oEPA
United States
Environmental Protection
Agency
Six-Year Review 3 Technical Support
Document for Chlorate
-------
Office of Water (4607M)
EP A-810-R-16-013
December 2016
-------
Disclaimer
This document is not a regulation. It is not legally enforceable, and does not confer legal rights
or impose legal obligations on any party, including EPA, states, or the regulated community.
While EPA has made every effort to ensure the accuracy of any references to statutory or
regulatory requirements, the obligations of the interested stakeholders are determined by statutes,
regulations or other legally binding requirements, not this document. In the event of a conflict
between the information in this document and any statute or regulation, this document would not
be controlling.
-------
This page intentionally left blank.
-------
Table of Contents
1 Introduction 1-1
2 Contaminant Background 2-1
2.1 Chemical and Physical Properties 2-1
2.2 Production, Use and Release 2-2
2.2.1 Commercial Production and Use in Industry and Agriculture 2-2
2.2.2 Incidental Production and Release 2-6
2.3 Environmental Fate 2-7
2.4 Regulatory and Non-Regulatory Actions for Chlorate 2-7
3 Health Effects 3-1
3.1 Summary of Health Effects 3-1
3.2 Derivation of the Health Reference Level 3-3
3.2.1 Considerations of Relative Source Contribution (RSC) from Drinking
Water for Chlorate 3-4
3.3 Additional Perspective on Chlorate in DBP Mixtures from Epidemiology
Studies 3-5
4 Analytical Methods 4-1
5 Occurrence and Exposure in Drinking Water 5-1
5.1 UCMR 3 Monitoring Program and Dataset 5-2
5.2 Summary of Analytical Results with UCMR 3 Data 5-4
5.3 Occurrence and Exposure Based on UCMR 3 Locational Average
Concentrations 5-5
5.4 Occurrence by Disinfectant Type 5-6
5.5 Occurrence of Chlorate by System Size 5-15
5.6 Variation of Occurrence from EPs to MRs 5-17
5.7 Comparing UCMR 3 and DBP ICR Data for Occurrence 5-20
6 Formation in Drinking Water 6-1
6.1 Chlorine Dioxide 6-1
6.2 Bulk Hypochlorite Solution 6-3
6.3 On-Site Generated Hypochlorite 6-5
6.4 Ozone 6-6
6.5 Gaseous Chlorine along with Other Disinfectant Types 6-7
7 Treatment 7-1
7.1 Reduction of Disinfectant Demand 7-1
Six-Year Review 3 i December 2016
Technical Support Document for Chlorate
-------
7.2 Modification of Disinfection Practices 7-2
7.2.1 Chlorine Dioxide 7-2
7.2.2 Bulk Hypochlorite Solution 7-3
7.2.3 On-Site Generation of Hypochlorite 7-4
7.2.4 Ozone 7-4
7.2.5 Gaseous Chlorine along with Other Disinfectant Types 7-5
7.2.6 Removal of Chlorate 7-5
8 References 8-1
Appendix A Supplemental Data Sources A-l
A. 1 Disinfection Byproducts Information Collection Rule (DBP ICR), 1997-1998 A-l
A. 1.1 Summary Analysis A-3
A. 1.2 Limitations of DBP ICR Data A-6
A.2 Community Water System Survey (CWSS), 2006 A-6
A.3 Environmental Working Group (EWG) Drinking Water Database, 2004-2009 A-7
Appendix B Additional UCMR 3 Occurrence Analyses B-l
B. 1 Additional Analyses on Occurrence by System Type B-l
B.2 Analyses on Samples with Detections B-6
B.3 Additional Analyses on Locational Averages B-l8
B.4 Analyses on Disinfectants Used B-22
B.5 Changes in Disinfection Practice B-28
Technical Support Document for Chlorate
-------
List of Exhibits
Exhibit 2.1: Chemical Structure of Chlorate 2-1
Exhibit 2.2: Physical and Chemical Properties of Chlorate 2-2
Exhibit 2.3: Reported Annual Manufacture and Importation of Sodium Chlorate and
Potassium Chlorate in the United States (pounds), from EPA's CDR Program 2-4
Exhibit 2.4: Estimated Annual Agricultural Use of Sodium Chlorate, 2012 2-5
Exhibit 4.1: Method Sensitivity Ratios (MSRs) for Chlorate 4-2
Exhibit 5.1. Comparison of the Number of Systems in the UCMR 3 Sample Design with
the Number of Systems with Chlorate Data in UCMR 3 Data 5-3
Exhibit 5.2. UCMR 3 Chlorate Data, By System Type 5-4
