.5 NTU or less,
In order to Met filtered water quality requlrtMnts. Also, any filters
reaoved froa service should be backwashed upon start-up. In som cases,
It My not be practical to backwash filters every tlM they are reaoved
froa service. This decision should be Mde by the PrlMcy Agency on a
case-by-case basis, based on the sue considerations as for conventional
systeas.
4.3.5 Slow Sand Filtration
Slow sand filters differ froa slngle-aedla rapid-rate filters 1n a
number of laportant characteristics. In addition to the difference of
flow rate, slow sand filters:
a. Function using biological aechanlsas as well as physical-che-
a1ca! aechanlsas
b. Use sMller sand particles
c. Are not backwashed, but rather are cleaned by reaoving the
surface aedla
d. Have auch longer run tlMS between cleaning
11 As with conventional treatMnt, direct filtration produces a relatively
poor quality filtered water at the beginning of filter runs and
therefore a f1lter-to-w"aste period is recoaMnded. In son eases, the
addition, of a filter aid or bringing filters on-line slowly will be
appropriate (Cleasby et al.r 1984).
4-18
-------
COAOULANTI
MfLUCNT
AAFC MIX
M SfC-t MM
DCTBNTICM
PLOCCUlATtOfl
tf«4S MM
t-4 HOURS
DURATION
¦AMD tANO: 2
OUAL A NO TM-MIXip
MIMA: 4-4 MMt'ftl
FIGURE 4-1-PLOW SHEET OP A TYPICAL CONVENTIONAL
WATER TREATMENT PLANT
-------
SINGLE STAGE SOFTENING (1|
LIME
'•ClUINT
¦ AffO Ml
it iic i mm
—*
PlOCCULATlON
QITlNTIOta
mm
[ff-
l«OMSNTATWN
MOUftS
PNJflAflOM
•AMD tAMO; i «¦•'«'
OUAL A NO MUCTI
«COtA *•« mmm t
(I] PH RANGE t-10
(21 ON ALTERNATE SOLIOS REMOVAL PROCESS
TWO STAGE SOFTENING (11
lime
INFLUENT
FLOCCULATOR
CLARIFIER
(1] RH RANGE 10*12
SOOA ASH
RECARBONATION
FLOCCULATOR
CLARIFIER
RECARBONATION
FILTRATION
SOFTENED WATER
FIGURE 4-2-FLOW SHEET OF TYPICAL SOFTENING TREATMENT PLAN1
-------
coagulants
MPi.UiNT<
RAPID MIX
30 Sf C • 2 WM
DETENTION
OUAL Oil MIXED
MEDIA FILTER
4-1 spin ft *
FIGURE 4*3 FLOW SHEET FOR A TYPICAL
DIRECT FILTRATION PLANT
COAGULANTS
INFLUENT'
RAPID MIX
39 SEC • 2 MIN
OETENTION
FLOCCULATION
15-30 MIN
OUAL OR MIXED
MEDIA FILTER
4-9 gpm ft 2
FIGURE 4-4-FLOW SHEET FOR A TYPICAL DIRECT
FILTRATION PLANT WITH FLOCCULATION
-------
e. Require « ripening period «t the beginning of each run
Although rapid rate filtration Is the water treatment technology
used most extensively In the United States, Its use has ofteo proved
Inappropriate for small communities since rapid-rate filtration Is a
technology that requires skilled operation by trained operators. Slow
sand filtration requires very little control by an operator. Consequent-
ly, use of this technology may be more appropriate for snail systems where
source water quality Is within the guidelines reconmended 1n Section
4.2.3.
As Indicated In this section, slow sand filtration also nay be
applicable to other source water quality conditions with the addition of
pretreatment such as a roughing filter or presedlmentatlon.
Design Criteria
The minimum design criteria presented In the Ten State Standards for
slow sand filters are considered sufficient for the purposes of Implemen-
tation of the SWTR with the following exceptions:
a. Raw water quality limitations should be changed to reflect the
values given In Table 4-2.12
b. The effective sand size should be between 0.15am and 0.35mro
rather than the current 0.30 on to 0.45 on.13
Additional guidance on the design of slow sand filtration 1s
available In the design Manual entitled Slow Sand Filtration for Community
Water Supplies Technical Paoer 24. 1987 published by the International
12 Without pretreatment, limitations exist In the auallty of water that
Is suitable for slow sand filtration (Logsdon, 1987b; Cleasby et al.,
1984; Bellamy et al., 1985; Fox et al., 1983).
11 Significant decreases 1n total coUform removals were shown at effective
sand sizes less than 0.35 on (Bellamy et al,, 1985). As defined in the
AWWA Standard for Filtering Material, effective size 1s the size opening
that will pass 10 percent by weight of a sample of filter material.
4-19
-------
Reference Centre for Community Water Supply and Sanitation (IRC),
P.O. Box 5500, 2280 HM Rljswijk, the Netherlands.
Operating Requirements
Maintenance of a slow sand filter involves two periodic tasks:
a. Removal of the top 2 to 3 cm (0.8 to 1.2 inches) of the
surface of the sand bed when the headloss exceeds 1 to 1.5 m.14
b. Replacement of the sand when repeated scrapings have reduced
the depth of the sand to approximately one-half of Its design
depth (Bellamy et al., 1985).
Following scraping, slow sand filters produce poorer quality
filtrate at the beginning of a run, and a filter-to-waste or ripening
period of one to two days 1s recommended before use to supply the system.
The ripening period 1s an Interval of time immediately after a scraped
filter 1s put back on-Hne, when the turbidity or particle count results
are significantly higher than the corresponding values for the operating
filter. During this time, the microorganisms multiply and attain
equilibrium 1n the "schmutzdecke." Filter effluent monitoring results
should be used to determine the end of the ripening period. For example,
a turbidimeter could be set at 1.0 NTU or less to initiate start of the
filter run.
When repeated scrapings of the sand have reduced the depth of the
sand bed to approximately one-half of its design depth, the sand should
be replaced. Filter bed depths of less than 0.3 to 0.5 « (12 to 20
Inches) have been shown to result 1n poor filter performance (Bellamy et
al., 1985). The replacement procedure should Include removal of the
remaining sand down to the gravel support, the addition of new sand to one
half of the design depth and placement of the sand previously removed on
top of the new sand."
14 Removal of this top layer of the "Schmutzdecke" should restore the
filter to Its operational capacity and initial headloss.
15 This procedure results in clean sand being placed 1n the bottom half
of the filter bed and biologically active sand in the top half reducing
the amount of time required for the curing period. It also provides
4-20
-------
The Mount of time for the biological population to Mature In a new
sand filter (also called curing) and to provide stable and full treatment
varies. The World Health Organization (1980) reported that curing
requires a few weeks to a few months. Fox et al., (1983) found that
"about 30 days" were required to bring particle and bacterial effluents
down to a stable level. All researchers agree that a curing time for a
new filter Is required before the filter operates at Its fullest potential
(Bellamy et al., 1985).
4.3.6 Olatomaceous Earth Filtration
Dlatomaceous earth (DE) filtration, also known as precoat or
dlatoalte filtration, 1s appropriate for direct treatment of surface
waters for removal of relatively low levels of turbidity and microorgan-
isms.
Dlatoalte filters consist of a layer of 0E about 3 am (1/8 Inch)
thick supported on a septum or filter element. The thin precoat layer of
0E must be supplemented by a continuous body feed of dlatoalte, which 1s
used to maintain the porosity of the filter cake. . If no body feed is
added, the particles filtered out will build up on the surface of the
filter cake and cause rapid Increases In headless. The problems Inherent
1n maintaining the proper film of DE on the septum have restricted the use
of diatomite filters for municipal purposes, except under certain
favorable raw water quality conditions, I.e., low turbidity and good
bacteriological quality. Specific upper limits of raw water quality
parameters are not well-defined because dlatomaceous earth filtration
performance depends on the nature, as well as the concentration, of the
raw water particles and the grades of dlatoalte employed. Logsdon (1987b)
reported that filtered water turbidities above 1 NTU and short filter runs
were observed for several diatooaceous earth plants having maximum raw
water turbidities above 20 NTU.
for a Complete exchange of sand over time, alleviating potential
problems of excessive silt accumulation and clogging of the filter bed
(Bellamy et al., 1985).
4-21
-------
Peiion CritiHi
The minimum design criteria presented 1n the Ten State Standards for
dlatomaceous earth filtration are considered sufficient for the purposes
of compllance with the SWTR with the following exceptions:
a. The recommended quantity of precoat Is 1 kg/ra2 (0.2 pounds per
square foot) of filter area, and the minimum thickness of the
precoat filter cake 1s 3m to 5on (1/8 to l/5-1nch).u
b. Treatment plants should be encouraged to provide a coagulant
coating (alum or suitable polyner) of the body teed.17
Operating Requirements
Operating requirements specific to 0E filters Include:
Preparation of body feed and precoat
Verification that dosages are proper
Periodic backwashlng and disposal of spent filter cake
Periodic Inspection of the septun(s) for cleanliness or damage
Verification that the filter Is producing a filtered water
that meets the performance criteria
4.3.7 Alternate Technologies
The SWTR allows the use of filtration technologies other than those
specified above provided that the system demonstrates to the Primacy
Agency using pilot studies or other means that the filtration technology
when combined with disinfection achieves at least 3-1og ei^rdia cyst and
4-log virus removal/inactlvatIon. Such technologies must also meet the
turbidity performance criteria for slow sand filtration. Guidance for
11 Studies have shown that a precoat thickness of 1 kg/m2 (0.2 lbs/ft2) was
most effective in eiardla cyst removal and that the precoat thickness
was more Important than the grade size 1n cyst removal (DeWalle et al.,
1984; Logsdon et al., 1981; Bellamy et al., 1984).
17 Although enhancement of the 0E is not required for GUrdia cyst removal,
coagulant coating of the body feed has been found to significantly
Improve removals of viruses, bacteria and turbidity. (Brown et al.,
1974; Bellamy et al., 1984).
4-22
-------
conducting pilot studios to demonstrate this effectiveness 1s provided in
Appendix N of this manual.
Reverse osmosis Is a membrane filtration method which 1s used for
desalination and/or the removal of organic contaminants. The treatment
processes effective for the removal of Giardia cysts and viruses and no
demonstration Is necessary.
Alternate filtration technologies which are currently available
Include, but are not limited to:
Package Plants
Cartridge Filters
Package plants 1n principle are not a separate technology from the
preceding technologies. However, in many cases they are different enough
1n design criteria, and operation and maintenance requirements that they
should be considered as an alternate technology. The package plant 1s
designed as a factory-assembled, skid-mounted unit generally incorporating
a single, or at most, a few process units. A complete treatment process
typically consists of chemical coagulation, flocculatlon, settling and
filtration. Package plants generally can be applied to flows ranging from
about 25,000 gpd to approximately 6 mgd (USEPA, 1968b). In cases where
the Primacy Agency believes that the design criteria of the package plant
corresponds to the design criteria of the processes established earlier
in this section (I.e., that the package plant qualifies as conventional
or direct filtration), the requirement of pilot testing may be waived.
The application of cartridge filters using either cleanable ceramic
or disposable polypropylene cartridges to small water systems may be a
feasible method for removing turbidity and some microbiological contami-
nants, such as Siardia cysts although no data are available regarding
their ability to remove viruses. Pilot studies are required to demon-
strate the efficacy of this technology for a given supply. However, 1f
the technology was demonstrated to be effective through pilot plant
studies at one site, then the technology could be considered to be
effective at another site which had similar source water quality
conditions. Therefore, pilot plant testing at the new site might not be
necessary.
4-23
-------
It Is Important to note that the demonstration of achieving the 3-
log filirdli cyst and 4-1og virus removal/inactivatlon requirements
Includes disinfection. Thus, If a cartridge filter Is demonstrated to
achieve a 3-log removal of Giardia cysts and It Is determined by CTs that
the disinfection achieves at least a 4-log virus inactlvation, the
effectiveness of the technology would be deaonstrated. The technology
oust also Maintain turbidities less than 1 NTU In 95 percent of the
monthly samples. Meeting this turbidity requirement assures a high
probability that turbidity will not Interfere with disinfection and that
the 1nact1vat1on efficiencies predicted by the CTs are reliable.
Design Criteria
After any necessary pilot studies are conducted and a successful
demonstration of performance has been made, design criteria should be
established and approved by the Primacy Agency. Eventually, a sufficient-
ly large data base will become available, making It easier to apply the
alternate technologies to other water supplies of similar quality.
Operating Requirements
After any necessary pilot studies are conducted and a successful
demonstration of performance has been made, operating requirements should
be established and approved by the Primacy Agency.
4.3.8 Nontreatment Alternatives
Under certain circumstances, some systems may have other alterna-
tives available. These alternatives Include regionalIzatlon and the use
of alternate sources.
For small water systems which must provide filtration, a feasible
option may be to join with other small or large systems to form a region-
al water supply system. In addition, alternative water sources located
within a reasonable distance of a community which would allow the system
to meet the requirements of the SWTR and other applicable drinking water
regulations, may be developed to provide a satisfactory solution to a
community water quality problem. The availability of alternative ground
4-24
-------
water sources will depend upon the size and location of the system and
the costs Involved.
4.4 Disinfection
4.4.1 general
The SWTR requires that disinfection be Included as part of the
treatnent of surface water for potable use. As noted earlier, EPA
recoanends that the nunber of treatment barriers be coaaensurate with the
degree of contaalnatlon In the source water In accordance with Table 4-2.
For example, as Indicated 1n Table 4-2, when the total collforns 1n the
source water are greater than 5,000/100 ml, conventional treataent with
predlslnfectlon 1s recommended. However, the selection of appropriate
disinfection requires consideration of other factors In addition to than
those Included 1n Table 4-2. These considerations Include:
a. Source water quality and the overall reaoval/lnactlvatlon of
fiilEdii cysts and viruses desired.
b. Likelihood of TTHM formation.
c. Potential need for an oxidant for purposes other than
disinfection Including control of taste; odor, Iron,
manganese, color, etc.
4.4.2 Recommended Removal/Inactlvatlon
The SWTR requires a minimum 3-1og reaoval/lnactlvatlon of Giardla
cysts and a alnlaua 4-1og reaoval/lnactlvatlon of viruses:
a. Well-operated conventional treataent plants which have been
optlalzed for turbidity reaoval can be expected to achieve at
least a 2.5-1og removal of 61ard1a cysts.
b. Hell-operated dlatomaceous earth, slow sand filtration and
direct filtration plants can be expected to achieve at least
2-log reaoval of filardla cysts.
EPA recoanends that:
a. Conventional filtration systems provide sufficient disinfec-
tion to achieve a mini nun of 0.5-1og Giardla cyst and 2-1og
virus 1nact1vat1on.
4-25
-------
b. Slow sand filtration systems provide sufficient disinfection
to achieve a minimum of 2-log 61ard1a cyst and 2-log virus
Inactlvatlon.
c. Systens using dtatomaceous earth and direct filtration, or
other filtration methods, should provide sufficient disinfec-
tion to achieve a minimum of 1-log filardl* cyst and 3-log
virus 1nact1vat1on.
Further guidance on the disinfection level to be provided Is
contained In Section 5. CT values for achieving these 1nact1vat1ons are
presented In Appendix E. As Indicated 1n this Appendix:
a. A comparison of Tables E-l through E-6 with Table E-7
Indicates that systens which achieve a 0.5-log Inactlvatlon
of 61*rd1a cysts, using free chlorine, will achieve greater
than a 4-log Inactlvatlon of viruses.
b. Ozone and chlorine dioxide are generally more effective at
Inactivating viruses than 61ard1a cysts. However, as
Indicated 1n Tables E-8 through E-U, there are some
conditions under which the disinfection needed to provide the
recommended virus Inactlvatlon 1s higher than that needed for
the recommended S1ard1a cyst Inactlvatlon. Therefore, a
system using ozone or chlorine dioxide for disinfection must
check the CT values needed to provide the recommended
Inactlvatlon of both Glardla cysts and viruses and provide the
higher of the two disinfection levels. Systems may demon-
strate their efficiency for overall removal/Inactlvatlon using
the protocol in Appendices G and M.
c. As indicated in Tables E-12 and E-13, chloraalnes are much
less effective for Inactivating 61ard1a cysts and viruses than
the other disinfectants. Also, chloraalnes may be applied to
the system in several ways, either with chlorine added prior
to ammonia, ammonia added prior to chlorine or preformed. For
systems applying chlorine ahead of aononla, the required level
of disinfection may be determined as follows:
determine the CT needed to provide the required
Inactlvatlon of 61ard1a and viruses and provide the
higher of the two levels or
4-26
-------
follow the protocol In Appendix 6 to demonstrate
effective Inactlvatlon to allow lower levels of
disinfection.
For systems applying ammonia ahead of chlorine or preformed
chloramlnes, the EPA recommends that the system demonstrate
effective virus Inactlvatlon according to the protocol 1n
Appendix 6, since the CT values for virus Inactlvatlon In
Table E-13 only apply to the addition of chlorine prior to
ammonia.
Although the SWTR requires a minimum of a 3-1og removal/Inactlvatlon
of Giardi* cysts and a minimum of a 4-1 og removal/Inactlvatlon of viruses,
1t may be appropriate for the Primacy Agency to require greater removals/-
1nactivat1ons depending upon the degree of contamination within the source
water.
Rose (1988) conducted a survey of water sources to characterize the
level of Glardla cyst occurrence for "polluted" and "pristine" waters.
Polluted waters are defined as waters In the vicinity of sewage and
agricultural wastes, while pristine waters are those originating from
protected watersheds with no significant sources of microbiological
contamination from human activities. EPA believes that treatment should
be provided to assure less than one case of microbiologically-caused
Illness per year per 10,000 people. In order to provide this level of
protection, 3, 4 or 5-1og Glardla cyst remova1/1nact1vat1on should be
provided for the following source water qualities:
Glardla Cyst Removal/Inactlvation Required Based11,1'
on Source Water Cvst Concentration
Glardla Inactlvatlon 3-10? 4-lop 5-loo
Allowable daily avg
cyst concentration/100 L <1 >1-10 >10-100
(geometric mean)
11 Rose, 1988.
19 10"4 annual risk per person based on consumption of 2 liters of water
dally.
4-27
-------
According to these guidelines, systems with sewage and agricultural
discharges to the source water should provide treatment to achieve an
overall S-log removal/1nact1vat1on of 61ardl* cysts, while the minimum
required 3-1og removal/InactlvatIon Is sufficient for sources'wlth no
significant microbiological contamination from human activities. A 4-log
removal/lnactlvatlon of cysts should be provided for source waters whose
level of microbiological contamination 1s between these two extremes. The
location of discharges or other activities polluting the water-supply with
respect to the location of the Intake should also be considered In
determining the level of removal/lnactlvatlon needed. For Instance, long
travel times and substantial dilution of a discharge will lessen the
Impact of the discharge on the Intake water quality, In which case less
of an Increase 1n the overall treatment recommended above, would be
warranted. It Is Important to note that these levels of treatment for
different generalized source water characterizations are presented only
as guidelines. The Primacy Agency could develop disinfection requirements
based on these or other guidelines. It could also require systems with
available resources to conduct raw water monitoring for Glardla cyst
concentrations to establish the appropriate level of overall treatment and
disinfection needed.
The Primacy Agency may also review the nature of occurrence of
Glardia-slzed particles 1n the raw water supply and the association with
turbidity occurrence. If 1t can be demonstrated that a higher degree of
removal of particles 1n the size range of filardla 1s accomplished when
turbidity levels and associated Glardla levels are elevated, then a log
removal credit higher than 3 could be allowed for that particular
treatment plant, during such occurrences. This credit should correspond
to the log particle removal efficiencies accomplished, as determined by
particle counting data, or turbidity data 1f properly qualified. In all
cases, a minimum of 0.5 log reduction of G1ard1a should be achieved by
disinfection 1n addition to the removal credit allowed for by other
treatment.
4-28
-------
Until a risk analysis for exposure to viruses Is developed, a rough
guideline for virus removal/1nactivat1on, can be considered as follows:
a. For a 4-log Glardia cyst removal/Inactlvat1on, a 5-log virus
removal/Inactivat1on 1s recommended.
b. For 5-1og G1ard1a removal/1nact1vat1on, a 6-log virus
removal/lnactlvatlon Is recommended.
These guidelines assume that virus occurrence In the source water
1s roughly proportional to Giardia cyst occurrence, and that
viruses occur at higher concentrations 1n source waters, or
are more Infectious than 61ardia cysts and
Infections from viruses may have more health risk significance
than Stardla cysts.
Based on these assumptions, higher levels of protection are warranted.
To meet the levels of Inactlvatlon recommended here, significant
changes 1n the system may be required. To avoid changes 1n the system
which may result In conflicts with future regulations, the Primacy Agency
may wish to establish Interim disinfection levels to provide protection
of the public health prior to the promulgation of the disinfection
by-product regulations and then reconsider whether these levels are still
appropriate after the disinfection by-product regulations are promulgated.
Guidance for establishing Interim disinfection requirements is provided
In Section 5.5.
4.4.3 Total Tribalomethane (TTHm Regulations
In addition to complying with disinfection requirements, systems
must meet the requirements of the TTtM regulations. Currently, this
regulation Includes an HCL for TTHMs of 0.10 mg/L for systems which serve
greater than 10,000 people. EPA expects to issue new regulations with a
lower MCI In the near future. These regulations may also pertain to
systems serving less than 10,000 people. Therefore, the selection of an
appropriate disinfectant or disinfection strategy must Include consid-
eration of current and future regulations.
4-29
-------
5. CRITERIA FOR DETERMINING IF FILTRATION
AND DISINFECTION ARE SATISFACTORILY
5.1 Introduction
Under the SWTR, new and existing filtration plants wist meet
specified monitoring and performance criteria in order to assure that
filtration and disinfection are satisfactorily practiced. These criteria
include:
Turbidity monitoring requirements
Turbidity performance criteria
Disinfection monitoring requirements
Disinfection performance criteria
The overall objective of these criteria is to provide control of:
Giardia cysts; viruses; turbidity; HPC; and Legionella by assuring a high
probability that:
a. Filtration plants are well-operated and achieve maximum
removal efficiencies of the above parameters.
b. Disinfection will provide adequate inactivation of Giardia
cysts, viruses, HPC and Legionella.
5.2 Turbidity Monitoring RMulrwrcnU
5.2.1 Sampling Location
The purpose of the turbidity requirements for systems which use
filtration is to indicate:
a. Giardia cyst and general particulate removal for conventional
treatment and direct filtration
b. General particulate removal for diatomaceous earth filtration
and slow sand filtration
c. Possible interference with disinfection for all filtration
processes
To accomplish the purposes of the turbidity requirements, the SWTR
requires that the turbidity samples be representative of the system's
5-1
-------
filtered water. The sampling locations which would satisfy this
requiresent include:
a. Combined filter effluent prior to entry into a clear-well,
b. Clearwell effluent;
c. Plant effluent or immediately prior to entry into the distri-
bution system; or
d. Average of measurements from each filter effluent.
The selection of sampling locations for demonstrating compliance
with the turbidity performance criteria is left to the system or the
preference of the Primacy Agency.
5.2.2 Samolino Frequency
The turbidity of the filtered water must be determined:
a. At least once every four hours that the system is in opera-
tion, or
b. The Primacy Agency may reduce the sampling frequency to once
per day for systems using slow sand filtration or filtration
treatment other than conventional treatment, direct filtration
or diatomaceous earth filtration, if it determines that less
frequent monitoring is sufficient to indicate effective
filtration performance. For systems serving 500 or fewer
people, the Primacy Agency may reduce the sampling frequency
to once per day regardless of the type of filtration used if
' it determines that less frequent monitoring is sufficient to
indicate effective filtration performance.
A system may substitute continuous turbidity monitoring for grab
sample monitoring if it validates the continuous measurement for accuracy
on a regular basis using a protocol approved by the Primacy Agency. EPA
recommends that the calibration of continuous turbidity monitors be
verified at least twice per week according to the procedures established
in Method 214A of the 16th Edition of Standard Methods.1
1 Although the 17th Edition of Standard Methods is available, the 16th
Edition is referred to in the SWTR. Continuous turbidity monitors must
be installed properly to prevent air bubbles from reaching the monitor.
-------
5.2.3 Additional Monitoring
As Indicated In Section 4.3.2, EPA reconmends that systems equip
each filter with a continuous turbidity monitor. This recommendation Is
not part of the requirements of the SWTR and 1s not required for
establishing compliance. Rather, It Is recommended as a tool for'systems
to use to better monitor their treatment efficiency and to provide a
method for detecting a deterioration 1n filter performance.
If continuous monitoring of each filter effluent cannot be
Implemented, then EPA recommends that at least the following be conducted
on a quarterly basis:
a. Monitor each filter, either by grab samples or continuous
monitors, through the course of a routine cycle of operation,
i.e.: from restart to backwash
b. Visually Inspect each filter where appropriate for Indications
of physical deterioration of the filter
These are general suggestions. The Primacy Agencies are encouraged
to work with the systems to determine the best overall monitoring
program(s) for their particular filtration plants In order to assess the
status of the filter units. Each filter within a system should be
maintained so that each filter effluent meets the turbidity performance
criteria for the combined filter effluent (I.e., the turbidity limits
specified In the SWTR).
5.3 Turbidity Performance Criteria
The SWTR establishes turbidity performance criteria for each of the
filtration technologies. As previously indicated, these criteria provide
an indication of:
a. Effective particle and microbial removal
b. Potential for Interference with disinfection
In filtration, effective particle removal depends on both physical
and chemical factors. The particles to be removed must be transported to
the surface of the media and they must attach to the media. When
efficient particle removal does not occur, the deterioration of filter
5-3
-------
performance,can be due to either physical problems with the filters or
problems with the treatment chemlsjtry.
Physical problems which can result In a deterioration of filter
performance Include:
a.
Media loss
b.
Media deterioration
c.
Mud ball formation
d.
Channeling or surface cracking
e.
(Jnderdraln failure
f.
Cross-connections
In addition, the treatment chemistry has a significant Impact on
filtration. Specifically, effective particle removal 1s a function of
the:
a. Raw water chemistry and the changes Induced by the chemicals
added
b. Surface chemistry of the particles to be removed
c. Surface chemistry of the media
Consequently, when a filter experiences particle or turbidity breakthrough
prior to the development of terminal headloss, the search for alternatives
to correct the problem must Include not only an evaluation of the
potential physical causes but the treatment chemistry as well. Generally
this involves an evaluation of one or more of the following:
a. Alternate coagulant type and/or dose
b. Alternate coagulant aid or flocculant aid type and/or dose .
c. Need for an alternate oxidant type and/or dose
d. Need for a filter aid or alternate dose
5-4
-------
5.3.1 Conventional Treatment or Pir»et mtratlnn
The Minimum turbidity performance criteria for systems using
conventional treatment or direct filtration are:
a. Filtered water turbidity must be less than or equal to 0.5 NTU
1n 95 percent, of the measurements taken every month.
b. Filtered water turbidity levels of less than or equal to 1 NTU
In 95 percent of the measurements taken every month nay be
permitted on a case-by-case basis If the Primacy Agency
determines that the system (filtration with disinfection) is
capable of achieving the minimum overall perforaance require-
ments of 99.9 percent removal/inactlvatlon of Glardia cysts at
the higher turbidity level. Such a determination could be
based upon an analysis of existing design and operating
conditions and/or performance relative to certain water quail-
ty characteristics. The design and operating conditions to be
reviewed Include:
the adequacy of treatment prior to filtration,
the percent turbidity removal across the treatment
train, and
level of disinfection.
Hater quality analysis which may also be used to evaluate the
treatment effectiveness Include particle size counting before
and after the filter. Pilot plant challenge studies simulat-
ing full scale operation may also be used to demonstrate
effective treatment. Depending on the source water quality
and system size, the Primacy Agency will determine the extent
of the analysis and whether a pilot plant demonstration 1s
needed. For this demonstration, systems are allowed to
Include disinfection in the determination of the overall
performance by the system.
c. Filtered water turbidity may not exceed 5 NTU at any time.
The Primacy Agency can assume that conventional treatment plants
that are meeting the minimum performance criteria are achieving at least
a 2.5-log removal of Glardia cysts and at least a 2-log removal of viruses
prior to disinfection.1
2 Recommended protocol for this demonstration 1s presented 1n Appendix M.
1 The literature Indicates that well operated conventional treatment
plants can achieve up to 3-log reduction of Glardia cysts and viruses
(Logsdson, 1987b and Roebeck et al.( 1962). Llmitina the credit to
2.5-logs for Glardia cysts and 2-logs for viruses provides a margin of
safety by requiring more disinfection. This 1s consistent with the
5-5
-------
The Primacy Agency can assume that direct filtration plants that are
¦eetlng the minimum performance criteria are achieving at least a 2-1og
removal of Glardla cysts and a 1-log removal of viruses.4
Although the minimum turbidity performance criterion allows for a
maximum filtered water turbidity of 0.5 NTU, treatment facilities using
conventional treatment or direct filtration, whose raw water supplies have
turbidity levels of 1 NTU or less, should be encouraged to achieve
filtered water turbidity levels of less than 0.2 NTU.S
Primacy Agencies may allow systems which believe that they are
actually achieving greater than a 2- or 2.5-1og Giardia cyst removal to
demonstrate the actual removal achieved using the protocol outlined 1n
Appendix M. It 1s reasonable to expect that systems using conventional
treatment for high turbidity source water (e.g., turbidities 1n excess of
100 NTU), and which optimize chemical treatment prior to filtration, may
be achieving a 3-log or greater Glardla cyst removal if their filter
effluent is substantially below the 0.5 NTU turbidity limit. Softening
plants using conventional processes and 2-stage treatment processes may
also achieve a 3-log Giardia cyst removal/1nactivat1on. The high pH of
softening may result 1n inactivation of Giardia cysts and viruses which
can be demonstrated according to the protocol outlined in Appendix G.
Appendix M can be used to demonstrate the Giardia cyst removal achieved.
multiple barrier concept.
4 Literature Indicates that well operated direct filtration plants can
achieve, up to a 3-log removal of Giardia cysts and up to a 2-log
removal of viruses (Logsdon, 1987b; Roebeck et al•, 1962). Limiting
the credit to 2-log for Giardia cysts and 1-log for viruses provides a
margin of safety by requiring more disinfection. This Is consistent
with the multiple barrier concept.
* Research has demonstrated that filter effluent turbidities substantial-
ly lower than 0.5 NTU are needed to obtain effective removals of
Giardia cysts and viruses with low turbidity source waters (Logsdon,
1987b; Al-Ani et al., 1985).
5-6
-------
5.3.2 Slow Sand Filtration
For systems using slow sand filtration, the turbidity performance
requirements are:
a. The filtered water turbidity must be less than or equal to
I NTU In 95 percent of the measurements for each mon.th.
b. At the discretion of the Primacy Agency, a higher filter
effluent turbidity may be allowed for well operated plants
(Section 4.3.5) on a case-by-case basis, It there Is no
interference with disinfection and the turbidity level never
exceeds 5 NTU. Noninterference with disinfection could be
assumed If the finished water entering the distribution system
Is meeting the conform MCL and HPC levels are less than 10/ml
during times of highest turbidity.
c. Filtered water turbidity may not exceed 5 NTU at any time.
Slow sand filtration plants, with appropriate design and operating
conditions and which meet the minimum turbidity performance criteria can
be considered to be well operated and achieving at least a 2-1og removal
of Glardla cysts and 2-log removal of viruses without disinfection.1
Primacy Agencies may allow systems which believe that they are actually
achieving greater than a 2-log Glardia cyst removal to demonstrate the
actual removal achieved using the protocol outlined in Appendix M.
5.3.3 piatofMCMus Earth filtration
For systems using dlatomaceous earth filtration, the turbidity
performance criteria are:
a. The filtered water turbidity must be less than or equal to
1 NTU In 95 percent of the measurements for each month.
b. The turbidity level of representative samples of filtered
water must at no time exceed 5 NTU.
Dlatomaceous earth systems, with appropriate design and operating
conditions and which meet the minimum turbidity performance criterion can
1 As Indicated 1n Section 4, pilot studies have shown that with proper
nurturing of the schmutsdecke, operation at a maximum loading rate of
0.2 m/hr will provide optimum removal of Giardia cysts ana viruses
(Logsdon, 1987b; Bellamy et al., 1985).
-------
be considered to be well operated and achieving at least 2-log removal of
Giardia cysts and at least 1-log removal of viruses without disinfection.
Systems which believe that they are actually achieving greater than a 2-
log Giardia cyst removal may demonstrate the actual removal achieved using
the protocol outlined In Appendix M.
5.3.4 Other filtration Technologies
The turbidity performance criteria for filtration technologies other
than those presented above, are the same as for slow sand filtration. The
Giardia cyst removal achieved by these systems must be demonstrated to the
Primacy Agency. The protocol outlined in Appendix H nay be used as a
basis for this demonstration.
Reverse osmosis is a membrane filtration method used to remove
dissolved solids from water supplies. Desalination is a typical use of
the process. Application to potable water treatment 1s limited to
extremely high quality raw water supplies of low turbidity (1 NTU or
less), or following pretreatment to produce a supply of low. turbidity.
The membrane excludes particles larger than 0.001 to 0.0001 um
range, thereby effectively removing bacteria, Slflrdti cysts and viruses.
Credit can be given for at least a 3-1 og Giardia cyst and 4-1 og virus
removal, with no demonstration. It should be noted that this removal
credit assumes the membranes are in tact with no holes 1n the membranes
allowing the passage of organisms.
5.4 Disinfection Monitoring Requirements
Each system must continuously monitor the disinfectant residual of
the water as It enters the distribution system and record the lowest
disinfectant residual each day. If there is a failure In the continuous
monitoring equipment, the system may substitute grab sample monitoring
every 4 hour's for up to 5 working days following the equipment failure.
Systems serving 3300 people or fewer may take.grab samples In 11 eu of
continuous monitoring at frequencies as follows:
5-8
-------
Syiteffl Population
£500
501-1,000
1,001 - 2,500
2,501 - 3,300
Samples/Dav
2
3
4
The grab samples must be taken at different tines during the day,
with the sampling Intervals subject to Primacy Agency review and approvals
If the residual concentration falls below 0.2 mg/L, the system must take
another sample within 4-hours and notify the Primacy Agency as soon as
possible, but no later than the end of the next business day, even If the
residual Is restored to 0.2 mg/L or greater within 4 hours. If the
residual 1s not restored to 0.2 mg/L or greater within 4 hours, the system
Is 1n violation of a treatment technique requirement. Each system must
also measure the disinfectant residual 1n the distribution system at the
same frequency and locations at which total coHform measurements are made
pursuant to the requirements 1n the revised Total Collform Rule (54 FR
27544; June 29, 1989)s For systems which use both surface and ground
water sources, the Primacy Agency may allow substitute sampling sites
which are more representative of the treated surface water supply.
5.5 Disinfection Performance Criteria
5.5.1 Minimum Performance Criteria Required bv the SWTR
For systems which provide filtration, the disinfection requirements
of the SWTR are:
a. Disinfection must be provided to ensure that the total
treatment processes of the system (Including filtration)
achieves at least a 3-1 og removal/lnactivation of Giardia cyst
and a 4-1og removal/1nact1vat1on of viruses. The Primacy
Agency must determine what level of disinfection is required
for each system to meet this criterion.
b. The system must demonstrate by continuous monitoring and
recording that a disinfectant residual in the water entering
the distribution system 1s never less than 0.2 mg/L for more
than 4 hours. If at any time the residual falls below 0.2
mg/L for more than 4 hours the system is 1n violation. The
5-9
-------
system must notify the Primaey Agency whenever the residual
falls below 0.2 mg/L before the end of the next business day.
c. The system nust demonstrate detectable disinfectant residuals
or HPC levels of 500 or fewer colonies/ml in at least 95
percent of the samples from the distribution system each month
for any two consecutive months.
5.5.2 Recommended Performance Criteria
Disinfection must be applied to assure that the overall treatment
provided achieves at least a 3-log removal/inactivatlon of Giardia cyst
and a 4-log removal/inactivation of viruses. As outlined in Section 5.3,
well operated filter plants achieve at least a 2 to 2.5-log removal of
fiiardia cysts and between a 1 to 2-log removal of viruses. IPA therefore
recommends that the Primacy Agencies adopt additional disinfection perfor-
mance criteria that include;
a. As a minimum, primary disinfection requirements that are
consistent with the overall treatment requirements of the
SWTR, or preferably,*
b. Primary disinfection requirements as a function of raw water
quality as outlined in Section 4.4.
Recommended Minimum Disinfection
The required minimum primary disinfection is the disinfection
needed for the entire treatment process to meet the overall treatment
requirement of 3-log Siardia and 4-1og virus removal/inactivation. The
following table provides a summary of the expected minimum level of
treatment performance In well operated filter systems and the recommended
level of disinfection.
Expected
Lo
-------
In cases where the systen believes that the treatment processes are
achieving greater removals than those listed above, the actual removal
provided by the processes can be demonstrated through the procedures
outlined In Appendix H. However, EPA recommends that, despite the
removals demonstrated, systems should provide a minimum of 0.5 log G1ard1a
cyst Inactlvatlon to supplement filtration and maintain a second treatment.
barrier for microorganisms.
P»eomnended Disinfection as a Function of Raw Water Quality
Although the SWTR requires the overall treatment to provide a
minimum of a 3-1 og Giardla cyst and a 4-log virus removal/lnactlvatlon, It
may be appropriate for the Primacy Agency to require greater removals/-
1nact1vat1ons depending on the degree of contamination In the source water
as presented In Section 4.4. Following 1s a summary of the recommended
overall treatment which should be provided based on an estimate of the
Glardla cyst concentration 1n the source water:
Allowable dally avg
cyst concentration/100 L
fgeometric mean! Sl_ AzlH ?>1Q-IQ0
Glardla cyst Removal/Inactivation 3-log 4-log 5-1og
Virus Removal/lnactlvatlon 4-log 5-1og 6-log
If a slow sand filtration plant must achieve a 4-log removal/lnactl-
vatlon of Glardla cysts and a 5-1og removal/1nact1vat1on of viruses, and
credit for 2-log Glardla cyst and 2-log virus removal by filtration is
granted, disinfection for a 2-log Glardla cyst Inactlvatlon and 3-1og
virus Inactlvatlon would be needed to meet the overall removal/inacti-
vate on. However, Primacy Agencies may allow systems which use particle
size analysis, outlined In Appendix N to demonstrate greater than a 2-log
Glardla cyst removal to provide less than 2-log Giardia cyst Inactivation
through disinfection.
5-11
-------
5.5.3 Disinfection By-Product Considerations
Although the EPA suggests increased levels of disinfection for
various source water conditions, a utility should not implement such a
change without considering the potential conflict with the requirements of
existing or future disinfection by-product regulations. EPA intends to
promulgate National Primary Drinking Water Regulations to regulate levels
of disinfectants and disinfection by-products when it promulgates
disinfection requirements for ground water systems (anticipated In 1992).
EPA is concerned that changes required in utilities' disinfection
practices to meet the recommended inactlvations for the SWTR might be
inconsistent with treatment changes needed to comply with the forthcoming
regulations for disinfectants and disinfection by-products. For this
reason, EPA recommends that Primacy Agencies exercise discretion,
sensitive to potential disinfection by-product concerns, in determining
the level of disinfection needed for filtered systems to meet the overall
treatment requirements specified in the rule or recommended based on
source water quality.
Until the promulgation of the disinfection by-product regulation,
EPA recommends that the Primacy Agency allow more credit for Giardia cyst
and virus removal by filtration than otherwise recommended if a) the
Primacy Agency determines that a system is not currently at a significant
risk from microbiological contamination at the existing level of
disinfection and b) less stringent Interim disinfection conditions are
necessary for the system to modify its disinfection process to optimally
achieve compliance with the SWTR as well as the forthcoming disinfection
by-product regulations. The following paragraphs outline the recommended
disinfection levels for systems meeting the above conditions.
For well-operated conventional filtration plants that meet the
minimum turbidity requirements at all times, the Primacy Agency may
consider giving the system credit for 3-1og Giardia cvst removal (in lieu
of the generally recommended 2.5-log credit). Also, for well-operated
direct filtration plants, the Primacy Agency may consider giving the
system credit for 2.5-log Giardia cyst removal in lieu of the generally
5-12
-------
recomended 2.0-log credit. EPA recomends that these additional credits
be given for conventional or direct filtration only if:
a. The total treatment train achieves 1) at least 99 percent
turbidity removal, or filtered water turbidities are consis-
tently less than 0.5 NTU, whichever 1s lower/ fir 2) a 99.9
percent removal of particles in the size range of 5 to 15 urn
Is demonstrated as outlined In Appendix M;f and
b. The level of heterotrophic plate count (HPC) bacteria In the
finished (disinfected) water entering the distribution system
Is consistently less than 10/ml.
Systems using slow sand filtration or dlatoaaceous earth filtration
may be given interim credit for up to 3-log Giardia cyst removal If the
system meets the recommended conditions listed above for conventional
systems. Pilot plant studies have demonstrated that these technologies,
when well operated, generally achieve at least 3.0-1og removals (USEPA,
1988a).
The EPA believes that interim level of disinfection requirements may
be appropriate In some cases depending upon source water quality,
reliability of system operation and potential increased health risks from
disinfection by-products. EPA Intends to regulate disinfectants and
disinfection by-products In 1992. At this time 1t will become apparent
how systems with disinfection by-product problems can optimally meet the
disinfection requirements of the SWTR and the disinfection by-products
regulations, concurrently.
7 For example, a system with a raw water turbidity averaging 20 NTU
maintaining a filtered water turbidity less than 0.2 NTU can be granted
3-1og Giardia cyst removal credit with no further demonstration.
1 In cases where the Primacy Agency has a data base which shows a
correlation between turbidity and Giardia cysts removal, turbidity may
be used 1n lieu of particle size analysis. Turbidity removal require-
ments should be set to assure 99.9 percent Giardia cyst removal. A
correlation between turbidity and Giardia cyst removal was shown in a
study reported by Hendricks et al (1984).
5-13
-------
5.5.4 Recommended Disinfection Svst^ Rft^mrtflnrY
The SWTR does not require a redundant disinfection system for
filtered supplies. However, In order to assure the continuous provision
of disinfection to meet the overall removal/lnactlvatlon requirements and
to maintain a residual entering the distribution system, EPA recomnends
that redundant disinfection equipment be provided. As contained In the
1907 edition of Ten State Standard*. where disinfection Is required for
protection of the supply, standby equipment is required. Automatic
switchover should be provided as needed, to assure continuous disinfectant
application.
Recommendations for providing redundant disinfection are outlined in
Section 3.2.4 and detailed in Appendix I.
5.5.5 Determination of Inactivation hv Disinfection
The desired level of inactivation can be achieved by disinfection at
any point in the treatment or distribution system prior to the first
customer. Disinfection provided prior to filtration Is referred to as
pre-disinfection while disinfection after filtration 1s referred to as
post-disinfection. As presented 1n Section 3.2, the Inactivation of
Giardia cysts and viruses provided by disinfection are Indicated by CT
values*
The SWTR defines CT as the residual disinfectant concentration(s) 1n
mg/l multiplied by the contact time(s) in minutes. The contact time 1s
measured from the point of disinfectant application to the point of
residual measurement or between points of residual measurement. The
inactivation efficiency can be determined by calculating CT at any point
along the process after disinfectant application prior to the first
customer.
A system may determine the Inactivation efficiency based on one
point of residual measurement prior to the first customer, or on a profile
of the residual concentration after the point of disinfectant application.
The residual profile is generated by monitoring the residual at several
points between the po1nt(s) of disinfectant application and the first
customer. The system can then use the method described 1n Section 3.2 for
determining the total inactivation credit. Profiling the residual allows
5-14
-------
for credit of significantly higher residuals which aay exist before the
water reaches the first customer. Methods for determining various
disinfectant residuals are described In Appendix D.
In pipelines, the contact tine can be assumed equivalent to the
hydraulic detention time and is calculated by dividing the internal* volume
of the pipeline by the peak hourly flow rate through the pipeline. In
mixing basins and storage reservoirs, the hydraulic detention time
generally does not represent the actual disinfectant contact time because
of short circuiting. The contact time 1n such chambers should be
determined by tracer studies or an equivalent demonstration; The time
determined from the tracer study to be used for calculating CT Is T10. TJ0
represents the time that 90 percent of the water (and microorganisms
within the water) will be exposed to disinfection within the disinfectant
contact chamber. Guidance for determining detention time 1n contact
chambers Is provided In Appendix C.
The residual disinfectant concentration should be measured dally,
during peak hourly flow, for each disinfectant section prior to the first
customer In the distribution system. Unless a system knows from
experience when peak flow will occur, a system can only Identify peak
hourly flow after It has occurred. Therefore, EPA suggests that residual
measurements be taken every hour. If It Is not practical to take grab
samples each hour, the system may take grab samples during the period peak
flow is expected to occur, or continuous monitors may be used. The
measurements taken during the hour of peak flow can then be used to
determine the CT for each section (CTcllt). The determination of CTs is
explained 1n Section 3.2.1.
Although the Inactlvatlon maintained In the system 1s determined
during peak hourly flow, the disinfectant dosage applied to maintain this,
inactlvatlon may not be necessary under lower flow conditions. Under
lower flow conditions, a higher contact time 1s generally available and
the CT needed to meet the required Inactlvatlon may be met with a lower
residual concentration. Continuing to apply a disinfectant dosage based
on the peak hourly flow may provide more disinfection than 1s needed,
Increasing costs and possibly resulting in Increased levels of disinfec-
tant by-products. However, the system should also maintain the required
5-15
-------
Inactlvatlon levels at non-peak hourly flows. The system should therefore
evaluate the dose needed to provide the CT necessary for maintaining the
required Inactlvatlon under different flow conditions and set the dosage
accordingly. The following example provides guidelines for determining
flow ranges and disinfection levels to maintain the required disinfection.
A 20-mgd direct filtration plant applying free chlorine as a
disinfectant has a contact time of 27.5 minutes under peak flow condi-
tions. As noted In Section 5.3, well-operated direct filtration plants
achieve 2-1og Glardla cyst removal and l-1og virus removal. Therefore,
disinfection for 1-log Glardla cyst Inactlvatlon and 3-log virus
1nact1vatlon 1s recommended. The pH and temperature of the water are 7
and 5 C, respectively. Using Table E-2, a CT of 55 Is required to achieve
1-log Glardla cyst Inactlvatlon at a residual of 2 mg/L. This level of
treatment Is more than adequate for 3-log Inactlvatlon of viruses
requiring a CT of 6, as Indicated In Table E-7. However, under low flow
conditions the available contact time Is longer, and lower residuals are
needed to provide the same level of Inactlvatlon. Based on the calculated
contact time under various flow rates and the CT values In Table E-2,
adequate disinfection would be provided by maintaining the following
chlorine residuals for the Indicated flows:
CT90
Contact (mg/L-m1n) Free Chlorine
Flow (MGDV
the CminV
Reouired
Residual
20
27.5
55
2.0
15
36
52.5
1.5
10
54
50
1.0
5
108
47
0.5
CT„ corresponds to a 1-log Inactlvatlon. If a different level of
Inactlvatlon were needed, CT values for that Inactlvatlon would be read
from the tables corresponding to the pH and temperature of the water.
Section 3.2.2 lists the percent Inactlvations corresponding to
log inactlvations, I.e., 0.5-log equals 68 percent requiring
CT„*
5-16
-------
In cases where the residual, pH or temperature of the water 1s
an Intermediate value not reported 1n the tables, linear
(stra1ght-11ne) Interpolation aay be used.
For example, 1n the above listing, 0.5 mg/L residuals are not
Included 1n- the Appendix E tables. The CT.0 value was
determined by Interpolating between the £0.4 mg/l value of 46
mg/L-m1n and the 0.6 mg/L value of 48 mg/L-mln.
CT values for Intermediate pH and temperature values may also
be interpolated; or
The CT values for the higher pH or lower temperature listed 1n
the table may be used Instead of Interpolation.
CT„ . tables In the SHTR can be used to calculate the CT
required to achieve any log 1nact1vat1on by:
log 1nact1vat1on
CT required « required x CT„ ,
3.0 log
The variation 1n CT required with respect to the residual for
chlorine makes 1t Impractical for the utility to continually change the
disinfectant dose as the flow changes. Therefore, EPA suggests that the
flow variation at the utility be divided into ranges and the residual
needed at the higher flow of the range be maintained for all flows within
the range to assure adequate disinfection. The following flow ranges and
residuals at the given pH and temperature are suggested for this plant:
Free Chlorine
Flow Range (MSP) Residual (mg/Ll
5-10 1.0
10-15 1.5
15-20 2.0
In this way, the utility 1s assuring the provision of the required
disinfection while minimizing the disinfectant costs and possibly lowering
disinfection by-products.
Although these residuals will meet the required CT, maintaining a
residual 1n the distribution system must also be considered. If there 1s
no other point of disinfection prior to the distribution system, the
residual for disinfection must be maintained at a level which will also
5-17
-------
provide a residual throughout the distribution system. The complete range
of flows occurring at the plant should be evaluated for determining the
required residual. The utilities may establish the residual needs for as
many flow ranges as is practical.
The Primacy Agency'should make periodic checks to assure that the
utility is maintaining adequate disinfection at both peak and non-peak
flow conditions.
In contrast to this close control of disinfectant addition and CT
monitoring, for filtered systems which have long detention, times and
regularly exceed the CT requirements for the inactivation needed, it may
be unnecessary for the system to calculate CTs each day of operation.
Unlike unflltered systems where CTs must be calculated each day, for
filtered systems, monitoring the residual at the end of the contact time
may be sufficient to Indicate that the required disinfection 1s provided.
However, this results in much higher CTs in the sunnier than 1s needed,
which adds to costs and possibly unnecessary Increased production of
disinfection by-products. The following example outlines one scenario for
which this would apply.
A utility disinfects with chlorine ahead of a reservoir prior to
direct filtration. The Primacy Agency may give a well-operated direct
filtration plant credit for 2-1og Giardia cyst removal and 1-log virus
removal. Therefore, 1-log Giardia cyst and 3-log virus inactivation
through disinfection is needed. For free chlorine, the CTs for 1-log
Giardia cyst inactivation exceed the CTs for 3-log virus inactivation.
Therefore, CTs for Giardia cyst inactivation are the controlling CTs. The
following water quality conditions occur in the reservoir during the year:
Example
pH
Temperature (° C)
Chlorine residual (mg/L)
7 - 7.5
5 - 20
0.2 - 0.8
The required CT for chlorine Increases with:
increasing residual,
Increasing pH, and
decreasing temperature
5-18
-------
0
Thus, for a residual of 0.8 mg/L the CT needed for a l-1og Giardia
cyst inactivatlon Is as follows:
CT,0
fill Tcmpcrflture (0 mg/L-min .
7.5 5 58 (Table E-2)
7 20 18 (Table E-5)
Tracer studies conducted on the reservoir Indicated a'Tl0 of 150
minutes at the system's maximum flow. For the maximum CT of 58 mg/L-m1n
required, the minimum residual needed to meet this requirement Is 0.4
mg/L, calculated as:
59 M/L-mln ¦ 0.4 mg/L
150 min
At a residual of 0.4 mg/L, CT90 Is 55 mg/L-min. Thus, any residual ^0.4
mg/L will provide the needed disinfection throughout the year and the
Primacy Agency nay require the system to report only the residual
maintained, reducing the effort needed to determine effective disinfec-
tion. Maintaining this residual In the summer, however, provides much
higher CTs than needed, possibly resulting 1n unnecessary costs and
Increased disinfection by-products.
Meeting the Recommended Inactivatlon Us1n9 free Chlorine
As previously Indicated 1n Section 3.2.1, the effectiveness of free
chlorine as a disinfectant 1s Influenced by both the temperature and pH of
the water and by the concentration of chlorine. The inactivatlon of
61ard1a cysts by free chlorine at various temperatures and pHs are
presented In Appendix E (Table E-l through Table E-6). The CT values for
the Inactivatlon of viruses by free chlorine are presented 1n Table E-7.
To determine whether a system 1s meeting these Inactlvatlons, the
free chlorine residual, pH and temperature must be measured, at one point
or several points prior to the first customer, where contact t1me(s) 1s
measured. The contact time should be determined from the point of
application of the disinfectant to the po1nt(s) where the residual is
measured for determining CTs prior to the first customer. The CTs
5-19
-------
actually achieved In the system should then be compared to the values in
the table for the pH and temperature of the water at the point(s) of
residual Measurement. Guidance on calculating the CT for chlorine is
presented 1n Section 3.2.1.
Meeting the Recommended Inartjyation Using Chlorine D4mHH»
CT values for the 1nact1vat1on of Giardia cysts by chlorine dioxide
are presented in Table E-8 and the CT values for the inactivation of
viruses are presented in Table E-9. As shown in Tables E-a and E-9, the
only parameter affecting the CT requirements for chlorine dioxide is
temperature. However, the disinfection efficiency of chlorine'dloxlde may
be significantly Increased at higher pHs. Since the CT values In Tables
E-8 and E-9 were based on data at pH 7 and 6, respectively, and chlorine
dioxide appears to be more effective at higher pHs, systems with high pHs
may wish to demonstrate that CT values lower than those presented in
Tables E-8 and E-9 may achieve the desired level of inactivation.
Chlorine dioxide residuals are short-lived. Therefore, sampling and
residual analysis at various points 1n the treatment process downstream of
the point of application may be necessary to establish the last point at
which a residual 1s present. Subsequent sampling and residual analyses
conducted upstream of this point can be used to determine the CT credit by
using the demonstrated detention time between the point of application and
the sampling location. Methods for calculating CT values are presented in
Section 3.2. Systems using chlorine dioxide may conduct pilot studies to
demonstrate effective disinfection 1n lieu of calculating CT, or for
determining that lower CT values than those 1n Appendix E are appropriate.
Guidelines for conducting these studies are presented in Appendix G.
Meeting the Recommended Inactivation using Ozone
CT values for the Inactivation of Giardia cysts by ozone are.
presented in Table E-10 for various temperatures and inactivation rates.
As Indicated In this table, the CTs required for Inactivation with ozone
are substantially lower than those required for free chlorine. This
reflects the fact that ozone Is a more powerful disinfectant. The CT
requirements for Inactivation of viruses using ozone are presented in
Table E-ll. In cases where only a 1-log or lower Giardia cyst inactiva-
tion 1s needed, the CT values for virus Inactivation may be higher than
5-20
-------
the CTs for Giardia cysts. Because of the reactivity of ozone, 1t Is
unlikely that a residual will exist for aore than a few minutes. As a
result, the application of a persistent disinfectant such as chlorine or
chloranlnes 1s needed to maintain the required disinfectant residual 1n
the distribution system. Guidance for calculating CT values for ozone are
presented In Section 3.2.1 and Appendix 0. In lieu of calculating the CT
for an ozone contactor or demonstrating that lower CTs are effective, the
disinfection efficiency can be demonstrated through pilot studies as
presented In Appendix G. •
Megtlno the Recommended Inactivation Requirements using Chloramines
CT values for the Inactivation of Giardia cysts by chloranlnes are
presented In Table E-12. The high CT values associated with the use of
chloranlnes may be unachievable for some systems. In these cases,
chlorine, ozone, or chlorine dioxide should be used for primary disinfec-
tion, and chloramines for residual disinfection, as necessary. Table E-13
presents CT values for the inactivation of viruses with chloramines. This
table is only applicable for indicating virus inactivation efficiencies if
chlorine 1s added prior to ammonia. Systems which add ammonia prior to
chlorine or ammonia and chlorine concurrently, can determine viral
inactivation efficiencies using the protocol given in Appendix G. For
systems applying chloramines to meet the virus inactivation requirements,
EPA recommends that they also monitor for HPC in the finished water, as
presented in Section S.6. Systems also say demonstrate effective
disinfection with chloramines in lieu of calculating CT, or to determine
that lower CT values than those indicated in Appendix E are appropriate.
The protocols outlined in Appendix G can be used for this demonstration.
Further guidance on chloramines is given In Section 3.2.1.
Meeting the Inactivation Requirement
Using (ilV) Ratiiiticn
Ultraviolet radiation is a method of disinfection which can be
applied to meet the virus inactivation requirements of the SWTR.
UV disinfectant dose, expressed in terns of UV intensity and
exposure time/unit area (mW-sec/cm*) incorporates the elements of the CT
concept and therefore can be considered as analogous or equivalent to a CT
value. UV disinfection usually employs conmercially available units
5-21
-------
designed to deliver doses of 25 to 35 aW-sec/cm2. The dose can be
Increased by. reducing water flow rate and/or by adding additional units 1n
series. " UV disinfection efficiency differs from that of chemical
disinfectants In' that 1t Is not affected by water temperature. UV
radiation does not effectively penetrate solids and 1s absorbed by'certain
dissolved substances. Therefore, turbidity and other water quality
factors are important determinants of UV disinfection efficiency, and UV
should be applied after turbidity removal.
CT values for the 1nact1vat1on of G1ard1a cysts by UV are not
included In Appendix E. The results of two studies (R1ct and Hoff, 1981;
Carlson si il. 1985) Indicate that Giardia cysts are extremely resistant
to Inactivation by UV with doses greater than 60 mW-sec/cm2 causing less
than 80% Inactivation. Because UV appears to be very Ineffective for
Giardia cyst Inactivation and in the absence of sufficient data showing
the doses needed to Inactivate 0.5 to 3.0 logs of cysts, UV must be used
in combination with other disinfectants to provide evidence of effective
cyst Inactivation.
CT values for the Inactivation of viruses by UV are presented in
Table E-14. Units used for UV disinfection should be equipped with fall-
safe devices that will provide automatic shutdown of water flow If UV dose
decreases to levels lower than those specified 1n Table E-14.
Meeting the Inactivation Requirement Using Alternate Disinfectants
For system using disinfectants other than chlorine, chloramines,
chlorine dioxide, or ozone, the effectiveness of the disinfectant can be
demonstrated using the protocol contained 1n Appendix G. The protocol In
Appendix G.3 for batch testing should be followed for any disinfectant
which can be prepared 1n an aqueous solution and will be stable throughout
the testing. For disinfectants which are not stable, the pilot study
protocol outlined In Appendix G.4 should be followed.
Examples for Determining the Disinfection to be Provided
1) Recommended 0.5-loo Giardia. 2-1oo Virus Inactivation
A community of 70,000 uses a river as Its drinking water source.
Ozonation prior to a conventional treatment plant is used to treat the
5-22
-------
water. The source has a protected watershed with limited human activity
and no sewage discharge. The river water has the following water quality
characteristics:
Turbidity 10 - 200 NTll
Total estimated Glardla cyst level <1/100 /L
pH 7.0 - 7.5 .
Temperature 5-15
The treatment plant has a design capacity of 15 mgd and treats an
average flow of 10 mgd. A three chamber ozone contactor precedes the
rapid nix. Alum and polymer are added as a coagulant and coagulant aid,
respectively. The finished water turbidity at -the plant Is maintained
within the range of 0.1 to 0.2 NTll. Chloramlnes are applied after the
filters, but prior to the clearwells, to maintain a residual entering and
throughout the distribution system.
Based on the raw water quality and source water protection, an
overall 3-log Glardla cyst and 4-log virus removal/inactlvation 1s
appropriate for this water source. However, as noted 1n Section 5.3,
Primacy Agencies may credit well operated conventional filtration plants
with 2.5-1og Glardia cyst removal and 2-log virus removal. Therefore,
disinfection for 0.5-1og Glardia cysts and 2-1 og viruses 1s recommended to
meet the overall treatment requirements of the SWTR.
On the day of this example calculation, the peak hourly flow rate of
the plant was 13 mgd. The contact time of the ozone basin, T)0 determined
from tracer study data Is 6 minutes for this flow. The water had a pH of
7 and a temperature of 5 C on the day of the calculation. For ozone under
these conditions of pH and temperature, the following CTs are needed for
the required 1nact1vat1on (Tables E-10, E-ll):
0,5-loa Glardla 2-io? virus
CT 0.3 0.6
The CT values indicate that viruses are the controlling parameter for
disinfection and the overall Inactivation provided will be calculated
based on viruses. The overall virus inactivation provided by the ozone
contactor 1s determined as follows:
Average
5-23
-------
Residual
(mlnutcs)
Chamber C (mfi)
1
2
3
0.1
0.2
0.2
2 0.2 0.9 0.
2 0.4 0.9 0.
2 0.4 0.9 0.
0.22
0.44
0.44
The sun of CT{|I{/CT„ , 1s 1.1. This corresponds to more than a 3-1 og virus
Inactlvatlon determined as 3 X CTlllc/CT„ , ¦ 3 X 1.1 - 3.3-log. Therefore,
the system exceeds the recommended Inactlvatlon.
2) Recommended l-loo Slardla Cvst. g-1oo virus Inactlvatlon
A 2 MGD slow sand filtration plant treating reservoir water, fed by
mountain streams with no nearby wastewater discharges, provides drinking
water for a community of 8,000 people. The water quality at the Intake
has the following water quality characteristics:
Turbidity 5-10 HTU
Total coliforms Not measured
Total estimated Slardia cyst level <1/100 L
pH 6.5 - 7.5
Temperature 5 - 15 C
The filtered water turbidity ranges from 0.6 - 0.8 NTU. Considering
the source water quality and plant performance, an overall 3-log Giardia
cyst and 4-log virus removal/Inactlvatlon Is considered sufficient for
this system. As noted In Section 5.3, the Primacy Agency may credit slow
sand plants with 2-log Glardla cyst and 2-1og virus removal. Therefore
disinfection for l-1og Glardla cyst and 2-log virus Inactlvatlon Is
recommended for the system to meet the overall treatment requirements.
Chlorine 1s added prior to the clearwells to provide disinfection.
The clearwells have a capacity of 80,000 gallons. A one mile, 16-Inch
transmission main transports the water from the treatment plant to the
first customer. The Inactlvatlon provided Is determined dally for the
peak hourly flow conditions*" Tracer studies have been conducted to
determine the Tl9 for the clearwells for different flow rates. For the
purposes of calculating the Inactlvatlon the system 1s divided Into tv*> sections.
Section 1 - clearwell
5-24
-------
Section 2 - transmission main
The flowrate at peak hourly flow from the clearwell was 1.5 mgd on
the day of this example. At this flowrate, the T10 of the clearwell 1s 67
minutes, as determined from the results of the tracer studies. At this
flowrate, water travels through the transmission main at 99 ft/min. The
data for the calculation of the 1nact1vat1on 1s as follows:
Section 1
Section 2
0
5280
0
53 *
67
0
67
53
chlorine
chlorine
1.0
0.6
5
5
7.5
7.5
length of pipe (ft)
contact time (min)
pipe
basin
total
disinfectant
residual (mg/L)
temperature C
pH
For free chlorine, a 1-1og Giardia cyst 1nact1vat1on provides greater than
a 4-log virus 1nact1vat1on; therefore, Giardia cyst 1nact1vat1on 1s the
controlling parameter, and the inactivatlon provided 1s determined based
on Giardia cysts. The calculation is as follows:
SWtiPn 1 - Chlorine
CTtllt ¦ 1.0 mg/L x 67 minutes ¦ 67 mg/L-m1n
From Table E-2, at a temperature of 5 C and a pH of 7.5, CT„ 9 is
179 mg/L-min
CT«.ic/CT99« ¦ 67 mg/l-min ¦ 0.37
179 mg/L-min
Section 2 - Chlorine
CTtil( ¦ 0.6 mg/L x 53 minutes ¦ 32 mg/L-min
From Table E-2, at a temperature of 5 C and a pH of 7.5, CT„ 9 is
171 mg/L-min
CT,m/CT.i.i ¦ 32 * 0.19
171 mg/L-min
The sum of CT{I|{/CT„ t 1s equal to 0.56. This is equivalent to a 1.7-log
Giardia cyst inactivatlon determined as 3-1og x CT{|I{/CT„ , ¦ 3 x 0.56 «
5-25
-------
I,7-logs. Therefore, the system exceeds the disinfection recommended to
¦eet the overall treatment requirements.
3) Recommended 2-log Siardia Cvst. 4-1o? Virus Inactivate
A coonunity of 30,000 people uses a reservoir treated by direct
filtration for its water supply. The reservoir 1s fed by a river which
receives the discharge from a wastewater treatment plant 10 miles upstream
of the reservoir. The reservoir water quality Is as follows:
Turbidity 5 - IS NTU
Total coliforis 100 - 1000/100 ml -
Total estimated Glardia cyst level 5/100 L
pH 6-7
Temperature i - 15 C
Based on the source water quality, an overall removal/1nact1vation
of 4-log Giardia cyst and 5-log virus is reconmended as outlined in
Section 4.4.
The source water flows by gravity to a 3 HG storage reservoir prior
to pumping to the water treatment plant. Chloramines are produced by
first adding chlorine then ammonia to the water within the Inlet of the
storage reservoir. Chlorine dioxide is added to the filtered water prior
to the clearwells. Chloramines are applied after the clearwells to
maintain a residual in the distribution system. The system design flow is
8 mgd with an average flow of 5 mgd. For the calculation of the overall
inactlvatlon, the system is divided into 2 sections.
Section 1 - the storage reservoir and the transmission to the
treatment plant
Section 2 - the clearwells
The overall inactlvation for the system is computed daily at the
peak hourly flow conditions. The pH, temperature, and disinfectant
residual is measured at the end of each section prior to the next point of
disinfectant application and the first customer.. The flow is measured in
the transmission main entering the plant and exiting the clearwells. On
the day of this example calculation, the peak hourly flow was 6 mgd in the
transmission mains entering and leaving the plant. If the flowrates were
different, the T,0 corresponding to the respective flowrate would be used
5-26
-------
1n the calculation. Guidance for determining CTs when flowrates vary
within a system 1s given 1n Section 3.2. The Mater velocity through the
20-Inch transmission main 1s 256 ft/mln at a flow of 6 mgd. Tracer
studies were conducted on the storage reservoir and clearwells. As
determined from the testing the detention tines« T,0, of the basins at a
flow of 6 ngd are 380 and 130 Minutes for the storage reservoir and
clearwells, respectively. The data for the calculation of 1nact1vat1on Is
as follows:
length of pipe (ft)
contact tine (min)
pipe
basin
total
disinfectant
residual (mg/L)
temperature C
pH
Section 1
4500
18
380
398
chloramines
1.5
5
7
Section 2
0
130
130
chlorine dioxide
0.2
5
7
For each of the disinfectants used, the following CTs are needed for
2-log Slardia and 4-log virus Inactivation for the pH and temperature
conditions of the system.
chloramines
chlorine dioxide
CT for 2-log
61ard1a
1430
17
CT for 4-log
Virus
1988
33.4
The CT required for the virus inactivation Is higher than that
needed for Giardla Inactivation for each of the disinfectants. Since the
viruses are the controlling parameter, the Inactivation calculation will
be based on the viruses. The calculation 1s as follows:
Section 1 - Chloramines
CTtlU ¦ 1.5 ng/L x 398 minutes - 597 ¦g/L-min
From Table E-13, at a temperature of 5 C and a pH of 7, CT„ „ is
1988 mg/L-min
CTtilt/CT(
99.19
597 fro/l-min ¦ 0.3
5-27
-------
1968 mg/L-mln
Section 2 - Chlorine Dioxide
CTtllt ¦ 0.2 og/L x 130 minutes - 26 mg/L-m1n
From Table E-9, at a temperature of 5 C and a pH of 7, CT.« aa Is
33.4 mg/L-m1n " 99
CTe»|e/CTfl» is ¦ 26 roo/L-iiHn ¦ 0.78
33.4 mg/L-mln
The sum of CT{ilt/CT„ „ Is equal to 1.08, which 1s equivalent to a 4.3-log
Inactlvatlon of viruses, determined as follows:
x • 4-log x CTca1c - 4 x 1.08 ¦ 4.3-logs
CT -
U,M 99
Therefore, the system provides sufficient disinfection to meet the overall
recommended treatment performance.
5.6 Other Considerations
Monitoring for heterotrophic plate count (XPC) bacteria 1s not ..
required under the SWTR. However, such monitoring may provide a good
operational tool for:
Measuring microbial breakthrough
Evaluating process modifications
Detecting loss of water main Integrity
Detecting bacterial regrowth conditions within the distribu-
tion system
Determining Interference with the collform measurements (AWWA,
1987)
Therefore, EPA recommends routine monitoring for HPC In the plant
effluent and within the distribution system whenever the analytical
capability Is available in-house or nearby. Systems which do not have
this capability should consider using a semi-quantitative bacterial water
sampler kit, although this Is not acceptable for compliance monitoring.
5-28
-------
As discussed 1n the pream&.e to the SWTR, EPA believes that It 1s
Inappropriate to Include HPC as a treatment performance criterion 1n the
rule since snail systems would not have 1n-house analytical capability to
conduct the measurement, and they would need to send the samples to a
private laboratory. Unless the analysis 1s conducted rapidly, HPC nay
multiply and the results may not be representative.
EPA recommends an HPC level of less than 10/ml 1n the finished water
entering the distribution system and levels of less than 500/ml throughout
the distribution system.
Legionella 1s another organism which 1s not included as a treatment
performance criterion. Inact1vat1on Information on Legionella Is limited.
EPA believes that treatment which complies with the SWTR will remove
and/or Inactivate substantial levels of Legionella which might occur 1n
source waters, thereby reducing chances that Legionella will be trans-
ported through the system and reducing the possibility that growth might
occur 1n the distribution system or hot water systems within homes and
institutions. Since Legionella are similar In size to collform organisms,
removals by filtration should be similar to those reported for total
conforms. In addition, the available disinfection information Indicates
that the CT requirements for inactivation of Legionella are lower than
those required for the inactivation of Giardia- cysts. EPA recognizes,
that regardless of the treatment provided, some Legionella may enter
plumbing and air conditioning systems and subsequently multiply (Muraca et
al., 1986). EPA believes that these concerns are best addressed through
guidance contained 1n Appendix B.
5-29
-------
6. mmm
6.1 Reporting Requirements for Public Hater Systems
Not Providing Filtration .
The 5WTR requires unfiltered systems to prepare monthly reports for
the Primacy Agency to determine compliance with the requirements for:
source water fecal and/or total col1form levels
source water turbidity levels
disinfection level
disinfectant residual entering the distribution system
disinfectant residuals throughout the distribution system.
The monthly reports must be prepared and submitted to the Primacy
Agency within 10 days after the end of the month. The utility must
maintain a daily or monthly data log used to prepare the monthly reports.
Tables 6-1 through 6-5 are examples of daily data sheets which the
utilities may find useful for logging the data needed to prepare reports
for the Primacy Agency.
Table 6-6 presents a concise format which can be used by the system
for the monthly reports to the Primacy Agency. Tables 6-3 and 6-4 must
also be submitted with the monthly report. After the initial 12 months
of reporting, the Primacy Agency may remove the requirement for reporting
the information contained in Table 6-3 1f it 1s satisfied that the system
is computing compliance with the CT requirements correctly. The
individual sample results summarized In the monthly reports should be kept
on file at the utility for a minimum of 5 years.
In addition to the monthly reporting requirements for source water
quality conditions and disinfection information, systems with unfiltered
supplies are also required to submit annual reports for the watershed
control program and the on-site Inspection, within 10 days after the end
of the federal fiscal year.
The Primacy Agency will review the reports to determine whether the
system is 1n compliance. A possible report format for the watershed
control program is:
6-1
-------
1. Summarize all activities in the watershed(s) for the previous
year.
2. Identify activities or situations of actual and potential
concern in the watershed(s).
3. Describe how the utility is proceeding to address activities
creating potential health concerns.
EPA recommends that the Primacy Agency submits the annual watershed
reports to the State Water Quality Managers. The reports will be useful
1n updating statewide assessments and management programs.
The SWTR requires each system to provide the Primacy Agency with a
report of the on-site inspection unless the inspection 1s conducted by the
Primacy Agency. EPA suggests that:
1. A report of the Inspection containing the findings, suggested
improvements and dates by which to complete improvements 1s
to be prepared followina the initial system review. When and
how system has resolved problems Identified in the previous
report should also be included.
2. To lessen the burden on utilities, a report containing results
of the general survey should be submitted in subsequent years.
In addition to these reporting requirements, the SWTR requires that
the reporting requirements of the Total Trlhalomethane Regulation and the
Collform Rule also be met.
Records of waterborne disease outbreaks also must be maintained.
In the event of a waterborne disease outbreak, as defined in part 141.2
of the SWTR, the Primacy Agency must be notified by the end of the next
business day.
The report of the outbreak should contain:
1. Date of occurrence
2. Type of illness
3. Number of cases
4. System conditions at the time of the outbreak, including
disinfectant residuals, pH, temperature, turbidity, and
bacteriological results.
The records of an outbreak should be maintained permanently or until
filtration is installed.
6-2
-------
6.2 Reporting Requirements fr>r Public Water Systems Us1n0 nitration
The SWTR requires filtered water systems to submit Monthly reports
to the Primacy Agency for determination of compliance with the require-
ments for:
treated water turbidity
disinfectant residual entering the distribution system
disinfectant residuals throughout the distribution system
Tables 6-7 and 6-8 present a format which the utility can use as a dally
data log and to submit monthly reports to the Primacy Agency.
Recommended Reporting Not Required bv the SWTR
The Primacy Agency may also want filtered water systems to report
some Information associated with recommendations made In this manual which
are not requirements of the SWTR. EPA recommends that, filtered water
systems:
1. Report the log Inactlvatlon of Glardla cysts and viruses,
required by the Primacy Agency.
2. Report point of application for all disinfectants used.
3. Report the dally CT(s) used to calculate the log Inactlvatlon
of Glardla cysts and viruses.
4. If more than one disinfectant Is used, report the CT(s) and
Inactlvatlon(s) achieved for each disinfectant and the total
percent Inactlvatlon achieved.
5. Note any difference between the measured CT(s) and the CT
required to meet the overall minimum treatment performance
.. requirement specified by the Primacy Agency.
Tables 6-3 and 6-4 can be used to maintain the records necessary for
numbers 2 through 5.
This Information can be used to determine the disinfection level
maintained by the system to assure that the overall removal/1nact1vat1oh
required 1s maintained.
6-3
-------
The Primacy Agency nay make provisions to minimize the reporting
requirements for systems with reservoirs, large amounts of storage or long
transmission mains which provide a long disinfectant contact time. Since
these systems typically provide inactlvatlon In excess of that needed, the
Primacy Agency may require the system only to report the minimum dally
residual at the end of the disinfectant contact time. The CT maintained
can then be estimated based on this residual and the contact time under
the system design flow. This method of CT determination will eliminate
the need for the system to determine the contact time under maximum flow
conditions each day.
-------
TABLE 6-1
I
SOURCE WATER QUALITY CONDITIONS FOR UNFILTERED SYSTEMS
(For iy«am im oafy)
Moeth _____ Symmt/Tnttamt FUat __
Year PWSID
Turbidity Mtf*surtncn:i
Date
2
Cotiform Measure menu
3
Maximum
Turbidity
(NTU)
T.;*:
'• 'Eve
No. of Sample*
No. of Sample* Meeting Specified Limit*
Fecal
Total
Fecal«- 20/100 mL)
Total«- 100/100 mL)
I
.
2
3
;
4
3
i
6
I
7
i
»
I
9
i
i
10
i
11
1
12
'• i
13
t
14
•
IS
16
17
IS
19
20
21
22
23
24
26
27
,
29
30
31
.
Total*:
Maximum daily turbiJit> *
Total number of turS.ii:> oc-.-.ts"
Notes:
1. Sample* are taken from the tourca water immediately prior to tba Rnt diiiafecboa point included in the CT deternnmi.. ::
2. At (pacified in 40 CFR 141.74(bXl), • focal or total coiiforn *ampie reus be taken on each day that the
*y*tem operate* and a *ource water turbidity measurement exceed* I NTU.
3. For each day that the maximum turbidity exceed* S NTU. the date should tl*o b« entered for the day that the Sutc
of thi* exceedance, e.g.. "7.3-22 Apr".
4. A "ye*" retponie i* required each day the maximum turbidity exceed* 5 NTU and the previous day did not. This i>
of the beginning of « turbidity'event*. The total number of "ye*' responses equal* the number of turbidity *cvcm» • .
-------
TABLE 6-2
LONG-TERM SOURCE WATER QUALITY CONDITIONS FOR
UN FILTERED SYSTEMS
(For quia «ee oaljr)
Year Syrtem/Treataeat Plant
FWSID
Turbidity Meaturtncnn
Month
Coliform Measurement* '
Day* with
Turbidity
>5 NTU
Nunbi
Turbv
Ever
No. of Samples No. of Sample* Meeting Specified Limits
Fecal
Total
Fecal(< ¦ 20/100 mL)
Total (< - 100/100 mL)
Jan wry
-
February
March
April
May
!
June
i
i
i
July
i
i
i
»
August
i
i
i
September
October
November
-
1
December
i.
i
i
i_—
Total:
-------
Month,
Ymj _
TABLE 6-3
_
CT DETERMINATION FOR UNFILTERED SYSTEMS - MONTHLY REPORT TO PRIMACY AGENCY
__ Syam/TiwtaMt Plant __
: rwsrn
Date
3
Disinfectant
Concentration,
C (mg/L)
3
Disinfectant
Contact Tim*.
T (mia.)
4
CTcalc
(¦C*T)
3.5
PH
3
Water
Temp.
«e«Q
6
CT99.9
(CTttlcCl
1
2
3
-
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
Is
19
20
21
22
23
24
25
26
27
28
29
30
31
Notes:
Pnpind by.
t. To be included in the monthly report for at I
may bo longer require this form.
2. Use i separate form for ench disinfectant/samplin| «it«
3. Measurement filrwi at peak hourly flow.
4. CTcalc ¦ C (mf/L) * T (mia.).
5. Only required if the diiinrectanl ii free chlorine.
6. From Tablet 1.1 • 1.6. 2.1. and 3.I.40CFR I4I.740>K3)
12 months after tha initiation of reporting After that time, the Primacy Agency
Enter disinfectant tad sequence position. *oxooe/lst" or "G02'3r
-------
TABLE M
DISINFECTION INFORMATION
FOR UNF1LTERED SYSTEMS - MONTHLY REPORT TO PRIMACY AGENCY
SymaraJT ft ran Plant
PWSID
Dai*
1
Minimum Disinfectant Residual
at Poitt-of-Eatry to
Distribution System (mg/L)
(CTcalc/CT99.9) (from Tabl* 6-3)
?
SUM (CTcalc/CT99.9)
SUM
-------
TABLE 6-5
DISTRIBUTION SYSTEM DISINFECTANT RESIDUAL DATA FOR UNFILTERED AND FILTERED SYSTEMS
MONTHLY REPORT TO PRIMACY AGENCY
j
j Month __Symm/Tmtmm PUot
jYcar PWSID
Data
No. of Sum Whm
No. of Sitae Wkeie no
No. of Sitae Wbm
No. of Sim Whata
No. of Sues Whart
Diiiofectut Rendu*!
Disinfectant Residual
Dieiafectaot Residual
Diiia&cttxtf Rniduii
Disinfectant Residual
mil Measured (»•)
Maamrad, but HPC
Not Detected, no HPC
Not Dateettd,
Not Measured,
MmnwI (»b)
Maasutad (>»e)
HPC > 500/ml (»d)
HPC > S00nl(-«)
1
-I
3 !
i 4 I
i s «
! « !
7 1
1 i
91
10 1
! 11 i
12 i
! 13 i
1 »
1 lS
1 16
! it
IS
19
20
21
22
23
24
' 23
26
27
28
29
30
31
Tatl
• "
b-
Cm
d«
a™
V »* (c+4 +«)/(.-b)* 100 - ( + + p( + ) a 100 - *
Pi«p*i«d by
Dm
-------
TABLE 6-6
MONTHLY REPORT TO PRIMACY AGENCY FOR
COMPLIANCE DETERMINATION - UNALTERED SYSTEMS
• /Twu—tM—t ¦
PWSID
w«" Ch»1'*y Condition*
A. Cumulative number of month* for which rewltt mi reported
For aource woter eoliform moaitoriai ___ (No. or month*)
For turbidity monitoring _____ (No. of month*)
1
B. Colifons Criteria
No. of Sample*
Focal
Total
No. of Samplea Meeting Specified Limit*
Preview* 6 month**;
Percentage of inmples < <
Percentage of aaaplee < ¦
UP <90% h Yaa:
»»
Focal (<¦ 20/100 mL)
Total«- lOfttOOmL)
' 20/100 mL focal coliforms, F • y/w * 100 -
> 100/100 mL total coiiformi, T * i/* * 100 ¦
No: N/A ; is T < 90* ?: Ya»:
.*
No:.
„ N/A:,
C. Turbidity Criteria
Mali mum turbidity level for reporting (current) month » .
. NTU
Date* of 5 NTU Euoedaiices Since Latest Month Recorded Abova
Beginning Data
Duration (day*)
Data Reported
Diiinfectiow Criteria
A. Poiat-of-Eiury Minimum DUinfectawt Reiidual Criteria
Day* the Reiidual wa* <0.2 mg/L
Day
Duratioo of Low Level (to.)
Deta Reported
to Primacy Agency
B. Distribution Syoam Disinfectant Ratidual Critoria
Tbe value of a. b, c, d, and a from Table 6-5, aa apocifwi in 40 CFR 141,73 {btfXiiiXAHE):
¦ ¦ ¦ b ¦ -c ¦ ¦ d ¦ ¦ • ¦ __
V" ' + * + • «»Q0" *
a + b
For previous month, V ¦ ___ *
C. Diinfectioo Requirement Criteria
Record tbe date and value of SUM (CTealc/CTW.#) for any SUM (CTealc/CTO J) < I (from Tabta *4):
If none, enter'none".
Date
SUM (CTcalc/CTW.9)
Prepared by _____________
Notea: 1. The current 6-month cumulative* are required to detrain# whether compliance with the eoliform criteria
hai been achieved. Thaaa total* are calculated from: the previou* t-moeth cumulative*, the current
month'*. and total* from the eerlictf of 6 previou* mondn.
-------
TABLE 6-7
J—~ ' DAILY DATA SHEET FOR FILTERED SYSTEMS
j (Foe syttata aee ooly)
I
, Month tymM/Ttmamt PUat
!Yeer Pite«do« T«dttotofjf _
! FWSID
1
1
1
1
iDate
1
Minimum Disinfectant Residual
at Point-of-Entry to
Distribution System (mg/L)
2
Maximum Filtered Water Turbidity
3
No. of Turbidity
Measurements
4
No. of Turbidity
Measurements < •
Specified Limit
No. of Tu
Measurei
•• > 5 N'
Filter
f
Combined Filter
Effluent
Clearwell
Effluent
Plant
Effluent
i I i
2 i
3 i
4 I
5
! 6 .
i
i " i
i s 1
i 9 1
I 10 |
1 n i
12 1
13 i
1 «
15 1
16 '
I 17 '
J 18 1
; 19 1
1 20 j
! 21 1
22 I
i
23 1
i
24 |
i
25
i
26 |
;
27 !
28
29
30
31
Totals:
Notcv
1. For multiple disinfectants. this column rauat only b« completed for the last disinfectant added prior to entering th« distribution
system. If 1cm than 0.2 mg/L, the dureboc of dM period mut be reported, e.j., *0.1 -3 hrt".
2. For systems using conventional treatment, direct filtration, or technologies other than slow send or diatomaceous earht filtration,
turbidity measurements may be taken et the combined filter effluent, cJmtwcU effluent, or plaat effluent prior to entry into the
distribution system. The turbidity may also be measured for each individual filter with a separate sheet maintained for each
3. For continuous monitors count each 4-hour period as 1 sample.
;4. Depending on the filtration technology employed, the number of turbidity samples meeting the following levels must be .-c^rJed:
conventional treatment or direct filtration-0.5 NTU, slow sand filtration-1 NTU. diatomaceous earth filtration-l NTL' 1>.e S-j:e rr
specify alternate performance levels for conventional treatment or direct filtration, not exceeding 1 NTU, and slow unJ •'..••ni 'n.
not exceeding 5 NTU, in which case the number of turbidity measurements meeting tbeaa levels must be recorded
5. In recording the number of turbidity measurements exceeding 5 NTU, the turbidity values should also be recorded, c ; ¦ * - 6
-------
TABLE 64
MONTHLY RETORT TO PRIMACY AGENCY FOR
COMPLIANCE DETERMINATION - FILTERED SYSTEMS
Moot* _______ Syea/Traaims* PUoc
Vm» Tvw iJ gjhrtina
Taibidify Lirak _________
rwsro
Turbidity Performsnce Criteria
A. Total millibar of filtered water turbidity measurements m _____
B. Total number of filtered water turbidity measurements that are leu than or equal to Um specified liraiu
for the flltratka technology employed - _____
C. Tba percentage of turbidity measurements mating tha specified limits ¦ l/A * 100 ¦ / » 100 • «
D. Record the data and turbidity valua for any measurements exceeding 5 NTU: II none, eater *aoaa*.
Date
Turbidity, NTU
Disinfection PerforHtintt Criteria
A. Point-of-Entry Minimum Disinfectant Residual Criteria
Date
Minimum Disinfectant Residual
at Point-of-Eiitry
to Distribution System {mg/L)
Date
Minimum Disinfectant Residual
at Point-of-Entry
to Distribution System (mg/L)
Data
Minimum Disinfectant Residual
at Point-of-Emry
to Distribution System (nj'L)
1
II
21
2
12
22
3
13
23
4
14
24
5
13
23
6
16
26
7
1?
27
t
it
21
9
19
29
10
20
30
31
Days tba Residual was <0.2 mg/L
Day
Duration of Low Level (bra.)
3ate Repotted to Primacy Agency
8. Distribution System Disinfectant Residual Criteria
Tba value of a, b, c, d, and a from Table 6-5, at specified in 40 CFR 141.75 {b)(2XiiiXsH*):
V" <~<«~>» >00- *
a + b
For previous month, V - *
Prepared by
Data
-------
7• COMPLIANCE
7.1 Introduction
This section provides guidance on when and how the requirements of
the SWTR will go Into effect, Including determinations made by- Primacy
Agencies.
7.2 SYSTEHS USING A SURFACE WATER SOURCE (NOT GROUND WATER
UNDER THE DIRECT INFLUENCE OF SURFACE WATER 1
The SOWA requires, within 18 months following the promulgation of
a rule, that Primacy Agencies promulgate any regulations necessary to
Implement that rule. Under S1413, these rules must be at least as
stringent as those required by EPA. Thus, Primacy Agencies must
promulgate regulations which are at least as stringent as the SWTR by
December 30, 1990. By December 30, 1991, each Primacy Agency must
determine which systems will be required to filter. If filtration 1s
required, 1t must be Installed within 18 months following the determina-
tion or by June 29, 1993, whichever is later. In cases where it is not
feasible for a system to Install filtration 1n this time period, the
Primacy Agency ir.ay allow an exemption to extend the time period (see
Section 9).
If a Primacy Agency falls to comply with this schedule for adopting
the criteria and applying them to determine who must filter, systems must
comply with the "objective11 or self-implementing criteria (I.e., the
requirements that are clear on the face of the rule and do not require the
exercise of Primacy Agency discretion). Unflltered supplies must comply
beginning December 30, 1991 and filtered supplies beginning June 29, 1993.