Exhibit 5.3: National Estimates of Sample Locations and Associated Population Served
with Locational Average Chlorate Concentrations Greater than Threshold Values (Based
on UCMR 3 Data) 5-6
Exhibit 5.4: Chlorate Occurrence by Form of Chlorine 5-9
Exhibit 5.5: Chlorate Occurrence, Chlorination versus Chloramination 5-10
Exhibit 5.6: Chlorate Occurrence when Chlorine Dioxide is in Use 5-11
Exhibit 5.7: Chlorate Occurrence when Ozone is in Use 5-12
Exhibit 5.8: Chlorate Occurrence when Other or No Disinfectants are in Use 5-13
Exhibit 5.9: UCMR 3 Chlorate Occurrence at Systems Using Chlorine Dioxide and
Hypochlorination, by System Size 5-16
Exhibit 5.10: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at All UCMR 3 Sampling Locations 5-17
Exhibit 5.11: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at UCMR 3 Sampling Locations Where Chlorine Dioxide was Reported
to Be in Use 5-18
Exhibit 5.12: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at UCMR 3 Sampling Locations Where Hypochlorination was Reported
to Be in Use 5-19
Exhibit 5.13: DBP ICR and UCMR 3 Comparison - Chlorate Occurrence in Common
Systems Using Chlorine Dioxide and Hypochlorite 5-21
Exhibit A. 1: Inventory of DBP ICR Systems Reporting Chlorate Occurrence Data A-3
Exhibit A.2: Chlorate DBP ICR Occurrence Data from Systems Required to Monitor -
Summary of Detected Concentrations A-4
Exhibit A.3: Chlorate DBP ICR Occurrence Data from Systems Required to Monitor -
Summary of Samples A-4
Exhibit A.4: Chlorate DBP ICR Occurrence Data - Summary of System and Population
Served Data from Systems Required to Monitor - All Detections A-5
Technical Support Document for Chlorate
-------
Exhibit A. 5: Chlorate DBP ICR Occurrence Data - Summary of System and Population
Served Data from Systems Required to Monitor - Detections > HRL A-5
Exhibit A.6: Chlorate DBP ICR Occurrence Data - Summary of System and Population
Served Data from Systems Required to Monitor - Detections > 2xHRL A-6
Exhibit A.7: Summary of EWG Chlorate Data, 2004-2009 A-8
Exhibit B. 1: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of
Samples with Detections Greater than Threshold Values (Community Water Systems) B-2
Exhibit B.2: Chlorate National Occurrence Measures Based on UCMR 3 Assessment
Monitoring Data - Summary of System and Population Served Data - Detections in
CWSs B-3
Exhibit B.3: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of
Samples with Detections Greater than Threshold Values (Non-Transient Non-Community
Water Systems) B-4
Exhibit B.4: Chlorate National Occurrence Measures Based on UCMR 3 Assessment
Monitoring Data - Summary of System and Population Served Data - Detections in
NTNCWSs B-5
Exhibit B.5: Chlorate Occurrence Data from UCMR 3 - Summary of Detected
Concentrations B-7
Exhibit B.6: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of
Samples with Detections Greater than Threshold Values B-8
Exhibit B.7: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
System and Population Served Data - Detections of Chlorate B-9
Exhibit B.8: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
System and Population Served Data - Detections of Chlorate > HRL (210 |ig/L) B-10
Exhibit B.9: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
System and Population Served Data - Detections of Chlorate > 2xHRL (420 |ig/L) B-l 1
Exhibit B. 10: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
System and Population Served Data - Detections of Chlorate > 3xHRL (630 |ig/L) B-l2
Exhibit B. 11: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
Sampling Locations and Proportional Population Served Data - Detections of Chlorate B-l3
Exhibit B. 12: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
Sampling Locations and Proportional Population Served Data - Detections of Chlorate >
HRL (210 ug/I.) B-l4
Exhibit B. 13: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
Sampling Locations and Proportional Population Served Data - Detections of Chlorate >
2xHRL (420 ug/I.) B-l5
Exhibit B. 14: National Occurrence and Exposure Based on UCMR 3 Data - Summary of
Sampling Locations and Proportional Population Served Data - Detections of Chlorate >
3xHRL (630 ug/I.) B-l6
Technical Support Document for Chlorate
-------
Exhibit B. 15: National Estimates of Systems and Population Served by Systems with At
Least One Detection of Chlorate Greater than Threshold Values (Based on UCMR 3
Data) B-17
Exhibit B. 16: National Occurrence Based on UCMR 3 Data - Summary of Sample
locations - Locational Average Chlorate Concentration > HRL (210 |ig/L) B-19
Exhibit B. 17: National Occurrence Based on UCMR 3 - Summary of Sample Locations -
Locational Average Chlorate Concentration > 2xHRL (420 |ig/L) B-20
Exhibit B. 18: National Occurrence Based on UCMR 3 - Summary of Sample Locations -
Locational Average Chlorate Concentration > 3xHRL (630 |ig/L) B-21
Exhibit B. 19: Use of Disinfectants by Source Water Type and System Size Based on
UCMR 3 Data in EPs (select categories) B-23
Exhibit B.20: Use of Disinfectants by Source Water Type and System Size Based on
UCMR 3 Data in EPs (mutually exclusive categories) B-24
Exhibit B.21: Use of Disinfectants by Source Water Type and System Size Based on
UCMR 3 Data in MRs (select categories) B-25
Exhibit B.22: Use of Disinfectants by Source Water Type and System Size Based on
UCMR 3 Data in MRs (mutually exclusive categories) B-26
Exhibit B.23: UCMR 3 Inventory of Chlorate Samples by Disinfectant Type B-27
Exhibit B.24: DBP ICR and UCMR 3 Comparison - Use of Disinfectants in Surface
Water Plants (Select Categories) B-29
Exhibit B.25: Comparison of Chloraminating Systems in UCMR 2 and UCMR 3 B-30
Exhibit B.26: DBP ICR and UCMR 3 Comparison - Hypochlorite Use in Surface Water
Plants B-30
Six-Year Review 3
Technical Support Document for Chlorate
v
December 2016
-------
Abbreviations
AMWA
Association of Metropolitan Water Agencies
AWWA
American Water Works Association
AwwaRF
American Water Works Association Research Foundation
BMD
Benchmark Dose
BMDL
Benchmark Dose Level
CAGC
Chloramine Formed from Gaseous Chlorine
CAOF
Chloramine Formed from Off-Site Hypochlorite
CAON
Chloramine Formed from On-Site Hypochlorite
CAS
Chemical Abstracts Service
CCL 3
Third Contaminant Candidate List
CCR
Consumer Confidence Report
CDR
Chemical Data Reporting
CLDO
Chlorine Dioxide
CLGA
Gaseous Chlorine
CLOF
Off-site Generated Hypochlorite Stored as Liquid
CLON
On-site Generated Hypochlorite with No Storage
CSFII
Continuing Survey of Food Intakes by Individuals
CWS
Community Water System
CWSS
Community Water System Survey
DBP
Disinfection Byproducts
DBP ICR
Disinfection Byproducts Information Collection Rule
D/DBPR
Disinfectants and Disinfection Byproducts Rule
DL
Detection Limit
EP
Entry Point to the Distribution System
EPA
U.S. Environmental Protection Agency
EWG
Environmental Working Group
G6PD
Glucose-6-Phosphate Dehydrogenase
GAC
Granular Activated Carbon
GW
Ground Water
GWUDI
Ground Water Under the Direct Influence of Surface Water
HSDB
Hazardous Substances Data Bank
HRL
Health Reference Level
2xHRL
Twice the HRL
3xHRL
Three times the HRL
IUR
Inventory Update Reporting
LC-MS/MS
Liquid Chromatography and Tandem Mass Spectrometry
LOAEL
Lowest Observed Adverse Effect Level
LRAA
Locational Running Annual Average
MAC
Maximum Acceptable Concentration
MCL
Maximum Contaminant Level
MCLG
Maximum Contaminant Level Goal
MDBP
Microbial and Disinfection Byproduct
MDL
Method Detection Limit
Six-Year Review 3
Technical Support Document for Chlorate
vi
December 2016
-------
MDPC
Microbial/Disinfection Products Council
MR
Location of Maximum Residence Time within the Distribution System
MRL
Minimum Reporting Level
MSR
Method Sensitivity Ratio
NCFAP