Monitoring requirements for unflltered systems must be met beginning
December 30, 1990 unless the Primacy Agency has already determined that
filtration 1s necessary. This coincides with the Agency's requirement to
promulgate regulations for making filtration decisions by that date under
the SDWA. Primacy Agencies may specify which systems should conduct the
monitoring necessary to demonstrate compliance with the criteria for
avoiding filtration. For sooe systems where an historical data base
exists, and where it 1s apparent that the system would exceed the source
7 - 1
-------
Mater quality criteria (or that some other criteria would not be met, such
as an adequate watershed control program), no monitoring may be necessary
for the Primacy Agency to determine that filtration Is required. If a
particular system (and/or the Primacy Agency) knows that it cannot meet
the criteria for avoiding filtration, there 1s no reason to require that
system to conduct the source water monitoring prior to the formal decision
by the Primacy Agency that filtration Is required. This 1s true because
the only purpose of that monitoring would be to demonstrate whether or not
the criteria to avoid filtration are being met.
In reviewing the data for determining which systems must filter, the
Primacy Agency will have to decide on a case-by-case basis the conditions
which will require filtration. For example, a system may not meet the
specified CT requirements for the first few months of monitoring and
upgrades Its disinfection to meet the CT requirements In subsequent
months. In this case, the Primacy Agency could conclude that the system
will be able to meet this criterion for avoiding filtration. The time
periods specified for 1n the criteria to avoid filtration (e.g., six
months for total conforms, one year and ten years for turbidity and one
year for CT requirements) do not begin until December 30, 1991 unless the
Primacy Agf-cy specifies an earlier date.
Beginning December 30, 1991 the requirements for avoiding filtration
specified 1n S141.71(a) and (b) and the requirements of S141.71(c) and
S141.72(a) go Into effect unless the Primacy Agency already has determined
that filtration Is required. Beginning December 30, 1991, if a system
falls to meet any one of the criteria for avoiding filtration, even if the
system were meeting all the criteria up to that point, it must install
filtration and comply with the requirements for filtered systems Includ-
ing the general requirements in S141.73 and the disinfection requirements
In S141.72(b), within 18 months of the failure. Whenever a Primacy Agency
determines that filtration is required, it may specify Interim require-
ments for the period prior to Installation of filtration treatment.
Following the determination that filtration Is required, the system
must develop a plan to implement its Installation. The plan must include
consideration for the following:
7 - 2
-------
Providing uninterrupted water service throughout the
transition period
Siting for the future facility
Financing options and opportunities
. - Scheduling of design and construction
Systems which are unable to Install filtration within the specified time
frame may apply for an exemption to extend the period for-Installing
filtration.
Table 7-1 summarizes the requirements for the SWTR for unflltered
systems noting conditions which require the installation of filtration.
It 1s Important to note that only treatment technique violations trigger
the requirement to install filtration while violations of monitoring,
reporting or analytical requirements do not. The monitoring requirements
for unfiltered supplies are presented in Section 3 and the reporting
requirements are presented 1n Section 6.
All systems with filtration in place must meet the treatment
technique requirements specified in S141.73 (filtration criteria) and
S141.72(b) (disinfection criteria), and the monitoring and reporting
requirements specified in 5141.74(c) and S141.75(b), respectively,
beginning June 29, 1993. Table 7-2 summarizes the SWTR requirements for
filtered systems, Including conditions needed for compliance with
treatment requirements. Monitoring requirements for filtered supplies are
enumerated 1n Section 5 and reporting requirements are presented in
Section 6.
7.3 Compliance Transition with Current NPDWR Turbidity Requirements .
The current (Interim) NPDWR for turbidity under S141.13 (MCL
requirements) and S141.22 (monitoring requirements) will apply for
unfiltered systems until December 30, 1991 unless the Primacy Agency
determines that filtration is required. In cases where filtration is re-
quired, the Interim NPOWR applies until June 29, 1993 or until filtration
is Installed, whichever is later. Unfiltered supplies will also be
7 - 3
-------
subject to the turbidity monitoring requirements of $141.74(b)(2)
beginning December 30, 1990 colncldently with the Interim requirements.
Beginning June 29, 1993, the turbidity performance criteria for filtered
systems ($141.73), and the monitoring requirements under S141.74 will
apply.
7.4 Systems Using a Ground Water Source
Under the Direct Influence of Surface Water
Part of the Primacy Agency's program revisions to adopt the SWTR
must Include procedures for determining, for each system 1n the Primacy
Agency served by a ground water source, whether that source is under the
direct Influence of surface water. By June 29, 1994 and June 29, 1999,
each Primacy Agency must determine which community and non-community
public water supplies, respectively, use ground water which is under the
direct influence of surface water. EPA recommends that these determina-
tions be made in conjunction with related activities required by other
regulations (e.g., sanitary surveys pursuant to the final coliform rule,
vulnerability assessments pursuant to the volatile organic chemicals rule,
the forthcoming disinfection requirements for ground water systems).. In
addition, EPA-approved wellhead protection programs required under the
Safe Drinking Water Act Section 1428 may contain methods and criteria for
determing zones of contribution, assessments of potential contamination,
and management of sources of contamination. These programs may be used
as a partial basis for the vulnerability assessment and for making the
determination of (a) whether a system is under the direct influence of
surface water and (b) if direct influence is determined, whether there is
adequate watershed control to avoid filtration. Guidelines for developing
and implementing a wellhead protection program are found in "Guidelines
for Applicants for State Wellhead Protection Program Assistance Funds
under the Safe Drinking Water Act" (U.S. EPA, 1987a).
A system using a ground water source under the influence of surface
water that does not have filtration in place must begin monitoring and
reporting in accordance with $141.74(b) and S141.75(a), respectively, to
determine whether it meets the criteria for avoiding filtration beginning
December 30, 1990 or six months after the Primacy Agency determines that.
7 - 4
-------
IMlf M
KQUiRCKais rot mruitiiD mum
Ualllttral Sapplias
(Sill.71)
a) laatta ffilff Qaallly
CMilllMi
1) fatal Call tar*
lalal Calilaia
t) fatklllly
I) SllaSptfllit
Caallllaas
11) Olslalstllaa fat
» »•« iltHil •
cyst I (li|
tins laatlitallaa
(SHI 12(a))
Rtlaalaal
Olilalacllaa
CMpaatals
(SHI, 72(a)(2))
Blsialatlaal
Rasllaal
•alirlai lla
SfllM
(SHI, 72(a)(1))
(II) Olslafttlaal Rasllaal
la Ika •tstfifeatlaa
Sfst«ai (Sill. 72(a)(4))
CfllUltk
HtwHiMti
R*tillttlaa
Ially
caasatalitt
Mailt (I)
aal <•. 2 a«/l
lat » 4
iaats
dalaclakla
la » 9SX al
Malkly
laaplas lat
aay lm taa-
OK al saa»las
If aa past f mi
« ctliatlaa
>MUI aal ass
*a»aal*
tlalallaa
lat >1 al 12
taayaatals
aal la plats
tlalaliaa II <(.2
a*/1 lai » 4 kaais
aalass Niaiiy
Apaaty Itlttaiaai
aaasaal aal
aapttllclakla
Malkly
tapatl
¦as I
feasiaass
i«y
Malkly
tapatl
aaaaal tapatl
•I aa-slla
laipictlaa
mil iaslaass
lay II <• ta«/l
lat aay ptilal
al tiM
Vts
Vas
tlalaliaa Malkly
aaliss lallata Is tapatl
aal taasil fey lallt-
laaty la saatci *alai
I it*lataI ' ,
steall*s
Malks
Tts
Y«« (II «• 2 a*/l
lat >4 kaats)
Vas
-------
1MK 7 1
•EQUIKKIIS IM IMriLKlID SVSKU (Caattaaai)
«'n"wil
J) lilmliil Caulv*l
Piaitaa
)) Oa-slla laspacllaa
4) IiIiiMih Dlsaasa
Oalbttak
S) 1•I•I CalIlata lull
t) talal trlkalaaalkaaa
¦•(•lailaa
cumin
la plata
aaaaal
aa aalbraaks
ratatiai
<1 paslllaa lai
ifiltai laklaf
<41 saaplas/aa,
40
plas/aa.
•. II a*/1
lar sfitaa
sartlai
>11. MS
¦mUhIm
acllvillas la
Mlaishai
aalatsbai
ptatraa >
llsiafacliaa
syslaa
pabllt baallk
IravHacy
laiai aa
papalatiaa
lililni far.
CwilHiir'
flllrallaa
(2)
Ma aalkiaak
¦ilk can tat
taallgaialiaa
aai saaica
aMat crllailaa
¦ »tk Milk
laartarly
lasafliclaal
ptagiaa as
lataraiaatf fey
Ptiaacy Afsacy
lasalfIciaal
pi a|i wi at
ialatalatl fey
Prtaaty A|»acy
aalbiaak allb
canaat caafiiai-
atlaa aal saaita
) ciliarlaa lai >1
al It taasacattaa
aaatks aalass
lailai* Is aal
caasal by a
ialUiaacy la
saarca Ml at
Haataaat
> crllailaa basal
aa aaaaal ataraga
BatlUf Maa ...
Iitmi Hnm tsllir '
aaaaal
rapail
aaaaal
rapatl
aail kaslaasi
aaalkly
rtpaii
Vas
1 ln.ffliiliiitfl nxiKi Ia i Itiilatii licktliai vlalillii
I. faliai* la laslall fIIIratlaa allkla II aaalks allar lallaia la aaal aallllatai sapply ttllarla rasalts la a traalaaal tacbalaa* alalallaa.
1. la latal aaaspapar allkla 14 iays al alalallaa aa* aall aallta allk bill ai by llsali ¦llkla 41 iays al alalallaa.
4. VI a I a 11aa Bay ka allaatl lar I al caasacatlta aaalks II Iba Ptiaacy A|a*cy ialaralaas aaa alalallaa la ba caasal by aaasaal aai aapraliclabla
(iitaatliacas.
5. Pdaati Afaacy aay iataralaa arbalbar alaqaata ilslalacllaa is praallal.
-------
"A
lititi
llliwi
ma i « •••«
m >
110 »/*«'•
!• MMIUHl
Hi
lIMM*
ma i < •••<•
«I1R I » "ifcwt
ssc
tl« t/Mll
It !••••! |M1
««»
litill
iiom
ma i « •¦••
1M » *•!•*«
xtt
i'l V/Mlt
I* IRMSIIWI
«A
ma i < »•••
1M > <•!*¦•(
«|f I«M ftts
tm i/i* ii
!• IMHIIHI
(I)
TFTP3
imim nwiij
nnntrrm
(0
isvrTTN>3
wriinwa
¦am aHini j aoi siamainftii
i-t navi
(t)(fiia« ••)
«a nia t n4to (»
ma i ii'*i t«**it«ii*ia. c»
«•» <•)
ma i Mti*iinj put mis «
(t)(nia i •! <•) ¦•M«niti
mai l i«iu '• (•
(inns) M'MHJ
H«|i»UI3 IHWllllli
-------
1MU 12
REQUItCMKNlS fM IHUIfB STIIll (Caal iaaal)
HmtfwiM
ftlilalatllaa tat riltiitl
Sappllas (SI41 »(»)}
CHlatlaa
jtaMiitm
Cfal iaata
(I)
Ptiiwit tetitt r«im
w
I) SapplMMit fllliallM
la a»ti Ovaiail
Tt*ala»at
it sptclfittf
Vtt
21) Blslaladaat lasliaal
Caltiiai SytlM
0. t af/l
caal laaaas;
syslHU « 31M
papain laa-|ial
saaplas
aal < i. I H|/l
lai > I Its
•ail
katlaass
<•»
Vcs
31) DUiilicital ¦•sllaal
la Disitlkaiiaa Systaa
ialttlailt
failiaal ff HfC
(¦¦pit
latallaa ft
Iftqaticy
tisil aa
papatattaa
•ppravitf iy
Slalt
aal < HCt la Matkly
> SB al aaalkly taparl
•(¦pits lai i
caasacalift aaalks
fas
EE ~
1 Vaa-CMpI laact malts la • tiaalatat lacfcaUaa tialaltaa.
t la lacal atvpaptr •llfcla 14 daft al tlalallaa aai Ml I aalica wllk kill ar k| llastlf talikla 41 lays al vlatatlaa.
). Pliancy Afiacy awy lilnalii aiatkai tflslaltcliaa Is piavliai.
4. II Pilaacy Aptacy aiaicltas tlictillaa.
-------
the ground water source is under the influence of surface water, whichever
Is liter. Within 18 months following the determination that a system is
under the influence of surface water, the Primacy Agency must determine,
using the sane , criteria that apply to systems using a surface water
source, whether the sys.tem must provide filtration treatment.,- As for
systems using a surface water source, the Primacy Agency must evaluate the
data on a case-by-case basis to determine conditions which will trigger
the need for filtration.
Beginning December 30,1991 or 18 months after the determination that
a system Is under the direct influence of surface water, -whichever Is
later, the criteria for avoiding filtration 1n S141.71(a) and (b) and the
requirements for unflltered systems in $141.71(c) and $141.72(a) go Into
effect, unless the Primacy Agency has determined that filtration is
required. As with systems using a surface water source, subsequent
failure to comply with any one of the criteria for avoiding filtration
requires the Installation of filtration treatment. Thus, beginning
December 30, 1991 or 18 months after the Primacy Agency determines that
a system 1s using a ground water source under the direct influence of
surface water, whichever is later, a system which fails to meet any one
of the criteria to avoid filtration must install filtration and comply
with the requirements for filtered systems within 18 months of the failure
or by June 29, 1993, whichever Is later. As for unflltered systems,
systems under the direct influence of surface water may apply for an
exemption to extend the time period for installing filtration.
Any system using a ground water source that the Primacy Agency
determines 1s under the direct influence of surface water and that already
has filtration in place at the time of the Primacy Agency determination
must meet the treatment technique, monitoring and reporting requirements
for filtered systems beginning June 29, 1993 or 18 months after the
Primacy Agency determination, whichever is later.
7 - 5
-------
7.5 Responses for Systems not Meeting ma Criteria
7.5.1 introduction
Systems *rfi1ch presently fill to meet the SWTR criteria say be able
to upgrade the system's design and/or operation and maintenance In order
to achieve compliance. The purpose of this section Is to present'options
which may be followed to achieve compliance.
7.5.2 Systems Hot filtering
Systems not filtering must meet the criteria to avoid, filtration
beginning December 30, 1991 and on a continuing basis thereafter or
Install filtration. Systems not filtering can be divided Into two
categories:
A. Those systems not currently meeting the SWTR criteria but with
the ability to upgrade to meet then.
B. Those systems not able to meet the SWTR criteria by December
30, 1991. If the Installation of filtration Is not possible
by June 29, 1993 the system may request an exemption and take
Interim measures to provide safe water to avoid violation of
a treatment technique requirement.
Systems 1n Category A
Example A - Response Situation
Condition; System 1s not meeting the source water fecal and/or
total conform concentrations but nas not received judgment on the
adequacy of Its watershed control.
Response Options••
- Monitor for fecal conforms rather than total col 1 forms 1f
this Is not already done. Fecal conforms are a direct
indicator of fecal contamination where total conforms are
not. If total conform levels are exceeded but fecal levels
•re not, the system meets the criteria.
- Take appropriate action 1n the watershed to assure fecal and
total conform concentrations are below the criteria, such as
elimination of animal activity near the source water Intake.
7 - 6
-------
Example B - Response Situation
Condition! System meets the source water quality criteria,
watershed control requlresents, and 1s maintaining a disinfectant
residual within the distribution system, but 1s not able to meet the
CT requirements due to lack of contact time prior to the first
customer.
Response Options;
Increase the application of disinfectant while monitoring THM
levels to ensure they remain below the MCL.
Add additional contact time through storage to obtain an
adequate CT.
Apply a more effective disinfectant such as ozone.
Systems 1n Category B
Example A - Response Situation
Condition; System meets the source water turbidity but not the
fecal conform requirements. A sewage treatment plant discharges
into the source water. A determination has been made that the
system does not have adequate watershed controls
Response Potions:
Purchase water from a nearby surveyor or use an alternate
source such as ground water If available.
Take steps to Install filtration, applying for an exemption
(time delay) as presented In Section 9 where appropriate.
Example I
Condition; The source water exceeds a turbidity of 5 NTU for more
than two periods 1n a year under normal weather and operating
conditions.
Response Options:
Purchase water from a nearby purveyor or use an alternate
source such as ground water If available.
Take steps to install filtration, applying for an exemption
(time delay) as presented 1n Section 9 where appropriate.
7 - 7
-------
In the Interim prior to adoption of either of the above options,
certain protective measures may be appropriate. One protective
measure which can be used would be the Issuance of a public notice
to boll all water for consumption during periods when the turbidity
exceeds 5 NTU. If such a notice Is Issued, the utility should
continue sampling the distribution system for chlorine residual and
total conforms, and Initiate measurement of HPCs In the distribu-
tion system. These data and the raw water turbidity should be used
to determine when to lift the boll water notice.
The notice could be lifted when:
The historical (prior to high turbidity) disinfectant residual
concentration 1s reestablished In the distribution system;
The total conform requirements are met;
The HPC count 1s less than 500/ml; and
The turbidity of the raw water Is less than 5 NTU.
7.4.3 Systems Currently Filtering
Systems which are currently filtering must meet the SWTR criteria
within 48 months of the SWTR to be In compliance, after which the criteria
must be continually met for the system to be In compliance.
Example A - Response Situation
Condition! A direct filtration plant Is treating a surface water
which 1s not compatible with this treatment process. The system 1s
not achieving Its required turbidity performance or disinfection
criteria.
Response Potions:
Optimize coagulant dose.
Reduce filter loading rates.
Evaluate the effect on performance of Installing flocculation
and sedimentation ahead of the filters.
7 - 8
-------
Example B ¦ Response
Condition! a filtration plant Is using surface water which Is
compatible with Its treatment system. The system 1s not achieving
disinfection performance criteria required by the Primacy Agency to
achieve a 1-log 1nact1vat1on of Giardi* cysts; however, It 1s
meeting the requirements of the Total Collform Rule.
Response Options;
Increase disinfectant dosage(s).
Install storage facilities to Increase disinfectant contact
time.
Ensure optimum filtration efficiency by:
Use of a filter aid.
Reduction 1n filter loading rates.
More frequent backwashlng of filters.
The Primacy Agency may grant additional removal credit for optimum
filtration.
EPA Intends to promulgate National Primary Drinking Water Regula-
tions to regulate levels of disinfectants and disinfectant by-product when
it promulgates disinfection requirements for ground water systems
(anticipated 1n 1992). EPA 1s concerned that changes required 1n
utilities' disinfection practices to meet the required Inactlvatlons for
the SWTR might be Inconsistent with treatment changes needed to comply
with the forthcoming regulations for disinfectants and disinfection
by-products. For this reason, the EPA Is allowing Primacy Agencies
discretion 1n determining the level of disinfection required for filtered
systems to meet the overall treatment performance requirements specified
In the rule or recommended based on source water quality.
During the Interim period, prior to promulgation of the disinfection
by-product regulation, EPA recommends that the Primacy Agency allow more
credit for 61ard1a cyst and virus removal than generally recoonended.
This interim level 1s recommended 1n cases where the Primacy Agency
determines that a system 1s not currently at a significant risk from
microbiological concerns at the existing level of disinfection and that
7-9
-------
a deferral Is necessary for the system to upgrade Its disinfection process
to optimally achieve compliance with the SWTR as well as the forthcoming
disinfection by-product regulations. Section 5.5.3 presents some
guidelines for establishing interim disinfection requirements.
7-10
-------
8. PUBLIC HQTITOTWI
The SNTR specifies that the public notification requirements of the
Safe Drinking Water Act (SDWA) and the inpleoentlng regulations of 40 CFR
Paragraph 141.32 must be followed. These regulations divide public
notification requirements Into two tiers. These tiers are defined as
follows:
1. Tier 1:
a. Failure to comply with MCL
b. Failure to comply with prescribed treatment.technique
c. Failure to comply with a variance or exemption schedule
2. Tier 2:
a. Failure to comply with monitoring requirements
b. Failure to comply with a testing procedure prescribed
by a NPOWR
c. Operating under a variance/exemption. This 1s not
considered a violation but public notification is
required.
The SWTR classifies violations of Sections 141.70, 141.71(c),
141.72 and 141.73 (i.e., treatment technique requirements as specified in
Section 141.76) as Tier 1 violations and violations of Section 141.74 as
Tier 2 violations. Violations of 141.75 (reporting requirements) do not
require public notification.
There are certain general requirements which all public notices must
meet. All notices must provide a clear and readily understandable
explanation of the violation, any potential adverse health effects, the
population at risk, the steps the system 1s taking to correct the
violation, the necessity of seeking alternate water supplies (1f any) and
any preventative measures the consumer should take. The notice must be
conspicuous, not contain any unduly technical language, unduly small print
or similar problems. The notice must Include the telephone number of the
owner or operator or designee of the public water system as a source of
additional Information concerning the violation where appropriate. The
notice must be b1- or multilingual if appropriate.
In addition, the public notification rule requires that when
providing information on potential adverse health effects 1n Tier 1 public
8-1
-------
notices and In notices on the granting and continued existence of a
variance or exemption, the owner or operator of a public water system must
include certain mandatory health effects language. For violations of
treatment technique requirements for filtration and disinfection, the
mandatory health effects'language is;
Microbiological Contaminants
The United States Environmental Protection Agency (EPA) sets drinking
water standards and has determined that microbiological contaminants are
a health concern at certain levels of exposure. If water 1s Inadequately
treated, microbiological contaminants in that water may cause disease.
Disease symptoms may include diarrhea, cramps, nausea, end possibly
jaundice and any associated headaches, and fatigue. These symptoms,
however, are not just associated with disease-causing organisms in
drinking water, but also may be caused by a number of factors other than
your drinking water. EPA has set enforceable requirements for treating
drinking water to reduce the risk of these adverse health effects.
Treatment such as filtering and disinfecting the water removes or destroys
microbiological contaminants. Drinking water which is treated to meet EPA
requirements is associated with little to none of this risk and should be
considered safe.
further, the owner or operator of a community water system must give
a copy of the most recent notice for any Tier 1 violations to all new
billing units or hookups prior to or at the time service begins.
The medium for performing public notification and the time period
in which notification must be sent varies with the type of violation and
1s specified in Section 141.32. For Tier 1 violations (I.e., violations
of Sections 141.70, 141.71, 141.72 and 141.73), the owner or operator of
a public water system must give notice:
1. ly publication in a local daily newspaper as soon as possible
but in no case later than 14 days after the violation or
failure. If the area does not have a daily newspaper, then
notice shall be given by publication in a weekly newspaper of
general circulation in the area, and
2. By either direct mail delivery or hand delivery of the notice,
either by Itself or with the water bill not later than 45 days
after the violation or failure. The Primacy Agency may waive
this requirement if it determines that the owner or operator-
has corrected the violation within the 45 days.
8 - 2
-------
Although the SWTR does not specify any acute violations, the Primacy
Agency say specify some Tier ! violations as posing an acute risk to human
health; for example these violations may include:
1. A waterborne disease outbreak In an unflltered supply.
2. Turbidity of the water prior to disinfection of an unflltered
supply or the turbidity of filtered water exceeds 5 NTU at any
time.
3. Failure to maintain a disinfectant residual of at least 0.2
mg/1 in the water being delivered to the distribution system.
For these violations or any others defined by the Primacy Agency as
"acute" violations, the system must furnish a copy of the notice to the
radio and television stations serving the area as soon as possible but in
no case later than 72 hours after the violation. Depending upon circum-
stances particular to the system, as determined by the Primacy Agency, the
notice may instruct that all water should be boiled prior to consumption.
Following the initial notice, the owner or operator must give notice
at least once every three months by mail delivery (either by itself or.
with the water bill), or by hand delivery, for as long as the violation
or failure exists.
There are two variations on these requirements. First, the owner
or operator of a comnunity water system in an area not served by a daily
or weekly newspaper must give notice within 14 days after the violation
by hand delivery or continuous posting of a notice of the violation. The
notice must be In a conspicuous place in the area served by the system and
wist continue for as long as the violation exists. Notice by hand
delivery must be repeated at least every three months for the duration of
the violation.
Secondly, the owner or operator of a noncommunitv water system
(i.e., one serving a transitory population) may give notice by hand
delivery or continuous posting of the notice in conspicuous places in the
area served by the system. Notice must be given within 14 days after the
violation. If notice 1s given by posting, then it must continue as long
8 - 3
-------
as the violation exists. Notice given by hand delivery must be repeated
at least every three months for as long as the violation exists.
For Tier 2 violations (I.e., violations of 40 CFR 141.74, analytical
and monitoring requirements) notice must be given within three months
after the violation by publication In a dally newspaper of .general
circulation, or 1f there Is no dally newspaper, then In a weekly
newspaper. In addition, the owner or operator shall give notice by mall
(either by Itself or with the water bill) or by hand delivery at least
once every three months for as long as the violation exists. Notice of
a variance or exemption must be given every three months from the date It
1s granted for as long as 1t remains In effect.
If the area 1s not served by a dally or weekly newspaper, the owner
or operator of a community water system must give notice by continuous
posting 1n conspicuous places 1n the area served by the system. This must
continue as long as the violation does or the variance or exemption
remains In effect. Notice by hand delivery must be repeated at least
every three months for the duration of the violation or the variance of
exemption.
For noncommunlty water systems, the owner or operator may give
notice by hand delivery or continuous posting 1n conspicuous places;
beginning within 3 months of the violation or the variance or exemption.
Posting must continue for the duration of the violation or variance or
exemption and notice by hand delivery must be repeated at least every
3 months during this period.
The Primacy Agency may allow for owner or operator to provide less
frequent notice for minor monitoring violations (as defined, by the
Primacy Agency If EPA has approved the Primacy Agency's substitute
requirements contained 1n a program revision application).
To provide further assistance In preparing public notices, several
examples have been provided. However, each situation Is different and
may call for differences 1n the content and tone of the notice. All
notices must comply with the general requirements specified above.
8 - 4
-------
Example 1 - Her 1 Violation-Unfilter»d S.mnlv
Following Is an example of a Tier 1 violation which nay be
considered by the Primacy Agency to pose an acute risk to human health.
A system which does not apply filtration experiences a breakdown In
the chlorine feed systems and the switchover system falls to activate the
backup systems. A number of hours pass before the operator discovers the
malfunction. The operator, upon discovery of the malfunction, contacts
the local television and radio stations and announces that the public Is
receiving untreated water. The announcement may read as follows:
We have just received word from the Aswan Water Board that a
malfunction of the disinfection system has allowed untreated water
to pass Into the distribution system. Thus, this system providing
drinking water 1s In violation of a treatment technique requirement.
The United States Environmental Protection Agency (EPA) sets
drinking water standards and has determined that microbiological
contaminants are a health concern at certain levels of exposure.
If water Is Inadequately treated, microbiological contaminants 1n
that water may cause disease. Disease symptoms may Include
diarrhea, cramps, nausea, and possibly jaundice and any associated
headaches, and fatigue. These symptoms, however, are not Just
associated with disease-causing organisms In drinking water, but
also may be caused by a number of factors other than your drinking
water. EPA has set enforceable requirements for treating drinking
water to reduce the risk of these adverse health effects. Treatment
such as filtering and disinfecting the water removes or destroys
microbiological contaminants. Drinking water which Is treated to
meet EPA requirements 1s associated with little to none of this risk
and should be considered safe.
The temporary breakdown In disinfection may have allowed micro-
organisms to pass into the distribution system. The operation of
the system has been restored so that no further contamination of
the distribution system will occur. Any further changes will be
announced.
Additional Information is available at the following number:
235-WATER.
A direct mailing of the notice is provided within 45 days of the
occurrence.
Example 2 - Tier 1 Violation-Unfiltered SudpIv
Following is an example of a Tier 1 violation which may be
considered by the Primacy Agency to pose an acute risk to human health.
8 - 5
-------
A system supplies an unfiltered surface water to its customers.
During a period of unusually heavy rains caused by a hurricane in the
area, the turbidity of the water exceeds 5 NTU. The turbidity data during
which the heavy rains occur 1s as follows:
Pay 1 ITU Day 2 NTU Day 3 NTU Dav 4 ntu Qav s ntu
0.4 0.8 0.7 0.7 7.6
0.4 0.5 0.4 7.6 3.1
0.5 0.5 0.4 11.3 2.7
0.7 0.4 0.5 9.6 0.7
1.1 0.4 0.4 7.2 0.8
0.9 0.6 0.6 5.0 0.5
The following public notice was prepared and submitted to the local
newspaper, television and radio stations within 72 hours of the first
turbidity exceedence of 5 NTU.
The occurrence of heavy rains in our watershed is causing a rise 1n
the turbidity of the drinking water supplied by Fairfax Water
Company.
Turbidity is a measurement of particulate matter in water. It is
of significance 1n drinking water because irregularly shaped
particles can both harbor microorganisms and interfere directly with
disinfection which destroys microorganisms. While the particles
causing the turbidity may not be harmful or even visible at the
concentrations measured, the net effect of a turbid water 1s to
Increase the survival rate of microorganisms contained In the water.
This 1s of concern because several diseases are associated with
waterborne microorganisms.
Because of the high turbidity levels, the Fairfax system 1s 1n
violation of a treatment requirement set by the Environmental
Protection Agency (EPA)»
The United States Environmental Protection Agency (EPA) sets
drinking water standards and has determined that microbiological
contaminants are a health concern at certain levels of exposure.
If water Is inadequately treated, microbiological contaminants in
that water may cause disease. Disease symptoms may Include
diarrhea, cramps, nausea, and possibly jaundice and any associated
headaches, and fatigue. These symptoms, however, are not just
associated with disease-causing organisms 1n drinking water, but
also may be caused by a number of factors other than your drinking
water. EPA has set enforceable requirements for treating drinking
water to reduce the risk of these adverse health effects. Treatment
such as filtering and disinfecting the water removes or destroys
. microbiological contaminants. Drinking water which is treated to
8-6
-------
Met EPA requirements 1s associated with little to none of this risk
and should be considered safe.
In order to protect yourself from Illness, all water from the
Fairfax system used for drinking, cooking and washing dishes should
be boiled at a rolling boll for one minute.
The system 1s being closely monitored and a notice will be issued
when the water returns to an acceptable quality and no longer needs
to be boiled.
The utility continues sampling the distribution system.for chlorine
residual and total coll forms, and Initiates measurement of the HPCs in the
distribution system. The notice Is lifted when all the following are met:
The historical (prior to high turbidity) disinfectant residual
concentration is reestablished in the distribution system.
The total conform requirements are met.
The HPC count is <500/al„
The turbidity of the raw water 1s less than 5 NTU.
The Primacy Agency most decide whether the turbidity event was unusual or
unpredictable and whether filtration should be installed.
Example 3 - Tier i Violation - Filtered Suoolv
A conventional treatment plant is treating a surface water. A
malfunctioning alum feed system resulted in an increase of the filter
effluent turbidities. The effluent turbidity was between 0.5 and 1.0 NTU
in 20 percent of the samples for the month. The utility Issued a notice
which was published in a local dally newspaper within 14 days after the
violation. The notice read as follows;
During the previous month, the Baltic Mater Treatment Plant
experienced difficulties with the chemical feed system. The
malfunctions caused an effluent turbidity level above 0.5 NTU in 20
percent of the samples for the month. The current treatment
standards require that the turbidity must be less than 0.5 NTU in
95 percent of the monthly samples. The Baltic drinking water system
has thus been in violation of a treatment technique requirement.
The United States Environmental Protection Agency (EPA) sets
drinking water standards and has determined that microbiological
8 - 7
-------
contaminants are a health concern at certain levels of exposure.
If water Is Inadequately treated, microbiological contaminants In
that water nay cause disease. Disease symptoms may Include
diarrhea, cramps, nausea, and possibly jaundice and any associated
headaches, and fatigue. These symptoms, however, are not just
associated with disease-causing organisms 1n drinking water, but
also may be caused by a number of factors other than your .drinking
water. EPA has set enforceable requirements for treating drinking
water to reduce the risk of these adverse health effects. Treatment
such as filtering and disinfecting the water removes or destroys
microbiological contaminants. Drinking water which Is treated to
meet EPA requirements 1s associated with little to none of this risk
and should be considered safe.
The chemical, feed and switchover components of the system have been
repaired and -are 1n working order and turbidity levels are meeting
the standard. It Is unlilcely that Illness will result from the
turbidity exceedences previously mentioned because continuous
stringent disinfection conditions were In effect and the system was
In compliance with other microbiological drinking water standards
pertaining to microbiological contamination. However, a doctor
should be contacted 1n the event of Illness. For additional
Information call, 1-800-726-WATER.
8 - 8
-------
9. EXEMPTIONS
9.1 Qytrylew of Rgqyj
Section 1416 pf the Safe Drinking Water Act allows a Primacy Agency
to exenpt any public water system within Its jurisdiction from any
treatment technique requirement Imposed by a national primary drinking
water regulation upon a finding that:
1. Oue to compelling factors (which may Include economic
factors), the public water system Is unable to comply with the
treatment technique requirement;
2. The pub.llc water system was In operation on the effective date
of the treatment technique requirement or, for a system that
was not In operation by that date, only 1f no reasonable
alternative source of drinking water 1s available to the new
system; and
3. The granting of the exemption will not result 1n an unreason-
able risk to health.
If a Primacy Agency grants a public water system an exemption, the
Agency must prescribe, at the time the exemption 1s granted, a schedule
for:
1. Compliance (Including Increments of progress) by the public
water system with each treatment technique requirement with
respect to which the exemption was granted; and
2. Implementation by the system of such control measures as the
Primacy Agency nay require during the period the exemption is
1n effect.
Before prescribing a schedule, the Primacy Agency must provide
notice and opportunity for a public hearing on the schedule. The schedule
prescribed oust require compliance by the public water system with the
treatment technique requirement as expeditiously as practicable, but in
no case later than one year after the exemption Is Issued (except that,
1f the system meets certain requirements, the final date for compliance
may be extended for a period not to exceed three years from the date the
exemption 1s granted). For systems serving less than 500 service
9 - 1
-------
connections, and meeting certain additional requirements, the Primacy
Agency My renew the exemption for one or more additional two-year
periods.
Under the SWTR, no exemptions are allowed from the requirement to
provide disinfection for surface water systems, but exemptions are
available to reduce the degree of disinfection required. Exemptions from
the filtration requirements are available. The following sections present
guidelines for evaluating conditions under which exemptions are appropri-
ate.
9.2 Recommended Criteria
In order to obtain an exemption from the SWTR, a System must meet
certain minimum criteria to assure no unreasonable risk to health. These
should be applied before looking at other factors such as economics..
Recommended minimum criteria for assuring no unreasonable risk to health
exists are listed below.
Systems which do not provide filtration
Practice disinfection to achieve at least a 2-log 1nact1vation
of Giardia cysts; or comply with the disinfection requirements
for the distribution system as defined 1n Section 141.72(b)
of the SWTR.
Comply with the monthly coliform MCI; or provide bottled water
or another alternate water source) or point of use treatment
evlces for their customers 1n which representee samples
comply with all the MCL National Primary Drinking Water
Regulations.
EPA recommends that 1n order to obtain an extension to the initial
1 year exemption period 1n addition to the required elements 1n Section
1416, the system would need to be 1n compliance with the monthly coliform
MCL, satisfy the above disinfection criteria and not have any evidence of
waterbome disease outbreaks attributable to the system at the end of that
first exemption period. If at any point during the extended exemption
period the system did not meet these conditions, the exemption should be
t withdrawn and the system should be subject to an enforcement action.
9 - 2
-------
Systems which provide filtration
Practice disinfection to achieve at least a 0.5 log Inactlva-
tlon of Giardia cysts; or comply with the disinfection
requirements for ' the distribution system as defined In
Section 141.72 of the rule.
Comply with the monthly coHform MCI; or provide bottled water
or another alternate water source) or point of use treatment
evlces for their customers 1n which representee samples
comply with all the MCL National Primary Drinking Water
Regulations.
Take all practical steps to Improve the performance of Its
filtration system.
In order to obtain an extension to the Initial exemption period, 1n
addition to the required elements In Section 1416, the system should be
In compliance with the conform MCL, satisfy the above disinfection
criteria and not have any evidence of waterbome disease outbreaks
attributable to the treatment system at the end of that first exemption
period. If at any point during the extended exemption period the system
did not meet these conditions, the exemption should be withdrawn and the
system should be subject to an enforcement action. In addition, the
system must continue to be taking steps to Improve the performance of Its
filtration system to achieve the criteria specified 1n the SWTR.
Once these minimum requirements are applied, the Primacy Agency
should look at the other factors as described In Sections 9.3, 9.4, and
9.5.
9.3 Compelling Factors
Compelling factors are often associated with small systems. The
major coapelllng factor tends to be economic. In some cases the
compelling factor may not be solely economic, but rather the contractual
and physical. 1nfeas1b111ty of having a required treatment Installed within
the time period specified In the regulation. For example, 1t may not be
feasible for a very large system to Install filtration by June 1993 if
required. In such cases exemptions are also appropriate. Additional
considerations for small systems are presented In Appendix L.
9 - 3
-------
If system Improvements necessary to comply with the SWTR incur costs
which the Primacy Agency determines pose an economic barrier to acquisi-
tion of necessary treatment, the system fulfills the criteria of
demonstrating a compelling hardship which makes it unable to meet the
treatment requirements. In such cases, the EPA believes It Is reasonable
to grant an exemption If the system also meets the criteria 1n 9.4 and
9.5.
The USEPA document, "Technologies and Costs for the Removal of
Microbial Contaminants from Potable Hater Supplies," contains costs
associated with available treatment alternatives (USEPA, 1988b). Costs
found 1n this document, or those generated from more site-specific
conditions, can be used as the basis for determining the ability of a
system to afford treatment. The total annual water production costs per
household for a system can be estimated based on the household water usage
and the production costs per thousand gallons. As estimated In the above
cited USEPA document, each cent per thousand gallons of treated water 1s
approximately equivalent to $1 per year per household 1f a household water
usage of 100,000 gallons per year 1s assumed.1 This estimate will need to
be adjusted according to water usage for cases where the household usage
differs from 100,000 gallons per year.
The following examples are presented to provide guidance In
estimating costs for a system to upgrade Its system or Install filtration.
This cost Information could be used for determining whether a system might
be eligible for an exemption.
Example 1
A water system which supplies an average daily flow of 0.05 mgd to
a small urban coamunlty receives Its water supply from a lake. The system
currently provides disinfection with chlorine but does not provide
filtration. The system reviewed Its source water quality and found the
characteristics to be as follows:
% ^
*
*t 1 This 1s the national average residential household consumption reported
in: Final Descriptive Summary - 1986 Survey of Community Water Systems.
October 23, 1987. USEPA: Office of Drinking Water.
9-4
-------
Total conforms 1,000/100 ml
Turbidity 10 - 13 NTU
¦Color 6 - 9 CU
Based upon.the criteria In the SWTR, this source requires filtration
and a review of the water quality criteria presented 1n Table 4-2
Indicates that the treatment technique best suited to these source
conditions Is conventional treatment. A conventional package treatment
plant with a capacity of 0.068 MGD may be purchased and put on line at
a cost of $277/household-year not Including real estate, piping or raw
water pumping costs which may be significant depending on .the plant
location.2 EPA has estimated that, on average, these costs might add
another 50% depending on site specific factors (USEPA, 1989)
Thus the cost estimate for implementing filtration Indicates that
the increase 1n the average annual household water bill would be
approximately $277 plus the cost of real estate, piping, and raw water
pumping as needed. The Incomes of people 1n the community and the current
water bills can be reviewed by the Primacy Agency along with these
estimated costs to determine if an undue economic hardship 1s Incurred by
these treatment methods. Upon determination that an economic hardship is
incurred, the Primacy Agency may grant an exemption from filtration,
provided that the system can assure the protection of the health of the
community. However, 1f the water supply system for a nearby community
meets the drinking water standards mi there is the ability to hook up to
that system, an exemption generally should not be granted unless such
costs also presented an economic hardship.
Example 2
A large urban community, with a median annual Income of $25,000 per
family, 1s supplied with water from lakes and reservoirs. The community
places an average dally demand of 3 mgd on the supply system. The
watershed of the system is moderately populated and used for fanning and
2 Table VI-3 ("Technologies and Costs for the Removal of Microbial
Contaminants From Potable Hater Supplies," USEPA, 1988b) lists the total
costs as 277.4 cents/1000 gal. Estimated costs for real estate, piping
and raw water pumping as a function of site specific conditions are
available in Table E-l, E-2, and E-3 of this same document.
9-5
-------
grazing. The system currently provides filtration using dlatomaceous
earth filtration and disinfection with chloramlnes.
A review of the source and finished water quality was conducted to
evaluate the plant's performance. The source water quality was determined
to be:
Total conforms 30 - 40/100 ml
Turbidity 2 - 3 NTU
Color 1 - 2 CU
Dlatomaceous earth Is therefore an acceptable filtration method.1
However, review of the finished water showed that a residual 1n the
distribution system 1s only maintained 80 percent of the time. In
addition to this, conforms were detected 1n 10 percent of the samples
taken over the twelve month period. Inspection of the chlorlnatlon
equipment showed the equipment 1s deteriorated. Review of the monthly
reports showed that the conforms appeared 1n the distribution system
shortly after the chlorlnators malfunctioned. This observation led to the
conclusion that new disinfection facilities were needed.
The source water quality and available contact time after disinfec-
tion were then used to determine the most appropriate disinfectant for the
system. As described 1n Section 5.5, ozone, chlorine or chlorine dioxide
can be used as primary disinfectants given these conditions. A prelimi-
nary review of costs for applying the various disinfectants showed
chlorine to be the most economical at a cost of $2.8/househo1d/year4
(USEPA, 1988b). This cost does not Include backup equipment; however,
even with providing duplicate equipment doubling this cost to $5.6/houSe-
hold/ year, the Improvement Incurs minimal cost and the Primacy Agency
should not grant the system an exemption based on economic hardship.
3 As determined from Table 4-2 of Section 4.
4 Table VI-12 (USEPA, 1988b) lists a total cost of 2.8 cents/1000 gal for
a plant capacity of 5.85 mgd.
(2.8 cents) YSl/household-vear^ - $2.8/household-year
(1,000 gal) (cents/1000 gal)
9-6
-------
9.4 Evaluation of Alternate Water Supply Sources
Systems which would Incur very high costs for Installing a required
treatment to comply with the SWTR, should evaluate the possibility of
using an alternate source. These alternate sources Include:
The use of ground water
Connection to a nearby water purveyor
Use of an alternate surface water supply
When considering the use of ground water, the purveyor must
determine the capacity of the underlying aquifer for supplying the demand.
The water quality characteristics of the aquifer must be evaluated to
determine what treatment may be needed to meet existing standards. The
cost of the well construction and treatment facilities must then be
determined and converted Into a yearly cost per household.
The connection to a nearby purveyor Involves contacting the purveyor
to determine their capacity and willingness to supply the water. Once It
has been determined that the alternate source meets all applicable
drinking water standards, the cost of the transmission lines, distribution
system, and other facilities (e.g. disinfection, repumplng, etc.) must
then be determined and amortized Into a yearly cost per household.
If the cost for using an alternate source 1s found by the Primacy
Agency to present an economic hardship, and the purveyor can demonstrate
that there will be no unreasonable risk to health, the Primacy Agency may
grant an exemption to the SWTR for the purveyor and develop a schedule of
compliance.
9.5 Protection of Public Health
Systems which apply for an exemption from the SWTR must demonstrate
to the Primacy Agency that the health of the conmunlty will not be put at
risk by the granting of such an exemption. A system should be able to
provide adequate protection for the public health by meeting the minimum
suggested EPA requirements In Section 9.2. However, a Primacy Agency may
specify additional measures or criteria a system must meet to protect
public health, depending on the particular circumstances. Systems with
currently unfiltered surface water supplies which fall to meet the source
9 - 7
-------
water quality criteria will be required to Install filtration as part of
their treatment process. However, 1t My take 3 to 5 years or Bore before
the filtration system can be designed, constructed and begin operation,
thereby justifying the granting of an exemption. Ourlng this period,
possible Interim measures which the system could take to further satisfy
the Primacy Agency's concern include one or more of the following:
a. Use of higher disinfectant dosages without exceeding the TTHM
MCL (even for systems not currently subject to this MCL)
b. Installation of a replacement or additional disinfection
system which provides greater disinfection efficiency and
which can be Integrated Into the new filtration plant
c. Increasing the monitoring and reporting to the Primacy Agency
d. Increasing protection of the watershed
e. Increasing the frequency of sanitary surveys
f. Temporarily purchasing water from a nearby water system
g. For small systems, temporary Installation of a mobile
filtration (package) plant
h. Increasing contact time by rerouting water through reservoirs
In some cases systems may be able to increase their disinfection
dosages during the Interim period to provide additional protection against
pathogenic organisms. This alternative should be coupled with a
requirement for Increased monitoring for conforms, HPC and disinfectant
residual within the distribution system. However, disinfectant dosage
should not be Increased If this would result 1n a violation of the TTHM
MCL, even for systems not currently subject to this MCL.
Systems which are planning to Install filtration may be able to
utilize a more efficient disinfectant that can later be Integrated Into
the filter plant. Currently ozone and chlorine dioxide are considered to
be the most efficient disinfectants.
For all systems which do not meet the source water quality criteria
iOd must Install filtration, EPA recommends that during the interim period
the Primacy Agency increase its surveillance of the system and require
9 - 8
-------
increased monitoring and reporting requirements to assure adequate
protection of the public health.
Any required Increases In watershed control and/or on-site
Inspections will not alleviate the need for more stringent disinfection
requirements and Increased monitoring of the effectiveness of the system
employed. Their purpose Mould be to Identify and control all sources of
contamination so that the existing system will provide water of the best
possible quality.
For some systems, It may be possible to purchase water from a nearby
system on a temporary basis. This may involve no more than the use of
existing interconnections or it may require the Installation of temporary
connections.
Trailer mounted filtration units (package plants) are sometimes
available from state agencies for emergencies or may be rented or leased
from equipment manufacturers.
Systems may also be required to supply bottled water or install
po1nt-of-entry (POE) treatment devices. For the reasons listed below,
these alternatives should only be utilized if the previously mentioned
alternatives are not feasible:
In many states bottled water is subject only to the water
quality requirements of the FDA as a beverage and not to the
requirements of the Safe Drinking Mater Act.
- . Point-of-entry treatment devices are not currently covered by
performance or certification requirements which would assure
their effectiveness or performance.
If the Installation of POE devices is required, the selection of the
appropriate treatment device should be based upon a laboratory or field
scale evaluation of the devices. A guide for testing the effectiveness
of POE units In the microbiological purification of contaminated water is
provided in Appendix N.
Several Issues arise with the use of POE devices. These include
establishing who or what agency (1) has the responsibility for ensuring
compliance with standards; (2) retains ownership of the treatment units;
(3) performs monitoring, analyses and maintenance; and (4) manages the
9 - 9
-------
treatment program and maintains Insurance coverage for damage and liabil-
ity. It should also be considered that there Is no significant Increase
In risk over centrally treated water.
These Issues should be borne 1n mind when POE as a treatment
alternative 1s being considered.
Systems with currently unflltered surface water supplies which meet
the source water quality criteria, but do not meet one or more of the
other requirements for watershed control, sanitary survey, compliance with
annual collform MCL or disinfection by-product regulatlon(s), will be
required to install filtration unless the deficiencies can -be corrected
within 48 months of promulgation of the SWTR. Interim protection measures
Include those previously listed.
Systems with currently unflltered surface water supplies which meet
the source water quality criteria and the site specific criteria but which
do not meet the disinfection requirements, will be required to Install
filtration unless the disinfection requirements (adequate CT and/or
disinfection system redundancy) can be met. During the Interim period,
available options Include:
a. Temporary Installation of a mobile treatment plant
b. Temporary purchase of water from a nearby purveyor
c. Increased monitoring of the system'
d. Installation of temporary storage facilities to Increase the
disinfectant contact time
Currently filtered supplies which fall to meet the turbidity or
disinfection performance criteria presented 1n Section 5 will be required
to evaluate and upgrade their treatment facilities In order to attain
compliance. During the Interim period available options for Improving the
finished water quality Include: ,
a. Use of a filter aid to Improve filter effluent turbidities
b. Increased disinfectant dosages
c. The addition of an alternate disinfectant 1s an option after
the disinfection by-products rule 1s promulgated
9 - 10
-------
d. Reduction In filter loading rates with subsequent reduction
In plant capacity
e. Installation of temporary storage facilities to increase
disinfectant contact tine
9.6 Notification to EPA
The SDWA requires that each Primacy Agency which grants an exemption
notify EPA of the granting of this exemption. The notification must
contain the reasons for the exemption, Including the basis for*the finding
that the exemption will not result In an unreasonable risk to public
health and document the need for the exemption.
9 - 11
-------
REFERENCES
-------
REFERENCES
Ali-Anl, M.; McElroy, J. M.; Hlbler, C. P.; Hendricks, 0. W. Filtration
Of Slanlli Cysts and ether Substances, Volume 3s Riold Rate Filtration.
EPA-600/2-85-027, U.S. Environmental Protection Agency, WERL, Cincinnati,
Ohio, April, 1985.
American Public Health Association; American Water Works Association;
Water Pollution Control Federation. Standard Methods for the Examination
of Water and Wastewater. 16th ed., pp. 134-6, 298-310, 827-1038, 1985.
American Public Health Association; American Water Works Association;
Water Pollution Control Federation. Standard Methods far the Examination
Bf Water and Wastewater. 17th ed., 1989.
American Water Works Association. Manual of Water Supply Practices and
Water Chlorination Principles and Practices, 1973.
American Water Works Association Research Foundation (AWWARF). A Summary
of State Orinking Water Regulations and Plan Review Guidance. June, 1986.
Bader, H.; Hoigne, J. Determination of Ozone 1n Water by the Indigo
Method, Water Research 15,* 449-454, 1981.
Bellamy, W. D.; Lange, K. P.; Hendricks, D. W. Filtration of Glardia Cysts
and Other Substances. Volume 1: Diatomaceous Earth Filtration.
EPA-600/2-84-114, U.S. Environmental Protection Agency, Cincinnati, Ohio,
1984.
Bellamy, W. D.; Silverman, G. P.; Hendricks, D. W. Filtration of Glardia
Cysts and Other Substances. Volume 2: Slow Sand Filtration. EPA-600/2-
-85-026, U.S. Environmental Protection Agency, MERL, Cincinnati, Ohio,
April, 1985.
Bishop, S.; Craft, T. F.; Fisher, D. R.; Ghosh, M.; Prendiville, P.W.;
Roberts, K. J.; Steimle, S.; Thompson, J. The Status of Direct Filtration,
Committee Report. J.AWWA, 72(7){405-411, 1980.
Bouwer, H. Ground Water Hvdroloov. McGraw Hill Book Co., New York,
pp. 339-356, 1978.
Brown, T. S.j Malina, J.- F., Jr.; Moore, B. D. Virus Removal by Diatoma-
ceous Earth Filtration - Part 1 & 2. J.AWWA 66(2):98-102, (12)5735-738,
1974.
Bucklin, K.; Amirtharajah, A.; Cranston, K. Characteristics of Initial
Effluent Quality and Its Implications for the Filter-to-Waste Procedure.
AWWA Research Foundation Report. November, 1988.
-------
Carlson, O.A.; Seatloom, R.W.; DeWalle, F.O.; Wetzler, T.F.; Evgeset, J.;
Butler, R.; Wangsuthachart, S.; Wang, S. Ultraviolet Disinfection of
Water for Small Water Systems. EPA/600/2-85-092, U.S. Environmental
Protection Agency, Water Engineering Reserach Laboratory, Drinking Water
Research Division, Cincinnati, Ohio, September, 1985.
Clark, R.M.; Reg11, S. -A Mathematical and Statistical Analysis for the
Inactlvation of Giardia lamb!1a by Free Chlorine. Submitted to the
Journal of Environmental Science Engineering, 1989.
Clark, R.; Regli, S.; Black, D. Inactlvation of Giardia lamb!la by Free
Chlorine: A Mathematical Model. Presented at AWWA Water Quality
Technology Conference. St. Louis, Mo., November 1988.
Cleasby, J. L.; Hllmoe, D. J.; Oimltracopoulos, C. J. Slow-Sand and Direct
In-L1ne Filtration.of a Surface Water. J.AWWA, 76(12):44-5S, 1984.
DeWalle, F. B.; Engeset, J.; Lawrence, W. Removal of Giardia Iambiia
Cysts by Drinking Water Plants. EPA-600/S2-84-069, United States En-
vironmental Protection Agency, MERL, Cincinnati, Ohio, May 1984.
Fox, K. R.; Miltner, R. J.; Logsdon, G. S.; Dicks, 0. L.; Drolet, L. F.
Pilot Plant Exploration of Slow Rate Filtration. Presented at the AWWA
Annual Conference Seminar, Las Vegas, Nevada, June 1983.
Fujioka, R.; Kungskulniti, N.; Nakasone, S. Evaluation of .the Presence
- Absence Test for Conforms and the Membrane Filtration Method for
Heterotrophic Bacteria. AWWA Technology Conference Proceedings, November,
1986.
Geldreich, E. Personal communication to Linda Averell, Malcolm Pirnie
Engineers, Paramus, New Jersey, July 1989.
Great Lakes-Upper Mississippi River Board of State Public Health and
Environmental Managers Committee. Recommended Standards for Water Works.
1987 Edition.
Hendricks, D.; Al-Ani, M.; Bellamy, W.; Hibler, C.; McElroy, J. Surrogate
Indicators for Assessing Removal of Giardia Cysts, AWWA Water Quality
Technology Conference, 1984.
Hoff, J. C. Inactlvation of Microbial Agents bv Chemical Disinfectants*
EPA-600/S2-86-067, U.S. Environmental Protection Agency, Water Engineer-
ing Research Laboratory, Drinking Water Research Division, Cincinnati,
Ohio, September 1986.
Hoffbuhr, J. W.; Blair, J.; Bartleson, M.; Karlin, R. Use of Particulate
Analysis for Source and Water Treatment Evaluation. AWWA Water Quality
Technology Conference Proceedings, November 1986.
Horn, J. B.; Hendricks, D. W. Removals of Giardia Cysts and other
Part'icles from Low Turbidity Waters Using the Culligan Mult1-Tech Filtra-
-2-
-------
tion System. Engineering Research Center, Colorado State University,
Unpublished, 1986.
Joost, R. 0.; Long, B. W.; Jackson, L. Using Ozone as a Primary
Disinfectant for the Tucson CAP Water Treatment Plant, presented at the
IOA/PAC Ozone Conference, Monroe, HI, 1988.
Kuchta, J. H.; States, S. J.; McNamara, A. M.; Uadowsky, R. M.; Yee, R. B.
Susceptibility of Legionella pneumophila to Chlorine in Tap Water. Appl.
Environ. Microbiol., 46(5): 1134-1139, 1983.
Letterman, R. 0. The Filtration Requirement In the Safe Orinklng Water
Act Amendments of 1986. U.S. EPA/AAAS Report, August 1986.
Logsdon, G. S.; Symons, J. M.: Hoye, Jr., R. L.; Arozarena, M. M.
Alternative Filtration Methods for Removal of Giardia Cysts and Cyst
Model. J.AWWA, 73:111-118, 1981.
Logsdon, G.; Thurman, V.; Frindt, E.; Stoecker, J. Evaluating Sedimenta-
tion and Various Filter Media for Removal of Giardia Cysts. J. AWWA,
77:2:61, 1985.
Logsdon, G. S. Report for Visit to Carrol Hon, Georgia, USEPA travel
report, February 12, 1987a.
Logsdon, G. S. Comparison of Some Filtration Processes Appropriate for
Giardia Cyst Removal. USEPA Drinking Water Research Division; Presented
at Calgary Giardia Conference, Calgary; Alberta, Canada/February 23-25,
1987b.
Long, R. L. Evaluation of Cartridge Filters for the Removal of Giardia
Iambi 1a Cyst Models from Drinking Water Systems. J. Environ. Health,
45(5):220-225, 1983.
Markwell, D. D., and Shortrldge, K. F. Possible Waterborne Transmission
and Maintenance of Influenza Viruses in Domestic Ducks. Applied and
Environmental Microbiology, Vol. 43, pp. 110-116, January, 1981.
Morand, J., M.; C. R. Cobb; R. M. Clark; Richard, G. S. Package Water
Treatment Plants, Vol. 1, A performance Evaluation. EPA-600/2-80-008a,
USEPA, MERL, Cincinnati, Ohio, July 1980.
Morand, J. M,; Young, M. J. Performance Characteristics of Package Water
Treatment Plants, Project Summary. EPA-600/52-82-101, USEPA, MERL,
Cincinnati, Ohio, March, 1983.
Muraca, P.; Stout, J. E.; Yu, V. L. Comparative Assessment of Chlorine,
Heat, Ozone, and UV Light for Killing Legionella pneumophila Within a
Model Plumbing System. ApplH Environ. Microbiol., 53(2)-.447-453, 1987.
Notestine, T., Hudson, J. Classification of Drinking Water Sources as
Surface or Ground Waters. Final Project Report. Office of Environmental
-------
Health Programs. Washington Department of Social and Health Services,
August 1988.
Poynter, S. F. B.; Slade, J. S. The Removal of Viruses by Slow Sand
Filtration, Prog. Wat. Tech. Vol. 9, pp. 75-88, Pergamon Press, 1977.
Printed in Great Britain.
Randall, A.D. Movement of Bacteria From a River to a Municipal Well - A
Case History. American Water Works Association Journal. Vol. 62, No. 11,
p.716-720., November 1970.
Rice, E.W.; Hoff J.C. Inactivatlon of Slardli Iambi1a cvsts by
Ultraviolet Radiation. Appl. Environ. Microbiol. 42: 546-547, 1981.
Robeck, 6. 6.; Clarke, N. A.; Dostal, K. A. Effectiveness of Water
Treatment Processes In Virus Removal. J. AWWA, 54(10):1275-1290, 1962.
Robson, C.; Rice, R.; Fujikawa, E.; Farver, B. Status of U.S. Drinking
Water Treatment Ozonation Systems, presented at 10A Conference, Myrtle
Beach, SC, December 1988.
Rose, J. Cryptosporidium in Water; Risk of Protozoan Waterborne Trans-
mission. Report prepared for the Office of Drinking Water, U.S. EPA,
Summer, 1988.
Rubin, A. Factors Affecting the Inactlvation of Giardia Cysts by
Monochloramlne and Comparison with other Disinfectants. Water Engineering
Research Laboratory, Cincinnati, OH, March 1988a.
Rubin, A. "CT Products for the Inactlvation of Giardia Cysts by Chlorine,
Chloramine, Iodine, Ozone and Chlorine Dioxide" submitted for publication
In J. AWWA, December 1988b.
Slezak, L.; Sims, R. The Application and Effectiveness of Slow Sand
Filtration 1n the United States. J.AWWA, 76(12):38-43, 1984.
Sobsey, M. Detection and Chlorine Disinfection of Hepatltus A in Water.
CR-813-024. EPA Quarterly Report. December 1988.
Stolarik, G.; Christie, D. Projection of Ozone C-T Values Los Angeles
Aqueduct Filtration Plant, 1988.
U. S. Environmental Protection Agency, Office of Drinking Water, Criteria
and Standards Division. Manual for Evaluating Public Drinking Water
Supplies, 1971.
U. S. Environmental Protection Agency, Office of Drinking Water. Public
Notification Handbook for Drinking Water Suppliers, May 1978.
U. S. Environmental Protection Agency, Office of Ground Water Protection.
Guidelines for Applicants for State Wellhead Protection Program Assistance
Funds Under the SDWA, June 1987a.
-------
U. I. Environmental Protection Agency, Office of Ground Water Protection.
Guidelines for Delineation of Wellhead Protection Area, June 1987b.
U. S. Environmental Protection Agency, Office of Drinking Water. Workshop
on Emerging Technologies for Drinking Water Treatment, April, 1988a.
(J. S. Environmental Protection Agency, Office of Drinking Water.
Technologies and Costs for the Removal of Microbial Contaminants from
Potable Water Supplies, October, 1988b.
World Health Organization Collaborating Center. Slow Sand Filtration of
Community Water Supplies in Developing Countries. Report of.an Interna-
tional Appraiser Meeting, Nagpur, India, Bulletin Series 16, September
15-19, 1980.
-------
APPENDIX A
EPA CONSENSUS METHOD
FOR msm CYST ANALYSIS
-------
reSTIVC FOR 11.••¦33? A IS WATER
To beg.6: wo^g-acupA on -Ctiling, Jaij l/aiconczfccM gauc a tlidz picic
abadt"(hi {eating method uaecf in the Reg-om 10 Labo-xatcvj. The jcWcrict^
paaw aiti Appzndlx C 4 uimaxizz hU talk.
Methods of Testing for Ciardia in Water
(George (Jay; ;sccncelos, Regional
Microbiologist. Region 10 Laboratory,
Manchester, Washington)
Background:
Although recent development of an excystation technique by Drs. Bingham,
Meyer, Rice and Schaefer could in future lead to developing cultural methods,
at this time no reliable methods exist for culturing Giardia cysts from water
sanples. At present, the only practical method for determining the presence
of cysts in water is by direct microscopic examination of sarple concentrates.
Microscopic detection in water-sanple concentrates isn't an ideal process.
Finding and identifying the cysts relies almost entirely on the training,
skill, experience and persistence of the examiner. (And it is a skill not
widespread among water-supply laboratories.) But despite its limitations,
microscopic identification is currently the best method we have.
Years ago, the basic assunption was made that in order to find Giardia cysts
in water, some form of sanple concentration was necessary. As early as i?So, •.
labs were using membrane filters with a porosity of 0.45 fjm. With few exceptions
these attempts were unsuccessful. The center for Disease Control has tried
particulate filtration, with diatomaceous earth as the medium, This removed
the cysts from the water, but the cysts couldn't be separated from the
particles of diatomaceous earth.
With the recent increase in the incidence of waterbome giardiasis, further
efforts have been made to improve the detection method. An ideal method would
be one that recovers all cysts in a water sanple rapidly, cheaply and singly;
allows rapid detection, identification and quantification; and provides
information on the viability of and/or infectivity potential of cysts detected.
Unfortunately, no such method exists. The methods presently available
can be broadly separated into two general stages: primary concentration and
processing (see Table 1 on next page), and detection and identification
(see Table 2 on next page).
-------
tbtixc k» giardia in mm
Methods of Testing for Giardia in Water (Continued...!
nut i: mmi eoHCtntiiATiON m MoetssiNS wctnoss
HCThQQ
l. ££j££f£LilLL£iii£l
Ctllylont
Pslyeirtlnttt
(U3m.lu*l
t. >«rtlcu>«t» nUMtiow
is»il»ieia«i aifin, tini,
iu.)
i. *)«» Ctnt»ifuat
4, *2t822£jm£]]iC«tie«i£
5. !ae»y-*'B«ralm lilitan
TyCI
ilgu.Swo,
I, mcfooorotft itfn»8»tn Otatft.
TTTTTa or I on i pelyprolylant)
7. >«)liti" Cnn'.n SyUi"
I. filt»r««»mnq AflBintm
UttlSTlgATO* (%)
Cmnf | MBlar
t,S*HS, Its«
fjttr, Qwfritti I M«i»rj In)
1912, (wnpwftHinaa)
SMm « «1, 1177
No'*l" t£il< If 13
DwiS, hiinington
I'IMf, »»««« Stit< UN,
(m«Pm81iib»8)
«i9S«, CONS Lib, lirtli/, a
(unpuBlitntd)
Jittifrowiin, f nekton, 117} i
III!, E?A.£meiM«i
mlltport Cerp.
(unpuOlt$ntd)
OuUiIlt, u. of uI01
J.IIX
|*trie:ion i*t. SIS
My Bi ustful for
procitsmg fiHir
Minings
Cliias 7S: r«co»»r/
frw orlon filttrs
TABLE 2; DETECTION HtTHQOi
HtTHOD
mvESTlG
-------
TPSTTK'C FOR CIARDIA IN WATER
Methods of Testing for Ciardia in Water (Continued...)
Copies of Table 1 and Table 2 are also shown in Appendix C, along with
further detail about the methods.
EPA Consensus Method:
In September, 1980, the EPA convened a workshop on Giardia methodology in
Cincinnati. Its main pmpose was to identify the best available methodology,
and to agree on a reference method. Hie five labs in attendance recognized
that any proposed method would be based in large part on opinions and personal
preferences rather than on hard data, but that agreeing on a consensus method
would promote uniformity and provide a basis for future comparisons. Cur
lab has modified the EPA consensus method slightly for our use. This re thec
is outlined below.
Filter unwound into quarters
Rinsed in distilled water with polysorbate 20
Settled overnight, or centrifuged
1
Collect sediment and add 2% Formaldehyde in PBS
Settled overnight, or centrifuged
4
Collect sediment
-"1*
41 g.
Sucrose or
Percoll-sucrose
gradient
InS04 Flotation
I
Microscopic observation of the entire
concentrate (Bri ghtfieId/Phase-contrast)
-------
APPENDIX B
INSTITUTIONAL CONTROL OF LEGIONELLA
-------
APPENDIX B
INSTITUTIONAL CONTROL OF LE6IQNPI1 A
Legionella is a genus name for bacteria commonly found in lake and
river waters. Some species of this genus have been Identified as the
cause of the disease leglonellosis. In particular, Legionella pneumophila
has been Identified as the cause of Legionnaires disease, the pneumonia
form of leglonellosis and with Pontlac Fever, a nonpneumonia disease.
Outbreaks of leglonellosis are primarily associated with inhalation of
water aerosols or, less commonly, with drinking water containing
Legionella bacteria with specific virulence factors not yet Identified.
Foodbome outbreaks have not been reported (USEPA, 1985).
As discussed in this document, treatment requirements for disinfec-
tion of a municipal water supply are thought to provide at least a 3 log
reduction of Legionella bacteria (see Section 3.2.2). However, some
recontamination may occur in the distribution system due to cross
connections and during installation and repair of water mains. It has
been hypothesized that the low concentrations of Legionella entering
buildings due to these sources may colonize and regrow 1n hot water
systems (USEPA, 1985). Although all of the criteria required for
colonization are not known, large institutions, such as hospitals, hotels,
and public buildings with recirculating hot water systems seem to be the
most susceptible. The control of Legionella in health care institutions,
such as hospitals, is particularly Important due to the increased
susceptibility of many of the patients. The colonization and growth of
Legionella 1n drinking water primarily occurs within the consumer's
plumbing systems after the water leaves the distribution system..
Therefore, the control of these organisms must be the consumer's
responsibility. This appendix Is intended to provide guidance to these
institutions for the detection and control of the Legionella bacteria.
B.l MONITORING
It is suggested that hospitals, and other institutions with
potential for the growth of Legionella, conduct routine monitoring of
B-l
-------
their hot water systems at least quarterly.1 The analytical procedures
for the detection of these organisms can be found In Section 912.1
"Legjonellaceae" of the 16th edition of Standard Methods. Samples should
be taken at, or closely following, the hot water storage reservoir and
from a number of shower heads. It 1s recommended that showers with the
least frequent usage be Included In the sampling program. Follow-up
testing 1s suggested for all positive Indications prior to the Initiation
of any remedial measures. If the the presence of Legionella is confirmed,
then remedial measures should be taken. Although the regrowth of
Legionella Is commonly associated with hot water systems, hot and cold
water Interconnections may provide a pathway for cross contamination. For
this reason, systems detecting Legionella In hot water systems shpuld also
monitor their cold water systems.
B.2 TREATMENT
Because the primary route of exposure to Legionella Is probably
Inhalation, rather than Ingestion, It Is recommended that disinfection
procedures Include an Initial shock treatment period to disinfect shower
heads and hot water taps where the bacteria may colonize and later become
airborne. The shock treatment period should also Include disinfection of
hot water tanks. After this time, a polnt-of entry treatment system can
be Installed to provide continual disinfection of the hot water system.
B.2.1 Initial Disinfection
The most applicable method for the Initial disinfection of shower
heads and water taps is heat eradication. The fittings can be removed and
held at temperatures greater than 60 C for at least 24 hours. Disinfec-
tion of fittings can also be achieved by soaking or rinsing with a strong
chlorine solution. When soaking the fittings, a minimum chlorine strength
of 50 mg/L should be used for a period of no less than 3 hours. Rinsing
•»
Monitoring frequency based on the reported rate of Legionella regrowth observed
during disinfection studies (USEPA, 1985).
B-2
-------
with chlorine should be performed with sore concentrated solutions, Cire
must be taken not to corrode the finished surface on the fittings..
Commercially available bleaches, for example, are typically 5.25 percent
chlorine by weight.
B.2.2 Long-Term Disinfection
Heat - Numerous studies have shown that increasing the hot water
temperature to 50 - 70 C over a period of several hours nay help to reduce
and inhibit Legionella populations. However, .some Instances of regrowth
after 3 to 6 months have been reported. In these cases, the authors have
concluded that a periodic schedule of short-tern temperature elevation In
the hot water nay be an effective control against legionellosis (USEPA,
1985; Muraca, 1986). Disinfection by this method also requires periodic
flushing of faucets and shower heads with hot water. Although heat
eradication is easily implemented and relatively inexpensive, a disadvan-
tage is the potential need for periodic disinfection. The potential for
scalding from the unusually hot water also exists (USEPA, 1985; Muraca, et
al. 1986}*
ChloHnation - Several studies have suggested that a free chlorine
residual of 4 mg/L will eradicate Lealone!la growth. There is, however,
a possibility for recontamination in areas of the system where the
chlorine residual drops below this level. A stringent monitoring program
is therefore required to ensure that the proper residual is maintained
throughout the system and under varying flow conditions. It may also be
necessary to apply a large initial chlorine dose to maintain the 4 mg/L
residual. This may cause problems of pipe corrosion and, depending on
water quality, high levels of trihalomethanes (THMs).
Ozone - Oione is the most powerful oxidant used in the potable
water Industry. One study indicated that an ozone dosage of 1 to 2 mg/L
was sufficient to provide a 5 log reduction of Legionella (Muraca, ft al.
1986). Ozone is generated by passing a high voltage current-of electrici-
ty through a stream of dry air or oxygen. The use of high voltage
electricity requires proper handling to avoid creating hazardous
conditions. The ozone is applied by bubbling the ozone containing gas
through the water in a chamber called a contactor.
B-3
-------
One of the disadvantages of this system Is Its complexity. It
requires a dry air or oxygen source, a generator, and a contactor sized to
provide 2 to 5 minutes of contact time and an ambient ozone monitor. All
materials In contact with the ozone must be constructed of special ozone
resistant materials to prevent leakage. Leak detection Is also required
because of the toxic nature of ozone and possible explosive conditions If
pure oxygen Is used for generation.
Another disadvantage of ozonation Is the rapid decooposltion of
ozone residuals. The half-life of ozone In drinking water 1s typically
around 10 minutes. ' This makes 1t difficult, 1f not Impossible, to
maintain a residual throughout the water system and may require the use of
a supplementary disinfectant such as chlorine or heat. For these reasons
1t 1s not thought that ozonation Is viable for Institutional applications.
ultraviolet Irradiation - Ultraviolet (UV) light, 1n the 254
nanometer wavelength range can be used as a disinfectant. UV systems
typically contain low-pressure mercury vapor lamps to maximize output In
the 254 nm range. Water entering the unit passes through a clear cylinder
while the lamp 1s on, exposing bacteria to the UV light. Because UV light
can not pass through ordinary window glass, special glass or quartz
sleeves are used to assure adequate exposure.
The Intensity of UV Irradiation 1s measured 1n microwatt-seconds per
square centimeter (uH-s/cm2). Several studies have shown a 90 percent
reduction of Legionella with a UV dosage of 1000 - 3000 uW-s/cm2, compared
to 2000 to 5000 uW-s/cm2 for L_£2ll, Salmonella and PsgudwwpaS (USEPA,
1985). In another study, a 5 log reduction of Legionella was achieved at
30,000 uW-s/cm2; and the reduction was more rapid than with both ozone and
chlorine disinfection (Muraca, et al. 1986).
The major advantage of UV disinfection Is that 1t does not require
the addition of chemicals. This eliminates the storage and feed problems
associated with the use of chlorine, chlorine dioxide and chloranlnes. In
addition, the only maintenance required 1s periodic cleaning of the quartz
sleeve and replacement of bulbs. UV monitors are available which measure
the light intensity reaching the water and provides a signal to the user
when maintenance 1s required. These monitors are strongly suggested for
any application of UV irradiation for disinfection. It should be noted,
6-4
-------
however, that these monitors Measure light Intensity which nay not be
directly related to disinfection efficiency. The UV lamps should
therefore not be operated past the Manufacturers use rating even with a
continuous UV monitor Installed.
Another disadvantage of UV disinfection, as with ozonation, 1s that
a residual is not provided. A supplementary disinfectant may therefore be
required to provide protection throughout the system. In addition,
turbidity may Interfere with UV disinfection by blocking the'passage of
light to the microorganisms.
B.3 OTHER CONTROL METHODS
In addition to chemical and heat disinfection, there are system
modifications which can be made to inhibit Legionella growth. Many
institutions have Urge hot water tanks heated by coils located midway In
the tank. This type of design may result 1n areas near the bottom of the
tank which are not hot enough to kill Legionella. Designing tanks for
more even distribution of heat may help limit bacterial colonization. In
addition, sediment build-up in the bottom of storage tanks provides a
surface for colonization. Periodic draining and cleaning may therefore
help control growth. Additionally, other studies have found that hot
water systems with stand-by hot water tanks used for meeting peak demands,
still tested positive for Legionella despite using elevated temperature
(55 C) and chlorlnation (2 ppm) (Fisher-Hoch, et al. 1984.) Stringent
procedures for the cleaning, disinfection and monitoring of these stagnant
tanks should be set up and followed on a regular basis.
In another study, It was reported that black rubber washers and
gaskets supported Legionella growth by providing habitats protected from
heat and chlorine. It was found, after replacement of the black rubber
washers with Proteus 80 compound washers, that it was not possible ta
detect Legionella from any of the fixtures (Colbourne, et al. 1984).
B-5
-------
B.4 CONCLUSIONS
Legionella bacteria have been Identified as the cause of the disease
leglonellosls, of which the most serious form Is Legionnaires Disease.
Although conventional water treatment practices are sufficient to provide
disinfection of Legionella, regrowth In buildings with large hot water,
heaters, and especially with recirculating hot water systems, Is a
significant problem. This problem Is of particular concern to-health care
Institutions, such as hospitals, where patients may be more susceptible to
the disease.
This guideline suggests a program of quarterly monitoring for
Legionella. If the monitoring program suggests a potential problem with
these organisms, a two stage disinfection program Is suggested consisting
of an Initial period of shock treatment followed by long term disinfec-
tion.
Four methods of disinfection for the control of Legionella were
presented In this appendix; heat, chlorlnatlon, ozonation, and ultraviolet
irradiation. All four of the methods have proven effective In killing
Legionella. Ultraviolet Irradiation and heat eradication are the
suggested methods of disinfection due, primarily, to advantages in
monitoring and maintenance. However, site specific factors may make
chlorlnatlon or ozonation more feasible for certain applications. In
addition, 1t 1s recommended that all outlets, fixtures and shower heads be
Inspected and all black rubber washers and gaskets replaced with materials
which do not support the growth of Legionella organisms.
One problem associated with the application of polnt-of-entry
treatment systems 1s the lack of an approved program for certifying,
performance claims. However, the National Sanitation Foundation (NSF),
Ann Arbor, MI an unofficial, non-profit organization, does have a testing
program to verify disinfection efficiencies and materials of construction.
Certification by the NSF, or other equivalent organizations, Is desirable
when selecting a treatment system.
B-6
-------
APPENDIX C
DETERMINATION OF DISINFECTANT
CONTACT TIME
-------
APPENDIX C
DETERMINATION OF DISINFECTANT CONTACT TIME
As Indicated In Section 3, for pipelines, ill fluid passing through
the pipe 1s assumed to have a detention tine equal to the theoretical or
mean residence t1ae at a particular flow rate. However, In nixing basins,
storage reservoirs, and other treatment plant process units, utilities
will be required to determine the contact tine for the calculation of CT
through tracer studies or other methods approved by the Primacy Agency.
For the purpose of determining compliance with the disinfection
requirements of the SWTR, the contact time of mixing basins and storage
reservoirs used 1n calculating CT should be the detention time at which 90
percent of the water passing through the unit 1s retained within the
basin. This detention time was designated as Tl# according to the
convention adopted by TMrumurthi (1969). A profile of the flow through
the basin over time can be generated by tracer studies. Information
provided by these studies Is used for estimating the detention time, Tl0,
for the purpose of calculating CT.
This appendix 1s divided Into two sections. The first section
presents a brief synopsis of tracer study methods, procedures, and data
evaluation. In addition, examples are presented for conducting hypo-
thetical tracer studies to determine the T|0 contact time In a clearwell.
The second section presents a method of determining Tl# from theoretical
detention times In systems where It is impractical to conduct tracer
studies.
Tracer Studies
c.i.i How conditions
Although detention time is proportional to flow, it is not generally
a linear function. Therefore, tracer studies are needed to establish
detention times for the range of flow rates experienced within each
disinfectant section.
As discussed In Section 3.2, a single flow rate may not characterize
the flow through the entire system. With a series of reservoirs,
C-l
-------
clear-wells, and storage tanks flow will vary between each portion of the
system.
In filter plants, the plant flow Is relatively uniform from the
Intake through the filters. An increase or reduction In the Intake
pumping capacity will Impart a proportional change 1n flow through each
process unit prior to and Including the filters. Therefore, at a constant
Intake pumping rate flow variations between disinfectant sections within
a treatment plant, excluding dearwells, are likely to be small, and the
the design capacity of the plant, or plant flow, can be considered the
nominal flow rate through each Individual process unit within the plant.
CTearwells may operate at a different flow rate than the rest of the
plant, depending on the pumping 'capacity.
Ideally, tracer tests should be performed for at least four flow
rates that span the entire range of flow for the section being tested.
The flow rates should be separated by approximately equal Intervals to
span the range of operation, with one near average flow, two greater than
average, and one less than average flow. The flows should also be
selected so that the highest test flow rate 1s at leaste 91 percent of the
highest flow rate expected to ever occur 1n that section. Four data
points will assure a good definition of the section's hydraulic profile.
The results of the tracer tests performed for different flow rates
should be used to generate plots of T10 vs. Q for each section In the
system. A smooth line 1s drawn through the points on each graph to create
a curve from which T10 may be read for the corresponding Q at peak hourly
flow conditions. This procedure 1s presented 1n Section C.1.8.
It may not be practical for all systems to conduct studies at four
flow rates. The number of tracer tests that are practical to conduct Is
dependent on site-specific restrictions and resources available to the.
system. Systems with limited resources can conduct a minimum of one
tracer test for each disinfectant section at a flow rate of not less than
91 percent of the highest flow rate experienced at that section. If only
one tracer test Is performed, the detention time determined by the test
may be used to provide a conservative estimate 1n CT calculations for that
section for all flow rates less than or equal to the tracer test flow
rate. Tl0 1s inversely proportional to flow rate, therefore, the T10 at a
C-2
-------
flow rate other than that which the tracer study was conducted (T,„) can
be determined by Multiplying the T„ from the tracer study (T,„) by the
ratio of the tracer study flow rate to the desired flow rate, I.e.,
Jios ¦ Tut X
-------
higher than the normal operating level, the resulting concentration
profile Mill predict an erroneously high detention time. Conversely,
extremely low water levels during testing may lead to an overly conserva-
tive detention tine. Therefore, when conducting a tracer study to
determine the detention time, a water level at or slightly below, but not
above, the normal minimum operating level 1s recommended.
For many plants, the water level 1n a clearwell or storage tank
varies between high and low levels 1n response to distribution system
demands. In such Instances, 1n order to obtain a conservative.estimate of
the contact time, the tracer study should be conducted during a period
when the tank level Is falling (flow out greater than flow 1n). This
procedure will provide a detention time for the contact basin which 1s
also valid when the water level Is rising (flow out less than flow 1n)
from a level which 1s at or above the level when the T10 was determined by
the tracer study. Whether the water level 1s constant or variable, the
tracer study for each section should be repeated for several different
flows, as described 1n the previous section.
For clearwells which are operated with extreme variations 1n water
level, maintaining a CT to comply with 1nact1vat1on requirements may be
Impractical. Under such operating conditions, a reliable detention time
1s not provided for disinfection. However, the system may Install a weir
to ensure a minimum water level and provide a reliable detention time.
Systems comprised of storage reservoirs that experience seasonal
variations 1n water levels may perform tracer studies during the various
seasonal conditions. For these systems, tracer tests should be conducted
at several flow rates and representative water levels that occur for each
seasonal condition. The results of these tests can be used to develop
hydraulic profiles of the reservoir for each water level. These profiles
can be plotted on the same axis of T10 vs. Q and may be used for calculat-
ing CT for different water levels and flow rates.
Detention time may also be Influenced by differences 1n water
temperature within the system. For plants with potential for thermal
stratification, additional tracer studies are suggested under the various
seasonal conditions which are likely to occur. The contact times
determined by the tracer studies under the various seasonal conditions
C-4
-------
should remain valid as long as no physical changes are lade to the Mixing
bas1n(s) or storage reservolr(s).
As. defined in Section 3.2.2, the portion of the system with a
Measurable contact tine between two points of disinfection or residual
¦onltorlng 1s referred to as a section. For systems which apply
dlslnfectant(s) at More than one point, or choose to profile the residual
froM one point of application, tracer studies should be conducted to
determine Tl# for each section containing process unit(s). The T,0 for a
section May or May not Include a length of pipe and is used along with the
residual disinfectant concentration prior to the next disinfectant appli-
cation or Monitoring point to determine the CTtlll for that section. The
inactivatlon ratio for the section 1s then determined. The total
inactivatlon and log 1nactivat1on achieved 1n the system can then be
determined by sunning the Inactivatlon ratios for all sections as
explained 1n Section 3.2.2.
For systems that have two or more units of identical size and
configuration, tracer studies only need to be conducted on one of the
units. The resulting graph of Tie vs. flow can be used to determine T,„
for all identical units.
Systems with more than one section in the treatment plant may
determine Tie for each section
by Individual tracer studies through each section, or
by one tracer study across the system
If possible, tracer studies should be conducted on each section to
determine the Ti0 for each section. In order to minimize the time needed
to conduct studies on each section, the tracer studies should be started
at the last section of the treatment train prior to the first customer and
completed with the first section of the system. Conducting the tracer
studies In this order will prevent the Interference of residual tracer
material with subsequent studies.
However, 1t May not always be practical for systems to conduct,
tracer studies for each section because of time and manpower constraints.
In these cases, one tracer study may be used to determine the Tjq values
C-5
-------
for til of the fictions it one flow rite. This procedure involves the
following steps:
1. Add tracer at the beginning of the furthest upstream disinfec-
tion section.
2. Measure the tracer concentration at the end of each disinfec-
tion section.
3. Determine the Tl0 to each monitoring point as outlined in the
data evaluation examples presented in Section C.1.7.
4. Subtract Tl# values of each of the upstream sections from the
overall T10 value to determine the T.„ of each downstream
section.
This approach is valid for a series of two or more consecutive
sections as long as all process units within the sections experience the
same flow condition. This approach 1s illustrated by Hudson (1975) in
which step-dose tracer tests were employed to evaluate the baffling
characteristics of flocculators and settling basins at six water treatment
plants. At one plant, tracer chemical was added to the rapid mix, which
represented the beginning of the furthest upstream disinfection section in
the system. Samples were collected from the flocculator and settling
basin outlets and analyzed to determine the residence-time characteristics
for each section. Tracer measurements at the flocculator outlet indicated
an approximate T„ of S minutes through the rapid mix, interbasln piping
and flocculator. Based on tracer concentration monitoring at the settling
basin outlet, an approximate Tl0 of 70 minutes was determined for the
combined sections, including the rapid mix, interbasln piping, floccu-
lator, and settling basin. The flocculator T,0 of 5 minutes was subtracted
from the combined sections' T,0 of 70 minutes, to determine the T,a for the
settling basin alone, 65 minutes.
This approach may also be applied in cases where disinfectant
application and/or residual monitoring 1s discontinued at any point
between two or more sections with known Tl0 values. These T,0 values may
be summed to obtain an equivalent Ti0 for the combined sections.
For ozone contactors, flocculators or any basin containing mixing,
tracer studies should be conducted for the range of mixing used in the
C-6
-------
process. In ozone contactors, air or oxygen should be added in lieu of
ozone to prevent degradation of the tracer. The flow rate of air or
oxygen used for the contactor should be applied during the study to
simulate actual operation. Tracer studies should then be conducted at
several air/oxygen to water ratios to provide data for the complete range
of ratios used at the plant. For flocculators, tracer studies should be
conducted for various nixing Intensities to provide data for the complete
range of operationss
c.i.3 Tracer Study Methods
This section discusses the two most common methods of tracer
addition employed In water treatment evaluations, the step-dose method and
the slug-dose method. Tracer study methods Involve the application of
chemical dosages to a system and tracking the resulting effluent
concentration as a function of time. The effluent concentration profile
1s evaluated to determine the detention time, Tl0.
While both tracer test methods can use the same tracer materials and
Involve measuring the concentration of tracer with time, each has distinct
advantages and disadvantages with respect to tracer addition procedures
and analysis of results.
The step-dose method entails Introduction of a tracer chemical at a
constant dosage until the concentration at the desired end point reaches
a steady-state level. Step-dose tracer studies are frequently employed In
drinking water applications for the following reasons:
the resulting normalized concentration vs. time profile Is
directly usea to determine, T10, the detention time required
for calculating CT
very often, the necessary feed equipment Is available to
provide a constant rate of application of the tracer chemical'
One other advantage of the step-dose method 1s that the data may be
verified by comparing the concentration versus elapsed time profile for
samples collected at the start of dosing with the profile obtained when
the tracer feed Is discontinued.
C-7
-------
Alternatively, with the slug-dose Mthod, a large instantaneous dose
of tracer is added to the Incoming water and samples are taken at the exit
of the unit over tine as the tracer passes through the unit. A disadvan-
tage of this technique 1s that very concentrated solutions are needed for
the dose in order to adequately define the concentration versus time
profile. Intensive mixing 1s therefore required to minimize potential
density-current effects and to obtain a uniform distribution of the
Instantaneous tracer dose across the basin. This is Inherently difficult
under water flow conditions often existing at inlets to basins. Other
disadvantages of using the slug-dose method include:
the concentration and volume of the instantaneous tracer dose
oust be carefully computed to provide an adequate tracer
profile at the effluent of the basin
the resulting concentration vs. time profile cannot be used to
directly determine Tl0 without further manipulation
a mass balance on the treatment section is required to
determine whether the tracer was completely recovered
One advantage of this method is that it may be applied where
chemical feed equipment is not available at the desired point of addition,
or where the equipment available does not have the capacity to provide the
necessary concentration of the chosen tracer chemical. Although, in
general, the step-dose procedure offers the greatest simplicity, both
methods are theoretically equivalent for determining T10. Either method
is acceptable for conducting drinking water tracer studies, and the choice
of the method may be determined by site-specific constraints or the
system's experience.
c.i.4 Tracer Selection
An Important step 1n any tracer study 1s the selection of a chemical
to be used as the tracer. Ideally, the selected tracer chemical should be
readily available, conservative (that is, not consumed or removed during
treatment), easily monitored, and acceptable for use in potable water sup-
plies. Historically, many chemicals have been used in tracer studies that
do not satisfy all of these criteria, including potassium permanganate,
alum, chlorine, and sodium carbonate. However, chloride and fluoride are
C-8
-------
the most common tracer chemicals employed in drinking water plants that
are nontoxic and approved for potable Mater use. Rhodamlne HT can be used
as a fluorescent tracer in water flow studies in accordance with the
following guidelines:
Raw water concentrations should be limited to a-ftaximum
concentration of 10 mg/L.