National Center for Food and Agricultural Policy
NOAEL
No Observed Adverse Effect Level
NODU
No Disinfection
NTNCWS
Non-Transient Non-Community Water System
NTP
National Toxicology Program
OPP
Office of Pesticide Programs
OSG
On-Site Generated or On-Site Generation
OTHD
All Other Types of Disinfectant
OZON
Ozone
PAC
Powdered Activated Carbon
PWS
Public Water System
PWSID
Public Water System Identification
RBC
Red Blood Cell
RD 3
Third Round of Regulatory Determinations
RfD
Reference Dose
RSC
Relative Source Contribution
RSD
Relative Standard Deviation
SM
Standard Method
SDWIS/Fed
Safe Drinking Water Information System/Federal Version
SW
Surface Water
THMs
Trihalomethanes
TRI
Toxics Release Inventory
TSC
Technical Service Center
TSH
Thyroid Stimulating Hormone
UC MR 1
First Unregulated Contaminant Monitoring Regulation
UCMR2
Second Unregulated Contaminant Monitoring Regulation
UCMR3
Third Unregulated Contaminant Monitoring Regulation
UF
Uncertainty Factor
ULVL
Ultraviolet Light
USACE
U.S. Army Corps of Engineers
USEPA
U.S. Environmental Protection Agency
USGS
United States Geological Survey
UV
Ultraviolet
WHO
World Health Organization
WRF
Water Research Foundation
Six-Year Review 3
Technical Support Document for Chlorate
vii
December 2016
-------
1 Introduction
The Safe Drinking Water Act requires the United States Environmental Protection Agency
(EPA) to conduct a periodic review of existing National Primary Drinking Water Regulations
and determine which, if any, are candidates for revision. The purpose of the review, called the
Six-Year Review, is to evaluate current information for each National Primary Drinking Water
Regulation to determine if there is new information on health effects, treatment technology,
analytical methods, occurrence and exposure, implementation and/or other factors that provide a
health or technical basis to support a regulatory revision that will improve or strengthen public
health protection.
Under Six-Year Review 3, EPA is reviewing the regulated chemical, radiological and
microbiological contaminants included in previous reviews, as well as the Microbial and
Disinfection Byproducts (MDBP) regulations that were promulgated under the following actions:
the Disinfectants and Disinfection Byproduct Rules (D/DBPRs), the Surface Water Treatment
Rules, the Ground Water Rule and the Filter Backwash Recycling Rule. The Surface Water
Treatment Rules consist of the Surface Water Treatment Rule, the Interim Enhanced Surface
Water Treatment Rule, the Long Term 1 Enhanced Surface Water Treatment Rule (LT1) and the
Long Term 2 Enhanced Surface Water Treatment Rule (LT2). This is the first time that EPA is
reviewing the MDBP rules. For more information about the Six-Year Review of the D/DBPRs,
the reader is referred to EPA's Six-Year Review 3 Technical Support Document for
Disinfectants/Disinfection Byproducts Rules (USEPA, 2016a). Under the SYR3, EPA also is
evaluating unregulated DBPs: for example, chlorate and nitrosamines.
Chlorate was included on EPA's Third Contaminant Candidate List (CCL 3) and evaluated as a
candidate for regulation under the Regulatory Determinations 3 (RD 3) program in 2014. In the
Federal Register notice for Preliminary Regulatory Determination 3 (79 FR 62715, USEPA,
2014a), the Agency stated that "because chlorate and nitrosamines are DBPs that can be
introduced or formed in public water systems (PWSs) partly because of disinfection practices,
the Agency believes it is important to evaluate these unregulated DBPs in the context of the
review of the existing DBP regulations. DBPs need to be evaluated collectively, because the
potential exists that the chemical disinfection used to control a specific DBP could affect the
concentrations of other DBPs. Therefore, the Agency is not making a regulatory determination
for chlorate and nitrosamines at this time."