Drinking water concentations should not exceed 0.1 ug/L.
Studies which results in human exposure to the dye must be
brief and infrequent.
Concentrations as low as 2 ug/L can be used in tracer studies
because.of the low detection level in the range of 0.1 to 0.2
ug/L.
The use of Rhodamine B as a tracer in water flow studies is not recom-
mended by the EPA.
The choice of a tracer chemical can be made based, in part,.on the
selected dosing method and also on the availability of chemical feeding
equipment. For example, the high density of concentrated salt solutions
and their potential for inducing density currents, usually precludes
chloride and fluoride as the selected chemical for slug-dose tracer tests.
Fluoride can be a convenient tracer chemical for step-dose tracer
tests of clearwells because it is frequently applied for finished water
treatment. However, when fluoride is used in tracer tests on clarifiers,
allowances should be made for fluoride that is absorbed on floe and
settles out of water (Hudson, 1975). Additional considerations when using
fluoride in tracer studies include:
it is difficult to detect at low levels
many states impose a finished water limitation of 1 mg/L
the federal secondary and primary drinking water
standards (MCLs) for fluoride are 2 and 4 mg/L, respec-
tively
The use of fluoride is only recommended in cases where the feed equipment
is already 1n place for safety reasons.
C-9
-------
In Instances where only one of two or more parallel units Is tested,
flow fron the other units would dilute the tracer concentration prior to
leaving the plant and entering the distribution system. Therefore, the
Impact of drinking water standards on the use of fluoride and other tracer
chemicals can be alleviated In some cases.
c.i.5 Tracer Addition
The tracer chemical should be added at the same po1nt(s) 1n the
treatment train as the disinfectant to be used 1n the CT calculations.
c.1.5.1 step-dose Method
The duration of tracer addition 1s dependent on the volume of the
basin, and hence, Its theoretical detention time. In order to approach a
steady-state concentration 1n the water exiting the basin, tracer addition
and sampling should usually be continued for a period of two to three
times the theoretical detention time (Hudson, 1981). It Is not necessary
to reach a steady state concentration 1n the exiting water to determine
T10, however, It Is necessary to determine tracer recovery. It 1s
recommended that the tracer recovery be determined.to Identify hydraulic
characteristics or density problems. Generally, a 90 percent recovery is
considered to provide reliable results for the evaluation of Tlfl.
In all cases, the tracer chemical should be dosed In sufficient
concentration to easily monitor a residual at the basin outlet throughout
the test. The required tracer chemical concentration, is generally depen-
dent upon the nature of the chosen tracer chemical, Including Its
background concentration, and the mixing characteristics of the basin to
be tested. Recommended chloride doses on the order of 20 mg/L (Hudson,
1975) should be used for step-method tracer studies where the background
chloride level 1s less than 10 mg/L. Also, fluoride concentrations as low
as 1.0 to 1.5-mg/L are practical when the raw water fluoride level 1s not
significant (Hudson, 1975). However, tracer studies conducted on systems
suffering from serious shortcirculting of flow may require substantially
larger step-doses. This would be necessary to detect the tracer chemical
and to adequately define the effluent tracer concentration profile.
C-10
-------
C.1.5.2 Slug-dose Method
The duration of tracer Measurements using the slug-dose method 1s
also dependent on the volume of the basin, and hence, Its theoretical
detention time. In general, samples should be collected for at least
twice the basin's theoretical detention time, or until tracer concentra-
tions are detected near background levels. In order to get reliable
results for T10 values using the slug-dose method, 1t Is recommended that
the total mass of tracer recovered be approximately 90 percent of the mass
applied. This guideline presents the need to sample until the tracer
concentration recedes to the background level. The total mass recovered
during testing will not be known until completion of the testing and
analysis of the data collected. The sampling period needed Is very site
specific. Therefore, 1t may be helpful to conduct a first run tracer test
as a screen to Identify the appropriate sampling period for gathering data
to determine T10.
Tracer addition for slug-dose method tests should be Instantaneous
and provide uniformly mixed distribution of the chemical. Tracer addition
Is considered Instantaneous 1f the dosing time does not exceed 2 percent
of the basin's theoretical detention time (Harske and Boyle, 1973). One
recommended procedure for achieving Instantaneous tracer dosing 1s to
apply the chemical by gravity flow through a funnel and hose apparatus.
This method Is also beneficial because 1t provides a means of standardiza-
tion, which 1s necessary to obtain reproducible results.
The mass of tracer chemical to be added 1s determined by the desired
theoretical concentration arid basin size. The mass of tracer added in
slug-dose tracer tests should be the minimum mass needed to obtain
detectable residual measurements to generate a concentration profile. As
a guideline, the theoretical concentration for the slug-dose method should
be comparable to the constant dose applied 1n step-dose tracer tests,
I.e., 10 to 20 mg/L and 1 to 2 mg/L for chloride and fluoride, respective-
ly. The maximum mass of tracer chemical needed 1s calculated by
multiplying the theoretical concentration by the total basin volume. This
1s appropriate for systems with high dispersion and/or mixing. This
quantity 1s diluted as required to apply an Instantaneous dose, and
minimize density effects. It should be noted that the mass applied is not
C-ll
-------
likely to get completely nixed throughout the total volume of the basin.
Therefore, the detected concentration might exceed theoretical concentra-
tions based on the total volune of the basin. For these cases, the nass
of chenlcal to be added can be determined by multiplying the theoretical
concentration by only a portion of the basin volume. An example of this
Is shown In Section C.1.7.2 for a slug-dose tracer study. In cases where
the tracer concentration 1n the effluent must be maintained below a
specified level, It may be necessary to conduct a preliminary test run
with a minimum tracer dose to Identify the appropriate dose for determin-
ing T10 without exceeding this level.
C.1.6 Test Procedure
In preparation for beginning a tracer study, the raw water
background concentration of the chosen tracer chemical must be estab-
lished. The background concentration Is essential, not only for aiding 1n
the selection of the tracer dosage, but also to facilitate proper
evaluation of the data.
The background tracer concentration should be determined by .
monitoring for the tracer chemical prior to beginning the test. The
sampling polnt(s) for the pre-tracer study monitoring should be the same
as the points to be used for residual monitoring to determine CT values.
The monitoring procedure 1s outlined 1n the following steps:
If the tracer chemical Is normally added for treatment,
discontinue Its addition to the water In sufficient time to
permit the tracer concentration to recede to Its background
level before the test 1s begun.
Prior to the start of the test, regardless of whether the
chosen tracer material Is a treatment chemical, the tracer
concentration In the water Is monitored at the sampling point
where the disinfectant residual will be measured for CT
calculations.
If a background tracer concentration 1s detected, monitor It
until a constant concentration, at or below the raw water
background level 1s achieved. This measured concentration 1s
the baseline tracer concentration.
C-12
-------
Following the determination of the tracer dosage, feed and monitoring
point(s), and a baseline tracer concentration, tracer testing can begin.
Equal sampling intervals, as could be obtained from automatic
sampling, are not required for either tracer study method. However, using
equal sample Intervals for the slug-dose method can simplify the analysis
of the data. During testing, the time and tracer residual of each
measurement should also be recorded on a data sheet. In addition, the
Mater level, flow, and temperature should be recorded during the test.
c.1.6.1 step-dose Method
At time zero, the tracer chemical feed will be started and left at
a constant rate for the duration of the test. Over the course of the
test, the tracer residual should be monitored at the required sampling
point(s) at a frequency determined by the overall detention time and site
specific considerations. As a general guideline, sampling at intervals of
2 to 5 minutes should provide data for a well-defined plot of tracer
concentration vs. time. If on-site analysis 1s available, less frequent
residual monitoring may be possible until a change In residual concentra-
tion 1s first detected. As a guideline, in systems with a theoretical de-
tention time greater than 4 hours, sampling may be conducted every 10
minutes for the first 30 minutes, or until a tracer concentration above
the baseline level is first detected. In general, shorter sampling
intervals enable better characterization of concentration changes;
therefore, sampling should be conducted at 2 to 5-mlnute intervals from
the time that a concentration change is first observed until the residual
concentration reaches a steady-state value. A reasonable sampling
interval should be chosen based on the overall detention time of the unit
being tested.
If verification of the test is desired, the tracer feed should be.
discontinued, and the receding tracer concentration at the effluent should
be monitored at the same frequency until tracer concentrations correspond-
ing to the background level are detected. The time at which tracer feed
is stopped 1s time zero for the receding tracer test and must be noted.
The receding racer test will provide a replicate set of measurements which
can be compared with data derived from the rising tracer concentration
versus time curve. For systems which currently feed the tracer chemical,
C-13
-------
the receding curve nay be generated from the tine the feed is turned off
to determine the background concentration level.
C.l.6.2 Sluo-dose Method
At tine zero for the slug-dose nethod, a large instantaneous* dose of
tracer will be added to the Influent of the unit. The sane sampling
locations and frequencies described for step-dose nethod tests also apply
to slug-dose nethod tracer studies. One exception with this nethod 1s
that the tracer concentration profile will not equilibrate to a steady
state concentration. Because of this, the tracer should be nonltored
frequently enough to ensure acquisition of data needed to Identify the
peak tracer concentration.
Slug-dose nethod tests should be checked by perforating a material
balance to ensure that all of the tracer fed is recovered, or,, nass
applied equals nass discharged.
c.1.7 Pata Evaluation
Data fron tracer studies should be summarized 1n tables of time and
residual concentration. These data are then analyzed to determine the
detention time, Tl0, to be used In calculating CT. Tracer test data from
either the step or slug-dose nethod can be evaluated graphically,
numerically, or by a combination of these techniques.
C.1.7.1 Step-dose Method
The graphical nethod of evaluating step-dose test data Involves
plotting a graph of dlnenslonless concentration versus tine and reading
the value for Tl0 directly fron the graph at the appropriate dlnenslonless
concentration. Alternatively, the data fron step-dose tracer studies may
be evaluated numerically by developing a seni-logarithnlc plot of the
dlnenslonless data. The semi-logarithmic plot allows a straight line to
be drawn through the data. The resulting equation of the line 1s used to
calculate the T10 value, assuming that the correlation coefficient
indicates a good statistical fit (0.9 or above). Scattered data points
from step-dose tracer tests are discredited by drawing a smooth curve
through the data.
C-14
-------
An Illustration of the Tl0 determination will be presented 1n an
txaflfile of the data evaluation required for a clearwell tracer study.
C.l.7.2 S1u?-dose Method
Data from slug-dose tracer tests Is analyzed by converting It to the
mathematically equivalent "step-dose data and using techniques discussed In
Section C.1.7.1 to determine Tl0. A graph of dimensionless concentration
versus time should be drawn which represents the results of a slug-dose
tracer test. The key to converting between the data forms Is obtaining
the total area under the slug-dose data curve. This area is found by
graphically or numerically Integrating the curve. The conversion to
step-dose data Is then completed 1n several mathematical steps Involving
the total area.
A graphical technique for converting the slug-dose data Involves
physically measuring the area using a planlmeter. The planlmeter 1s an
Instrument used to measure the area of a plane closed curve by tracing Its
boundary. Calibration of this Instrument to the scale of the graph Is
required to obtain meaningful readings.
The rectangle rule Is a simple numerical Integration method which
approximates the total area under the curve as the sum of the areas of
Individual rectangles. These rectangles have heights and widths equal to
the residual concentration and sampling Interval (time) for each data
point on the curve, respectively. Once the data has been converted, T,0
may be determined 1n the same manner as data from step-dose tracer tests.
Slug-dose concentration profiles can have many shapes, depending on
the hydraulics of the basin. Therefore, slug-dose data points should not
be discredited by drawing a smooth curve through the data prior to Its
conversion to step-dose data. The steps and specific details involved
with evaluating data from both tracer study methods are Illustrated In the
following examples.
Example for Determining T,0 In a Clearwell
Two tracer studies employing the step-dose and slug-dose methods of
tracer addition were conducted for a clearwell with a theoretical
detention time, T, of 30 minutes at an average flow of 2.5 MGD. Because
fluoride Is added at the Inlet to the clearwell as a water treatment
chemical, necessary feed equipment was in place for dosing a constant
C-15
-------
concentration of fluoride throughout the step-dose tracer test. Based on
this convenience, fluoride Mas chosen as the tracer chealcal for the
step-dose aethod test. Fluoride was also selected as the tracer chenical
for the slug-dose aethod test. Prior to the start of testing, a fluoride
baseline concentration of 0.2 ng/L was established for the water exiting
the clearwell.
StBB-tfost Method Test
For the step-dose test a constant fluoride dosage of 2.0 ag/L was
added to the clearwell inlet. Fluoride levels In the clearwell effluent
were monitored and recorded every 3 alnutes. The raw tracer study data,
along with the results of further analyses are shown In Table C-l.
The steps 1n evaluating the raw data shown 1n the first column of
Table C-l are as follows. First, the baseline fluoride concentration,
0.2 ng/L, Is subtracted froa the measured concentration to give the
fluoride concentration resulting from the tracer study addition alone.
For example, at elapsed time - 39 minutes, the tracer fluoride concentra-
tion, C, 1s obtained as follows;
C * C»iiuri< " C|isiiln<
- 1.85 ag/L - 0.2 ng/L
- 1.6S mg/L
This calculation was repeated at each time Interval to obtain the data
shown In the third column of Table C-l. As Indicated, the fluoride
concentration rises froa 0 mg/L at t • 0 alnutes to the applied fluoride
dosage of 2 ag/L, at t ¦ 63 alnutes.
The next step Is to develop dlmenslonless concentrations by dividing
the tracer concentrations 1n the second column of Table C-l by the applied
fluoride dosage, Co - 2 mg/L. For time « 39 alnutes, C/Co Is calculated
as follows:
C/Co ¦ (1.65 ag/L)/(2.0 ag/L)
- 0.82
The resulting dimensionless data, presented In the fourth column of
Table C-l, Is the basis for completing the determination of T,0 by either
the graphical or numerical method.
C-16
-------
Uimilfii
TABLE C-l
CLEARWELL PATA.-STEP-OOSC TRACER
fluoride Concentration
Dlmensipnless. C/Ce
0
0.20
0
0
3
0.20
0
0
6
0.20
0
0
9
0.20
0
0
12
0.29
0.09
0.045
15
0.67
0.47
0.24
18
0.94
0.74
0.37
21
1.04
0.84
0.42
24
1.44
1.24
0.62
27
1.55
1.35
0.68
30
1.52
1.32
0.66
33
1.73
1.53
0.76
36
1.93
1.73
0.86
39
1.85
1.65
0.82
42
1.92
1.72
0.86
45
2.02
1.82
0.91
48
1.97
1.77
0.88
51
1.84
1.64
0.82
54
2.06
1.86
0.93
57
2.05
1.85
0.92
60
2.10
1.90
0.95
63
2.14
1.94
0.96
Notes:
"
1.
Baseline conc. - 0.2 mg/L, fluoride dose • 2.0 ng/L
2.
Measured conc. * Tracer conc.
<1- Baseline conc.
3.
Tracer conc. ¦ Measured conc.
- Baseline conc.
-------
In order to determine T|0 by the graphical method, a plot of C/Co vs.
tine should be generated using the data In Table C-l. A snooth curve
should be drawn through the data as shown on Figure C-l.
T10 Is read directly from the graph at a dimenslonless concentration
(C/Co) corresponding to the time for which 10 percent of the tracer has
passed at the effluent end of the contact basin (T|0). For step-dose
method tracer studies, this dimenslonless concentration Is C/Co ¦ 0.10
(Levensplel, 1972).
Tl# should be read directly froa Figure C-l at C/Co - 0.1 by first
drawing a horizontal line (C/Co ¦ 0.1) froa the Y-axis (t ¦ 0) to Its
Intersection with the smooth curve drawn through the data. At this point
of Intersection, the time read froa the X-ax1s 1s T10 and may be found by
extending a vertical line downward to the X-axis. These steps were
performed as Illustrated on Figure C-l, resulting In a value for T10 of
approximately 13 minutes.
For the numerical method of data analysis, several additional steps
are required to obtain T10 from the data 1n the fourth column of Table C-l.
The forms of data necessary for determining T10 through a numerical
solution are log10(l-C/Co) and t/T, the elapsed time divided by the
theoretical residence time. These are obtained by performing the required
mathematical operations on the data In the fourth column of Table C-l.
For example, recalling that the theoretical detention time, T, Is 30
minutes, the values for 1og10 (1-C/Co) and t/T are computed as follows for
the data at t - 39 minutes:
1og10(l-C/Co) - log10 (1-0.82)
- log,, (0.18)
- -0.757
t/T • 39 mln/30 mln -1.3
This calculation was repeated at each time Interval to obtain the
data shown In Table C-2. These data should be linearly regressed as
1°g10(l-C/C°) versus t/T to obtain the fitted straight-line parameters to
the following equation:
C-17
-------
FIGURE C-1
C/Co vs.. Time
Graphical Analysis for T10
0.9
o.a
0.7
0.6
0.6
o
0.4
0.3
0.2
40
20
TIME (MINUTES)
-------
TABLE C-2
DATA FOR NUMERICAL DETERMINATION OF T1Q
t/T lflflio ( 1~C/Cq)
0 0
0.1 0
0.2 0
0.3 0
0.4 .0.020 ~
0.5 -0.116
0.6 -0.201
0.7 -0.237
0.8 -0.420
0.9 -0.488
1.0 -0.468
1.1 -0.629
1.2 -0.870
1.3 -0.757
1.4 -0.854
1.5 -1.046
1.6 -0.939
1.7 -0.745
1.8 -1.155
1.9 -1.125
2.0 -1.301
2.1 -1.532
-------
1ogl#(l-C/Co) ¦ «(t/T) ~ b
(I)
In equation 1, a and b are the slope and intercept, respectively,
for a plot of log,0(l-C/Co) vs. t/T. This equation can be used to
calculate T10, assuring that the correlation coefficient for the fitted
data Indicates a good statistical fit (0.9 or above).
A linear regression analysis was performed on the data 1n Table C-2,
resulting 1n the following straight-line parameters:
slope • a ¦ -0.774
Intercept - b - 0.251
correlation coefficient ¦ 0.93
Although these numbers were obtained nuserlcally, a plot of
1og10(l-C/Co) versus t/T is shown for illustrative purposes on Figure C-2
for the data in Table C-2. In this analysis, data for tiae ¦ 0 through 9
minutes were excluded because fluoride concentrations above the baseline
level were not observed in the clearwell effluent until t - 12 minutes.
Equation 1 1s then rearranged in the following form to facilitate a
solution for Tl0:
T.0/T % (log,, (1 - 0.1) - b)/m (2)
In equation 2, as with graphical aethod, T1# is determined at the
tine for which C/Co ¦ 0.1. Therefore, in equation 2, C/Co has been
replaced by 0.1 and t (tine) by Tt0. To obtain a solution for Tl0, the
values of the slope, Intercept, and theoretical detention time are
substituted as follows:
TJO/30 Bin. % (logl5(l - 0.1) - 0.251)/(-0.774)
Tl0 - 12 minutes
In sumary both the graphical and numerical aethods of data
reduction resulted In comparable values for Tl0. With the numerical
aethod, Tl# was determined as the solution to an equation based on the
C-18
-------
straight-line parameters to a linear regression analysis of the tracer
study data Instead of an "eyeball* estimate from a data plot.
SlMfl-tiQH HtthQd TftSt
A slug-dose tracer'test was also performed on the clearwell at a
flow rate of 2.5 mgd. A theoretical clearwell fluoride concentration of
2.2 ng/L was selected. The fluoride dosing volume and concentration were
determined from the following considerations;
Paging Vol mm
The fluoride Injection apparatus consisted of a funnel and a
length of copper tubing. This apparatus provided a constant
volumetric feeding rate of 7.5 liters per minute (L/min) under
gravity flow conditions.
At a flow rate of 2.5 mgd, the clearwell has a theoretical
detention time of 30 minutes. Since the duration of tracer
Injection should be less than 2 percent of the clearwell1 s
theoretical detention time for an instantaneous dose, the
maximum duration of fluoride injection was:
Max. dosing time ¦ 30 minutes x .02 ¦ 0.6 minutes
- At a dosing rate of 7.5 L/m1nf the maximum fluoride dosing
volume 1s calculated to be:
Max. dosing volume - 7.5 L/m1n. x 0.6 minutes - 4.5 L
• For this tracer test, a dosing volume of 4 liters was select-
ed, providing an instantaneous fluoride dose in 1.8 percent of
the theoretical detention time.
fluoride Concentration
The theoretical detention time of the clearwell, 30 minutes,
was calculated by dividing the clearwell volume, 52,100
gallons or 197,200 liters, by the average flow rate through
the clearwell, 2.5 mgd.
- Assuming the tracer is completely dispersed throughout the
total volume of the clearwell, the mss of fluoride required
to achieve a theoretical concentration of 2.2 mg/L is calcu-
lated as follows:
Fluoride mass (initial) - 2.2 mg/L x 197,200 L x _Lfl ¦ <34g
1000 mg
C-19
-------
FIGURE C-2
1-C/CO VS. t/T
Numerical Analysis for T10
o
o
0.01
2
2.5
0
0.5
1.5
1
t/T
Slop*, m--0.774
lnt»rc#pt. b¦ 0.251 Correlation Cotfflcfcnt • 0
-------
Tht concentration of the instantaneous fluoride dose is
determined by dividing this mss by the dosing volune, 4
liters;
Fluoride concentration ¦ <34 a ¦ 109 g/L
4 L
Fluoride levels in the exit to the clearwell were aonitored and
recorded every 3 minutes. The raw slug-dose tracer test data are shown in
Table C-3.
The first step in evaluating the data for different tines is. to
subtract the baseline fluoride concentration, 0,2 ag/L, froa the measured
concentration at each sampling interval (Table C-3). This 1s the sane as
the first step used to evaluate step-dose aethod data and gives the
fluoride concentrations resulting froa the tracer addition alone, shown in
the third column of Table C-3. As indicated, the fluoride concentration
rises froa 0 og/L at t - 0 nlnutes to the peak concentration of 3.6 ag/L
at t - 18 nlnutes. The exiting fluoride concentration gradually recedes
to near zero at t - 63 ainutes. It should be noted that a aaxinum
fluoride concentration of 2.2 ag/L Is based on assuaing complete nixing of
the tracer added throughout the total clearwell volune. However, as shown
in Table C-3, the fluoride concentrations In the clearwell effluent
exceeded 2.2 ag/L for about 6 ainutes between 14 and 20 ainutes. These
higher peak concentrations are caused by the dispersion of tracer
throughout only a portion of the total clearwell volume. If a lower
tracer concentration 1s needed in the effluent because of local or federal
regulations, the aass to be added should be decreased accordingly.
The diaenslonless concentrations In the fourth column of Table C-3
were obtained by dividing the tracer concentrations in the third column by
the clearwell's theoretical concentration, Co ¦ 2.2 ag/L. These
dimensionTess concentrations were then plotted as a function of time, as
is shown by. the slug-dose data on Figure C-3. These data points were
connected by straight lines, resulting in a somewhat jagged curve.
The next step 1n evaluating slug-dose data is to deteraine the total
area under the slug-dose data curve on Figure C-3. Two aethods exist for
finding this area — graphical and numerical. The graphical method is
C-2Q
-------
TABLE C-3
ciearweu pata - slug-pose mm test(I,2,3)
Fluoride Concentration
t. winutes HCMItreti. PW/1 Tracer. »g/L Piwensionless. C/Co
0
0.2
0
0
3
0.2
0
0
6
0.2
0
0
9
0.2
0
0
12
1.2
1
0.45
15
3.6
3.4
1.55
18
3.8
3.6
1.64
21
2.0
1.8
0.82
24
2.1
1.9
0.86
27
1.4
1.2
0.55
30
1.3
1.1
0.50
33
1.5
1.3
0.59
36
1.0
0.8
0.36
39
0.6
0.4
0.18
42
1.0
0.8
0.36
45
0.6
0.4
0.18
48
0.8
0.6
0.27
51
0.6
0.4
0.18
54
0.4
0.2
0.09
57
0.5
0.3
0.14
60
0.6
0.4
0.18
63
0.4
0.2
0.09
Notes;
1. Measured conc. = Tracer conc. ~ Baseline conc.
?. Baseline conc. = 0.2 ag/L, fluoride dose = 109 g/L, theoretical conc. « 2.2 »g/L
J. Iracer conc. = Measured conc. - Baseline conc.
-------
FIGURE C-3
C/Co vs. Time
Conversion of Slug-to Step-Dose Data
Slug-dose dot'
Step-dose dc*.
O
CJ
CJ
0.8
0.6
0.4
0.2
20 30 40 50
60
70
0
10
TIME (MINUTES)
-------
based on a physical measurement of the area using a planlmeter. This
Involves calibration of the Instrument to define the units conversion and
tracing the outline of the curve to determine the area. The results of
performing this procedure nay vary depending on Instrunent accuracy and
measurement technique. Therefore, only an Illustration of the numerical
technique for finding the area under the slug-dose curve will be presented
for this example.
The area obtained by either the graphical or numerical method would
be similar. Furthermore, once the area Is found, the remaining steps
Involved with converting the data to the step-dose response are the same.
Table C-4 summarizes the results of determining the total area using
a numerical Integration technique called the rectangle rule. The first
and second columns 1n Table C-4 are the sampling time and fluoride
concentration resulting from tracer addition alone, respectively. The
steps In applying these data are as follows. First, the sampling time
Interval, 3 minutes, Is multiplied by the fluoride concentration at the
end of the 3-mlnute Interval to give the Incremental area, 1n units of
milligram minutes per liter. For example, at elapsed time, t • 39
minutes, the Incremental area Is obtained as follows:
Incremental area ¦ sampling time Interval x fluoride conc.
¦ (39-36) minutes x 0.4 mg/L
> 1.2 mg-m1n/L
This calculation was repeated at each time Interval to obtain the data
shown In the third column of Table C-4.
If the data had been obtained at unequal sampling Intervals, then
the incremental area for each Interval would be obtained by multiplying
the fluoride concentration at the sod of each Interval by the. time
duration of the Interval. This convention also requires that the
Incremental area be zero at the first sampling point, regardless of the
fluoride concentration at that time.
As Is shown In Table C-4, all Incremental areas were summed to
obtain 59.4 mg-min/L, the total area under the slug-dose tracer test
curve. This number represents the total mass of fluoride that was
C-21
-------
detected during the course of the tracer test divided by the average flow
rate through the clearwell.
To . complete the conversion of slug-dose data to Its equivalent
step-dose response requires two additional steps. The first Involves
sunning, consecutively, the Incremental areas In the third column/of Table
C-4 to obtain the cumulative area at the tnd of each sampling Interval.
For example, the cumulative area at time, t - 27 minutes 1s found as
follOWSI
Cumulative area -0+0+0+0+3* 10.2 ~ 10.8 + §.4 + 5.7 ~ 3.6
¦ 38.7 mg-min/L
The cumulative areas for each interval are recorded in the fourth column
of Table C-4.
The final step in converting slug-dose data involves dividing the
cumulative area at each Interval by the total mass applied. Total area
based on applied mass Is calculated as follows:
Total area mass applied/average flow ¦ 434 g x 1000 ag/6,570 _L_
g nin
¦ 66.1 mo-win
L
For time ¦ 39 minutes, the resulting step-dose data point 1s calculated as
follows:
C/Co ¦ 49.5 mg-min/L / 59.4 mg-mln/L
-0.83
The result of performing this operation at each sampling Interval is the
equivalent step-dose data. These data points are shown in the fifth
column of Table C-4 and are also plotted on Figure C-3 to facilitate a
graphical determination of T10. A smooth curve was fitted to the step-dose
data as shown on the figure.
Tt# can be determined by the methods Illustrated previously in this
example for evaluating step-dose tracer test data. The graphical method
illustrated on Figure C-3 results in a reading of Tl0 ¦ 15 minutes.
C-22
-------
0
3
6
9
12
15
18
21
24
27
30
33
36
39
42
45
48
51
54
57
60
63
TABLE C-4
EVALUATION OF SLUG-DOSE DATA
Equivalent
Incremental Cumulative Step-Dose
Fluoride. PW/L Area. wp-win/L Area. wa-min/L Data
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
3
3
0.05
3.4
10.2
13.2
0.22
3.6
10.8
24.0
0.40
1.8
5.4
29.4
0.49
1.9
5.7
35.1
0.59
1.2
3.6
38.7
0.65
1.1
3.3
42.0
0.71
1.3
3.9
45.9
0.77
0.8
2.4
48.3
0.81
0.4
1.2
49.5
0.83
0.8
2.4
51.9
0.87
0.4
1.2
53.1
0.89
0.6
1.8
54.9
0.92
0.4
1.2
56.1
0.94
0.2
0.6
56.7
0.95
0.3
0.9
57.6
0.97
0.4
1.2
58.8
0.99
0.2
~ 0.6
59.4
1.00
Total Area « 59.4
-------
C.1.7.3 Additional Considerations
In addition to determining Tl0 for use In CT calculations, slug-dose
tracer tests provide a sore general Measure of the basin's hydraulics In
terms of the fraction of tracer recovery. This number 1s representative
of short-circuiting and-' dead space In the unit resulting froo poor
baffling conditions and density currents Induced by the tracer chemical.
A low tracer recovery 1s generally Indicative of Inadequate hydraulics.
However, Inadequate sampling 1n which peaks 1n tracer passage are not
measured will result 1n an under estimate of tracer recovery. The tracer
recovery Is calculated by dividing the mass of fluoride detected by the
mass of fluoride dosed.
The dosed fluoride mass was calculated previously and was 434 grams.
The mass of detected fluoride can be calculated by aultlplylng the total
area under the slug-dose curve by the average flow, In appropriate units,
at the time of the test. The average flow In the clearwell during the
test was 2.5 mgd or 6,570 L/m1n. Therefore, the mass of fluoride tracer
that was detected Is calculated as follows:
Detected fluoride mass ¦ total area x average flow
¦ 59.4 mo-min x 1 g x 6,570 JL_
L 1000 mg m1n
• 390 g
Tracer recovery Is than calculated as follows:
Fluoride recovery • detected mass/dosed mass x 100
• 390 g / 434 g x 100
- 90 %
This 1s a typical tracer recovery percentage for a slug-dose test, based
on the experiences of Hudson (1975) and Thlrumurthl (1969).
C.I.* -flow Dependency of Tt„
For systems conducting tracer studies at four or more flows, the T10
detention time should be determined by the above procedures for each of
the desired flows. The detention times should then be plotted versus
flow. For the example presented In the previous section, tracer studies
C-23
-------
were conducted at additional flows of
values at the various flows were:
Flow
' 1.1
2.5
4.2
5.6
T10 data for these tracer studies were plotted as a function of the flow,
Q, as shown on Figure C-4.
If only one tracer test 1s performed, the flow rate for the tracer
study should be not less than 91 percent of the highest flow rate
experienced for the section. The hydraulic profile to be used for
calculating CT would then be generated by drawing a line through points
obtained by nultlplylng the Tl0 at the tested flow rate by the ratio of the
tracer study flow rate to each of several different flows 1n the desired
flow range.
For the example presented 1n the previous section, the clearwell
experiences a maximum flow at peak hourly conditions of 6.0 mgd. The
highest tested flow rate was 5.6 ngd, or 93 percent of the maximum flow.
Therefore, the detention time, T]0 ¦ 4 minutes, determined by the tracer
test at a flow rate of 5.6 ngd may be used to provide a conservative
estimate of Tl0 for all flow rates less than or equal to the maximum flow
rate, 6.0 mgd. The line drawn through points found by multiplying T10 -
4 minutes by the ratio of 5.6 mgd to each of several flows less than 5.6
mgd 1s also shown on Figure C-4 for comparative purposes with the
hydraulic profile obtained from performing four tracer studies at
different flow rates.
C.2 Determination of Tl0 Without Conducting a Tracer Study
In some situations, conducting tracer studies for determining the
disinfectant contact time, T10, may be Impractical or prohibitively
expensive. The limitations may Include a lack of funds, manpower or
equipment necessary to conduct the study. For these cases, the Primacy
Agency may allow the use of "rule of thumb" fractions representing the
1.1, 4.2. and 5.6 MGD. The Tl0
Iio
25
13
7
4
C-24
-------
FIGURE C-4
Detention Time vs. Flow
4-Flow profile
1 -Flow profile
Average ¦
5
7
8
0
2
3
6
A
1
FLOW (MGD)
-------
ratio of T)0 to T, and the theoretical detention tlae, to determine the
detention tlM, T10, to be used for calculating CT values. This aethod for
finding Tja Involves Multiplying the theoretical detention tlae by the rule
of thuab fraction, T10/T, that Is representative of the particular basin
configuration for which T10 Is desired. These fractions provide rough
estimates of the actual T,g and are recommended to be used only on a
Halted basis.
Tracer studies conducted by Marske and Boyle (1973) and Hudson
(1975) on chlorine contact chaabers and flocculators/settllng basins,
respectively, were used as a basis In deteralnlng representative Tl0/T
values for various basin configurations. Marske and Boyle (1973) performed
tracer studies on 15 distinctly different types of full-scale chlorine
contact chambers to evaluate design characteristics that affect the actual
detention tine. Hudson (1975) conducted 16 tracer tests on several
flocculatlon and settling basins at six water treatment plants to Identify
the effect of flocculator baffling and settling basin Inlet and outlet
design characteristics on the actual detention t1ae.
c.2.i Impact of Design Characteristics
The significant design characteristics Include: length-to-width
ratio, the degree of baffling within the basins, and the effect of Inlet
baffling and outlet weir configuration. These physical characteristics of
the contact basins affect their hydraulic efficiencies In terns of dead
space, plug flow, and alxed flow proportions. The dead space zone of a
basin Is basin voluae through which no flow occurs. The reaalnlng volume
where flow occurs 1s comprised of plug flow and alxed flow zones. The
plug flow zone Is the portion of the remaining voluae In which no alxlng
occurs In the direction of flow. The alxed flow zone Is characterized by
coaplete alxlng In the flow direction and Is the coapleaent to the plug
flow zone. All of these zones were Identified In the studies for each
contact basin. Comparisons were then aade between the basin configura-
tions and the observed flow conditions and design characteristics.
The ratio T)0/T was calculated from the data presented 1n the studies
and compared to Its associated hydraulic flow characteristics. Both
studies resulted 1n Tig/T values which ranged from 0.3 to 0.7. The results
C-25
-------
of th« studies Indicate how basin baffling conditions can Influence the
T,0/T ratio, particularly baffling at the inlet and outlet to the basin.
As the basin baffling conditions laproved, higher Tl0/T values Mere
observed, with the outlet conditions generally having a greater i^act
than the inlet conditions.
As discovered from the results of the tracer studies performed by
Harske and Boyle (1973} and Hudson (197S). the effectiveness of baffling
in achieving a high Tl0/T fraction Is sore related to the geometry and
baffling of the basin than the function of the basin. For tills reason,
Tl0/T values may be defined for three levels of baffling conditions rather
than for particular types of contact basins. 6enera> guidelines were
developed relating the T„/T values froa these studies to the respective
baffling characteristics. These guidelines can be used to determine the
T1# values for specific basins.
C.2.2 Baffling Classifications
The purpose of baffling is to aaxialze utilization of basin volume,
Increase the plug flow zone In the basin, and ainiaize short circuiting.
Some fora of baffling at the inlet and outlet of the basins is used to
evenly distribute flow across the basin. Additional baffling may be
provided within the interior of the basin (Intra-basln) in circumstances
requiring a greater degree of flow distribution. Ideal baffling design
reduces the inlet and outlet flow velocities, distributes the water as
uniformly as practical over the cross section of the basin, minimizes
mixing with the water already in the basin, and prevents entering water
from short circuiting to the basin outlet as the result of wind or density
current effects. Three general classifications of baffling conditions —
poor, average, and superior — were developed to categorize the results of
the tracer ftudies for use in determining Tl# from the theoretical
detention time of a specific basin. The Tu/T fractions associated with
each degree of baffling are summarized In Table C-5. Factors representing
the ratio between T}0 and the theoretical detention time for plug flow in
pipelines and flow In a completely mixed chamber have beerv included in
Table C-5 for comparative purposes. However, in practice the theoretical
TI0/T values of 1.0 for plug flow and 0.1 for alxed flow are seldom
C-26
-------
TABLE C-5
tuffHnn Condition
Unbaffled (mixed flow)
Poor
Average
Superior
Perfect (plug flow)
IioZI
0.1
0.3
0.5
0.7
1.0
Baffling D«crfpt1qn
None, agitated basin, very low length
width ratio, high inlet and outlet
velocities
Single or multiple unbaffled inlets an<
outlets, no Intra-basin baffles
•
Baffled inlet fit outlet with some in
basin baffles
Perforated inlet baffle, serpentine
perforated intra-basin baffles, outlet i
or perforated launders
Very high length to width ratio (pipe
flow), perforated inlet, outlet, and in
basin baffles
-------
achieved because of the effect of dead space. Conversely, the Tl8/T values
shown for the Intermediate baffling conditions already Incorporate the
effect of the dead space zone, as well as the plug flow zone, because they
were derived enplrlcally rather than from theory.
As Indicated 1n Table C-5, baffling conditions consist of an
unbaffled Inlet and outlet with no 1ntra-bas1n baffling. Average baffling
conditions consist of 1ntra-basin baffling and either a baffled Inlet or
outlet. Superior baffling conditions consist of at least a baffled Inlet
and outlet, and possibly some 1ntra-bas1n baffling to redistribute the
flow throughout the basin's cross-section.
The three basic types of basin Inlet baffling configurations are:
a target-baffled pipe Inlet, an overflow weir entrance, and a baffled
submerged orifice or port Inlet. Typical 1ntra-bas1n baffling structures
Include: dlffuser (perforated) walls; launders; cross, longitudinal, or
naze baffling to cause horizontal or vertical serpentine flow; and
longitudinal divider walls, which prevent mixing by increasing the
length-to-width ratio of the basln(s). Commonly used baffled outlet
structures Include free-discharging weirs, such as sharpcrested and
V-notch, and submerged ports or weirs. Weirs that do not span the width
of the contact basin, such as Clpolletl weirs, should not be considered
baffling as their use may substantially Increase weir overflow rates and
the dead space zone of the basin.
C.2.3 Examples of Baffling
Examples of these levels of baffling conditions for rectangular and
circular basins are explained and Illustrated 1n the following section.
Typical uses of various forms of baffled and unbaffled Inlet and outlet
structures are also Illustrated.
The-plan and section of a rectangular basin with poor baffling con-
ditions, which can be attributed to the unbaffled Inlet and outlet pipes,
1s Illustrated on Figure C-5. The flow pattern shown In the plan view
Indicates straight-through flow with dead space occurring In the regions
between the Individual pipe Inlets and outlets. The section view reveals
additional dead space from a vertical perspective In the upper Inlet and
lower outlet comers of the contact basin. Vertical mixing also occurs as
C-27
-------
bottom density currents induct a counter-clockwise flow In the upper water
layers.
The Inlet flow distribution is Markedly improved by the addition of
an inlet diffuser wall and 1ntra-basin baffling as shown on Figure C-6.
However, only average baffling conditions are achieved for the basin as a
whole because of the inadequate outlet structure — a Cipolleti weir. The
width of the weir is short In comparison with the width of the basin.
Consequently, dead space exists in the corners of the basin, as shown by
the plan view. In addition, the snail weir width causes a-high weir
overflow rate, which results in short circuiting in the center of the
basin.
Superior baffling conditions are exemplified by the flow pattern and
physical characteristics of the basin shown on Figure C-7. The inlet to
the basin consists of submerged, target-baffled ports. This Inlet design
serves to reduce the velocity of the incoming water and distribute it
uniformly throughout the basin's cross-section. The outlet structure is
a sharpcrested weir which extends for the entire width of the contact
basin. This type of outlet structure will reduce short circuiting and
decrease the dead space fraction of the basin, although the overflow weir
does create some dead space at the lower comers of the effluent end.
These inlet and outlet structures are by themselves sufficient to attain
superior baffling conditions; however, maze-type intra-basin baffling was
Included as an example of how this type of baffling aids In flow
redistribution within a contact basin.
The plan and section of a circular basin with poor baffling
conditions, which can be attributed to flow short circuiting from the
center feed well directly to the effluent trough 1s shown on Figure C-8.
Short circuiting occurs In spite of the outlet weir configuration because
the center feed inlet is not baffled. The inlet flow distribution is
Improved somewhat on Figure C-9 by the addition of an annular ring baffle
at the Inlet which causes the inlet flow to be distributed throughout a
greater portion of the basin's available volume. However, the baffling
conditions In this contact basin are only averaoe because the inlet center
feed arrangement does not entirely prevent short circuiting through the
upper levels of the basin.
C-28
-------
SECTION
FIGURE C-5 POOR BAFFLING CONDITIONS --
RECTANGULAR CONTACT BASIN
-------
y.
/ ' ?
'/ I ^
71
it:
V
N
•/ '
i m
: ' ~ , :
, j V* . ! j
i'V/ ^ - xz-
zz
I !
PLAN
r
A
/
ZZX
A
A
r,\
¦s
SECTION
FIGURE C-S AVERAGE BAFFLING CONDITIONS --
RECTANGULAR CONTACT BASIN
-------
SECTION
FIGURE C-7 SUPERIOR BAFFLING CONDITIONS --
RECTANGULAR CONTACT BASIN
-------
PLAN
//////////
SECTION
FIGURE C-i POOR BAFFLING CONDITIONS
CIRCULAR CONTACT BASIN
-------
PLAN
7
SECTION
FIGURE C-9 AVERAGE BAFFLING CONDITIONS -
CIRCULAR CONTACT BASIN
-------
Superior baffling conditions are attained In the basin configuration
shown on Figure C-10 through the addition of a perforated Inlet baffle and
submerged orifice outlet ports. As Indicated by the flow pattern, more of
the basin's volume Is utilized due to uniform flow distribution created by
the perforated baffle. Short circuiting Is also minimized because only a
small portion of flow passes directly through the perforated baffle wall
from the inlet to the outlet ports.
C.2.4 Additional Cons1derations
Flocculation basins and ozone contactors represent water treatment
processes with slightly different characteristics from those presented in
Figures C-5 through C-10 because of the additional effects of mechanical
agitation and mixing from ozone addition, respectively. Studies by Hudson
(1975) indicated that a single-compartment flocculator had a Tl6/T value
less than 0.3, corresponding to a dead space zone of about 20 percent and
a very high mixed flow zone of greater than 90 percent. In this study,
two four-compartment flocculators, one with and the other without
mechanical agitation, exhibited T,0/T values in the range of 0.S to 0.7.
This observation Indicates that not only will compartmentation result in
higher Tie/T values through better flow distribution, but also that the
effects of agitation intensity on T,0/T are reduced where sufficient
baffling exists. Therefore, regardless of the extent of agitation,
baffled flocculation basins with two or more compartments should be
considered to possess average baffling conditions (Tl0/T ¦ 0.5), whereas
unbaffled, single-compartment flocculation basins are characteristic of
poor baffling conditions (Tl0/T ¦ 0.3).
Similarly, multiple stage ozone contactors are baffled contact
basins which-show characteristics of average baffling conditions. Single
stage ozone contactors should be considered as being poorly baffled.
However, circular, turbine ozone contactors may exhibit flow distribution
characteristics which approach those of completely mixed basins, with a
Tl0/T of 0.1, as a result of the intense mixing.
In many cases, settling basins are directly connected to the
flocculators. Data from Hudson (1975) Indicates that poor baffling
conditions at the flocculator/settling basin interface can result in
C-29
-------
backmixing from the settling basin to the flocculator. Therefore,
settling basins that have Integrated flocculators without effective Inlet
baffling should be considered as poorly baffled, with a T10/T of 0.3,
regardless of the outlet conditions, unless 1ntra-bas1n baffling 1s
employed to redistribute flow. If 1ntra-bas1n and outlet baffling Is
, utilized, then the baffling conditions should be considered average with
a T,0/T of 0.5.
Filters are special treatment units because their design and
function 1s dependent on flow distribution that 1s completely uniform.
Except for a small portion of flow which shortclrcults the filter media by
channeling along the walls of the filter, filter media baffling provides
a high percentage of flow uniformity and can be considered superior
baffling conditions for the purpose of determining T10. As such, the T0
value can be obtained by subtracting the volume of the filter media,
support gravel, and underdralns from the total volume and calculating the
theoretical detention time by dividing this volume by the flow through the
filter. The theoretical detention time Is then multiplied by a factor of
0.7, corresponding to superior baffling conditions, to determine the T10
value.
c.2.5 conclusion?
The recommended T,0/T values and examples are presented as a
guideline for use by the Primacy Agency In determining T10 values In site
specific conditions and when tracer studies cannot be performed because of
practical considerations. Selection of T10/T values In the absence of
tracer studies was restricted to a qualitative assessment based on
currently available data for the relationship between basin baffling
conditions and their associated T10/T values. Conditions which, are
combinations or variations of the above examples may exist and warrant the
use of Intermediate T10/T values such as 0.4 or 0.6. As more data on
tracer studies become available, specifically correlations between other
physical characteristics of basins and the flow distribution efficiency
parameters, further refinements to the Tl0/T fractions and definitions of
baffling conditions may be appropriate.
C-30
-------
PLAN
Q
iy
TzzzzzzzhzA)
I
iip
v
/
iZL
rrrx
3"
SECTION
FIGURE C-10 SUPERIOR BAFFLING CONDITIONS
CIRCULAR CONTACT BASIN
-------
Buferences
Hudson, H. E., Jr., "Residence T1aes In Pretreataent", J. AWWA, pp. 45-52,
January, 1975.
Hudson, H. E., Jr.. Water Clarification Processes; Practical Design and
Evaluation. Van Nostrand Relnhold Conpany, New York, 1981.
Ltvtnspltl, 0.. Chemical Reaction Enolneerlno. John HI ley ft Sons, New
York, 1972.
Marske, D. M. and Boyle, J. 0.. "Chlorine Contact Chamber Oeslgn - A Field
Evaluation", Hater and Sewage Works, pp. 70-77, January, 1973
Th1runurth1, 0.. "A Break-through In the Tracer Studies of Sed1Dentation
Tanks", J. WPCF, pp.. R405-R418, November, 1969.
C-31
-------
APPENDIX C:
CONCENTRATING, PROCESSING, DETECTING AND IDENTIFYING
GIARDTA CYSTS IN WATER
Thi
-------
APPENDIXC: CONCENTRATING, PROCESSING
DETECTING AMD IDENTIFYING GIARHIA CYSTS TN WATER
"ETHOD
INVESTIGATOR (S)
RESULTS
1. Menhrane nitration
Cellu^slc
(*7rnm-0.45um)
DolvcarMnate
(29>m-Sum)
2. Particulate Filtration
Mlatoroaceous earth, sanH,
etc.)
3. Alga* (Foerst) Centrifuge
4. Anionic and Cat1on1c
Exchange Resins
5. Epoxy-Flherglass Ralston
Tube FiIters
(10"-Hum)
e>. M1crooorous Yarnwnven Depth.
F'1ters
(/ *n
-------
APPENDIX C: CONCENTRATING, PROCESSING,
DETECTING AND IDENTIFYING GIAflDIA CYSTS IN WATER
PRIMARY CONCENTRATION AND PROCESSING METHODS
1. MEM3RANE FILTER (MF) METHODS
a. f>lu1os1c (m
-------
APPENDIX :: CONCENTRATING, PROCESSING,
DETECTING AND IDENTIFYING GIARDIA CYSTS IN WATER
'• A'.GAE CENTRIFUGE
a. Was foun<< to recover more cyst.s (10X) than a series of MF-fllters and
nylon screens: 5 vs. 1 day by MF.
b. May be Impractical 1n field because of power requirement.
c. If used 1n lab, 1 large single sample collected 1n the field could miss
cyst.
d. May find application for concentration cysts from orlon f1lt.gr washings.
*• ANIONIC AND CATIOUTC EXCHANGE RESINS (Brewer - unpublished
a. Based on hypothesis that cysts could be attracted to charged surfaces,
cysts have a charg® of approximately 25mV at pH 5.5 which Increases In
electro-negativity as the pH rises to B.O.
b. Charge attraction techniques have been used for concentration of both
bacteria and viruses 1n water.
c. Five exchange resins were tested:
(1. 4° percent recovery from anionic Dowex 1-XY columns
(2. 38 percent recovery from cationic Dowex 50W-X8 columns
d. Compared' to parallel tests w/dlatomaceous earth, exchange resins less
efficient in retention.
5. BALSTOH EPQW-FIBERGLASS TUBE FILTERS
a. Riggs of CSHD, Viral and Rick. Lab., can filter 500 gallons drinking
water thru 10" - 8 um Balston tube filter,
b. Baclcflushes w/1 L 3 percent beef extract or solution of 0.5 percent
potasslun citrate.
c. Concentration Is centrlfuged w/40 percent potassium citrate and middle
layer filtered thru 5 u polycarbonate filters.
d. Uses direct Immunofluorescence antibody technique for detection and
identification.
%
e. Claims ?0-S0 percent efficiency In collection, preprocessing and ID.
6. MICROPOP/IUS YARMWOVEN DEPTH FILTERS
a. In 1975 EPA develooed a concentration-extraction method Involving rge
volumes of water thru mlcroporous yarnwoven orlon-flber fiUe-s.
b. This method has been tenatlvely adopted as the "method of cho1::*
concentnti ng cysts from water supplies.
-------
APfWIX C: CONCENTRATING. PROCESSING,
DEt£CTIM20 percent.
1. Gone from 7 to 1 um porsity filter
2. United the ra"e of flow to 1/2 gallon/mln
3. Limited the pressure head to 10 PSI
4. Have gone to po'yproylsne filters 1n lieu of orlon
d. It was the first methol successfully used to detect cysts 1n the
distribution systen of a coraiunlty water supply.
e. Is the recommended filter to be used by the EPA consensus method.
7. PC.IUCAH CASSETTE SYST?1
a. Is a pUte and frame style holder which accepts both ultra thin and iepti
t»p* filters.
h, Has fron 0.5 to 25 ft* of filter area.
c. Has not been Investigate*4 thoroughly but has had some success In virus
concentration.
d. Its main application for cyst recovery nay lay with the processing of
filter washings.
• 8. FILTERWASHW APPARATUS
a. This is a proposed device by OuWalle, 1982 from U. of W., for unwinding
the fibers *rom the filter cartridge while repeatedly brushing and
squeezing them while 1n a bath solution.
b. Bath could contlin either a surfactant or pH controlled solution.
c. Potential claims are as high as 75 percent extraction of cysts from the
fibers.
TASLE 2;
DETECTION METHODS
IHVESTICATOR(S)
Riggs, CSDHS Lab, Berkley, CA
1983
method
RESULTS
1. Inmunof! uorescence
a.~lfl
Good prep..
Cross Rx
b. I FA
Sauch, *PA-C1nc1nnat1
Riggs, CSDS
Still under stu
c. Monoclonal Antibodies Rigqs, CSDHS
Sauch, EPA-Cincinnati
(unpublished!
Still under stu
2. EL ISA Method
Hungar, J. Hopkins MD, 1983 Feces samoles o
3. Briqhtfleld/Phase Contrast EPA Consensus method
Ongoing
-------
APPENDIX C: CONCENTRATING. PROCESSING,
DETECTING AMD IDENTIFYING 6lAP.PIA CYSTS IN WATER
DETECTION METHODS
1.8. niRECT FLUORESCENT ANTIBODY (DRA) TECHNIQUE
1. R1ggs has oroduced a high titer purified Immune sera to Giardla lamblia
cysts 1n guinea pigs and labeled it with Huorecein Isothio cyanate. Jen
Is purified thru NH4OH and DEAE se^adex fractionation.
2. Obtained cross reactions with ChUonastlx aesnlH cysts but claims 1t can
be easily distinguished from Giardia by its smaller size..
l.b. INDIRECT FLUORESCENT ANTIBODY (IFA) TECHNIQUE
1. Sauch using IFA with Iranune sera from rabbits (unourlfied). It 1s reactei
with commercially available fluorescent-labeled goat ant1-rabb1t gamma
globulin. ,
2. Some cross-reactions with certain algal cells.
1.e. MONOCLONAL ANTIBODIES
1. Using clones of hybrldoma cell lines obtained by fusing mouse myeloma
cells with spleen cells from mice (BALB/c) Immunized with G. lanblia
troohozoltes. "~
2. Produced eight monoclonal antibodies evaluated by IFA against both trophs
an* cysts.
a. 3/8 stained the ventral disk
b. 2 stained the nuclM
c. 2 stained cytoplasmic granules
ri. 2 stained membrane components
3. Variability 1n staining may be due to differences 1n stages of encystme'nt
4. Preliminary results Indicate nonodonal ABs may give rapid and specific I
0* cysts.
5. Rx may be too specific, not reacting with all human forms of 6. 1ambl1 a
may have to go to polyclonal ABs.
2. EL ISA *€TH0D
a. Hungar at John Hopkins (unpublished) has produced a detection method by
EL ISA using a intact "sandwich" technique 1n 96-weli microtiter plates.
b. Using antlsera from 2 different animals (may present problem).
c. Me«d a nlnlmno of 12 cysts/wel'' for color Rx.
-------
APPENDIX 0
ANALYTICAL REQUIREMENTS OF THE SWTR AND
A SURVEY OF THE CURRENT STATUS OF RESIDUAL DISINFECTANT
MEASUREMENT METHOOS FOR ALL CHLORINE SPECIES AND OZONE
-------
APPENDIX 0
ANALYTICAL REQUIREMENTS
Only the analytical method(s) specified in the SWTR, or otherwise
approved by EPA, may be used to demonstrate compliance with the
requirements of the SWTR. Measurements of pH, temperature, turbidity, and
residual disinfectant concentrations must be conducted by a party approved
by the Primacy Agency. Measurements for total coliforms, ftcal xoliforms,
and heterotrophic bacteria as measured by the heterotrophic plate count
(HPC), must be conducted by a laboratory certified by the Primacy Agency
or EPA to do such analysis. Until laboratory certification criteria are
developed for the analysis of HPC and fecal coliforms, any laboratory
certified for total col 1 form analysis is acceptable for HPC and ftcal
coliform analysis. The test methods to be used for various analyses are
listed below;
(1) Fecal coliform concentration - Method 908C (MPN Procedure),
9080 (Estimation of Bacterial Density), or 909C (Membrane
Filter Procedure) as set forth in Standard Methods for the
Examination of Water and Wastewater. American Public Health
Association, 16th edition.
(2) Total coliform concentration - Methods 908A, B, 0 (MPN
Procedure) or 909A, B (Membrane Filter Procedure) as set forth
in Standard Methods for the Examination of Water and
Wastewater, American Public Health Association, 16th edition;
Autoanalysis Colilert (EPA refers to this as Minimal Medium
0NP6-MU6 Method), as set forth In Applied and Environmental
Microbiology, American Society for Microbiology, Volume 54,
No. 6, June 1988. pp. 1595-1601.
(3) Hettrotorphic Plate Count - Method 907A (Pour Plate Method),
as set forth in Standard Methods for the Examination of Water
and Wastewater. American Public Health Association, 16th
edition.
(4) Turbidity - Method 214A (Nephelometric Method) as set forth in
Standard Methods for the Examination of Water and Wastewater,
American Public Health Association, 16th edition.
(5) Residual disinfectant concentration - Residual disinfectant
concentrations for free chlorine and combined chlorine must be
measured by Method 408C (Amperometric Titration Method),
Method 4080 (DPD Ferrous Titrimetric Method), Method 408E (DPD
Color-metric Method), or Method 408F (leuco Crystal Violet
0-1
-------
Method) as set forth In Standard Methods for the Examine
-------
PREFACE
The AWWA paper entitled "A survey of the current status of residual
disinfectant measurement method for all chlorine species and ozone" will
be included in the final document. It has not been Included here for the
sake of brevity. However, the publication is available from the AWWA
Customer Services Department, 6666 W. Quincy Avenue, Denver, Co. 80235;
Telephone (303) 794-7711. The document publication number is 90529.
The above publication summarizes the AWWA Research foundation's 816
page publication entitled " Disinfectant Residual Measurement Methods",
publication number 90528. This document Is also available from the
customer services department listed above.
-------
A SURVEY OF THE CURRENT STATUS OF RESIDUAL DISINFECTANT
ttASUREHEHT METHODS FOR ALL CHLORINE SPECIES AND OZOHE
by
Gilbert Gordon
Oepartatnt of Otcsni stry
Mi ami University
Oxford, OH 45055
Will fan J. Coooer
Drinking Water Restarc.1 Center
Florida International University
Miami, Florida 33159
Rip 6. Rice
R1ce, Incorporated
Asnton, Maryland 20861
Gilbert £. Pacey
Department of Chemistry
Miami University
Oxford, Ohio 45056
Prepared for:
AWWA Research Foundation
6665 W. Ouincy Avenue
Denver, CD 30225
November 1987
Published by the American Water Works Association
-------
DISCLAIMER
This study was fund id by the American Water Works Association
Research Founcation (AWWARF). AWWARF assumes no responsibil-
ity for the cancans of the research study resortaa 1n this
publication, or for the opinions or statements of fact
exoressed in the resort. T!i« stent ion of tradenames for
commercial products does not represent or Imply the approval
or endorsement of AWWARF. This report is presented solely
for informational purposes.
Although the research described In this document has been
funded in part by tne United States Environmental Protection
Agency tnrougn a Coooerative Agreement, CR-311225-01, ta
AWWARF, it has not been subjected to Agency review and
tnerefcre does not necessarily reflect the views of the
Agency a no no official endorsement snouid be inferno.
C:pyricnt 9 1S87
by
American Water wor*s Association Research Foundation
Printed in U.S.
-------
FOREWORD
7h11 resort is oart of the on-going research program of the AWWA Research
Founcation. The research described 1n tne following pages was funded by
the Founcation in benalf of 1t3 memoers and subscribers in particular anc
tne water supply industry in general. Selected for funding by AWWARF's
Board of Trustees, We project was identified as a practical, priority need
of tne incustry. It is hooed that this publication will receive wide and
serious attention and that its findings, conclusions, and recommenoations
will be applied in communities througnout the United States ano Canada.
The Research Foundation was created by the water suoply industry as its
center for cooperative researcn and development. Tne Foundation Itself
does not conduct research; 1t functions as a planning and management
agency, awarding contracts to other institutions, such as water utilities,
universities, engineering fira$, and other organizations. The scientific
and technical exoertise of the staff is further enhanced by industry
volunteers wno serve on Project Advisory Co ran ctees and on otner standing
committees and councils. - An extensive planning process involves many " •
hundreds of water professionals in the important task of keeoing the
Foundation's"program resoonsive to the practical, ccerational neecs, of
local utilities and to the general research and development needs of a
progressive incustry.
All aspects of water suoply are served Oy AWWARF's research ace.nca:
resources, treatment anc operations, distribution and storage, water
quality and analysis, economics and management. The ultimate pursose of
this effort is to assist local watar suopliers to provide tne hignest
possible quality of water, economically and reliably. The Foundation's
Trustees art pleases to offer tr.is puoiication as contribution towar- tnat
end.
This project reviewed all disinfectant residual measurement methods for
free chlorine, chloramines, ozone and chlorine dioxide with special
attention to interferences tnat could be experienced by tne water utility
industry. Reccsnenoations, practical guidance, and cautions on tne
selection of acoropriate residual measurement techniques are summarized
(Please see Preface for information cn full resort).
«e/*:.te i. ji ,aert
£nairrnan, Scare of T
-uwa Rasearo.*? F cunca
fJmes r. Manwarmg, ?.i.
^-f!Tecutive Director
AWWA Researcn Founcation
-------
.PREFACE
This document summarizes the AWWA Research Foundation's 315 pace
puo11cation "Disinfectant Residual Measurement Methods." That
puo11 cation (Publication Numoer 90523) can be oraerea frsm tne A«WA
Cultsmer Services Qepareaent, Sooe W. Quincy Avenue, Oenver, CO 30235;
telepnone, (303) 794-77U.
Trie purpose of this sunwary document 1s to provide the water utility '
lafioratory analyst with guidance in selecting disinfectant residual
measurement methods. Either this document or the full resort is
reccnaended as a comoanlon to Standard Methods for the Examination of
Water ana Wastewater.
-------
AOCWWLED SEME NTS
The authors w1sn iz txaress t.lt1r appreciation ta tne American Watir Wor
-------
ITTrTTTTtf* CJ1VWADV
iMUVilvCi dUflllMZ
8ACKCR0TO
The objeeclve of this Report 1* co reviev and suaaarize all dlsinfeeeanc »•
sidual aeasureaent techniques currently available foe free chlorine (along wich
cha various ehloraalnas), eoabined ehlorlna, chlorite ion, ehlorlna dioxide,
chlorate Ion, and oxon*.
Presently, both chlorine dloxida and ozone art gaining considerable favor as
alternatives eo ehlorlna disinfection (I). The analytical cheatstry for chest
disinfectants when compared vleh ehlorlna Is even nora complex and lass readily
understood as evidenced by various aurveys <2-5> and dacalled research carried
ouc in varioua laboraeorlaa (6*10).
Chlorine dloxida is aanufaccured ac cha siee of lea uaa by raacclons involv-
ing aodiua ehlorice, chloraca ion, ehlorlna gaa and/or hypochlorite ion and sul-
furic acid or hydrochloric acid (11-12). Consequently/chlorate ion, ehlorice
ion, hypochlorite Ion and/or hypochlorous acid frequently will ba found occur-
ring as by-produces or unraacced scarclng aacarlals. Theae aacerlals are strong
oxidizing agencs vhieh ara vary raaceiva and behave in aany ways siallar so
ehlorlna dioxide itself.
There ara aora than 2,000 vacer creacaanc planes today using ozone, and less
chan half of ehea are applying ozone solely for disinfection. The large oajor-
icy of vacer ereacaenc planca use ozone as a cheaieal oxidanc. Many of the
planes applying ozone for dislnfaeeion also are using ozone, in ehe *aae plant,
for ehealcal oxidation. Analyses for residual ozone in vacer are applicable
only in cha treatment plane,. aicher in ehe osona coneactor(s) or ac their
ouclaes. Residual ozone la never present in ehe discribution systeo; however,
its by-produces say be.
There have been nuaeroua accaapcs eo evaluate ehe relative advaneagas and
dlsadvaneages associated vleh ehe aeasureaent of free and eoabined chlorine.
Different crleerla are frequently uaed for the evaluaeion of ehe analytical
aeasureaenes and ofeen suggesclona for the laproveaent of eese procedures have
gone largely ignored. No coaprehenaive and objactive review of ehe licaraeure
appears eo ba available. This Report la aiaad ae providing such a reviev along
vleh guidance and recoaaendaeions aa eo vhae erieeria vacer utilities should use
in selecting residual aonlcoring cechniques based on clreuascanees by category.
OBJECTIVES
•¦¦anAdUUHl
1. To review and suaaarize all residual aeasureaent eechniques
currently available for fraa chlorine--caking Into account
the roles of ehloraalnas.
2. To reviev and suoaarisa all residual aeasuraaenc eechniques
currencly available for eoabined chlorine.
I
-------
3. To briefly review the present understanding of the chlorine-
aaaonia chemistry and In particular, in relationship eo the
aeasureaent of chlorine and combined chlorine.
4. To review and suaaarize all residual aeasureaent techniques
currently available for chlorine dioxide, chlorite ion and
chlorate Ion.
5. To review and suaaarize the analytical procedures currently
used by operating water utilities to control ozone treatment
processes, considering disinfection as well as the aany oxid-
ative applications of ozone in water treataent applications.
6. To discuss coaaon Interferences associated with the aeasureaent
of each of the dlsinfectants/oxidants described above (free
chlorine, coablned chlorine, chlorite ion, chlorine dioxide,
chlorate ion, and ozone).
7. To provide guidance and recoaaendation for water utilities in
selecting residual aonitoring techniques for each of the above .
dlsinfectants/oxidants.
8. To recoaaend future research for developaent of aonitoring and
analytical aethods to laprove accuracy, and reduce tiae and cost
requireaents for the aeasureaent of the above disinfectants.
In Che full report, we present as coaplete as possible an exaalnation of the
world-wide body of literature on analytical aethods used by the wecer utility
industry in order co elaborate on the various probleas, advantages. disadvan-
tages and known interferences for each of the currently used analytical aethods.
Foreaost In our objectives has been a better understanding of the reliabil-
ity of various aeasureaents which have been carried out. Since there are inher-
ent limitations in all aeasureaents, it becoaes apparent that there are specific
needs for some indication of the reliability of the result, i.e., what is she
precision and accuracy of the reported value, and are these acceptable?
The volatility of most of the disinfectants makes sampling and sample
handling a major contributor to laprecision and Inaccuracies. "Standard
additions" is a questionable technique; it should be avoided If possible, since
the pipetting and dilution process causes potential loss of disinfectant.
The relative usefulness of each aethod, along with descriptions of known
Interferences such as turbidity, organic aatter, ionic materials, solids, color,
buffering capacity, as well as the nature of the sample and the time between
collection of the sample and the actual analysis, are described in this report.
It must be emphasized, however, that almost invariably each of the methods
described is based on the total oxidizing capacity of the solution being
analyzed and is readily subject to interferences from the presence of other
potential oxidizing agents and/or intermediates from concomitant' chemical
reactions. Under ideal conditions some of the methods are accurate co better
2
-------
than Si**••specially In eh# absence of coaaon interferences••whereas other
aethod* are alaose seai-quantitative in nature with aany coaaon species
interfering with both eh* precision and accuracy o£ the aeasureaents.
tfe have also included chlorate ion as a residual species in ehae only
recently have reliable analytical aethods begun to appear in the literature
(5,6.10). . We also report on the cheaistry of the chlorine*aaaonia system ar.d
the associated breakpoint reactions, because one of the aost coaaon inteferences
in the aeasureaent of free chlorine is aonochloraaine.
The aost laportant davelopaent for this report has been the decision to in-
clude a preliminary section describing an "idealized" analytical aethod. The
need for this section becaae apparent as our writing progressed describing .each
of the analytical aethods for chlorine. Specific lteas included in this "ideal*
lzed" aethod are accuracy, precision, reproducibllty, lack of interferences,
ease of use of ehe aethod, lack of false positive values, and so forth.
The benefit of the "idealized" analytical aethod is to allow individual coo*
parisons and to allow the eholce between various aethods based on individual
aethod shorteoaings. For exaaple, a particular aethod aight have as its only
difficulty Interference by aanganese and iron. In aany circuastanees. this type
of interference aight be a aajor problea. However, should the water supply
under consideration not have any aanganese or iron, it is quite likely that ;he
aethod aight be very usable*-and as a aatter of face well aight be the best
aethod of choice.
In other cases, speed of analysis rather than potential Interferences (or
choice of soae other laportant characteristic) aight be the guiding factor in
choosing an analytical aethod. In this way rational choices can be made based
on potential and/or real difficulties and/or interferences and as compared co an
"idealized" aethod •• rather than a possibly controversial existing method.
Table I has been constructed as a quick reference guide to the available
aethods for the determination of water disinfection cheaicals and byproduct.
Included are pertlnenc analytical characteristics such as detection limits,
working range, Interferences, accuracy and precision estlaates. The current
status of the aethod, as gleaned froa this report, is given. The necessary
operator skill level is given to aid the laboratory aanager in assessing the
aanpower requireaents for a particular aethod. Additional information
concerning the reasons for Che current status is contained in the Recommendation
Section of the Executive Suoaary and the coaplete report.
As each of ehe Mthods is described in detail in the full report, specific
conclusions are drawn*-along with appropriate recommendations- *by comparing the
aethod agalnse ehe "idealized" analytical aethod for that species.
One additional benefit of this direct coaparlson is the possibility of add-
ing or subtracting a aethod to the list of Standard H«ehod» for ehe Examination
of Water and Vntwit»r (13), based on a rational set of criteria. It should
also be possible in ehe fueure eo decide on ehe viability of various methods
based on their aeeting specific criteria rather than based only on comparisons
between analytical laboratories (and personalized subjective reactions co ehe
various aethods theaselves
3
-------
TABU X. CHARACTERISTICS AND COKfARISONS OF ANALYTICAL METHODS®
Sp.cl.i' DETECTION »ORXlNG EXPECTED EXPECTED
TYPE OF TEST MEASURED LIKXT RANGE ACCURACY PRECISION SKILL*
(METHOD)' DIRECTLY (± I) (± %) LEVEL
FREE CHLORINE
"Ideal" CI, + HOCI/OCI" 0,001 0.001-10 0.3 0.1 1
UV/VISIBLE CI, + HOCl/OCl' - 1 1 • 100 NR NR 3
Continuous CI, + HOCl/OCl" 1.3 1.5 - 300 NR * . NR 3
AMPEROMETRIC TITRATION:
Forward CI, * HOCl/OCl" 0.0018* >10 NF NF 2
0.02 • 0.03* >10 NF 3 - SO 2
Back CI, ~ HOCl/OCl" 0.002 > 10 3 - .50 SF 2
Continuous CI, + HOCl/OCl" 0.005 > 10 NR 1.0 2/3
IODOMETRIC TITRATION:
Standard (Total Chlorina) 0.07® 0.1 • 10 NR Ml 2
0.35* 0.5 - 10 NR NR 2
DPD .
FAS Tlt'n CI, ~ HOCl/OCl" 0.004» 0.01 - 10 NF 2-71
0.01l« 0.01 - 10 NF 2-7 1
Color*atre CI, ~ H0C1/0C1" 0.01* 0.01 - 10 1-15 1-14 I
Setadlfac CI, + HOCl/OCl" 0.01* 0.01 - 10 NF NR 1/2
LCV
Blaek and
tfhittla CI, ~ HOCl/OCl" 0.01 0.25 - 3 NF NR .1
Whittle &
Lapctff CI, ~ HOCl/OCl" 0.01 0.25-10 NR 0-10 2
4
-------
TABLE I. CHARACTERISTICS (cont'd)
STABILITY
FIELD
CURRENT
REAGENT
PRODUCTS
INTERFERENCES
pH RANGE
TEST
AUTOMATED
STATUS
5 YRS
> 1 DAY
NONE
Independent
YES
YES
RECOMMENDED
NA
NA
C1NH, - CI jN
pH Dependent
NO
NO
RECOMMENDED
btekpid Abs
(LAB TEST)
NA
NA
CINHj « C1,N
pH Dependent
NO
YES -
CONT'D STUDY
1-2 yrs
HA
ClNHa - C1,N
pH Dependent
YES
YES
RECOMMENDED
1-2 yrs
NA
CINHj - C1,N
pH Dependent
YES
YES
RECOMMENDED
1-2 yrs
NA
CINHj - CI,SI
pH Dependent
YES
YES
RECOMMENDED '
1-2 yrs
NA
ClNHa - CI,*
pH Dependent
YES
YES
RECOMMENDED
1 yr
10 ain
All oxidizing
pH Dependent
NO
NO
RECOMMENDED
or less
species
(LAB TEST)
I yr
10 sin
All oxidising
pH Dependent
NO
NO
RECOMMENDED
or less
species
(LAB TEST)
powder
30 ain
cum, - ci,tt
Requires
NO
NO
RECOMMENDED
stable*
oxid species
buffer
(LAS TEST)
powder
30 ain
cum, - Cl,lt
Requires
HO
80
RECOMMENDED
stable*
oxid species
buffer
(LAB TEST)
powder
30 Bin
C18H, - Cl,N
Requires
YES
NO
RECOMMENDED
stable*
oxid species
buffer
(FIELD TEST)
powder
30 aln
C1MH, • C1,N
Requires
YES
. NO
RECOMMENDED
•cable*
oxid species
buffer
(FIELD TEST)
aonths
NR
C1KH, • C1,N
Requires
YES
NO
ABANDON
oxid species
buffer
aonchs
NR
Oxidizing
Buffering
YES
NO
RECOMMENDED
species
(LAB TEST)
5
-------
TAILS I. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS* (cont'd)
Special DETECTION WORKING EXPECTED EXPECTED
TYPE Or TEST MEASURED LIMIT RANGE ACCURACY PRECISION SKILL*
(METHOD)* DIRECTLY (¦g/L) («g/L) (± I) (± t) LEVEL
FACTS
Color'acre CI, + HOCI/OC1" " O.l 0.25 - 10 5 • 20 1-11 1
Space'phoco CI, + H0C1/0CI" 0.012 0.05 • 10 NF MR 1
METHYL ORANGE CI, + HOC1/OCI" NR NR NR NR 2
O-TOLIDINE CI, * H0C1/0C1" NR NR NR NR I
3,3'-DIMETHYLNAPHTHIDINE
CI, ~ HOCl/OCl" 0.05 NR NR 2-6 2/3
O-DIANISIDINE CI, ~ HOCl/OCl" 0.1 NR NR NR 2/3
CHEMILUMINESCENCE
Hydrogen
Peroxide CI, ~ HOCl/OCl" NR NR NR NR 3
Luainol OCL" 0.0007 NR NR NR 3
Lophina OCL" 0.14 0.2 - 20 NR NR 3
ELECTRODE METHODS
Maabrana HOCL 0.004 0.04 - 1 NR 1.6 3
Bara-vlra CI, ~ HOCl/OCl* 0.1 0.1-3 NR 1-25 3
Pocane'acre CI, + HOCl/OCl" 0.005 0.01 - I I - 6 7-10 2
Agl
Volc'acrc CI, + HOCl/OCl" * 0.01 0.1 - 10 NR NR 3
6
-------
TABLE X. CHARACTERISTICS (eons'd>
STABILITY •
REAGENT PRODUCTS INTERFERENCES pH RANGE
FIELD
TEST
CURRENT
AUTOMATED STATUS
2 y«ariT 30 aln Oxidizing
ac high CI, spaeias
2 yaarsT 30 Bin Oxidizing
ic high CI, spaeias
NF NF Oxidizing
Spaeias
NF NF Oxidizing
spaeias
NF 13*20 Bin Oxidizing
spaeias
NF 55 ain Oxidizing
spaeias
NR <1 sac Nona
NR <1 sac Oxidizing
spaeias
NR <1 sac Nan*
NA NA Oxidizing
Cm spaeias
NA HA Oxidizing
spaeias, CI"
3 months NA Oxidizing
spaeias CI"
NA NA Oxidizing
spaeias, CI*
Buffaring YES
critical
Buffaring YES
critical
Buffaring YES
raquirad
Buffaring YES
raquirad
NR NO
NR NO
Indapandanc NO
pH Dapandane NO
pH Dapandane NO
Dapandant POSSIBLE
on pK
NR POSSIBLE
pH Dapandane YES
Buffar POSSIBLE
raquirad
7
NO RECOMMENDED
NO _ RECOMMENDED
NO ABANDON
NO ABANDON
NO ABANDON
NO ABANDON
POSSIBLE ABANDON
POSSIBLE CONT'D STUDY
YES CONT'D STUDY
POSSIBLE CONT'D STUDY
POSSIBLE CONT'D STUDY
YES RECOMMENDED
POSSIBLE CONT'D STUDY
-------
TAILS X. CHARACTERISTICS AND COMPARISONS Of ANALYTICAL METHODS® (conc'd'.
Species' DETECTION UORKINC EXPECTED EXPECTED
TYPE OF TEST MEASURED LIMIT RANGE ACCURACY PRECISION SKI
(METHOD)' DIRECTLY <«I/L> <«g/L> 10 NF NF 2
NH,C1 NHCl, NCI,
• CI, ~ H0C1/0C1* 0.02 -0.03* >10 NF 3 ~ 50 2
NH,C1 NHCl, NCI,
Sack CI, ~ HOCl/OCl" 0.002 > 10 3 - 50 :!F 2
NH,C1 NHCl, NCI,
Concinuouj CI, * HOCl/OCl* 0.005 >10 NR 1.0 2/3
NH,C1 NHCl, NCI,
I0D0METRIC TITRATION:
Standard CI, + HOCl/OCl* 0.07s 0.1 - 10 NR NR 2
NH,Cl NHCl, NCI,
CI, ~ HOCl/OCl* 0.35* 0.5 - .100 NR NR 2
NH,C1 NHCl, NCI,
DPD
FAS Tlc'n
CI, +
HOCl/OCl-
0.006*
0.01 - 10
NF
2 - 7
I
NH,C1
NHCl, NCI,
CI, *
HOCl/OCl*
0.1l«
0.01 - 10
NF
2 ~ 7
I
UK,CI
NHC1, NCI,
Color'sere
CI, ~
HOCl/OCl*
0.00l»
0.01 - 10
I- 15
1-14
I
NH,C1
NHCl, NCI,
LCV
Black &
Vhletia CI, ~ HOCl/OCl" 0.005 0.25 • 3 NF 4 - 10
NH,Cl NHCl, NCI,
8
-------
TABLE X. CHARACTERISTICS (cont'd)
STABILITY FIELD CURRENT
REACENT PRODUCTS INTERFERENCES pH RANGE TEST AUTOMATED STATUS
5 YRS > I DAY
HOKE
Independent YES YES RECOMMENDED
of pH
I - 2 yrs NA
1 - 2 yrs MA
1 • 2 yrs NA
I • 2 yrs HA
Oxidising
Species
Oxidizing
Species
Oxidizing
Species
Oxidizing
Special
pH Dependent YES
pH Dependent YES
pH Dependent YES
pH Dependent YES
YES RECOMMENDED
YES RECOMMENDED
YES RECOMMENDED
YES RECOMMENDED
I yr 10 Bin All oxidizing pH Dependent NO NO RECOMMENDED
species (LAB TEST)
1 yr 10 Bin All oxidizing pH Dependent NO NO RECOMMENDED
species (LAB TEST)
povdar 30 ain
stabla*
povdar 30 ain
stable*
povdar 30 ain
stable*
Oxidizing
Spacias
Oxidizing
Species
Oxidizing
Spacias
Requires NO NO RECOMMENDED
buffer (UB TEST)
Requires YES NO RECOMMENDED
buffar (FIELD TEST)
Raquiras YES NO RECOMMENDED
buffer (FIELD TEST)
nonchs
NR
Oxidizing
Spacias
Raquiras
buffar
YES
NO
ABANDON
9
-------
TABLE I. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS* (conc'd)
Spaciat* DETECTION WORKING EXPECTED EXPECTED
TYPE OF TEST MEASURED LIMIT RANGE ACCURACY PRECISION SKILL*
(METHOD)* DIRECTLY (ag/L) (4 %) (± %) LEVEL
Hhltela &
Up caff CI, ~ HOCI/OC1* O.Ol 0.23 * 10 NF 4-10 2
NH,Cl NHCI, NCI,
FACTS
Color'mere CI, * HOCl/OCl* O.I 0.25 -10 3 • 20 I*- ii I
NH,Cl NHCI, NCI,
Space'phoeo CI, ~ HOCl/OCl" 0.012 0.03 - 10 NF IJR I
NH,C1 NHCI, NCI,
ELECTRODE METHOOS
Poe'aaeric CI, + HOCl/OCi* 0.003 0.01 -1 1-6 7-10 2
NH,C1 NHCI, NCI,•
MONOCHLORAMINE*
"Idaal* NH,C1 0.001 0.001 - 10 0.5 0.1 1
UV/VISIBLE NHjCl - 1 1-100 NR NR 3
AMPEROMETRIC TITRATION;
Forward NH,C1 NR > 10 NF 0 - 10 2
Sack NH,C1 NR >10 NF NF 2
DPD
FAS Tlt'n NH,Cl NR 0.01 - 10 NF 2-7 1
Color'acre NH,C1 NR 0.01 * 10 NF S - 75 1
10
-------
TABLE I. CHARACTERISTICS (cont'd)
STABILITY FIELD CURRENT
REAGENT fROMICTS INTERFERENCES pH RANGE TEST AUTOMATED STATUS
aonehs NR
Oxidizing
Species
Buffering YES
NO RECOMMENDEDD
(Las test)
2 YRS 30 ain Oxitflzlns
at high Species
CI,
2 YRS 30 aln Oxidizing
at high ipulu
CI,
Buffering
critical
Buffering
critic*!
YES
YES
NO
NO
RECOMMENDED
RECOMMENDED
3 aonehs NA
Oxidizing pH Dependent YES YES RECOMMENDED
Species. CI"
3 YRS >1 DAY
NONE
Independent YES
YES RECOMMENDED
NA
NA C1,NH • CljN
backgnd Abs
pH Dependent NO
NO RECOMMENDED
(LAB TEST)
1-2 yrs NA C1,NH - C1,N pH Dependent YES YES RECOMMENDED
1-2 yrs NA Ci,NH • Cl,N pH Dependent YES YES RECOMMENDED
powder 30 ain
¦cable*
powder 30 ain
stable*
C1NH,
CI,9
oxid species
C1NH,
oxid speeies
Requires NO NO RECOMMENDED
buffer (LAB TEST}
Requires YES NO RECOMMENDED
buffer (FIELD TEST'
11
-------
TAALE I. CHARACTERISTICS AND COMPARISONS Of ANALYTICAL METHODS* (canc'd)
TYPE OF TEST
(METHOD)»
Sp«cU*» DETECTION VORXING EXPECTED EXPECTED
MEASURED LIMIT RANGE ACCURACY PRECISION SKILL*
DIRECTLY (as/L) (ng/L) (± 1) (£ %) LEV5L
LCV
Uhltel* &
Lapt«££
NH,Cl
m
0.25 - 10 HF 0 . 43 2
ELECTRODE METHODS
Silver lodld*
Vole*aa«erie
KH,Cl
MR
0.1 - 10 NR NR
DICHLQRAM1N1*
UV/VISIBLE
:«ci,
NHCl,
0.001 0.001 . 10 0.5 0.1
- 1
1 - 100
MR NR
AMPEROMETRIC TITRATION:
Forward
Back
NHCl,
NHCl,
NR
NR
> 10
> 10
NF 0
3 • 50 NF
DPD
FAS Tie'n
Co lor'acre
NHCl,
NHCl*
NR 0.01 - 10 NF NF 1
NR 0.01 - 10 NF 0 - 100 1
LCV
uhletl« &
Laptoff
NHCl,
NR
0.25 - 10 NF 10 - 150 2
12
-------
TABLE I. CHARACTERISTICS (cons'd>
STABILITY . FIELD CURRENT
REAGENT PRODUCTS INTERFERENCES pH RANGE TEST AUTOMATED STATUS
months
NR
Oxidizing
species
Requires
buffer
YES
NO
RECOMMENCED
(Las tests
NA NA Oxidizing Requires POSSIBLE POSSIBLE CONT'D STUDY
species buffer
5 YRS > 1 DAY NONE Independent YES YES
of pH
NA NA C1NH, & CIjN pH Dependent NO NO
baekgnd Abs
RECOMMENDED
RECOMMENDED
CIA3 TEST)
1-2 yrs NA C1NH, & ClsN pH Dependent YES YES RECOMMENDED
L-2 yrs NA C1NH, & Cl,N pH Dependent YES YES RECOMMENDED
powder 30 sis
stable*
povder 30 sin
stable*
CINH, & CIjN Requires
oxid species buffer
CUM, 6 CI jN
oxid species
Requires
buffer
NO NO RECOMMENDED
(LAB TEST)
YES NO RECOMMENDED
(FIELD TEST)
aonehs
Nit
Oxidizing Requires YES NO RECOKMENDEDD
species buffer (LAB TEST)
13
-------
TAILS X. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS® (cont'd)
TYPE OF TEST
(METHOD)>
TRICHLORAMINE*
"Idaal"
Spaeias'
MEASURED
DIRECTLY
NCI,
DETECTION VORXINC EXPECTED EXPECTED
UMIT RANGE ACCURACY PRECISION SKILL*
(»I/L) (ag/L) (S %) (± %) LEVEL
0.001 0.001 • 10 0.5 0.1
UV/VISIILE
NCI,
NR
NR
NR NR
AMPEROMETRIC TITRATION:
Forward NCI,
NR
> 10
NF 5 - 100
DPD
FAS Tit'n
Color'atre
NCI,
NCI,
NR 0.01 - 10 NR NR 1
NR 0.01 - 10 N*R :."R I
LCV
VhiceU &
Lapeaff
NCI,
NR
0.25 • 10 NR
.NR
CHLORINE DIOXIDE
"Ideal"
IODOMETRIC
AMPEROMETRIC
DPD
UV
Manual
CIO,
CIO,
0.001 0.001 • 10 0.5 0.1 1
0.002 0.002 - 95 1 - 2 1-2 2
CIO,19 0.012 0.02 • ?f 1 • IS 1 • 15 3
CIO,1 *"11 0.008 0.008 - 20 10 7-15 2
CIO,
0.05 0.05 • 500
FIA
CIO,
0.25 0.25 - 142
14
-------
TAILS I. CHARACTERISTICS (cont'd)
STABILITY FIELD CURRENT
REAGENT PRODUCTS INTERFERENCES pH RANGE TEST AUTOMATES STATUS
3 YRS > 1 DAY
NONE
Independent YES
YES
RECOMMENDED
NA
NA CINH, • CljNH pH Dependent NO
backgnd Abs
HOCi/OCL-
510 RECOMMENDED
(LAB TEST)
1-2 yrs
NA
C1VH, - C1,NH pH Dependent NO
YES RECOMMENDED
(LAB TEST)
powder 30 aid
stable*
powder 30 ain
stable*
C1NH, - C1,NH Riqulru SO NO RECOMMENDED
ox id species buffer (LAB TEST)
CINHj - C14NH Requires YES NO RECOMMENDED
oxid species buffer (LAB TEST)
aonths
NR
Oxidizing Requires YES NO RECOMMENDED
species buffer (LAB TEST)
5 YRS
I YR
good
> 1 DAY
NONE
Subject eo Oxidizing
oxidation species
Subject to Metal ions &
oxidation nitrite ion
Independent YES
NO
solid < 30 ain
stable*
none
none
none
none
Oxidizing
species
Other UV
absorbers.
none
2 • 3
?
1
NO
NO
Independent NO
Independent NO
YES RECOMMENDED
NO NOT RECOMMENDED
NO CURRENTLY USED
NO SOT RECOMMENDED
YES
YES
RECOMMENDED
(US TEST)
RECOMMENDED
(LAB TEST)
15
-------
TABLE I. CHARACTERISTICS AND COSFARISONS Of ANALYTICAL METHODS® (cont'd)
TYFE OF TEST
(METHOD)•
Spool**'
MEASURED
DIRECTLY
ACVK1' CIO,
CHLOROFHENOL RED CIO,
DETECTION WORKING EXFECTED EXFECTED
LIMIT RANGE ACCURACY PRECISION "SKILL*
(ag/L) (ag/L) (it) (t %) LEVEL
o-TOLIDINE
CIO,
0.04
0.003
O.I
0 - 25
0.003 • I
NR
NR
10
NR
NR
5
NR
INDICO BLUE CIO,
CHEMILUMINESCENCE
Lunlnol CIO,
CDF1A*' CIO,
ELECTROCHEM.
FC Microoloc. CIO, + CIO,"
Vic. Carbon CIO,
Vole*a. M«a.
Rotating Vole.