Chlorate, like the related compound chlorite, is an oxidation state of chlorine. Chlorate and
chlorite are chemically inter-convertible (see Chapter 6 for a detailed discussion.) They occur,
and can co-occur, when hypochlorite solution and/or chlorine dioxide are applied during the
drinking water treatment process. Chlorite is a regulated DBP (USEPA, 2016a). The potential
common health effects and co-occurrence of chlorate and chlorite are discussed in the Six-Year
Review 3 Technical Support Document for Disinfectants/Disinfection Byproducts Rules (USEPA,
2016a).
Six-Year Review 3
Technical Support Document for Chlorate
1-1
December 2016
-------
The remainder of this document provides a summary of available information and data relevant
to EPA's understanding of the contaminant background, health effects, analytical methods,
occurrence and exposure, formation and treatment/control strategies for chlorate. The
information cutoff date for Six-Year Review 3 was December 2015. That is, information that
was published after December 2015 was not considered for this document. The Agency
recognizes that scientists and other stakeholders are continuing to investigate and publish
relevant information subsequent to the cutoff date. While not considered as part of Six-Year
Review 3, the Agency anticipates providing consideration of that additional information in
subsequent activities.
Six-Year Review 3
Technical Support Document for Chlorate
1-2
December 2016
-------
2 Contaminant Background
This chapter presents background information on chlorate that EPA is evaluating under the
SYR3 program. The following topics are covered in the chapter: physical and chemical
properties; production, use and release; environmental fate; and regulatory and non-regulatory
actions.
The chlorate anion (CIO3") forms a variety of salts (e.g., sodium chlorate, calcium chlorate,
potassium chlorate and magnesium chlorate) that are collectively known as chlorates. Chlorate
and its salts are powerful oxidizers. Sodium chlorate is registered for use as an herbicide and to
generate chlorine dioxide for multiple uses, including bleaching paper and disinfecting drinking
water (USEPA, 2006a). Disinfection practices are an important source of chlorate in drinking
water; this includes formation as a disinfection byproduct (DBP) and presence in disinfectants as
an impurity (USEPA, 2006a).
2.1 Chemical and Physical Properties
Exhibit 2.1 presents the structural formula for chlorate. Physical and chemical properties and
other reference information are listed in Exhibit 2.2.
Exhibit 2.1: Chemical Structure of Chlorate
i
0"
Source: ChemlDPIus/National Library of Medicine
Six-Year Review 3
Technical Support Document for Chlorate
2-1
December 2016
-------
Exhibit 2.2: Physical and Chemical Properties of Chlorate
Property
Data
Chemical Abstracts Service (CAS)
Registry Number
14866-68-3 (CAS registry number for "chlorates")
EPA Pesticide Chemical Code
073301 (sodium chlorate), 073302 (calcium chlorate), 073303
(potassium chlorate), 530200 (magnesium chlorate)
Chemical Formula
ClOs"
Molecular Weight
83.45 g/mol (Calculated)
Color/Physical State
Colorless or white crystals (Lide, 1984)
Boiling Point
Decomposes when heated above melting point (Lide, 1984)
Melting Point
Varies with the salt (Lide, 1984); 248 deg C (NaCIOs) (HSDB,
2015)
Density
2.5 g/cm3 (NaCIOs) (HSDB, 2015)
Freundlich Adsorption Coefficient
--
Vapor Pressure
Negligible at room temperature (NaCICb) (HSDB, 2015)
Henry's Law Constant
--
Log Kow
--
Koc
--
Solubility in Water
1,000,000 mg/L @25°C (NaCIOs) (HSDB, 2015)
Other Solvents
Slightly soluble in ethanol (NaCIOs) (HSDB, 2015)
Note:indicates that no information was found.