Moabrano
CHLORITE ION
"Idoal"
AMFEROMETRIC
lodoaocrie
IODOMETRIC
Sequential
Modified
CIO,
CIO,
C10,"
ClO,"
C10,
CIO,
0.01
0.3
O.OOS
1.3
32
0.25
0.30
0.001
0.0S
0.011
0.3
NR
0.3-1 •
0.005 • 74
NR
NR
NR
0.30 * 3
0.03
NR
NR
2
7
SR
SR
NR
1.5
0.001 - 10 0.5
95
SR
SR
NR
6.4
O.l
> I
O.S . 20 0.5 1 . 3
i
1
2/3
3
2
2/3
DFD
CIO,
0.01
0.01 - 10 5
16
-------
TA3LE X. CHARACTERISTICS (eone'd)
STABILITY FIELD CURRENT
REAGENT PRODUCTS INTERFERENCES pH RANGE TEST AUTOMATED STATUS
m
6 aonths
NR
good
NR
NR
NIC
good
ainiaal
unknown
Oxidizing
spaeias
0, CI,
8.1 - 8.4 NO NO CONT'D STUDY
7 YES NO TOT RECOMMENDED
NR NO NO NOT RECOMMENDED
>4 NO NO NOT RECOMMENDED
1 DAY
1 DAY
< 1 sae
< 1 sac
NR
CI,
NR
> 12
NO NO NOT RECOMMENDED
NO YES RECOMMENDED
CONT'D STUDY
none
none
none
none
none
non«
ClO,"
C10a-
HOCi
5-5.5 NO NO . CONT'D STUDY
3.5-7 NO NO CONT'D STUDY
7.8 NO NO CONT'D STUDY
nona
nona
H0C1
5 - 5.5
NO
NO
CONT'D STUDY
5 YRS > 1 DAY
NONE
Independent YES
YES RECOMMENDED
i YR Subjaee eo Oxidizing
oxidation spaeias
2 - 5
NO
NO SOT RECOMMENDED
good Subjaee eo Maeal ions & 7
oxidation nieiea ion
good Subjaee eo Maeai ion* & 2
oxidation nieiea ion
Solid
stable*
< 30 ain
Oxidizing
spaeias
NO
NO
NO
NO RECOMMENDED AT
HIGH CONC.
NO
CONT'D STUDY
NO NOT RECOMMENDED
17
-------
TAILS Z. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS* (conc'd)
TYPE or TEST
(METHOD)~
CHLORATE IOS
"Xdaal*
IODOMETRIC
S«qu«nci*l
Modified'
HA
DPD
OZONE
IODOMETRIC
ARSENIC BACK
TITRATION
FACTS
DPD
INDIGO
SpicC photo
Sp«ct«»'
MEASURED
DIRECTLY
cio,-
ClO,-
CiOj-
ClO,-
C10,"
0,
0,
DETECTION UORJCINC EXPECTED EXPECTED
LIMIT RANGE ACCURACY PRECISION SKILL*
(ag/L) (ag/L) it %) <± %) LEVEL
0,001
0.3
0.08
0.02
0.1
0.001
O.OOi
0.1
0.001 - 10 0.5
0.064 > 1
0.3 - 20
0.08 - 0.8 3.5
0.1
2 • 5
1 • 3
0.01 0.01 - 10 5 5
0.01 0.01 - 10 0.5 0.1
0.002 0.5 - 100 1 - 35 1 - 2
0.002 0.5-65 1-5 1-2
0.5-5 5-20 1-5
0.2-2 5-20 5
0.01 - .1
0.05 - .5
> 0.3
1 0.5
I 0.5
1 0.5
18
-------
TABLE I. CHARACTERISTICS (cont'd)
STABILITY •
REAGENT PRODUCTS
INTERFERENCES pH RANGE
FIELD CURRENT
TEST AUTOMATED STATUS
S YRS
> 1 DAY
NONE
Indapandanc
YES
YES
RECOMMENDED
good
SubJace eo
oxidation
Macal ions &
nicrica ion
7
SO
NO
.~RECOMMENDED AT
HICH CONC.
good
Subjact co
oxidation
Macal ions &
nicrica ion
2
NO
NO
CONT'D STUDY
I yaar
I day
Oxidizing
spacias
< i
NO
YES
USED AFTER ALL
cio, cio,- gc:;e
Solid
cable*
< 30 ain
Oxidizing
spacias
7
NO
NO
N'OT RECOMMENCED
5 YRS
> I DAY
NONE
Indapandane
YES
YES
RECOMMENDED
I YK
tubjacc co
oxidaeion .
All ozona
by produces
and oxidants
< 2
NO
NO
ABANDON
i m
subjset Co
oxidaeion
Oxidizing
spacias
1.8
NO
NO
CONT'D STUDY
2 YHS
no fading
firsc 3 ain
Oxidizing
spacias
6.6
NO
NO
NOT RECOMMENDED
Solid
seabia*
< 30 ain
Oxidizing
spacias
6.4
NO
NO
NOT RECOMMENDED
good
good
good
¦°od
good
good
CI,, Mn ions
lrs I,
CI,, Mn ions
Bfj 1^
CI,, Mn ions
St, I,
2
"2
2
NO
NO
NO
YES
YES
YES
RECOMMENDED
RECOMMENDED
RECOMMENDED
19
-------
TABLE I. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS* (cont'd)
TYPE or TEST
(METHOD)•
Spaeias* DETECTION WORKING EXPECTED EXPECTED
MEASURED LIMIT RANGE ACCURACY PRECISION SKILL*
DIRECTLY (ag/L) (ag/L). (t %) (± %) LEVEL
INDICO (eont'd)
Visual
0.1 0.01 • 0.1 5
> O.i 5
5
5
I
I
GDFIA
LCV
ACVK
0,
o»
0.03
0.005
0.25
0.03 - 0.4 I
othar rangaa
possibla
NR
0.05 • I
NR
NR
0.5
SR
:rs
o-TOLIDINE
1ISTERFYRIDINE
0,
Oi
NOT QUANTITATIVE NR I
0.004 0.05 • 20 2.7 2.1 3
CARMINE INDICO
< 0.5
NR
NR
SR
ELECTROCHEM
Aaparoaatrie
Aaparoaatrie
lodoaatric
Bar* alactroda
Maabrana alaec.
Diffarantial
Pulsa Dropping
Marcury
Diffarantial
Pulsa Polar*
ography
Poeanclomaerlc .
Tocal
oxidants
Total
Oxidants
Oi
- 1
- 0.5
0.2
0.062
NR
0.003
NR
NR
NR
NF
NF
NR
NR
NR
5
5
NR
NR
NR
5
5
NR
NR
NR
2
I
3
1
20
-------
TABLE X. CHARACTERISTICS (eonfd)
STABILITY
REACENT PRODUCTS
good food
good good
good good
Scabl* Seabls
MR MR
sr NR
Cood Cood
SR MR
none SA
1 YR Subject co
oxidation
Ron* HR
non« lift
non« HR
non« NR
non« NR
INTERFERENCES pK RANCE
FIELD CURRENT
TEST AUTOMATED STATUS
CI,, Mn ions 2
Br, I,
CI,, Hn iona 2
It, X,
CI, at > lag/L 2
S«- SO1" Cr** 2
Hn > I ag/L 2
CI, > 10 ag/L
Metal ion*. NO," 2
CI, • <7
YES 510 RECOMMENCED
YES SO RECOMMENDED
SO YES * COMPARISON
STUDIES
NEEDED
SO HO CONT'D STJSY
SO SO COST'D STUDY
YES SO ABANDON
SO YES RECOMMENDED
(LAB'TEST)
NR
SO
SO
CONT'D STJT:
Oxidizing
*p«ei«*
SO
YES
RELATIVE
KONITORINC
Oxidizing
sp«cl««
m
m
4 - 4.5
HR
HR
NO
SO MOT RECOMMENDED
HO YES
HO ?OSSISLE
COST'D STUDY
CONT'D STJCY
HR
NR
NO
NO
RESEARCH LAJ
NR
NR
4
NR
NO NO CONT'D STUDY
NO YES CONT'D STUDY
21
-------
TAIL£ X. CHARACTERISTICS AND COMPARISONS OF ANALYTICAL METHODS*
TYPE OF TEST
(METHOD)*
Species' DETECTION WORKING EXPECTED EXPECTED
MEASURED LIMIT RANGE ACCURACY PRECISION SJC
DIRECTLY (ag/L) (t %) (± %) L2*.
uv
0.02
> 0.02
0.51* 0.1
ISOTHERMAL
PRESSURE CHANGE
4 x 10'* 4 x 10*» - 10 0.5
0.3
0,
0»
OZONE GAS PHASE
'Ideal*
UV
Stripping
Absorpeion
Xodeaatry 0,
Cheailuninescence 0,
Gas phase titration 0,
0
Rhodaaina B/
Gallic Acid
Aaparoaetry
s
0>
1 1 - 50.000 1 1
0.5 0.5 - 30,000 2 2.5
0.002 0.5 • 100 1 - 35 *. - 2
0.003 0.005 -1 7 5
0.005 0.005 - 30 S 3.5
0.001
NR
NR
NR
NR
NR
1
1/2
2
i/:
* for page nuabers In the full report, refer to ehe Alphabetical Index
' direct determination of eha species aaaaurad without interferences
* Oparater Skill Lavala: 1 - ainiaal, 2 - good technician,
3 - axparianead cheaist
NA Not applicable
NR Noe raporead
NF Noe found
1 Using rasaareh grade alactrochaaieal equipaent
2 Using eoaaarcial cicraeor
3 Spectrophotometry andpoinc daeaecion
4 Visual endpoint daeaecion.
5 Using test kit
5 Liquid raagane is unseabla
7 Stablility is vary dependant on the purity of eha 2-propanol us«d
22
-------
TABLE I. CHARACTERISTICS (cont'd)
STABILITY .
reagent moDucTs
INTERFERENCES
pH RANGE
FIELD
TEST
CURRENT
AUTOMATED STATUS
nona
NA
Othar
Absorbar
Indapandant
NO
YES
ESTABLISH
HOUR ABSORB-
TIVITY
nona
good
nona
Indapandanc
NO
YES
COMPARISON
STUDY
nona
nona
nona
Indapandane
YES
YES
RECOMMENDED
nona
nona
nona
na
YES
YES
RECOMMENDED
good
good
SO, 310,
UK
YES
NO
ABANDON
seabla
< 1 sac
nona
NA
YES
YES
RECOMMENDED
itabla
seabla
nona
HA
YES
NO
NOT RECOMMENDED
problaaa
nr
NA
YES
POSSIBLE NOT RECOMMENDED
nona
nona
nr
NA
YES
YES
NOT RECOMMENCED
8 Total Chlorina it all chlorina spacias vlch +1 oxidation ataca
9 Vary lieela actual work has baan carriad out on aalaetlva daearainacion
of chloraalnaa. Tha valuas raporcad ara froa axerapolaead scudlaa that
had objactlvas ochar than tha salaetiva datarmination of chloraainea.
Most aathoda ara Indlraee procaduras which ara not racoaaandad
10 Indlraee aathod
11 1/3 ot CIO, datarainad
12 Acid ehroaa vlolae potasaiua (ACVK)
13 Ca* diffusion flov injaeeion analysis (GDFIA)
14 Baaad on currant molar abaorbtivicy and propar saapla handling taeniquas,
Curranc base aaciaaeas of aolar absorbeiviey of 2900-3300 giva a
possibla arror of > 101.
• Takan froa Gordon, Coopar, Rica, and Patay, AWUA-RF Rsvlaw on
"Disinfactanc Residual Haasuraaants Hachods" (1917)
23
-------
Chapter 4 (Indexed Reference Citations) has been included In ehis report in
order eo assise readers In locating particular papars of inearese. The 43
categories for ehlorine, chloraainei. and ehe oxy-chlorine species. along with
eha additional 60 categories for ozone. should taaka cha cask of finding in-
dividual papars of interest considerably lass cuabersoaa. Papars which describe
savaral aachods hava baan ineludad in aaeh of cha appropriaca caeagorias. All
together, cha 1,600 references cicad in Chapears 1-3 nuabar sora than 2.COO
individual eieaeions vhan distributed in eha indexed fora of Chapcar 4.
Chapter S is an alphabaeieal liseing of eha individual r«far«ncas eieaeions.
finally, a daeailad Index has baan ineludad in ordar eo assise readers in
locaeing subjaces of specific inearase. Va hopa eha raadars will find ehase
addieional chapears as useful as hava va in preparing this report.
RECQfflgPATIQWS
Canaral Statements on Comparisons.
There hava been and will continue to be rapores of aethods eesaarison. -ne
of tha aost important considerations for a aethod is accuracy, i.e. ;he ability
of eha aathod to determine tha correct concentration of a disinfectant in
solution. An equally iaportant consideracion is precision, i.e. how wall does
eha analyeical aeehod reproducibly aeasura cha saae conceneracion. Frequently
experiments ara conducted to deteraine the "equivalency" of the sechocs. Fraa
such results, aathods aay be found to be equivalent, but she only analytical
considerations tested were accuracy, as judged by a Referee Metr.od. ar.e
precision, judged for each aethod based on tha axperiaental design.
No considerations ware given to specificity or .analyst preference. Vet ore
of the aost difficult tasks in eha araa of disinfection analyeical aethods
development is coaparison casting. It is recoaaandad that a protocol be
developed to .inieiaea coaparison of eha disinfeceanes. This proeoeol should
inelude all of eha faccors dellnaaeed in cha 'Ideal Method" and should ba
undareakan in boch laboraeory controllad conditions and ac selected water
eraataent planes around cha counery.
Chlorine Chaaisery.
Clearly, eha conversion eo aolas, equivalents, or nonaliey from units of
ag/L (as Cl4) or ag/L (as oehar oxidanes) can easily be confused (and
confusing). Our racoaaendaeion is ehae ail oxidizing agenes be raporeed in molar
units (M) and, if necessary, in ag/L of that oxidizing agent as aeasured (i.e.
ag/L (as Cla) or ag/L (as C10#") or ag/L (as C10s"). Furtheraore. we recomnend
that oxidizing equivalents par aola of oxidant be reported to minimise
additional poeancial confusion. For exaapla, when C10t is reduced co CIO.",
ehis corresponds co one aquivalone/aole; on che ocher hand, whan CIO, is reduced
co CI", ehis corresponds to fiva equivalents/aola. A sunaary of eolecuUr
weights and oxidizing equivalents for the various ehlorina species, oxychiorir.e
species and ozone is given in Tabla II.
24
-------
TABLE XX. EQUIVALENT WEIGHTS FOR CALCULATING CONCENTRATIONS ON THE
BASIS OF MASS.
Speeias
Molecular
Weight
g/aol
Elecerons
Transferred
Equivalent
Weight
g/eq
Chlorine
70.906
2
35.453
Monochloraaina
51.476
¦2
25.738
Dichloraaine
85.921
4
21.480
Trichloraaine
120.366
6
20;061
Chlorine dioxide
67.452
I
67.452
Chlorine dioxide
67.452
S
13.490
Chlorite ion
67.452
4
16.863
Chlorate ion
83.451
6
13.909
Ozone
47.998
2
23.999
Ozone
47.998
6
8.000
Several aechanisas have bean proposed for cha daeonposlcion of dichloranine,
buc che coaplece aechanisa at cha breakpoint has hoc bean resolved. Clearly, che
eheaistry Is coaplicated and varies markedly vich solution conposition. A
decalled underscanding of cha specific reaecions involved requires a decailed
knowledge of cha coneencracion of all chloraaina species in che systea.
Nitrogen-containing organic coapounda say be prasenc in surface water and
ground-wacer. Because of analytical coaplexicies, very few decailed studies
have been undertaken Co de canine the individual coapounds present and che
concentration at which they exist. KJeldahl nitrogen analysis is used
frequently, but this does net provide any decailed inforaation wich regard to
individual coapounds. The area of organic nitrogen atld che dateraination of
specific coapounds in natural wacers is one of cha increasing interest and
requires considerably aore research in characterization and aathods development.
Ultraviolet Methods.
In general, because the aolar absorptivities are quite low for chlorine and
chleraaine species, ultraviolet aethods are not considered useful in routine
aonitoring of chlorine residuals. In addition to the low aolar absorptivities,
there is often background absorbance that nay interfere wich the aeasureaenc in
various natural waters. However, these aeasureaents are of use in standardising
the chlorine species in distilled wacers and are often used in •xperinencai work
25
-------
ralated to chlorine speelation. This aethod does have considerable pottr.ci*!
for tha determination of relatively high concentrations of halogens,
particularly in relatively clean water. this aethod aight fifld USS in
aonicoring ehlorina species in water craaeaane planes. However, with a sore
elaborate aultivavelength spaccrophoeoaacar and eoaputer-controlled spaecral
analysis, ic alght ba possibla co analyze several halogans siaulcanaously.
le is also possibla ehac additional aachods using paraaabla aeabranes could
ba davalopad for cha siaultanaoua dacarainacion of ehlorina species' in aqueous
soluclon. Additional work is nacassary in this araa. Although cha aoiar
absorptivisias of cha spaeias is not of a aagnituda as to land it to tha routine
dataraination of tha dilute (lass than 10"* M) ehlorina and chlorine«aaaonia
spaeias, it is potantially helpful in dataraining tha concantration of standard
solutions. Absorption spaetrophotoaatrie analysis, has and vill eontinua to ba
vary iaportant* in tha araa of ehlorina ehaaistry. It can ba usad in tha
unaabiguous dataraination of ralativaly high eoneantrations of tha spaeias in
ralativaly pura water.
Continuous Aaparoaetrlc Titration Mathod.
Interferences appaar to ba raduead using tha continuous aaperoeetric aethod
baeausa tha raagants ara addad to tha sasple Just prior to contacting the
indicating alactroda. Thus, vhan eoaparad to tha aaperoaetric titration, tha
aaount of intarfaranca by iodata ion, broaata ion, eoppar(II). iron(III). and
aanganasa
-------
Colorlaatric Methods.
Xe 1« reporeed in Standard Merhads (13) chac nicregan trichloride can ba
aeasured using eha DPD aechod: howavar. cha aechod has not baan eonflraed by
Independent investigations and should ba used only aa a qualitative aachod.
Additional raaaarch is necessary to deteraine cha effectiveness of tha D?3
aathod for nitrogan trichlorida. Tha effect of tha prasanea of aercurle
ehlorida in tha raagants for alnialzlng tha braakthrough of aonoehloraaina ir.co
tha fraa ehlorina raading with tha OPD aathod has baan shown. It is vary
laportant that tha addition of aarcuric ehlorida to tha buffar ba follovad to
alnialze tha diraet reaction of aonoehloraaina with DPD. This phanoaanon is not
thoroughly undarstood. This effect should ba studiad aora thoroughly and. the
prineipla aay ba applieabla to all of tha eoloriaatrie aathods.
Tha usa of thioaeecaalde vas evaluated for aonoehloraaina (using C?D>
Staadifae). It was shown undar thasa conditions to aliainata any posiciva
intafaranea in cha fraa rasidual aaasuraaant. Thasa rasults ara not as y«c
undarstood, bue tha iaplieation is chat tha ehaaistry of oxidation is different
for aonoehloraaina and fraa ehlorina. Thasa rasults suggast that aora work ij
naeassary to battar dafina cha raaecions involvad, and chis aay laad co a aora
usabla analytical proeadura. This proeadura is raeoaaandad for usa in waters
chac ara suspaccad co ba ralacivaly high in combined ehlorina.
. Tha DPD-Ethyl Acacaca Excraeclon Proeadura is a aodifieacion of cha C?D
eheaistry. Tha aachod is basad on cha oxidacion of iodida ion by aeciva
ehlorina followad by axcraecion oI cha iodina spaeias into achyl acacaca. This
proeadural aodifieacion aay ba of usa in cha dacarainacion of cocal residual
ehlorina in boch cha fiald and laboracory. Additional work is naeassary before
ic ean ba usad co any graac axcenc. Ic does noc appear co offer substantial
advantages co cha already wall cesced eoloriaecrie aechod for laboratory
aaasureaencs.
The DPD aechods have beeoae che aosc widely used procedures for che measure -
aenc of chlorine. This is noc likely co change. The DPD eolor reagenc, in
liquid fora, has been shown co be quite unstable and is noc reeoaaended for use.
Ic is sensitive co oxidacion by oxygen and chus requires a eoncrol measurement.
Clearly, ic is beccer co usa dry reagents.
Leuco Crystal Violet, LCV.
No scudles have been reporced chac exaalne che incerferenee of chlorine
dioxide and/or ozone in che LCV aechod. Ie is ancieipaced chac chase oxidants
would incerfere in che aechod, and scudies should be eondueced co quancify chase
potential incerferencs.
Syringaldazine; FACTS.
A study using syringaldazine in a continuous aechod co differentiate free
froa combine ehlerine has been reported. It was eoneluded that it eould be used
and was useful in controlling free chlorinacion. Furcher work would have co be
eondueced to use this or any eoloriaatrie aathod in continuous analyzers.
27
-------
Cheailusd.nescence.
Several papers have appeared that detail Che reaction of hydrogen peroxiie
and hypochlorous aeid and the reauleing cheailuaineseenee. lhe aechanisa has
been relatively wall established and che cheailuaineseenee is thought to occur
as a result of the formation of singlet oxygen. The light enitted is red <635
nn). and occurs aost readily in alkaline solution. This reaction is rather
insensitive to lov concentrations and is not suitable for the determination of
hypochlorous acid in aqueous solution. However, the studies that have been
reported ean serve as a guide for those interested in pursuing other aethods for
the determination of hypochlorous acid by cheailuaineseenee. It is not sensitive
enough to be considered as an analytical aethod for chlorine in water treatment.
A study has been reported that details the use of luainol for the
aeasureaene of hypochlorite ion. The optimum pH for analysis was between 9.0 anc
11.0 Luainol also has baen used for the determination of hydrogen peroxide.
4,5,6,7,-tetraaethoxyluainol is 30 t aore sensitive than luainol. Sither of
these eoopounds aay be aore sensitive in the determination of free chlorine. As
these coapounds have not been tried it appears that additional studies are
necessary. Froa the Halted data available, it appears that this reaction has
considerable proaise as an analytical aethod. It aay very well be the zosz
sensitive aethod to date.
It is reported chat lophine, in a reaction with hypochlorite ion. produces
light. Very few details ware given in the study for this reaction. It appears
that lophine also aay be good as. a cheailuaineseenee reaction systea for free
chlorine. Additional work should be undertaken to better characterise che
details of this reaction.
Luainol and soaa of its derivatives, or lophine, aay be well suited far the
very sensitive aeasuraaents of chlorine species. Additional research should be
undertaken to develop the use of cheailuaineseenee for use in the determination
of chlorine in water. The potential exists for rapid, sisple. ar.d specific
aethods for chlorine and possibly other oxidants. tilth the advent of fiber
optie sensors and their application in cheailuaineseenee oethoas, this
technology will ba important in the future.
Fluorescence.
The use of rhodaaine S has been reported as a lov level fluoroaetrie method
for the determination of bromine. This aethod is qualitatively specific for
bromine, although chlorine will react to decrease the fluorescence. The advant-
age of this aethod is that it is capable of determining oxidants at very low
concentrations. This aethod could ba applied to chlorine analysis by first
using the froa chlorine to exidiie tha broaide ion to broaine, an irreversible
reaction, followed by tha determination of broaina. This oethod was not
developed fully and vary litcla work has been undertaken since the first
publication. It does appear to have considerable potential apd future research
in tha area of aethods development should not exclude additional work on this
fluoroaetrie procedure.
28
-------
Other Electroda Method*.
Additional studies art required to becter understand che liaicacions of
aeabrana eleccrode aethods. Xc appears chat thav aav hava proainant roles to
play in ehlorina residual aeasureaents in the future.
In a series of experiaents carried out for the determination of f:i«
chlorine in cap vacer, it was observed chac chare was a statistically
significant difference becveen the results of ehe anperoaetric titration and the
aeabrana electrodes. Zc was thought co be a problea in che aeabrane eleccrodes.
However, on reeonsideraclon. it is possible chac the electrodes were actually
living a free chlorine reading and the aaperoaecric titration was reading the
sua of free and organically coabined chlorine. The study was conducted on water
which is relatively high in organic nitrogen. It is possible chac considerable
chlorine is present as organically coabined chlorine and -Interferes in the
aoperoaetrie titration procedure, bue does noe Interfere with the aeabrane
electrode aeasureaents. This question suae ba resolved. Carefully designed
experiaents co explcltly resolve these differences would be aosc appropriate.
There have been no reports of experiaents using bare-electrode anperoaetric
analyzers where other oxidants such as chlorine dioxide, chlorite ion, chlorate
ion or ozone have been tested with the bare-electrode. Additional studies are
required to expand these bare-electrode aaperoaecric studies to quancitate
Interferences with oxidants other than those tested, and to expand to other
natural waters.
Since the accuracy of the potentioaetric electrodes is affected, if
teaperature corrections are not used, it is recommended that temperature se
either controlled or aeasured simultaneously. Additional Independent aeasure-
aents of accuracy should be undertaken for the potentioaetric electrodes.
It appears that the potentioaetric electrode can be used for the
determination of total residual oxidant. It is suitable for continuous
aeasureaents and appears to give results that are acceptable when compared ::
the aaperoaecric cicracor.
Ceneral Suaaary and Recoaaendacions for Chlorine.
In comparing all of the aethods to the "Ideal Method" we find that none cone
very close to our ideal standard. Continued developaent of che various aethods
will, however, coae closer and closer to the ideal.
For the present, the aoperoaetrie titration techniques will remain the
laboratory standard used for the basis of eoaparisons of aceuraey. These
aethods, with proper precautions ean differentiate between the conaon inorganic
ehorine/chlorine aaaonla species, and in general suffer froa as few inter-
ferences as any of the aethods.
Of the three eoaaon coloriaetric procedures, OPD, LCV, and FACTS, the OPD is
by far che nose commonly used aechod. Froa the available licarature it is clear
thac che DPD procedure has a number of weaknesses. In particular, the colored
product is a free radical whieh Halts the stability of the colored reaction
product. The direct reaction with aonochloraaine, co fora a product identical
29
-------
co eha reaecion wieh free ehlorina, la also * drawback. This problea car. se
reduced by cha addleion of ehioacecaaida. Liquid reagent instability produces
chair use in «oit e«i«i; eiri should be taken eo daeeraine blanks frequently.
His present l£V aathod thae appears in se»nd»Fd *«ehod» (13) is outdated and
has been substantially laproved upon by Whietle and Lapteff (14). This aeti-.od
allows Cor the differentiation of cha eoaaon free and eoabined inorjanie
ehlorina spaeias. Kovavar, baeausa only ona comparison scudy has ";«en
conducted, additional collaboraciva eescing is recoaaended.
Tha FACTS case procedure appaars eo ba vary usaful for cha deeerainacion of
fraa ehlorina in cha prasanea of ralacivaly high eoneancracions of coabir.ed in-
organic ehlorina. A savara drawback of cha FACTS ease procedure is cha insoiu-
biliey of cha syringeIdaslne in aichar 2-propanol or wacar. This laads eo di£-
fieuleias in raagane praparaeion, and prasuaably eo cha eolor scabiliey problem
aneounearad ac tha higher eoneancracions of ehlorina (greacer ehan 6 - I ai/l
(as CI,)). Alchough a aachod for cha usa of cha FACTS case for cocal chlorine
has baan raporcad, it should ba eascad furthar.
Eleccrode aachods hava baan davalopad aaploying savara1 different concepts.
Tha aeabrane alaecrodas appaar co hava poeaneial as spaeifie aachods Cor hv?o-
ehlorous aeid. Coaaon inearfaraneas ara ochar nonionizad aoiaeulas such as
ehlorina dioxide and ozone. Pocaneioaeerie alaecrodas for cha dacarainaclon of
Cocal ehlorina ara laproving in boch dacaecion liaie and scabiliey. These
alaecrodas appaar co hava proaisa in cha araa of proeass conerol. Thai;
inclusion as aachods for roueina-use in eha laboracory and fiald is warranted.
Bo eh fluoraseanea and ehaailuainaseanea aachods also show proaisa far
spaeifie dacarainacion of fraa ehlorina ae vary low eoneancracions. Within
araa of spaecrofluoroaaerie aachods, chara is eonsidarabla work vac co ba
iniciacad. Coneinuad davalopaanc work is warrancad and raeoaaandad in zr.is
proaising araa.
Froa eha raviaw of analyeieal proeaduras for eha dacarainacion of ehlorir.a
in aqueous solution, ic is raadily apparanc chae only a faw of cha aachods ara
usad roucinaly. Navarchalass. chara is eareain eo ba a continued incarase in
davaloping naw and baccar aaehoda of analysis. Va would serongly receaaand chac
nav aaehods ba presented in earns of eha "Ideal Method" and chac whenever pos-
sible. eoaparisona wich real saaples and inter laboratory eoaparisons ba aada.
Flow injection analytical taehnlques ara becoming vary eoaaon. Coneinuad
davalopaanc should laad eo the automation of aany eoloriaacrie and fluoronecrlc
analyeieal aathods for tha aaasuraaant of fraa and eoabinad ehlorina and its
various spaeiaa in water. With tha eurrant emphasis on autoaatlon. eha aachods
chac are ee ba developed and ehosa already developed can raadily exceed present
standards of aeeuraey and precision. Aucoaaeion will also laad co operator
independent aethoda and should lead to iaproveaents in process control and
aonicoring.
Chlorine Analytical Method* Coaparaeiva Studies.
Tha reader is caucloned against aeeepcing eha resules of any or ail o£ cha
abova cases wtchoue soae reservations. Where possible wa hava cried co add ccrz-
30
-------
¦•net. parenthetically, based upon eur knowledge of the field. It is very la-
porcant in reviewing data froa comparison cases chat che analyse be awsre of the
objectives of cha coaparison tasting. For example. a case aay be judged
unaecapeabla bacausa of an unacceptable lovar limit of dacaecion chac is bavond
cha naad for eoncarn for ochar investigators.
In ganaral vhan casting savaral cast proeaduras it is laporcanc co identify
cha objective of cha casting. Equally important is cha usa of cha daca. In
raporting tha results of cha above tests, ic should be kept in aind that many
manufacturers of cheaieals for analytical aechods and Test Kits change their
procedures as a resulc of the testing. The concerned analyst needs co decersine
if che results are still valid. This change is not necessarily applicable to
ocher studies where che chealsczy of an analycieal aethod is exaained. In
general, the aore the cesc studies cheaisczy and noc aerely che .pesc procedures,
che aore applicable che results are for future reference.
Another area of confusion concerns precision and accuracy. An analytical
aethod aay be judged acceptable based on the precision of the results, while the
saae aethod aay give poor accuracy. These statistical paraaeters are separate
and Bust be tested using different experimental designs. Coaparisons with the
"Ideal Method* would require that both be at acceptable levels.
In general, there is a lack of cooprehensive studies co better understand
the cheaistry associated vich che individual cesc procedures. Invescigacions of
this nature are necessary on a continuing basis, because of the advances in ana-
lytical instrumentation and our-continued improvements in understanding the de-
tails of che underlying cheaiscry.
Chlorine Dioxide Analycieal Methods.
The iodoaetric aethod is e questionable aethod even for carefully controlled
research laboratory chlorine dioxide standards. In real samples where a large
number of potential interferences can exist, the aethod is destined to produce
erroneous results. Newer, aore species specific aethods are better choices.
Any aethod which determines concentrations by difference is potentially
inaccurate and subject co large accumulative errors--both in terns of accuracy
and precision. The subcracclon of two large nuabers co produce a saall number
aeans chat che errors assoeiaced with chose large numbers are propagated to the
small number. The resulc in aany cases is chat the error is larger than the
saaller number, therefore, giving aeaningless information. Methods such as
this, whieh obtain values by differences, should be avoided.
The DPD aechod uses che difference aechod in che evaluation of concen-
trations. The direet aeasureaent of species by aeans of a aore reliable and
accurate aechod co deceraine chlorine dioxide is needed. The saae questions
raised about che 0P0 aechod for chlorine also apply here.
Ultraviolet spectrophotoaetry, utilizing continuous flow automated methods,
has a great potential for accurate and precise aeasurements with che added
advantage of ease of operation and high sample throughput. Flow injection
analysis aethods (FIA) should be carefully evaluated against existing methods
for accuracy and precision. The aethod should be field tested and the potential
31
-------
problea of aeabrane reliability should be evaluated for long cera operations.
Additional bench studies using continuous flow aethods wlch cheailuainescer.s
detection wise -be carried ouc. The suparior selectivity of this aaehod needs so
ba utilitad. Coaparison lab tsseing and fiald study should ba carriad out.
Chlorlee/Chlorate Ion Analytical Method*.
The lodoaetrle/eaperoaecrie aethods arc indirect daterminations of chlorite
Ion and cannot ba recoaaended. The DPO aachod for chlorita ion can noc ba
recoaaended bacausa ie is unraliabla.
Tha lodoaetric sequential aathods appaar to bo vary workable on saaples
containing graaear than 1 ag/L chlorita ion or chlorata ion with good precision
and accuracy resulting. Tha aathod requires considarabla operator skill and
axparianca to obcain good pracision and accuracy for saaplas containing lass
than 1 ag/L chlorita ion or chloraca ion. Tha aaehod should ba fiald tested
with oehar aathods using both high and low racios of chloraes ion co chlorite
ion. Tha aathod should bo usad with caution on low level staples of drinking
water and/or wastewater. although dirace aathods requiring lass specialized
skills ara prafasrad.
Intsrlaboratory coaparison* should b« carried out for tha aodified
iodoaatric aaehod for eha dirace analysis of chlorita ion and chlorata ion. Tha
dataHad affacts of various potantial intsrfaroncas naad to ba evaluated.
Tha srgentoaetric titration aaehod is to ba racoaaandad only for relatively
high concentrations of oxy-chlorine spaciss (10-100 ag/L) but aay ba very usecui
in establishing inter*laboratory bench nark coaparison* at these high concen-
tration ranges. No such eoaparisons ara currently available.
A highly precise, autoaated FIA aathod for low level chlorate ion needs to
be developed possibly using various aasking agents such as glycine, oxalic acid,
salonie acid, and nitrite ion eo initially reaove other possible oxy-halo$en
interfering species. Tha aathod appears to be very proaising in that it can ba
used to directly determine low leval chlorate ion concentrations.
Difficulties Vlth Ozena Measurements: Need For Ideal Method.
As a consequence of the nature of ozone, its continuous self-decomposition,
volatility froa solution, and tha reaceion of otone and its decoaposition
products with aany organic and inorganic eontaalnants in water, the deter-
aination of dissolved residual ozone is very difficult. A detailed knowledge of
the aechanisa of aqueous otone decoaposition and the potential role of the
various highly reactive intemediates, is iaparative in order to accurately
evaluate the analytical aeehods CIS). In this context it should be noted that
aost ozone aethods are aodifications of chlorine residual aethods which
determine total oxidants in the solution. Therefore, ozone decoaposition
products such as hydrogen peroxide and the like are also aaasured.
Iodoaetry can be used as an example of the difficulties encountered in
asking aqueous ozone neasursments (16). Iodide ion is oxidized to iodine by
ozone in an unbuffered potasslua iodide solution. The pH then is adjusted co :
32
-------
with sulfuric acid and eh* liberated iodina is titrated with sodiua ehiosulfsra
to a search and point, The osone/iodine steichioaetry for this raaeeion has bean
found co range froa 0.63 co l.S. Faccors affaccing eha aeoichioaecry include:
pH. buffar eoaposition, buffar concentration. iodlda ion concentration, ssapling
techniques, and reaction tia*. The pH during eha itiiciaL ozone/iodide ion
reaction and eha pH during eha iodine daearainaelon have been shown eo aarkedly
alear eha ozone/iodine scoiehioaecry. Tha foraation of lodaea ion and hydro;en
paroxida hava been iapiieaced spaclfieally as factors affaccing eha ozona/iodine
stoichiomatry (17). Modifications in eha iodina deceraination include changas
in and poine dacaeeion, pH. and back-deration techniques. Ilona of ehaaa
aodificaeions has baan daaonseracad eo ba totally satisfactory.
Tha biggasc difficulty in inesrpracing eha axlscing ozena liearacura is ehac
no ona aaehod has baan accepted as eha Rafaraa Method. Therefor*. comparison
baevaan savaral diffarant aathods ean eraaca falsa conclusions abouc tha
accuracy of eha aethods. Tha aaehod aosc ofcan used for eoaparaciva purposes in
tha research laboratory is UV aeasureaent of ozona at 260 na. Ivan with this
aaehod ehara* is apparanc confusion ovar eha aolar absorptivity for aqueous
ozona, with tha valuas ranging froa 2900 eo 3600 M*1 ca*-.1 (16).
All analytical aathods raporcad, particularly ehosa of aarly vir.taja, should
ba reevaluated, considering tha racane information about oxidativa by-products
froa ozona dacompoaition and eha ozonation procasa itsalf. Sons of ehasa
* faccors may not hava baan considarad during davalopaane of :ha original
analytical proeaduras, Cartainly, mora daeailad information and coaparisor.s
should bo availabla. Baeausa of eh* difficultias of astablishing a raliabla
Rafaraa Method wa proposa ehae eha axiseing and fueura aachods ba coaparad
againse an "Ideal Mathod". This "Ideal Method" would ineorporaea all of the
charactariselcs chat ara desired for an ozona aathod, eaking into account all
. other pocancial intarfarancas, daeoaposicion products, and saaplas originating
froa various sources. Finally, autoaaeion, while not an absolute necessity, can
add co tha selaeeivity and idaal nature of a aathod for ozone determination.
Ozone Keasureaant: Caa Phase.
Die aany uses of ozonation in the treatment of drinking water ara controlled
by aonitoring a nuab*r of parameters. Oissolvad residual ozone is only ona of
these paraaaeers, and its aeasuraaant controls only disinfection conducted after
fileraeion, but before addleion of a residual disinfectant for the distribution
syseea. However, ie is vary clear ehae eh* cost, efficiency, safety and
iaproveaenta in design of ozona watar purification systems is extremely
dependant on eha accurate determination of gas phaaa ozone. Therefore,
analytical aathods Bust be developed thae will accurately aeasura ozena in the
gas phase and residual ozon* in eh* aquaous phase. At this point it is
unrealistic eo believe ehat one single aathod will b* aecapeabla for both sample
matrices.
lodoaetry, UV absorption and chemiluninascene* are the three nose common
aechods enployed for gas phaaa aeasuraaenes (16). Each of chase has been applied
to determine che aaount of ozona present in generator axit gases, when stripped
froa solution to the gas phase, or tha aaount of ozone in a coneaceor exhaust
gas.
33
-------
These techniques of aonicoring concentrations in contactor exhaust {itn are
quiet promtsinj n a aethod of controlling the produceion of adequate quantities
of ozon*. This provides considerable savings in *ieetrieal energy costs for
ozone generaeion. Direct inter-coaparisons of eh* various gas phasa aeasureeent
techniques are needed in order eo evaluate accuracy.
Determination of stripped ozone in eha gasaous scaca vas raporead in the
liSb Edition of standard Methods (13) for measuring osona dissolved in waur.
However, in addition co eha procadura baing subject eo eha saaa limitations of
UV absorption and chsailuainssctnes procedures in aqueous solution, sh* affects
of eha gas stripping procass ieself oust ba takan ineo considaration.
Although tha iodoaetric stripping/aquaous absorption aathod has been
approvad in Standard Methods (13), va quastion eha accuracy of tha aethod. All
avidanca would suggase ehae eh* aathod is problaaaeie. Evan chough tha
iapuritias ara subseantially left bahind by eha seripping, eha aceuai procadura
and eha continual dacoaposieion of ozone doas Introduce inaccuracies ineo this
aathod. This aathod can b* us*d as a ralaciva aaasura of ozona for control
purposes.
This basie stripping approach followed by absorption in aqueous solution
(and coioriaetric measurement) aay deserve eo be studied furcher. However, the
biggest potential problea appears eo ba that at high concentrations of osona the
coioriaetric compounds aay reaet by a aechanisa different froa that used for
residual ozona aeasureaents. Research should be concentrated on the reager.ts
that have already exhibited ozona selectivity.
lodoaetry (Aqueous Phase).
If the performance of ozone in a specific treataent application is not de-
pendent only on the ozone, but is instead a collective function of its reactive
decomposition products as well, then iodoaetry can give a representative and
reproducible reading of the total oxidants. For example, aost European drinking
water treatment plants employing ozonation as eha primary disinfectant, have
relied on iodomeeric measurements as the basis for insuring adequate
disinfection, attaining a residual "ozona" level of 0.4 ag/L in the first
contact chaabar and maintaing this lavel for at least four ainutes),
Howavar, it is now abundantly clear that th* 0.4 ag/L value is a measure of
the aaount of coeal oxidants present, and not necessarily ozone alone.
Therefore, either eh* absolute l*v*l of ozon* required co aecain eha expected
degree of disinfeeeion is lower ehan 0.4 ag/L over th* required period of cise.
or soae of eha dacoapos it ion/oxidation products foraed upon ozonation also have
disinfecting properties, or both. Clearly, detailed experiments need to be
carried out to demonstrate tha efficacy of disinfection by the decomposition
produces of ozone. Similar efficacy data for ozona dacoaposieion products could
b* developed for oehar uses of ozona (e.g., cheaical oxidation) when measurement
of residual ozone levals must ba aada eo control the process. Such data would
help to juscify the continued use of lodoaetry to aeasur* "total oxidants",
rather than only ozone.
Historically, lodometry has been used as tha reference method for deter-
mining ozone, . and against which other analytical procedures have been
34
-------
"standardized". Ic is now quit# clear that because of its lack of selectivity,
eh* us* of iodoaetry should b* Halted co that of only a control procedure. In
cans of ozonation processes, measureaent-for control purposes-of the production
rata of oxona generators and bactarial diainfaction/viral inactlvation aay be
basad upon iodoaetry, providad tha usar recognizee tha aany liaitations of tha
aathod. Tha reevaluation of this aathod auat ba carriad out with tha specific
goal being to defihe whan tha aathod ia raliabla and tha aituationa whara it is
not accurata.
Many authors hava tactfully pointad out tha aany disadvantages of iodoaetry.
laaving it to tha raadar to dacida whether or not to usa tha procadura. In a
datailad coaparison of aight analytical aathods for .the deteraination of
raaidual ozone it was concluded (16):
"No iodoaatric aathod is racoaaandad for tha dataraination
of ozone in aqueous solution bacauaa of the unreliability
of the aethod and because of the difficulty of the coa-
pariaon of reaults obtained with ainor aodificationa in
the iodoaetric aethod itself."
Arsenic(IZZ) Direct Oxidation.
In the, direct oxidation of arsenic(IZZ). ozone reacts with inorganic
arsenic(XXX) at pH 4-7, tha pH is adjusted to 6.5*7 and the excess arsenic(Z:i)
species is back-titrated with atandard iodine to a starch end point. Values for
residual ozone deterained by the.araenic direct oxidation aethod and by the
indigo aethod agreed within 6% of the UV values. The priaary advantages of the
arsenic direct oxidation procedure are ainiaal interferences, good precision in
the hands of experienced operators, and apparently good overall accuracy. This
procedure continues to be recoaaended along with the indigo aethod. Additional
coaparisons of this aethod ahould be Bade with the Indigo aethod under various
conditions.
Syrlngaldazine, FACTS.
The FACTS procedure, which was developed for the selective determination of
free available chlorine (hypochlorous acid + hypochlorite ion) in the presence
of coabined chlorine (chloraaines), has been adapted for the deteraination of
residual ozone (19). In this procedure, an aqueous solution of ozone is added
to a solution of potassiua iodide, and the liberated iodine ia added to a 2-
propanol solution of syrlngaldazine at pH 6.6. The resulting color is aeasured
spectrophotoaetrically at 530 na.
The FACTS procedure has the aajor advantage of providing a spectrophoto-
aetric procedure for the deteraination of ozone. However, the aajor liaitations
of the FACTS aethod are still those of the iodoaetric procedure. Due to the
observed changes in slope and intercept which are probleas caused .by the
interferences, self•decoaposition of ozone, and stoichioaetry, this method could
be reviewed in order to fully evaluate its potential usefulness. However,
considering the other coloriaetric methods that are available further
development of the FACTS aethod does not seea to give any promise of the
iaproved selectivity that is needed.
35
-------
N,t(-Dlethyl*p>phenylenedlaaine, DPD.
The DPD procedure is based en the ozone oxidation of iodide ion present in
excess phosphate buffer at pH 6.4 eo produce iodine, which then oxidizes che DPD
cation to a pink tfurster cation which it measured spectrophotoaetrically, or
titrated. . The incerferencef include all oxidants capable of oxidizing iodide
ion to iodine, including ozone decoaposition products, halogensand aanganese
oxides (20).
One advantage of the DPD aethod is that detetainations can be aade by
ferrous aaaoniua sulfate (FAS) titriaetry, spectrophotoaecrically or by a color
coaparator. Ozone concentrations of less than or equal to 2 ag/L ean be
deterained coloriaetrically. Clearly, the procedure requires the difference of
differences and is liaited by the saae faetors which liait iodoaetry, specific-
ally the presence of aaceriais which can oxidize iodide ion to iodine. v
Although evaluation of this procedure versus the standard ultraviolet and
indigo procedures would seea to be necessary eo aaka a aore educated decision
about the continued use or abandonaent of this aethod. the recoaaendation is
that other coloriaetric aethods are considerably aore reliable than D?0.
Therefore developaent or testing is neither recoaaended nor considered necessary
at this tiae.
Indigo Trisulfonace.
The indigo aethod is subject to fewer interferences ehan most coloriaecric
aethods and fewer interferences than all iodometric procedures (21 • 23). Ac ?M
2, chlorite, chlorate, and perchlorate ions, and hydrogen peroxide do r.oc
decolorize Indigo Reagent when observed within a few hours and when the
concentrations of the interferents are within a factor of 10 of that of che
ozone to be deterained.
Ozone decoaposition products and the products of ozonolysis of organic
solutes do noc appear to interfere. However, chlorine, broaine, and iodine do
cause soae interference, as do the oxidized foras of manganese. The addition of
aalonic acid to the saaples will aask the interference of chlorine.
For the Indigo Trisulfonate Method, it should be noted that uhen che
ultraviolet absorption aethod is used to standardize the indigo aethod (or
aethod) for ozone, the choice of aolar absorptivity is very critical. It is
recoaaended that the equations of Hoigne continue to be used since they are
based on a aolar absorptivity of 2950 IT'ca"4. If and when a different value
for aolar absorptivity is reported and confirmed, the (calibration) equations
would hav* to be appropriately changed. In this aanner. all current
measureaents using the indigo aethod would continue to be coaparable.
The advantages of the indigo procedure is that it is based on a measure of
discoloration which is rapid and stoichioaetric. This analytical procedure is
recoaaended for use over any other procedure for the determination of residual
ozone. Its priaary attributes are its sensitivity, selectivity, accuracy,
precision, speed, and simplicity of operation.
36
-------
The gas diffusion flew Injecclon analysis (CD-FIA) proeedure eliainaces cha
incarfaranea of oxidized fonts of aanganase, and aarkadly radueas cha incerfer-
anea of ehlorlna (24). Ochar than Incarfaranea of chlorlna which can ba raducad
co zaro by addleion of aalonic acid, chara ara no known Incarfaraneas co cha
determination of ozone by this CD-FIA proeadura using eha Indigo aachod.
Tha prlaary advantages of cha CD*FIA proeadura ara lea. accuracy,
salacclvicy, laek of Incarfaraneas, raproduelblllcy, and rapldlcy. Thus, cha
aachod la wall sulcad for laboracory rasaarch scudlas and for usa as an
aueoaacad analycleal proeadura.
Mora seudlas should ba eonduecad with spaelfle gas-permeable maabrants,
parcleularly with raspaec co rapaacad and/or eonclnuous exposur# co ozona solu-
clons. Tha usa of FIA aqulpaenc In a proeass eoncrol environment.also ausc ba
avaluacad. Tha CD-FIA Indigo proeadura alghc wall ba adopcad as cha analycleal
aachod of eholea.
o-Tolidine
Tha o-colldlne aachod (addleion of 1-2 drops of o»colldlne solucion to
orone-containing wacar co davalop cha yallow eolor) is vary siapla, and easily
adapcad co flald eolor eoaparacors, sulcabla for unsklllad analyses. However,
chls advancaga eannoc eoapansaca for cha lack of quancicacion of eha aachod, r.or
for cha earelnoganlcley of cha raaganc (o-colidina). Tha recommendation is co
abandon chls aachod.
Caralna Indigo.
Tha canine indigo proeadura has baan usad in Canadian vacar works planes
for cha pasc 15 yaars. Tha ozona eoncalning wacar is derated with a solucion
of earaina indigo uncil a fainc blua eolor parslscs indieacing chac all of eha
ozona has baan dascroyad. Spaelfle incarfaraneas ara unknown, buc any oxidanc
capable of daeolorlzing cha earaina Indigo dya aosc likaly will incarfara.
Effaces of incarfarancs should ba dacarainad, as should praeision. accuracy,
and affaccs of raaganc scoraga and pH. Tha aachod should ba scudiad in direct
coaparlson with ochar aachods, such as cha indigo and UV absorption machods.
Automation of chls aachod could laad co iaprovad salacclvicy for ozona.
Aoparoaacry.
Vleh bara alaccroda aaparoaacars, aichar cha solucion qv cha electrode is
rocacad co ascablish a diffusion layar, and cha alaecrleal curranc aaasurad is
diraccly proporeional Co cha concancraelon of dissolvad oxidanc (25). - Commer-
cial aaparoaacrle analyzars giva saclsfaecory rasulcs provided ehara is no
oxidanc ochar chan ozona praaanc in cha saapla. In aany situations thay provide
adaquata monitoring of eocal oxidanc. Tha bara alaccroda syscaa has good
sansicivlcy,. and is applicabla as a eonclnuous nonsalaeciva aonicor for ozone.
When ochar oxidants such as chlorlna, chlorlna dioxlda, bromine, and iodine are
present, che cachnlque has diffleulcias. Tha axaec nature and magnitude of
these interferences requires addicional research.
37
-------
Out CO ehe accumulation of surface impurities at eh* electrode surfaces, all
bar* aaperoaetric *l*cerod* systaas ar* subjace to loss of sensitivity with us*.
With uncovered electrode surfaeas, fouling has been observed eo b* a signifieant
problea as was eh* cas* in earlier eeses with oxygen electrodes. Additionally»
the response is influenced by nuaeroua surface*active agents and also halogens
and oxygen.
An iaproveaene in the developaene of aaperoaetric aechods for ozone analysis
has been the application of gas-peraeable aeabranes for increasing selectivity
and preventing electrode fouling (26-27). Th*s* Teflon aaabran* *l*ctrod*i
exhibie less ehan 2% interference (in teras of current response) froa broaine,
hypobroaous acid, chlorine dioxide, hydrogen peroxide, nicrogen erichloride, anc
hypochlorous acid (26-27).
this type of aaperoaetric aeabrane sensor needs to be developed furthe:
based on the exhibieed seleccivieles. The aost disturbing aeeribuce is sh<
eeaperaeure dependence. If differene aeabranes could aaineain selectivity whili
ainiaizing the teaperaeure effect, this type of sensor could becoae highh
recoaaended.
The application of positive voltage potentials and the use of polyserie sea-
branes that are selectively peraeable to gases has enhanced the opportunity fot
selective aeasureaene of ozone. This is a vary significant iaproveaent ovai
bare aaperoaetric electrodes as well as aost older coloriaetric/speetrophoco-
ne trie and titriaetric aeehods. With an applied voltage of +0.6 V (v« SCE) a:
Che cathode, only the aose -powerful oxidizing agents can overeoae :hi
"resistance" of this anodie volcage and cause electron flow cathodieally chrougr
the electrocheaical circuit. This general approach should continue zo oe usee
in future electrocheaical developaents.
Other Electrocheaical Methods.
In the differential pulse polarography procedure (DP?), a predetermines
aaount of phenylarsine oxide (PAO) is added in excess to an ozone solution zc
reduce the levels of dissolved ozone present. Excess PAO then is measures
quantitatively by pulse polarography. The DPP aethod aay under son*
circumstances be useful in ehe research laboratory. The prospects of its us* if
the plant or field are not as proaising since a higher degree of operator skill
is required.
Poteneioaatry Involves cha cathodic reduction of dissolved ozone. The
diffusion*Halting current aeasured is proportional eo the concentration of
ozone in cha water. FUreher evaluation of potentioaetric systeas say be ir
order. Hevavar, ehe fundaaental probleas of electrode fouling must be
addressed. Perhaps a coabination of aeabranes and potentioaetric deteccior
would produce . a proaising systea for ozone deterainations. The systea appears
to have aodest poeential for devalopaent.
Uleraviolee Kaasuraaanes.
Ultraviolet absorption aeasureaents also can be used for residual aqueous
ozone ae 258-260 na. There is uncertainty with respect to the roUt
absorptivity for aqueous ozone. In ehe literature, values ranging froa 2900 zs
38
-------
3600 M*1 ca*1 are reported. This uncertainty In cha solar absorptivity is
critical to eh* future us* and calibration uses of th* UV aethods. Clearly,
further work eo v*rify this valu* it strongly recoaaended.
If tha aolar absorptivity for ozone is known unaabigiously, UV absorption i«
in principle an absolute aethod for the determination of ozone, which is r.oe
dependent upon calibration er standardization against ether analytical aethods.
Thar*for*, it can b* usad for calibration of other analytical naehods for osona.
It is specific to cha dataralnation of ozona, and is applicable to aeaeureaent
in gaseous and aqueous phases.
Physical Methods.
The caloriaecrie aeehod is based on eh* enthalpy af the catalyzed
deeoaposition of ozone (AH - 144.41 KJ/aole). The ealoriaatric deteraination
of ozone is calibration-independent. The technique is specific to she
determination of aolecular osona, but is applicable eo aeasureaent only in the
gas phase. However, th* higher the concentration of ozone in the gas phase, the
aoxe accurate the aethod appears to be, since a greater teaperature difference
is observed. Potential interferents hav* not been reported.
The aethod has been shown eo agree with iodoaetrie and UV absorption pro-
. cedures, particularly for the aeasureaent of ozone in ehe gists exiting ozone
generators. Therefore, ehe procedure can be used to Monitor ' applied ozone
dosages. Additional dacailad interlaboratory coaparisons need to be carried
out.
The isotheraal differential pressure procedure is based on ehe generation o
an increased nuaber of gas aolacules during ehe UV destruction of ozone at
constant teaperature. When this reaction is carried out isothereally in a
closed vessel, ehe increase in pressure of the concainad gas is proportional to
th* ozone concentration. In principle, this procedure achieves a totally
physical ozone aeasureaent without requiring calibration using a cheaieal
aeehod. Various aueoaaced instruaental checks such as the stored solar
absorptivity, tha age of th* UV light source, the zero poine reading,
aeasureaent of th* flow of th* test gas and the flushing gas, and the reading of
th* diagnostic display are possible.
No specific coaperisons are reported. How*v*r, in principle it appears that
this physical aethod is the besc candidate for calibrating the gas phase ozone
inscruaents currently being used for ozonation conerol. As long as pure oxygen
is used for ozone generation this aethod would be free of interferences and
would b* aubjece only eo serlce teaparatur* control of the aeasuremene cell.
Further study of ehis syseea would be necessary before it could be recommended
for further consideration.
General Sanaary end Recoaaendations for Ozone.
In coaparing all th* aethods to the "Ideal Method" we find that none com®
close to our ideal standard. Continued developaent of the various selective
aethods will, however, coae closer and closer to the ideal.
39
-------
In Mill of |u phase nuuriunti. none of the existing aethods can b«
recoaoended for accurate determinations of ozone. if a ralaelva valua of tha
oxona concentration is naadad for control purposas. aosc of cha aethods reportee
could ba applicable.
Tha aeeurata deteraination of ozona in tha aqueous phase is conplicated =•
the decomposition of ozone, its reactivity to the other species present, ana :hi
by-products of the ozonation reactions. Host current aethods vera developed
without a elaar knowledge of the associated osone chemistry. Therefore note o:
the aathods are uiiaeeeptable or cannot ba reeoaaended. In particular, nc
iodometrie based chemistry is acceptable for tha dateraination of aqueous ozor.e.
Indigo trisulfonata and arsenic(I1Z) direct oxidation are acceptable aeeheds.
Aaperometery continues to iaprove •• especially as an automated'control aethod.
The stripping techniques have soaa aaric in terms of improved ozem
selectivity. However, autoaated cheaieal systaas such as flow injecsiot
analysis offer considerably aora proaisa. Tha currant GD-FIA indigo proceeun
is superior for residual ozone aeasureaents due to its selectivity for ozone.
The aost important aspect of any potential new or iaproved ozone analysis*:
aethod will ba spaed of analysis and selectivity of the detacsion lyseea fa:
only ozone. As a point of eoaparison, wa strongly recommend that all future ant
existing aethods be coapared against tha "Ideal Method".
LITERATURE
1. Symons, J.H.; »t el "Ozone, Chlorine Dioxide and Chloraalnes as
Alternatives to Chlorine for Disinfection of Drinking ««cer" i:
CMorlnaclon: SnVitflfffltnSil IffigiSt «K* Httlth gfgtSSI,
2; Jolley, R.L. *, Corchev, H. and Kaailton, O.K., Jr., Editors. (Am
Arbor, HZ: Ann Arbor Science Publishers, Inc., 1979) pp. 555-56i
and Coaplete Report entitledi "State of the Art ..." (Cincinnati
OH: U.S. EPA, November. 1977), 84 pp.
2. Proceedings of Seainar on "Tha Deaign and Operation of Drinking Water
Facilities Using Ozone or Chlorine Dioxide", Rice. R.C., Editor
(Dedhasi, MA: Haw England (later Works Assoc., 1979).
3. Miller. G.V.; Rica, R.C.; Robson, C.M.; Scullin, R.L.: Kuhn, V. and
Wolf, H., "An Assessment of Ozone and Chlorine Dioxidi
Technologies for Treataenc of Municipal Water Supplies". U.S
Cnvironaental Protection Agency. EPA Project Report, SPA-600/2
71/018. 1978, 571 pp.
4. Miltner, R.J. "Measurement of Chlorine Dioxide and Related Products".
in Proceeding of che U*r.r nualltv Ttchnolostcal rgnftrtr;*
(Oanver, CO; Aaerican Water Wotks Assoc., 1976), pp. 1-11.
40
-------
I
5. Gordon, C. "laproved Methods of Analysis for Chlorate, Chloric*,
and Hypochlorite Ions at eh* Sub-ag/L L*v*l-, U.S. Environaental
Protection Agency, CPA Taehnieai Report. EPA-600/4-85/079, October.
1983, 35 p. and Prasanead at AUWA WQTC, tn »roe awa Water Otiall-v
Technology Canfmrmnem. Deceaber. Nashville, IS, 1982, pp. 175-189.
6. Aiata, E.M.; Roberts, P.V. "Chlorine Dioxida Cheaiatry: Canaraclon
and Residual Analysis" in Chemistry tn Hater Reuse. V^. X,
Coopar, W.J,, Editor (Ana Arbor, HI: Ann Arbor Scianca Publishers.
Inc., 1911), pp. 421.452.
7. Holgni, J.; lader, H. "Bestiaaung von Ozon und Chlordloxid in Wassar
¦it dar Indigo-Methods" ("Determination of Ozone and Chlorine
Oioxida in Water With the Indigo Method"), Voa Wassar, 1980, 15.
261-280.
8. Gilbert, E.; Holgni, J. "Messung von Ozon in Wassarverkan; Verglaieh
dar DPD- und Indlgo-Methoda" ("Ozone Meaauresient in Water Treataenc
Plants; Coaparison of the DPD and Indigo Methods"), CFW-
Waaser/Abvassar. 1983, Hi, 527-531.
9. Schaiekaap. H. "European Alternatives and Experience- in Proceeding
af ehe Satlonal fCanadian^ Conference an Critical tnues in
Drinking Water Quel icy. (Ottawa, Ontario, Canada: Federation of
Associations on Canadian Envlronaant, 1984), pp. 140-169.
10. Ikeda, Y.; Tang, T-F.; Cordon, C. "Iodometrie Method of Determination
of Trace Chlorate Ion", Anal. Chaa., 1984, 71-73.
11. Eaaenegger F.; Cordon, C. "Tfca Rapid Interaction between Sodiua
Chlorite and Dissolved Chlorine", Inorg. Chaa., 1967, £, 633-635.
12. Aiata, E.M.; Berg, J.D. "A Review of Chlorine Dioxide in Drinking
Water Treataent", J. Aa. Water Works Aasoc., 1986, If, 62-72.
13. Standard Methode for The anamination of Water end Wastewater. 16th
Editign. Creanbarg, A.E.; Trussell, R.R.; Clascari. L.S.; Franson,
M.A.H., Editors (Washington, D.C.: Aaarlcan Public Health Assoc.,
198S), 1268 pp. and 15th Edition. Creanbarg, A.E.; Connors. J.J.;
Jenkins, D.; Franson, M.A.H., Editors (Washington, DC; Aaarlcan
Public Health Assoc., 1980), 1134 pp.
14. Whittle, C.P.; Lapteff. A., Jr. "Haw Analytical Techniques for the
Study of Water Disinfaetion" in ch«ml«trv af Water Supply
TrtlSmiS. end Pt»erlhuelon. Rubin, A.J., Editor. (Ann Arbor. MI:
Ann Arbor Scl. Pub., Inc., 1974), pp. 63-88.
15. Toaiyasu, H.; Fukutoai, H.; Cordon. C. "Kinetics and Mechanlsa of
Ozone Decoapositlon in Basic Aqueous Solution". Inorg. Chea.. 1985,
IU, 2962-2966.
41
-------
Grunvell, J.; Benga, J.; Cohen, H., Cordon, C. "A Detailed Conparison
of Analytical Methods for Residual Oxone Measurement*, Ozena Sci.
Eng., 1983. i, 203-223.
Flaaa, D.L.; Anderson, S.A. "Iodate Formation and Decoaposicion In
Zodoaetric Analysis of Ozone". Environ. Sei. Technol., 1975. 2,
• 660*663.
Rehae, K.A.; Puzak, J.C.; Board, M.E.; Smith. C.F.; Paur. R.J.
"Evaluation of Ozone Calibration Procedures", U.S. Environaental
Protection Agency, EPA Project Suaaary, EPA-600/S4-10-050.
Fabruary. 1980, 277 pp.
Lieberaann, J., Jr.; Rescher, K.H.; Kaiar, E.P.; Cooper,.W.J, Develop*
aeat of tha FACTS Proeadura for Coabinad Forma of Chlorina and
Ozena in Aqueous Solutions*, Environ. Sei. Taehnol., 1980, i£,
1395-1400.
Palin, A.T.; Derreuaaux, A. "Ddeerainacion da I'Ozone Risiduel dans
l'aau" ("Determination of Ozena Rasidual in Water"), L'Eau *c
1'Industrie, 1977, 1£, 57-60.
Badar, H.; Hoigni, J. "Coloriaecric Mashod for eha Measurement
of Aqueous Ozone Based on the DacoLorization of Indigo
Derivatives", in Ozanization Sinual for Peter and 'Jnttwmr
Traatmanc. Haasehelein, U.J., Editor, (New York, SV: John
Wiley & Sons. 1982), pp. 169-172.
Badar. H.; Hoigni, J. "Determination of Ozone in Water by eha
Xndigo Method", Waeer Research 1981, 12, 649-456.
Badar, H.; Hoigni, J. "Dacarainacion of Ozone in Water by tha Indigo
Method; A Subaittad Standard Method", Ozone: Science and Eng.,
1982, 4, 169-176.
Scraka, M.R.; Cordon, G.; Paeey, C.E. "Residual Aqueous Ozone Deter-
ainaeien by Gas Diffusion Flow injection Analysis", Anal. Chen.,
1985, SI, 1799*1803.
Masschalein, W.J. "Continuous Aaperoaetrle Rasidual Ozena Analysis
la eha Tailfer (Brussels, Balgiua) Plane", ia Qzcmization Manual
for Water end Trmmrmmnc. Masschalein. W.J., Editor, (N'ev
York, KY: John Wiley & Sons, 1982), pp. 187-188.
Seaaley, J.H.; Johnson, J.D. "Aaperometric Meabrane Electrode for
Measureaent of Ozone in Water", Anal. Chaa., 1979, SI, 2144-2147.
Stanley, J.W.; Johnson, J.D. "Analysis of Ozone in Aqueous Solution",
In Handbook of Ozonm Technology and ApplicationsVftl- 1, Rice.
R.C. and Netzer, A., Editors (Ann Arbor, MI: Ann Arbor Sci. Pub.,
Inc.. 1982), pp. 235-276.
42
-------
a cuide r» imcim est of this ufoit (and a ikisf glossary OF THUS)
This Raporc contains a very detailed review of ail disinfectant residual
aeasuraaene aeehoda. Tha Cxaeuclva Suaaary Is intended co give readers a brief
overview of eha advantages and disadvantages of aaeh aethod. To that and. TabIa
X (Characteristics and Coaparlsons of Analytical Methods) has baan .incLudad :o
suaaarize aaeh of our findings and to racoaaend possible directions for fucura
rasaareh. In addition, labia II (Equivalent Weights for Calculating
Coneancraeions on eha Basis of Mass) daseribas eha aquivaiane vaighes of aaeh of
eha disinfection spaelas in earns of eha aeeual raaeeions involved in eha
disinfaeeion process.
iaeh ehapear contains individual raco—andations following eha discussion of
eha aaehod. A aiiaairy of all of eha raco—endatlons is also givan ae eha and of
aaeh ehapear. Additional halp is givan by aaaits of an alphabaeieal Index
containing aora ehan 2500 individual earns. Specific eross rafaraneing for all
raeoaaandaeiona can ba found in eha Index aiehar undar eha "recosendation*, or,
in earns of eha subject of eha nuabarad reeoanendation iesalf.
Tha eara Rafaraa Haehod is usad to daseriba appropriaca comparisons with
axiseing aaehods and Standard Methods rafars eo a spaeifieally recoomended
aaehod. Tha Indax should ba an addieional aid eo finding eha daeails of
spacifie aaehoda.
In ehis concaxc, ie should bo noead ehae eha'individual liearacura citations
ara spacifie eo aaeh individual ehaptar -- and ara aiehar nuabarad individually
within chapears 2 and 3, or alphabaeieally sequenced within chapters 4 and 5.
Chapter 4 (Indexed Reference Cieaeions) haa been included in ehis report in
order to assist readers in locating particular papers of ineerest. Tha 43
categories for chlorine, ehloraaines, and eha oxy-chlorine speeies, along with
eha addieional 60 cacegorles for osone, should aaka eha task of finding in-
dividual papers of inearese considerably lass cuabersoae. Papers which describe
several aethods have baan included in aaeh of eha appropriate eaeagories. All
eogeehar, the 1,400 reference* eitad in Chapters 1-3 nuaber aore than 2,000
individual cltaeions when distribueed in the Indexed fora of Chapter 4,
Chapter S is an alphabetical listing of eha individual references cieaeions.
Finally, a detailed Indax has bean included in order to assist readers in
locating subjects of spaelfie ineerest. Va hope the readers will find these
additional chapters as useful as have we In praparlng ehis report.
A brief Clossary follows on eha next page in order to assist readers in the
various specialised terns and abbreviations used in ehis rapore. For additional
teras, the reader is referred to the Index.
43
-------
GLOSSARY
Accuracy • • the ability to determine the correct concentration
BAKX •• boric acid buffarad pocaaaiua iodida aached for ozone
Breakpoint -• eha inorganic raaccion of chlorine with aaaonia nitrogen
CDFV • • chlorina deaand fraa water
Coabinad Chlorina •• inorganic and organic chloraainas
Dataction Liaic •• a signal chat is 3 ciaas cha noisa level of tha system
DOC -- dissolved organic carbon
DPD • • (N.N-diethyl-p-phenylenediaaine)
FACTS -- fraa available chlorine tasc vich syringaldarine
FIA •• flow injection analysis, an aueoaated analysis procedure
Free Chlorine -• the species. Cla + H0C1 + 0C1*
KI -- poeassiua iodide aethod for ozone
LCV •• leuco crystal violet
aL -- milliliter(«), standard unit of voluae
Molar Absorptivity (<) reported in units of H*'ca*1
NBKI •• neutral buffered potassiua iodide aethod for ozone
Precision •« how well the aethod reprodueibly aeasures the saae
concentration
Reactive Interaediate •• species such as 0," , HO,". H02, OH, 04", ate.
Referee Method •• the aethod aqainst which a working aethod is compared
Sensitivity • • eha change in signal par unit concentration (i.e. Aaps/aol]
Standard Maehoda -- the book. Standard Maehoda far cha Examlratlon £
tfatar and tfntmttr published by APHA. AUUA. and WPCF
THM's •• trlhaloaethanes
Total Chlorina •• Cha coabinaeion of Free Chlorine and Coabined Chlorine
TOC •• total organic carbon
TOX - - total organic halogen
uu
-------
APPENDIX E
INACTIVATIONS ACHIEVED
BY VARIOUS DISINFECTANTS
-------
TABLE E l
CT VALUES FOR IN ACTIVATION
OF (MAMMA CYSTS BY FREE CHLOfUNB
AT OS COR LOWER (t)
CHLORINE
pH<->6
gU
-6.5
#»
-7.0
PH
-7.5
CONCENTRATION
Loglaa
ellvilia
M
Im
ctrrHioi
Utlm
CtMtiMM
<•*«-)
0.S
1.0
1.5
2.0
2.5
3.0
0.5
1.0
I.S
2.0
2.5
3.0
05
1.0
I.S
2.0
2.5
3.0
0.5
1.0
is
2.0
15
3.0
<-0.4
23
4§
m
91
114
137
27
54-
12
109
136
163
33
65
M
130
163
I9S
40
79
119
isi
191
237
0.6
24
4?
n
94
III
Ml
21
56
M
112
140
161
33
C7
190
133
167
200
40
10
120
159
19*
239
0.1
M
41
n
9?
121
145
29
57
M
IIS
143
172
34
46
109
137
171
205
41
12
123
164
20S
m
1
2S
49
n
99
123
Ml
29
59
M
117
147
176
35
70
101
140
175
210
42
M
127
169
211
253
1.2
25
SI
n
Ml
12?
152
30
60
m
120
150
IM
36
72
101
143
19
215
43
16
130
173
214
2S9
1-4
26
S2
71
M»
129
ISS
31
61
92
123
153
IM
37
74
111
147
IM
221
44
19
133
ITT
222
266
1.6
26
at
79
MS
131
1ST
32
a
95
126
m
119
31
7S
113
151
III
226
4§
91
137
in
221
273
It
27
54
II
toa
I3S
162
32
64
97
129
161
193
39
T7
116
154
193
231
47
93
140
IM
233
279
2
n
SS
13
110
131
ICS
33
66
99
131
164
197
39
79
III
157
197
236
41
9S
MS
191
231
2*6
2.2
n
56
•5
113
Ml
169
34
67
101
134
161
201
40
II
121
161
202
M2
SO
99
M9
m
Ml
297
2.4
29
ST
•6
IIS
143
172
34
61
103
137
171
205
41
12
124
165
206
247
50
99
M9
m
MS
291
2.6
29
51
as
117
m
175
35
70
IQS
139
174
209
42
•4
126
161
2H>
252
SI
Ml
152
209
253
304
2.1
30
S9
•9
119
Ml
171
36
71
107
142
171
213
43
16
129
171
214
257
S2
NS
155
207
251
310
3
30
10
91
121
ISI
III
36
72
109
145
III
217
44
17
131
174
211
261
53
MIS
151
211
263
31*
CHLORINE
ftt>I.O
ptl
•1J
pH<-9.0
CONCENTRATION
L
Im
cbvatio
M
Lflg iMCtlWtllMft
0.5
1.0
1.5
2.0
2.5
3.0
0.5
1.0
IS
2.0
2.5
3.0
0.S
1.0
1.5
2.0
2.S
3 J»
<¦0.4
44
92
139
IIS
231
277
55
110
165
219
274
329
65
130
I9S
»0
325
390
o.«
4*
95
143
191
231
2*6
57
114
171
221
215
342
61
136
204
271
339
407
0.1
49
ft
141
197
246
295
S9
III
IT*
236
295
354
70
Ml
211
m
352
422
1
SI
101
152
303
253
304
61
122
113
243
301
365
73
146
219
291
364
437
1.2
52
IM
157
209
261
313
63
125
in
251
313
376
75
ISO
226
301
31V
4SI
.
1.4
34
107
161
214
261
321
65
129
194
251
323
317
77
155
232
309
317
464
1.6
•SS
110
165
219
274
329
66
in
199
265
331
397
•0
159
239
311
m
477
t
I.B
56
113
m
225
m
331
61
136
204
271
339
407
12
163
245
326
401
419
2
a
IIS
ITS
231
m
346
70
139
209
271
341
417
13
167
250
333
417
500
2.2
39
III
IT7
235
IflkJI
1
353
71
142
213
2M
355
426
IS
170
256
341
426
511
2.4
60
120
Itt
241
301
361
73
145
211
290
363
435
17
174
261
341
435
522
2.6
<1
123
114
245
307
361
74
141
222
MU
370
444
19
171
267
355
444
S3)
2.1
63
125
IU
2S0
313
375
75
151
226
301
377
452
91
III
272
362
453
543
1
64
127
191
255
311
312
77
153
230
307
311
460
92
IM
276
361
460
552
No
-------
TABLE E-2
CT VALUES FOR IN ACTIVATION
OF OlARDt A CYSTS BY FREE CHLORINE
CHLORINE
1
A
-6
pH-6.5
I*"
7.0
-7.5
CONCENTRATION
L
BglMt
M
LqB laactwaltoM
Log iMdnntMNM
Ughm
ctivtfiaaa
(•^L)
0J
1.0
IJ
2.0
2J
3.0
0.5
1.0
15
2.0
2.5
3.0
0.5
1.0
1.5
2.0
2.5
3.0
OJ
1.0
1.5
2.0
23
3.0
<*0.4
10
32
49
«S
Bl
97
20
39
59
7S
98
117
23
4*
70
93
II*
139
18
S3
83
III
1*
14*
0.4
17
33
90
C7
83
too
20
40
*0
BO
100
120
24
48
72
95
119
143
' 29
S7
8*
114
143
171
0.1
n
34
12
«•
Bi
K»
20
41
*1
SI
102
122
24
49
73
97
122
14*
29
S8
SB
117
M*
I7S
1
it
3»
S3
10
SB
MS
21
42
«
83
104
125
25
50
7S
99
124
149
30
*0
90
119
m
179
1.2
IB
3*
14
71
B9
107
21
42
*4
SS
106
127
25
SI
7*
101
127
IS2
31
*1
92
122
m
183
1.4
IB
3*
Si
73
91
M9
22
43
*5
S7
108
130
2*
52
78
103
III
ISS
31
*2
*4
I2S
IS*
187
1.*
19
37
54
74
93
III
22
44
*6
BS
110
132
2*
S3
79
105
132
ISS
32
*4
9*
I2B
1*0
192
I.I
1*
3B
17
7»
9S
IM
23
45
*8
90
113
135
27
54
SI
108
135
1*2
33
45
98
131
1*3
19*
2
19
39
SB
77
97
lit
23
4*
*9
92
115
138
28
SS
83
110
138
l*S
33
m
MO
133
1*7
100
2.2
20
39
S9
79
9B
IIS
23
47
70
93
117
140
28
5*
85
113
141
1*9
34
*8
raz
13*
170
304
2.4
20
40
<0
B0
100
120
24
4S
72
95
119
143
29
57
8*
115
143
172
35
70
105
139
174
309
2.*
20
41
«l
SI
102
122
24
49
73
97
122
14*
29
58
88
117
14*
I7S
3*
71
m
142
178
213
2.1
21
41
«
S3
103
124
25
49
74
99
123
148
30
59
B9
119
148
ITS
3*
72
m
MS
1BI
217
3
21
42
43
B4
105
12*
25
SO
7*
101
12*
ISI
30
*1
91
121
152
182
37
74
in
M7
184
221
CHLORINE
fM«8.0
pH
-8.5
1
A
-9.0
CONCENTRATION
iv^aa
Logk.
imm
0.S
1.0
IJ
2.0
2.S
3.0
0.S
1.0
1.5
2.0
2.5
3.0
0.5
1.0
1.5
2.0
2.S
3.0
<-0.4
31
<*
99
132
its
I9B
39
79
IIS
157
197
23*
47
93
140
IS*
233
279
0.6
34
«
t02
m
170
204
41
Bl
122
1*3
203
244
49
97
14*
194
243
291
0.1
35
10
108
140'
I7S
210
42
S4
12*
I6S
210
252
50
MO
ISI
201
251
301
1
3*
72
MB
144
110
21*
43
B7
130
173
217
2*0
S2
104
15*
208
210
312
1.2
37
74
III
147
IS4
221
45
19
134
178
223
2*7
S3
107
1*0
213
2*7
320
1.4
3S
7#
IM
ISI
IB9
227
4*
*1
137
1S3
228
294
SS
110
MS
219
274
329
1.*
39
77
II*
ISS
193
232
47
94
141
IS7
234
281
s*
112
149
225
2*1
337
is
40
79
119
119
190
231
4B
9*
144
191
239
287
S8
IIS
173
230
288
345
2
41
SI
122
142
203
243
49
9S
147
19*
24S
294
59
118
177
235
294
353
2.2
41
•3
124
ICS
207
24S
50
100
ISO
200
250
300
*0
120
ISI
241
301
3*1
'
2.4
42
B4
127
l«9
211
2S3
51
102
153
204
2SS
30*
*1
123
IS4
245
307
368
2.6
43
M
129
172
215
251
52
104
15*
20>
2*0
312
*3
12S
III
250
313
375
2.1
44
U
m
175
219
2*3
53
10*
159
212
2*5
318
*4
127
191
255
SIS
382
)
43
19
134
179
223
2*S
54
I0S
1*2
21*
270
324
*5
130
195
259
324
319
%•
• Ni«n:
(I) CT » CT for 3-log iwdBiiw
99.9
*
-------
TABLE E-3
err values for in activation
Of* OIAROIA CYSTS BY FREE CHLORINE
CHLORINE
CONCENTRATION
0.S
L
1.0
pH<—6
m IncCkMao
IS 2.0
M
2-5
3.0
0.5
pH-6.5
IsN&VltMNM
1.0 1.5 2.0 2.5
3.0
0.5
fH*=7.0
Log iMCtiwtilM
1.0 1.5 2.0
2.5
3.0
0.5
1.0
fH
1.5
-7.5
2.0
IS
3.0
<-0.4
12
M
m
49
<1
73
IS
29
44
59
73
II
17
35
52
69
17
KM
21
42
<3
S3
KM
121
o.«
13
23
m
SO
<3
75
IS
30
45
60
75
90
II
36
54
71
19
m
21
43
64
«
m
121
0.1
13
26
39
12
65
71
15
31
46
61
77
92
11
37
SS
73
92
110
22
44
66
17
m
131
1
13
26
40
S3
<6
79
16
31
47
63
71
94
19
3?
S6
7S
93
112
22
45
67
•9
112
134
1.2
13
21
40
S3
67
M
1*
32
4S
63
79
95
19
31
St
76
95
IM
23
46
69
91
IM
137
1.4
M
21
41
SS
<1
12
16
33
49
65
12
91
19
39
m
77
97
116
23
47
70
93
117
MO
1.6
M
¦2t
42
55
69
13
17
33
50
66
13
99
20
40
to
79
99
119
24
41
72
96
120
M4
l.t
14
29
43
S7
72
•6
17
34
51
67
14
101
20
41
61
II
102
122
25
49
74
N
123
M7
2
IS
29
44
SI
73
17
17
35
52
69
17
104
21
41
62
13
103
124
25
SO
75
too
125
ISO
2.2
IS
30
45
$9
74
•9
II
35
S3
70
U
105
21
42
64
15
106
127
26
SI
77
MS
121
IS3
2.4
IS
30
45
10
75
90
II
36
54
71
19
107
22
43
65
M
KM
129
26
92
79
MS
131
IS7
2.6
IS
31
46
61
77
92
II
37
SS
73
92
110
22
44
66
17
109
131
27
53
10
107
133
160
2.1
I*
31
4?
62
71
93
19
37
56
74
93
111
22
45
67
¦9
112
134
27
94
12
109
136
163
3
l«
32
41
63
79
*5
19
31
57
75
94
113
23
46
•
91
114
137
21
55
13
111
I3B
166
CHLORINE
CONCENTRATION
(¦««-)
0.S
L
1.0
pH-1.0
Bg laactmtia
1.5 2.0
as
2.5
3.0
0.5
fH-1.5
1.0 1.5 2.0 2.5
3.0
0 J
1.0
pH<—9.0
I.S 2.0
25
3.0
<-0.4
25
90
n
99
124
149
30
99
19
III
141
177
35
70
MS
139
174
209
0.6
26
SI
n
102
121
153
31
61
92
122
153
113
36
73
109
145
112
211
o.«
26
S3
n
I0S
132
151
32
63
95
126
151
119
31
75
113
ISI
in
226
t
27
S4
ii
KM
I3S
162
33
65
91
130
163
195
39
71
117
156
195
234
1.2
2S
SS
13
III
131
166
33
67
100
133
167
200
40
10
120
160
200
240
1-4
2t
S7
IS
113
142
170
34
m
103
137
172
206
41
12
124
165
206
247
1.*
29
51
17
ll«
145
174
3S
70
106
Ml
176
211
42
M
127
169
211
253
I.I
30
m
90
119
149
179
36
72
101
143
179
215
43
16
130
173
216
259
2
30
61
91
121
IS2
112
37
74
111
147
114
221
44
U
133
177
221
265
2.2
31
62
93
124
IS5
116
31
75
113
150
IS!
225
45
90
136
III
226
271
2.4
32
63
95
127
IS*
190
31
77
115
IS3
192
230
46
92
131
IM
230
276
2.6
32
65
97
129
162
194
39
71
117
156
195
234
47
94
141
117
234
211
2.1
3 J
66
99
131
164
197
40
10
120
159
199
239
41
96
144
191
239
2*7
1
34
67
101
134
I6S
201
41
ai
122
162
203
243
49
97
146
19S
243
292
No»c«:
(I) CT - CT for Meg ¦ftivllna
99.9
-------
TABLE E-4
CT VALUES FOR INACTIVATION
OF GIAKOIA CYSTS BY FREE CHLORINE
AT ISC (t)
CHLORINE
pH<
>6
P"
-6.5
pH-7.0
PH
-7.5
CONCENTRATION
L
of laact
tafia
M
Leg l««rti—lia
M
Laf IwrtiwtinM
Lcfi-
ctwrtioM
("Rt)
0.5
1.0
li
2.0
15
3.0
0.5
1.0
1.5
2.0
2.5
3.0
0.5
1.0
I.S
2.0
2.5
3.0
0.5
1.0
15
2.0
2.S
3J>
o
H
V
•
14
25
33
41
49
10
20
30
39
49
59
12
23
3S
47
51
70
M
2S
42
SS
•
S3
0.4
S
17
2S
33
42
so
10
20
30
40
50
60
12
24
36
41
60
72
M
29
43
57
72
S6
0.1
9
17
24
35
43
S2
10
20
31
41
SI
41
12
34
37
49
41
73
IS
29
44
59
73
SS
1
9
It
27
35
44
S3
21
32
42
S3
63
13
25
31
50
63
75
IS
30
4S
40
75
90
1.2
9
II
27
36
45
S4
II
21
32
43
S3
64
13
25
31
51
63
76
IS
31
44
61
71
92
1.4
9
It
21
37
44
SS
II
22
33
43
54
6S
13
26
39
52
6S
71
14
31
47
63
71
94
1.6
9
19
21
37
47
56
II
22
33
44
SS
66
13
26
40
53
66
79
16
32
41
64
to
96
1.1
10
19
29
31
41
57
II
23
34
4S
57
61
14
27
41
54
41
II
16
33
49
45
12
94
2
10
19
29
39
41
SI
12
23
35
44
51
69
14
21
42
SS
69
13
17
33
SO
67
S3
too
2.2
10
20
30
39
49
S9
12
23
35
47
51
70
14
21
43
57
71
IS
17
34
SI
41
SS
Ktt
2.4
10
20
30
40
50
40
12
24
36
41
60
72
14
29
43
57
72
16
It
35
S3
70
SS
MS
2.4
10
20
31
41
SI
61
12
24
37
49
61
73
IS
29
44
59
73
tl
II
36
54
71
19
M7
2.1
10
21
31
41
S2
62
12
25
37
49
62
74
IS
30
4S
59
74
•9
II
36
SS
73
91
M»
3
II
21
32
42
S3
63
13
25
31
SI
63
76
IS
30
44
41
76
91
19
37
54
74
93
III
CHLORINE
pH-
1.0
P«
-1.5
f«<
-9.0
CONCENTRATION
LobIm*
UbH>
LocImc
0.3
1.0
1.5
2.0
15
3.0
0.5
1.0
I.S
2.0
2.5
3.0
0.S
1.0
I.S
2.0
2.5
3.0
<¦0.4
17
.. n
SO
44
S3
99
20
39
59
79
91
III
23
47
70
93
117
140
0.4
17
34
SI
41
IS
102
20
41
61
II
102
122
24
49
73
97
122
144
0.1
It
35
S3
70
It
105
21
42
63
14
105
126
25
50
76
101
126
151
1
II
36
54
72
90
KM
22
43
6S
17
101
130
26
52
71
104
130
156
1.2
19
37
54
74
93
III
22
45
67
19
112
134
27
S3
10
107
133
160
1.4
I*
31
57
76
95
114
23
46
69
91
114
• 17
21
SS
13
110
131
165
1.4
. 19
39
SI
77
97
116
24
47
71
94
III
141
21
S4
IS
113
Ml
Itt
1.1
20
40
40
79
99
119
24
41
72
94
120
144
29
51
17
IIS
144
173
2
20
41
41
•1
102
122
25
49
74
91
123
147
30
99
19
III
141
177
2.2
21
41
42
13
103
124
25
50
75
100
125
ISO
30
40
91
121
151
III
2.4
21
42
44
15
106
127
26
SI
77
102
121
153
31
41
92
123
153
IM
2.4
22
43
65
•6
106
129
26
52
71
104
130
156
31
63
94
I2S
IS7
in
2.1
22
44
44
M
110
132
27
53
•0
106
133
159
32
64
96
127
159
191
3
22
45
67
•9
112
134
27
54
II
104
135
142
33
45
94
130
163
195
Nolo:
(I) CT » CT foe Hog
99.9
-------
TABLE E-5
CT VALUES FOR IN ACTIVATION
OF GIARXHA CYSTS BY FREE CHLORINE
AT 20 C (I)
CHUMUNB
pH<-6
I*
-6.5
yll>7.0
PH
-7J
CONCENTRATION
L
ag Im
m
Log iMClfVfldOM
Lflg IwKiivMtiQM
CtfWtlMM
0J
1.0
I.S
2.0
2J
3.0
0.S
1.0
IS
2.0
2.5
3.0
0.5
1.0
I.S
2.0
2.5
3.0
0.S
1.0
IJ
2.0
23
3.0
<*0.4
*
12
It
U
30
36
7
IS
22
29
37
44
9
17
26
35
43
52
M
21
31
41
52
62
0.*
*
IS
19
25
32
31
•
IS
23
30
31
45
9
11
27
36
45
54
II
21
32
43
S3
64
0.1
7
IS
20
2*
1)
39
1
IS
23
31
31
46
9
11
21
37
46
SS
11
22
33
44
SS
66
1
7
13
20
26
33
39
I
16
24
31
39
47
9
19
21
37
47
56
II
22
34
45
56
67
1.2
7
13
20
27
S3
40
1
16
24
32
40
41
10
19
29
31
41
57
12
23
IS
46
SI
69
1.4
7
14
21
27
34
41
1
16
25
33
41
49
10
19
29
39
41
SI
12
23
35
47
SI
70
1.*
7
14
21
2S
3S
42
1
17
2S
33
42
50
10
20
30
39
49
59
12
SI
36
41
60
72
1.1
7
14
22
29
36
43
9
17
26
34
43
51
10
20
31
41
51
61
12
25
37
49
62
74
2
7
IS
22
29
37
44
9
17
26
35
43
52
10
21
31
41
S2
62
13
25
31
SO
63
TS
2.2
7
IS
22
29
37
44
9
II
27
35
44
S3
II
21
32
42
S3
63
13
2*
39
SI
64
77
2.4
1
IS
23
30
M
45
9
II
27
36
45
54
II
22
33
43
54
65
13
26*
39
s
«S
n
2.6
1
IS
23
31
M
46
9
II
21
37
46
55
11
22
33
44
SS
<6
13
27
40
S3
67
m
2.1
1
It
24
31
39
47
9
19
21
37
47
S*
11
22
34
45
S«
67
14
27
41
54
61
ii
3
1
l«
24
31
39
47
10
19
29
31
41
57
II.