2.2 Production, Use and Release
2.2.1 Commercial Production and Use in Industry and Agriculture
According to Bommaraju and O'Brien (2015), most commercially-produced sodium chlorate
(over 95 percent) is used to generate chemicals (e.g., chlorine dioxide) used for bleaching in the
pulp and paper industry. The remainder is used in agriculture as an herbicide, in the manufacture
of chlorites and potassium chlorate, in the hydraulic mining of uranium, and in the production of
perchlorate for pyrotechnics, rocketry, and matchheads. Estimates of actual U.S. production and
importation of both sodium chlorate and potassium chlorate, based on data gathered under EPA's
Chemical Data Reporting (CDR) program, are presented below in Exhibit 2.3.
No industrial release data for chlorate or any of its salts are available from EPA's Toxics Release
Inventory (TRI). (More precisely: the list of compounds for which TRI reporting is required has
never included a compound with "chlorate" in its name (USEPA, 2016b)).
Sodium chlorate is approved for use on cotton, rice, corn, soybeans, dry beans, potatoes,
sunflowers, flax, safflower, chili peppers, grain sorghum and wheat. It is also registered for use
as a nonselective herbicide to kill grasses and weeds in industrial and non-agricultural sites such
as uncultivated areas/soils and around ornamentals. There are 30 active product registrations
containing sodium chlorate as an active ingredient (USEPA, 2016c). Data on the application of
sodium chlorate compounds as a pesticide are available from several sources, as described in the
Six-Year Review 3
Technical Support Document for Chlorate
2-2
December 2016
-------
sections below. (No information about sodium chlorate usage was found in EPA's Pesticide
Industry Sales and Usage Reports.)
2.2.1.1 Inventory Update Reporting (IUR) / Chemical Data Reporting (CDR) Program
In compliance with the Toxic Substances Control Act, EPA gathers information on the
manufacturing (including both domestic manufacture and importation) of chemical substances.
Under the Inventory Update Rule (IUR), manufacturers (including importers) provided
information on a periodic basis between 1986 and 2006. Under the CDR Rule that superseded
the IUR in 2011, manufacturers (including importers) are continuing to provide information once
every four years (reporting under this rule began in 2012). CDR data gathered in 2012 cover
reporting years 2010 and 2011.
Production data from EPA's CDR program are available for sodium chlorate and potassium
chlorate for the years 2010 and 2011. No data on chlorate-related compounds are available from
the IUR program in earlier years. Under CDR, the minimum reporting threshold was 25,000
pounds (USEPA, 2014b).
Eighteen industrial sites are reported as having "manufactured" (i.e., domestically manufactured
or imported) sodium chlorate in both 2010 and 2011. The available data indicate that at least five
of those sites domestically manufactured sodium chlorate, nine sites imported it, and one did
both. Over 1.9 billion pounds were reported as manufactured each year in both 2010 and 2011.
The actual quantities produced were likely higher, as some production figures were redacted as
confidential business information. Reported per-facility imported quantities ranged from 0 to
over 583 million pounds, and reported per-facility manufactured quantities ranged from 3.5
million to 311 million pounds.
There was no reported domestic manufacture of potassium chlorate in 2010 or 2011. Only one
industrial site imported potassium chlorate, in quantities of 119,048 pounds in 2010 and 198,414
pounds in 2011.
Six-Year Review 3
Technical Support Document for Chlorate
2-3
December 2016
-------
Exhibit 2.3: Reported Annual Manufacture and Importation of Sodium Chlorate
and Potassium Chlorate in the United States (pounds), from EPA's CDR Program
Contaminant
Type of Activity
Chemical
Inventory Update
Reporting Cycle
(2010)
Chemical
Inventory Update
Reporting Cycle
(2011)
Sodium Chlorate
Domestic
Manufacture
748,053,800
1,023,390,997
Sodium Chlorate
Importation
730,746,281
913,070,966
Sodium Chlorate
Total
1,936,490,474
1,936,461,963
Potassium Chlorate
Domestic
Manufacture
0
0
Potassium Chlorate
Importation
119,048
198,414
Potassium Chlorate
Total
119,048
198,414
Source: USEPA, 2015a. Note: Because some reports do not specify whether production volumes represent
manufacture or importation, values may not add up to totals.