23
34
45
57
61
14
2S
42
55
69
13
CHLORINE
-1.0
ptt-1.5
|M<-9.0
CONCENTRATION
Lo*«-
ctiratao
L
of Im
utivflbQi
m
L
«*U«
iAflffllSMI
<-*«.)
0.5
1.0
I.S
2.0
2.5
3.0
0.S
1.0
1.5
2.0
2.5
3.0
0.5
1.0
IJ
2.0
2.S
3.0
<-9.4
12
2$
37
4f
«z
74
IS
30
45
59
74
19
11
3S
S3
70
n
MIS
-
o.«
1)
26
39
SI
64
77
IS
31
46
61
77
92
II
36
SS
73
91
109
0.1
13
26
40
SJh.
66
79
16
32
a
63
79
95
19
31
57
75
94
113
1
14
27
41
54
N
II
16
33
49
65
12
91
20
39
59
71
91
117
1.2
14
2S
42
SS
69
13
17
33
SO
67
13
100
20
40
60
IB
100
120
1.4
14
n
43
57
71
6
17
34
52
69
16
103
21
41
62
12
103
123
1.6
IS
39
44
SI
73
17
11
35
S3
10
II
105
21
42
63
M
KB
126
l.t
IS
30
4S
S9
H
•9
It
36
54
72
90
101
22
43
65
16
101
129
2
IS
30
46
61
76
91
II
37
55
73
92
110
22
44
66
II
110
132
2.2
16
31
47
62
71
93
19
31
57
75
94
113
23
45
61
90
113
135
24
16
32
4»
63
79
95
19
3t
SI
77
96
IIS
23
46
69
92
IIS
131
26
1*
32
49
65
11
91
20
39
59
n
91
117
24
47
71
94
til
141
2 •
If
S3
SO
66
•3
99
20
40
60
19
99
119
24
41
72
95
119
143
3
17
34
SI
67
84
101
20
41
61
•1
102
122
24
49
73
97
122
146
N«
-------
TABLE E-6
CT VALUES FOR IN ACTIVATION
OFGIARDIA CYSTS BY FREE CHLORINE
AT 25 C (I)
CHLORINE
ptl<-4
pH
—4.5
pti-
7.0
pH
1
CONCENTRATION
Log lnrti»
15
20
23
30
4
12
II
24
30
34
7
14
22
29
3ft
43
o.i
4
9
13
IT
22
2ft
5
10
M
21
26
31
4
12
19
25
31
37
T
13
22
29
37
44
I
4
9
13
IT
22
2ft
5
10
1ft
21
2ft
31
4
12
19
23
31
37
1
IS
23
30
31
45
1.2
S
9
14
11
23
27
5
11
Ift
21
27
32
4
13
19
25
32
31
1
IS
23
31
31
44
1.4
3
9
14
II
23
27
«
11
17
22
21
33
7
13
20
2ft
33
39
1
14
M
31
39
47
1.*
3
9
M
19
23
2S
6
II
17
22
21
33
7
13
20
27
33
40
I
14
M
32
40
41
l->
S
10
IS
19
24
29
ft
II
17
23
21
34
7
14
21
27
34
41
I
14
25
33
41
49
2
5
10
IS
19
24
29
ft
12
II
23
29
33
7
14
21
27
34
41
I
17
25
33
42
SO
2.2
S
10
IS
20
23
30
ft
12
II
23
29
35
7
14
21
21
35
42
9
17
2ft
34
43
51
2.4
3
10
IS
20
23
30
ft
12
II
24
30
36
7
14
22
29
34
43
9
17
24
3S
43
32
2.*
3
10
1*
21
2*
31
ft
12
19
23
31
37
7
IS
22
29
37
44
9
11
27
35
44
S3
2.8
3
10
1*
21
2ft
31
ft
12
19
23
31
37
I
IS
23
30
31
43
9
11
27
3ft
45
S4
3
3
11
1*
21
27
32
ft
13
19
2S
32
31
1
IS
23
31
31
44
9
11
21
37
4ft
55
CHLORINE
-8.0
»"
-1.3
1
A
-9.0
CONCENTRATION
Log 1—rthliuM
1 laactmtioas
Log liwth'rtiwi
(mm
0J
1.0
1.5
2.0
2 J
3.0
0.3
1.0
1.5
2.0
2.5
3.0
0.S
1.0
I.S
2.0
2.5
3.0
<-0.4
1
17
23
33
42
90
10
20
30
39
49
59
12
23
35
47
SI
70
o.«
9
IT
2*
M
43
51
10
20
31
41
31
61
12
24
37
49
41
73
o.»
9
IS
27
33
44
53
II
21
32
42
33
63
13
25
31
30
63
73
i
9
II
27
31
43
54
II
22
33
43
' 34
65
13
2ft
39
52
45
71
1.2
9
II
2S
3?
44
33
11
22
34
43
56
67
13
27
40
S3
47
10
1.4
10
19
29
31
4S
37
12
23
3S
44
SI
69
14
27
41
SS
61
12
1.*
10
19
29
39
41
a
12
23
35
47
31
10
14
n
42
56
70
14
I.I
10
20
30
40
30
40
t2
24
3ft
41
«0
72
14
29
43
57
72
M
.
2
10
20
31
41
31
41
12
23
37
49
62
74
IS
29
44
59
73
M
2.2
10
21
31
41
32
62
13
23
31
SO
63
75
IS
30
4S
60
75
90
•
2.4
II
21
32
42
S3
63
13
26
39
SI
64
77
IS
31
4ft
61
77
92
2.6
II
22
33
43
34
ftS
13
26
39
32
65
71
16
31
47
43
71
94
2.S
II
22
33
44
S3
66
13
27
40
S3
67
to
16
32
41
44
10
94
I
II
22
34
45
36
67
14
27
41
54
61
II
16
32
49
45
II
97
Note*:
(I) CT * CT far 34og iaactivatioa
-------
TABLE E-7
CT VALUES FOR
fNACTIVATION OF VIRUSES BY FREE CHL0R1NE<'>
Lm Imct 1 Yfltlon
MQ 3,0
_aa m m
m
90
60
45
30
22
Temoerature (C)
Jfct
m
Jhi
m
i=l
O.S
6
45
9
66
12
5
4
30
6
44
8
10
3
22
4
33
6
15
2
15
3
22
4
20
1
11
2
16
3
25
1
7
1
11
2
Notes:
1. Basis for values given in Appendix F.
-------
TABLE i-8
CT VALUES FOR
I MOTIVATION OF GIARDIA CYSTS
BY CHLQBTWF D10XIPf<'>
Temperature (C\
iMrtlYittwi.
£±1
is
jm
22
25
0.5-log
10
4.3
4
3.2
2.5
2
1-leg
21
®.7
7.7
6.3
5
3.1
1,5-log
32
13
12
10
7.5
5.!
2-log
42
17
15
13
10
7.:
2.5-log
52
22
19
16
13
9
3-1 og
63
26
23
19
15
11
Note:
1. Basis for values given in Appendix F.
-------
IfiOSXll
2-log
3-1og
4-log
TABLE 1-9
CT VALUES FOR
INACTIVATION OF VIRUSES
§Y CHLORINE PIOXIPC PH 6-9(,)
Ttinpcraturc
<¦1
i_
m
11
m
8.4
5.6
4.2
2.8
2.1
25.6
17.1
12.B
8.6
6.4
SO.l
33.4
25.1
16.7
12.5
25
1.4
4.3
8.4
Notes.-
1.
Basis for values given In Appendix F.
-------
TABU E-10
Inactlvatlgn
0,5-log
1-log
1.5-log
2-log
2.5-1og
3-leg
CT VALUES FOR
INACT1VATI0N OF 6IARDIA CYSTS
BY OZONE
.
•
Tenoeratur*
fC)
Ssl
1IL
15
2L
2i.
0.48
0.32
0.23
0.16
0.12
0.08
0.97
0.63
0.48
0.32
0.24
0.16
1.5
0.9S
0.72
0.48
0.36
0.24
1.9
1.3
0.95
0.63
0.48
0.32
2.4
1.6
1.2
0.79
0.60
0.40
2.9
1.9
1.43
0.95
0.72 .
0.48
Note:
1. Basis for values given in Appendix F.
-------
TABLE E-ll
CT VALUES FOR
INACTIVATIOH OF VIRUSES BY OZONE<'>
Inactivation
2-lOfl
3-log
4-log
ssi
1G-
laL
2SL
2S-
0.9
0.6
0.5
0.3
0.25
0.15
1.4
0.9
0.8
0.5
0.4 "
0.25
l.S
1.2
1.0
0.6
0.5
0.3
Note:
1. Basis for values given in Appendix F.
-------
TABLE 1-12
CT VALUES FOR
INACTIVATION OF G1ARDIA CYSTS
8Y CHLORftHIHE PH 6»9{T)
Ttrotrituri (C)
<-1
_J5_
•12—
-IS—
20
-ZL
0.5-log
. 635
365
310
250
185
125
1-log
1,270
735
615
500
*370
250
1.5-log
1,900
1,100
930 '
750
550
375
2-log
2,535
1,470
1,230
1,000
735
•50C
2.5-log
3,170
1,130
1,540
1,250
91S
62E
3-log
3,800
2,200
1,850
1,500
1,100
75C
1. Basis for values given in Appendix F.
-------
TABLE E-13
CT VALUES FOR
INACTIVATION OF VIRUSES BY CHLORAM!NE<'>
Twnoerature tc\
Inactlyation
-SsL.
__5—
-1IL-
-liL—
22-
22-
2-log
1,243
857
643
428
321
214
3-log
2,063
1,423
1,067
712
.534
356
4.log
2,883
1,988
1,491
994
746
497
Notes:
1. Bails for values given 1n Appendix F.
-------
TABLE E-14
CT VALUES FOR
1NACTIVATION QF VIRUSES BY UV<'>
Lop I reactivation
_ _
21 36
Basis for values given In Appendix F.
-------
APPENDIX F
BASIS FOR CT VALUES
-------
APPENOIX F
BASIS Qf CT VALUE'S
F.l Inactivation of Giardla Cvsts
F.i.i fret Chlorine
The CT values for free chlorine in Tables E-l through E-6 are based
on a statistical analysis (Clark et al., 1988; attached to this appendix),
which considered both animal infectlvlty studies (Hlbler et a.1.-, 1987) and
excystatlon studies (Jarroll et al., 1981; R1ce et al., 1982; Rubin et
al., 1988). A multiplicative model was selected to best represent the
chemical reactions during the inactivation process. This model was
applied to each of the data sets, listed above, and 1n various comblna-
tlons. The animal infectlvlty data were Included 1n all combinations
studletf. The animal Infectlvlty data was considered essential for
Inclusion 1n all the analysis of combined data sets because 1t included
many more data points than the other data sets, all of which represented
Inactivation levels at 99.99 percent. Because of limitations with the
excystatlon methodology, only data for achieving less than 99.9 percent
Inactivation was available from such studies.
Statistical analysis supported the choice of combining the Hlbler et
al. and the Jarroll et al. data (and excluding the R1ce et al. (1981) and
Rubin et al. (1987) data), to form the best fit model for predicting CT
values for different levels of inactivation. As a conservative regulatory
strategy, Clark and Regl1 (1990) (attached at the end of thsi appendix),
recommended that CT values for different levels of inactivation be
determined by applying first order kinetics to the 99 percent upper
confidence Interval of the CTM „ values predicted by the model.
The model was applied using the above strategy as a safety factor,
to determine the CT values ranging from 0.5-log to 3-log inactivation at
0.5 and 5 C. CT values for temperatures above 5 C were estimated assuming
a twofold decrease for every 10 C. CT values for temperatures at 0.5 C
were estimated assuming a 1.5 times increase to CT values at 5 C. This
general principle is supported by Hoff (1986). It is Important to note
that the CT values for free chlorine are sensitive to the residual
F-l
-------
% concentration, C. For example, at a pH of 7 and a temperature of 10 C, a
3-lofl Giardia cyst inactivation results from a CT of 107 ng/L-«1n with a
free residual of 0.6 g/l and a CT of 124 ng/L-nin with a free residual of
2.0 ag/L.
Application of the aodel to pHs above 8, up to 9, Mas considered
reasonable because the aodel is substantially sensitive to pH (e.g., CTs
at pH 9 are over three tiaes greater than CTs at pH 6 and over two tiaes
greater than CTs it pH 7). At a pH of 9, approximately four percent of
the hypochlorous acid fraction of free chlorine is still present. Recent
data Indicate that in terns of H0C1 residuals (versus total free chlorine
residuals Including H0C1 and 0C1*> the CT products required for inactlva-
tion of Siardia rouris and MMil lamb Ha cysts decrease with Increasing
pH from 7 to 9 (Leahy et al., 1987; Rubin et al., 1988b). However, with
increasing pH, the fraction of free chlorine existing as the weaker
oxidant species (0C1) increases. In terns of total free chlorine
residuals (i.e., H0C1 and OCT) the CT products required for inactivation
of Siardia muris cysts increase with Increasing pH from 7 to 9 by less
than a factor of 2 at concentrations of less than 5.0 ag/L (see
Table F-l). Also, the significance of pH on the value of CT products
achieving 99 percent inactivation appears to decrease with decreasing
temperature and free chlorine concentration. The relative effects of pH,
temperature, and chlorine concentration, on inactivation of Giardia muris
cysts appears to be the sane for Siardia laifclii cysts (Rubin et al.,
1988b), although not as nuch data for fiiindii Iambi la cysts for high pH
and temperature values as for Siardia ituris cysts Is yet available.
F-l-2 Ozone and Chlorine Dioxide
The CT values for ozone in Table E-10 are based on disinfection
studies using in vitro excystation of
61 irdli 1 Mlb 118
(Wlckramanayake, S. B., et al., 1985). CTti values at 5 C and pH 7 for
ozone ranged from 0.46 to 0.64 (disinfectant concentrations ranging from
0.11 to 0.4a ag/L). Mo CT values were available for other pHs. The
highest CT„ value, 0.64, was used as a basis for extrapolation to obtain
the CT values at 5 C, assuaing first order kinetics and applying a safety
factor of 2, e.g., (0.64 X 3/2 X 2 - 1.9). CT values for temperatures
-------
TABLE F-l
CT VALUES TO ACHIEVE 99 PERCENT
INACTIVATION OF 6IARDIA MURIS CYSTS BY FREE CHLORINE
(Source: Rubin, et «K, 1968b)
fitt
7
Temperature
Concentration fmo/L\
1.0-2.0
1
500
760
1,460
1,20
IS
200
290
360
29
8
I
510
820
1,580
1,30
15
220
32
9
1
440
1,100
1,300
2,20
IS
310
420
620
76
-------
above 5 C were estimated assuming a twofold decrease for every 10 C. CT
values for temperatures at 0.5 C were estimated assuring a 1.5 tines
Increase to CT values at 5 C.
The CT values for- chlorine dioxide 1n Table E-8 are based on
disinfection studies using la vitro excystatlon of Glardla tiur'U CTM
values at pH 7 and 1 C, 5 C, 15 C and 25 C (Leahy, 1985 and Rublrt, 1988b).
The average CTff value at each temperature (27.9 at 1 C, 11.8 at 5 C, 8.5
at 15 C, and 4.7 at 25 C) was extrapolated using first order kinetics and
multiplied by a safety factor of 1.5 to obtain the CT„ , values, e.g.,
at 1 C, C„ , - 27.9 x 1.5 x 1.5 ¦ 63.
Because of the United data available at pHs other than pH 7, the same CT
values are specified for all pHs. Although most of the CT„ data were
determined at pH 7, It Is known that chlorine dioxide Is more effective at
pH 9. Thus, the CT values 1n the rule are more conservative for higher
pHs than for lower pHs.
A lower safety factor Is used for chlorine dioxide than for ozone,
because the data was generated using Giardia muris cysts which are more
resistant than Glardla Iambi1a cysts. CT values at other temperatures
were estimated, based on the same rule of thumb multipliers assumed for
ozone.
A larger safety factor was applied to the ozone and chlorine dioxide
data than to the chlorine data because:
a. Less data were available for ozone and chlorine dioxide than
for chlorine;
b. Data available for ozone and chlorine dioxide, because of the
limitations of the excystatlon procedure, only reflected up to
or slightly beyond 99 percent Inactlvatlon. Data for chlo-
rine, based on animal Infectlvlty studies rather than excysta-
tlon procedures, reflected Inactlvatlon of 99.99 percent.
Extrapolation of data to achieve CT values for 99.9 percent
Inactlvatlon with ozone and chlorine dioxide, involved greater
uncertainty than the direct determination of CT values for
99.9 percent Inactlvatlon using chlorine.
c. The CT values for ozone and chlorine dioxide to achieve 99.9
percent Inactlvatlon are feasible to achieve; and
d. Use of ozone and chlorine dioxide Is likely to occur within
the plant rather than In the distribution system (versus
chlorine and chloramines which are the likely disinfectants
-------
for use 1n the distribution system). Contact time measure-
ments within the plant will Involve greater uncertainty than
Measurement of contact tine In pipelines.
EPA recognizes that the CT values for ozone and chlorine dioxide are
based on United data. Therefore, EPA encourages the generation of
additional data 1n accordance with the protocols provided In Appendix 6 to
determine conditions other than the specified CT values,' for providing
effective disinfection at a particular system.
F.1.3 Chloramlnes
The CT values for chloramlnes in Table £-12 are based on disinfec-
tion studies using preformed chloramines and in vitro excystatlon of
fiiardia muris (Rubin, 1988). Table F-2 sumoarizes CT values for achieving
99 percent Inactlvatlon of fiiardia muris cysts. The highest CT values for
achieving 99 percent inactlvatlon at 1 C (2,500) and S C (1,430) were each
multiplied by 1.5 (i.e., first order kinetics were assumed) to estimate
the CT„ , values at 0.5 C and 5 C, respectively, in Table £-12. The CT„
value of 970 at 15 C was multiplied by 1.5 to estimate the CT„ 8 value.
The highest CT„ value of 1,500 at 15 C and pH 6 was not used because it
appeared anomalous to the other data. Interesting to note is that among
the data 1n Table F-2 the CT values in the lower residual concentration
range (<2 mg/L) are higher than those 1n the higher residual concentration
range (2-10 mg/L). This is opposite to the relationship between these
variables for free chlorine. For chloramlnes, residual concentration may
have greater influence than contact time on the inactlvatlon of Giardia
cysts within the range of chloramine residual concentrations practiced by
water utilities (less than 10 mg/L). No safety factor was applied to
these data since chloramination, conducted in the field, is more effective,
than using preformed chloramlnes. Also, S1ard1a mrts appears to be more
resistant than fiiardia lamb Ha to chloramlnes (Rubin, 1988b).
The protocol in Appendix G can be used to demonstrate if less
stringent disinfection conditions than those cited 1n Table E-12 can
achieve comparable levels of inactlvatlon for specific system characteris-
tics.
-------
TABLE F-2
CT VALUES FOR 95 PERCENT
INACTIVATfflN OF GIARDTA HUftIS CYSTS i¥ MONOCHLQRAMINE*
(Source: Rubin, 1988}
Temperature Honochloramlne Concentration fmn/n
fill (S3 1,500 >880
1 >1,500 >880
7 15 >970 970
5 >970 1,400
1 2,500 >1,400
8 15 1.000 530
5 >1.000 1,430
1 >1,000 1,880
15 890
5 >890 >560.
1 >890 >560
*CT values with ">" signs are extrapolated from the known data.
-------
F.2 Inactivation of Viruses
F.2.1 Free Chlorine
CT values for fret chlorine ire based on data by Sobsey (1988) for
inactivation of Hepatltus A virus (HAV), Strain HM175, at pH 6,7,8,9 and
20, chlorine concentrations of 0.5 to 0.2, and a temperature of 5 C, as
contained in Table F-3, The highest CT value for the pH range 6-9 for
achieving 2, 3, and 4-log inactivation of HAV were multiplied by a safety
factor of 3 to obtain the CT values listed in Table E-7. (e.g., the CT
value for achieving 4-log inactivation at pHs 6-9 was determined by
multiplying 2.55 X 3 ¦ 7.6 ¦ 8). The CT values at pH 10 Mere significant-
ly higher than those for pHs 6-9 and are considered separately. The CT
values in Table E-7 for pH 10 also include a safety factor of 3. CT
values at temperatures other than 5 C were determined assuming a two fold
decrease for every 10 C increase. CT values for inactivating viruses in
general are based on HAV data since they give higher CT values than those
for inactivation of polio and rotaviruses under similar conditions of pH
and temperature (Hoff, 1986).
F.2.2 Chlorine Dioxide
Data by Sobsey (1988) for inactivation of Hepatltus A virus, strain
HM 175, by a chlorine dioxide concentration of 0.5 mg/1 at pH 6 and 5 C is
shown in Table F-4. The CT values in Table E-9 for pHs 6-9 and tempera-
ture ¦ 5 C were determined by applying a safety factor of 2 to the average
CT values presented in Table F-4 at pH 6. This safety factor is lower
than that used to determine CT values for chlorine because chlorine
dioxide appears to be significantly more effective at higher pHs and most
waters are assumed to have a higher pH than 6.
CT values at temperatures other than 5 C in Table E-9 were
determined by applying a twofold decrease for every 10 C increase. The
data for pH 9 was not considered because it is very limited and other
viruses are more resistant to chlorine dioxide than Hepatltus A 1s at pH
9. According to Hoff (1986) at a pH of 9 and a temperature of 21 C, a CT
of 0.35 provides a 4-log Inactivation of poliovirus 1. Applying the same
safety factor and rule of thumb multipliers to this data results in a CT
F-5
-------
. of 2.8 for a 4-log virus Inactlvatlon at 0.5#C, In contrast to a CT of
50.1 resulting from the Hepatltus A data at pH 6. Therefore, in order to
assure Inactlvatlon of Hepatltus A, the higher CT values are needed.
Systems with high pHs may wish to demonstrate the effectiveness of
chlorine dioxide at lower CT values based on the protocol In Appendix G.
Chlorine dioxide Is ouch sore effective for inactivating rotavirus and
polio virus than It is for Inactivating HAV (Hoff 1966).
f.2.3 Chi oramines
The CT values in Table E-13 at 5 C were based directly on data by
Sobsey (1988) using preformed chloramlnes at pH 8. No safety factor was
applied to the laboratory data since chloramination'1n the field, where
some transient presence of free chlorine would occur, 1s assumed more
effective than preformed chloraaines.
HAV 1s less resistant to preformed chloraaines than are other
viruses. For example, CTs of 3,800-0,500 were needed for 2-log Inactlva-
tlon of slalan rotavirus at pH - 8.0 and temperature - 5 C (Beruan and
Hoff, 1984) „ However, these same viruses are very sensitive to free
chlorine. CT values ranging from less than 0.025 to 2.16 were required to
achieve 99 percent Inactlvatlon of rotavirus by free chlorine at pH ¦ 6-10
and temperature • 4-5 C (Hoff, 1986). HAV 1s more resistant to free
chlorine than are rotaviruses%
The CT values 1n Table E-13 apply for systems using combined
chlorine where chlorine Is added prior to annonla in the treatment
sequence. This should provide sufficient contact with free chlorine to
assure inactlvatlon of rotaviruses. CT values Table E-13 should not be
used for estimating the adequacy of disinfection in systems applying
preformed chloraaines or ammonia ahead of chlorine, since CT values based
on HAV inactlvatlon with preformed chloraaines may not be adequate for
destroying rotaviruses. In systems applying preformed chloramlnes, 1t 1s
recommended that inactlvatlon studies as outlined In Appendix G be
performed with Bacteriophage HS2 as the Indicator virus to determine
sufficient CT values. Also, the protocol in Appendix G can be used by
systems applying chlorine ahead of ammonia to demonstrate less stringent
disinfection conditions than those Indicated In Table E-13.
F-6
-------
TABLE F-3
CT VALUES FOR INACTIVATION OF HEPATITUS A VIRUS
^ FREE CHLORINE
(Source: Sobsey 1988)
m INACTIYATIQH
2
3
4
dH
2
fi
1
11
1.18
0.70
1.00
1.25
19
1.75
1.07
1.S1
1.9
14
2.33
1.43
2.03
2.55
9
-------
TABLE F-4
CT VALUES FOR INACTIVATION OF HEPATITUS A VIRUS
BY CHLORINE DIOXIDE rSOBSFV \9M\
CIQj Concentration froo/H
Experiment
—Hsu Initial Average
pH6 1 0.49 - 0.32
2 0.50 0.33
3 0.51 0.36
4 0.51 0.37
pH9 1 0.5 - 0.5
2 0.5 0.5
Inactivation Time
Exoeriment
NOt
Exoeriment No.
Log
A\
Inactivation
1
2
3
4
" 1
2 3 4
—
pH6
2
12
9
5
7
3.8
3.0 1.8 2.6
3
30
29
22
20
9.4
9.6 7.9 7.4
4
55
59
43
39
17
20 16 14
•
pH9
>2.5
0.33
m m
<0.17
4
>3.6
0.33
—
—
mm
<0.17
mm mm mm
«
Note;
1. CT values were obtained by multiplying inactivation time by the avi
concentration shown above for each experiment.
-------
F.2.4 Ozone
No laboratory CT values based on Inactivatlon of HAV virus are yet
available for ozone. Based on data from Roy (1982), a Man CT value of
0.2 achieved 2-log Inactivatlon of pollovlrus 1 at 5 C arid pH 7.2. Much
lower CT values are needed to achieve a 2-log Inactivatlon of rotavirus
(Vaughn, 1987). No CT values were available for achieving greater than a
2-log Inactivatlon. The CT values In Table E-ll for achieving 2-1og
Inactivatlon at 5 C were determined by applying a safety factor of 3 to
the data from Roy (1982). CT values for 3 and 4-log Inactivatlon were
determined by applying first order kinetics and assuming the sane safety
factor of 3. CT values were adjusted for temperatures other than 5 C by
applying a twofold decrease for every 10 C Increase. Based on the
available data, CT values for ozone disinfection are not strongly
dependent on pH. Therefore, data obtained at pH • 7.2 1s assumed to apply
for pHs In the range of 6.0 to 9.0. However, 1t should be noted that the
maintenance of an ozone residual Is affected by pH.
F.2.5 ultraviolet light (UY)
The CT values for Inactivatlon of viruses by UV are based on studies
by Sobsey (1988) on Inactivatlon of Hepatitis A virus (HAV) by UV. These
data were used because HAV has been established as an Important cause of
waterbome disease. The CT values were derived by applying a safety
factor of 3 to the HAV Inactivatlon data. The CT values 1n Table E-14 are
higher than the CT values for UV Inactivatlon of pollovlrus 1 and simian
rotavirus from previous studies (Chang et al., 1985).
F.2.6 Potassium Permanganate
Potassium permanganate 1s a commonly used oxidant 1n water
treatment. Preliminary testing by Yahya, et al 1988, Indicates that
potassium permanganate may contribute to virus Inactivatlon. The test
data Included in Table F-5 indicates the Inactivatlon of bacteriophage
MS-2 using potassium permanganate with a pure water-buffer solution.
These data da not include safety factors. It 1s likely that CT values for
actual water treatment processes will differ from these values. This data
has only been provided here as an indication of the potential of potassium
-------
TABLE F-5
CT VALUES FOR 2-LOG IKACTIVATION
OF MS-2 BACTERIOPHAGE WITH POTASSIUM PERMANGANATE
KMnO.
(mg/L) PH 6.Q PH 8.0
0.5 27.4 a(1) 26.1 a
1.5 32.0 a 50.9 b
2.0 NDW 53.5 c
5.0 63.8 a 35.5 c
Notes:
1.
2.
Letters indicate different experimental conditions.
Not determined.
-------
THE BASIS FOR 6IARDIA C T VALUES IN THE SURFACE WATER
TREATMENT RULE: INACTIVATION BY CHLORINE
by
Robert M. Clark, Director
Drinking Water Research Division
Risk Reduction Engineering Laboratory
Cincinnati, OH 45268
and
Stlg Reg!1
Office of Drinking Mater
U.S. Environmental Protection Agency
Washington, DC 20460
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH ANO DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
February 1991
-------
THE BASIS FOR SIARDIA C T VALUES IN THE SURFACE HATER
TREATMENT RULE: INACTIVATION BY CHLORINE
by
Robert M. Clark,* and Stlg R«g11b
INTRODUCTION
The 1986 amendments to the Safe Drinking Hater Act (SDWA) require EPA to
promulgate primary drinking water regulations (a) specifying criteria under
which filtration would be required, (b) requiring disinfection as a treatment
technique for all public water systems, and (c) establishing maximum
contaminant levels (MCLs) or treatment requirements for control of Siardia
lamb!1a. viruses, Legionella, heterotrophic plate count bacteria, and
turbidity. EPA has promulaged treatment technique requirements to fulfill the
SDWA requirement for systems using surface waters and ground waters under the
direct Influence of surface water.1 Additional regulations specifying
disinfection requirements for systems using ground water sources not under the
direct Influence of surface water will be proposed and promulgated at a later
• date. This paper presents a model that relates pH, temperature, chlorine
concentration, and Inactlvatlon level to Glardia inactlvatlon by free
chlorine. Because Giardia lamblia Is known to be one of the most resistant
organisms to disinfection by chlorine found 1n water, much Interest and effort
'Director, Drinking Water Research Division, Risk Reduction Engineering
Laboratory, 26 W. Martin Luther King Drive, Cincinnati, Ohio 45268
bUSEPA, Office of Drinking Water, 401 M Street, S.W., Washington, DC 20460
-------
has been devoted to determination of C't values for 61ard1a lamblla. The
model has been used to predict "C t" values that have been Included as part of
EPA's Surface Water Treatment Rule (SWTR).
BACKGROUND
Under the SWTR all conmunlty and non-counun1ty public Mater systems using
surface vater, or ground water under the direct Influence of surface water,
are required to provide minimum disinfection to control filardia Iambi la.
enteric viruses and bacteria.1 In addition, unless the source water 1s well
protected and meets certain water quality criteria (total or fecal conforms
and turbidity limits), treatment must also Include filtration. The treatment
provided, 1n any case, Is required to achieve at least 99.9 percent removal
and/or 1nact1vat1on of Glirdla llfflfelii cysts and at least 99.99 percent
¦ removal and/or Inactlvatlon of viruses (I.e., virus of fecal origin and
Infectious to humans). Unflltered systems are required to demonstrate that
disinfection alone achieves the minimum performance requirements by monitoring
disinfectant resldual(s), disinfectant contact tlme(s), pH (If chlorine Is
used), and water temperature. These data must.be applied to determine If their
"Ct' value [the product of disinfectant concentration (mg/L) and disinfectant
contact (minutes)] equals or exceeds the C't values for filardia Iambi la
specified In the SWTR.1 With the exception of chloramlnes, where ammonia 1s
added prior to chlorine, these C't values are also adequate to achieve greater
than 99.99 percent 1nact1vat1on of viruses. For filtered systems, states are
required to specify the level of disinfection for each system to ensure that
their overall treatment achieves at least 99.9 and 99.99 percent removal
and/or Inactlvatlon of Glardla Iambi la cysts and viruses, respectively.1
In the Guidance Manual to the SWTR, EPA recommends C't values for
-------
different disinfectants to achieve levels of Inactlvatlon for unflltered
systems, filtered systems will be required to achieve less Inactlvatlon then
required for unflltered systems. The percent Inactlvatlon that filtered
» ¦» •
systems should achieve as a function of the filtration technology In place and
source water quality conditions 1s also recommended.2
PROBLEM
The destruction of pathogens by chlorlnatlon 1s dependent on a number of
factors, Including water temperature, pH, disinfectant contact time, degree of
mixing, turbidity, presence of Interfering substances, and concentration of
chlorine available. The pH has a significant effect on Inactlvatlon
efficiency because 1t determines the species of chlorine found 1n solution,
each of which has a different inactlvatlon efficiency.
The Impact of temperature on disinfection efficiency Is also significant.
For Example, Clarke's work in virus destruction by chlorine Indicates that
contact time must be Increased two to three times when the temperature 1s
lowered 10°C.3 Disinfection by chlorlnatlon can Inactivate Glardla cysts, but
only under rigorous conditions. Host recently, Hoff et al. concluded that (1)
these cysts are among the most resistant pathogens known, (2) disinfection at
low temperatures Is especially difficult, and (3) treatment processes prior to
disinfection are Important.4
Typical C t values for 99 percent Inactlvatlon of Siardia laiblla by free
chlorine at different temperatures and pH values are shown In Table 1.
3
-------
TABLE 1. C T VALOES FOR f« INACTIVATION OF 6IARDIA4
MBfiLM CYSTS BY FREE CHLORINE
Ranoe
Disinfectant
Tenp
Concentration
Tine
Mean
No. of
(*c
pH
(ng/L)
(¦In)
C*t
Ct
Experiments
5
6
ko-s%a
f-47
47-14
65
4
7
2.0-8.0
7-42
56-152
97
3
8
2.0-8.0
72-164
72-164
110
3
15
6
2.1-3.0
7
18-21
20
2
7
2.5-3.0
6-18
18-45
32
2
8
2.5-3.0
7-21
21-52
37
2
21
6
1.5
<.s
< 9
< 9
1
7
1.5
< 7
<10
<10
1
8
1.5
< 8
<12
<12
1
Jarroll el al., using in vitro excystatlon to determine cyst viability,
showed that greater than 99.8 percent of Siardla Iambi 1a cysts can be killed
by exposure to 2.5 ng/L of chlorine for JO minutes it 158C and pH 6, or after
60 minutes at pH 7 or 8. At 5°C, exposure to 2 ng/L of chlorine killed at
least 99.8 percent of all cysts at pH 6 and 7 after 60 Minutes.* While It
required 8 ng/L to kill the sane percentage of cysts at pH 6 and 7 after 10
minutes, 1t required 8 ng/L to Inactivate cysts to the same level at pH 8
after 30 nlnutes. Inactlvatlon rates decreased at lower temperatures and at
higher pH values as Indicated by the higher C t values.
Because of the obvious Interactions among these variables It Is essential
that a model be developed for predicting C t values under the various
conditions that nay exist in drinking water systems.
4
-------
OBJECTIVE
As Indicated, many factors Influence filirtili Iambi 1a reaction kinetics.
The objective of the study described 1n this paper therefore Is to develop an
equation that will relate C t values for Slardla Inactivated by chlorine to
such factors as pH, temperature, level of Inactlvatlon and chlorine
concentration. As mentioned previously, this equation ultimately provided the
values presented In the SWTR and associated Guidance Manual for disinfection
of Glirdl! Iambi 1a by free chlorine.
SIGNIFICANCE
The significance of these efforts relates to the fact that EPA's Office
of Drinking Hater has adopted the C't concept to quantify the 1nact1vat1on of
61ard1a Iambi 1a by disinfection with free chlorine. Whether or not a utility
1s forced to Install surface water treatment will depend on Its ability to
meet the C't values specified by the SWTR. Even 1f the utility 1s not
required to Install filtration a utility may have to make significant
Investments 1n holding basins and disinfection capacity 1n order to meet these
requirements. Therefore C't values established under the SVTR will be
extremely Important to the drinking water Industry and the authors believe It
1s Important that the Industry understand the basis for the procedures used to
estimate these values. This paper describes the way In which C't values were
calculated for the SVTR. It Is unlikely that utilities can directly use the.
models developed In this paper, although It Is Important that they understand
the mechanism by which C't values have been derived. Tables generated from
the model will be useful as they provide the C't values for Glardla Inactl-
vatlon by chlorine that utilities must achieve. These tables are presented at
the end of the paper.
5
-------
THEORY
Current disinfection theory 1s based on the Chick or Chick-Watson model.
Chick's 1m expresses the rate of destruction of microorganisms based on a
first-order chemical reaction.1
dlf/dt - -kt (1)
which when Integrated yields
In (VN0) - -kt (2) .
where
Nt - number of organisms present at time t (minutes)
N0 - number of organisms present at time 0
k - rate constant characteristic of type of disinfectant,
microorganism, and water quality aspects of system (minutes'1)
t - time (minutes)
Watson, using Chick's data, refined this equation to produce an empirical
relation that Included changes 1n the disinfectant concentration:7
In (N/No) - r Cnt (3)
where
C ¦ concentration of disinfectant [(m11Hgrams/11ter),/n]
r - coefficient of specific lethality (I1ters/m1 IHgram "minutes)
n - coefficient of dilution (Uters/mllllgrams s minutes)
* or
(1/r) In (Nt/N#) - Cnt (4)
For a given level of survival such as ¦ 0.001 (3 log reduction) the left
hand side of equation 4 1s a constant K, or
K - Cnt (5)
The value K will vary depending on the level of 1nact1vat1on.
6
-------
EFFECT OF OTHER VARIABLES
As Indicated previously C't values have been found to be a function of
pH, temperature, disinfectant concentration and level of Inactlvatlon.1
Therefore In this study equation S was reforaulated as follows:
C t - C'(n*l) K (6)
where
K - f (pH, temp, I)
I ¦ ratio of organisms at tlM t to the organlsas at tine 0 (Nt/N0)
temp • temperature at which experiment was conducted In °C
pH - pH at which experiment was conducted In pH units
Equation 6 can be rewritten In the form:
C t - R I*CbpHetempd (7)
where
R,a,b,cl and.d are coefficient to be determined.
A more convenient form for coefficient estimation and the one used In this
paper Is as follows:
t - R I'Cb"1pHctempd (8)
As will be discussed In the following sections these coefficients will be
determined by a statistical analysis using appropriate data bases.
COEFFICIENT ESTIMATES
Several data sets are available for estimating the coefficients 1n .
equation 8. Data sets have been developed by Jarroll, Hlbler, R1ce and
Rubin.5,,,10,u
Much of the available Slardla Inactlvatlon data Is based on excystatlon
rather than animal 1nfect1v1ty since 1t 1s an easier measure of cyst
viability.11 Hoff et al. compared mouse 1nfect1v1ty and excystatlon for
7
-------
determining the viability of fi. mris cysts exposed to chlorine ind reported
thit both methods yielded similar results." Hlbler it al. used Mongolian
gerblls to detemlne the effects of chlorine on fi. Iambila cysts.* In a
series of expedients, cysts were exposed for various tlM periods to free
chlorine concentrations ranging from 0.4 to 4.2 ng/L it 0.5, 2.5, and §.Q°C
and pH 6t 7, and 8. Each of I gerblls was fed I x 10* of the chlorine exposed
cysts and subsequently examined for evidence of Infection. Since the test
animals had each received a dose of 5 x 104 of the chlorine exposed cysts and
subsequently examined for evidence of Infection and since 1nfect1v1ty studies
with unchlorlnated cysts showed that approximately I cysts usually constituted
an infective dose, the following assumptions were made depending on the
1nfect1v1ty patterns occurring In the animals. If all five animals were
Infected, 1t was assumed that Ct had produced less than 99.99 percent
Inactlvatlon and If no animals were Infected, that it had produced greater
than 99.99 percent Inactlvatlon.9 If, however, 1-4 animals were Infected It
wis assumed that the level of viable cysts were 5 per animal and that 99.99
percent of the original cyst population had been Inactivated. Hlbler
Interpolated from the results and provided comprehensive tables showing C t
values at 0.5°C temperature Intervals.1 Because of observations Indicating
that C t values Increased as chlorine concentration Increased within the range
of chlorine concentrations used, Hlbler et al. advised against use of the C t.
values for chlorine concentrations above 2.5 mg/l.
Table 2 siumarlzes Hlbler's data for the different experimental
conditions examined. Column 3 shows the range of chlorine concentrations in
rog/L to which cysts were exposed before being fed to the gerblls, and Column 7
shows the number of experiments which yields 1-4 infected gerblls out of 5.
8
-------
Column 4 shows the range of cyst exposure tines and Column 5 contains the
range of C't values that are the product of the chlorine concentration and
cyst exposure tine.
TABLE 2. C'T VALUES FOR 99.99 PERCENT INACTIVATION BASED
ON ANIMAL INFECTIV1TY DATA
Range of
Range of
Cyst Exposure
Range of
Range of
Number of
Temp
Cone.
Time
C't values from
Predicted
Observa-
pH
°C
(»g/L)
(m1n)
Data
C't Values
tions
6
0.5
6.56-3.66
3$-366
113-263
136-lii
25
6
2.5
0.53-3.80
18-222
65-212
107-151
15
6
5
0.44-3.47
25-287
50-180
93-134
26
7
0.5
0.51-4.05
76-600
156-306
205-295
14
7
2.5
0.64-4.23
55-350
124-347
169-235
14
7
5
0.73-4.08
47-227
144-222
156-211
15
8
0.5
0.49-3.25
132-593
159-526
294-410
22
8
2.5
0.50-3.24
54-431
175-371
233-324
21
8
5
0.84-3.67
95-417
200-386
209-299
15
Hlbler's data set, based on animal 1nfect1v1ty, Is appealing because It
1s a more direct Indicator of cyst viability than data based on excystatlon.
However the C't values In this data set are based solely on 99.99 percent
1nact1vat1on. The other three data sets, based on excystatlon, have values
calculated for all four parameters In equation 8. Table 3 contains a summary
characterization of the studies on which these data sets were based. Because
no one Individual experiment provided the exact characteristics required for
this study an attempt was made to find the 'most consistent' set of data for
parameter estimation, which might Include several of the data sets discussed.
9
-------
TABLE 3. CHARACTERIZATION OF fi. IAMBI1A FREE CHLORINE*
1NACTIVATION STUDIES USED IN PREDICTIVE MODELS
Reference
No.
T
Cyst
Source
Viability
Assay
Comments
Symptomatic
human
excystatlon Conventional :
survival curves
based on Multiple
samples.. End
point - 0.1%
survival
gerbU infec- No survival curves,
tlvlty (I Endpolnt sought
animals/sample) - 0.01% survival
Gerbils, adapted
from Infected
humans. (CDC
Isolate)
Symptomatic and
nonsymptomatic
humans
Gerbils adapted
from Infected
humans. (Several
Isolates used)
excystatlon
excystatlon
Conventional
survival curves
based on multiple
samples. End
point - 0.1%
survival
Conventional
survival curves
based on multiple
samples. End
*Data provided by Dr. John Hoff formerly of USEPA
The Hlbler data set was Included In all combinations considered because
It was the largest data set, the data set was based on animal 1nfect1v1ty, and
the data reflected higher percent Inactivation than required under the SWTR.
Since the data based on excystatlon, with the exception of a few data points,
only reflected percent Inactivation up to 1 log or less than that required
under the SWTR, inclusion of the Hlbler data was considered essential for
-developing a model that could predict disinfection conditions for achieving
10
-------
99.9 percent Inictlvitlon with minimum uncertainty. Filtered systems will
need to know disinfection conditions for achieving less than 99.9 percent
1nact1vat1on. Therefore data from at least one of the excystatlon studies was.
considered essential since the C t values In the SVTR nay be used for
calculating partial Inactlvatlon levels (1.«.« less than 99.9 percent).
A fundamental question that needed to be addressed was the statistical
compatibility of the data sets. Initial regression estimates for each of the
data sets were made using equation 8.13 High *r'* were obtained for these
fits but significant differences were found for the coefficient or slope.
This Indicated that the basic model was adequate but that there were
differences 1n the coefficients as defined by the Individual estimates using
equation 8. It was decided to 'anchor* all of the data sets to the Hlbler
data set. The approach used was to construct an Indicator random variable to
move the regression Intercept or slope to compensate for data set
differences.19 The significance of the Indicator random variable would
support the hypothesis of different regression surfaces, I.e., Incompatibility
of the data sets chosen. The Indicator random variable was created In such a
way as to always differentiate between the Hlbler data set and other data sets
considered and to move the regression Intercept not the slope. The Indicator
random variable was defined as follows:
0 if Hlbler data
2-1 1 (9)
1 1f other data
Therefore equation 8 was modified as follows:
t - R I*Cb"lpHctempd10*x (10)
where t, I, C, pH, temp are defined as in equation 8, and R.a.btC.d.e are
constants determined from regression.
11
-------
Equation 10 cin be transformed as follows:
log t % log R ~ a log I ~ (b-1) log C * c log pH ~ d temp ~ ez (11)
:In aquation 11 when z - 0 equation 10 Is defined over the Hlbler data set,
and
t - R la C*"1 pH® tempd (12)
When z ¦ 1 equation 10 1s defined over the renalnlng data and
t - (R * 10*) Ia Cb_1 pH® teapd (13)
Table 4 displays the data set combinations and regression diagnostics. Note
that z 1s the indicator random variable.
In Table 4, the first column shows the various data sets considered In
the analysis. Column two contains the "r2" values based on equation 13 for
each of the data combinations. Column three indicates major results of the
analysis. For example it was found, for the first data set combination, that
the Intercept, and temperature variable were not significant. Column 4 shows
the test that was used to determine whether or not the equation yields biased
results.
As Indicated In Column 4 of Table 4 residual plots were used to determine
constant variance and normality. Fortunately a strict assumption of normality
1s not required. As stated In Neter, Wasserman and Whltmore 'Small departures
from normality do not create any serious problems.14 Major departures, on the
other hand, should be of concern*. Further they write, 'Unless the departures
from normality are serious, particularly with respect to skewness, the actual
confidence coefficients and risks of error will be close to the levels for
exact normality*. In addition because of the large sample size one would
expect the central limit theories would apply and symnetry would not be an
Issue.
12
-------
It was found that 90% of the data fell within plus/or/minus 1.64 standard
deviations of the Man. In addition 75% of the data fell within plus or minus
1 nlnus standard deviation which gives support for the normality assumption.
[For a perfect normal distribution we would expect 68% of the data to 11e
within plus or minus 1 standard deviation. Similarly, we would expect 90% to
He within plus or minus 1.64 standard deviation of the Man].
The Indicator random variable for the Intercept variable using the
Hlbler, Jarroll data base was not significant (p-value - 0.3372). All other
data bases considered had a significant Indicator random variable at the 0.095
level of significance. A formal test.for differences, of Intercept and/or
slope between the Hlbler and Jarroll data sets was conducted and no difference
was detected.
As mentioned previously the Hlbler data set does possesses some desirable
characteristics and 1t Is the largest data set among all data sets available.
However one might argue that by forcing the Hlbler data set into the analysis
the possibility has been Ignored that the other data sets may be mutually
consistent, and the Hlbler data set may represent an 'outlier*. In addition,
one might hypothesize that data from different experimental situations
prohibits us from making a reasonable comparison among these excystatlon
studies. Table 4 shows that the Hlbler and Jarroll data sets are compatible.
Since Table 4 also shows that H1bler-R1ce and H1bler-Rub1n 1s not consistent,,
then 1t Is reasonable to assume that the Jarroll date Is not consistent with
the R1ce and Rubin data so that the Hlbler data 1s not alone 1n being
Inconsistent with the other data sets. It seems reasonable therefore to start
with the Hlbler data set, the largest one, then Incorporate other smaller data
sets Into the modeling process. Thus logic supports the use of the Hlbler,
13
-------
Jarroll data base for extending the model development and the coefficients In
equation 8 Mere estimated using these data as shown In Table S In the log
transformed form.11
TABLE 4. DIAGNOSTIC RESULTS FROM DATA SET COMBINATION ANALYSIS
Data sets considered R-Souare Variables Plots
Hlbler, Rice, Jarroll, Rubin 0.6801 Intercept, temp non-normal data
not-significant non-constant var
Hlbler,
Rice, Jarroll, Rubin, z
0.7316
Intercept, temp
non-normal data
non-constint var
Hibler,
Rice, Rubin
6.6*49
Intercept, temp
not-significant
non-normal data
non-constant var
Hlbler, R1ce, Rubin, z
0.7899
Intercept
not-sl9n1f1cant
non-normal data
non-constant var
Hlbler,,
, Jarroll, Rubin
0.6424
Intercept, temp
not-significant
non-normal data
non-constant var
Hlbler, Jarroll, Rubin, z
0.6879
Intercept, temp
not-significant
non-normal data
non-constant var
Hlbler,
Rice, Jarroll
0.8619
all variables
significant
non-normal data
non-constant var
Hlbler,
Rice, Jarroll, z
0.865
all variables
significant
non-normal data
non-constant var
Hlbler,
Rubin
.0.6483
temp
not-significant
non-normal data
non-constant var
Hlbler,
Rubin, z
6.7*9*
intercept
not-s1gn1f1cant
non-normal data
constant var
Hlbler,
Rice
0.8548
all variables
significant
non-normal data
constant var
Hlbler,
Rice, z
0.8678
all variables
significant
non-normal data
constant var
Hlbler,
Jarroll
0.8452
all variables
significant
non-normal data
constant var
Hlbler, Jarroll, z 0.8459 2 not significant non-normal data
constant var
14
-------
TABLE 5. COEFFICIENT ESTIMATES FOR EQUATION 8.'
Statistical Analysis
Standard
T for HO:
Variance
Variable
DF
Coefficient
Error
Parameters
PROB > 0 |
Inflation
. Factor
INTERCEP
1
-0.902
0.200
-4.518
0.0001
0.000
L06I
1
-0.268
0.014
-19.420
0.0001
1.183
LOGCHLOR
1
•0.812
0.042
-19.136
0.0001
1.033
LOGPH
1
2.S44
0.221
11.535
0.0001
1.032
LOGTEMP
1
-0.146
0.028
-5.117
0.0001
1.179
In Table 5 column 7 entitled the 'Variance Inflation Factor (VIF)* 1s
defined as (l-R^) where R^2 1s the coefficient of multiple determination when
Is regressed on the other variables 1n the model. The minimum value of VIF
1s.1 1f there 1s no multlcolHnearlty. As shown 1n column 7 all of the
variance Inflation factors are close to one.
DISCUSSION OF MODEL
As discussed 1n the previous sections the coefficients for equation 8
were determined by a combination of log transformation and linear regression.
An Issue to consider 1s the probability that there 1s measurement error 1n the
model's Independent variables and the effect that this could have on estimates
of the parameters.
Regression 1s Intended to fulfill the dual purposes of prediction and
explanation. The purpose of equation 8 Is primarily to predict by providing
water utilities guidance as to what C t values will be needed for a desired .
level of 1nact1vat1on. The purpose of this model 1s to predict C't values and
will not be hampered by measurement error as long as consistency 1s
maintained.15 Since any measurement Is subject to some type of error, the
approach taken to deal with this Issue was to provide safe or 'conservative
estimates" of C't values.
15
-------
As one of the diagnostic procedures applied to the analysis equation 13
was evaluated for nultlcolInearlty. As can be see from Table 5 all of the
coefficients are highly significant and there Is no aultlcolInearlty.
TABLE 6. COLUNEARITY DIAGNOSTICS
Condition
VAR PROP
VAR PROP
VAR PROP
VAR PROP
VAR PROP
Number
Intercep
LOGI
10GCHL0R
LOGPH
LOGTEMP
1.000
0.0002
0.0031
0.0214
0.0003
0.0174
2.495
0.0001
0.0063
0.0138
0.0001
0.7833
2.801
0*0003
0*0067
0.9285
0.0004
0.0005
10.662
0.0147
0.9266
0.0029
0.0253
0.1918
45.636
0.9847
0*0574
0.0334
0*9739
0.0071
In Table 6 VAR PROP 1s the variance-decomposition proportion (VDP) and
has a maximum value of 1. A high condition number coupled with high VDP
values for two or more coefficients 1s an Indication of multicolllnearlty
between those variables. A condition of 45.636 In conjunction with an
Intercept VOP of 0.9847 and Log(pH) VDP of 0.9739 Indicated a dependency
between the Intercept and Log(pH) variable, however, multlcollInearlty among
the other coefficients were nonexistent.
The final equation used for predicting C't values In the SWTR was based
on equation 8 as follows:
C t - RI*CbpHctempd (14)
Confidence Intervals of the coefficients estimate for equation 14 based
on the Bonferronl method at a 99% confidence Interval are:14'"
R: ( 0.384, 0.4096}
a: (-0.2321, -0.3031)
b: ( 0.0792, 0.2977)
c: ( 1.9756, 3.1117)
d: (-0.2192, -0.0724)
16
-------
RESULTS
Thert in many uncertainties regarding the various data sets that night
be considered for calculating Ct values. The random variable analysis shows ;
the statistical Incompatibility among most of these data sets. More work
needs to be done to define the Impact of strain variation and in vivo versus
ill vitro techniques on C t values. In order to provide conservative estimates
for C t values In the SVTR and the guidance document the authors used the
approach Illustrated In Figure 1.
In Figure 1 the 99% confidence Interval of the 4 log 1nact1vat1on level
Is calculated. First order kinetics are then assumed so that the Inactlvatlon
'line* goes through 1 at Ct ¦ 0 and a C t value equal to the upper 99% con-
fidence Interval at 4 logs of Inactlvatlon. As can be seen the Inactlvatlon
line consists of higher C t values than all of the mean predicted Ct values
from equation 14, most of the Jarroll et al., and most of the Hibler data
points. Conservative Ct values, for a specified level of Inactlvatlon, can
be obtained from the Inactlvatlon line prescribed by the disinfection condi-
tions. For the example Indicated in Figure 1, the appropriate Ct for
achieving 99.9% Inactlvatlon would be 105. This approach (assumption of first
order kinetics) also provides the basis for establishing credits for sequen-
tial disinfection steps allowed under the SVTR. It should be noted that this
approach provides very conservative estimates at mid range levels of Ct.
Note in Figure 1 that some of the individual data points fall outside the
99% confidence Interval estimated at the four logs of Inactlvatlon. This Is
to be expected since the confidence Intervals constructed were for mean Ct
values, but also Indicated the high variability of the Hibler data.
Equation 14 was applied using the above strategy, as a safety factor, to
determine the Ct values for 99.9 percent inactlvatlon at 0.5°C and 5°C 1n the
17
-------
00
I
N
A
C
T
1
V
A
I
O
N
L
E
V
E
L
¦ Ct-PRED.
ACTUAL C»
A 99% CONF. INTERVAL
AT 4 LOGS OF INACTIOTION
1.0000
0.1000
0.0100
0.0010
0.0001
0.0000
20 40 60 80 100 120 140 160 180 200 220 240
Ct VALUES
FIGURE 1 99% CONFIDENCE LEVELS USING
EQUATION 14 FOR CHLORINE « 2 mg/I;
PH » 6; TEMPERATURE - 5°C
-------
final SWTR.1 C't values for temperatures above 5°C were estimated assuming a
twofold decrease for every 10°C Increase in temperature since all the Hlbler
data was generated at 5°C or less. This general principle Is supported by
Hoff.
Application of equation 14 to pHs above 8, up to 9, was considered
reasonable because the model Is substantially sensitive to pH (e.g., C'ts at
pH 9 are about three times greater than C'ts at pH 6 and about two times
greater than C'ts at pH 7). At a pH of 9. approximately four percent of the
hypochlorous acid fraction of free chlorine Is still present. Other data
Indicate that 1n terms of H0C1 residuals (versus total free chlorine residuals
Including H0C1 and 0C1") the C't values required for Inactlvatlon of filardla
muHs and Slardla Iambi 1a cysts decrease with Increasing pH from 7 to 9.10
However, with Increasing pH, the fraction of free chlorine existing as the
weaker oxidant species (OCT) Increases. In terms of total free chlorine
residuals (I.e., HOC! and 0C1") the C't values required for Inactlvatlon of
Slardla tnurls and Slardla lamblla cysts Increase with Increasing pH from 7 to
9 but generally less than by a factor of 2 at concentrations of less than 5.0
mg/L.10 Table 7 compares the C't values In the proposed SVTR to those given
1n the SWTR. The C't values In the proposed SWTR were based only on the
Hlbler data and Included different safety factors.2'*
19
-------
TABLE 7. COMPARISON BETWEEN MODIFIED APPROACH (MEANS) AND RULE C TS
AT 19.9% INACTIVATION AND 5°C IN THE PROPOSED AND FINAL SWTR.
M.
Concentration 6 7 8 f
wq/i Proposed Final Proposed Final Proposed Final Proposed Final
1 105 108 149 16| 211 238 329 312
2 116 122 161 186 243 269 371 353
The Ct values in the final SWTR are 0-10 percent lower than in the
proposed SWTR. Table 8 presents representative C t values determined by
application of the above described approach.
TABLE 8. CALCULATED C T VALUES FOR GIARDIA INACTIVATION
USING USING EQUATION 14 AT 0.5°C and 5°
Values for Inactlvatlon of Giardia Cysts
by Free Chlorine at 0.S°C
£^on-ne pH ¦ 6 pH - 7 pH ¦ 8 pH ¦ 9
Concentration jgf IftlStiyitfgn in IPICtlvitlPH Log Inactlvatlon Log Inactlvatlon
Btg/L O.S 1.0 2.0 3.0 O.i 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.8
0.4 23 46 91 137 33 65 130 195 46 92 185 277 65 130 260 390
1 25 49 99 148 35 70 140 210 51 101 203 304 73 146 291 437
2 28 55 110 165 39 79.157 236 58 115 231 346 83 167 333 500
3 30 60 121 181 44 87 174 261 64 127 255 382 92 184 368 552
ya1ues for inactlvatlon of Giardia Cysts
by Free Chlorine at 5°C
£l^or^ne pH ¦ 6 pH • 7 pH - 8 pH ¦ 9
Concentration Log Inactlvatlon Uf Inactivity log InstttYltfCft in Iwt1vat1pn
mg/L 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.8
0.4 16 32 65 97 23 46 93 139 33 66 137 198 47 93 186 279
1 18 35 70 105 25 50 99 149 36 72 144 216 52 104 208 312
2 19 39 77 116 28 55 110 165 41 81 162 243 59 118 235 353
3 21 42 84 126 30 61 121 182 45 89 179 268 65 130 219 389
20
-------
4
Because calculations for the SVTR C't values art the upper Unit on the
error bounds associated with equation 14 (Table 8), an equation was developed
to estimate these C't values for 0.5 and S°C directly. C't values above 5°C
can be estimated by using the Mthod given below to estimate C't values at
5°C, then the assumption that there Is a twofold decrease In C't values for
every 10°C Increase In temperature can be applied. The equation for the
estimated C't values at 0.5 and 5°C Is as follows:
C t - 0.36 pH*-89temp"0'15C°'15(-log I)1'00 (R* - 0.998) (15)
where the variables In equation 15 are as defined previously.
Table 9 compares the values estimated by equation 15 and the SVTR values
shown In Table 8.
TABLE 9. CALCULATED CT VALUES FOR GIARDIA I MOTIVATION
USING EQUATION 15 AT 0.5 AND 5°C
Values for Inactlvation of G1ard1a Cysts
bv Free Chlorine at O.S°C
Chlorine pH-6 pH-7 pH-8 pH-9
Concentration Log InKtlVltlOn Log IniCtlYitlon too Inactlvation Loo Inactlvation
mg/L 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.:
o 22 4* 66 12S 33—65 131 196 47—$4 167 261 £4 124 257 is
1 25 49 99 148 37 75 149 224 54 107 214 321 74 147 294 44
2 • 27 55 109 164 41 83 165 248 59 118 137 355 81 163 325 48
3 29 58 116 174 44 88 175 263 63 126 251 377 86 173 345 51
Values for Inactlvation of Glardla Cysts
bv Free Chlorine at 50C
Chlorine pH • 6 pH - 7 pH ¦ 8 pH - 9
Concentration Loo Inactlvation Lgg iMCtlVltlW Log Inactlvation Log
mg/L 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.0 0.5 1.0 2.0 3.1
0 n—51—SI—51 —92 136 33—66 132 196 46 *1 182 27
1 17 35 70 104 26 53 106 158 38 76 151 227 52 104 208 31
2 19 39 77 116 29 58 117 175 42 84 167 251 58 115 320 34
3 20 41 82 123 31 62 124 186 44 89 178 266 61 122 244 36
21
-------
FUTURE WORK .
Because of the tiportance from an econonic and a public health viewpoint
of the calculation of C t values for the Inactlvatlon of 61ard1a Iambi la by
free chlorine* auch effort has been expended In developing nodels that.
Interrelate the important variables effecting these values.* The work
reported 1n this paper reflects the authors attempts to develop such t
relationship for Inclusion In the StfTR. However, It also raises a-very
Interesting point regarding the application of statistical aethodology to
public policy decision problems. There Is no perfect 'regulatory* experiment
that answers all of the textbook questions that could be raised regarding
regulatory decision Baking. One has to use available data and Incorporate the
best judgment that can be brought to bear on a given Issue to insure that
public health and welfare Is protected. The need to combine data sets from
different investigations and then develop a decision rule based on the data,
as shown in this paper, as an example of the this process.
There is no doubt In the authors' Bind that other better models may be
developed. For example, Haas' work in applying the Horn model to Inactlvatlon
data and incorporating the Bethod of Maximum Likelihood for estimating
parameters 1s promising.17 The authors believe that the public is best served
by examining problems from many different points of view and encourage others
to pursue these difficult, frustrating but extremely challenging problems.
SUMMARY AND CONCLUSIONS
Amendments to the Safe Drinking Water Act clearly require that all
surface water suppliers In the U.S. filter and/or disinfect to protect the
health of their customers, fi. Iambi 1a has been Identified as one of the
.leading causes of waterborne disease outbreaks 1n the U.S. £. Iambi 1a cysts
are also one of the most resistant organisms to disinfection by free chlorine.
22
-------
EPA's Office of Drinking Water has adopted the Ct concept to quantify the
Inactlvatlon of fi. limblla cysts by disinfection. If a utility can assure
that a large enough Ct can be maintained to ensure adequate disinfection
then, depending upon site specific factors, It say not be required to install
filtration. Similarly, the C t concept can be applied to filtered systems for
determining appropriate levels of protection. *
In this paper, an equation has been developed that can be used to predict
C t values for the Inactlvatlon of fi. UmbHa by free chlorine based on the
Interaction of disinfectant concentration, temperature, pH, and Inactlvatlon
level. The parameters for this equation have been derived from a set of
animal 1nfect1v1ty and excystatlon data. The equation can be used to predict
Ct values for achieving O.S to 4 logs of inactlvatlon, within temperature
ranges of 0.5 to 5°C, chlorine concentration ranges up to 4 mg/L, and pH
levels of 6 to 8. While the model was not based on pH values above 8, the
model Is still considered applicable up to pH level of 9. The equation shows
the effect of disproportionate Increases of C t versus Inactlvatlon levels.
Using 99% confidence Intervals at the 4 log Inactlvatlon levels and applying
first order kinetics to these end points a conservative regulatory strategy
for defining C t at various levels of Inactlvatlon has been developed. This
approach represents an alternative to the regulatory strategy previously
proposed.
23
-------
ACKNOWLEDGMENTS
The authors Mould like to acknowledge Ms. Patricia Plerson and Ms. Diane
Routledge for their assistance In preparing this Manuscript. The authors are
grateful to Dr. John Hoff, formerly of USEPA, Ms. Shirley Plen and Ms. Eleanor
Read of the Computer Sciences Corporation, Mr. Dennis Black of the University
of Nevada, Las Vegas, and Dr. Charles Haas of the Illinois Institute of
Technology for their review and suggestions to Inprove the Manuscript. The
authors would like to extend a special acknowledgenent to Ms. Dlanne Wild for
her assistance In the preparation of this aanuscrlpt.
24
-------
REFERENCES
1. National Primary Drinking Water Regulations: Filtration, Disinfection,
Turbidity, 61ird1l lMlil. Viruses, Legionella. and Heterotrophic
Bacteria. Final Rule, 40 CFR parts 141 and 142. Fed. Reg. 54:124:27486
(June 29, 1989).
2. U.S. Environmental Protection Agency, Office of Drinking Water, Criteria
and Standards Division. Guidance Manual for Compliance with the Filtra-
tion and Disinfection Requirements for Public Water Systems Using Surface
Water Sources, October 1987.
3. Clarke, N. A., Berg, C., Kabler, P. W., and Chang, S. L., "Human Enteric
Viruses In Water: Source, Survival, and Removability". International
Conference on Water Pollutions Research, Landar, September, 1962.
4. Hoff, J. C., R1ce, E. W., and Schaefer III, F. W. "Disinfection and the
Control of Waterborne Giardiasis', In Proceedings of the 1984 Specialty
Conference, Environmental Engineering Division, ASCE, June 1984.
5. . Jarroll, E. L., Bingham, A. K., and Heyer, E. A. "Effect of Chlorine on
GUrdli Umblli Cyst Viability". Applied and Environmental Microbiology.
Vol. 41, pp. 483-48, February, 1981.
6. Chick, H., "an Investigation of the Laws of Disinfection:, J. Hygiene, 8,
92 (1908).
7. Watson, W. E., "A Note on the Variation of the Rate of Disinfection with
Change 1n the Concentration of the Disinfectant", J. Hygiene, 8, 536
(1908).
8. Clark, R. M., Read, E. J., and Hoff, J. C. "Analysis of Inactlvatlon of
G1ard1a Lamb!1a bv Chlorine*. Journal of Environmental Engineering, ASCE,
Vol. 115, No. 1, February, 1989, pp. 80-90.
9. Hlbler, C. P., Hancock, C. M., Perger, L. M., Wegrzyn, J. G. and Swabby,
K. D. "Inactlvatlon of Glardla Cysts with Chlorine at 0.5°C to 5.0°C.
amerlcan Water Works Association Research Foundation, 6666 West Qulncy
Avenue, Denver, Colorado 80235, 1987.
10. Rice, E. W., Hoff, J. C. and Schaefer III, F. W. "Inactlvatlon of Glardla
Cysts by Chlorine", Applied and Environmental Microbiology. Jan. 1982,
Vol. 43, No. 1, pp. 250-251.
11. Rubin, A. J., Evers, D. P., fyman, C. M., and Jarroll, E. I.,
"Inactlvatlon of Gerbll-Cultured Glardla Lamb!1a Cysts by Free Chlorine",
Applied and Environmental Microbiology. Oct. 1989, Vol. 55, No. 10, p. .
2592-2594.
12. Hoff, J. C., "Inactlvatlon of Microbial Agents by Chemical Disinfectants"
EPA/600/2-86-067.
13. Draper, N. and Smith, H. (1981) Second Edition, Applied Regression
Analysis. Wiley: New York.
25
-------
14. Ntter, J. and Vasseraan, ¥. (1974), Applied Llneir Statistical Models,
Irwin: Honewoed, IL.
15. Full«r, Wayne, Measurement Error Models. John lilley I Sons, 198?..
16. Belsley, D. A., Kuh, •. and Welsch, R. E. (1980), RMrtHlOfl PliWOrtlC?,
Wiley: New York.
17. Haas, Charles W., and Millar, B. Statistical Analysis of Data on
Chlorine Inactlvatlon of Slirdii JjaULii, Final Report prepared for U.S.
IPA Office of Drinking Water, January 6, 1988.
26
-------
GLOSSARY
«
d Nt/dt - rate of change of organisms with respect to time
k • Inactlvatlon rate 1n minutes*1 , .
t • tine In Minutes
Nt - number of organisms at tine t
H9 • number of organism at time 0
r » coefficient of specific lethality (liters/milligram - ainutes)
C - concentration of disinfectant [milligrams/liter]1'"
n • coefficient of dilution
K • constant at given level temperature, pH and inactlvatlon level
pH « pH 1n water phase
temp ¦ temperature in °C
I • level of inactlvatlon
C t * concentration in mg/L times time In minutes
R - coefficient to be determined
a - coefficient to be determined
c • coefficient to be determined
d - coefficient to be determined
e » coefficient to be determined
z - coefficient to be determined
VIF ¦ variance inflation factor. If VIF is 1 there is no multicollnearUy
VDP * variance decomposition number. If VDP is high for two or more
variables there is an Induction of *ultico11nearity between
variables
Bonferroni technique • a conservative method of estimating confidence
Intervals
27
-------
permanganate as a disinfectant. It 1s not meant to be used as a basis for
establishing CT requlrenents.
References
Berman, D.; Hoff, J. Inactlvatlon of Simian Rotavirus SA 11 by Chlorine,
Chlorine Dioxide and Monochloramlne. Appl. Environ. Microbiol.,
48:317-323. 1984.
Chang, J.C.H.; Ossoff, S.F.; Lobe, D.C.; Dorfman, M.H.; Dunals, C.M.;
Quails, R.G.; Johnson, J.O. Inactlvatlon of Pathogenic and Indicator
Microorganisms. Applied Environ. Micro., June 1985, pp. 1361-1365.
Clark, R.M.; Read, E.J.; Hoff, J.C. Inactlvatlon of Giardia Iambi 1a by
Chlorine: A Mathematical and Statistical Analysis. Unpublished Report,
EPA/600/X-87/149, DWRD, Cincinnati, OH, 1987.
Clark, R.; Regli, S.; Black, D. Inactlvatlon of G1ard1a Iambi1a by Free
Chlorine: A Mathematical Model. Presented at AWWA Hater Quality
Technology Conference. St. Louis, Mo., November 1988.
Clark, M.R.; Regli, S. The Basis for Glardla CT Values 1n the Surface
Water Treatment Rule: Inactlvatlon by Chlorine. Submitted to Journal of
Water Supply Research and Technology-Aqua, August 1990.
Hibler, C.P.; C.M. Hancock; L.M. Perger; J.G. Wegrzn; K.D. Swabby Inactl-
vatlon of Glardla cysts with Chlorine at 0.5 C to 5.0 C American Water
Works Association Research Foundation, In press, 1987.
Hoff. J.C. Inactlvatlon of Microbial Aoents bv Chemical Disinfectant!;.
EPA/600/52-86/067, U.S. Environmental Protection Agency, Water Engineering
Research Laboratory, Cincinnati, Ohio, September, 1986.
Jarroll, E.L.; A.K. Binham; E.A. Meyer Effect of Chlorine on Giardia
lamb!ia Cvst Viability. Appl. Environ. Microbiol., 41:483-487, 1981.
Leahy, J.G.; Rubin, A.J.; Sproul, O.J. Inactlvation of Giardia murls
Cysts by Free Chlorine. Appl. Environ. Microbiol., July 1987.
Rice, E.; Hoff, J.; Schaefer, F. Inactlvatlon if Giardia Cysts by
Chlorine. Appl. and Environ. Microbiology, 43:250-251, January 1982.
Roy, 0., R.S. Engelbrecht, and E.S.K. Chian. Comparative Inactivation of
Six Enteroviruses by Ozone. J. AWWA, 74(12):660, 1982.
Rubin, A. Factors Affecting the Inactivation of Giardia Cysts by Mono-
chloramine and Comparison with other Disinfectants. Water Engineering Re-
search Laboratory, Cincinnati, OH March 1988a.
F-8
-------
Rubin, A. *CT Products for .the Inact1vat1on of 61ard1a Cysts by Chlorine,
Chloramlne, Iodine, Ozone and Chlorine Dioxide" submitted for publication
In J. AWWA, December, 1988b.
Sobsey, M. Detection and Chlorine Disinfection of Hepatltus A 1n Hater.
CR-813-024. EPA Quarterly Report. December 1988.
Vaughn, J.; Chen, Y.j Undburg, K.; Morales, D. Inact1vat1on of Hunan and
Simian Rotaviruses by 02one. Appl. Environ. Microbiol., 53(9)$2218-2221,
September 1987.
Hlckramanayake, G.; Rubin, A.; Sproul, 0. Effects of Ozone -and Storage
Temperature on Slardla Cysts. J.AHHA, 77(8):74-77, 1985.
Yahya, M.T., Landeen, L.K., Forsthoefel, N.R., Kujawa, K., and Gerba, C.P.
Evaluation of Potassium Permanganate for Inactlvatlon of Bacteriophage
MS-2 1n Hater Systems. Copyright 1988, Cams Chemical Company, Ottawa,
Illinois.
F-9
-------
APPENDIX G-l
DETERMINING CHLORAMNE INACTIVATIOTI OF SflARDIA
FOR THE SURFACE WATER TREATMENT RULE
Microbiological Treatment Branch
Risk Reduction Engineering Laboratory
and
Parasitology and Immunology Branch
Environmental Vonltorlng Systens Laboratory
U.S. Environmental Protection Agency
26 West Martin Lutfjer King Drive
Cincinnati, Ohio 45268
-------
2
TABLE OF CONTENTS
1. Materials ....3
II. Reagents 4
III. S1ard1a murls Assay 7
IV. Disinfection Procedures for Glardla 10
V. Procedure for Determining Inact1vat1on 12
VI. Bibliography 13
VII. Technical Contacts 14
Appendix
A. Use of the Hemocytometer 15
B. Preparation and Loading of Chamber Slides 20
-------
3
The Surface Water Treatment Rule requires 99.91 or greater removal/
1nact1vat1on of Slardli. The following protocol may be used to determine
the percentage of Glardla Inactlvatlon obtained by a treatment plant
using chloramlne disinfection.
I. MATERIALS
A. Materials for Disinfection
1. Stock chlorine solution
2. Stock ammonia solution
3. Stirring" device
4. Incubator or Mater bath for temperatures below ambient
5. Mater from treatment plant
6. Glardla murls cysts
7. Assorted glassware
8. Assorted pipettes
9. Reagents and instruments for determining disinfectant residual
10. Sterile sodium thlosulfate solution
11. Vacuun filter device, for 47mm diameter filters
12. 1.0 urn pore size polycarbonate filters, 47 mm diameter
13. Vacuum source
14. Crushed 1ce and Ice bucket
15. Timer
3. Materials for Excystatlon
1. Exposed and control Slardla murls cysts
2. Reducing solution
3. 0.1 M sadlum bicarbonate
4. Trypsln-Tryode's solution
5. 15 ml conical screw cap centrifuge tubes
6. Water bath, 37°C
7. Warn air Incubator or slide warming tray, 37°C
8. Aspirator flask
9. Vacuum source
10. Assorted pipettes
11. Vortex mixer
12. Centrifuge with swinging bucket rotor
13. Chamber slides
14. Phase contrast microscope
15. Differential cell counter
16. Timer
-------
REAGENTS
A. Reducing Solution
Ingredient . Amount
glutathione (reduced form) 0.2 g •
L-cyste1ne-HCl 0.2 g
IX Hanks' balanced salt solution 20.0 ml
Dissolve the dry Ingredients 1n the IX Hanks' balanced salt
solution and warm to 37*C before use 1n the experiment.
Prepare fresh, w1thin 1 hour of use.
3. Sodium Bicarbonate Solution, 0.1 M
Ingredient Amount
Soalum bicarbonate : 1 0.42 g
Dissolve the salt 1n 10 to 15 ml distilled water. Adjust
the volume to 50 ml with additional distilled water and
warm to 37®C before use 1n the experiment. Prepare fresh,
within 1 hour of use.
C. Sodium Bicarbonate Solution, 7.51
Ingredient Amount
Soaium bicarbonate 7.5 g
Dissolve the sodium bicarbonate 1n 50 ml distilled water.
Adjust the volume to 100 ml with additional distilled
water. Store at room temperature.
D. Sodium Thlosulfate Solution, 105
Ingredient Amount
Soalum thlosul fate 10.0 g
Dissolve the sodium thlosulfate 1n 50 ml distilled water.
Adjust the volume to 100 ml with additional distilled
water. Filter sterilize the solution .. through a 0.22 um
porosity membrane or autoclave for 15 minutes at 121°C.
Store at room temperature.
-------
5
E. Tyrode's Solution, 20X
Ingredient Amount
wTi : rarrg
KC1 4.0 g
CaCl? - 4,0 9
MgCl2*6H,0 2.0 g ,
NaHjpOj^rUQ 1.0 g
Glucose 20.0 g
Dissolve the dry Ingredients in the order listed 1n 750 ml
distilled water. Adjust the volume to 1.0 liter with addi-
tional distilled Mater. If long tern storage {up to 1
year) Is desired, filter sterilize the solution through a
0.22 um porosity membrane.
F. Tyrode's Solution, IX
Inqredlent Amount
25X Tyrode's solution 0~mT
Dilute 5 ml of the 20X Tyrode's solution to a final volume"
of 100 ml with distilled water.
S. Trypsln-Tyrode's Solution
Inqredlent Amount
Trypsin, 1:160, U.S. biochemical to. OFg~
HaHCOt 0.15 g
IX Tyrode's solution 100.00 ml
With continuous mixing on a stlrplate, gradually add 100 ml
IX Tyrode's solution to the dry ingredients. Continue
stirring until the dry ingredients are completely dissolved.
Adjust the pH of the solution to 8.0 with 7.55 NaHC0i%
Chill the trypsin Tyrode's solution to 4°C. NOTE: Trypsin
lots must be tested for their excystatlon efficiency.
Prepare fresh, within 1 hour of use.
H. Polyoxyethylene Sorbltan Konolaurate (Tween 20) Solution, 0.01%
(v/v)
Ingredient Amount
tween 26 VTmT
Add the Tween 20 to 1.0 liter of distilled water. Mix
well.
-------
s
I. Vaspar
Ingredient ¦ Amount
Paraffin 1 part
Petroleum jelly 1 part
Heat the two Ingredients 1n a boiling Mater bath until melt-
ing and mixing Is complete.
-------
GIARDIA HURTS ASSAY
A. Cysts
Glardla murls cysts may be available from commercial sources.
The cysts may be produced 1n Mongolian gerblls (Herlones ungulcu-
latus) or 1n mice. Hus musculus. the laboratory mouse, CF-1,
BALSc, and C3H/he strains have been used to produce J». murls
cysts. The method 1s labor Intensive and requires a good animal
facility.
In order for the disinfection procedure to work properly, the jj.
murls cysts used must be of high quality. Evaluation of a cyst
suspension 1s a subjective procedure involving aspects of morpho-
logy and microbial contamination as well as excystment. A good
quality murls cyst preparation should exhibit the following:
1. Examine cyst stock suspension microscopically for the presence
of empty cyst walls (ECU). Cyst suspensions containing equal
to or greater than 1% ECU should not be used for determining
1nact1vat1on at any required level. However, 1f a 99.95
level of disinfection 1nact1vat1on Is required, the stock
cyst suspension must contain <0.15 ECU.
2. Excystatlon should be 90S or greater. •
3. The cyst suspension should contain little or no detectable
microbial contamination.
4. Good Jj. murls cysts are phase bright with a defined cyst wall,
perltrophlc space, and agranular cytoplasm. Cysts which are
phase dark, have no detectable perltrophlc space, and have a
granular cytoplasm may be non-viable. There generally should
be no more than 4 to 5X phase dark cysts 1n the cyst prepara-
tion.
Good G. murls cyst preparations result when the following
guidelines are followed during cyst purification from feces:
a. Use feces collected over a period of 24 hours or less.
b. The Isolation of the cysts from the feces should be done
immediately after the fecal material 1s collected.
c. Initially, G. murls cysts should be purified froni the
fecal material by flotation using 1,0 M sucrose.
d. If the G. murls cyst suspension contains an undesirable
¦dens1ty~of contaminants after the first sucrose float,
further purification Is necessary. Two methods for
further purification are suggested.
1) Cysts may be reconcentrated over a layer of 0.85 M
sucrose 1n a 50 mi conical centrifuge tube. If this
-------
3
second exposure to sucrose 1s not done quickly, high
cyst losses can occur due to their Increased bouyant
density In the hyperosmotic sucrose medium. The
cysts must be thoroughly washed free of the sucrose
Immediately after collection of the Interface.
2) Cysts can be separated from dissimilar sized contami-
nants by sedimentation at unit gravity, which will
not adversely affect cyst bouyant density, morphology,
or viability.
B. Maintenance of Cysts
1. Preparation of stock suspension
Deternlne the suspension density of the G. murls cyst prepara-
tion using a hemocytometer (see Appendix"A). Adjust the cyst
suspension density with distilled water to approximately 3-5
x 10® cysts/ml.
2. Storage
Store cysts 1n distilled water 1n a refrigerator at 4#C.
Cysts should not be used for disinfection experiments 1f they
are >nore than 2 weeks old (from time of feces deposition).
C. Excystatlon Assay
A number of S. murls excystatlon procedures have been described In
the scientific literature (see B1bllography, Section VI). Any of
these procedures may be used provided 90S or greater excystatlon
of control, undlslnfected £. murls cysts Is obtained. The
following protocol Is used to" evaluate the suitability of cysts 1n
the stock suspension, and to determine excystatlon 1n control and
disinfected cysts.
1. For evaluating a cyst suspension or for running.an unexposed
control , transfer 5 x 10* G. murls cysts fron the stock
preparation to a 15 ml conical screw cap centrifuge tube. An
unexposed control should be processed at the.same time as the
disinfectant exposed cysts.
2. Reduce the volume of G. murls cyst suspension 1n each 15 ml
centrifuge tube to 0.5^ml or less by centrlfugatlon at 400 x
g for 2 minutes. Aspirate and discard the supernatant to no
less than 0.2 ml above the pellet.
3. Add 5 ml reducing solution, prewarmed to 37#C, to each tube.
4. Add 5 ml 0.1 M NaHC03, prewartned to 37eC, to each tube. NOTE:
Tightly close the caps to prevent the loss of COj. If the
CO2 escapes, excystatlon will not occur.
5. Mix the contents of each tube by vortexlng and place in
a 37°C water bath for 30 minutes.
-------
9
6. Remove the tubes' from the water bath and centrifuge each for
2 minutes at 400 x g.
7. Aspirate and discard the supernatant to no less than 0.2 ml
above the pellet*and resuspend the pellet In each tube 1n 10
ml trypsln-Tyrode's solution chilled to 4°C.
8. Centrifuge the tubes for 2 minutes at 400 x g.
9. Aspirate and discard the supernatant to no less than 0.2 ml
above the pellet.
10. Add 0.3 ml trypsln-Tyrode's solution, prewamed to' 378C, to
each tube. . Resuspend the £. murls cysts by low speed vortex-
1ng. ~
11. Prepare a chamber slide for each tube (see Appendix B).
12. Seal the coversllp on each chamber slide with melted vaspar
and Incubate at 37®C for 30 minutes 1n an Incubator or on a
si1de warmer.
13. After Incubation, place a chamber slide on the stage of an
upright phase contrast mlc'roscope. Focus on the slide with a
low power objective. Use a total magnification of 400X or
more for the actual quantitation. NOTE: Be careful to keep
the objectives out of the vaspar.
14. While scanning the slide and using a differential cell coun-
ter, enumerate the number of empty cyst walls (ECW), partial-
ly excysted trophozoites (PET), and Intact cysts (IC) observed
(see Section V for a further description of these forms and
the method for calculating percentage excystatlon). If the
percentage excystatlon In the stock suspension Is not 90" or
greater, do not continue with the disinfection experiment.
-------
10
IV. DISINFECTION PROCEDURES FOR GIARDIA
A. The treatment plant water to be used should be the water Influent ¦
Into the chloramlne disinfection unit process used 1n the plant. , \
If chloramlne disinfection 1s performed at more than one point In
the treatment process, e.g., preflltratlon and postflltratlon,
the procedure should simulate as closely as possible actual
treatment practice.
B. Prepare stock ammonia and chlorine solutions to be added to the
treatment plant water to achieve the same stoichiometric relation-
ship between chlorine and ammonia that Is used 1n the water
treatment plant. These solutions should be concentrated enough
so that no more than 2 ml of each solution will be added to the
treatment plant water being disinfected.
C. Determine the contact time by the methods described In the Surface
Water Treatnent Rule and/or the associated Guidance Manual.
0. Rinse a 600 ml beaker with treatnent plant water to remove any
extraneous material that may cause disinfectant demand. Then
add 400 ml treatment plant water to the beaker.
E. Mix the contents of the beaker short of producing a vortex 1n the
center and continue until the conclusion of the experiment.
F. Equilibrate the 600 ml beaker and Its contents as well as the dis-
infectant reagents to the desired experimental temperature.
G. Adjust the stock J3. murls cyst suspension with distilled water so
that the concentration 1s 2-5 x 10® cysts/ml.
H. Add 0.5 ml of the adjusted cyst suspension to the contents of the
600 ml beaker.
1. Add the disinfectant reagents to the beaker using the same rea-
gents, the same sequence of addition of reagents, and the same
time Interval between addition of reagents that 1s used 1n the
disinfection procedure In the treatment plant.
J. Just prior to the end of the exposure time, remove a simple ade-
quate for determination of the disinfectant residual concentra-
tion. Use methods prescribed 1n the Surface Water Treatnent 3uTe
for the determination of combined chlorine. This residual should
be the sane (±205) as residual present 1n the treatment plant
operation.
K. At the end of the exposure time, add 1.0 ml 10% sodium thlosulfate
soljtlon to the contents of the 600 ml beaker.
L. Concentrate the G. murls cysts 1n the beaker by filtering the
entire contents through a 1.0 urn porosity 47 mm diameter polycar-
bonate filter.
-------
u
N. Place the filter, cyst side up, on the side of * ISO ml beaker.
Add 10 nil 0.01S Tween" 20 solution to the beaker. Using a Pasteur
pipette, nash the JS. wurls cysts from the surface of the filter
by aspirating and expel ling the O.OH Tween 20 solution over the
surface of the filter.
N, Transfer the contents of. the 150 ml beaker to an appropriately
labeled 15 ml screw cap conical centrifuge tube,
0. Keep the tube on crushed Ice until the excystatlon assay Is
performed (see Section III, C) on the disinfectant exposed cysts
and on an unexposed control preparation obtained fron -the stack
cyst suspension.
-------
12
PROCEDURE FOR DETERMINING IMACTIVATIQM
A. Glardla murls Excystatlon Quantitation Procedure
The percentage excystatlon Is calculated using the following for-
mula:
• excystatlon ¦ .ECW,*J* 10° •
ECU ~ PET ~ IC
where ECU 1s the number of empty cyst walls,
PET 1s the number of partially excysted trophozoites, and
IC Is'the number of Intact cysts.
An ECU 1s defined as a cyst wall which Is open at one end and 1s
completely devoid of a trophozoite. A PET Is a cyst which has
started the excystatlon process and progressed to the point where
the trophozoite has either started to emerge or has completely
emerged and Is still attached to the cyst wall. An IC 1s a
trophozoite *h1ch 1s completely surrounded with a cyst wall
showing no evidence of emergence. For the control, generally 100
forms are counted to determine the percent excystatlon.
The number of cysts that must be observed and classified (ECW,
PET, IC) 1n the disinfected sample 1s dependent on the level of
1nact1vat1on desired and on the excystatlon percentage obtained
1n the control ,
For 0.5, 1 and 2 logig reductions, (681, 90? and 99S 1nact1-
vatlon, respectively), the minimum number of cysts to be
observed and classified 1s determined by dividing 100 by the
percentage excystatlon (expressed as a decimal) obtained 1n
the control.
For a 3 log^Q reduction (99.9X 1nact1vat1on) the minimum
number of cysts to be observed and classified 1s determined
by dividing 1,000 by the percentage excystatlon (expressed
as a decimal) obtained 1n the control.
B. Determining Inact1vat1on
The amount of 1nact1vat1on 1s determined by comparing the percent-
age excystatlon of the exposed cyst preparation to the percentage
excystatlon 1n the control preparation using the following for-
mula:
I 1nact1vat1on ¦ 100X - [(exposed Z excysted/control 2 excysted) x 100]
If the percentage excystatlon In the exposed preparation 1s zero.