2.2.1.2 National Center for Food and Agricultural Policy (NCFAP) Pesticide Use Database
The National Center for Food and Agricultural Policy (NCFAP) maintains a national Pesticide
Use Database, primarily for herbicides. Pesticide use estimates are based on state-level
commercial agriculture usage patterns and state-level crop acreage. NCFAP listed uses of
sodium chlorate on six crops totaling approximately 8,293,000 pounds active ingredient per year
in 14 states in 1992. In 1997, NCFAP listed uses of sodium chlorate on seven crops totaling
approximately 7,262,000 pounds active ingredient per year in 16 states (NCFAP, 2009).
2.2.1.3 EPA Office of Pesticide Programs (OPP) Registration Review Program
In 2006, EPA's Office of Pesticide Programs (OPP) estimated that approximately 2.8 million
pounds of sodium chlorate active ingredient were used annually in the United States (USEPA,
2006a). A 2015 screening level usage analysis covering the 2004-2013 timeframe indicated that
approximately 1.2 million pounds of sodium chlorate were applied annually in agriculture, with
the bulk applied to cotton and rice (USEPA, 2016d). This estimate does not include anti-
microbial applications.
2.2.1.4 United States Geological Survey (USGS) Pesticide Use Maps
The United States Geological Survey (USGS) has produced maps of pesticide use for several
hundred compounds used in United States crop production. The pesticide use maps show the
average annual pesticide use intensity expressed as average weight (in pounds) of a pesticide
applied to each square mile of agricultural land in a county. The USGS maps were created using
data from NCFAP and county-level information on harvested crop acreage from the Census of
Agriculture. The maps are complemented by bar graphs showing trends in total quantity applied
annually, broken out by crop.
Six-Year Review 3
Technical Support Document for Chlorate
2-4
December 2016
-------
Exhibit 2.4 (USGS, 2015) shows the geographic distribution of estimated average annual sodium
chlorate agricultural use in the United States in 2012. A breakdown of annual use by crop from
1992 to 2012 is presented in Exhibit 2.5. USGS used two methods to estimate sodium chlorate
usage, since pesticide usage information was not available in some districts. On the left of
Exhibit 2.4 and Exhibit 2.5, the "EPest-High" estimates were generated by projecting usage
estimates for such districts based on usage in neighboring districts. On the right of Exhibit 2.4
and Exhibit 2.5, the "EPest-Low" estimates were generated by assuming no usage in such
districts. According to these USGS estimates, annual usage peaked in the mid-1990s with a high
of at least -7.5 million pounds (in the low-usage estimate) of sodium chlorate. Since that time,
annual use has exhibited a general decline, with approximately one million pounds having been
applied annually in 2010, 2011 and 2012. The maps in Exhibit 2.4 indicate that the greatest use
of sodium chlorate is in California, Arizona, Wisconsin and Georgia. Cotton is the major crop
treated with sodium chlorate. (The anomalous spike in pasture and hay usage in 1996 appears to
be an artifact of the "EPest-High" methodology.)
Exhibit 2.4: Estimated Annual Agricultural Use of Sodium Chlorate, 2012
Estimated use on
agricultural land, in
pounds per square mile
I I < 0.17
I 10.17- 1.25
EPest-Low
EPest-High
I I No estimated use
Source: USGS, 2015
Six-Year Review 3
Technical Support Document for Chlorate
2-5
December 2016
-------
Exhibit 2.5: Estimated Annual Agricultural Use of Sodium Chlorate by Year and
Crop, 2012
3D
25
= 20-
Use by Year and Crop
EPest-High E Pest-Low
30
25
20-
15
10-
NPJ'fUllflNOnOi-IMCJ'tlOlflNfflOlO'-N
fflfflfflfflCPlfflfflJlOOOOOOOOOQrrr
Cr>CT>CDCDCr>o-iCr>CnOOOOOOOOO OOOO
fllih
llln..
i—i Other crop6
i—i Pasture ana nary
1=1 Alfalfa
¦¦ Ofciarde and graces
I—I Rice
i i vegetatxee ana Trull
^ Cotton
B wt>eat
RgBi SovCesrB
I I Corn
N(0^f««}NO)CT>CDCDCr>CTJCr>cnOOOOOOOOO OOOO
Source: USGS, 2015
2.2.2 Incidental Production and Release
Disinfection practices are an important source of chlorate in drinking water (USEPA, 2006a).