I.e., only IC (no ECW or PET) are observed and counted, use <1 as
the value for "exposed i excysted" 1n the formula for.calculating
X Inactlvatlon.
-------
13
BIBLIOGRAPHY
American Public.Health Association; American Water Works Association;
Water Polutlon Control Federation. Standard Methods for the Examlna-
tlon of Mater and Wastewater. 16th ed. (1985).
Belosevlc, .1. & G.H. Faubert. G1ard1a murls: correlation between
oral dosage, course of Infection, and.trophozoite distribution 1n the
mouse small Intestine. Exp. Parasltoi., 56:93 (1983).
Erlandsen, L.S. and E.A. Meyer, Glardla and Giardiasis. Plenum
Press, Hew York, (1984).
Faubert, G.M. et al. Comparative studies on the pattern of infec-
tion with G1ard1a spp. 1n Mongolian gerblls. J. Parasltoi.. 69:802
(1983).
Feely, D.E. A slmpHfed method for In vitro excystatlon of Glardla
muHs. J. Parasltoi., 72:474-475 (195T).
Feely, D.E. Induction of excystatlon of Glardla murls by C0;>. 62nd
Annual Meeting of the American Society of Parasitologists, Lincoln,
Nebraska, Abstract No. 91 (1987).
Gonzalez-Castro, J., Bermejo-Vlcedo, M.T. and Palaclos-Gonzalez, F.
Desenau1stan1ento y cultlvo de Glardla murls. Rev. Iber. Parasltoi.,
46:21-25 (1986).
Melvln, D.M. and M.M. Brooke. Laboratory Procedures for the Diagnosis
of Intestinal Parasites. 3rd ed., HHS Publication No. (CDC) 82-8282
(1982).
M1ale, J.B. Laboratory Medicine Hematology, 3rd ed. C. V. Mosby
Company, St. Louis, Missouri (1967).
Roberts-Thomson, I.C. et al.. Giardiasis 1n the mouse: an animal
model. Gastroenterol., 71:57 (1976).
Sauch, J.F. Purification of Glardla murls cysts by velocity sedi-
mentation. Appl. Environ. Microbiol., 48:454 (1984) .
Sauch, J.F. A new method for excystatlon of Glardla. Advances 1n
Glardla Research. University of Calgary, Calgary, Canada (In Press).
Schaefer, III, F.W., R1ce, E.W., & Hoff, J.C. Factors promoting
In vitro excystatlon of Glardla murls cysts. Trans. Roy. Soc. Trop.
fiedTHyg., 78:795 (1984).
-------
TECHNICAL CONTACTS:
A. Eugene U. Rice
Microbiological Treatment Branch
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
26 Uest Martin Luther King Drive
Cincinnati, Ohio 45268
Phone: (513) 569-7233
B. Frank W. Schaefer, III
Parasitology and Immunology Branch
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
26 Uest Martin Luther King Drive
Cincinnati, Ohio 45268
Phone: (513) 569-7222
-------
15
Appendix A: Use of the Hemocytometer
Suspension Density Determination Using the Improved Neubauer (Bright-line)
Hemocytometer
The hemocytometer consists of two chambers separated by a transverse
trench and bordered bilaterally by longitudinal trenches. Each chamber
1s ruled and consists of nine squares, each 1 x 1 x 0.1 mm with a
volume of 0.1 huh. Each square mm Is bordered by a triple line. The
center line of the three 1s the boundary line of the square. (See
Figure 1).
According to the U. S. Bureau of Standards' requirements, the cover
glass must be free of visible defects and must be optically plane on
both sides within plus or itlnus 0.002 mm. ONLY HEMOCYTOMETER COVER
GLASSES MAY BE USED. ORDINARY COVER 6LASSES AMD SCRATCHES HgH6CyT6H£TEft$
ARE UNACCEPTABLE, as they Introduce errors Into the volume relationships.
The suspension to be counted must be evenly distributed and free of
large debris, so that the chamber floods properly. The suspension to be
counted should contain 0.011 Tween 20 solution to prevent S1ard1a cysts
from sticking and causing Improper hemocytometer chamber flooding. Cyst
suspensions should be adjusted so that there are a total of 60 to 100 cysts
1n the four corner counting squares. Counts are statistically accurate
1n this range. If the suspension 1s too numerous to be counted, then 1t
must be diluted sufficiently to bring 1t Into this range. In some cases,
the suspension will be too dilute after concentration to give a statisti-
cally reliable count 1n the 60-100 cyst range. There 1s nothing that can
be done about this situation other than to record the result as question-
able.
To use the hemocytometer:
1. Dilute or concentrate the suspension as required.
2. Apply a clean cover glass to the hemocytometer and load the
hemocytometer chamber with 8-10 pi of vortexed suspension per
chamber. If this operation has been properly executed, the
liquid should amply fill the entire chamber without bubbles or
overflowing into the surrounding moats. Repeat this step with a
dean, dry hemocytometer and cover glass, 1f loading has been
Incorrectly done. See step (12) below for the hemocytometer
cleaning procedure.
3. Do not attempt to adjust the cover glass, apply clips, or 1n any
way disturb the chamber after 1t has been filled. Allow the
G1ard1a cysts to settle 30 to 60 seconds before starting the
count.
4. The G1ard1a cysts may be counted using a magnification 200-600X.
5. Move the chamber so the ruled area 1s centered underneath 1t.
6.
Then, locate the objective close to the cover glassi while watch-
ing ft fron the side of ratner than tnrougn the microscope.
-------
16
7. Focus up from the coversllp until the hemocytometer ruling
appears.
8. At each of the four corners of the chamber Is a 1 mm? divided •
Into 15 squares In which Glardla cysts are to be counted (see '
Figure 1). Beginning with the top row of four squares, count
with a hand tally counter In the directions Indicated In Figure
2. Avoid counting Glardla cysts twice by counting only those
touching the top and left boundary lines and none of those touch-
ing the lower and right boundary lines. Count each square mm in '
this fashion.
9. The formula for determining the number of Glardla cysts per ml
suspension 1s:
§ of cysts counted „ 10 „ dilution factor w 1.000 mm^ _
# of sq. mm counted * OF X — 1 x TrnT
f cysts/ml
10. Record the result on a data sheet similar to that shown In
Figure 3.
11. A total of six different hemocytometer chambers .nust be loaded,
counted, and then averaged for each Glardla cyst suspension to
achieve optimal counting accuracy.
12. After each use, the hemocytometer and coversllp must be cleaned
Immediately to prevent the cysts and debris from drying on 1t.
Since this apparatus 1s precisely machined, abrasives cannot be
used to clean 1t as they will disturb the flooding and volume
relationships.
a. Rinse the hemocytometer and cover glass first with tap
water, then 70% ethanol , and finally with acetone.
b. Ory and polish the hemocytometer chamber and cover glass
with lens paper. Store It 1n a secure place.
13. A number of factors are known to Introduce errors Into hemocyto-
meter counts. These Include:
a. Inadequate suspension mixing before flooding the chamber.
b. Irregular filling of the chamber, trapped air bubbles,
dust, or o11 on the chamber or coversllp.
c. Chamber coversllp not flat.
d. Inaccurately ruled chamber.
e. The enumeration procedure. Too many or too few Glardla
cysts per square, skipping or recounting some Glardla cysts.
-------
17
f. Total number of 61ard1a cysts counted 1s too low to
give statistical confidence 1n result.
g. Error In recording tally.
h. Calculation errorj failure to consider dilution factor,
or area counted.
' 1. Inadequate cleaning and removal of cysts fro* the previous
count.
j. Allowing filled chamber to sit too long so that chamber sus-
pension dries and concentrates.
-------
18
MM
_
3
MM
—
"
L—
IT*
•*
L
J
Zj
t/t
mm.
c
Q
¦
¦
¦
~
¦
¦
¦
¦
¦
T
-
g
¦
13
n
r
1
*
~
1
L-
1
af CiwMtar • 0.1mm.
Figure I. Hemocytometer platform ruling. Squares I, 2, 3, and 4 are
used to count 61ard1a cysts, (From Wale, 1967)
J •
o
•
3
3
•I 4
>
•
1 ft.
#
•
c
1 I
•
— 9
*
o|
• i
c
>
1 *
f •
1
t
' ~
• 0
I 1
i <
1
•
o c
Figure 2. Manner of counting 61ard1a cysts In 1 square mm. Dark cysts
are counted and light cysts ire omitted, (After Mfale, 1967)
-------
19
lat*
Nnen
Counting
Count
I
1 Cells
Counted
# ¦»*
Counted
Dilution
Faster
# C»«t»*
«H
• MUftl
1
..
2
3
«
S
C
?
B
•
9
10
11
12
13
1*
11
16
1?
.
*
18
19
•
20
*§ cysts/ml . * of cysts counted x 10 x dilution factor x 1,000 ntm3
ff of sq. mm counted 1 urn I 1 ml
Figure 3. Hemocytometer Data Sheet for S1ard1a Cysts
-------
m ¦*»
Appendix 6. Preparation and Loading of Excystatlon Chamber Slides
1. Using tape which 1s sticky on both sides, cut strips approximately 12
x 3 mm.
2. Apply a strip of the tape to one side of a 22 x 22 mm covefsl-1?.
3. Apply a second strip of tape to the opposite edge but same side of
the covers!1p.
4. Handling the covers! 1p by the edges only, attach the coversl.lp to the
center of a 3x1 Inch glass slide by placing the taped sides of the
coverslfp down along the long edge of the glass slide.
5. Make sure the coversl1p 1s securely attached to the slide by lightly
pressing down on the edges of the coversl1p with your fingers. Cart
should be taken to keep finger prints off the center of the coversl1 p.
6. To load the chamber slide, place a Pasteur or microliter pipette
containing at least 0.2 ml of the Siardia cyst suspension about 2 m
from an untaped edge of the coverslip. Slowly allow the cyst suspen-
sion to flow toward the coversl 1p. As it touches the coversl 1p it
will be wicked or drawn rapidly under the coversl1p by adhesive forces.
Only expell enough of the cyst suspension to completely fill the
chamber formed by the tape, slide, and covers!ip.
7. Wipe away any excess cyst suspension which is not under the coversl1p
with an absorbant paper towel, but be careful not to pull cyst
suspension from under the covers!Ip.
8; Seal all sides of the coversl1p with vaspar to prevent the slide from
drying out during the incubation.
Figure 1. Excystatlon Chamber Slide
NOTE; Prepared excystatlon chamber slides may be commercially avail-
able from Spiral Systems, Inc., 6740 Clough P1ke, Cincinnati,
Ohio 45244, (513) 231-1211 or 232-3122, or from other sources.
-------
appendix G-2
DETERMINING CHLORAHIHE INACTIVATION OF VIRUS
FOR THE SURFACE MATER TREATMENT RULE
Microbiological Treatment Branch
Risk Reduction Engineering Laboratory
and
Parasitology and Immunology Branch
Environmental Monitoring Systems laboratory
U.S. Environmental Protection Agency .
26 West Martin Luther King Drive
Cincinnati, Ohio 45268
-------
2
TABLE OF CONTENTS
I. Materials... 3
II. Reagents and Media ..4
III. MS2 Bacteriophage Assay 6
IV. Disinfection Procedure.. 8
V. Procedure for Determining Inact1vat1on, 9
VI. Bibliography 10
VII. Teehnleal Contacts.................................................11
-------
3
The Surface Water Treatment Rule requires 99.99S or greater removal/
1nact1vat1on of viruses. The following protocol may be used to determine
the percentage of virus 1nact1vat1on obtained by a treatment plant using
chloramlne disinfection.
I. MATERIALS
A. Materials for Disinfection
1.
Stock chlorine solution
2.
Stock ammonia solution
3.
Stirring device
4.
Incubator or water bath for less than ambient temperature
5.
Water from treatment plant
6.
MS2 bacteriophage
7.
Assorted glassware
8.
Assorted pipettes
9.
Aqueous, sterile sodium thlosulfate solution
10.
Refrigerator
U.
Vortex mixer
12.
Timer
Materials for MS2 Assay
1.
MS2 bacteriophage and Its Escherichia coll host
2.
Assorted glassware
3.
Assorted pipettes
4.
Incubator, 37°C
5.
Refrigerator
6.
Petri dishes, 100 x 15 mm, sterile
7.
Vortex mixer
8.
Water bath, 45°C
9.
Sterile rubber spatula
10.
EOTA, dlsodlum salt
11.
Lysozyme, crystallized from egg white
12.
Centrifuge with swinging bucket rotor
-------
4
II. REASENTS AND MEDIA
A. Tryptone-Yeast Extract (TYE) Broth
Ingredient Amount
Bacto tryptone 16.0 g .
feast extract 1.0 s
Glucose 1.0 §
NaCI 8.0 g
1.0 N CaCI2 2.0 ml
Dissolve 1n distilled water to a total volume of 1.0 liter,
then add 0.3 ml of 6.0 M NaOH. This medium should, be steri-
lized either by autodavlng for 15 minutes at 121"C or
filtration through a 0.22 um porosity membrane and then
stored at approximately 4*C. " It 1s used In preparing
bacterial host suspensions for viral assays.
B. Tryptone-Yeast Extract (TYE) Agar
Inqredlent Amount
TV? broth 1.6 liter
Agar 15.0 |
The agar should be added to the broth prior to steriliza-
tion. The medium should be sterilized by autodavlng for
15 minutes at 121°C. This medium 1s used to prepare slant
tubes for maintenance of bacterial stock cultures. The
prepared slant tubes should be stored at approximately 4°C,
C. Bottom Agar for Bacteriophage Assay
•Ingredient Amount
Bacto tryptone ' 1 • 10.0 g
Agar 15.0 g
NaCI 2.5 g
KC1 * 2.5 g
1.0 H Ca CI 2 1.0 ml
Dissolve the ingredients 1n distilled water to a total
volume of 1 liter. The medium should be sterilized by
autoclavlng for 15 minutes at 121°C. After autodavlng and
cooling, store at 4*C. Immediately prior to use, liquefy
the medium by heating. Add approximately 15 ml of lique-
fied agar Into each Petri dish. This bottom layer serves
as an anchoring substrate for the top agar layer.
-------
5
Top Agar for Bacteriophage Assay
Ingredient
Amount
Bacto tryptone
Agar
Ma CI
Yeast extract
Glucose
1.0 M CaCl 2
TCTT
8.0 g
8.0 g
1.0 g
1.0 g
1.0 ml
Dissolve the Ingredients 1n distilled water to a total
volume of 1 liter. This medium should be sterilized by
autoclavlng 15 minutes at 121*C. After cool1ng,' store at
4"C until needed 1n bacteriophage assays. Immediately
prior to. use 1n assays, liquefy the medium by heating and
then cool to and maintain at a temperature of 45°C.
Salt Diluent for Bacteriophage Assay
Ingredient Amount
flat! OT
1.0 M CaCl2 2.0 ml
Dissolve 1n distilled water to a total volume of 1 liter.
This diluent should be sterilized either by autoclavlng
for 15 minutes at 121®C or filtration through a 0.22 jm
porosity membrane. Store at room temperature.
CaCl 2 • 1.0 M
Inqredlent Amount
TOTi : ITTg
Dissolve 1n distilled water to a total volume of 100 ml.
Autoclave 15 minutes at 121°C or filter sterilize the
solution through a 0.22 urn porosity membrane. Store at
room temperature.
Sodium Thlosulfate, If
Ingredient Amount
Solium thiosuWate HlTg
Dissolve the sodium thlosulfate 1n 50 ml distilled water.
Adjust the volume to 100 ml with additional distilled
water. Filter sterilize the solution through a 0.22 ym
porosity membrane or autoclave 15 minutes at 121°C. Store
at room temperature.
-------
~
III* MS2 BACTERIOPHAGE ASSAY '
A. Microorganisms
1. MS2 bacteriophage: catalog number 15597-B1, American Type
Culture Collection, 12301 Parklawn Drive, Rockvllle, MD -20352
2. Bacterial host: Escherichia col 1, catalog number 15597,
American Type Culture Collection.
B. Growth and Maintenance of Microorganisms
1. Preparation of bacterial host stock cultures
Inoculate host bacteria onto TYE agar slant tubes, incubate
24 hours at 37°C to allow bacterial growth, and then refriger-
ate at 4#C. At monthly intervals the cultured bacterial
hosts should be transferred to a new TYE agar slant.
2. Preparation of bacteriophage stock suspension
Melt top agar and maintain at 45°C. Add 3 ml of the agar to
a 13 x 100 mm test tube cpntalned 1n a rack 1n a 45°C water
bath. Add 0.5 to 1.0 ml of the bacteriophage suspension
diluted so that the host bacterial "lawn" will show nearly
complete lysis after overnight Incubation. Add 0.1 to 0.2 ml
of a TYE broth culture of the host bacteria that has been
Incubated overnight. Mix gently and pour the contents on the
surface of bottom agar contained 1n a Petri dish that has'
been prepared previously. Rock the Petri dish to spread the
added material evenly over the agar surface. After the top
agar solidifies (about 15 minutes). Invert the Petri d 1 sh
and Incubate overnight at 37°C. Repeat the above procedure
so that a minimum of 5 but no more than 10 Petri dishes are
prepared.
Following this Incubation and using a sterile rubber spatula,
gently scrape the top and bottom agar layers into a large
beaker. Add to this pool of agar layers an amount of TYE
broth sufficient to yield a total volume of 80 ml. To this
mixture add 0.4 g of EDTA (dlsodlum salt) and 0.052 g of
lysozyme (crystallized from egg white). Incubate this mixture
at room temperature for 2 hours with continuous nixing. Then
centrifuge the mixture for 15 minutes at 3,000 x g. Carefully
remove the upper fluid layer. This fluid layer constitutes a
viral stock suspension for use 1n subsequent testing and
assays. The viral stock suspension may be divided Into
allquots and stored either frozen or at 4°C.
C. Performance of Bacteriophage Assay
A two-week supply of Petri dishes may be poured with bottom agar
ahead of time and refrigerated inverted at 4°C. If stored in a
refrigerator, allow agar plates to equilibrate to room temperature
-------
?
before use. Eighteen hours prior to beginning i bacteriophage
assay, prepare a bacterial host suspension by Inoculating 5 ml of
TYI broth with a small amount of bacteria taken directly from a
slant tube culture. Incubate the broth containing this bacteria!
inoculum overnight (approximately IS hours) at 37°C immediately
prior to use in bacteriophage assays as described below. This
type of broth culture should be prepared freshly for each "day's
bacteriophage assays. If necessary, a volume greater than 5 ml
can be prepared in a similar manner.
On the day of assay, melt a sufficient amount of top agar and
.laintain at 45'C 1n a water bath. Place test tubes (13 * 100 mm)
In a rack in the same water bath and allow to warm, then"add 3 ml
of top agar to each tube. Inoculate the test tubes containing
top agar with the bacteriophage samples (0.5 to 1.0 ml of the
sample/tube) plus 0.1 to 0.2 ml of the overnight bacterial host
suspension. Dilute the bacteriophage samples from 10"! to 10"*
in salt diluent prior to inoculation and assay each dilution in
triplicate. In addition, assay the unlnoculated salt diluent as
a negative control. Agitate the test tubes containing top agar,
bacteriophage inoculun, and bacterial host suspension gently on a
vortex mixer, and pour the contents of each onto a hardened
bottom agar layer contained In an appropriately numbered dish.
Quickly rock the Petri dishes to spread the added material evenly,
and place on a flat surface at room temperature while the agar
present 1n the added material solidifies (approximately 15 min-
utes). Invert and incubate the dishes at 37°C overnight (approxi-
mately 13 hours). The focal areas of viral Infection which
develop during this incubation are referred to as "plaques" and,
if possible, should be enumerated 1mmed1at1y after the incubation.
If necessary, the incubated Petri dishes can be refrigerated at
4°C overnight prior to plaque enumeration. As a general rule,
count only those plates that contain between 20 and 200 plaques.
-------
3
IV. DISINFECTION PROCEDURE
A. The treatment plant water to be used should be the water Influent
Into the chloramlne disinfection unit process used 1n the plant.
If chloramlne disinfection 1s performed at more than one point 1n
the treatment process, e.g. preflltratlon and postflltratlon, the
procedure should simulate as closely as possible actual treatment
practice.
B. Prepare stock ammonia and chlorine solutions to be added to the
treatment plant water to achieve the same stoichiometric relation-
ship between chlorine and ammonia that 1s used 1n the water
treatment plant. These solutions should be concentrated enough
so that no more than 2 ml of each solution will be added to the
treatment plant water being disinfected.
C. Determine the contact time by the methods described 1n the Surface
Water Treatment Rule and/or the associated Guidance Manual.
D. Rinse two 600 ml beakers with treatment plant water to remove any
extraneous material that may cause disinfectant demand. Then add
400ml treatment plant water to the beaker. The first beaker
will be seeded with H92 before the contents are chloramlnated.
The second beaker will be an Indigenous virus control and will
be chloramlnated without addition of extraneous phage.
E. Mix the contents of the beaker short of producing a vortex 1n the
center and continue until the conclusion of the experiment.
F. Equilibrate the 600 ml beakers and their contents as well as the
disinfectant reagents to the desired experimental temperature.
G. Dilute the stock MS2 bacteriophage so that the bacteriophage con-
centration 1s 1 to 5 x 10® PFU/ml. ,
H. Add 1.0 ml of the diluted MS2 bacteriophage to the contents of the
first 600 ml beaker.
I. Remove a 10 ml sample from the contents of the first beaker after
2 minutes of mixing. Assay the HS2 bacteriophage concentration
1n this sample within 4 hours and record the results as PFU/ml.
This value 1s the Initial MS2 concentration.
J. Remove a 10 ml sample from the contents of the second beaker
after 2 minutes of mixing. Assay the Indigenous bacteriophage
concentration In this sample within 4 hours (at the same time as
you assay the sample from the first beaker) and record the .
results as PFU/ml. This value 1s the Initial unseeded concentra-
tion.
K. . Add the disinfectant reagents to the contents of both beakers
using the same sequence, time, and concentrations as are used in
the actual treatment plant operations.
-------
9
L. Just prior to the end of the contact time, remove a volume of sam-
ple adequate for determination of the disinfectant residual con-
centration from both beakers. Use methods prescribed 1n the
Surface Water Treatment Rule for the determination of combined
chlorine. This residual should be the same (*20t) as the
residual present In the treatment plant operation.
H. 'At the end of the exposure time, remove a 10 ml sample from the
first 600 ml beaker and neutralize with 0.25 ml of 1.01 aqueous,
sterile sodium thlosulfate. Assay for the MS2 bacteriophage
survivors and record the results as PFU/ml. This value Is the
exposed HS2 concentration.
N. At the end of the exposure time, remove a 10 ml sample from the
second 600 ml beaker and neutralize with 0.25 ml of 1.01 aqueous,
sterile sodium thlosulfate. Assay for the Indigenous bacterio-
phage survivors and record the results as PFU/ml. This value 1s
the exposed unseeded concentration.
V. PROCEDURE FOR DETERMINING INACTIVATION
A. Calculation of Percentage Inact1vat1on
Use the following formula to calculate the percent 1nact1vat1on
of MS2:
1. 1 1nact1vat1on - 1001 - [(exposed HS2/1n1t1al MS2) x 100]
'Jslng values from Section IV steps I, J, M and N calculate Initial
MS2 and exposed HS2 as follows:
2. Initial MS2 (PFU/ml) - I - J.
3. Exposed MS2 (PFU/ml) ¦ M - N.
If the number of PFU/ml 1n exposed MS2 1s zero, I.e., no plaques
are produced after assay of undiluted and diluted samples, use <1
PFU/ml as the value 1n the above formula.
B. Comparison of Percentage Inact1vat1on to log^Q of Inact1vat1on
681 1nact1vat1on 1s equivalent to 0.5 logio 1nact1vat1on
901 1nact1vat1on 1s equivalent to 1 log^o 1nact1vat1on
991 1nact1vat1on 1s equivalent to 2 log^o 1nact1vat1on
99.91 1nact1vat1on 1s equivalent to 3 log^Q 1nact1vat1on
-------
10
VI. BIBLIOGRAPHY
Adams, M.H. Bacteriophages. Intersclence Publishers, New York (1959).
American Public Health Association; American Water Works Association;
Water Pollution Control Federation. Standard Methods for the Examina-
tion of Water and Wastewater. 16th ed. (1985).
Grabow, W.O.K. et al. Inact1vat1on of hepatltus A virus,.other enter-
ic viruses and Indicator organisms In water by chlorlnatlon. Water
Sc1. Technol., 17:657 (1985)
Jacangelo, J.D.; 0l1v1er1, V.P.; & Kawata, K. Mechanism of Inactlva-
tlon of microorganisms by combined chlorine. AWWARF Kept., Denver,
CO (1987).
Safe Drinking Water Committee. The disinfection of drinking water.
In: Drinking Water and Health. National Academy Press, Washington,
D.C., 2:5 (1980).
Shah, P. & McCamlsh, J. Relative resistance of pollovlrus 1 and col 1 -
phages and T2 1n water. Appl. Microbiol. 24:658 (1972).
U.S. Environmental Protection Agency. Guidance Manual for COmpllance
with the Filtration and Disinfection Requirements for Public Water
Systems Using Surface Water Sources. Appendix G. U.S. EPA, Office
of Water, Criteria and Standards Division, Washington, D.C. (1983).
Ward, N.R.; Wolfe, R.L.; & Olson, S.H. Effect of pH, application
technique, and chlorine-to-n1trogen ratio on disinfectant activity of
Inorganic chloranlnes with pure culture bacteria. Appl. Environ.
Microbiol., 48:508 (1984).
-------
TECHHICAL CONTACTS:
A. Donald Be man
Microbiological Treatment Branch
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
25 West Martin Luther King Drive
Cincinnati, Ohio 41268
Phone: (513) 569-7235
8. Chrlston J. Hurst
Microbiological Treatment Branch
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
26 West Martin Luther King Drive
Cincinnati, Ohio 45263
Phone: (513) 569-7331
-------
G.3 DETERMINING CHLORINE DIOXIDE INACTIVATION
OF filARPIA CYSTS AND VIRUS
Giardia Cysts .
The basis for the chlorine dioxide CT values for Giardia cysts In
the Guidance Manual is given in Appendix F.1.2. The CT values are based
on data collected mainly at pK 7. Very little data was available at other
pHs. A review of data from Hoff (1966) indicates that the disinfection
efficiency of chlorine dioxide for bacteria and viruses' increases
approximately 2 to 3 fold as pH Increases froo 7 to 9. Data on which the
CT values 1n the SWTR are based Indicate that at 25 C, £. muris cyst
Inactlvation CTs were approximately 2 fold higher at pH 7 than at pH 9
(Leahy, 1985). In addition, the data also Indicate that chlorine dioxide
efficiency Increases as disinfectant concentration increases within the
range studied.
Because the data on effects of chlorine dioxide concentration and
water pH on Giardia cyst inactlvation efficiency were very limited, they
were not considered 1n calculating the Giardia cyst CT values In Appendix
E. However, the data suggest that site specific conditions, I.e. water
pH and disinfectant concentration, can have significant effects on
chlorine dioxide effectiveness. Therefore, the option of allowing the
Primacy Agency to consider the use of lower CT values by Individual
systems has been provided.
This approval should be based on acceptable experimental data
provided by the system. The data should be collected using the protocol
provided in Appendix G-l for determining Giardia cyst Inactlvation by
chloramlne with appropriate changes in Section IV A,„B, I and J to reflect
the use of chlorine dioxide rather than chloramines. This procedure can
be used for any disinfectant which can be prepared 1n an aqueous solution
and 1s stable over the course of the testing. To do this, chloramine
should be replaced with the test disinfectant in the above noted sections.
G.3 - 1
-------
Virus
The basis for the chlorine dioxide CT values for virus In Appendix
F.2.2 consists of United data from Sobsey (1968). Because the pH 9 data
available were very limited, the CT values are based on the pH 6 data with
a safety factor of 2 applied. As indicated previously, review of data
from a number of studies (Hoff, 1986) shows that chlorine dioxide
efficiency Increases 2 to 3 fold as pH Increases from 7 to 9.
Because the virus CT values for chlorine dioxide are very conserva-
tive and most systems operate at water pHs higher than those jon which the
CT values are based, the option of allowing the Primacy Agency to consider
the use of lower CT values has been provided.
This approval should be based on acceptable experimental data
provided by the system. The data should be collected using the protocol
provided in Appendix G.2 with appropriate changes 1n Sections I A, 1 and
2 and IV A, B, D, K, and L to reflect the use of chlorine dioxide rather
«
than chloramlnes. This procedure can be used for any disinfectant which
can be prepared 1n an aqueous solution and 1s stable over the course of
the testing. To do this, chloramine should be replaced with the test
disinfectant 1n the above noted sections.
REFERENCES
Hoff, J.C. Inactlvation of Microbial Agents By Chemical Disinfectant's,
EPA/600/52-86/067, U.S. Environmental Protection Agency, Water Engineering
Research Laboratory, Cincinnati, Ohio, September, 1986.
Leahy, J.G. Inactlvation of Giardla Murls Cysts by Chlorine and Chlorine
Dioxide. Thesis, Department of C1v11 Engineering, Ohio State University,
1985.
Sobsey, N.D. Detection and Chlorine Disinfection of Hepatitis A Virus 1n
Water. CR813024, EPA Quarterly Report, Dec. 1988.
G.3 - 2
-------
G.4 DETERMING OZONE INACTIVATION OF 6IARDIA CYSTS AND VIRUS
G.4.1 BACKGROUND
The basis for the ozone CT values are given In Appendices F.1.2
(Giardia cysts) and F.2.4 (Virus). As Indicated, both sets of CT values
are based on limited data and because of this, the values established are
conservative and employ large safety factors. In addition, the difference
between the way the laboratory experiments used to develop the CT values
and how ozone Is used In water treatment presents a problem with
translating the data for field use. The laboratory studies were conducted
using steady state ozone concentrations with ozone continually added
during the contact period. In contrast, steady state ozone concentrations
are not maintained In field use. Also, the effectiveness of ozone
contactors used In field applications may vary from each other and from
the mixing efficiencies applied In the laboratory experiments used to
establish the CT values.
The net effect of all of these differences 1s to limit the appli-
cability of the CT values 1n the SWTR and Guidance Manual to Individual
systems. Therefore, the option of allowing the Primacy Agency to consider
the use of lower CT values by Individual systems has been provided.
This approval should be based on acceptable experimental data
provided by the system. In general, the procedures provided In Appendix
G.l for determining Giardia cyst inactivation and Appendix G.2 for
determining virus Inactivation can be used. However, unlike chloramlnes
ozone is not a stable disinfectant. Because of ozone's rapid dissipation,
a pilot study must be used In lieu of the batch system to demonstrate the
disinfection efficiency. General considerations for conducting pilot
studies to demonstrate the disinfection ability of ozone or any other
unstable disinfectant are enumerated below.
G.4.2 GENERAL CONSIDERATIONS FOR PILOT TEST
A. All microorganisms, reagents and media are prepared as in-
dicated in sections G.l for Giardia and G.2 Tor virus.
G.4-1
-------
8. The disinfectant should be prepared, Measured and added to the
test water as it would be added to the water at the water
treatment plant.
C. Specific reactor design should be a function of the disinfec-
tant and reflect how the disinfectant Is added at the water
treatment plant. Provisions should be Bade to determine
concentration of disinfectant and Microbial Survival to be <
¦easured with contact tiM..
An example of conducting a pilot test for a plug flow reactor using
ozone or another unstable disinfectant Is provided below.
Bmnpk * Plug Flow Reactor Protocol
The size of the plug flow reactor can be approxlnated from the table
below. Glass, stainless steel, copper, plastic tubing or other Material
compatible with the disinfectant can be used to construct the plug flow
reactor. Table 1 shows the approximate length of pipe for a" plug flow
reactor to yield 10 minutes contact at flow rates between 50 and 500
ml/min. Depending on pipe size and Material an economical reactor can be
constructed.
TAIlE 1. A'HOXIIUTE UMTH MB DIAMETER OF *IPE
IASI0 ON F LOW
LINEAI PlfE-lEISTH, MlfllS
NOMINAL PIH OlAMETil. 01
no* time mmt o.« 1.2 1.1 in • s.n 1.01
¦t/niR NIK. LITERS CC 0.21 . I 11 I, 14 1.07 11.49 10.27
10
10
0.1
100
17.7
4 4
2.0
1.0
0.4
0.2
too
10
1
1000
11.4
I.I
. 1.1
2.0
O.t
0.1
200
10
2
2000
70.7
17.7
7. t
1.1
11
1.0
100
10
1
1000
101.1
26.1
11.1
l.t
2. i
1.1
100
10
4
4000
141. S
11.4
11.7
7.1
J.l
2.0
I0S
10
1
1000
171.1
44.2
It 1
It
4. 4
2.1
Additional information on the design of specific pilot studies can
be found in the following references by Thompson (1982), Montgomery (1985),
and A1-Ani (1985).
Additional Materials to those in G.l and/or G.2
plug flow reactor
cyst suspension, 2xl07 cysts/trial
-------
cyst quantity - cysts are prepared as Indicated In 6.1.
10® cysts/ml X 20,000 ml % 2xl07 cysts required/trial.
MS2 stock, 2xl0'°/tr1a1
2-20 liter (5 gal) carboy
test water pump, mid range 200 ml/m1n
disinfectant generator
disinfectant pump, aid range 10-20 ml/mln
disinfectant residual reagents and equipment
A. Reactor conditions
1. Test Hater Flow rate- 200 ml/mln (this may vary from 50 to 500
ml/mln with 20 1 reservoir total experimental time- 100 m1n)
2. Disinfectant flow
gas-requ1res specific contactor designed for disinfectant
L1qu1d»l0 to 20 ml/mln
3. Temperature
controlled
4. Prepare 20 liter reservoir (5 gal) of test water at the pH and
temperature of the CT trial. Do not add microorganisms
5. Prepare 20 liter reservoir (5 gal) of test water and equi-
librate to the temperature of the CT trial. Add 61ard1a murls
cysts at an Initial density of 101 cysts/ml and/or MS2 bacter-
ial virus at an Initial density of 10* PFU/ml. Nix thoroughly
and adjust pH to the pH of the CT trial. Continuous mixing of
the test water feed stock should be carried out over the
course of the CT trial to prevent the Glardla cysts from
settling.
B. Disinfection Procedure - Prior to Disinfection Trial
1. Determine contact time for the sample ports In the plug flQw
reactor under conditions of the CT trial by methods described
1n the SWTR.
2. Determine disinfectant concentration with no microorganisms 1n
the feed test water.
6.4-3
-------
C. CT Trial Procedure
1. Start test water feed without cysts and or virus (approx. 200
¦1/nln), start disinfectant feed (gas or liquid).
Allow system to equilibrate.
Monitor disinfectant residual by appropriate aethod during
this time. Samples for disinfectant residual should be taken
directly Into tubes or bottles containing reagents to fix the
disinfectant at the tlae the sample Is collected. * Keep a plot
of disinfectant residual vs running tlae to evaluate steady
state conditions.
2. After the disinfectant residua! has stabilized, switch to the
reservoir containing the test eicroorganism(s).
3. Allow system to equilibrate for a tine ¦ 3 X final contact
tine.
example
final contact time -10 Bin, allow 30 nin.
4. Monitor disinfectant residual by appropriate method during
this time. If the disinfectant residual 1s stable begin
chemical and biological sampling for calculation of CT. -
5. Sampling
a. Chemical
A sufficient volume (about 250 ml should be collected
from the sampling tap prior to the biological composite
to determine:
pH
Residual disinfectant - Samples
should be collected directly
into tubes or bottles containing
reagents to fix the disinfectant
at the time the sample Is collected.
fc. Biological
Samples for microbial analysis are collected as short
time composite samples over a 10 to 20 minute time
period. Several trials may run for a given 20 liter
test water preparation as long as sufficient equilibra-
G.4-4
-------
tion and flow recovery tines are allowed between,
trials.
- Zero time samples should be collected as 250 *1
composite samples either directly from the test
water feed reservoir or In line prior to the addi-
tion of the disinfectant.
• Four 250 ml samples are collected separately Into
a 2 1 sterile bottle containing a neutralizing
agent for the particular disinfectant. Each
sample 1s thoroughly nixed upon collection and
stored at 4 C. If Multiple saaple ports are used,
the order of collection should be froa longest to
shortest contact time to nlnlnlze flow changes due
to sampling.
6. Giardia cyst recovery and assay.
Concentrate the 1000 nl composite sample by filtration
according to the method given 1n section G.l. Record and
report the data as described 1n section 6.1. The expected
cysts/sample Is given below:
Cysts/sample « 4 x 250 ml X 103 cyst/ml • lxlO'cyst/sample.
7. Virus Assay
Before filtration for Giardia. remove 10.0 ml from the
biological composite sample to a sterile screw cap culture
tube containing 2 to 3 drops chloroform. Assay for MS2,
record and report the virus data according to the methods and
procedures described In G.2. Be sure to correct the Giardia
sample volume to 990 ml.
8. Calculation of CT _•
Calculate CT In a manner described In Section G.l for Giardia
and Section G.2 for virus. The residual disinfectant should
be the average of the four residual determinations performed
prior to the Individual samples collected for the biological
composite and the time should be the time determined for. the
sample port under similar flow conditions.
G .4-5
-------
REFERENCES
Al-Ani, C.S.U., Filtration of filardla Cysts and other substances:
Volume 3. Rapid Rate Filtration (EPA/600/2-85/027) 1985.
Montgomery, Janes M. Consulting Engineers Inc., Water Treatment Principle
and Oeslon. John Wiley and Sons, May 1982.
WalUs, P.M., Oavles, J.S., Nuthonn, R.,B1chan1n-Mapp1n, J.M., Roach, P.O.,
and Van Roodeloon, A. Removal and Inactlvatlon of 61ard1a~ Cysts 1n a
Mobile Water Treatment Plant Under Field Conditions: Preliminary Results.
In Advances In Glardla Research. P.M. Wall Is and B.R. Hanand, eds, Union
of Calgary Press, p. 137-144, 1989.
Wolfe, R.L., Stewart, M.H., Llange, S.L., and McGuIre, M.J., Disinfection
of Model Indicator Organisms In a Drinking Water Pilot Plant by Using
PEROXONE, Applied Environmental Microbiology, Vol 55, 1989, pp 2230-2241.
OHvlerl, V.P. and Sykora, J.L., Field and Evaluation of CT for Determining
the Adequacy of Disinfection. American Water Works Association Water
Quality Technology Conference. In press, 1989.
6.4-6
-------
APPENDIX H
SAMPLING FREQUENCY FOR TOTAL COLI FORMS
IN THE DISTRIBUTION SYSTEM
-------
TABLE H-l
TOTAL COLIFORM SAMPLING REQUIREMENTS
BASED UPON POPULATION
Minimum
Minimum
Number
Number
Population
of Samples
Per Month
Population
of Samples
Served
Served
Per Month
25
to
1,000
1
59,001
to
70,000
70
1,001
to
2,500
2
70,001
to
83,000
80
2,501
to
3,300
3
83,001
to
96,000
90
3,301
to
4,100
4
96,001
to
130,000
100
4,101
to
4,900
5
130,001
to
220,000
120
4,901
to
5,800
6
220,001
to
320,000
150
5,801
to
6,700
7
320,001
to
450,000
180
6,701
to
7,600
8
450,001
to
600,000
210
7,601
to
8,500
9
600,001
to
780,000
240
8,501
to
12,900
10
780,001
to
970,000
270
12,901
to
17,200
15
970,001
to
1,230,000
. 300
17,201
to
21,500
20
1,230,001
to
1,520,000
330
21,501
to
25,000
25
1,520,001
to
1,850,000
360
25,001
to
33,000
30
1,850,001
to
2,270,000
390
33,001
to
41,000
40
2,270,001
to
3,020,000
420
41,001
to
50,000
50
3,020,001
to
3,960,000
450
50,001
to
59,000
60
3,960,001
or
more
. 480
Notes:
1. Non-community systems using all or part surface water and community'
systems oust monitor total coliform at this frequency. A non-
community water system using ground water and serving 1,000 persons
or fewer must monitor quarterly, beginning 5 years after the rule's
promulgation, although this can be reduced to yearly if a sanitary
survey shows no defects. A non-community water system serving more
than 1,000 persons during any month, or a non-community water system
using surface water, must monitor at the same frequency as a like-
sized community public water system for each month the system
provides water to the public.
2.
Unfiltered surface water systems must analyze one coliform sample
each day the turbidity exceeds 1NTU.
-------
TABLE H-l
TOTAL COLZFORM SAMPLING REQUIBEMENTS
BASED UPON POPULATION (Continued)
Systems collecting fewer than 5 samples per month on a regular basis
must conduct sanitary surveys. Community and non-community systems
must conduct the initial sanitary surveys within 5 and 10 years of
promulgation, respectively. Subsequent surveys must be conducted
every 5 years# except for non-community systems using protected and
disinfected ground water, which have up to 10 years to conduct
subsequent surveys.
-------
TABLE H-2
MONITORING AND
REPEAT SAMPLE FREQUENCY
System Size
ff Routine
Samples
# Repeats More Monitorina For
NCWS(1)
Quarterly*2'
4 5/mo for
1
additional
mo
25 - 1,000
Monthly*2'
4 5/mo for
1
additional
mo
1,001 - 2,500
2/mo
3 5/mo for
1
additional
mo
2,501 - 3,300
3/mo
3 5/mo for
1
additional
mo
3,301 - 4,100
4/mo
3 5/mo for
1
additional
mo
4,101 - 4,900
5/mo
3 None
>4,900
Table 1
3 None
Notes:
1.
2.
Non-community Water system*.
For exceptions, see Table 1.
-------
APPENDIX I
MAINTAINING REDUNDANT
DISINFECTION CAPABILITY
-------
APPENDIX I
REDUNDANT DISINFECTION CAPABILITY
The SWTR requires that unflltered water systems provide redundant
disinfection components to ensure the continuous application of a
disinfectant to the water entering the distribution system. In aany
systems, both filtered and unflltered, a primary disinfectant is used to
provide the overall Inactlvatlon/removal and a secondary -residual Is
applied to maintain a residual In the distribution system. As outlined 1n
Sections 3.2.4 and 5.5.4, redundancy of the disinfection system(s) Is
recommended to ensure that the overall treatment requirement of 3-1 og
Giardla cyst and 4-1og virus removal/inactlvatlon Is achieved, and a
residual Is maintained entering the distribution system. This 1s
particularly Important for unflltered supplies where the only treatment
barrier 1s disinfection. Redundancy of components Is necessary to allow
for disinfection during routine repairs, maintenance and inspection and
possible failures.
In reviewing water disinfection facilities for compliance with
redundancy requirements, the following Items should be checked:
I. General
A. Are the capacities of all components of both the primary
system and the backup system equal to or greater than the
required capacities?
Some systems may have two or more units that provide the
required dosage rates when all units are operating. In these
cases, an additional unit 1s needed as backup, during the
downtime of any of the operating units." The backup must have
a capacity equal to or greater than that of the largest
on-line unit.
B. Are adequate safety precautions being followed, relative to
the type of disinfectant being used?
C. Are redundant components being exercised or alternated with
the primary components?
D. Are all components being properly maintained?
E. Are critical spare parts on hand to repair disinfection
equipment?
1-1
-------
F. Arc spare parts avallabia for components that art inch spons-
ible for disinfecting the water?
II. Disinfectant Storage
A minimum of two storage units capable of being used alternately
should be provided. The total combined capacity of the storage units
should provide as a minimum the systea design capacity.
A. Chlorine
Storage for gaseous chlorine will normally be in 150-lb cylinders,
2,000-lb containers, or larger on-site storage vessels.
1. Is there automatic switchover equipment If one cylinder
or container empties or becomes inoperable!
2. Is the switching equipment 1n good working order,
(manually tested on a regularly scheduled basis), and
are spare parts on hand?
3. Are the scales adequate for at least two cylinders or
containers.
b. Hypochlorite
Storage of calcium hypochlorite or sodium hypochlorite is normally
provided in drums or other suitable containers. Redundancy requirements
are not applicable to these by themselves, as long as the required minimum
storage quantity is on hand at all times.
C. Ammonia
An hydrous ammonia 1s usually stored in cylinders as a pressurized
liquid. Aqua ammonia is usually stored as a solution of ammonia and water
in a horizontal pressure vessel.
1. Is the available storage volume divided Into two or more
usable units?
2. Is automatic switching equipment in operation to change
over from one unit to another when one is empty or
inoperable?
3. Are there spare parts for the switching equipment?
1-2
-------
111. ^tncrttton
Ozone and chlorine dioxide are not stored on-site. Rather, because
of their reactivity, they are generated and used Immediately.
To satisfy the redundancy requirements for these disinfectants It Is
recommended that two generating units, or two sets of units, capable of
supplying the required feed rate be provided. In systems where there Is.
more than one generation system, a standby unit should be available for
times the on-line units need repair. The backup unit should have a
capacity equal to or greater than the un1t(s) It mays replace.
A. Chlorine Dioxide
Chlorine, sodium chlorite, or sodium hypochlorite should be stored
In accordance with storage guidelines previously described.
B. Ozone
Are all generation components present and In working order for both
the primary and the redundant units (whether using air or oxygen)?
C. Common
Is switchover and automatic start-up equipment Installed and
operable to change from the primary generating unit(s) to the redundant
unit (s)?
iv. feed Systwis
Redundancy In feed systems requires two separate units, or systems,
each capable of supplying the required dosage of disinfectant. If more
than one unit 1s needed to apply the required feed rate, a spare unit
should be available to replace any of the operating units during times of
malfunction. The replacement unit should, therefore, have a capacity
equal to tr greater than that of the largest unit which 1t may replace*
This requirement applies to all disinfection methods, and 1s best
implemented by housing the on-line and redundant components 1n separate
rooms, enclosures, or areas, as appropriate.
In reviewing these systems for redundancy, the following components
should be checked:
A. Chlorine
1. Evaporators
2. Chlorinators
3. Injectors
1-3
-------
B. Hypochlorite
1. Mixing tanks and mixers
2. Chemical feed pumps and controls
3. Injectors
C. Ozone
1. Dissolution equipment, including compressor and delivery
piping systems
D. Chlorine Dioxide
1. Chlorine feed equipment
2. Sodium chlorite nixing and metering equipment
3. Day tank and mixer
4. Metering pumps
5. If a package C102 unit 1s used, two must be provided
I. Chloramlnation
1. Chlorine feed equipment
2. Amnonla feed equipment, Including applicable equipment
for either:
a. Anhydrous ammonia (gas
b. Aqua ammonia (solution
v. Residual Monitoring
The best method of monitoring a disinfection facility for continuous
operation 1s by continuous recording equipment. To improve reliability,
It 1s suggested that duplicate continuous monitors are present for backup
In the event-of monitor failure. However, if there Is a failure 1n the
monitoring system for Indicating that a continuous residual Is being
maintained, the SWTR allows systems to take grab samples every four hours
for up to five days during monitor repair. For systems without 24 hour
staffing 1t will not be practical to take grab samples and redundant
monitoring equipment is recommended. Failure of continuous monitoring
would be a violation of a monitoring requirement, not a treatment
requirement.
a. Chlorine
1. Does the facility have a continuous monitor for chlorine
residual at the disinfection system site with an alarm
1-4
-------
or Indicator to show when the Monitor 1s not function-
. Ing? For added assurance, the provision of a backup
nonltorlng unit 1s also recommended.
2. Is there instrumentation In place to automatically
switch from one Monitor to the other If the first one
falls?
B. Hypochlorite
Sane as for chlorine system.
C. Ozone
1. Does the facility have a continuous ozone monitor with
automatic switchover capability and alarms?
2. Does the facility have a continuous ozone residual
nonltor with automatic switchover capability and alarms?
D. Chlorine Dioxide
1. Does the facility have a continuous chlorine dioxide
monitor with automatic switchover capability and alarms?
2. Does the facility have a continuous chlorine dioxide
residual monitor with automatic switchover capability
and alarms?
E. Chloramination
1. Does the facility have a continuous aamonia Monitor with
automatic switchover capability and alarm?
2. Does the facility also have a continuous chlorine
residual monitor on-site with automatic switchover
capability and alarms?
vi. Power SmbbIy
A permanently Installed standby generator, capable of running all
electrical equipment at the disinfection station, and equipped for
automatic start-up on power failure, should be on-site and functional.
Alternatives to a standby generator, such as a feed line from a
different power source, are acceptable 1f they can be shown to have equal
reliability.
1-5
-------
VII. Alarms
Indicators.and alarms, both local and remote, should be capable of
promptly alerting operating and supervisory personnel of problem
conditions.
A. Local
Lights, buzzers, and horns should be Installed and functioning to
alert on-site personnel to problem conditions.
b. Remote
Alarm signals should be relayed to a central control panel which 1s
manned 24 hours per day and whose operators can notify response personnel
Immediately.
C. Problem Conditions
A minimum 11st of problem conditions which should have Indicators
and alarms, both locally and at a 24-hour per day switchboard, are as
follows:
1. Disinfectant leak
2. Feeder pump failure
3. Power outage
4. Generator or alternate power source on
5. Disinfectant residual less than setpolnt value
viii. Uvwt
Maximum reliability is ensured when redundant units are separated
from primary units. The type of separation should be appropriate to the
type of potential malfunction. For example, any area within a building
subject to a chlorine leak should have primary components separated from
redundant components by an airtight enclosure, i.e., separate rooms of
varying sizes.
IX. Separate Facility
Under certain conditions, such as location of a disinfection
facility in an area of high earthquake potential, the most reliable means
of providing redundant facilities may be to house them in a completely
separate structures at a different site.
1-6
-------
APPENDIX J
WATERSHED CONTROL PROGRAM
-------
APPENDIX J
WATERSHED CONTROL PROGRAM
The following 1s a guideline for documenting a watershed control
program. The SWTR only requires a watershed control program for
unflltered supplies. A watershed control program can also benefit a
filtered system by providing protection for maintaining the source water
quality, minimizing the level of disinfection to be provided. It 1s
therefore reconanended that all systems conduct the basic elements of a
watershed control program. However, the scope of the program should
Increase as the complexity and size of the watershed/ system Increases.
The program could be more or less comprehensive than this outline, and
will be determined on a case-by-case basis by the utility and the Primacy
Agency. In addition to the guidelines below, a wellhead protection
program could be the basis of a watershed control program In many states.
All of the elements found below would also be part of a local wellhead
protection program.
A. Watershed Description
1. Geographical location and physical features of the
watershed.
2. Location of major components of the water system in
relationship to the watershed.
3. Hydrology: Annual precipitation patterns, stream flow
characteristics, etc.
4. Agreements and delineation of land use/ownership.
B. Identification of the Watershed Characteristics
and Activities Detrimental to Water Quality
1„ Naturally Occurring:
a. Effect of precipitation, terrain, soil types and
land cover
b. Animal populations (describe) — include a dis-
cussion of the Giardia contamination potential,
any other microbial contamination transmitted by
animals
J-l
-------
c. Other - any other activity which can adversely
affect water quality
Han-Made:
a. Point sources of contaolnation such as wastewater
treatment plant, Industrial discharges, barnyard,
feedlots, or private septic systems
The Impact of these sources on the microbiological
quality of the water source should be evaluated. In
cases resulting In Identifiable degradation, the
discharges should be eliminated In order to minimize the
treatment of the water needed.
b. Nonpolnt Source of Contamination:
1) Road construction - major highways, rail-
roads
2) Pesticide usage
3) Logging
4) Grazing animals
5) Discharge to ground water which recharges
the surface source
6) Recreation activities
7) Potential for unauthorized activity 1n the
¦ watershed
8) Describe any other human activity in the
watershed and Its potential Impact on water
quality
It should be noted that grazina animals In the watershed
may lead to the presence of Cryptosporidium in the
water. Cryptosporidium 1s a pathogen which may result
1n a disease outbreak upon Ingestion. No Information Is
available on Its resistance to various disinfectants,
therefore It Is recommended that grazing should not be
permitted on watersheds of non-filtering systems:
Sewage discharges will Introduce viruses into the water
source which may be occluded 1n solids and protected
from inactivation through disinfection. It is, there-
fore, recommended that sewage discharges should not be
permitted within watersheds of non-filtering supplies.
Although it is preferable to not have grazing or sewaae
discharges within the watershed, Primacy.Agencies will
-------
need to evaluate the Inpact.of these activities on a
case-by-case basis. In cases where there Is a. long
detention tine and a high degree of dilution between the
point of the activity and the water Intake, these activ-
ities nay be pernlsslble for unflltered supplies. The
utility should set priorities to address the Inpacts 1n
B.l. and 2., considering their health significance and
the ability to control then.
C. Control of Detrimental Activities/Events
Depending on the activities occurring within the watershed,
various techniques could be used to ellnlnate or nlnlnlze
their effect. Describe what techniques are being used to
control the effect of activities/events Identified In B.l. and
2. In Its yearly report.
Example:
Activity: Logging In the watershed.
Management Decision: Develop program to minimize Impact
of logging.
Procedure: Establish agreements with logging companies
to maintain practices which will nimnize adverse
Impacts on water . quality. These practices should
Include:
limiting access to logging sites
ensuring cleanup of sites
controlling erosion fron site.
Monitoring: • Periodically review logging practices to
ensure they are consistent with the agreement between
the utility and the logging conpanles.
Example:
Activity: Point sources of discharge within the
watershed.
Management Decision: Eliminate those discharges or
minimize their Impact.
Procedures: Actively participate 1n the review of
discharge permits to alert the reviewing agency of the
potential (actual} Impacts of the discharge and lobby
for its elimination or strict control.
J-3
-------
Monitoring: Conduct special Monitoring to ensure
conditions of the peralt are Mt and to document adverse
effects on water quality.
o. Monitoring
1. Routine: Minimum specifications for monitoring several
raw water quality parameters are listed In Section 3.1.
Describe when, where and how these samples will be
collected. These results will be used to evaluate
whether the source may continue to be Used without
filtration.
2. Specific: Routine monitoring may not provide Informa-
tion about all parameters of Interest. For example, 1t
may be valuable to conduct special studies to measure
contaminants suspected of being present (Giardia. pesti-
cides, fuel products, enteric viruses, etc.). Frequent
presence of either Giardia or enteric viruses In raw
water samples prior to disinfection would Indicate an
Inadequate watershed control program. Monitoring may
also be useful to assess the effectiveness of specific
control techniques, and to audit procedures or opera-
tional requirements instituted within the watershed.
Utilities are encouraged to conduct additional monitor-
ing as necessary to aid then 1n controlling the quality
of the source water.
e. Management/Operations
1. Management
a. Organizational structure
b. Personnel and education/certification requirements
2. Operations
a. Describe system operations and design flexibility.
b. The utility should conduct some form of ongoing
review or survey In the watershed to Identify and
react to potential Impacts on water quality. The
scope of this review should be documented and
agreed upon by the utility and Primacy Agency oh
a case-by-case basis.
c. Specifically describe operational changes which
can be made to adjust for changes In water quali-
ty. Example: Switching to alternate sources;
Increasing the level of disinfection; using
J-4
-------
settling basins. Discuss what triggers, and who
decides to Make, those changes.
3. Annual Report: As part of the watershed program, an
annual report should be submitted to the Primacy Agency.
The contents of the report should:
a. Identify special concerns that occurred 1n the
watershed and how they were handled (example:
herbicide usage, new construction, etc.).
b. Summarize other activities In the watershed such
as logging, hunting, water quality monitoring,
etc.
c. Project what adverse activities are expected to
occur in the future and describe how the utility
expects to address them.
Agreements/Land Ownership
The goal of a watershed management program 1s to achieve the
highest level of raw water quality practicable. This 1s
particularly critical to an unflltered surface supply.
1. The utility will have maximum opportunity to realize
this goal 1f they have complete ownership of the
watershed. Describe efforts to obtain ownership, such
as any special programs or budget. When complete
ownership of the watershed 1s not practical, efforts
should be taken to gain ownership of critical elements,
such as, reservoir or stream shoreline, highly erodable
land, and access areas to water system facilities.
2. Where ownership of land 1s not possible, written
agreements should be obtained recognizing the watershed
as part of a public water supply. Maximum flexibility
should be given to the utility to control land uses
which could have adverse effect on the water quality.
Describe such agreements.
3. Describe how the utility ensures that the landowner
complies with these agreements.
-------
APPENDIX K
SANITARY SURVEY
-------
APPENDIX K
SANITARY SURVEY
The SWTR requires that an on-site Inspection be conducted each year
as outlined 1n Section 3. It 1s recommended that at the onset of
determining the classification of a source water that a detailed sanitary
survey be conducted. In addition, It 1s recommended that a sanitary
survey such as contained In this appendix be conducted every 3 to 5 years
by both filtered and unflltered systems to ensure that the quality of the
water and service Is maintained. This time period 1s suggested since the
time and effort needed to conduct the comprehensive survey makes 1t
Impractical for It to be conducted annually. A periodic sanitary survey
is also required under the Total Coliform Rule for systems collecting
fewer than 5 samples/month. The survey must be conducted every 5 years
for all systems except for protected ground water systems which disinfect.
These systems must conduct the survey every 10 years.
The sanitary survey Involves three phases, Including planning the
Survey, conducting the survey and compiling the final report of the .
survey, as will be presented 1n the following pages.
1. Planning the Survey
Prior to conducting or scheduling a sanitary survey, there
should be a detailed review pf the water system's file to
prepare for the survey. The review should pay particular
attention to past sanitary survey reports and correspondence
describing previously Identified problems and their solutions.
These should be noted, and action/Inaction regarding these
problems should be-specifically verified in the field. Other
Information to review Includes: any other correspondence,
water system plans, chemical and microbiological sampling
results, operating reports, and engineering studies. This
review will aid In the familiarization with the system's past,
history and present conditions, and the agency's past interac-
tions with the. system.
The Initial phase of the water quality review will be carried
out prior to conducting the survey as well, and will consist
of reviewing the water system's monitoring records. Records
should be reviewed for compliance with all applicable microbi-
ological, inorganic chemical, organic chemical, and radiologi-
cal contaminant MCLs, and also for compliance with the
monitoring requirements for those contaminants. The survey
K-l
-------
will provide an opportunity to review these records with the
utility, and to discuss solutions to any MCL or Monitoring
violations. The survey will also provide an opportunity to
review how and where samples are collected, and how field
measurements {turbidity, chlorine residual, fluoride, etc.)
are made. Points to cover include:
a. Is the system in compliance with all applicable MCLs
(organic chemical, inorganic chemical, aicrobiological,
and radiological)?
b. Is the system In compliance with all monitoring require-
ments?
The pre-survey file review should generate a list of items to
check in the field, and a list of questions about the system.
It will also help to plan.the format of the survey and to
estimate how much time it may take. The next step is to make
the initial contact with the system management to establish
the survey date(s) and time. Any records, files, or people
that will be referenced during the survey should be mentioned
at the outset. Clearly laying out the intent of the survey up
front will greatly help in manaaing the system, and will
ensure that the survey goes smoothly without a need for repeat
trips.
Conducting the Survey
The on-site portion of the survey is the most important and
will involve interviewing those in charge of managing the
water system as well as the operators and other technical
people. The survey will also review all major system compo-
nents from the source(s) to the distribution system. A
standard form is frequently used to ensure that all major
components and aspects of each system are consistently
reviewed. However, when in the field, it is best to have an
open mind and focus most attention on the specifics of the
water system, using the form only as a guide. The surveyor
should be certain to be on time when beginning the survey.
This consideration will help get the survey started smoothly
with the operator and/or manager.
As the survey progresses, any deficiencies that are observed
should be brought to the attention of the water system
personnel, and tne problem and the corrective measures should
be discussed. It is far better to clarify technical details
and solutions while standing next to the problem than it is to
do so over the telephone. Points to cover include:
a. Is the operator competent in performing the necessary
field testing for operational control?
-------
b. Are testing facilities and equipment adequate, and do
reagents used have an unexpired shelf life?
c. Are field and other analytical instruments properly and
regularly calibrated?
d. Are records of field test results and water quality
compliance monitoring results being maintained?
e. Conduct any sampling which will be part of the survey.
Also, detailed notes of the findings and conversations should
be taken so that the report of the survey will be an accurate
reconstruction of the survey.
Specific components/features of the system to review and some
pertinent questions to ask are:
A. Source Evaluation
ATI of the elements for a source evaluation enumerated below
may also be part of a Wellhead Protection Program.
1. Description: based on field observations and
discussion with the operator, a general charac-
terization of the watershed should be made.
Features which could be Included 1n the descrip-
tion are:
a. Area of watershed or recharge area.
b. Stream flow.
c. Land usage (wilderness, farmland, rural
housing,, recreational, commercial, Indus-
trial, etc.).
d. Oegree of access by the public to watershed.
e. Terrain and soil type.
f. Vegetation.
g. Other.
2. Sources of contamination in the watershed or
sensitive areas surrounding wells or well fields
should be Identified. Not only should this be
determined by physically touring and observing
the watershed and its daily uses, but the survey-
or should also actively question the water system
manager about adverse and potentially adverse
K-3
-------
activities In the watershed. An example of types
of contamination Includes:
a. Man Made.
1. Point discharges of sewage, storm-
water, and other wastewater.
2. On-site sewage disposal systems.
3. Recreational activities (swlonlng,
boating, fishing, etc.).
4. Hitman habitation.
5. Pesticide usage.
6. Logging.
7. Highways or other roads from which
there might be spills.
8. Commercial or Industrial activity.
9. Solid waste or other disposal facili-
ties.
10. Barnyards, feed lots, turkey and
chicken farms and other concentrated
domestic animal activity.
11. Agricultural activities such as graz-
ing, tillage, etc., which affects
soil erosion, fertilizer usage, etc.
12. Other.
b. Naturally Occurring.
1. Animal populations, both domestic and
wild.
2. Turbidity fluctuations (from precipi-
tation, landslides, etc.).
3. Fires.
4. Inorganic contaminants from parent
materials (e.g., asbestos fibers).a
5. Algae blooms.
K-4
-------
6. Other.
This 11st 1s by no means all Inclusive.
The surveyor should rely principally on his
observations and thorough questioning
regarding the unique properties of each
watershed to completely describe what aay
contaminate the source water.
3. Source Construction.
a. Surface Intakes.
1. Is the source adequate in quantity?
2. Is the best quality source or loca-
tion in that source being used?
3. Is the Intake protected from icing
problems if appropriate?
4. Is the intake screened to prevent
„ entry of debris, and are screens
maintained?
5. Is animal activity controlled within
the 1 mediate vicinity of the Intake?
6. Is there a raw water sampling tap?
b. Infiltration Galleries.
1. Is the source adequate in quantity?
2. Is the best quality source being used?
3. Is the 11d over the gallery water-
tight and locked?
4. Is the collector In sound condition
and maintained as necessary?
5. Is there a raw water sampling tap?
c. Springs.
1. Is the source adequate in quantity?
2. Is there adequate protection around
the spring such as fencing to control
the area within 200 feet?
K-5
-------
3. Is the spring constructed to best
capture tne spring flow and exclude
surface water Infiltration?
4. Are there drains to divert surface
water from the vicinity of the
spring?
5. Is the collection structure of sound
construction with no leaks or cracks7
6. Is there a screened overflow and
drain pipe?
7. Is the supply Intake located above
the floor and screened?
8. Is there a raw water sampling tap?
Catchment and Cistern.
1. Is source adequate In quantity?
2. Is the cistern of adequate size?
3. Is the catchment area protected from
potential contamination?
4. Is the catchment drain properly
screened?
5. Is the catchment area and cistern of
sound construction and In good condi-
tion?
6. Is catchment constructed of approved
non-toxic, non-leaching material?
7. Is the cistern protected from contam-
ination — manholes, vents, etc?
8. Is there a raw water tap?
Other Surface Sources.
1. Is the source adequate In quantity?
. 2. Is the best possible source being
used?
3. Is the Immediate vicinity of the
source protected from contamination?
-------
4. Is the structure 1n good condition
and properly constructed?
5. Is there a raw water sampling tap?
4. Pumps, Pumphouses, and Controls.
a. Are all Intake pumps, booster pumps, and
other pumps of sufficient capacity?
b. Are all pumps and controls operational and
maintained properly? •
c. Are check valves, blow off valves, water
meters and other appurtenances operated and
maintained properly?
d. Is emergency power backup with automatic
start-up provided and does it work (try
it)?
e. Are underground compartments and suction
wells waterproof?
f. Is the interior and exterior of. the pump-
house in good structural condition and
properly maintained?
g. Are there any safety hazards (electrical or
mechanical) in the pumphouse?
h. Is the pumphouse locked and otherwise
protected against vandalism?
i. .Are water production records maintained at
the pumphouse?
5. Watershed Management (controlling contaminant -
sources): The goal of the watershed management
program is to identify and control contaminant
sources in the watershed (see Section 3.3.1 of.
this document, "Watershed Control Program").
Under ideal conditions each source of contamina-
tion identified in 2 will already have been
identified by the utility, and some means of
control instituted, or a factual determination
made that its impact on water quality is insig-
nificant. To assess the degree to which the
watershed management program is achieving its
goal, the following types of inquiries could be
made:
K-7
-------
a. If the watershed Is not entirely owned by
the utility, have written agreements been
made with other land owners to control land
usage to the satisfaction of the utility?
Are appropriate regulations under the
contract of state/local -department of
health In effect?
b. Is the utility making efforts to obtain as
complete ownership of the watershed as
possible? Is effort directed to control
critical elements?
c. Are there means by which the watershed Is
regularly Inspected for new sources of
contamination or trespassers where access
Is limited?
d. Are there adequately qualified personnel
employed by the utility for Identifying
watershed and water quality problems and
who are given the responsibility to correct
these problems?
e. Are raw water quality records kept to
assess trends and to assess the impact of
different activities and contaminant con-
trol techniques in the watershed?
f. Has the system responded adequately to
concerns expressed about the source or
watershed in past sanitary surveys?
g. Has the utility Identified problems in its
yearly watershed control reports, and if
. so, have these problems been adequately
addressed?
h. Identify what other agencies have control
or jurisdiction In the watershed. Does the
utility actively Interact with these agen-
cies to see that their policies or activi-
ties are consistent with the utility's goal
of maintaining high raw water quality?
Treatment Evaluation
1. Disinfection.
a. Is the disinfection equipment and disinfec-
tant appropriate for the application
(chloramines, chlorine, ozone, and chlorine
K-8
-------
dioxide are generally accepted disinfec-
tants)?
Are there back-up disinfection units on
line in case of failure, and are they
operational?
Is there auxiliary power with automatic
start up in case of power outaae? Is it
tested and operated on a regular basis,
both with and without load?
Is there an adequate quantity of disinfec-
tant on hand and 1s it properly stored
(e.g., are chlorine cylinders properly
labeled and chained)?
In the case of gaseous chlorine, is there
autonatic switch over equipment when cylin-
ders expire?
Are critical spare parts on hand to repair
disinfection equipment?
Is disinfectant feed proportional to water
flow?
Are daily records kept of disinfectant
residual near the first customer from which
to calculate CTs?
Are production records kept from which to
determine CTs?
Are CTs acceptable based on the level of
treatment provided (see Surface Water
Treatment Rule for CT values, and
Sections 3 and 5 of this guidance manual
for calculation of CT).
Is a disinfectant residual maintained 1n
the distribution system, and are records
kept of daily measurements?
If gas chlorine is used, are adequate
safety precautions being followed (e.g.,
exhaust fan with intake within six inches
of the floor, self-contained breathing
apparatus that is regularly tested, regular
safety training for employees, ammonia
bottles and/or automatic chlorine detec-
tors)? Is the system adequate to ensure
-------
the safety of both the public and the
employees In the event of a chlorine leak?
2. Other.
a. Are other treatment processes appropriate
and are they operated to produce consis-
tently high water quality?
b. Are pumps, chemical feeders, and other
mechanical equipment 1n good condition and
properly maintained?
c. Are controls and Instrumentation adequate
for the process, operational, well main-
talned and calibrated?
d. Are accurate records maintained (volume of
water treated, amount of chemical used,
etc.)?
e. Are adequate supplies of chemical on hand
and properly stored?
f. Are adequate safety devices available and
precautions observed?
Sections of a sanitary survey pertaining to
systems containing filtration facilities have
been omitted, as this section of the guidance
document pertains to non-f1lter1ng systems.
Distribution System Evaluation
After water has been treated, water quality must be
protected and maintained as it flows through the
distribution system to the customer's tap. The follow-
ing questions pertain to the water purveyor's ability to
maintain high water quality during storage and distribu-
tion.
1. Storage.
a. Gravity.
1. Are storage reservoirs covered and
otherwise constructed to prevent,
contamination?
2. Are all overflow lines, Vents, drain-
lines, or cleanout pipes turned down-
ward and screened?
K-10
-------
3. Are all reservoirs inspected regular-
iy?
4. Is the storage capacity adequate for
the system?
5. Does the reservoir (or reservoirs)
provide sufficient pressure through-
out the system?
6. Are surface coatings within the res-
ervoir 1n good repair and acceptable
for potable water contact?^
7. Is the hatchcover for the tank water-
tight and locked?
8. Can the reservoir be isolated from
the system?
9. Is adequate safety equipment (caged
ladder, OSHA approved safety belts,
etc.) In place for climbing the tank?
10. Is the site fenced, locked, or other-
wise protected against vandalism?
11. Is the storage reservoir disinfected
after repairs are made?
12. Is there a scheduled program for
cleaning storage reservoir sediments,
slime on floor and side walls.
Hydropneumatic.
1. Is the storage capacity adequate for
the system?
2. Are instruments, controls, and equip-
ment adequate, operational, and main-
tained?
3. Are the Interior and exterior surfac-
es of the pressure tank In good con-
dition?
.4. Are tank supports structurally sound?
5. Ooes the low pressure cut 1n provide
adequate pressure throughout the
entire system?
K-ll
-------
6. Is the punp cycle rate acceptable
(not more than 15 cycles/hour)?
Cross Connections.
a. Is the system free of known uncontrolled
cross connections?
b. does the utility have a cross connection
prevention program, including annual test-
ing of backflow prevention devices?
c. Are backflow prevention devices. Installed
at all appropriate locations (wastewater
treatment plant, Industrial locations,
hospitals, etc.)?
es of the year?
b. Oo all construction materials meet AWWA or
equivalent standards?
c. Are all services metered and are meters
read?
d. Are plans for the system available and
current?
e. Does the system have an adequate mainte-
nance program?
Is there evidence of leakage In the
system?
Is there a pressure testing program?
Is there a regular flushing program?
Are valves and hydrants regularly
exercised and maintained?
Are AWWA standards for disinfection
followed after all repairs?
Are there specific bacteriological
criteria and limits prescribed for
new line acceptance or following line
repairs?
Other.
a.
pressures and flows maintained
K-12
-------
Describe the corrosion control pro-
gran.
Is the system Interconnected with
other systems?
Management/Operation
1. Is there an organization that 1s responsible for
providing the operation, maintenance, and manage-
ment of the water system?
2. Does the utility regularly summarize both current
and long-term problems identified In their water-
shed, or other parts of the system, and define
how they Intend to solve the problems I.e., Is
their planning mechanism effective; do they
follow through with plans?
3. Are customers charged user fees and are collec-
tions satisfactory?
4. Are there sufficient personnel to operate and
manage the system?
5. Are personnel (Including management) adequately
trained, educated, and/or certified?
6. Are operation and maintenance manuals and manu-
facturers technical specifications readily avail-
able for the system?
7. Are routine preventative maintenance schedules
established and adhered to for air components of
the water system?
8. Are sufficient tools, supplies, and maintenance
parts on hand?
9. Are sufficient operation and maintenance records
kept and readily available?
10. Is an emergency plan available and usable,, and
are employees aware of it?
11. Are all facilities free from safety defects?
When the survey Is completed, 1t 1s always preferable to
briefly summarize the survey with the operator(s) and
management. The main findings of the survey should be
reviewed so 1t is clear that there are not misunder-
standings about findings/conclusions. It Is also good
K-13
-------
to thank the utility for taking part In the survey,
arranging Interviews with employees, gathering and
. explaining their records, etc. The Information and help
which the utility can provide an Invaluable to a
successful survey, and every attempt should be made to
continue a positive relationship with the "system.
Reporting the Survey
A final report of the survey should be completed as soon as
possible to formally notify the system and other agencies of
the findings. There Is no set or necessarily best format for
doing so, and the length of the report will depend on the
findings of the survey and size of the system.." Since the
report may be used for future compliance actions and Inspec-
tions, .1t should Include as a minimum: 1) the date of the
survey; 2) who was present during the survey; 3) the findings
of the survey; 4) the recommended Improvements to Identified
problems; and 5) the dates for completion of any Improvements.
Any differences between the findings discussed at the conclu-
sion of the survey and what's Included In the final report
should be discussed and clarified with the utility prior to
sending out the final report. In other words, the utility
should be fully aware ot the contents of the final report
before receiving It.
K-14
-------
APPENDIX L
SMALL SYSTEM CONSIDERATIONS
-------
APPENDIX L
SMALL SYSTFM CONSIDERATIONS
Introduction
Under the provisions of the SWTR, systems with fewer than 500
service connections may be eligible for an exemption. Guidance on the
requirements for an exemption 1s provided In Section 9. For systems which
are not eligible for an exemption, compliance with the SWTR 1s/mandatory.
It 1s recognized that the majority (approximately 75 percent) of people In
the United States are served by a relatively small number of large
systems. However, most water systems In the United States are small. For
small systems, compliance with the various provisions of the SOMA has
traditionally been a problem. Records show small systems have a
disproportionately higher Incidence of drinking water quality and
monitoring difficulties. The*reasons for these difficulties can generally
be broken down Into the following three categories:
Economics
Treatment Technologies
Operations (lack of qualified personnel)
Small water systems typically face severe economic constraints.
Their lack of operating revenues results in significant limitations on
their ability to respond to the requirements of the SDWA. These systems
cannot benefit from the economies of scale which are available to larger
systems.
The second difficulty facing the small systems has been the lack of
appropriate treatment technologies. Although methods for removing most of
the contaminants known to occur In drinking water are available, many of
these technologies have only recently been scaled down for the smaller
systems.
The third problem which has traditionally plagued small systems is
the lack of well trained operators. This deficiency Is the result of many
combined factors. First of all, many of these operators are employed only
on a part-time basis or if they are employed on a full-time basis they
have a myriad of additional duties. In addition, the operator's technical
L-l
-------
background nay be United as well. This results from the low salary of
the position, which 1s uninviting to qualified operators. Also, In spite
of the requirement of retaining certified operators upheld in aany states,
It seems to be difficult to enforce this requirement in small systems.
The purpose of this appendix 1s to provide assistance to the Primacy
Agency 1n defining the problems and potential solutions typically
associated with small systems. It Is beyond the scope of this document to
provide an Indepth dlcusslon of the needs of small systems. However, over
the past several years the needs of the small water systems have been
recognized to be of primary concern and numerous workshops, seminars and
committees have been attempting to more clearly define workable solutions.
A partial listing of the papers, reports and proceedings which discuss
problems and solutions pertaining to small systems beyond that which Is
possible 1n this manual Is presented 1n the reference list of this
appendix.
Economics
One of the most severe constraints of small systems Is the small
economic base from which to draw funds. Certain treatment and services
must be provided for a community regardless of how few people are served.
Thus, as the number of connections to the system decrease, the cost per
connection Increases. The economic limitations of small utilities makes
1t difficult to provide needed upgrading of existing facilities or an
adequate salary to maintain the employment of a qualified operator to
monitor and maintain the system. Adding to the severity of the economic
hardships of small systems Is the fact that many of the small water
systems are privately owned, with private ownership Increasing as system
size decreases. The ownership of the plant presents difficulties since
privately owned systems are subject to rate controls by the local public
utility commission, are not eligible for public grants and loans, and may
find commercial loans hard to obtain.
Financing options for small systems Include; federal and state loan
and grant programs, federal revenue sharing and revenue bonds (for
municipal systems) and loans through the United States Small Business
Administration (SBA) and use of Industrial development bonds or prlvatlza-
-------
Hon (for private utilities). These options are explained 1n. greater
detail 1n the "Guidance Manual - Institutional Alternatives for.Snail
Hater Systems" (AWWA, 1986). The following paragraphs will explain some
existing options which nay ease the hardship of financing snail water
treatment facilities.
The najor cause of small system difficulties arises fron the lack of
funds and resources. It 1s therefore In the best Interest of snail
utilities to expand their economic base and the resources available to
them, to achieve the economies of scale available to larger systems.
RegionalIzatlon 1s the physical or operational union of snail systems to
effect this goal. This union can be accomplished through the physical
Interconnection of two or more small systems or the connection of a
smaller system to a pre-existing larger system. Hater supply systems can
also join together for the purchase of supplies, materials, engineering
services, billing and maintenance. The union of the small systems
Increases the population served, thereby dispersing the operational costs
and decreasing the cost per consumer.
The creation of utility satellites Is another form of reglonallza-
tion. A satellite utility Is one which taps Into the resources of an
existing larger facility without being physically connected to, or owned
by, the larger facility. The larger system may provide any of the
following for the smaller system:
1. Varying levels of technical operational, or managerial
assistance on a contract basis.
2. Hholesale treated water with or without additional services.
3. Assuming ownership, operation and maintenance responsibility,
when the small system is physically separate with a separate
source.
The formation of a satellite offers many advantages for both the
satellite and the parent utility. These advantages Include: an Improved
economy of scale for satellites, an expanded revenue base for the parent
utility, provisions of needed resources to satellites, the retention of
the satellites' local autonomy, Improved water quality management of the
satellite, improved use of public funds for publicly owned satellites.
-------
In order to create & sore definite structure for the union of
resources of water treatment facilities, water districts Bay be created.
Mater districts are formed by county officials and provide for the public
ownership of the utilities. The utilities in any given district would
combine resources and/or physically connect systems so that one or two
facilities would provide water for the entire district. The creation of
water districts creates eligibility for public monies, has the potential
for economies of size, facilitates the takeover or contract services with
publicly owned non-community systems and snail privately owned systems,
and offers a tax advantage. Drawbacks include subjection to politics, a
strong local planning effort is needed for success, and competition with
private enterprises.
The centralization of utilities can be taken one step further
through the creation of county utilities or even state utilities. The
government will create a board which may then act to acquire, construct,
maintain and operate any public water supply within Its district, the
system may provide water on its own or purchase water from any municipal
corporation. The board may adopt and administer rules for the construc-
tion, maintenance, protection and use of public water supplies and the
fixation of reasonable rates for water supplies. The cost of construction
and/or upgrading of facilities may be defrayed through the issuance of
bonds and/or property assessment. As with all the alternatives, the
creation of government control of the utilities has its advantages and
disadvantages. The advantages include: the creation of central
management, creation of economy of scale for utilities, eligibility for
public grants and loans, savings through centralized purchasing, manage-
ment, consultation, planning and technical assistance, and possible
provision for pool of trained operators. The disadvantages include the
subjectivity to politics, the slow response caused by bureaucracy, and
competition to private contractors.
Treatment Technologies
The high cost of available treatment technologies has limited their
use in small water supply systems. Recently prefabricated package plants
and individual treatment units have been developed to lessen these costs.
L-4
-------
At the present tine, the treatment technologies which are available to
enable systems to comply with the Safe Drinking Hater Act are Identified
to be the following:
Package plants
Slow-sand filters
Olatomaceous earth filters
Cartridge filtration
A brief discussion of each treatment method 1s provided below.
Pflckaqg Plants
Clarification and filtration units which require minimal assembly In
the field can now be manufactured. To minimize required operator skill
level and operational attention, the equipment should be automated.
Continuous effluent turbidity and disinfectant residual monitoring systems
with alarms and emergency shutdown provisions are features that safeguard
water quality and should be provided for unattended plants.
Slow-Sand Filters
Slow-sand filters are applicable to small water supply systems.
Their proven record of effective removal of turbidity and Giardia cysts
makes them suitable for application where operational attention Is
minimal. Since no chemicals other than a disinfectant are needed, and no
mechanical equipment 1s Involved, the required operator skill level 1s the
lowest of the filtration alternatives available to small systems.
Platofliaceous Earth Filters
Diatomaceous earth (DE) pressure and vacuum filters can be used on
relatively low turbidity surface waters (less than 1 to 2 NTU) for removal
of turbidity and Glardia cysts. DE filters can effectively remove
particles as small as 1 micron, but would require coagulating chemicals
and special filter aids to provide significant virus removal.
Cartridge Filters
Cartridge filters using mlcroporous ceramic filter elements with
pore sizes as small as 0.2 um'may be suitable for producing potable water,
In combination with disinfection, from raw water supplies containing
moderate levels of turbidity, algae, protozoa and bacteria. The advantage
to a small system, 1s, with the exception of chlorlnatlon, that no other
chemicals are required. The process is one of strictly physical removal
-------
of small particles by straining as the water passes through the porous
membranes. Other than occasional cleaning or membrane replacement.,
operational requirements are not complex and do not require skilled
personnel *
Selection of a Filtration Technoloov
The criteria for selection of a filtration technology for a small .
community are essentially the same as those for a larger community. That
Is, the utility must first screen the complete 11st of available
alternatives to eliminate those which are either not technically suited to .
the existing conditions (Table 4-1) or not affordable by the utility.
Remaining alternatives should then be evaluated based on both cost
(capital, annual, and life-cycle) and non-cost bases (operation and
maintenance, technical requirements versus personnel available; flexibili-
ty regarding future needs; etc.). In these evaluations it should be noted
that even though automated package plants are cost-competitive with slow
sand filters, their operation requirements to achieve optimum performance
could be complicated. Also, the maintenance requirements for package
plants would be mechanically and electrically oriented and might require
a maintenance agreement with the manufacturer.
During the process of Installing the treatment system, interim
measures should be taken to ensure the delivery of a reasonably safe water
to the consumers. In addition to the available interim measures listed In
Section 9.3, temporary Installation of mobile filtration plants may be
possible. These trailer-mounted units are sometimes available from state
agencies for emergencies, but more often may be rented or leased from an
equipment manufacturer.
Modification of Existing Filtration Systems
Small treatment systems that are already In existence should comply
with the performance criteria of the SWTR. If the systems are not found
to be performing satisfactorily, modifications to the existing process may
be required. Improvement 1n treatment efficiency depends on the type of
filtration system 1n use. Operation of slow sand filters could be checked
for bed depth, short-circuiting, excessive hydraulic loading, and for the
need to pretreat the raw water. Infiltration galleries, or sometimes,
roughing filters ahead of a slow sand filter may provide for better
L-6
-------
performance by reducing the solids load on the filters. However, the
design criteria and costs for this alternative have not yet been defined.
Site specific studies may be required before roughing filters could be
used to achieve compliance with the regula- tlons. Dlatoaaceous earth
(DE) filters should be checked for appropriate precoat and body feed
application, hydraulic loading, grade (size) of DE being used, and
possible need for chemical pretreatment. Package plants would have to be
checked process-by-process, similar to the system used for a conventional
plant. Other filtration processes would have to be checked for hydraulic
loading rate, appropriateness of the filter material (pore size), and
possible need for additional pretreatment;
Disinfection
Disinfection (CT) requirements for small systems can be met 1n
several different ways. The most obvious method of maintaining a
disinfectant residual 1n the distribution system 1s to add disinfectant at
one or more additional locations. An alternate method 1s to Increase the
disinfectant dose at the existing application po1nt(s). The latter
alternative, however, may Increase disinfectant byproducts, Including
THMs, in the system.
t •
If it is a relatively short distance between the treatment system
and the first customer, additional contact time can be provided so that
the disinfectant dose does not have, to be Increased beyond desirable
residuals. Two specific methods of Increasing contact time for small
systems are 1) Installing a pressure vessel or closed storage vessel,
baffled to provide adequate contact time, or 2) constructing a looped
pipeline, on the finished water line between the filtration-disinfection
system and the first customer. The feasibility of either of these methods
would depend on system specifics that Include size, physical conditions,
and cost.
If It 1s not practical to provide additional storage time to achieve
the desired CT, an alternate, «ore effective disinfectant may be used. An
alternate disinfectant may provide a sufficient CT without altering the
system configuration.
L-7
-------
Qpcratlgns
Water treatment facilities need to be operated properly in order to
achieve maximum treatment efficiencies. There Is currently a lack of well
trained operators at many snail treatment plants. The main cause 1s lack
of awareness of the Importance of correct plant operation, lack of
training programs, lack of enforcement of the requirement for employment
of a certified operator and lack of funds to employ such an operator.
Small systems may wish to Implement a circuit rider/operator
program. In this program a qualified, certified, experienced operator
works for several water supply systems. The rider can either directly
operate the plants, or provide technical assistance to Individual plant
operators, by acting as a trainer through on-the-job supervision. The
latter would be preferable since it could create a pool of well trained
operators.
The main cause of inadequately trained operators is the lack of well
established training programs. Until such training programs are begun,
systems must depend on other training means, such as seminars and books.
One resource which may be helpful 1n running the plant Is "Basic
Management Principles for Small Water Systems • An AWWA. Small-Systems
Resource Book", 1982.
Most package plant manufacturers' equipment manuals Include at least
brief sections on operating principles, methods for establishing proper
chemical dosages, instructions for operating the equipment, and trouble-
shooting guides. An individual who studies these basic Instructions and
receives comprehensive start-up training should be able to operate the
equipment satisfactorily. These services are vital to the successful
performance of a package water treatment plant and should be a requirement,
of the package plant manufacturer. The engineer designing a package plant
facility should specify that start-up and training services be provided by
the manufacturer, and also should consider requiring the manufacturer to
visit the plant at 6-month and 1-year Intervals after start-up to adjust
the equipment, review operations, and retrain operating personnel.
Further, this program should be ongoing and funds should be budgeted every
year for at least one revisit by the package plant manufacturer.
L-8
-------
Another way for snail systems to obtain qualified plant operation
would be to contract the services of administrative, operations, and/or
maintenance personnel from a larger neighboring utility, government
agencies, service companies or consulting firms. These organizations
could supply assistance In financial and legal planning, engineering,
purchasing accounting and collection services, laboratory support,
licensed operators or operator training, treatment and water quality
assurance, regulatory liaison, and/or emergency assistance. .Through the
contracting of these services the utility provides for the resources
needed, Improves water quality management, and retains Its autonomy.
However, 1f and when the contract Is terminated, the utility returns to
Its original status.
References
American Water Works Association. Basic Management Principles for Small
Water Systems. 1982.
American Water Works Association. Design and Construction of Small Water
Systems. 1984.
Kelly, Gldley, Blair and Wolfe, Inc. Guidance Manual - Institutional
Alternatives for Small Water Systems. AWWA Research Foundation Contract
79-84, 1986.
L-9
-------
APPENDIX M
PROTOCOL FOR DEMONSTRATION
OF EFFECTIVE TREATMENT
-------
APPENDIX M
PROTOCOL FOR DEMONSTRATION
OF EFFECTIVE TREATMENT
This appendix presents approaches which can be taken to demonstrate
overall effective removal and/or inactivation of fiiardia cysts.
M.l Demonstration for Alternate Technology
Systems using a filtration technology other than those enumerated
In the SWTR may demonstrate the effectiveness of the treatment process
through pilot or full scale testing. As a minimum, testing should be
conducted when the source exhibits its worst case annual conditions. Some
systems may have two periods of "worst case" water quality Including the
cold water 1n winter or algae blooms during the summer.
Pilot units should Include the following:
filtration rate of the pilot system equal to filtration
rate on full scale unit
pilot filter diameter greater than or equal to 50 times
the media diameter, (Robeck, et al 1959)
- media diameter, depth, and size gradation should be
Identical to full scale,
coagulant dosing Identical to full scale
any mixing and settling occurring before filtration In
the full scale plant should be reproduced as closely as
possible in the pilot. Mixing should be of the same G
value(s), and the detention time for settling should be
close to the average flow detention time for the
projected full scale plant.
According to the SWTR, alternate technologies must be capable of
meeting the same turbidity performance criteria of slow sand filtration
systems. Thus the filtered water from the process should be monitored
continuously or with grab samples every four hours for turbidity.. The
requirement for meeting turbidity performance has been established to
M-l
-------
ensure that there **111 be no Interference of turbidity with virus
inactivation through disinfection.
Following the demonstration of neeting the turbidity requirements,
the level of Giardia cyst removal achieved must be determined. The
protocol in M.2 may be followed for this demonstration.
M.2 Particle Si2e Analysis Demonstration for Giardia Cvst R
trtfwvfll Credit
Particle size analysis may be used to demonstrate the level of
actual Giardia cyst removal provided by the system. This demonstration
can be done using samples from the full scale plant or a pilot unit.
In the case of either a full scale or pilot scale demonstration,
removal of particles in the range of 5 to IS urn In diameter should be
determined using an electronic particle counter that has been calibrated
with latex spheres. If a light blockage device is used (e.g. MAC) this
calibration should have been done during installation of the device. The
calibration should be checked before taking measurements for the purposes
of this demonstration. Samples should be diluted appropriately to ensure
that measurements do not reflect coincident error. Coincident error
results when more than one particle passes the detector at one time,
causing an inaccurate particle count and diameter measurement. An
electrical sensing zone device (e.g. Coulter Counter or Elzone) may also
be used. Appropriate dilutions, electrolyte strength, and calibration
procedures should be followed (these are scheduled to be outlined in the
17th edition of Standard Methods). When using an electrical sensing zone
Instrument, an orifice no larger than 125 urn and no smaller than 40 urn
should be used since only particles between 2% and 40% of the orifice
diameter are accurately sized and counted (Karuhn et al 1975)„
Samples of the filter Influent and effluent should be taken 5
minutes after the backwashed filter is placed In operation, and every 30
minutes thereafter for the first 3 hours of operation, followed by hourly
samples up until backwash (Wiesner et al 1987). All samples should show
at least a 2-log removal. The SWTR establishes an overall treatment
requirement of 3-log Giardia cyst removal/inactivation. Thus, disinfec-
M-2
-------
tlon «w$t be provided to supplement the particulate removal and meet this
requirement.
Samples from repeated filter runs may be averaged at each sampling
time, but samples should not be averaged within one filter run.
Additional suggestions on particle counting technique (Wiesner
1985):
1) If particle counts are not determined immediately upon
sampling (within 10 minutes) samples should be diluted.
2) For an electrical sensing zone measurement, samples
should be diluted 1:5 to 1:20 with a "particle-free"
electrolyte solution (approximately 1% NaCI) containing
100 particles per ml or fewer.
3) For a light blockaae measurement, particle free water
should be used to dilute samples.
4) Dilutions should be done, to produce particle concentra-
tions as close to the tolerance for coincident error as
possible to minimize background counts.
5) Particle counts should be determined within 8 hours of
sampling.
6) All sampling vessels should be washed with laboratory
detergent, double rinsed 1n particle free water, and
rinsed twice with the water being sampled at the time
of sampling.
The log reduction of particles in the size range of 5 to 15 um In
size can be assumed to correspond to the log reduction of Glardia cysts
which would be achieved.
M.3 Demonstration for Increased Turbidity Allowance
Based upon the requirements of the SWTR, the minimum turbidity
performance criteria for systems using conventional treatment or direct
filtration is filtered water turbidity less than or equal to 0.5 NTU in
95 percent of the measurements taken each month. However, at the dis-
cretion of the Primacy Agency, filtered water turbidity levels of less
than or equal to 1 NTU in 95 percent of the measurements taken every month
M-3
-------
My be permitted on a case-by-case basis depending on the capability of
the total system to remove and/or Inactivate at least 99.9 percent of
Glardia lamb!la cvsts.
Treatment plants that use settling followed by flltratton, or direct'
filtration are generally capable of producing a filtered water with a
turbidity of 0.2 NTU or less. The most likely cause of high turbidities
In the filtered water 1s Incorrect coagulant dosing (O'Hella, 1974).
Regardless of the turbidity of the raw or finished water, coagulant
addition at some point prior to filtration Is required to destabilize
particles for removal In the filter. Only plants documenting continuous
coagulant feed prior to filtration should be eligible for being allowed
higher filtered water turbidities than the 0.5 NTU requirement. At plants
that continuously feed coagulant and do not meet the 0.5 NTU requirement,
a series of jar tests, and perhaps sand column filtration tests (1n batch)
should be performed to evaluate the optimum coagulant dose for turbidity
removal.
In the event that plants can document continuous coagulant feed,
and, after running the plant under conditions determined 1n batch testing
to be optimal for turbidity removal, still do not meet the 0.5 NTU
requirement, effective filtration status may still be appropriate. This
would further be supported 1f it can be shown that the full scale plant
1s capable of achieving at least a 2-log reduction 1n the concentration
of particles between 5 and 15 um in size through particle size analysis
as outlined 1n Section M.2.. Where a full scale plant does not yet exist,
appropriately scaled-down pilot filters might be used for such a
demonstration.
Disinfection
The level of disinfection could also be considered for determining
when to allow a higher turbidity performance criterion for a system. For
example, 1f a system achieves 3-log Giardia cyst inactivation through
disinfection, as determined by CT values, It may be appropriate to allow
higher filtered water turbidities (i.e., greater than 0.5 NTU but less
than 1 NTU in 95 percent of the measurements and never exceeding 5 NTU).
M-4
-------
The expected level of fecal contamination and Siardia cyst
concentrations 1n the source water should be considered 1n the above
analysis. High levels of disinfection (e.g., 2 to 3-log Inactlvatlon of
Giardia cysts), in addition to filtration which achieves less than 0.5 NTU
In 95 percent of the measurements may be appropriate, depending upon
source water quality. Further guidance on the level of disinfection to be
provided for various source water conditions 1s provided 1n Section 4.4.2.
In all cases the minimum disinfection to be provided must supplement the
particulate removal to ensure at least a 3-log 61ard1a cyst remov-
al/1nact1vat1on.
References
American Public Health Association; American Hater Works Association;
Water Pollution Control Federation. Standard Methods for the Examination
of Water and Wastewater. 17th ed. (supplement), September 1989.
Coulter Electronics 600 W. 20th Street, Hlaleah, FL 33010-2428
Karuhn, R.; Davles, R.; Kaye, B.; Clinch, M. Studies on the Coulter
Counter Part I. Powder Technology Volume II, pp. 157-171, 1975.
0'Helia, C. The Role of Polvelectrolvtes In Filtration Processes. EPA -
67012-74-032, 1974.
Robeck, G.; Woodword, R. L. Pilot Plants for Water Treatment Research,
Journal of Sanitary Engineering ASCE Vol. 85;SA4; 1, August 1959.
Wiesner, M. "Optimum Water Treatment Plant Configuration Effects of Raw
Water Characteristics," dissertation John Hopkins University, Baltimore,
MD, 1985.
Wiesner, M.; Rook, J. J.; Fiesslnaer, F. Optimizing the Placement of GAC
Filters, J. AWWA VOL 79, pp. 39-49, Dec 1987.
M-5
-------
APPENDIX N
PROTOCOLS FOR POINT-OF-USE
OEVICES
-------
Preface
The protocol presented in this paper can be applied to denonatrate the
effectiveness of new technologies as vail as point-of-use devices. The
avaluation presented hara daals with tha reaoval of particulates and
disinfection. In areas which pertain to disinfection, the guidelines
contained in Appendix G take precedence.
-------
CONTENTS
PREFACE
1. GENERAL
2. PERFORMANCE REQUIREMENTS
3. MICROBIOLOGICAL HATER PURIFIER TEST PROCEDURES
APPENDIX N-l SUMMARY FOR BASIS OF STANDARDS AND
TEST HATER PARAMETERS
APPENDIX N-2 LIST OF PARTICIPANTS IN TASK FORCE
APPENDIX N-3 RESPONSE BY REVIEH SUBCOMMITTEE TO
PUBLIC COMMENTS
-------
UNITED STATES
ENVIRONMENTAL PROTECTION AGENCY
Registration Division
Office of Pesticide Programs
Criteria and Standards Division
Office of Drinking Hater
GUIDE STANDARD AND PROTOCOL FOR
TESTING MICROBIOLOGICAL WATER PURIFIERS
Report of Task Force
Submitted April, 1986
Revised April, 1987
-------
1.1 Introduction
1. GENERAL
The subject of microbiological purification for waters of unknown micro-
biological quality repeatedly presents Itself to % variety of governmental and
non-governmental agencies, consumer groups, manufacturers and others. Exam-
ples of possible application of such purification capabilities irtclude:
- Backpackers and campers
- Non-standard military requirements
- Floods and other natural disasters
- Foreign travel and stations (however, not for extreme contamination
situations outside of the U.S.)
- Contaminated individual sources, wells and springs (however, not for
the conversion of waste water to microbiologically potable water)
- Motorhomes and trailers
Batch methods of water purification based on chlorine and iodine disin-
fection or boiling are well known, but many situations and personal choice-
call for the consideration of water treatment equipment. Federal agencies
specifically Involved in responding to questions and problems relating to
microbiological purifier equipment include!
- Registration Division, Office of Pesticide Programs (OPP), Environ-
mental Protection Agency (EPA)i registration of microbiological
purifiers (using chemicals);
- Compliance Monitoring . Staff, EPA: control of microbiological
purifier device claims (non-registerable products such as ultra-
violet units, ozonators, chlorine generators, others);
- U.S. Army Medical Bioengineering Research and Development Laboratory
(USAMBRDL), U.S. Army Natlck Research and Development Center and
other Army and military agencies: research and development for
possible field applications;
- Criteria and Standards Division, Office of Drinking Water (ODW),
EPA: Consideration of point-of-use technology as acceptable tech-
nology under the Primary Drinking Water Regulations; consumer
Information and service;
N-l
-------
- Drinking Water Research, Watar Engineering Research Laboratory
(WERL), EPA; responsible for water treatment technology rasa arch)
• Microbiology Branch, Haalth Effects Research Laboratory (HERL), EPA;
raaponaibla for study of haalth affacta ralatad to drinking watar
filtars.
A number of representatives of tha abova mentioned aganciaa provldad
axcallant participation in tha task forca to davalop microbiological tasting
protocols for water purifiars. Major participation was also provldad by the
following:
- A technical representative from the Vatar Quality Association;
- A technical representative from the Environmental Haalth Canter,
Department of Haalth and Welfare of Canadaj and
- An associate professor (microbiology) from the University of
Arizona.
1.2 Basic Principles
1.2.1 Definition
As set forth in EPA Enforcement Strategy and as supported by a Federal
Trade Conaission (FTC) decision (FTC v. Sibco Products Co., Inc., et al.,
Nov. 22, 1965) , a unit, in order to be called a microbiological water
purifier, must remove, kill or inactivate all types of disease-causing micro-
organisms from the water, including bacteria, viruses and protozoan cysts so
as to render the processed watar safe tox drinking. Therefore, to qualify, a
microbiological water purifier must treat or remove all types of challenge
organisms to meet specified standards.
¦
1.2.2 General Guide
The standard and protocol will be a general guide and, in some casas, may
present only the minimum features and framework for testing. While basic
features of the standard and protocol have been tested, it was not feasible to
conduct full-fledged testing for all possible types of units. Consequently,
protocol users should Include pre-testing of their units in a testing rig,
including the sampling techniques to be used. Where users of the protocol
find good reason to alter or add to the guide in order to meet specific
operational problems, to use an alternate organism or laboratory procedure, or
to respond to Innovative treatment units without decreasing the level of
N-2
-------
tasting or Altering the intent of the protocol, they should feel free to dc
so. For example, the OPP Registration Division might find it necessary to
amend the guide somewhat for different types of treatment units. Another
example would be ultraviolet (U.V.) units, which nay have specific require-
ments In addition to the guide protocol.
1.2.3 Performance-Based
The standard will be performance-based, utilizing realistic worst case
challenges and test conditions and use of the standard shall result in water
quality equivalent to that of a public water supply Meting the
microbiological requirements and intent of the National Primary Drinking Hater
Regulations.
1.2.4 Exceptions
A microbiological water purifier must remove, kill or inactivate all °
types of pathogenic organisms if claims are made for any organism. However,
an exception for limited claims may be allowed for units removing specific
organisms to serve a definable environmental need (i.e., cyst reduction units
which can be used on otherwise disinfected and microbiologically safe drinking
water, such as a disinfected but unfiltered surface water containing cysts.
Such units are not to be called microbiological water purifiers and should not'
be used as sole treatment for an untreated raw water.)
1.2.5 Wot to Cover Non-Microbiological Reduction Claims
The treatment of water to achieve removal of a specific chemical or other
non-microbiological substances from water will not be a part of this standard.
National Sanitation Foundation (NSF) Standards 42 (Aesthetic Effects) and 53
(Health Effects) provide partial guides for chemical removal and other claims
testing.
1.2.6 Construction and Information Exclusions
While the standard recommends safe, responsible construction of units
with non-toxic materials for optimum operation, all such items and associated
operational considerations are excluded as being beyond the scqpe of .the
standard. Included in the exclusion are materials of construction, electrical
and safety aspects, design and construction details, operational instructions
and information, and mechanical performance testing.
1.2.7 Research Needs Excluded
N-3
-------
The guide standard and protocol Bust represent a practical tasting
program and not include researSR reeosBJendations. For example, consideration
of outant organisms or diffarantiation between injurad and daad organisms
would be rasaarch items at this tine and not appropriate for inclusion in tha
standard.
1.2.8 Mot to Consider Sabotage
Esoteric problans which could ba prasantad by a variety of hypothetical
terrorist (or wartime) situations, would provide an unnecessary conplication,
and are not appropriate for inclusion in tha standard.
1.2.9 Continuity
The guide standard and protocol will be a living document, subject to
revision and updating with the onset of new technology and knowledge. Zt is
recooaended that the responsible authorities for registration and drinking
water quality review potential needs every two to three years and reconvene
the task force upon need or upon request from the water quality industry, to
review and update the standard and testing protocol.
1.3 Treatment Pnlts Coverage
' l*3*1 Universe of Possible Treatment Pnits
A review of treatment units that Bight be considered aa aicrobiological
purifiers discloses a number of different types covering treatment principles
ranging froa filtration and chemical disinfection to ultraviolet light ra-
diation.
1.3.2 Coverage of This Standard
Zn view of the limited technical data available and in order to expedite
the work of the task force, the initial coverage is limited, on a priority
basis, to three basic types of aicrobiological water purifiers or active
components with their principal means of action as followsi
1.3.2.1 Ceraaic Filtration Candles or Onits (may. or
may not contain a chemical bacteriostatic agent)
Filtration, and adsorption, and chemical anti-microbial activity if a
chemical is included.
1.3.2.2 Haloqcnated Resins and Units
N-4
-------
Chemical disinfection and possibly filtration. (Notet Mill* not
included in this guide standard, halogen products for disinfection or systems
using halogen addition and fine filtration say be tested using many of its.
elements, i.e., test water parameters, microbiological challenge and reduction
requirements, analytical techniques and other pertinent elements.)
1.3.2.3 Ultraviolet (UV) Pnits
UV irradiation with possible add-on treatment for adsorption and filtra-
tion (not applicable to UV units for treating potable water from public water
supply systems).
1.3.3 Application of Principles to Other Units
While only three types of units are covered in this standard, the princi-
ples and approaches outlined should provide an initial guide for the testing
of any of a number of other types of units and/or systems for the microbiplog-
ical purification of contaminated water.
N-5
-------
2. PgRTORMAWCg WEQUIREMEHTS
t
x 2.1 Microbiological Water Purifier
In order to make the claim of "microbiological water purifier," units
. mat be tested and demonstrated to meet the microbiological reduction require-
ments of Table 1 according to the test procedures described in Section 3
(Appendix N-l) for the specific type of unit involved.
2.2 Chemical Health Limits
Where silver or some other pesticidal chemical is used in a unit, that
chemical concentration in the effluent water must meet any National Primary
Drinking Water Maximum Contaminant Level (MCL), additional Federal guidelines
or otherwise be demonstrated not to constitute a threat to health from con-
sumption or contact where no MCL exists.
2.3 Stability of Pesticidal Chemical
Where a pesticidal chemical is used in the treatment unit, the stability
of the chemical for disinfectant effectiveness should be sufficient for the
potential shelf life and the projected use life of the unit based on manufac-
turer* s data. Where stability cannot be assured from historical data and
information, additional tests will be required.
2.4 Performance Limitations
2.4.1 Effective Lifetime
1 The manufacturer must provide an explicit indication or assurance of the
unit's effective use lifetime to warn the consumer of potential diminished
treatment capability either through:
a. Having the unit terminate discharge of treated water, or
b. Sounding an alarm, or
c. Providing simple, explicit instruction for servicing or replacing
units within the reconoended use life (measurable in terns of volume
throughput, specific time frame or other appropriate method).
2.4.2 Limitation on Use of Iodine
N-6
-------
EPA policy initially developed in 1973 and reaffirmed in 1982 (memo of
March 3, 1982 from J. A. Cotxuvo to C. A. Jones, subjectt "Policy on Xodiri*
Disinfection") is that iodine disinfection is acceptable for short-tern or
Halted or emergency use but that it is not recommended for long-term or
routine community water supply application where iodine-containing species may
remain in the drinking water.
N-7
-------
3; MICROBIOLOGICAL WATER PURIFIER TEST PROCEDURES
3.1 Purpose
These testa are performed on ceramic filtration candles or units, halo-
genated resins and units and ultraviolet (UV) units in order to substantiate
their microbiological removal capabilities over the effective use life of the
purifier as defined in Table 1 and, where a pesticidal chemical is used, to
determine that said chemical is not present In the effluent -at excessive
levels (see Section 3.5.3.4, Appendix N).
3.2 Apparatus
Three production units of a type are to be tested, simultaneously, if
feasible; otherwise, in a manner as similar to that as possible.
Design of the testing rig jnust parallel and simulate projected field use-
conditions. For plumbed-in units a guide for design of the test rig may be
taken from "Figure 1: Test Apparatus-Schematic" (p. A-2 of Standard Number S3
"Drinking Hater Treatment Units — Health Effects," National Sanitation
Foundation). Otherwise, the test rig must be designed to simulate field use
conditions (worst case) for the unit to be tested.
3.3 Test Waters — Non-Mlcroblological Parameters . .
In addition to the microbiological influent challenges,'the various test
waters will be constituted with chemical and physical characteristics as
follows:
3.3.1 Test water »1 (General Test Water)
This water is intended for the normal non-stressed (non-challenge) phase
of testing for all units and shall have specific characteristics which may
easily be obtained by the adjustment of many public system tap waters, as
follows:
a. It shall be free of any chlorine or other disinfectant residual;
b. pH -- 6,.5 - 8.5;
c. Total Organic Carbon (TOC) 0.1 - 5.0 mg/L;
d. Turbidity 0.1-5 NTU;
N-0
-------
•. Temperature 20 C " 5 C» and
f. Total Dissolved Solids (TDS) 50 - 500 mg/L.
3.3.2 Test Water *2 (Challenge Test Water/Halogen Plalnfaction)
This watar la lntandad for tha straaaad challenge phaaa of testing whare
units involva halogen disinfectants (halogen resins or other units) and shall
have the following specific characteristicss
a. Free of chlorine or other disinfectant residuali
b. (1) pH 9.0 " .2, and
(2) for iodine-based units a pH of 5.0 * .2 (currant information
indicates that the low pH will be the Dost severe test for virus
reduction by iodine disinfection)j
c. Total Organic Carbon (TOO) not less than 10 mg/L;
d. Turbidity not less than 30 NTU;
e. Temperature A C " 1 C> and
f. Total Dissolved Solids (TDS) 1,500 mg/L * 150 mg/L.
3.3.3 Test Hater #3 (Challenge Test Hater/Ceramic Candle
or Units With or Without Silver Impregnation)
This water is intended for the stressed challenge phase of testing for
the indicated units but not for such units when impregnated with a halogen
disinfectant (for the latter, use Test Water #2). Zt shall have the following
specific characteristics]
a. Zt shall be free of any chlorine or other disinfectant residual;
b. pH 9.0 * .2»
c. Total Organic Carbon (T0C) — not less than 10 mg/L>
d. Turbidity — not less than 30 NTUj
e. Ten$erature 4 C 1 Ct and
f. Total Dissolved Solids (TDS) — 1,500 mg/L * 150 mg/L.
3.3.4 Test Water #4 (Challenge Test Water for Pltraviol'et Units)
This water is intended for the stressed phase of testing for UV units and
shall have the following specific characteristics:
N-9
-------
a. Free of chlorine or other disinfectant residual;
b. pH 6.5 - 8.5;
c. Total Organic Carbon (TOCJ — not less than 10 mg/L;
d. Turbidity — not less than 30 NTU;
e. Temperature 4 C * 1 C;
f. Total Dissolved Solids (TDS) -- 1,500 ag/1 * 150 mg/Lr
g. Color 0.V. absorption (absorption at 254 no) — Sufficient para-
hydroxybenzoic acid (PHBH) to be just below the trigger point of the
warning alarm on the U.v. unit. (Note that Section 3.5.1.1 provides
an alternative of adjusting the U.V. lamp electronically, especially
when the U.V. lamp is preceded by activated carbon treatment.)
3.3.5 Test Water *5 (Leaching Test Water for Units Containing Silver)
This water is Intended for stressed leaching tests of units containing
silver to assure that excess levels of silver will not be leached into the
drinking water. It shall have the following specific characteristics:
a. Tree of chlorine or other disinfectant residual;
b. pH — 5.0 * 0.2)
c. Total Organic Carbon (TOC) — approximately 1.0 mg/L;
d. Turbidity -- 0.1 - 5 NTUj
e. Temperature — 20 C * 5 C» and °
f. Total Dissolved Solids-(TDS) — 25 - 100 mg/L.
3.3.6 Recomnended Materials for Adjusting Test Water Characteristics
a. pHs Inorganic acids or bases (i.e., HC1, NaOH);
b. Total Organic Carbon (TOC): humlc acids; .
c. Turbidity: A.C. Fine Test Dust (Park No. 1543094}
from: A.C. Spark Plug Division
General Motors Corporation
1300 North Dort Highway
Flint, Michigan 46556;
d. Total Dissolved Solids (TDS): sea salts, Sigma Chemical Co., S9863
(St. Louis,MO) or another equivalent source of TDSj
N-10
-------
• . Color U.V.. Absorption! p-hydroxybenxoic acid (grade: general
purpose reagent).
3.4 Analytical Methods
3.4.1 Microbiological Methods
Methods in this section are considered "state-of-the-art" at the tine of
its preparation and subsequent improvements should be expected. Methods used
for nicrobiologlcal analyses should be ccepatible with and equal to or better
than those given below.
3.4.1.1 Bacterial Tests
a. Chosen Organism: Klebsiella terrigena (ATCC-33257).
b. Method of Production: Test organism will be prepared by overnight
growth in nutrient broth or equivalent to obtain the organism in the
stationary growth phase (Reference: Asburg, £.0./ Methods of
Testing Sanitizers and Bacteriostatic Substances Xm Disinfection,
Sterilization and Preservation, Seymour S. Block, ed., pp. 964-980,
1983). The organism will be collected by centrifugation and washed
three times in phosphate buffered saline before use. Alternatively,
the organisms may be grown overnight on nutrient agar slants or
equivalent and washed from the slants with phosphate buffered
saline. The suspensions should be filtered through sterile Whatman
Number 2 filter paper (or equivalent) to remove any bacterial
clumps. New batches of organisms oust be prepared daily for use in
challenge testing.
c. State of Organism: Organisms in the stationary growth phase and
suspended in phosphate buffered saline will be used.
d. Assay Techniques: Assay may be by the spread plate, pour plate or
membrane filter technique on nutrient agar, M.F.C. or m-Endo medium
(Standard Methods for the Examination of Water and Wastewater, 16th
edition, 1983, APHA). Each sample dilution will be assayed in
triplicate.
3.4.1.2 Virus Tests
a. Chosen Organisms: Poliovirus type 1 (LSc) (ATCC-VR-59), and Rota-
virus Strain SA-11 (ATCC-VR-899) or WA (ATCC-VR-2018).
b. Method of Production: All stocks should be grown by a method
described by Smith and Gerba (in Methods in Environmental Virology,
pp. 15-47, 1982) and purified by the procedure of Sharp, et al.
(Appl. Microbiol., 29:94-101, 1975), or similar procedure (Berman
and Hoff, Appl. Environ. Microbiol., 48:317-323, 1984), as these
methods will produce largely monodispersed virion particles.
N-ll
-------
e. Stat* of tha Organism> Preparation procedure vill largely product
monodlspersed particles.
d. Assay Techniques Foliovirua typa 1 may ba grown in tha BGM, MA-104
or othar call lina which will support tha growth of this virus. Tha
rotaviruses are bast grown in tha KA-104 call lina. finca both
viruses can be assayed on tha MA-104 cell line, a challenge test may
consist of equal amounts of both viruses as a mixture (i.e., tha
mixture Bust contain at least 1.0 x 10 /mL of each virus) % Assays
nay be aa plaque forming units (PFO) or as ianunofluorascence foci
( IF) (Smith and Gerba, Im Methods in Environmental Virology,
pp. 15-4?# 1982). Each dilution will be assayed in triplicate.
3.4.1.3 cyst Tests
a. Chosen organism
1. Glardla lamblla or tha related organism, Glardla murls, may be
used as tha challenge cyst.
2. Where filtration is involved, tests with 4-6 micron spheres or
particles have been found to be satisfactory and may be used as
a substitute for tests of occlusion using live organisms (sea
Table 1). Spheres or particles may only be used to evaluate
filtration efficacy. Disinfection efficacy can only ba evalu-
ated with the use of viable Glardla cysts.
b. Method of Productioni Glardla murla may be produced in laboratory
mica and Glardla lamblla may ba produced In Mongolian gerbilsi
lnactivation results based on excystation measurements correlate
well with animal infactivity results.
c. State of the Organisms Organisms may be separated from fecal
material by tha procedure described by Saudi (Appl. Environ.
Microbiol., 48t454-4S5, 1984) or by the procedure described by
Bingham, at al. (Exp. Farasitol., 47>284-281, 1979).
d. Assay Techniquesi Cysts are first reconcentrated (500 ml., minimum
saaple size) according to the method of Rice, Hoff and Schaefer
(Appl. Environ. Microbiol., 43»250-251, 1982). The excystation
method described by Schaefer, et al. (Trans., Royal Soc. of Trop.
Med. c Hyg. 781795-800, 1984) shall be used to evaluate Glardia
murls cyst viability. For Giardla laablla cysts, the excystation
method described by Bingham and Mayer (Nature, 277i301-302, 1979) or
Rice and Schaefer (J. Clin. Microbiol., 14.709-710, 1981) shall be
used. Cyst viability may also be determined by an assay method
involving the counting of trophozoites as well as intact cysts
(Bingham, at al., Exp. Farasitol., 47i284-291, 1979).
3.4.2 Chemical and Physical Methods
N-12
-------
All physical and chemical analysts shall be conducted in accordanca with
procaduras in Standard Mathoda for tha Examination of Water and Wastewatef.
16th Edition. American Public Haalth Association, or aquivalant.
3.5 Tast Procaduras
3.5.1 Procadura - Plumbed-ln Units
a. 1. Install thraa production units of a typa as shown in Figure 1
and condition each unit prior to tha start of tha tast in
accordanca with tha aanufacturar' s instructions with tha tast
watar without tha addition of the tast contaminant. Measure
the flow rate through each unit. The unit shall be tested at
the maximum system pressure of 60 psig static and flow rate
will not be artificially controlled.
2. Test waters shall have the defined characteristics continuously
•xcept for test waters 2, 3 and 4 with respect to turbidity.
The background non-sampling turbidity level will be maintained
at 0.1-5 NTU but the turbidity shall be increased . to the
challenge level of not less than 30 NTU in the following
Banner:
- In the "on" period (s) prior to the sampling "on" period.
- In the sampling "on" period when the sample actually will
be taken. (Notei at least 10 unit void volumes of the 30
NTU water shall pass through the unit prior to actual
sampling so as to provide adequate seasoning and uni-
formity before saaple collection.) .
1. Use appropriate techniques of dilution and insure continual
mixing to prepare a challenge solution containing the bacterial
contaminant. Then spike test water continuously with the
influent concentration specified in Table 1.
2. Use appropriate techniques to prepare concentrated virus and
Glardia suspensions. Feed these suspeosicins into the influent
stream so as to achieve the influent concentrations specified
in Table 1 in the following manneri
- in the "on" period(s) prior to the sampling "on" period.
- in the sampling "on" period when the sasgple actually will
be taken. (Note: at least 10 unit void volumes of seeded
water shall pass through the unit prior to sampling so as
to provide adequate seasoning and uniformity before sample
collection.)
N-13
-------
c. Pur?* the system of the uncontaminated water with a sufficient flow
of contaminated test water. Start an operating cycle of 10 percent
on, 90 percent off with a 15 to 40 ainute cycle (Exaqplet 3 minutes
on, 27 minutes off.) with the contaminated test water. This cycle
shall be continued for not more than 16 hours per day (minimum daily
rest period of 8 hours). The total program shall extend to 100% of
estimated volume capacity for halogenated resins or units and for
10-1/2 days for ceramic candles or units and U.V. units.
d. Samplings Samples of influent and effluent water at the specified
sampling points shall be collected as shown below. for the various
units; these are minimum sampling plans which may be increased in
number by the Investigator. All samples shall be collected in
duplicate frota the flowing water during the sampling "on" portion of
the cycle and they shall be one "unit void volume" in quantity (or
of appropriate quantity for analysis) and' represent worst case
challenge conditions. Effluent samples shall usually be collected
near the middle of the sampling "on" period (or the whole volume
during one "on" period) except for samples following the specified
"stagnation" periods, for which sampling shall be. conducted on the
first water volume out of the unit. Each sasple will be taken in
duplicate and shall be retained and appropriately preserved, if
required, for chemical or microbiological analysis in the event
verification is required. (for units where the volume of a single
"on" period is insufficient for the required analysis, samples from
successive "on" periods may be accumulated until a sufficient volume
has been collected.)
1(a). Sampling Plan: Halogenated Resins or Onits (Non-iodine Based)
Test Point
(% of Estimated
Capacity)
Start
25%
50%
After 48 hours
stagnation
General
Influent
Background
Tests
Active
Agent/
Residual ,
X
X
X
Microbiological
MM*
X
X
X
60% Chal- * x
75% lenge X x
After 48 hours pH -
stagnation 9.0 * 0.2 X X
100% x x
1(b). Sampling Plan: lodinated Resins or Units
N-14
-------
Tests
Test Point
(% of Estimated
Capacity)
Start
25%
50%
After 48 hours
stagnation
General
Influent
Background
Active
Agent/
Residual
X
X
X
Microbiological
X
X
X
60%
75%
After 40 hours
stagnation
Chal-
lenge
pH -
9.0 * 0.2
X
X
X
X
90%
100%
After 40 hours
stagnation
Chal-
lenge
pH -
5.0 * 0.2
X
X
X
X
2. Sampling Plan: Ceramic Candles or Units and D.V. Onits
Tests
Test Influent
Test Point Water Background Microbiological
Start General X X
Day 3 (middle) X
Day 6 (middle) X
After 40 hours
stagnation X
Day 7 (middle) X
Day 8 (near end) Chal- - X
After 48 hours lenge
stagnation X
Day 10-1/2 X
(Notei All days are "running days" and exclude stagnation periods. When
the units contain silver, a leaching test shall be conducted as shown in -
Section 3.S.l.e and silver residual will be measured at each microbiological
sailing point.)
N-15
-------
teaching Tests for Silvarized Units: Where the unit contains
silver, additional tests utilizing Tast Watar #5 will be eonductad
as followsi
Tasts
Start
Day 2
After 48 hours
Tast Point
Xnfluant
Background
X
Silver/Residual
X
X
stagnation
x
f. Altarnata Sampling Plans:
1. Sinca soma laboratories nay find it inconvaniant to tast some
units on a 16 hour on/8 hour off cycle, two alternates are
recognized:
- Go to a shorter operational day but lengthen the days of
test proportionally
- Use up to 20 percent "on"/80 percent "off" for a propor-
tionally shorter operational day
2. Sampling points must be appropriately adjusted in any alternate
sampling plan.
g. Application of Test Waters: The application of test waters is
designed to provide information on performance under both normal and
stressed conditions; it should be the sane or equivalent to the
following:
1. a.. Halogenated Resins or Units (Non-iodine based) —
First 50% of tast period: Test Water 1 (General)
Last 50% of tast period: Test Water 2 (Challenge)
(pH - 9.0 * 0.2)
b. lodinated Resins or Units
First 50% of test period; Test Water 1 (General)
Next 25% of test period: Test Water 2 (Challenge)
(t>H - 9.0 * 0.2)
Last 25% of test period: Test Water 2 (Challenge)
(but with pH - 5.0 * 0.2)
2. Ceramic Candles or Units -
First 6 days of testing:
'Test Water 1 (General)
N-16
-------
Last 4-1/2 days of tattingi
Taat Water 3 (Challenge)
3. Ultraviolet (U.V.) Units
First 6 days of tastings
Last 4-1/2 days of tastingt
Tast Water 1 (General)
Test Water 4 (Challenge)
Analyses and monitoring:
1. Microbiological sampling and analysis shall be conducted of the
specified Influent and effluent saspling points during each
Indicated sampling period.
2. Tast Water Monitoring: The specified paraneters of the various
tast waters (see Section 3.3) will be measured and recorded at
•ach microbiological sampling point; the specified parameters
will be measured at least once on non-sampling days when the
units are being operated.
3. Background chemical analyses of influent water shall be con-
ducted at least or.ee at the start of each test period to
determine the concentration of the U.S. EPA primary inorganic
contaminants, secondary contaminants and routine water para-
meters, not otherwise covered in the described test waters.
4. In addition, quality assurance testing shall be conducted for
the seed bacteria under environmental conditions on the first
and last days of testing to make sure that there is no signifi-
cant change over the test day. Populations will be measured
(for example, as dispersed in the supply tank) at the beginning
and end of the test day to detect possible incidental effects
such as proliferation, die-off, adsorption to surfaces, etc.
Relatively stable bacterial seed populations are essential to
an acceptable test program.
5. Whan a .unit contains a halogen or silver, the active agent
residual will be measured in the effluent at each microbiologi-
cal test (sampling) point.
6. Silver will additionally be measured three times in the efflu-
ent as specified in Section 3.S.l.e.
Neutralization of Disinfection Activity: Immediately after col-
lection, each test san^le must be treated to neutralize residual
disinfectant. For halogen- and silver-based disinfectants this may
be done by addition of thioglycollate-thiosulfate neutralizer
solution (Chambers, et al., J. Amar. Water Works Assoc., 54:208-216,
1962). This solution should be prepared daily. All results are
invalid unless samples are neutralized immediately upon collection.
Special Provisions for Ceramic Candles or Units:
N-17
-------
1, Provisions for slow flowi Ceramic unit* may ba subjset to
clogging slid greatly reduced flow over the test period. An
attempt should be ud« to aaintain manufacturer rated or
claimed flow rates, but even st raducad flows tha sampling
program sat forth in Saetion 3.5.1.d.2 shall ba maintained.
2. Cleaning of ceramic unitsi Units should ba cleaned according
to manufacturer's directions. two cleanings should occur
during the period of test (in order to prove the unit's
durability through the cleaning procedure). However, near the
time of Bicrobiological sampling, the units should not be
cleaned until after the saapling. further, no ainti-microbial
chemical (for cleaning or sanitising) say ba applied to the
units during the test period unless the manufacturer specifies
the sane as part of routine maintenance.
k. Halogenated units or U.V. units with mechanical filtration processes
separate from the microbiological disinfection components shall have
the mechanical filtration conswnants replaced or serviced when
significant flow reduction (clogging) occurs in accordance with the
manufacturer's instructions in order to maintain the test flow rate.
Units with non-removable mechanical filtration components will be
run until flow is below that considered acceptable for consumer
convenience. (If premature clogging presents a problem, some
specialized units may require a customized test plan.)
1. Special Provisions for Ultraviolet (U.v.) Unitsi
1. The units will be adequately challenged by the prescribed test,
waters; consequently they will be operated at normal intensity.
However, where the U.V. treatment component is preceded by
activated carbon treatment, the output of the U.V. lamp shall
be adjusted electronically, such as by reducing tha current to
the lamp or other appropriate swans, to be just above the alarm
point. This option shall be available for use under other U.V.
configurations, at the choice of the persons responsible for
testing, as an alternative to the use of the U.V. absorbent,
p-hydroxybenzoic acid.
2. Pail/safe: Units will provide and will be tested for fail/safe
warnings in the event of water quality changes or equipment
failures which may interfere with its microbiological purifica-
tion function.
3. Cleaningi Manufacturer's guidance with respect to cleaning
will be followed.
3.5.2 .Procedure» Non-Plumbed Units
a. General; The basic procedures given in Section 3.5.1 shall be used
with necessary adaptations to allow for the specific design of the
N-18
-------
unit.' Xit any event, the testing procedures shall provide a test
challenge equivalent to those for* pluabed-ia units.
«•
b. Test conditions and apparatus should be adapted to reflect proposed
or actual use conditions in consultation vith the Banufacturer,
including flow rate and nuaber of people to be served per day. Zn
sons cases variable flow or other non-standard conditions Bay be
necessary to reflect a worst-case test.
1.5.3 Acceptance and Records
3.3.3.1
To qualify as a microbiological water purifier, all thraa production
units of a type Bust continuously aeet or exceed the reduction requirements of
fable 1, within allowable measurement tolerances for not aore than ten percent
of influent/effluent sample pairs, defined as followsi
The geometric mean of all Bicrobiological reductions Bust aaet or axceed
the requirements of Table 1. An example is given as follows;
- Unitt iodinated resin.
- Number of sasple pairs over the completed test program:
10 per unit — 3 units ¦ 30.
- number of allowable sample pairs where log reduction is insuffi-
cient! 10% of 30 - 3 sample pairs.
- Allowable minimum log reductions in these 3 pairs<
- Conclusion! If the geometric mean of all reductions Beets or
•xceeds the requirements of Table 1, the indicated insufficient
sample pairs will be allowed.
3.5.3.2 Records
Virusi
Bacteriat
Cystsi
one order of aagnitude
one order of magnitude
one/half order of aagnitude
Bacteria
Virus
Cyst
5 log
3 log
2-1/2 log
N-19
-------
All pertinent procedures. and data shall be recorded in a standard format
and ratalnad for possible review until the report of results has been com-
pletely accepted by review authorities, in no case for less than a year.
3.5.3.3 Scaling Op or Down
Where a manufacturer has several similar units using the same basic
technology and parallel construction and operation, it may sometimes be
appropriate to allow the test of one unit to be considered representative of
others. Where any serious doubt exists, all units of various sizes may
require testing. A "rule of three" is suggested as a matter of judgment.
Scaling up to three tines larger or on-third, based on the size of either the
test unit or of its operative element, may be allowed. However, for UV units,
any size scale-up must be accompanied by a parallel increase in radiation
dose.
3.5.3.4 «
Where silver or some other chemical is used in the unit, concentrations
in the effluent water must meet any National Primary Drinking Water Maximum
Contaminant Level (MCL), additional Federal guidelines, or otherwise must not
constitute a threat to health where no MCL exists.
N-20
-------
APPENDIX H-i
SUMMARY FOR BASIS OP STANDARDS AND TEST WATER PARAMETERS
Microbiological Reduction Regulreaenta
1. Bacteria
Currant standards for the Microbiological safety of drinking
water ara baaad on the preaence of colifora bactaria of which
Klebsiella la a aaabar. ^ Maabers of tha ganua Klebsiella ara also
potantlalpathogana of nan (Vlassof, 1977). Klabalaila tarrigena la
daalgnatad as tha teat organism alnca It is cenaonly found in
aurfaca watera (Izard, at al., 1981).
Experience with tha uae of colifora bactaria to eatlaate tha
praaanca of antarlc bactarial pathogana in drinking watar aa per-
formed ovar tha laat 73 years indieatas a high dagraa of reliabil-
ity. Required tasting of aora than ona bactarial pathogan appaars
unjustified alnca viral and Glardla tasting will ba raqulrad.
Entaric viruses and Glardla ara known to ba aora rasistant to canton
disinfectants than antarlc bactarial pathogana and vlrusas ara aora
rasistant to removal by traataenta auch aa filtration. Thus, any
traataant which would give a good raaoval of both virus and Glardla
pathogens would aost likely reduce enteric bactaria below levels
considered infectious (Jarroll, at al., 1981; Liu, at al., 1971) .
lite eventration of colifon bactaria in raw sewage ia approx-
imately 10/100 al. Concentrations in polluted stream waters have
been found to exceed 10 per 100 al (Culp, et al., 1978, Table 10).
Baaad on tha over 105/100 al concentrations observed in highly
polluted stream watar and a target affluent concentration of less
than 1/100 al, a 6 log reduction Is recoaaended.
2. . Vlrua
Xn tha United States concantrationa of enteroviruses ara esti-
aated to range from 10 -10 /liter in raw aawaga (Farrah and Schaub,
1971), Baaed on this observation it la aatiaatad thatj naturaj
watara contaalnatad with raw aewage nay contain froa 10 to "10
enteric viruses per liter.
There are currently no standards for viruses in drinking water
in the United States. However, EPA has proposed a non-enforceable
health-baaed recommended maximum contaminant level (RMCL) of zero
for viruses (EPA, 1985). Several individuals and organizations have
developed guidelines for the presence of viruses in drinking water
and various experts have proposed standards (WHO# 1979, 1984? Berg,
19711 Melnick, 1976). It has generally been felt that drinking
N-21
-------
water should be free of infectious virus sines even ons virus is
potentially infectious and suggested standards are largely based on
technological limits of our detection Mthodology. Guidelines
suggested by the World Health Organization (1984) and others
recoanend that volumes to be tested be in the order of-100-1,000
liters and that viruses be absent in these volunes.
Assuming a target effluent level of less than, one virus in 100
liters of water and a concentration of 10 enteric viruses in 100
liters of sewage-contaminated waters, the water purifier units
should achieve at least 4 logs of virus removal.
The relative resistance of enteric viruses to different dis-
infectants varies greatly among the enteric viruses and even among
members of the same group (i.e., enteroviruses). For exaople, while
f2 coliphage is one of the most resistant viruses to inactivation by
chlorine it is one of the most susceptible to inactivation by ozone
(Harakeh and Butler, 1984). Zonic conditions and pH can also affect
the relative resistance of different viruses to a disinfectant
(Engelbrecht, et al., 1980). On this basis it is felt that more
than one enteric virus should be tested to ensure the efficacy of
any disinfection system. Pollovlrus type 1 (Strain LSc) was chosen
as one of the test viruses because it has been extensively used .in
disinfection and environmental studies as representative of the
enterovirus family. It is recognized that it is not the most
resistant virus to inactivation by chlorine, but is still resistant
enough to serve as a useful Indicator. Rotavirus is selected as the
second test enteric virus since it represents another group of
enteric viruses in nucleic acid coaposltion and size. Zt is also a
major cause of viral gastroenteritis and has been documented as a
cause of water borne gastroenteritis (Gerba, et al., 1985). The
human rotavirus or the similar Simian rotavirus may be used in the
test procedure. A net 4-log reduction for a joint challenge of
1 x 10 /L each for pollovirus and rotavirus is recoanended.
Cysts (Protozoan)
Over the past several years, giardiasis has consistently been
one of the most frequently reported waterbome diseases transmitted
by drinking water in the United States (Craun, 1984). EPA has
proposed a RMCL of zero for Glardia (EPA, 1985). Zts occurrence has
generally been associated with treatment deficiencies including
either inadequate or no filtration. Giardiasis has not been known
to occur from drinking water produced by well-operated filtration
treatment plants. De Walle, et al. (1984), in a study of filtration
treatment plant efficiencies, cited percent removals for Glardia in
pilot plant tests as follows:
- Rapid filtration with coagulation-sedimentation: 96.6-99.9%;
- Direct filtration with coagulation: 95.9-99.9%.
N-22
-------
rroa . this research and from th« lack of Clardia caaea in
systems where adequate filtration exists, a 3-log (99.9%) reduction
requirement la considered to be conservative and to provide a
comparable laval of protection for water purifiera to ' a
wall-operated filtration treatment plant.
Data on environmental lavala for cyata in natural watera ia
limited bacauaa of tha difficulties of sampling and analysis.
Unpublished data indicata vary low lavala from laaa than 1/L to laaa
than 10/L. Mara a 3-log raduction would provida an affluant of laaa
than 1/100 t, comparable to tha recoanended virus raduction requlra-
menta. *
Either Glardla lamblla or tha related organism, Giardla murls.
which ia raportad to ba a satisfactory taat organ!am (Hoff, at al.,
1985), may ba uaad aa tha challenge organism. Teats will ba con-
ducted with a challenge of 10 organisms par litar for a 3-log
reduction*
Where the treatment unit or component for eyata ia baaed on the
principle of occluaion filtration alone, teating for a 3-log raduc-
tion of 4-6 micron particlea or apherea (National Sanitation Founda-
tion Standard 53, aa an example) ia acceptable. Difficultiea in tha
cyat production and measurement technologies by lesser-equipped
laboratories may require the use of such alternative testa where
applicable.
Microbiological Purifier Test Procedures
1. Test Waters
a. The general teat water (test water II) ia designed for the
normal, non-stressed phase of testing with characteristics that
may easily be obtained by the adjustment of many public aystam
tap waters.
b. Test water #2 is Intended for tha etressed phase of testing
where unita Involve halogen disinfectants.
1. Since the disinfection activity of aome halogens falls
with a rising pH, it ia important to atraaa test at an
elevated pH. The recommended level of 9.0 * 0.2, while
exceeding the recoasaended aecondary level (Environmental
Protection Agency, 1984J ia atill within a range seen in
some natural watera (Environmental Protection Agency,
1976). However, for iodine-baaad units, a second stress-
ful condition is provided — a pH of, 5.0 "0.2 aince
current Information indicates that tha diainfection
activity of iodine falls with a low pH (National Research
Council, 1910). While beneath the recommended aecondary
level (Environmental Protection Agency, 1984) a pH of 5.0
K-23
-------
la not unusual in natural waters (Environmental Protection
Agency, 1976).
2. Organic matter as total organic carbon (TOO) is known to
interfere with halogen disinfection. While this TOC is
higher than levels in many natural waters, the designated
concentration of 10 mg/L is cited as typical in stream
waters (Culp/Wesner/Culp, 1978).
3. High concentrations of turbidity can shield aicroorganisas
and interfere with disinfection. While the recommended
level of not less than 30 MTU is in the range' of turbidi-
ties seen in secondary wastewater effluents, this level is
also found in many surface waters, especially during
periods of heavy rainfall and snow nelt (Culp/Wesner/Culp,
1978).
4. Studies with Glardla cysts have shown decreasing halogen
disinfection activity with lower tesiperatures (Jarroll,
et al., 1980); 4 C, a coaroon low tenperature in many
natural waters, is reconmended for the stress test.
5. The amount of dissolved solids (TDS) nay impact the
disinfection effectiveness of units that rely on displace-
able or exchange elements by displacement of halogens or
resins, or it may interfere with adsorptive processes.
While TDS levels of 10,000 mg/L are considered unusable
for drinking, many supplies with over 2,000 mg/L are.used,
for potable purposes (Environmental Protection Agency,
1984). The reconmended level of 1,500 mg/L represents a
realistic stress challenge.
Test water #3 is intended for the stressed phase of testing of
ceramic filtration candles or units with or without silver
impregnation.
«
1. Since viruses are typically eluted from adsorbing media at
high pHs (Environmental Protection Agency, 1978) it may be
concluded that a high pH will provide the most stressful
testing for a ceramic-type unit; consequently, the high
natural water pH of 9.0 is reconmended.
2. Expert opinion also holds that organic material will
interfere with adsorption of viruses. Thus, a high total
organic carbon level of not less than 10 mg/L is recom-
mended.
3. Turbidity may enhance the entrapment and removal of
microorganisms but it also may stimulate "short-
circuiting" through some units. A turbidity level of
30 NTU will provide stress at time of sampling but the
N-24
-------
non-saaqpling level of 0.1-5 NTU will allow routlna opera-
tion of units.
*•
4. Expert opinion holds that low watar temperatures and high
TDS would most likely intarfara with virus raduetion by
adsorption; consequently, a 4 C temperature and 1,500 mg/L
TDS ara recoonended.
d. Tast watar 14 is intandad for the stressed phasa of tasting for
ultraviolat (UV} units.
1. In general, high TOC, turbidity and TDS and' low tempera-
tura ara considarad most strassful for UV, and tha in-
dicated challanga lavals ara tha same as for tast
watar #2.
2. Tha pH is not critical and nay rang* from 6.5 to 8.5.
3. In ordar to tast tha UV units at thair nost vulnerabla
staga of operation, a color challanga (light absorption at
254 nm) is to ba maintained at a level where uv light
intensity is just above tha unit's low intensity warning
alarm point. However, an alternate to the absorption
challenge is provided through adjusting the light intensi-
ty output of the UV lamp electronically by reducing
currant to the lamp, or other appropriate means, to be
just above the alarm point; this approach would be
particularly necessary where the UV lamp is preceded by
activated carbon treatment.
e. Test water #5 is intended for the stressed leaching tests of
units containing silver. Low pK, TOC, turbidity, and TDS and
higher temperature are felt to be the characteristics associ-
ated with increased leachability. The recoonended pH of
5.0 * .2, while being beneath the recosraended secondary range
of 6.5*8.5 (Environmental Protection Agency, 1984) is still
found in some natural waters.
Test Procedures
The plan for tasting and sampling is designed to reveal unit
performance under both "normal" and "stressed" operating conditions.
Tha stressed phase would utilize a set of water quality and opera-
tions conditions to give the units a realistic worst case challenge.
Testing plans for a specific model might involve modifications to
the recomnended plan; more samples•could be taken and analyzed; more
units could be studied. The principle of demonstrating adequate
performance even under realistic worst case conditions should be
maintained and the final selected test procedures should be agreed
as between investigators and reviewers or regulators.
N-25
-------
Mhlla SOOM aiptcta of tha tasting procaduras hava bMn utlllzad
in actual axpariaants, tha proposed protocol lias not bmmn varifiad
or titilirai for tha various wits that My ha considarad. Conse-
quently, investigators and users of this protocol aay find raasons
to altar soma aspacts through thair practical axperlance* naadad
changas should ha discussed and claarad with involved revievers/-
regulators.
N-26
-------
REFERENCES:
Bar?, C. Integrated approach to the problem of viruses in water. J. ASCE,
Sanit. En?. Olv, 97t867-882, 1971.
Culp/Wesner/Culp. Guldanca for planning tha location of water supply intakes
downstream from municipal wastewater treatment facilities. EPA Report, Office
of Drinking Hater. Washington, DC, 1978.
Craun, G. F. 1984. Waterbome outbreaks of giardlasist Current status, in:
Clardla and giardiasis. D. L. Erlandsen and E. A. Meyer Eds., .Plenum Press,
New York, pp. 243-261, 1984.
DeWalle, P. B.i J. Engesetj Lawrence, W. Removal of Glardia lamblla cyst by
drinking water treatment plants. .Report No. EPA-600/52-84-069, Office of
Research and Development, Cincinnati, OR, 1984.
Engelbrecht, R. S., et al. Comparative inactivaticn of viruses by chlorine.
Appl. Environ. Microbiol. 40:249-256, 1980.
Environmental Protection Agency. Quality criteria for water. Washington, DC,
1976.
Environmental Protection Agency. National secondary drinking water
regulations. EPA-570/9-76-000, Washington, DC, 1984.
Environmental Protection Agency. National primary drinking water regulations;
synthetic organic chemicals, inorganic chemicals and microorganisms; Proposed
rule. Federal Register, Nov. 13, 1985.
Farrah, S. R., and S. A. Schaub. Viruses in wastewater sludges. Xn> Viral
Pollution of the Environment, G. Berg, Ed. CRC Press, Boca Raton, Florida,
pp. 161-163, 1983.
Gerba, C. P.j Rose, J. B.f Singh, S. N. Waterbome gastroenteritis and viral
hepatitis. CRC Critical Rev. Environ. Contr. 15:213-236, 1985.
Harakeh, M.j Butler, M. Inactivatlon of human rotavirus, SA-11 and other
enteric viruses in effluent by disinfectants. J. Hyg. Camb. 93:157-163, 1984.
Hoff, J. C.i Rice, E. W.i Schaefer, F. W. Comparison of animal lnfectivity
and excystation as measures of Glardia murls cyst inactivatlon by chlorine.
Appl. Environ. Microbiol. 50:1115-1117, 1985.
Izard, D.; Farragut, C.iGavinl, F.i Kersters, K.i DeLey, J.j Leclerc, H.
Klebsiella terrigena, a new species from water and soil. Intl. J. Systematic
Bacteriol. 31:116-127, 1981.
Jakubowski, W. Detection of Glardia cysts in drinking water. In: Glardia
and Giardiasis, Erlandsen, S. L. t Meyer, E. A. Eds., Plenum Press, NY.
pp. 263-206, 1904.
N-27
-------
Jarroll, E. L.t Bingham, A. K.j Meyer* E. A. Glardla cyst destruction:
Effectiveness of six small-quantity water disinfection methods. Am. J. Trop.
Med. 29<8*11/ 1980
Jarroll, E. L.t Binghaa, A. K.t Meyer, E. A. Effect of chlorine Qn Glardia
cyst viability. Appl. Environ. Microbiol. 43»483-487, 1981. ———
Liu, 0. C., et al. Relative resistance of 20 human enteric viruses to free
chlorine in Potomac River water. Proceedings of 13th Water Quality Conference
Snoeyink, V.; Griffin, V. Eds., pp. 171-195, 1971.
Melnick, J. L. Viruses in water. Im Viruses in Water Berg, G.»
Bodily, H. L.j Lennette, E. H.j Melnick, J. L.t Metclaf T. G., Eds. Amer.
Public Hlth. Assoc., Washington, DE, pp. 3-11, 1976.
National Research Council. The disinfection of drinking water, Ini Drinking
Water and Health, Volume 2. Washington, DC, pp. 5-137, 1980.
National Sanitation Foundation. Drinking water treatment units: Health
effects. Standard 53. Ann Arbor, Ml, 1982. .
Vlassoff, L. T. Klebsiella. Zn: Bacterial Indicators/Health Hazards
Associated with Water Hoadley, A. W.; Dutka, B. J., Eds. American Society for
Testing and Materials, Philadelphia, PA. pp. 275-288, 1977.
World Health Organization. Human Viruses in Water, Technical Support
Series 639, World Health Organization, Geneva, 1979.
World Health Organization. Guidelines for Drinking Water Quality. Volume 1.
Recommendations. World Health Organization, Geneva, 1984.
N-28
-------
APPENDIX N-2
LIST OF PARTICIPANTS! TASK FORCE ON GUIDE STANDARD AND PROTOCOL TOR
TESTING MICROBIOLOGICAL WATER PURIFIERS
Stephen A. Schaub, Chairman — U.S. Army Medical Bloengineerlng Research and
Development Laboratory (USAMBRDL), Fort D«trick, Maryland 21701, FTSr
8/935-7207 — Comm: 301/663-7207.
Frank A. Ball, Jr., Secretary — Criteria and Standards Division, Office of
Drinking Water (WH-550), Environmental Protection Agency^ Washington,
D.C. 20460, Phonet 202/362-3037.
Paul Berger, Ph.D. — Criteria and Standards Division, Office of Drinking
Water (WH-550), Environmental Protection Agency, Washington, D.C. 20460,
Phonei 202/382-3039.
Art Castillo — Disinfectants Branch, Office of Pesticide Programs (TS-767C0,
Environmental Protection Agency, Washington, D.C. 20460, Phonex 703/557-
3695.
Ruth Douglas — Disinfectants Branch, Office of Pesticide Programs (TS-767C),
Environmental Protection Agency, Washington, D.C. 20460, Phone: 703/557-
3675.
A1 Dufour — Microbiology Branch, Health Effects Research Laboratory,
Environmental Protection Agency, 26 W. St. Clair Street, Cincinnati, Ohio
45268, Phone: FTSi 8/684-7870 — Coma: 513/569-7870.
Ed Geldreich — Chief, Microbiological Treatment Branch, Water Engineering
Research Laboratory, Environmental Protection Agency, 26 W. St. Clair
Street, Cincinnati, Ohio 45268, Phonei FTS; 8/684-7232 — Comm:
513/569-7232.
Charles Gerba — Associate Professor, Department of Microbiology and
Immunology, University of Arizona, Tucson, Arizona 85721, Phone:
602/621-6906.
John Hoff — Microbiological Treatment Branch, Water Engineering Research
Laboratory, Environmental Protection Agency, 26 W. St. Clair Street,
Cincinnati, Ohio 45268, Phone: FTS: 8/684-7331 — Coann: 513/569-7331.
Art Kaplan — office of Research and Development (RD-681) Environmental
Protection Agency, Washington, D.C. 20460, Phone: 202/382-2583.
Bala Krishnan — Office of Research and Development (RD-681) Environmental
Protection Agency, Washington D.C. 20460, Phone: 202/38^-2583.
John Lee — Disinfectants Branch, Office of Pesticide Programs (TS-767C)
N-29
-------
Environmental Protection Agency, Washington, D.C. 20460, Phone:
703/557-3663.
Dorothy Portner — Disinfectants Branch, Office of Pesticide Programs
(TS-767-C), Environmental Protection Agency, Washington, D.C. 20460,
Phonei 703/557-0484.
Don Reasoner — Microbiological Treatment Branch, Water Engineering Research
Laboratory, Environmental Protection Agency, 26 W. St. Clair Street,
Cincinnati, Ohio 45268, Phonet 312/654-4000.
P. Reguanthan (Regu) — Everpure, Inc., 660 N. Blackhavk Drive, Westmont,
Illinois 60559, Phonet 312/654-4000.
David Stangel — Policy and Analysis Branch, Office of Compliance Monitoring,
Environmental Protection Agency, Washington, D.C.,. Phone: 202/382-7845.
Richard Tobin -- Monitoring and Criteria Division, Environmental Health
Center, Department of Health and Welfare of Canada, Tunney's Pasture,
Ottawa, Ontario, K1A 0L2, Canada, Phone: 613/990-8982.
N-30
-------
APPENDIX N-3
RESPONSE BY REVIEW SUBCOMMITTEE ^ TO PUBLIC COMMENTS ON GUIDE STANDARD
AND PROTOCOL FOR TESTING MICROBIOLOGICAL WATER PURIFIERS
A. Recommendation for the us* of Glardla lamblia cysts as a replacement for
Glardla aurls cysts as the protozoan cyst test organisms.
Recommendationi
The subcommittee concurs with the recommendation and further endorses the
use of Giardia lamblia as the preferred cyst test for evaluation of all
treatment units and devices. Obviously the use of the protozoan orga-
nisms of actual health concern in testing is most desirable. Anyone
finding the Giardia lamblia strain feasible for testing and cost-
effective to work with is encouraged to use same instead of Giardia
rauris.
B. Substitution of 4-6 micron bead or particle tests as an alternate option
instead of the Glardla cysts for evaluating devices that rely strictly on
occlusion filtration for microbiological removal: Several commenters
criticized the use of beads or particles (e.g., A.C. fine dust) and
recommended only use of live Glardla cysts for performance tests.
Discussion?
The subcommittee recognizes and favors the use of the natural human
parasite, Glardla lamblia, but was not aware of any convincing scientific
data which would disallow the optional use of testing with beads or
particles for units or devices using only occlusion filtration to remove
microorganisms. Previous development of the National Sanitation Standard
(NSF) S3 (1982) requirement for cyst reduction (using 4-6 micron parti-
cles as cyst models) was based on engineering and scientific opinion and
experimental evidence at that time. Specifically, Logsdon used
radioactive cyst. models in the initial phase of a study of removal
efficiencies for diatomaceous earth filters; subsequent experiments with
Glardla muris cysts confirmed the efficacy of the diatomaceous earth
filters. Further studies by Hendricks and DeWalle with Glardla
lamblia cysts also showed comparable reduction efficiencies for
diatomaceous earth filters.
l.S.A. Schaub; F.A. Bell, Jr.; P. Berger; C. Gerba; J. Hoff;
P. Regunathan; and R. Tobin. (Includes additional revision pursuant to
Scientific Advisory Panel review (Federal Insecticide, Fungicide, and
Rodenticide Act).]
N-31
-------
Subsequently confirmatory parallel testing results have been developed
vis-a-vis 4-6 micron particles as coopered to Glaxdla lamblia cysts.
Specifically, two..units listed by NSF for cyst reduction (using 4-6
micron particles)1 ' have also been tested and listed for 1001 efficiency
reduction (using Clardla lamblia cysts) by fflblar i
1. Everpure Model QC4-SC
2. Royal Boulton Nodal F303.
Again we prefer the use of the human pathogen, Clardla laabllay however,
no experimental data has bean provided regarding the lack of validity or
of failure in previous tests utilizing beads or particles of 4-6 microns.
Zn most cases the bacterial or viral challenges to occlusion filters ill
represent a greater problem in terms of microbiological reduction
requirements than will cysts. Therefore, without substantiation of
deficiencies, the use of 4-6 micron beads or particles is considered to
be as feasible as the use of live cysts for routine performance testing
of water filtration (occlusion) devices.
Recommendationt
%
Recownend retaining the optional use of 4-6 micron particles or beads for
cyst reduction testing in occlusion filtration devices only.
Referencess
f 1)
Logsdon, G. S., et al. Alternative Filtration Methods for Removal
of Glardla Cysts and Cyst Models, JAWWA, 73(2)111-118, 1981.
Logsdon, G. S.; Hendricks, D. w., et al. Control of Glardia Cysts
by Filtratlom The Laboratory's Rose. Presented at the Awwa Water
Quality Technology Conference, December, 1983.
DeWalle, et al. Removal of Glardia lamblia Cysts by Drinking Hater
Treatment Plants, Grant No. R806127, Report to Drinking Hater
Research Division, U.S. EPA (ORD/MERL), Cincinnati, phio.
(4)
National Sanitation Foundation, Listing of Drinking Hater Treatment
Onits, Standard 53. May, 1986.
*5' Hibler, C. P. An Evaluation of Filters in the Removal of Glardia
lamblia. Hater Technology, pp. 34-36. July, 1984.
C. Alternate assay techniques for cyst tests (Jensen)s Proposed alterations
in cyst tests include a different method for separating cysts from fecal
material and an assay method involving the counting of trophozoites as
well as intact cysts. Both alterations have been used by. Bingham, et al.
(Exp. Parasitol., 47»284-291, 1979).
Recommendation t
N-32
-------
Thai* alterations appear to ba raaaonabla laboratory procedures, support-
ad by a peer-reviewed articla and will ba Included in tha Report as
options for posslbla development and usa by intarastad laboratorias.
D. Tha uae of pour plata .tachniquas as an option for Klebsiella. terriqena
bacteria analyse!.
Raconnendatlon;
The pour plata technique adds a heat stress factor to tha bacteria which
constitutes a possible deficiency. However, it is a recognized standard
nethod and probably will not adversely affect the Klebsiella terriqena.
Consequently, it will ba added to the Report as one of the acceptable
techniques.
E. ^tion of using Escherichia coli in lieu of Klebsiella terriqena for the
bacterial tests.
Discussion;
Appendix N-l, Section A.l. of the Guide Standard and Protocol sets forth
the basis for selection of K. terriqena as the test bacteria. The
selection was nade along pragmatic line emphasizing the occurrence of K.
terriqena in surface waters and that it would represent the enteric
bacteria. It was also pointed out that the tests with virus and Giardla
were expected to be more severe than the bacterial tests. For comprehen-
siveness, bacterial tests were included in the protocol but were not felt
to be as crucial as the virus and Giardla tests.
E. coll, or any number of other generally accepted indicator bacteria,
could be used for the test program if they were shown to have good
testing and survival characteristics (equivalent to K. terriqena) by the
interested research laboratory.
Recoannendation:
•
The intent of the Guide Standard and Protocol is to provide a baseline
program subject to modification when properly supported by an interested
laboratory. Consequently, any laboratory could propose and with proper
support (demonstrating challenge and test equivalency to K. terriqena)
usa Escherichia coll or one of the other enteric bacteria. This idea
will be included in revised working in Section 1.2.2, "General Guide."
F. Performance requirements for Giardla cysts and virus in relation to the
EPA-Recommended Maximum Contamination Levels (RMCLs) of zero.
Discussion:
The RMCLs of zero for Giardla and viruses which have been proposed by EPA
are health goals. They are no enforceable standards since to assure the
presence of "no organisms" would require an infinite sample. The
N-33
-------
rational* for th« recommended performance requirements for Glardla eyata
and virus ia sat forth in Saetiona A.2 and A. 3 of Appandix A. We faal
that these requirements together with tha application of raaliatic worst
caaa tast conditiona will provide a conaarvativa taat for unita rasultir.g
in traatad affluant watar equivalent to that of a public wafer aupply
aaating tha microbiological raquirananta and intant of tha National
Primary Drinking Water Ragulations.
Racoanandatloni
Retain recoanended parformanca (log raduction) requirements for cyat and
virus raduction.
C. Rotavirus and ita proposad assay: Ona cotnmenter atataa that tha rota-
virus taata ara impractical bacauaa Amirtharajah (J. AWWA, 78(3):34-49,
1976) cltas "no satisfactory culture procedures avallabia for analysis of
thaaa pathogans and, therefore, monitoring would not ba faasibla."
Dlseuaaloni
Section 3.4.1.2, "Virus Tasta" of the Report, praaanta means for cul-
turing and aasaying rotaviruses. This swans for doing tha rotavirus
tasta are available and are practical for application in tha laboratory.
Or. Amirtharajah was referring to the field collection, identification in
the presence of a wide variety of microorganisms, and quantification as
not baing "satisfactory." Laboratory analysis of rotaviruses is practi-
cal but their field monitoring may not yet ba feasible.
Further, the aalactlon of both poliovirua and rotavirus as tast viruses
was necessitated by the fact that tha surface adsorptiva properties and
disinfection resistance of the varloua anteric viruses hava been shown to
differ aignificantly by virus group and by atrains of a specific virus.
While all anteric viruses and their atraina could not ba economically
teated, it waa determined by the taak force that at least two distinctly
different virua typea ahould be taated to achieve soma idea of the
diversity of removal by tha various types of water purifiers. Polio and
rota viruses hava distinctly different phyaical and chemical charac-
terlatics representative of tha viruaes of concern. Polioviruses ara
small aingla stranded RNA viruses with generally good adaorptive proper-
ties to aurfacea and filter media while rotaviruaes ara over twice as
large, are double stranded RNA and in soma atudies hava been found to
posaesa lass potential for adsorption onto aurfaces or filter media.
These two viruaas alao have been daaonatratad to hava somewhat different
diainfectlon kinetics.
Recownendatlon;
Retain the rotavirus test raquirementa.
H. Definition of microbiological water purifiers One general comment
requested redefinition based on "lack of any virus removal "requirement
N-34
-------
In the CPA primary drinking water regulations, so that no virus reduction
requirement should be included. Also, it was claimed that th* separation
of purifiers from non-purifiers would ba a "disservice to consumers and
othar users."
Discussion:
Viruses ara recognized in tha EPA regulations vis-a-vis a proposed recom-
mended maximum contaminant level of zero. Since virus monitoring for
compliance with a possible MCL is not yet feasible, a treatment require-
ment is necessary. Virus control will be considered in the Safe Drinking
Water Act filtration and disinfection treatment regulations. The reduc-
tion of viruses by treatment is discussed by Amirtharajah (J. awwa,
78i3i34-49, 1986).
With respect to consumers and other users, we feel that the current
definition is appropriate and necessary.' the average consumer cannot be
expected to know the difference between viruses, bacteria and cysts, or
when a raw water will or will not contain any of these organisms. Zn
order to protect the average consumer, the subject units either alone or
with supplementary treatment, should be able to cope with all of the
specified organisms.
Recommendationi
Retain the current definition for microbiological water purifier.
Z. Coverage of units< Several comments related to the coverage of units.
These questions are addressed individually as follows:
1. Ultraviolet units that are used for supplemental treatment of water
from public water system taps would not be covered. We agree that
such units are not covered and parenthetical language has been
included in Section 1.3.2.3 to clarify this point.
2. A special status should be given to units which remove Glardla and
bacteria but not virus. Specifically, the meaning of Section 1.2.4,
"Exceptions," was addressed. The Exceptions" section was specif-
ically developed to relate to the problem of public water, systems
having disinfection but no filtration on a surface supply. Cysts
alone have been found to survive disinfection treatment and could be
present in such treated waters. Zn this case an effective cyst
filter serves an Independent, beneficial purpose and should not be
required to be a microbiological water purifier. However, such a
unit should not be used as sole treatment for untreated raw water.
Additional parenthetical language has been added to Section 1.2.4.
3. The entire treatment unit or system should be tested, not just a
single component. We agree but believe that it is sufficiently
clear without providing additional language.
N-35
-------
4. the protocol should be expanded to cover units for the reduction of
TCE, EDB and other chemical pollutants. Ne felt that the introduc-
tion of non-microbiological claims to the standard would Bake it
large, unwieldy and. duplicative of an existing third-party standards
and testing program (see Section 1.2.5).
J. Alleged preference of National Sanitation foundation (NSF) over other
laboratories for conducting the microbiological water purifier testing
protocol. The consent indicated that we were giving NSF preferential
treatment "to the detriment of other laboratories well qualified to
perform the required protocol."
Discussion:
He have made appropriate references to existing standards (142 and f53)
developed by the NSF standards development process. Standard 53, the
health effects standard, was developed by a broadly based Drinking Water
Treatment Units Committee, including representatives from local, State
and Federal health and environmental agencies, universities, professional
and technical associations, as well as water quality industry
representatives. Zt was adopted in 1982 and the only test from it
utilized in our Report has been substantiated as described in Part B of
this "Response."
Nowhere in our report have we advocated NSF (or any other laboratory) as
the prime or only laboratory for implementing "the required protocol."
Recommendation!
No action needed.
K. Instruction concerning effective lifetime. One comment described an
alternate means for determining lifetime where a ceramic unit is
"brushed" to renew its utility and is gradually reduced in diameter. A
gauge is provided to measure diameter and to determine when replacement
is needed.
Recomnendatloni
Where a manufacturer provides a satisfactory "other" means of determining
lifetime, this should be accepted. Appropriate words have been added to
Section 2.4.l.C.
L. Ceramic candles should not be cleaned during testing because some
consumers would not clean them and this would provide the "worst case
test." One comment asserted this point.
Discussion:
There is some truth to this proposition. However, the other approach may
also have validity. Frequent brushing may reduce filtration efficiency.
N-36
-------
In any event, where a manufacturer prescribes filter cleaning and how te
do it, and providea a gauge to determine lifetime, we feel the teating
program ia bound to follow tha manufacturer's diraetlona.
Recommendation»
Ho change naadad.
N. Scaling up or down. Ona comment pointa out that ona or more manufac-
turara may vary piza of treatment unite by increasing or daeraaainf tha
number of oparativa units rathar than tha aiza of tha operative unit,
Tha conaant suggests allowing aealiitf baaad on size of oparativa unit.
Recoapendatlon > '
Or
We agraa with tha eoanant and hava addad clarifying word* to sec-
tion 3.3.3.3.
II, Turbidity laval of "not laaa that 30 NTO" for caranic candles or unita.
Ona coaanent atatas that "Such levela ara inpoasibla to utilize in taating
nechanical filtration devices which will clog entirely or require auch
frequent bruahing aa to render tha teat impoaaibla as a practical
siatter."
Dlscussloni
We recognized the potential "clogging problems" in Section 3.5.1.a(2)
where tha 30 MTU water is only to be applied immediately before and
during each aampling event> the non-sampling turbidity level, which will
be applied over 90% of the "on" time, is currently aat at no less than
10 NTO.
Turbidity levels of 30 NTU are commonly found in surface waters during
heavy rainfall or snow malt. Treatment units nay be used under these
circuastaneea, ao thia challenge level ahould be retained. However, most
uaage will occur under background conditions so tha non-aampling
turbidity levala ahould be 0.1-5 NTO.
Racommendatlonai
1. Retain aaapling turbidity level of not leaa than 30 HTU, and
2. Change non-saapling turbidity to 0.1-5 NTU. Appropriate wording
changes hava bean introduced in Section 3.5.1.a(2) and in Appen-
dix H-l, Section B.
O. Chlorine in test water #5. One comment aaaerts that chlorine "tends to
increase silver ion leaching activity" and that a high chlorine level
ahould be included in the ailver leaching test? but no reference or
evidence, however, ia provided to back this assertion.
N-37
-------
Discussion:
We have no couponing evidence or reason to expect that chlorine will
enhance the leaching of silver. However, the prescribed low pH and TDS
levels will provide a clearly severe test for silver leaching.
Recommendation :
No change needed.
P. Unnecessary difficulty and expense of test protocols. Several comments
were nade under this general heading. These coanents are" outlined and
discussed as followsi
1. Too many sampling events are required; sampling of a few units at
start, middle and finish should be satisfactoryi The comnittee has
carefully laid out the standard and protocol and we feel the minimum
sampling plan must be maintained for the consumers' health pro-
tection.
2. Three units are too many to study; parallel testing of .two units
should be satisfactory: For consumer protection, the Disinfectants
Branch, Office of Pesticide Programs, has traditionally required the
testing of three units. The coanittee recognizes . the additional
cost involved in testing a third unit but feels that this will
provide a minimum level of assurance to prevent infectious disease
and recommends retention of the 3-unit requirement.
3. The protocol requires large tanks and microbiological reseeding on a
daily basis: We feel that the tank size requirements are not
extreme and can be met by an interested laboratory. With respect to
reseeding, it should be pointed out that virus and cyst seeding need
only be conducted immediately before and during the sampling "on"
period (see Section 3.5.1.b(2)), equivalent to less that 10% of the
"on" time. This "spot" seeding for viruses and cysts recognized the
expense and .difficulty of maintaining large populations of these
organisms. Continuous seeding was provided for bacteria because
they are easier to grow and maintain and might have the capacity to
grow through some units, given enough time and opportunity.
4. Challenge levels of contaminants are too high compared to known
environmental conditions and the required log reductions exceed Safe
Drinking water Act requirements: As explained in a footnote to
Table 1, Section 2, the influent challenges may constitute greater
concentrations than would be anticipated in source waters. These
levels are necessary to test properly for the required log reduc-
tions without having to utilize sample concentration procedures
which are time/labor intensive and which may, on their own, intro-
duce quantitative errors to the microbiological assays. As men-
tioned in Part I of this paper, the log reductions for bacteria,
virus and Giardla have been suggested for public water system
N-38
-------
treatment in * papet by Aoirtharajah (1986, JAMWA/ 78:3«34-49)s The
reductions in the aierobioiofical purifier standard are entirely
ccnpatible with the reductions cited for public water supply
treatment%
N-39
-------
APPENDIX 0
GUIDELINES TO EVALUATE OZONE DISINFECTION
Principal Technical Author: Dr. Ovadia Lev
Division of Environmental Sciences
Hebrew University of Jerusalem
Jerusalem, Israel
-------
APPENDIX 0
GUIDELINES TO EVALUATE OZONE DISINFECTION
TABLE OF CONTENTS
Paoe
0.1 INTRODUCTION 0.1-1
0.1.1 Background 0.1.1
0.1.2 Objectives of the Recommended Guidelines 0.1-2
0.1.3 EPA's Approach 1n Setting the Recoonended
Guidelines 0.1-3
0.1.4 Typical Ozone Disinfection Units 0.1-4
0.2 DETERMINATION OF CONTACT TIME (TJ 0.2-1
0.2.1 Background 0.2-1
0.2.2 Tip Analysis 0.2-1
0.2.3 Additional Considerations for T10:
Multiple Chamber Contactors . 0.2-3
0-2-4 Alternative Analysis of Disinfection Kinetics 0.2-7
0.2.S Continuously Stirred Tank Reactor (CSTR) Approach 0.2-9
0.2.6 Segregated Flow Analysis (SFA) 0.2-10
0.2.7 Relative Inact1vat1on of G1ard1a Cysts and Viruses 0.2-12
0.2.8 Examples of Determining Contact Time (T) 0.2-13
0.2.8.1 Evaluation Using Tl(, 0.2-13 '
0.2.8.2 Evaluations Using CSTR Calculations 0.2-16
0.2.8.3 Evaluations Using SFA 0.2-18
0.2.9 Estimating T 0.2-21
0.3 DETERMINATION OF OZONE CONCENTRATION (C) 0.3-1
0.3.1 Introduction 0.3-1
0.3.2 Direct Measurement of C 0.3-2
0.3.3 Estimating C Based on Residual Measurements
at the Outlet 0.3-6
0.3.3.1 First Chambers 0.3-6
0.3.3.2 Subsequent Chambers 0.3-7
0.3.4 Estimating C 0.3.-10
0.4 SITE SPECIFIC EVALUATION OF OZONE CONTACTORS 0.4-1
0.4.1 Introduction 0.4-1
0,4.2 Site Specific Correlations of C with an
Observable Variable 0.4-2
0.4.2.1 Utilizing Off-Gas Measurements 0.4-4
0.4.3 Modeling the Performance of Full Scale Operations 0.4-7
0.4.4 Microbial Indicator Studies to Model Inact1vat1on
Contactors 0.4-9
REFERENCES
-------
TABLE OF CONTENTS (Continued)
LIST OF FIGURES
Figure Following
No. Title Page
0-1 Turbine Contactor, Haworth Water Treatment Plant 0.1-5
Hackensack, NJ
0-2 Multiple-Chamber Ozone System, Los Angeles
Aqueduct Filtration Plant 0.1-5
0-3 Multiple-Chamber .Ozone System, City of Tucson, Arizona 0.1-5
0-4 Multiple-Chamber Ozone System. East Bay Municipal
Utility District, Oakland, California 0.1-5
0-5 Schematic of Oeep U-Tube Ozone Contactor 0.1-6
0-6 Schematic of In-Llne Static Mixer 0.1-6
0-7 Principal of Segregated Flow Analysis 0.2-11
0-8 Segregated Flow Analysis of an Ozone Contactor
Tracer Study 0.2-14
0-9 Tracer Study of Sturgeon Bay Ozone Contactor 0.2-15
0-10 Segregated Flow Analysis of Ozone Contactor
- Integration of Survival Efficiency. 0.2-20
0-11 Segregated Flow Analysis of Ozone Contact Chamber 0.2-20
0-12 Decision Tree for Estimating T 0.2-21
0-13 Flow Configurations In Ozone Contact Chambers. 0.3-1
0-14 Direct Measurements for Determining T 0.3-5
0-15 Ozone Concentration Profiles 0.3-8
0-16 Decision Tree for Estimating T 0.3-10
0-17 Example of Empirical Correlation of Residual Ozone
and Off-Gas 0.4-6
-------
TABLE OF CONTENTS (Continued)
LIST OF TABLES
Table Following
No Title Page
O-l Recommended Procedures to Calculate Contact Time (T) 0.2-2
0-2 CT Values for Inactlvatlon by Ozone 0.2-7
0-3 k Values for Inactlvatlon by Ozone 0.2-7
0-4 Spread Sheet Notations for SFA 0.2-18
0-5 Segregated Flow Analysis of an Ozone Disinfection
Contactor at Hackensack 0.2-20
0-6 Correlations to Predict C Based on Outlet Ozone
Concentrations 0.3-6
0-7 Henry's Constants for Ozone 0.4-5
0-8 Empirical Correlation Between C^ and yM 0.4-6
-------
0.1 INTRODUCTION
0.1.1 Background
Tht" Surface Water Treatment Rule (SWTR) specifies overall minimal
removal/inactivation efficiencies by filtration and disinfection for
Giardia cysts and viruses. The SWTR uses the "CT" concept to predict
inactivation efficiencies of microorganisms by disinfection. "CT"
represents the product of contact or exposure time ("T"j and the
concentration of disinfectant ("C") during disinfection. The Guidance
Manual suggests design, operating and performance criteria for specific
surface water quality conditions to provide compliance with the SWTR.
Appendix C of the Guidance Manual recommended guidelines for the
determination of contact time (T,0) for the disinfection of drinking water.
T10 is the time defined to assure that 90 percent of the water that enters
the disinfection chamber will remain for at least T,0 minutes. This
appendix recommends additional procedures which may be used for consistent
determination of the C and T for systems using disinfection by ozone.
Ozone has unique characteristics and warrants special consideration
for estimating inactivation efficiencies. In developing these recommended
procedures, EPA addressed the following complications that are specific to
ozone disinfection and distinguish it from other typical disinfection
processes.
Despite the long operational experience with ozone disinfec-
tion, the data regarding performance of ozone as a disinfectant
are rather limited. Most of the available inactivation rate
data are derived from laboratory conditions which are sub-
stantially different than full scale continuous operation,
generally more so than for other disinfectants.
From a technical point of view, disinfection of drinking water
by ozone is more complicated than disinfection by other common
disinfectants because of ozone's unique gas-liquid mass
transfer characteristics. Ozone requires sophisticated mass
transfer equipment to Introduce It into water, because of the
relatively low ozone concentration in the feed gas. Ozone is
a powerful oxidant, that reacts rapidly with organic and
inorganic substances present in the water and undergoes auto-
decomposition. Therefore, it's residual is much less stable
than that of other disinfectants and dissipates rapidly.
0.1-1
-------
• Ozone contactors •xhlblt more diversified types of flow
configurations relative to the flow pattern 1n contactors for
the other disinfectants. The flow configuration often ranges
from an almost continuously stirred tank reactor (CSTR1 to an
almost Ideal plug flow configuration, making the determination
of contact time for ozonation more complex than for other
disinfectants.
• Ozone contactors are closed vessels because of ozone's toxici-
ty. The contactors have United access for measurement of the
ozone concentration profile within the contactor. Gas bubbles
also may interfere with the determination of the dissolved
ozone concentration. If the bubbles are entrapped during
sampling*
Ozone technology Is still evolving and new types of ozone
contactors are being developed. These guidelines should not
set unnecessary obstacles that will Inhibit engineering
progress and prevent Innovative designs of disinfection
systems.
CPA's procedures for determining C and T for disinfection with ozone
differ from those recommended for systems using chlorine, chloramlnes or
chlorine dioxide as disinfectants. The CT evaluation procedures
presented 1n previous chapters of the Guidance Manual are not appropriate
for ozone disinfection because they would result in excessive ozone
dosages. Excessive ozone doses result 1n high energy requirements and
costs and may lead to unnecessary production of ozonation by-products
which may have associated health risks. Additionally, excess dissolved or
entrained ozone should be destroyed or removed before reaching the first
drinking water consumer or plant personnel, In order to prevent health
risks. Therefore, excessive dosage of ozone may require an additional
unit operation to destroy the remaining residual ozone.' This process 1s
expensive and may not be necessary 1f guidelines such as those presented
1n this section are used for compliance with the SWTR.
0.1.2 Obl&ctives of the. Recommended Guidelines
The recommended guidelines were developed to assure compliance with
the SWTR for the whole range of flow rates, flow configurations and water
quality conditions that may be encountered with ozone disinfection of
drinking water. The primary goal of these guidelines 1s to assure
0.1-2
-------
compliance with the SWTR even under "worst cast" conditions. Without
compromising this primary goal, these guidelines were developed to.meet
the following criteria:
1. Simplicity: The guidelines for selecting contact time (T) and
concentration (C) have to be easily understood by practitio-
ners, even by those who do not have an engineering background.
2. Implementation: The procedure to estimate concentration and
time should be easily Implemented, even by water utilities that
have only limited engineering and technological means.
3. Economics: The guidelines should be designed to minimize
capital and operating costs and to minimize ozone consumption.
The guidelines should be flexible enough to allow systems to
take advantage of site specific characteristics of the treated
water and the various designs of ozone contactors.
0.1.3 EPA's Approach 1n Setting the Recoronended Guidelines
EPA 1s aware that the current technological knowledge 1s Insufficient
to formulate a consistent and efficient single set of general rules that
will achieve these conflicting goals and still guarantee compliance with
the SWTR. Therefore, EPA developed two alternative sets of guidelines
that systems may use depending on their technological resources:
Alternative 1: General guidelines which assure compliance with
the SWTR regardless of the site specific conditions,
Alternative 2: A sophisticated evaluation procedure that water
utilities may use to take advantage of their site specific
conditions.
These guidelines are considered to be state-of-the-art. As more
Information becomes available, more accurate approaches and models may be
developed. A brief description of the current alternatives follows.
Alternative 1 - General Guidelines
This alternative consists of a simple set of general guidelines that
assure compliance with the SWTR even under worst case conditions. These
guidelines were developed to emphasize generality and simplicity.
However, they may not result 1n the lowest cost alternative(s).
0.1-3
-------
The second and third sections of this Apptndlx contain detailed
descriptions of the general guidelines. Section 0.2 contains procedures
to estimate the contact time (T) and Section 0.3 contains procedures to
calculate the concentration (C) In ozone contactors based on simple
Measurements of some parameters. The basis for these general guidelines
Is discussed 1n two papers (Lev and Regl 1, 1990a,b).
Alternative 2 - Site Specific Evaluation Procedures
This alternative consists of a more sophisticated set of evaluation
procedures to characterize the performance of ozone contactors and thereby
take advantage of site specific conditions. IPA recommends that systems
be given opportunity to prove by further experimental and analytical data
that the performance of their ozone contactors are better than the
performance predicted by the first alternative, thereby allowing a system
to minimize costs while providing adequate treatment.
Section 0.4 outlines recommended procedures for demonstrating that
ozone contactors achieve better performance than that predicted by the
first alternative.
0.1.4 Tvolcal Ozone Disinfection Units
Several types of ozone contactors are currently In use for disinfec-
tion of drinking water 1n the United States. Other types of contactors
are being designed or are being used for disinfection of treated sewage
effluents. The following characteristics Illustrate the diversity of
ozone contactors:
The capacity of ozonation systems ranges from less than 1
¦illion gallons per day (ragd) up to GOO mgd.
The volume of ozone contactors ranges from less than 35 cubic
feet up to more than 35,000 cubic feet for a single chambers.
The ozone gas stream may be introduced Into the water by
several ways Including porous dlffusers, submerged turbines and
gas Injectors.
Ozone contactors include single or multiple gas/1iquld contact
chambers.
0.1-4
-------
Four typical ozone contactors currently In use or In design In the
United States are shown on Figures 0-1 through 0-4. Figure 0-1 presents
a schematic of an aspirating turbine contactor, operating 1n coun.tercur-
rent flow. A turbine agitator 1s used to Introduce the ozone into the
contactor and to mix the liquid phase. This unit say serve as the first
ozone chamber In a series of chambers or as a single chamber. The unit
shown 1n this f1gur*e 1s from the Hackensack Water Company's Haworth Plant
at Haworth, New Jersey. The turbine chamber Is followed by a reactive
chamber to provide additional contact time. Studies conducted 1n the full
scale turbine agitated contactor demonstrated that even when the ozone
demand was high, the dissolved ozone concentration was almost constant
throughout the contactor as a result of the vigorous action of the turbine
(Schwartz et al, 1990).
The 600 mgd ozone system of the city of Los Angeles 1s comprised of
four parallel contactors each consisting of six chambers. A schematic of
one of these contactors 1s presented on Figure 0-2. (Stolarlk and
Christie, 1990) As Indicated on this figure:
An oxygen stream containing a few percent by weight of ozone
1s compressed through bubble dlffusers Into the first and third
chambers of the contactor.
The second and fourth chambers are used to provide contact
time, without supplying additional gas to the liquid stream.
The size of the first three gas/11qu1d contact chambers 1s
20,400 cubic feet each.
The fifth and sixth chambers are the ozonated water channel and
the rapid mixer basins.
The liquid and gas streams 1n the first and third chambers flow
In a counter-current pattern; the gas stream flows upward and
the water stream flows downward.
As Illustrated on Figure 0-3, a similar design approach was taken by
the City of Tucson, Arizona. This contactor Is comprised of five
chambers, all of which are equipped with gas dlffusers. The sixth chamber
has no dlffusers. The flow 1n all six chambers 1s counter-current flow.
These counter-current chambers are separated by narrower co-current liquid
channels 1n which the water flows upward to the Inlet of the next chamber.
0.1-5
-------
The East Bay Municipal Utility District Oakland, California 1s
currently designing two 60 mgd ozone contactors, the first of which 1s to
be operational 1n 1991. As Illustrated on Figure 0-4, the contactor
Includes three ozone gas/liquid chambers followed by three sore reactive
chambers to provide additional contact time. The first and third chambers
are counter-current and the second chamber 1s co-current. In the latter,
the water and the gas bubbles flow 1n the same direction.. Hydrogen
peroxide can be added at the outlet of the contactor to dissipate any
residual dissolved ozone.
The following types of contactors are already used 1n other parts of
the world, but have not yet been Installed 1n the United States:
The Deep U Tube contactor shown on Figure 0-5, 1s comprised of
two concentric flow tubes. Water ana gas streams are intro-
duced at the top of the inner tube and the mixture is pumped
10 to 30 meters downwards at a velocity greater than the rise
rate of the gas. After reaching the very bottom of the
contactor the mixture flows up in the outer section of the
contactor. The Deep U-tube Is basically a co-current operation
taking advantage of the Increased mass transfer at high
pressures.
The Static Mixer (shown on Figure 0-6) consists of a flow tube
equipped with baffles to produce efficient contact between the
liquid and the gas streams. This installation 1s gaining
popularity 1n Europe particularly for small and medium size
disinfection units. Here the flow 1s basically co-current, the
liquid and gas flow is in the same direction, through a tube
equipped with baffles that create turbulence and thus increases
the rate of gas-Hqu1d mass transfer. The ozone is applied to
the water prior to the mixer either through an eductor or a
dlffuser. Following dissolution through the mixer, the water
flows through a pipeline in plug flow.
Some contactors, particularly for disinfection of wastewater
effluents, use packed beds to increase mass transfer. Co-
current or counter-current flow configuration may be used.
The guidelines were developed to represent four different flow
conditions in ozone contactors. However, other types of contactors or
flow conditions may still use the same guidelines 1f the features of the
gas-Hquid flow configuration as presented 1n Section 0.4 of this appendix
are taken into account.
0.1-6
-------
RAW WATER INLET
EFFLUENT
SELF
ASPIRATING
TURBINE
REACTIVE
CHAMBER
FIGURE 0-1 - TURBINE OZONE CONTACTOR,
HAWORTH WATER TREATMENT PLANT
HACKENSACK. NJ
-------
OZONE CONTACT BASIN
WATER INLET
t i
o o o o o o
o o o o o o
WATER
OUTLET
OZONC FEED CAS SPLIT
FIGURE 0-2 - MULTIPLE-CHAMBER OZONE SYSTEM, LOS-ANGELES AQUEDUCT
FILTRATION PLANT, STOLARIK #! al. (198S)
-------
fUPV
OZONE DFRJSESS
RAW WA7EM
MPOUNCwerr
FIGURE 0-3 - MULTIPLE-CHAMBER OZONE SYSTEM, CITY
OF TUCSON, ARIZONA, JOOST «t al. (1960)
-------
t-INHACIM
IKtttJM. 10
moNi
OISINUCI
oiomi rfcocAi
—> /JA
]WMUIC
ui inn ion
VIMi
tMMtfli
MIlNTiUN 11 OCTIKTKM
TMC
ItMMUfl
billHlMM
lIMt
9MWUVI
DtltNIMM
IMk
cm *
till 1
Ctll »
IMIMPatMMM
miwtliiMt
mmmictmm
IWIMUIWM
DMMKIMM I llWHIPAliUUI11UJIItWHIMM
FIGURE 0-4 - MULTIPLE-CHAMKVI OZONE SYSTEM. EAST BAY IfUMCIPAL
UTILITY DISTRICT. OAKLAND. CALIFORtlA. CAfdiS (!•#«)
-------
INLET OF WATER
OUTLET
INLET OF OAS
OUTLET OF WATER
DESCENDING
TUBES
EXTERNAL-
TUBING CUVE
FIGURE 0-5 - SCHEMATIC OF THE DEEP U-TUBE OZONE
CONTACTOR, ROUSTAN «t al. (1987)
-------
FIGURE 0-« - SCHEMATIC OF IN-LINE STATIC MIXER
-------
0.2 DETERMINATION OF CONTACT TIME (T)
0.2.1 Background
The hydraulic characteristics 1n ozone contactors range from an
almost Continuous St1rred-Tank Reactor (CSTR) to an Ideal plug flow
configuration. Because the T10 approach nay not be adequate for determin-
ing the 1nact1vat1on provided for systems resembling a CSTR, and because
the T10 approach 1s overly conservative 1n other cases, EPA recommends the
following three numerical methods to predict the contact time (T) In ozone
contactors:
T10: The Tt0 method discussed In Appendix C (and In Section 0.2.2) Is
a good measure to characterize the contact time 1n most cases.
However, this method reduces the possibility of complying with the
SWTR for systems that fjave relatively high back-m1x1ng and require
high 1nact1vat1on levels.
Segregated Flow Analysis (SFA): (See Section 0.2.6) This 1s an alter-
native procedure to calculate the disinfection contact time. This
procedure 1s applicable only to systems that have good data from
tracer studies of high resolution as explained 1n Section 0.2.6.
CSTR: The Continuously St1rred-Tank Reactor (CSTR) method described
1n Section 0.2.5, assumes the ozone contactor behaves as a CSTR.
This procedure 1s extremely conservative. However, no apparent
simplified analysis 1s currently available to make It less conserva-
tive. The CSTR approach should be used only when:
Other predicting techniques are not recommended,
The required 1nact1vat1on level 1s very low, or
Systems cannot afford to get good tracer study data for other
methods.
Systems may choose the optimal method for their situation based on
the available data to perform the calculations. A discussion of each 1s
presented belqw.
0.2.2
The simplest method of calculating the contact time, T, of microor-
ganisms In a contactor Is by the T10 approach. T10 1s defined as the
detention time to assure that 90 percent of the liquid that enters the
0.2-1
-------
contactor will rtmaln at least T,0 minutes In the contactor. A system
achieving a CT10 corresponding to X percent 1nact1v«t1on, will assure that
90 percent of the water passing through the contactor Is receiving at
least X percent 1nact1vat1on, while 20 percent of the water will receive
less than X percent 1nact1vat1on.
When conducting a step-Input tracer study, T10 is the time Interval
required for the outlet tracer concentration to achieve 10 percent of Its
ultimate response, following an Inlet step addition. Appendix.C of this
manual contains procedures to conduct and evaluate tracer studies for the
determination of T10. Appendix C also contains procedures to estimate the
T10 of contactors based on their baffling conditions and flow configura-
tion.
The results of tracer studies conducted on several ozone contactors
(Stolarlk and Christie, 1990, Schwartz et al, 1990, Rosenbeck et ai, 1989)
Indicate that high quality tracer data on ozone contactors can be obtained
and that T10 can be estimated with high precision, but to a lesser degree
when T10 1s less than one minute.
T10 1s a good measure of the contact time In most contactors and the
safety margin provided by using T10 compensates for the Inferior perfor-
mance of contactors with a high degree of short-c1rcu1t1ng and backmixing
relative to contactors that approach plug flow conditions, (see Lev and
Regli, 1990a, for further detail.) However, for contactors with a high
degree of short-c1rcu1t1ng jui a need to provide a high level of Inactiva-
tion, this safety margin falls to compensate for the effect of backmixing.
In such cases, approximately 10 percent of the water passing through the
contactor receives significantly less than the 1nact1vat1on Indicated by
CT,0. In these cases, either the SFA or the CSTR approach should be used
for determining the contact time.
The recommended alternatives for determining the contact time (T) for
various conditions of T,0 versus hydraulic detention time (HOT) are
presented 1n Table 0-1. HOT 1s determined by dividing the liquid volume
of the contactor by the rate of flow through the contactor. As illustrat-
ed in this table:
0.2-2
-------
TABLE 0-1
Rtcomended Procedures to Calculate the
Disinfection Contact Tine (T)
Condition: T10<(HDT)/3 T.o<(H0T)/3 T.0>(HDT)/3
4 4
•Log(I/I0)tn < 2.5 -Log(I/I0)<1> >- 2.5
Recommended
Methods: T-T,0
SFA(2)
CSTR13'
T-T„
sfa(2) sfa(2)
cstr<3) cstr(3)
Notes:
1. Required level of 1nact1vat1on in logs of either Glardia
Iambi 1a cysts or viruses whichever value 1s greater;
I - I live organisms 1n outlet of ozone contactor and
I0 ¦ f live organisms 1n Inlet to ozone contactor
2. High resolution tracer characterization of the ozone contactor
must be available.
3. The CSTR method 1s extremely conservative and should be avoided
when alternative approaches are possible.
-------
The T1# method Is applicable for systems that art required to
achieve ltss than a 2.S-1og 1nact1vat1on of Slardii cysts even
If the flow configuration In their ozone contactor approaches
that of a CSTR, such as disinfection In contactors using
turbine agitators.
• Likewise, the T10 approach Is appropriate for systems demon-
strating T10/HOT greater than 1/3 regardless of the required
level of disinfection.
Systems for which the T10 approach 1s appropriate to have the
option of applying either the SFA or CSTR analysis. The method
resulting 1n the highest T value, or thereby the lowest C value
may then be followed.
The SFA or CSTR should be used 1n Heu of T10 when the:
- Level of 1nact1vat1on required for Glardla cysts and/or viruses
1s 2.5-log or higher
- T10/HDT 1s less than 1/3.
Systems should be aware that the 2.5-log 1nact1vat1on guideline refers to
the 1nact1vat1on provided by the ozone system alone regardless of
1nact1vat1on provided by other disinfectants. For example, 1f a system
requires an overall 1nact1vat1on of 3-1og and provides l-1og 1nact1vat1on
by chlorine, then a 2-log 1nact1vat1on 1s required by ozone and the T10
approach can be used.
Examples for applying the different methods of calculation for T are
Included 1n Section 0.2.8.
0.2.3 Additional Considerations for T10: Multiple Chamber Contactors
This section provides guidelines for computing T10 for several
contactors 1n series. The main shortcoming of the T10 approach 1s the
Inherent non-linearity of this measure. In contrast to the HDT, which 1s
a linear measure, T10's of Individual subunlts do not sum up to give the
T10 of the overall unit. For example:
- The HDT of two equal CSTRs 1n series 1s exactly twice the HDT
of each CSTR.
The T^ for the same two CSTRs 1n series 1s more than twice the
sum of the Individual T10's.
0.2-3
-------
This raises some practical questions:
How should the T10 of a multiple-chamber contactor be deter-
mined using tracer studies?
Is 1t necessary to conduct Individual tracer studies for each
chamber or 1s It sufficient to conduct an overall study of the
whole contactor?
How can'the contact time of one chamber be determined based on
the T10 of the overall system?
Conducting tracer studies of Individual chambers 1n a multiple
chamber ozone contactor 1s likely to be difficult. In addition, an
analysis conducted by Lev and Regll (1990a) Indicates that the computation
of the contact time (T) based on tracer studies of the Individual chambers
Is likely to lead to over design. The excess volume of a system designed
by summing the T1Qs of the separate chambers may be up to 9.5 times higher
than one designed by the overall T,j approach. Therefore, EPA recommends
the use of an overall tracer study of the whole contactor, 1n order to
lower operation costs and to avoid overly complex tracer studies.
Disinfection credits for a multiple chamber contactor should be based
only on the active chambers, those which have a detectable ozone residual.
Based on the recommendation to use overall tracer studies, guidelines are
needed for determining the disinfection credit for the active part of a
system based on overall tracer studies. The average concentration 1n the
Individual chambers of a multiple-chamber system may deviate considerably
from one another. Therefore, systems must be able to assign contact
times for each chamber.
Lev and Regll, (1990a) evaluated the consequences of using a linear
approximation based on relative contact chamber vplumes and overall T10 of
the contactor to determine the contact time of Individual chambers 1n an
ozone contactor:
0.2-4
-------
TlO.cfcMbtr " (Vdtwter) 1^10,total ) /(V,.t.i) (1)
Where:
Tio,en««Mr * An approximation for the contact time of one
chamber.
T„ total " T,0 of the entire multl-chamber ozone contactor as
determined by tracer studies
Vr,-t11- - Volume of the Individual chamber
vtot*i " Ov^all volume of the multl-chamber ozone conta-
ctor
9
They demonstrated that such 1Inear extrapolation may lead to an underesti-
mate of the required T. This underestimate can be significant when the
concentration In the different chambers deviate considerably from each
other. This would be the case when the residual ozone concentration 1n
one chamber 1s zero.
Considering the various safety margins that are Included 1n the T10
approach, and considering the practical complexity Involved In conducting
separate tracer studies, EPA recommends the use of the linear approxima-
tion described 1n Equation 1 provided that the volume of the portion of
the contactor that has zero residual ozone 1s less than half of the
overall volume of the ozone contactor:
^Jnoetivo dia^or/^total *
Where:
v, , ....rtnf ¦ The volume of the chambers 1n the contactor
where the ozone concentration 1s zero
V.. , - The volume of the chambers with a residual
tote I
The following examples Illustrate the computation of the overall
1nact1vat1on performance of jnultlpie-chamber systems using the linear
approximation of Equation 1:
Example 0.2-1 linear approximation to predict T«„
An ozone contactor has three chambers In series. Each chamber
has a volume of 353 cubic feet.
0.2-5
-------
The average ozone concentration In each chamber Is:
First chamber: C,-0 ng/L ozone.
Second chamber: C2«l ng/L ozone.
Third chamber: C, «0.5 ng/L ozone.
C,, C. and Cj are the average concentrations, determined as
described In Section 0.3.
The utility measured T10 - 5 mln for the entire ozone con*
tactor.
The volumetric fraction of the chamber which has no ozone
residual 1s V,/(V,+V2+Vj)) - 0.33 which 1s less than the 0.5
guideline. Therefore ft 1s permissible to use Equation i 1n
order to estimate the CT achieved In the ozone contactor.
* '
The total CT achieved by the ozone contactor Is:
" - (C2)[(T10>toti;)(V2)/(Vtotll)] + (C3)[(T10(tot-l)(V,)/(Vtot.l)]
CT - (1)[(5)(10)/(30)] + (0.5)[(5)(10)/(30)] - 2.5
The CT achieved by the ozone contactor 1s 2.5 mg-m1n/L.
Example 0.2-2 Linear approximation not applicable
An ozone contactor consists of:
A chamber with a volume of 70 cubic feet and equipped
with a turbine agitator
Followed by a second chamber with a volume of 200 cubic
feet.
The first chamber has an ozone residual of 0.5 mg/L and the
second chamber has an ozone residual of zero
The T10 tot-l - 8 m1n for both chambers at the peak flow rate
The volumetric fraction of the chamber with no ozone residual
1s 200/270 • 0.74 which Is greater then 0.5 of the total
volume. Therefore, the use of Equation 1 to approximate the
T10 of the chamber that contains an ozone residual Is not
recommended.
The system may estimate Its performance by either the CSTR
approach taking Into account only the detention time of the
first chamber or conduct tracer studies of the first chamber.
0.2-6
-------
0.2.4 Alternative Analysis of Disinfection Kinetics
The CSTR and the SFA approaches utilize the Chick-Watson inactlvatlon
rule directly rather than relying on the CT approach. The following
section describes this alternative approach to represent the disinfection
kinetics.
The Guidance Manual recommends that systems should calculate the
1nact1vat1on level in their disinfection contactors by the CT approach.
Table 0-2 presents CT data corresponding to specified inactlvatlon levels
of Ciardla cysts and viruses by ozone. An alternative nay to present the
same Information is by tables of the kinetic coefficients used to
calculate the CT values.
The CT values presented in Table 0-2 were calculated based on batch-
reactor experimental Information that was fitted into a logarithmic
correlation according to a first order Chick-Watson's rule (Chick, 1907;
Watson 1908; and Hoff, 1987):
log(I/I0) - - k CT (2)
Where:
1/I0 - Survival ratio of the SiirdU cysts or viruses
C - Residual concentration of ozone 1n mg/l
J - Exposure time in m1n.
k A kinetic coefficient which characterizes the
specific rate of inactlvatlon of the microorgan-
isms at the appropriate temperature and pH.
Solving Equation 2 for k yields:
k - -loo IVM (3)
CT
Equation 3 can be used to calculate k values corresponding to the CT
values in Table 0-2. Table 0-3 summarizes these k values. Equation 3 may
also be used to transform inactlvatlon levels (I/IB) to CT values and vice
versa.
The following example illustrates the use of the values presented in
Table 0-3 to calculate the performance of multiple-chamber ozone
contactors:
0.2-7
-------
Example 0.2-3 Multiple-chamber Ozone Contietor
An ozone contactor consists of three chambers 1n series .
Temperature 1s 5 C.
The first chamber has a 10 percent survival ratio for 61ard1a
cysts, or (I/I0) - 0.1, which also corresponds,to 90 percent
1nact1vat1on
The second chamber has an I/I0 • 0.07
The third chamber has an I/I0 - 0.03
The total 1nact1vat1on may be calculated by either summing CT's
or summing logs of Inactlvatlon, as presented below.
Summing CT's:
At 5 C the k for Glardla cvsts - 1.58
The survival fractions are:
First Chamber -0.1
Second Chamber - 0.07
Third Chamber - 0.03
Therefore, the CT values 1n each of the chambers are:
First chamber:
CT - -log(I/I0)/k -'-log (0.1) /(I.58) - 0.63
Second chamber:
CT - -log (I/I0)/k - log (0.07)/1.58 - 0.73
Third chamber:
CT - -log(I/I0)/k - -log <0.03)/l.58 - 0.96
Total CT 1s : 0.63 + 0.73 ~ 0.96 • 2.32
As Indicated in Table 0-2, a CT of 2.32 1s sufficient to
achieve a 3-log Inactlvatlon of fillElll cysts.
Summing logs of Inactlvatlon:
First chamber: -log (I/I0) ¦ -log(O.l) - 1
Second chamber: -log(I/I0) - -log (0.07) ¦ 1.15
0.2-8
-------
TABLE 0-2
CT VALUES FOR
INACTIVATION BY OZONE
G1«rd1a
Inactivatlon
0.5 log
1 log
1.5 log
2 log
2.5 log
3 log
Virus
Inactivatlon
2 log
3 log
4 log
S±1
0.48
0.97
1.5
1.9
2.4
2.9
0.9
1.4
1.8
-i.
0.32
0.63
0.95
1.3
1.6
1.9
0.6
0.9
1.2
UL
0.23
0.48
0.72
0.95
1.2
1.4
0.5
0.8
1.0
13L
0.16
0.32
0.48
0.63
0.79
0.95
0.3
0.5
0.6
2JL
0.12
0.24
0.36
0.48
0.60
0.72
0.25
0.4
0.5
2L.
0.06
0.1(
0.2-
0.3!
0.41
0.4
0.]
o.J
0.:
-------
TABLE 0-3
k Values for Ozone Inact1vat1on<1}
TEMPERATURE (C) M- 5 UL- JJL_ 21— 2S
Inactivation 1.03 1.58 2.08 3.12 4.17 6.25
of fiiiidii cysts
Inactivation 2.22 3.33 4.00 6.67 8.00 13.3
of Viruses
(1) k - -log(I/I0)/(CT) 1n L/mg-mln. When Chick's rule 1s repre-
sented by the formula In(I/I„) • -K CT (In stands for the
natural logarithm) then k should be calculated by k - 2.303(K)
-------
- Third chamber: -log(I/I0) - -log (0.03) - 1.12
Th§ total logs of 1nact1vat1on is;
•log (1/I0) - 1 + 1.15 ~ 1.12 - 3.67,
- Th« 3.67-log Inactlvatlon of Slardli cysts fs higher than
the required 3-log inactlvatlon
0-2-5 Continuously Stlrred-Tanlc Reaetor (CSTRi Approach
The CSTR aiethod assumes that the flow configuration in the ozone
contactor approaches that of completely stirred reactor. In most cases,
this calculation method 1s the most conservative approach. Studies by
Schwartz et al (1990) suggest that well-operated turbine contactors
approach Ideal CSTR characteristics and the CSTR calculation is appropri-
ate. In some cases, CSTR calculations offer the only apparent method to
evaluate the performance of the ozone contactors. CSTR calculations
should be used under the following conditions If systems have no other
means for demonstrating the inactlvatlon efficiency.
Tracer data are not available.
The required inactlvatlon level 1s greater than 2.5-log, and
ozone disinfection Is applied in a single chamber contactor
with T10/HDT < 1/3.
If either the required Inactlvatlon level 1s less than 2.5-log
T10/HDT > 1/3 then the inactlvatlon predicted by CT,g Is
appropriate provided that tracer data are available. If nigh
resolution tracer data are available then the SFA method can
be applied regardless of the level of Inactlvatlon required or
the ratio of T10/HOT.
In some cases, systems may actually receive more credit by using the
CSTR approach then by using the T10 approach. Hlghe'r credits result when
a low level of ozone disinfection such as 0.5-log 1s required and mixed
contactors are used.
When using the CSTR approach, the inactlvatlon performance should be
evaluated for viruses and Glardla cysts, regardless of which required CT
1s higher. This recommendation results from the influence of flow
characteristics on contactor performance, as discussed 1n Section 0.2.7.
The performance equation for a CSTR is based on two important
assumptions:
0.2-9
-------
1. The concentritlon of disinfectant and artcroorganIsms 1$
homogeneously distributed 1n the contactor.
2, First order Chick-Watson's law appllas. That 1s, tht rata of
1nact1vat1on of the microorganisms Is approximately proportion-
al to tha concentration of the Microorganisms and the concen-
tration of disinfectant.
The performance of a CSTR contact chamber Is given by:
(I/I,) %1/[1 ~ 2.303(k)C(WT)] ' (4)
Where:
k - kinetic coefficient for microorganism Inactlvatlon
[k values are listed 1n Table 0-4 (t/mg-min)j
(I/I0) ¦ Survival ratio of organisms
C • Average concentration of disinfectant (mg/L)
HOT ¦ Hydraulic detention time (mln)
Equation 4 nay also be used to calculate the ozone concentration that
Is required to achieve a specified level of inactlvatlon for a given HOT
or to compute the HOT required to achieve a desired.inactlvatlon level for
a given ozone concentration. Equation S restates Equation 4 for use in
determining C or HOT
C(HDT) - [l-(I/I0)]/[2.303 k (I/I0)] (5)
The effects of nixing on improving disinfection effectiveness nay be
very significant in CSTR contactors, and are not accounted for 1n this
model.
Examples demonstrating how to calculate the operating conditions
necessary to meet the required Inactlvatlon levels by the CSTR approach
are Included in Section 0.2.8.2.
0.2.6 Segregated Flow Analysis fSFAl
SFA is a method that is often used to characterize chemical
reactions. Better approximations may be determined through analysis and
modelling of the specific details of the flow pattern in the ozone
0.2-10
-------
contactor, but such noddling cannot be done based on tracer studies
alone, as the SFA can. Comprehensive descriptions of the SFA can be found
In several references Including Levensplel (1972) and Seinfeld and Lipldus
(1984). The SFA assumes that the Inactlvatlon In a contactor can be
determined by the product of the probabilities of two events: the
probability distribution for water to renaln 1n the contactor; and the
probability distribution for organisms to survive as they pass through the
contactor.
The first probability function describes the chances of a microor-
ganism remaining 1n the contactor for a specified time period. The water
passing through the contactor has a probability distribution, determined
by tracer studies which Indicate the detention time for each fraction of
the flow through the contactor.
The second probability function describes the chances of a microbio-
logical species surviving following exposure to a disinfectant for a
certain amount of time. This probability function Is given by the
modified Chick's equation: (I/I0) %10"Kt. Each fraction of the flow would
have a different "t" for which this equation would apply. For example, a
virus that 1s exposed for 1 minute to C>1 mg/l ozone when k-l L/mg-m1n has
0.1 (10 percent) chances to survive.
The following illustrates the intuitive origin of the SFA approach:
The flow in an imaginary contactor may be viewed by flow lines.
A microorganism that 1s Introduced at time t"0 will follow one
of these flow lines.
For simplicity, consider that only four flow lines exist as
represented on Figure 0-7.
A microorganism that 1s Introduced in the feed to the contactor
has some probability (PI) of following any one of these four
lines.
The microorganism will then remain for a specific detention
time, characteristic of each flow line, In the contactor.
This concept is presented schematically on Figure 0-7, where
the flow lines are represented by four different tubes whose
lengths (or detention times) correspond to the lengths of the
flow lines on Figure 0-7.
0.2-11
-------
- Microorganisms that arc Introduced Into various tubas have
different probabilities of survival (P2), because of different
susceptibilities to disinfection in each of these tubes.-
- The product of the probability that a Microorganism will be
carried Into a specific tube (PI) times the probability of
survival after being exposed to the hostile environment for the
appropriate time (P2) Is the probability that a microorganism
Introduced Into the feed Inlet will get out alive from a
specific tube
-------
A. PLOW LINES IN A CONTACTOR
MPUT
OUTPUT
B. SCHEMATIC REPRESENTATION OF THE PLOW LINES
•£
¦£
T5
C. SURVIVAL PROBABILITY FOR AN ORGANISM
P P P P
Flow Friction of Flow (I/10) Ovlrill
Um Into thg Hpwllnt SurytYil Ritles survwti Ratios
1 2/10 1/4 16/320
2 4/10 1/8 16/320
3 1/10 1/16 2/320
4 3/10 1Z21 2ZIM
SUM 1 37/320
FIGURE 0-7 - PRINCIPLE OF SEGREGATED FLOW ANALYSIS
-------
assure compliance with the virus Inactlvatlon requirements. Specifically,
this 1s true when:
kvlr-Aeyt > 1'°9( 1°9( I/I0)ey.t («)
The SWTR, however, requires a higher level of Inactlvatlon of viruses
than 6lard1i cysts." Therefore, ozone contactors that are characterized by
a high degree of turbulence will find that, as the flow configuration
approaches that of a CSTR (T,2.5-toq
The Haworth Water Treatment Plant, Hackensack, New Jersey, uses a
turbine ozone chamber followed by a contact chamber to provide additional
contact time. A schematic of the contactor 1s shown on Figure 0-1. The
treatment plant provides filtration after the ozone contactor. For the
purposes of this example, although It 1s not the case for Hackensack, the
0.2-13
-------
ozone system must provide disinfection for 2-log 6iar
-------
The HOT of the contactor at the flow rate of the study was 150
seconds.
Thus T10/HDT ¦ 30/150 - 0.2, Is less than 1/3, however, because
the required 1nact1vat1on 1s less than 2.5>log, the T10
evaluation for this system 1s appropriate.
Based on the T10 evaluation, the residual needed to meet the CT
requirement 1s determined as follows:
CT - 0.16 mg-m1n/L
C - 0.16 mo-m1n/L - 0.32 mg/l
0.5 m1n
Thus, according to this approach, the system must provide an
ozone concentration of 0.32 mg/l to meet the 1nact1vat1on
requirements.
Because of the low T10/HDT value for this system, the CSTR
approach 1s an alternative for determining C. This example 1s
presented 1n Section 0.2.8.2.
Example 0.2-6 Low Petentlon Time. Inactlvatlon Required >2.5-1oq
An unflltered water system must provide disinfection for a 4-log
1nact1vat1on of viruses and a 3-1og 1nact1vat1on of Glardla cysts. The
ozone system uses a single chamber turbine contactor for disinfection:
The hydraulic detention time measured at peak flow rate 1s 30
minutes and T10 determined by a tracer study 1s 9 minutes.
The T.» approach Is not recommended for this system because
T.j/HDt of 0.3 1s less than 1/3 and the required level of 4-log
virus 1nact1vat1on 1s higher than the 2.5-1og level.
SFA or the more conservative CSTR calculations may be used to
determine the required ozone concentration for this system.
Examples of the CSTR and SFA calculations are presented 1n
Sections 0.2.8.2 and 0.2.8.3, respectively.
Example 0.2-7 High Detention Time. Inactlvatlon Required <2.5-1oo
The Sturgeon Bay Water Treatment system (Rosenbeck, 1989) uses a
series of two submerged turbine ozone contactors followed by a reactive
chamber to disinfect ground water:
The results of a tracer study conducted on one of the mixed
contactors 1s shown on Figure 0-9.
The T1(, from this study is approximately 30 seconds while ihe
hydraulic detention time Is 62 seconds.
0.2-15
-------
- T.q/HDT ¦ 30/62 - 0.48 which 1s areater than 1/3. Therefore,
the T10 approach 1s appropriate for this system.
In this cast, the SFA method Is not recommended as an alternative to
the T10 approach because of the Minimal detention tines In the contactor.
With such a short period for the collection of samples, the data are
Insufficient for the SFA method. The resolution of the tracer studies,
apparent In Figure 0-9, will lead to an overly conservative estimate of
the 1nact1vat1on if differentiation 1s conducted by a forward algorithm.
o.z.8.2 EvilmHQ"? Mnq.CSTti
The following two examples demonstrate the CSTR approach. One Illus-
trates the benefit of the CSTR analysis over the T,0 analysis. The other
Identifies conditions for which the CSTR approach is not practical.
Example 0.2-8 U* PftCfrtlOn T1ft. Ifligt 1 vit 10" RM^rtd <2,5-193
The system Identified In Example 0.2-5 1s « slow sand filtration
plant, using ozone to provide for a i-log Giirdla cyst Inactivatlon.
Chlorine provides the 2-log virus inactivation. Because the level of
Inactivatlon required from ozone disinfection 1s less than <2.5-log, the
system may choose any method for the determination of the contact time.
A tracer study conducted on the ozone contactor resulted 1n a
T,0 of 30 sec for a"HOT of 150 sec.
The fraction of T10/HDT 1s 0.2, which 1s less than 1/3,
Indicating that the CSTR approach may be appropriate.
Chlorine provides disinfection for the viruses, therefore the
CSTR calculation for the ozone disinfection requirements will
be based on Giardla cyst inactivatlon.
- The following conditions apply:
Mater Temperature ¦ 25 C
CT for 1-1 og Glardia cyst - 0.16 mg-min/L
l
- Equation 5 from Section 0.2.5 applies for the CSTR calculation:
C(HDT) - [1 - (I/Io)]/t(2.303)k (1/1,)]
.0.2-16
-------
too
•0
M
w
8
&
o
«0
i20
3
10
1S
30
2S
39
20
«0
rME (WMMUTES)
FIGURE 0-8- SEGREGATED FLOW ANALYSIS OF AN OZONE CONTACT CHAMBBMRACEfl STUDY
-------
0.9
0.8
0.7
0.6
»
i
—
I
0.4
0.1
170
130
150
30
70
110
90
TIME IN SECONDS
FIGURE 0-9 - TRACER STUDY OF STURGEON BAY OZONE
CONTACTOR, G.L. ROSENBECK, (1989)
-------
The parameters are determined as follows:
1. From Table 0-3, k^,, • 6.25 for CT *0.16 n1n/L it 25c
2. For 1-log 1nact1vat1on, I/I0 *0.1
3. HOT ¦ 150 sec or 2.5 m1n
C 1s determined as follows:
C(HDT) - [0.9]/[(2.303)(6.25) (0.1)] - 0.625 mg-m1n/L
C- 0.625/2.5 - 0.25 mg/L
Thus, according to the CSTR approach, the system must provide an
ozone concentration of 0.25 mg/L to meet the 1nact1vat1on requirements.
For this case, the system would prefer to use the CSTR approach rather
than the T10 approach since the T10 approach would require a 0.32 mg/L
ozone residual, as shown 1n Example 0.2-5.
Example 0.2-9 Low Detention Time. Inact1vat1on Required >2.S-loo
An unflltered water system must provide disinfection for 4-1og
1nact1vat1on of viruses and 3-1og 1nact1vat1on of G1ard1a cysts. The
system uses a single chamber turbine ozone contactor. Hydraulic detention
time measured at peak flow rate 1s 30 minutes and T10 determined by tracer
studies 1s 9 minutes. T10/HOT 1s less than 1/3 and greater than 2.5-1og
1nact1vat1on 1s required, therefore the T10 approach should not be used.
The CSTR or SFA methods are appropriate.
The CSTR calculation must be conducted for both 61ard1a cysts
and viruses to determine the controlling parameter
Compute the C required for 1nact1vat1on of G1ard1a cysts:
k cysts ¦ 6.25 (Table 0-3).
For 3-1og 1nact1vat1on, I/Ic • 0.001
Using the CSTR equation:
C(HOT) - [1-0.001]/[2.303(6.25)(0.001)] • 69.5 mg-mln/L
C ¦ (69.5 mg-m1n/L)/(30 min) - 2.3 mg/L
0.2-17
-------
• Compute the required C for Inactlvatlon of viruses:
* kyiru* ¦ 13-3 (Tible 0-3)
For 4-log Inactlvatlon I/I0 ¦ 0.001
Applying the CSTR Equation:
C(HDT) • [1 - 0.0001]/[(2.303) (6.25) (0.0001)] - 326 mg-m1n/L
C - (326 mg-n1n/L)/(30 n1n) - 10.8 ®g/L
As Indicated, virus Inactlvatlon Is the controlling parameter,
requiring a C of 10.8 mg/L. Because of the higher ozone residual needed
for the virus 1nact1vat1on, this example Illustrates why systems should
verify compliance with the Inactlvatlon requirements for viruses as well
as for the Inactlvatlon requirements for Glardla cysts. Since obtaining
an ozone residual of 10.8 mg/L 1s unrealistic, this example Illustrates
how stringent disinfection conditions can become assuming CSTR character-
istics. Consequently, the SFA would result 1n a more feasible residual
requirement for this system.
0.2.8.3 Evaluations Using SFA
The SFA method can be conducted on spread sheets. Table 0-4 presents
the calculation procedure 1n spread sheet notations for a step tracer
Input:
The first column of Table 0-4 represents the sequential
numbering of consecutive tracer study measurements or digital
measurement points fed Into the computer.
The second column represents the time Interval that elapsed
between the step change In tracer concentration and the
sampling of the specific tracer point.
The third column represents the tracer effluent concentration
at' a point 1n time determined by the analyzer (spectropho-
tometer conductivity meter, etc.) reading.
The fourth column represents the tracer response on a scale of
0-1, where 0 corresponds to background reading of the analyzer
and 1 to ultimate response after a long time Interval. In
other words, 1t Is C^/C,,, where C^ 1s the tracer concentra-
tion in the outlet of the contactor and C,n 1s the baseline
tracer concentration 1n the Inlet.
0.2-18
-------
1(1
liwc height Llii
ti h| f ("h,/h(
t, h, Fj-h^h,
Mil
ICll
IClt
ICll
ICll
llln
| I/I, - SIM ([s 6 tj
TRACER CURVE FROM STEP INPUT
»•
Ca/CI
•i
mi
T« TS
Et « f« rs
i» ii
tm
TS ft
f MINUTES
MOTE: GUIDANCE FOR CONDUCTING TRACER
STUDIES IS GIVEN IN APPENDIX C
TABLE 0-4 -SPREAD SHEET NOTATIONS OF SEGREGATED FLOW
ANALYSIS FOR A STEP TRACER
-------
The fifth column represents the forward derivative of the
F(t) response. It 1s the slope of the tracer curve at a
specific tine Interval, or the rate at which C-^/C,. changes
with respect to time at different intervals In tine. Note that
by forward evaluation of the derivative: E(t) - [F(tfdt)-
F(t)]/dt the E(t) curve 1s shifted by half a dt toward the
origin.
This method of differentiation Introduces an Inherent safety'
margin to the calculation. Systems can reduce this safety
margin by collecting more tracer points at the Initial period
of the tracer response, when the response 1s starting to
Increase.
This period has the largest effect on the accuracy of the
tracer analysis because most of the contribution to the total
survival of microorganisms comes from the organisms that remain
only for short time Interval 1n the contactor.
The sixth column represents Chick's 1nact1vat1on rule, computed
at the concentration and the appropriate 10'kCt.
The seventh column represents the survival expectancy function
(Es(t) ¦ E(t)(10 ct) which Is the product of columns 5 and 6.
The eighth column represents the organism survival 1n each
segment passing through the contactor. It Is also known as the
Integral of the survival expectancy function (Es presented 1n
the 7th column).
The survival ratio (I/I0) Is 'the sum of column 8. This
represents the sum of organism survival In all the water
segments passing through the contactor.
Table 0*4 Illustrates only one form of performing the Integra- .
tlon (I.e., quadratic Integration). Other Integration methods
can also be used.
The corresponding log 1nact1vat1on and the corresponding
calculated CT may be computed by the procedures outlined 1n
Section 0.2.4.
The following examples Illustrate the use of the SFA method to
calculate conditions In ozone contactors, and a situation where SFA cannot
be used.
Example 0.2-10 Turbine Contactor
As noted In Example 0.2-4, the ozone system at Haworth Water
Treatment Plant, uses a turbine ozone chamber followed by a reactive
0.2-19
-------
chamber to provide additional contact time. A tricer study was conducted
on one of the contactors resulting 1n a T,0 value of 11 minutes for a HOT
of 20 minutes. Using the same conditions as the above cited example, the
SFA will be conducted on the tracer data. The following illustrates a
step by step procedure for conducting a SFA:
The digitized tracer response 1s depicted in Figure 0-10
as a function of t(1) where:
1 stands for the consecutive numbering of randomly chosen
points from the tracer study chart, and
t(1) Is the corresponding time coordinate.
The slope of the tracer curve, also known as the density of the
expectancy function, E(t) approximated by the following
equation is depicted In Figure 0-10.
E(1) - [F(1+l)-F(1)]/[t(i+l)-t(i)]
The digitized points were not translated into a smooth curve
in order to avoid numeric compromises.
The survival expectancy (Es(t)) was then calculated by Es(i)«Et
(i)(10'*cio>) and summed to give the survival ratio (I/I0) as
shown in Table 0-5.
Figure 0-11 depicts the integration for conditions where the
ozone residual is C - 0.15 mg/L.
The cumulative survival ratio is 0.00982 which 1s below 0.01
assuring compliance with the 2-log or 99 percent Inactivation
requirement for Siardia cysts. A survival ratio of <0.01
corresponds to an inactivation of greater than 99 percent or
2-log.
The residual value determined from this method Is lower than C-0.17
mg/L predicted by the T10 approach presented in Example 0.2-4. Although
this example only shows a small difference in C values needed, other cases
may result in a greater reduction of C compared to the C resulting from
the T10 approach.
0.2-20
-------
SEGREGATED FLOW ANALYSIS
DIGITIZED TRACER RESPONSE, F(l», E(l)
lima (minute
miSE0-1(1. SEGIIBlATmRiW AMAUWO^^NECCmCT
CHAMBER-fJTEGRAT^~^ Of SUHVTOM. HfCOICY
-------
TABLE 0-5
Segregated Flow Analysis
of an Ozone Disinfection Contactor at Hackensack
10"kct
time
(min)
height
(mm)
F(t)
E(t)
(C-0.16
k-1.03)
Es-
¦E(t)10
(EsUt
0
0.0
0.000
0.000
1.000
0.00000
0.00000
1
0.0
0.000
0.000
0.708
0.00000
0.00000
2
0.0
0.000
0.000
0.502
0.00000
0.00000
3
0.0
0.000
0.000
0.355
0.00000
0.00000
4
0.0
0.000
0.000
0.252
0.00000
0.00000
5
0.0
0.000
0.002
0.178
0.00043
0.00013
6
0.3
0.002
0.014
0.126
0.00173
0.00035
6.5
0.5
0.004
0.013
0.106
0.00143'
0.00215
7
2.0
0.016
0.008
0.089
0.00072
0.00072
8
3.0
0.024
0.016
0.063
0.00102
0.00204
9
5.0
0.040
0.024
0.045
0.00108
0.00324
10
8.0
0.065
0.032
0.032
0.00102
0.00408
11
12.0
0.097
0.056
0.022
0.00127
0.00889
12
19.0
0.153
0.056
0.016
0.00090
0.00630
13
26.0
0.210
0.040
0.011
0.00045
0.00225
14
31.0
0.250
0.040
0.008
0.00032
0.00160
15
36.0
0.290
0.048
0.006
0.00029
0.00174
16
42.0
0.339
0.073
0.004
0.00027
0.00243
17
51.0
0.411
0.065
0.003
0.00018
0.00144
18
59.0
0.476
0.081
0.002
0.00016
0.00160
19
69.0
0.556
0.044
0.001
0.00006
0.00066
21
80.0
0.645
0.040
0.001
0.00003
0.00030
23
90.0
0.726
0.016
0.000
0.00001
0.00004
25
94.0
0.758
0.016
0.000
0.00000
0.00000
27
98.0
0.790
0.017
0.000
0.00000
0.00000
34
113.0
0.911
0.004
0.000
0.00000
0.00000
36
114.0
0.919
0.016
0.000
0.00000
0.00000
41
124.0
1.000
0.000
0.000
0.00000
0.00000
45
124.0
1.0000
0.000
0.000
0.00000 .
0.00000
£(Ea)4t + - I/I0 - 0.00982
-------
SEGREGATED FLOW ANALYSIS
SURVIVAL OF CYSTS
-------
0.2.9 ESTIMATING T
The results of this section a re sumarized in Figure 0-12. The
decision tree shows the applicable methods of estimating T for each
approach, and provides a quick Beans to compare alternatives and sake a
selection.
0.2-21
-------
ESTIMATING 'T
DETENTION
NO
DATA
AVAILABLE
CALCULATE la
CALCULATE la
YES
LOG
T»
OR
NO
YES
CONOUCT
sec. aow
ANALYSIS
CONOUCT
CSTR
CALCULATIONS
FIGURE 0-f2 - DECISION TREE FOR ESTIMATING T
h
-------
0.3 DETERMINATION OF OZONE CONCENTRATION (C)
0.3.1.inlroductlon
This section presents ways to Measure or estimate the- ozone
concentration, C, for the calculation of CT. An alternative, more
elaborate concept, requiring better characterization of the hydrodynamics
of the ozone contactor is presented In Section 0.4 of this appendix.
EPA recommends use of the average dissolved ozone concentration in
the water for C for all types of ozone contactors. The average concentra-
tion may be determined by one of following Methods:
1. Direct measurement of the concentration profile of dissolved
ozone 1n each contact chamber
2. Indirect prediction of the average concentration by assuming
a set of conservative correlations between an observed variable
such as the concentration of ozone 1n the outlet from the ozone
chamber and the average concentration within the ozone chamber.
The application of these methods to estimate the average concentra-
tion should take into account the gas/11quid flow configuration 1n the
ozone contactor. The next section presents a short discussion of the
types of I1qu1d/gas contact In ozone chambers, followed by two sections
that describe the methods to estimate the average concentration in the
chamber based on simple measurements.
Classification of Ozone Chambers
Ozone contactors currently 1n use or 1n design stage 1n the US may
be classified into four types of flow configurations as Illustrated on
Figure 0-13. This, of course, does not preclude the use of other types of
contactors. The four configurations are as follows:
1. Continuously St1rred-Tank Reactor (CSTR):
Ozone contactors using turbine agitators, where the water may
be considered uniformly mixed as shown on Figure 0-13, diagram
1. Studies conducted 1n a full scale turbine contact chamber
Indicate that turbine contactors may be considered uniformly
0.3-1
-------
nixed (Schwartz et, al%, 1990)s This study was conducted 1n the
first contact chamber under conditions of high ozone demand.
Therefore, It Is assumed that under less stringent kinetic
conditions, turbine contactors can still be considered uniform-
ly mixed. •
2. Counter-Current Flow Chambers
In these chambers, the water flows opposite the direction of
the gas bubbles. For example, the first and third chambers 1n
the Los-Angeles ozone treatment system, as shown on Figure 0-2.
3. Co-Current Flow Chambers
In these chambers, the gas bubbles and the water flow 1n the
same direction. For example, the Deep ll-Tube contactor shown
1n Figure 0-5 and the Static Mixer contactor. This 1s the case
also for the conventional gas/11qu1d contact chambers such as
the second contact chamber In the configuration designed for
the East Bay MUD water disinfection system, as shown on Figure
0-4„
4. Reactive Flow Chambers %
In these chambers, no gas (and ozone) 1s being Introduced into
the chamber or conduit. The second and fourth Chambers of the
Los Angeles water disinfection system are reactive chambers
(Figure 0-2).
0.3.2 Direct Measurement of C
Direct measurement of the dissolved ozone concentration 1s the
preferred method to determine the ozone concentration 1n ozone contact
chambers. However, very little full scale experience 1s currently
available with this type of measurement. Some guidelines were developed
based on the limited studies conducted at the Haworth, NJ (Schwartz et al.
1990) and Los Angeles water treatment systems (Stolarlk and Christie, .
1990). The guidelines developed for direct .measurement of ozone
concentration 1n the liquid phase are detailed In the following sections.
' Analyze Each Chamber Separately
Every chamber of a multiple-chamber unit should be analyzed separate-
ly. Different chambers 1n series exhibit different ozone consumption rates
and reactivities and, therefore, are likely to have different dissolved
ozone profiles.
0.3-2
-------
t TURBINE CHAMBER
2. COUNTER-CURRENT
CHAMBER
3 CO-CURRENT CHAMBER
4. REACTIVE FLOW
CHAMBER
L: Liquid
Q: Gas
FIGURE 0 13- FLOW CONFIGURATIONS IN OZONE CONTACTOR CHAMBERS
-------
Avoid Inttrftrtrrce frow fiis Bubbles
Gas bubbles nay strongly Interfere with the Measurement of ozone
concentration, particularly If some bubbles are carried Into the sampling
taps. This Interference may be reduced by directing the sampling port
opposite to the direction of the bubble flow 1n order to prevent gas from
entering the sampling tube. Additionally, the operator should verify, by
visual Inspection, that the sample water does not contain gas bubbles.
Systems using In-sltu ozone analyzers should be careful to prevent
direct contact of gas bubbles with the measuring probe which 1s usually a
gas permeable membrane. Such contact may bias the measurements and give
high results.
Minimize Distance to Qzo"e Analyzers
Minimize the distance from the sampling ports to the ozone analyzer
to limit ozone consumption by reducing agents 1n the water. This
consideration 1s particularly Important when evaluating the concentration
profile 1n chambers with high ozone demand such as the first chamber 1n
multiple-chamber units.
Provide Proper Soaclal Distribution
The vertical profile of the ozone concentration 1n ozone contact
chambers should be measured 1n at least five vertical locations and at
least two different horizontal locations for each vertical sampling point
within the contact chamber. Each sample should represent the time
averaged concentration at the specific location. This may be achieved by
sampling a large volume of water Into a container and analyzing the water
by the 1nd1go tr1sulfonate method (Bader and Holgne, 1982). In-situ
measurement of ozone should be carried out over a sufficient time Interval
to suppress temporal fluctuations. Such Instruments should be Initially
calibrated by the 1nd1go tr1sulfonate method. Facilities that have more
than 25 percent deviation between the average concentration at two
horizontal locations should collect additional measurements at a third
location. The average of all measurements may be taken as the average
concentration of dissolved ozone In the ozone contact chamber. For
systems with a symmetrical vertical distribution of ozone concentration
0.3-3
-------
the vertical. sampling points should be tquldlstant. Systems with an
asymmetrical distribution of available sampling points can perform an
Integration of the data to estimate the average concentration In the
chamber. An example of this 1s given at the end of this section.-
Some contact chambers, such as the Oeep U-Tube chambers, static
mixers and reactive flow chambers have a high length to width ratio, where
the length of the chamber In the direction of fluid flow 1s greater than
four times the cross section length. These chambers have more uniform
radial concentration profiles, eliminating the need to measure the
concentration at various vertical or horizontal locations. Therefore,
measuring the concentration profile at several points along the flow path
should be sufficient to accurately determine the average concentration.
4
Select Representative Locations
All sampling positions should be placed 1n representative locations,
avoiding stagnant zones and zones near the wall. Measurements 1n stagnant
locations will lead to low values of the average residual concentrations.
While measurements at the wall may result 1n either an underestimate or
overestimate of the residual depending on the ozone flow pattern.
Systems having two or more identical parallel ozone contact chambers
may determine the average ozone concentration by measuring the concentra-
tion profile at one horizontal location in each contact chamber. These
systems should, however, show by dual or triple horizontal measurements in
at least one of the parallel chambers that the measurement 1n the
particular horizontal location adequately represents the concentration
profile In the contact chamber.
Example 0.3-1
A system with a co-current chamber with dimensions of 10' X 10' x 20'
was sampled to determine the average concentration 1n the chamber, in
accordance with the recommended guidelines, the following samples were
taken:
0.3-4
-------
Mater
Ozone Residual (mg/L)
P«°th 'ft)
2
0.1 Xl2
6
10
14
18
0.15 0.17
0.1S 0.1#
0.3 0.21
0.6 0.65
The horizontal sampling point measurements are within 25
percent of each other Indicating that no additional horizontal
sampling Is needed. Figure 0-14a shows the sampling locations
and the resulting ozone profile.
Average the H, and H, sampling points to determine CM:
Cayfl - (0.1 + 0.15 +1>.15 ~ 0.3 + 0.6 ~ 0.12 ~ 0.17 A.14 *
0.Z5 + 0.65)/10 - 0.26. C#vf' equals 0.26 mg/L, which is C for
the chamber._
Eximplg
A system with a co-current chamber and the same dimensions of the
system in Example 0.3-1 has sampling results as follows:
m Average • (H, + H2)/2
The sampling points are not vertically equidistant so the
system will plot the average ozone concentration of the
horizontal sampling points versus depth to calculate the area
under the curve. This approach should only be used If the
sampling points cover the range of the water depth.
As shown on Figure 0-14b, the area under the curve is deter-
mined for the range of depths sampled from 2 to 18 ft. .
Several methods can be used for calculating the area includ-.
Ing:
2
8
14
,16
18
Water
0.16 0.14 0.15
0.27 0.3 0,285
0.70 0.73 0.715
0.62 0.51 0.615
Measurement with a planimeter
Mathematical methods such as:
Simson's Rule
Runge Kutta
0.3-5
-------
SAMPLING LOCATIONS
OZONE PROFILE
Mi
o
m
ui
5
>
0 J 4 • •
EQUIDISTANT SAMPLING
]• It
HI
O
<
*
10
15
99
0 0.2 9.9 1.9
OZONE CONCENTRATION
SAMPLING LOCATIONS
OZONE PROFILE
Q
K
HI
5
*
19
IS
29
H. „
i
~ i
1
i
i
T
1
1
i
\
1
1 ill
0 2 « ¦
to. SKEWED SAMPLING
HI
o
<
*
10 ft
«•
AREA FOR •
CALCULAf MQ C •*«
0.2 0.» 1.0 (wf/ll
OZONE CONCENTRATION
FIGURE 044-DIRECT MEASUREMENTS FOR DETERMINING C
-------
- The area under the curve Is 1n units of mg/l-ft. C 1s deter-
mined is:
irei fma/L-fU
range of depth sampled (ftJ
For this data, use of a planlmeter results 1n an area of 5.44
mg/l-ft, with the concentration determined as follows:
5.44 ma/l ¦ ft - 0.34 mg/L
18 ft - 2 ft
0.3.3 Estimating C Based on Residual Measurements at tht Outlet
For many systems, measuring ozone profiles 1ri their ozone chamber may
be impractical because of physical constraints. These systems may
estimate C in the chamber based on measurements of the ozone residual at
the outlet from the chamber. EPA has established correlations for
different types of gas-liquid contact configurations currently in use in
ozone contactors. These relationships were derived based on conservative
assumptions regarding the type of flow configuration in the contactor.
Due to the highly reactive nature of ozone the values for C vary slightly
between first chambers and subsequent chambers. The recommended
concentrations for first and subsequent chambers are summarized in Table
0-6.
0.3.3.1 First Chambers
, A first chamber 1s the chamber in which ozone is initially intro-
duced. In establishing guidelines for determining C values for the first
ozone contact chamber, the following Items were considered:
1. The relationship between C and the outlet concentration in the
first chamber of a multiple-chamber system (or single chamber).
may be very sensitive to the reaction order of the ozone
consumption kinetics.
The average concentration in the contactor may be less than 10
percent of the outlet concentration. This was demonstrated in
pilot plant studies conducted in a multiple chamber system by
Stolarik and Christie, 1990. Therefore, general relationships
. between the residual ozone concentration at the outlet from a
first (or single) ozone contact chamber and the average
concentration in this chamber cannot be developed.
0.3-5
-------
TABLE 0-6
CORRELATIONS TO PREDICT C BASED
ON OUTLET OZONE CONCENTRATIONS"'1*
FLOW CQNH6URATIQN
CO-CURRENT COUNTER-CURRENT REACTIVE
TURBINE FLOW FLOW FLOW
)er
Subsequent Chambers
PARTIAL"'
CREDIT
'out
or
'out
PARTIAL®2*
CREDIT
Cout/2
C - (C„ + CJ/2
im*
1. Definitions:
C Characteristic Concentration (mg/L)
NOT
APPLICABLE
"out
'out
Dissolved ozone concentration at the outlet from the chamber (mg/L)
Cjn Concentration of ozone at the Inlet to the chamber (mg/L)
2. 1-log of virus 1nact1vat1on providing that C^, > 0.1 mg/L and 1/2-1og Glardia
cysts 1nact1vat1on providing that > 0.3 og/L.
3. Alternatively, C may equal the average concentration as evaluated by the direct
measurement method (Section 0.3.2).
-------
2. The rate of disinfection of viruses (collphage) by ozone often
decreases with respect to contact time whereby the Initial
Inactlvatlon rate Is very fast and deteriorates afterwards.
3. Pilot plant experiments reported by Wolfe et al, (1989)
suggest that the Inactlvatlon of organisms Including HS2
bacteriophages, Glardla our1s cysts, R2A bacteria and E. Col 11
In the first chamber of a multiple-chamber reactor Is very
rapid even when high ozone demand waters are used.
Considering these Items, EPA recommends a general guideline of
crediting the first ozone chamber with CT credits equivalent to 1-log
virus Inactlvatlon and 0.5-log 61ardia cyst Inactlvatlon, provided that
the residual concentration measured at the outlet from the first contact
chamber exceeds 0.1 mg/L and 0.3 mg/L, respectively, regardless of the
contactor configuration. However, this guideline does assume that the
volume of the first chamber 1s€ equal to the volume of subsequent chambers.
The credit for 1-log virus Inactlvatlon at an outlet residual of 0.1 mg/L
may appear conservative with respect to HS2 bacteriophage data, however,
only limited data for ozone Inactlvatlon of the animal viruses of concern
1s currently available. Preliminary test results Indicate that bacterio-
phage may not be an appropriate Indicator for virus Inactlvatlon by ozone
(Finch, 1990).
Systems may prove higher performance of their first contact chambers
by measuring the concentration profiles In the first chamber, as outlined
1n Section 0.3.2 or by applying the more sophisticated methods that are
presented 1n Section 0.4.
0.3.3.2 Subsequent Chambers
The correlations 1n Table 0-6 are based on analysis of the dissolved
concentration profile In liquid/gas contacting chambers. All correlations
rely on the accurate measurement of ozone concentration outside of the
gas/11 quid contacting regime. Concentrations at the outlet from the ozone
contact chambers can be measured accurately without Interferences from the
ozone bubbles. The correlations represent the highest possible estimate
of C that can be supported without s1te-spec1f1c test data. These
estimates are conservative and systems may choose to determine C based on
0.3-7
-------
direct measurement of the concentration profile 1n the contact chamber, or
use one of the procedures recommended In Section 0.4.
Correlations were developed for the four types of flow configura-
tions:
Turbine
Counter Current Flow
Co-current Flow
Reactive Flow
Turbine
For turbine chambers or rigorously mixed chambers, the flow
characteristics 1n the chamber approach that of a CSTR and, therefore, the
concentration at the outlet from the contactor (CMt) 1s assumed to be
representative of the dissolved concentration of ozone 1n the liquid phase
(C). Currently, contactors using turbine agitators appear to approximate
CSTR characteristics (Schwartz et al, 1990). Other systems with T10/HDT
values less than 0.33 may use the same correlations. This correlation 1s
applicable to every chamber, Including turbine contactors used for first
chambers or as a single chamber contactor.
The measurement of ozone concentration 1n the gas phase 1s a possible
alternative for determining C although such correlations will be highly
site specific. A procedure to develop site specific correlations between
the average ozone concentration and the off-gas concentration 1s presented
1n Section 0.4.2.1.
Counter-Current Flow
In counter-current flow, the water flows opposite to the direction
of bubble rise. Measurement of the concentration profile 1n such systems-
revealed that the concentration 1n the liquid phase uniformly Increased
with depth 1n the ozone chamber as shown 1n Figure 0-15. The maximum
concentration 1n the chamber 1s achieved near the water outlet from the
ozone chamber.
Measurement of the ozone concentration 1n an Ideal plug flow chamber
reveals that the average concentration 1s only 25 to 50 percent of the
outlet concentration for these chambers under typical operating condi-
tions. Additional contributions to the average concentration that are not
accounted for by the plug flow analysis, Include the contribution of
0.3-8
-------
TOP
a.
w
O
<
S
iOTTOM
INCREASES
OISSOLVEO OZONE RESIDUAL
A. COUNTER-CURRENT FLOW PROFILE
TOP
m
O
a
UJ
<
*
iOTTOM
INCREASES
OISSOLVEO OZONE RESIDUAL
i. CO-CURRENT FLOW PROFILE
FIGURE 0-15 * OZONE CONCENTRATION PROFILES
-------
4
turbulence and the contribution of the Inlet concentration. Based on these
considerations, EPA recommends the use of one-half the outlet concentra-
tion of ozone as an estimate for C.
The measurement of ozone concentration In the off gas 1s a possible
alternative for determining the average ozone concentration although the
correlations will be highly site specific. A procedure to develop site
specific correlations between the average ozone concentration and the off-
gas concentration Is presented 1n Section 0.4.2.1.
Co-Current flow
In co-current flow, both the water and gas flow 1n the same
direction. The ozone concentration profile In co-current operation
Increases until It reaches a maximum and then decreases along the contact
chamber as shown on Figure 0-15. The dissolved ozone concentration
Increases at the beginning of the column due to dominant mass transfer
from the ozone rich bubbles. Then the gas phase becomes depleted of ozone
and the Impact of ozone consumption In the liquid phase dominates the
ozone profile. C can be estimated as the concentration of dissolved ozone
at the outlet or by the average of the Inlet and outlet concentrations of
dissolved ozone, whichever 1s higher. This estimate should still be
conservative, particularly for systems exhibiting high transfer efficien-
cies.
The measurement of ozone concentration 1n the off gas 1s a possible
alternative for determining the average ozone concentration although the
correlations will be highly site specific. A procedure to develop site
specific correlations between the average ozone concentration and the off-
gas concentration 1s presented 1n Section 0.4.2.1.
Reactive Flow
In ozone chambers operated 1n a reactive flow configuration, the
water contains dissolved ozone residual from previous chambers but no
additional ozone 1s being Introduced. Reactive flow chambers are used:
for other disinfectants, such as chlorine, chlorine dioxide and chloramin-
es; for the decay of ozone following a contactor or a static mixer; and
for combining ozone with hydrogen peroxide.
0.3-9
-------
For static mixers, the Mixer acts as a turbine chamber with the
pipeline following the mixer acting as the reactive chamber. The pipeline
Is In effect the second chamber and the guidelines 1n Table 0-6 apply for
the determination of C. " The contact time 1n the pipeline can be
calculated by assuming plug flow.
In order to be consistent with the recommendations for monitoring
other disinfectants In reactive flow chambers, and.in order to assure
compliance under worst case conditions, the use of the residual outlet
from the chamber (C^) is recommended as a conservative measure of C. The
CT for reactive flow chambers may be estimated by dividing the chamber
Into subunlts, measuring the concentration at the end of each subunlt, and
adding the CT credits.
Estimates of C based on the outlet concentration were conservatively
developed based on available test data. EPA's recommended values for C
are summarized 1n Table 0-6. A system may choose to perform additional
testing for direct measurement of ozone residuals to support a higher
value, 1f appropriate. In addition, a reactive flow chamber may be
subdivided Into smaller units with ozone measurements at the end of each
unit to Improve CT credit.
0.3.4 Estimating C
The results of this section are summarized 1n Figure 0-16. The
decision tree shows the applicable methods of estimating C for each flow
configuration, and provides a quick means to compare alternatives and make
a selection.
0.3-10
-------
ESTIMATING 'C
COUNTER
CURRENT
FLO*
CO-CURRENT
ROW
TURBINE
CHAMBER
Cmn OR Cm*
C - 0
nil
Un
C mm (V*
1 ST
CHAMBER
t
t ST
CHAMBER
f
C an OR
C i» * Coal
jH
Cm* i;i
«» Cm, m
REACTIVE
FLOW
CHAMBER
C-Cw
a*
OR Cmm
NOTES: 1. CREDIT FOR l-log VIRUS WHEN C« > 0,1 mg/L
ANO 0.5 -tog QABOA INACT1VAT1QN
WHEN CM > 0.3mgA FOR FIRST CHAMBERS.
2. DETERMINATION OF Cavo IN SECTION 0.3.2 AND Youi/H IN SECTION 0.4.2
3. FOR FIRST OR SUBSEQUENT CHAMBERS.
FIGURE 0-16- DECISION THEE FOR ESTIMATING C
-------
0.4 SITE-SPECIFIC EVALUATION OF OZONE CONTACTORS
0.4.1 Introduction
The second set of guidelines Is designed to prevent systems from
costly over-design and use of overdoses of ozone, by performing site
specific characterization of their ozone contactors. This approach was
partially utilized 1n the previous two sections by recommending a direct
measurement of the ozone concentration profile and by allowing systems to
use the SFA or CSTR approaches. In this section the site specific
evaluation procedure will be further developed by presenting additional
options to Improve disinfection credits or simplify monitoring procedures.
EPA recommends the following three alternatives for site specific
evaluations:
Estimating C by measurement of another variable
Modeling performance of field scale operation
Use of microbial Indicator studies
C may be estimated by measuring an easily monitored (observable)
variable. Systems should develop site specific correlations between C and
another observable parameter such as the gas or liquid concentration
exiting (C^) the chamber and monitor this observable parameter Instead
of C. Guidelines to develop such site specific correlations are presented
1n Section 0.4.2
Modelling the performance of full scale operations 1s an alternative
to the separate C and T approach. The first procedure separated the
analysis Into two separate Issues related to determining C and T.
Extensive modelling of the system may predict higher 1nact1vat1on levels,
even for the same C and T. EPA recommends that systems construct
mathematical models of their ozone contactors to predict the disinfection
performance, provided that the models are confirmed by experimental
observation of the actual ozone concentration profile 1n the contact
chambers, as discussed in Section 0.4.3.
Microbial Indicator studies may be used to determine the 1nact1vation
of viruses and Glardia cysts 1n ozone contactors. EPA recommends that
systems be allowed to evaluate the performance of their disinfection
systems by spiking a pilot of the contactor with an indicator microorgan-
ism and predicting the actual inactivation of Glardia cysts and viruses
0.4-1
-------
based on the 1nactivat1on of the Indicator microorganisms. Guidelines to
conduct such pilot scale performance evaluations are presented In Section
0.4.4.
0.4.2.Site Specific Correlation of C with an ObstrvahU
Section 0.3 recommends determining the concentration of ozone In
contactors by one of the following ways: . .
1. Measure the concentration profile 1n the chambers and determine
the average dissolved ozone concentration for C.
2. Measure the dissolved concentration of ozone 1n the water
outlet from each chamber (C ) and estimate C by the correla-
tions presented 1n Tables 0-6.
This section presents an alternative method to determine C.
The SWTR requires unflltered systems to report a dally CT for their
disinfection systems. Similar requirements may be specified by the
Primacy Agency for filtered systems. Measuring the concentration 1n the
ozone chambers each day may be difficult. Determining the ozone
concentration 1n a chamber by continuous or dally measurements of other
variables is probably preferable. Likewise, many systems may prefer to
monitor the ozone concentration 1n the off gas (YM) or via the applied
ozone dose rather than monitor CM. However, based on available data, a
non-site specific correlation between the average ozone concentration in
the chamber and an observable variable other than could not be
developed.
EPA encourages systems to develop such site specific correlations and
use them instead of the general procedures. These correlations may be
developed in one of the following ways:
1. Determine site specific correlations between and another
variable that can be easily monitored. Measure the variable,
estimate Z f and then use the correlations presented 1n Tables
0-6 to predict C.
2. Determine site specific correlations directly between C and
another variable such as the ozone concentration 1n the off gas
(Yout) or ^out- Measure that variable and estimate C.
0.4-2
-------
Correlations between C or C^, and a measurable parameter nay vary 1n
complexity from a simple empirical linear correlation to a highly
sophisticated mathematical model accounting for the ozone concentration
profile 1n the contact chamber. Development of appropriate correlations
depends on the engineering capabilities of the utility. Therefore, EPA
does not recommend any particular mathematical relationships. However,
the following sections present guidelines to assist systems Indeveloping
appropriate correlations.
Correlations for Specific Chambers
The correlations should refer to a specific contact chamber and
should be verified to fit the performance of this chamber. For example, a
correlation for the first chamber should hot be used to predict C 1n the
second chamber of a multiple-chamber system.
Developing the Correlation
When fitting the correlation with experimental data, a record of the
following variables should be kept:
a. Water flow rate
b. Gas flow rate
c. Ozone concentration 1n the gas feed
d. Ozone transfer efficiency
d. Water temperature and pH
e. Concentrations of all major-Inorganic reducing agents,
1f they constitute a substantial, proportion of the total
ozone demand, such as Iron(II) and manganese, TOC,
alkalinity and turbidity.
f. Cpyt or whatever 1s being correlated
g. The measurable variables such as ozone dosage or CM
The system should also record the dependent (C or C^) and Indepen-
dent measurable variables.
0.4-3
-------
ftppHcitlM ,of the Corrtl«t1on
The correlation should b« evaluated with at least i 90 percent '
confidence level. Since confidence margins are very sensitive to the
number of observations used to develop the relationship, this requirement
will prevent the use of correlations that are based on a United amount of
observations. On the other hand, because systems usually make dally
records of most of the parameters needed to develop a correlation, the
number of observations will usually be very high, thereby, providing a
high confidence level for the correlation. Simple procedures to determine
confidence Intervals are presented 1n statistical textbooks.
The correlation must be checked periodically, such as monthly, as an
additional precaution against unexpected shifts 1n water conditions.
The correlation should be applied only to conditions that are within
the parametric range for which the correlation was developed, as noted 1n
the second guideline. Interpolation 1s permitted but extrapolation 1s
not. Correlations developed during the winter t1mt should not be used to •
evaluate performance 1n the summer.
EPA believes that by permitting such correlations, systems will be
encouraged to apply sophisticated mathematical models 1n order to decrease
the confidence interval and administer smaller doses of ozone. EPA also
expects that systems will develop correlations between C in the contactors
and measurable parameters to simplify their operations. Small or lesser
equipped systems will then be able to use these relationships to estimate
the performance of their ozone contactors. EPA intends to follow advances
in this field and Issue updated examples and guidelines regarding the
selection of efficient site specific correlations.
0.4.2.1 Utilizing Off-Sas Measurements
In ozone contactors, the gas and liquid streams equilibrate when the
contact between the gas and liquid Is Intimate enough and for sufficient
time, otherwise the concentration 1n the water phase will be much lower
than the equilibrium concentration. It can be assumed that close to
equilibrium conditions are reached, when the transfer efficiency In the
contactors is greater than 85 percent ((¥,„-¥«*«)/%n > 0.85). When the
transfer efficiency is greater than 85 percent, systems may use solubility
0.4-4
-------
constant data to calculate from the contactor, based on the ozone
concentration in the off gas. This may lead to a slight over estimate of
the concentration in the liquid phase but this over estimate is justified,
in view of the better reliability of gas phase measurements.
Henry's constants for ozone at various temperatures are presented In
Table 0-7. The residual concentration of ozone may be estimated by:
Where:
Y§ut * The concentration of ozone in the gas phase (ppm -
volume or partial pressure-atm)
(w ¦ The concentration of ozone in the liquid phase
(mg/L)
H - Henry's constant (atm/mg/L)
When applying off-gas modelling, liquid phase measurements must be
made periodically to check the correlation, as the ozone transfer
efficiency has a high impact on the results of this correlation.
Systems must be cautioned against the use of off-gas measurements
for multiple chamber contactors with a common headspace. As noted
previously, modelling must be specific to individual chambers. Thus, if
a contactor 4ias a common head space between chambers, no distinction can
be made as to the concentration in each chamber. Therefore, off-gas
measurements for modelling are recommended for use with single chamber
contactors.
Example 0.4-1
The Metropolitan Water District of Southern California conducted
off-gas monitoring on a single chamber co-current flow pilot contactor to
determine the dissolved ozone concentration:
- Operating conditions were as follows:
source water: Colorado River
feed gas ozone concentration ¦ 2 percent by weight
off gas ozone concentration - 0.185 percent by weight
(or 0.123 percent by volume)
0.4-5
-------
transfer efficiency • 90.8 percent
temperature - 16.5 C
observed ozone residual • 1.04 ng/L
Henry's constant 16.5'C - 0.001179 atm/mg/L
The ozone residual estimated from the off gas concentration
1s:
C-ut-Yout/H ¦ 0.00123/0.001179
» 1.04 mg/l
The measured residual 1s the same as that predicted by the
off-gas measurement Indicating that this approach 1s appropri-
ate for this system.
Example 0.4-2 Empirical Correlation between Cout and Yaut
A system using two counter-current contact chambers 1n series wants
to predict 1n the second chamber by the concentration of ozone In the.
off-gas (Y0Mt). Dally observations of the pertinent parameters during the
first month of operation are presented 1n Table 0-8.
The system chose to correlate (L.t and Y t by linear empirical
correlation.
The dally observations, and the best linear fit are presented
In Figure 0-17.
The 90 percent confidence interval Is presented by the lower
line 1n Figure 0-17.
The system may use the 90 percent confidence level line to
estimate C^,. based on measurements of
For example when - 0.4 percent then the system may use C^,
- 0.36 mg/L.
- Although the best estimate 1s - 0.4 mg/l, the system
should predict - 0.36 mg/L.
Now, according to.Table 0-6, the system may predict C using
the recommended guideline of C ¦ C^/2 - (0.36)/2 ¦ 0.18 mg/L.
The system measures the ozone concentration at the chamber
outlet monthly, to check the model correlation.
0.4-6
-------
TABLE 0-7
HENRY'S CONSTANTS FOR OZONEm
Water
Temperature Henry's Constant Henry's Constant
iWtlQlg friction Utm/ma/1 ozone)
0 1,940 0.00073
5 2,180 0.00082
10 2,480 0.00093
15 2,880 0.00108 •
20 3,760 0.00141
25 4,570 0.00171
30 5,980 0.00224
*
NOTE: m EPA, 1986
-------
V
TABLE 0-8
Empirical Correlation
Between and
U
£«ut
COO
Tmid. 0|
0.5
0.5
2.0
20
0.47
0.43
2.8
15
0.38
0.41
2.5
17
0.39
0.4
2.3
18
0.28
0.32
2.4
18
0.2
0.17
2.6
20
0 25
0.23
2.0
20
0.32
0.27
2.Q
21
0.29
0.27
2.0
18
0.2
0.18
2.0
17
0.22
0.2
1.9
18
0.30
0.33
1.8
20
0.32
0.34
1.9
17
0.28
0.27
1.9 . •
18
0.29
0.32
2.5
18
0.4
0.42
2.4
19
0.47
0.45
2.3
19
0.35
0.37
2.4
21
0.30
0.29
1.9
19
0.20
0.17
1.9
19
0.15
1.19
2.0
19
0.12
0.20
1.9
17
0.17
0.17
1.9
19
0.14
0.16
2.0
19
0.13
0.12
1.9
18
0.25
0.27
1.9
17
0.29
0.32
1.9
18
0.30
0.29
1.8
17
0.22
0.20
1.9
17
0.22
0.20
1.9
18
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.9
1.9
2.0
2.1
2.0
2.0
1.8
. 1.9
1.9
1.8
1.9
1.8
1.8
2.0
2.0
1.9
2.0
2.0
2.0
2.1
2.0
2.0
1.9
-------
»).!»
:ur>
%
0.4 -
0.35 -
0.3 -
0.2
0.13 -
OJ -
II
u
~
n
ri
ci
U
8
(j
u o
Sa
~ ~
n -
ii
' n
~
y
n
O.05
OJ
O.?
—r~
o.i
(ill yds commiti dtinn (fel
"~r~
o.4
FIGURE 0-17- EXAMPLE OF EMPIRICAL CORRELATION OF
RESIDUAL OZONE AND OFF GAS
a
-------
If this system hid the means to monitor the concentration
profile In the contactor and determine C directly it could
develop a correlation between C and Instead of using Table *
0-6.
0.4.3 Modeling the Performance of full Scale Operations
More extensive site specific Mathematical modelling of the actual
performance of the ozone contactor may determine higher 1nact1vat1on
levels than those determined by the separate C and T approach. Therefore,
systems should be allowed to use such advanced modelling, provided that
these models are confirmed by direct measurement of the dissolved ozone
profile in the contactor. Only after the model 1s confirmed to correctly
estimate the concentration profile In the contactor can 1t be used to
estimate the 1nact1vat1on performance of the contactor. Systems with
multiple chamber contactors must develop models for each of the chambers.
Various types of mathematical models for reaction-diffusion systems
were reported (Danckwerts, 1876) and some were shown to be applicable for
ozone contactors (Gurol and Singer, 1982). This section deliberately •
avoids giving preference to any type of mathematical modelling 1n order to
encourage engineering Innovations. The guidelines presented below may
help systems to select appropriate modelling that will be consistent with
the requirements of the SWTR.
The model should account for the ozone demand of the water being
treated in the contactor. The rate of ozone reaction and decomposition
should be based on batch experiments, on-site pilot plant columns, or
full-scale measurements.
The model should represent the actual flow distribution In the ozone
contact chambers by Incorporating a dispersion term and/or a three,
dimensional velocity distribution term 1n the contactor.
The modelled profile of the concentration of dissolved ozone 1n the
contactor should fit the actual distribution of dissolved ozone, as
verified by direct measurements, with a variation of less than 10 to 20
percent. This difference between the model and measured residual allows
for the inherent Inaccuracies 1n measuring the actual ozone residual, The
mathematically modelled concentration profile should not be used without
comparing it with actual measurements. Even elaborate mathematical models
0.4-7
-------
art not considered reliable enough to estimate the concentration
distributions of dissolved gasses In complex gas/11 quid operations,
without additional verification of the actual concentration profile In the
contactor.
In addition to the above guidelines, the nodel Bay also account for
other phenomena that may affect the performance of the ozone contact
chambers, such as: the effects of varying bubble diameter during Its
movement through the contactor, the effect of stagnant regions 1n the
contactor and the variation of the hydrostatic pressure.
For example, a system may use the two film theory coupled with
reaction kinetics to estimate the performance of an ozone contact chamber.
Using the two film theory the relevant differential equations are:
L dC/dz • Mt + Mr ~ Md
G dy/dz ¦ Mt
L dl/dz - Md - KCI
Where:
C - Concentration of dissolved ozone (mg/L)
G > Gas flow rate per cross section of the contactor (m2.Kg
. gas/m1n)
I ¦ Concentration of the target microorganism (Glardla or viruses)
L ¦ Water flow rate per cross section area of the reactor (Kg wat-
er/ml n.m2)
y ¦ Concentration of ozone 1n the gas phase (mg/L)
z ¦ Length coordinate of the contactor
Mt ¦ An expression for ozone transfer from the bubble phase to the
water phase. For example, l^afC.-C) where kta stands for the
volumetric mass transfer coefficient, C, represents the
1nterfac1a1 concentration of ozone, given by solubility data
(Table 0-10).
Mr ¦ An expression for the rate of ozone consumption in the water
due to auto-decompos1t1on and the ozone demand of the treated
water. For example, Mr- k,C -k2(C)(R). Where k, and k2 are
kinetic coefficients, and R represents the variable ozone
demand, such as TOC. An additional equation may be required to
represent the variation of R along the contactor.
0.4-8
-------
Md % An expression for the dispersion by turbulence and bubble flow
of dissolved ozone In the specific contactor. For example, ;
Od'C/dz* the dispersion coefficient (D) nay be evaluated by %
analysis of tracer study data. The third equation describing
the microorganism concentration (dl/dz) should Incorporate the
same dispersion coefficient (D).
KCI- Chick's 1nact1vat1on term (K-2.303k, where k ¦ Chick-Watson's
1nact1vat1on coefficient presented 1n Table 0-4, C represents
the local concentration of ozone and I represents the concen-
tration of microorganisms)%
The validity of these equations 1s subject to the appropriate
boundary conditions at the bottom and top of the contactor. The signs of
the various terms depend on the definition of coordinates and the type of
flow configuration (co-current or counter-current flow configuration),
0.4.4 Microbial Indicator Studies to Hodel Inactlvatlon Contactors
According to the recommendations In Appendix 6, systems may
demonstrate the actual performance of a disinfection system rather than
rely on the CT approach. The procedures outlined 1n Appendix G recommend
the use of Siardia murls cysts as indicators of Giardia Inactlvatlon and
bacteriophage (MS2) as Indicators for virus Inactlvatlon by disinfection
in general. However, recent data Indicate that HS2 phages may be
substantially more sensitive to ozone disinfection than pathogenic
viruses, and therefore are not a good Indicator for determining adequate
ozonation conditions for Inactivating pathogenic viruses (Finch, 1990).
Additional research is needed to determine which coliphage species, if
any, can be used as an appropriate indicator for virus inactlvatlon by
ozone. Pilot scale Inactlvatlon experiments using appropriate Indicator
microorganisms can serve as powerful tools to Indicate the performance of
the ozone contactors. This section contains guidelines for conducting
Indicator studies. At this time, full-scale testing with Indicator
organisms 1s not feasible because of the high volume of organisms needed
and the concern for introducing organisms Into the finished water.
However, with the development of naturally occurring indicators such as
resistant species of coliphage, demonstration on the full-scale level may
be feasible in the future.
0.4-9
-------
Systems may determine the performance of their disinfection basins
by demonstrating levels of inactlvatlon of Indicator microorganisms such
as Glardla murls cysts, or other Indicator microorganisms provided that
such demonstrations are based on solid engineering principles. The
following steps can be used for conducting Indicator studies:
1- Batch Experiments
On-site batch disinfection experiments are recoomended with treated
water spiked with Indicator microorganisms to determine the 1nact1vat1on
kinetics of the Indicator used 1n the pilot scale experiments. Microor-
ganisms should be used as Indicators preferably In the range where the
Inactlvatlon kinetics approximate Chick's law. This protocol assumes that
within the desired Inactlvatlon range, the Inactlvatlon kinetics will
approximate Chick's law. It-1s Important to note that other disinfection
kinetic models, not yet apparent, may be developed to more accurately
predict ozone Inactlvatlon efficiency than the Chick-Watson model.
Evidence that other models may be more appropriate 1s shown with data
generated by several researchers for different organisms (Wolfe, R.L. et
al, 1989; Finch 6., et al 1988; Finch G. and Smith, D.W. 1989).
2. Pilot Scale Indicator Experiments
Pilot-scale experiments should then be conducted using Identical
strains of biological Indicators to those used 1n the batch experiments.
The pilot-scale experiments should be repeated under Identical gas and
water flow conditions with and without Introducing ozone Into the gas
stream. The actual performance may then be calculated by subtracting the
inactlvation achieved 1n the control experiment (without ozone) from the
Inactlvatlon achieved 1n the ozone disinfection experiments.
3. Evaluation of Inactlvatlon Performances
Systems may choose direct or Indirect methods to Interpret the
Inactlvatlon performance of ozone contactors based on Indicator studies.
The direct method 1s more conservative and simple while the Indirect
method 1s more accurate but requires mathematical modelling of the
contactors. The two procedures are outlined below:
0.4-10
-------
P1r*ct prediction of inactivity perform™™
i. Determine k, (where k, is Chick-Watson's inactlvation
coefficient of the Indicator microorganism) from bitch
t«st data with the expression:
1 °9(I/Io^Indicator " *kjCt
where:
(I/lo) indicator % Survival ratio of Indicator microorganism
•* determined by batch experiments.
C * Dissolved ozone concentration 1n the batch
experiment (mg/L)
t ¦ time (minutes) elapsed from the beginning
of the batch experiment
Note; This assumes that the 1nact1vat1on data *#111 provide a
reasonable fit for this equation. If this 1s not true,
then the following 1s not applicable and other relation-
ships should be developed.
b. Determine the disinfection performance of the pilot
scale disinfection system on the Indicator microorganism
(I/Ie) Indicator'
c. Calculate the 1nact1vat1on of S1ard1i cysts or viruses
(I/I0) using the appropriate k' values from Table 0-3:
log(I/IQ> ¦ 1og(I/It)
Indicator (k'/k,) I Oil > k) {!)
log (I/IO)log k). then the plug flow operation represents the more
conservative prediction approach. Equation 7 Is based on the assumption
that the flow configuration In the chamber approaches plug flow. When the.
Indicator microorganism 1s more vulnerable then the target microorganism
(k, > k) then the CSTR approach provides a more conservative estimate.
Equation 4 represents a conservative approximation to the CSTR similarity
0.4-U
-------
rule. A «ort accurate determination of the Inactlvatlon performance of
th« contactor nay be calculated by the following approach:
2.
ndlrect determination of the disinfection performance
Determine k, (where k. 1s Chick-Watson's 1nact1vat1on
coefficient of the Indicator microorganism) from batch
test data with the expression:
Og(I/Io) indicator *
here:
l/Io)titfiHtor " Survival ratio of Indicator micro-organisms
as determined by batch experiments.
Olssolved ozone concentration 1n the batch
experiment (mg/L)
% time (minutes) elapsed from the beginning
of the batch experiment
Determine the disinfection performance of the pilot
scale Inactlvatlon level of the Indicator microorganism
indicator*
Determine the actual concentration profile 1n the
disinfection chamber (see Section 0.3.2).
Construct a mathematical model that estimates the
concentration profile 1n the contactor as discussed 1n
Section 0.4.3
Confirm the Mathematical model by fitting Its parameters
such as dispersion or kinetic coefficients to describe
accurately the concentration profile of ozone 1n the
contactor and the overall Inactlvatlon of the Indicator
microorganism. A model that predicts within 10-20.
percent the Inactlvatlon of the Indicator microorganism
and the concentration profile of dissolved ozone tn the
contactor would be considered to be valid and can be
used by Incorporating k values from Table 0-3 to
estimate the Inactlvatlon of G1ard1a cysts or viruses 1n
the contactor.
-------
REFERENCES
Bider, H. and J. Holgne, Determination of Ozone 1n Water by the Indigo
Method, A Submitted Standard Method, Ozone: Science and Engineering, pp
449-456, 1982, A
Cams, K, Design of East Bay Municipal Utility District ozone disinfection
plants in Oakland Ca., presented at EPA workshop on Ozone CT, Cincinnati,
OH., Feb. 1990
Chick, H., An Investigation of the laws of disinfection, J. Hygiene, 8, 92
(1908)
Danckwerts, P.V. Gas Liquid Reactions, McGraw H111 Inc. (1976)
Finch, G.R., Smith, D.W., Stiles, M.E., Dose Response of Escherichia Col 1
In Ozone Demand-free Phosphate Buffer, Water Research Vol. 22, No.12, pp.
1563-1570, 1988.
Finch, G., University of Alberta, Canada, Private Communication (1990).
Hoff, J. C., Strengths and weaknesses of using CT values to evaluate
disinfection practice, In assurance of adequate disinfection, or C-T or
not C-T, AWWA seminar proceedings, Annual AWWA Conference, Kansas C1ty,
pp. 9 (1987)
Gurol M.D. and P.C. Singer, Kinetics of ozone decomposition, A dynamic
approach, Env. Sci. & Tech., 16, 377 (1982) .
Joost, R.D., L.A. Jackson and L.J. Bollyky, Optimization of ozone
contactors for drinking water disinfection, Proceedings;of the 9th Ozone
World Congress, L.J. Bollyky, Editor, New York, NY, vol 1, pp. 551 (1989).
Lev, 0. and S. Regll, Compliance of ozone disinfection systems with the
Surface Water Treatment Rule, Selection of characteristic exposure time
(T), submitted to J. of Env. Eng. ASCE (1990a)
Lev, 0. and S. Regll, Compliance of ozone disinfection systems with the
Surface Water Treatment Rule, Selection of characteristic concentration
(C), submitted to J. of Env. Eng. ASCE (1990b)
Levensplel, 0., Chemical Reaction Engineering, 2nd ed. Wiley, New York, NY
(1972).
Perry R.H., Perry's Chemical Engineering Handbook, McGraw Hill Book
Company, New York, NY (1963)
Richards, 0. A. and Flelschman, M., Ozone transfer to aqueous systems in
a static mixer, 2nd Symposium on ozone technology, Montreal, Canada, R. G.
Rice, P. Pichet and M. A. Vincent, eds, p. 57 (1975).
-------
Rice, R. EPA, 1n preparation, Source Document: Ozone Treatment of
OrInking Water, EPA/625/1-86/021.
Rosenbeck, G.L, McMahon Associates, Inc., Private communlcation (1989)
Schwartz, B., Hackensack Hater Company, Private communication (1989)
Schwartz, B., L.C. Fung and C. N. Weng, Determination of ozone concentra-
tion in turbine contactor, presented at EPA Workshop on Ozone C.T,
Cincinnati, OH., Feb-. 1990
Seinfeld J. H. and Lapldus L., Mathematical methods 1n Chemical Engineer-
ing, Vol. 3, Process Modeling Estimation and Identification, Prentice Hall
Inc. (1984).
Stolarlk, G.L. and J. 0. Christie, Projection of ozone C-T values, Los
Angeles Aqueduct Filtration Plant, presented at Int. Ozone Assoc. Conf.
Monroe, Mich., April 1987.
Stolarlk, G.L. and J. 0. Christie, Projection of ozone C-T values, Los
Angeles Aqueduct Filtration Plant, presented at EPA workshop on Ozone CT,
Cincinnati, OH., Feb. 1990
Treybal, R.E., Mass-transfer operations, McGraw-Hill Book Company, 3rd ed.
(1980)
Watson, H.E., A note on the variation of the rate of disinfection with
change 1n the concentration of the disinfectant, J. Hygiene, 8, 536 (1908)
Wolfe, R.L, M.H. Stewart, K.N. Scott and M. J. McGuire, Inactivation of
Giardia Murls and Indicator organisms seeded 1n surface water supplies by
peroxone and ozone, Env. Sci. and Tech., 23, 744 (1989)
Yuteri C. and Gurol M.6. Evaluation of kinetic parameters for the
ozonation of organic micropollutants, Wat. Sc1. & Tech. 21, 465 (1989)
------- |