The chlorate ion may be present as an impurity in sodium chlorite, the most common feedstock
used to generate chlorine dioxide for drinking water treatment. Less frequently, chlorate
compounds are used to generate chlorine dioxide for disinfection. Chlorate may persist and be
carried through to finished water in either case. In addition, chlorate is one of a number of DBPs
that can form during and after chlorine dioxide use in water treatment (USEPA, 2006a).
Chlorate may also be present as an impurity in hypochlorite solutions (sodium hypochlorite and
calcium hypochlorite) used for drinking water disinfection. Concentrations of chlorate in
hypochlorite solutions typically increase with storage time, and increase more quickly at higher
temperatures. Chlorate may be introduced into water when the solutions are used for disinfection
(Gordon et al., 1995; USEPA, 1999; USEPA, 2006a).
A report by the American Water Works Association Research Foundation (AwwaRF, since
renamed the Water Research Foundation or WRF) (Gordon et al., 1995) found that chlorate
concentrations in finished water were higher at facilities that use hypochlorite solutions for
disinfection (mean concentration of 0.49 mg/L) than at facilities that use chlorine dioxide (mean
concentrations of 0.25-0.29 mg/L). Additional information on the formation of chlorate during
drinking water disinfection is presented in Chapter 6.
Chlorate may be present as an impurity or generated as a byproduct in other contexts as well. For
example, just as it may be present in hypochlorite solutions used for drinking water disinfection
as noted above, chlorate may also be present as an impurity in commercial sodium hypochlorite
Six-Year Review 3
Technical Support Document for Chlorate
2-6
December 2016
-------
used for cleaning, for pool disinfection and for medical and other uses. ICIS (2006) estimates
that in 2005, 553 million gallons of household strength (5.25 percent) sodium hypochlorite and
310 million gallons of industrial strength (12.5 percent) sodium hypochlorite were used in the
U.S. Use and disposal of this sodium hypochlorite could lead to the presence of chlorate in
wastewater and the ambient environment.
2.3 Environmental Fate
Sodium chlorate is not expected to volatilize from soil or water, and sodium chlorate has low
potential to bioaccumulate. Sodium chlorate is reported to persist in soil for 0.5 to 5 years,
depending on soil type, application rate and weather conditions (USEPA, 2006a).
Chlorate salts readily dissolve in water. In the absence of redox reactions, the chlorate ion would
be expected to partition predominantly into water and to be highly mobile in water. However,
under most environmental conditions, chlorate is subject to redox reactions. Factors affecting
oxidation and reduction in soil and water include temperature, pH, chlorate concentration, the
nature and concentration of reductants, and the degree of moisture in soils. Alkaline conditions
favor chlorate stability (USEPA, 2006a). As a strong oxidizing agent, chlorate is typically
reduced to chlorine species in lower oxidation states, such as chloride. In the environment,
extensive redox reactions are expected to reduce the concentration of chlorate in the water
column (USEPA, 2006a).
2.4 Regulatory and Non-Regulatory Actions for Chlorate
Some domestic and foreign agencies have established regulatory actions or non-regulatory
advisories to address chlorate contamination of drinking water.
In 2002, the State of California proposed an action level of 200 ng/L for chlorate as a drinking
water contaminant, based on a 20 percent relative source contribution (RSC) (CalEPA, 2002).
Later, in 2007, the state set a notification level of 800 ng/L, based on an 80 percent RSC (CDPH,
2007). The State of California maintained this level in 2015, with a note that chlorate may be
produced during the disinfection process (CalEPA, 2015). Health Canada adopted an individual
maximum acceptable concentration (MAC) (an enforceable standard) of 1 mg/L for chlorate and
chlorite in 2008, based on an 80 percent RSC from drinking water (Health Canada, 2008). Using
this MAC, NSF/ANSI60 set a Single Product Allowable Concentration of 0.3 mg/L chlorate for
products used in drinking water treatment (Stark, 2013).
The American Water Works Association (AWW A) conducted an assessment of the potential
regulatory implications of chlorate in the United States in 2014, based on a health reference level
of 210 |ig/L (AWW A, 2014). AWWA recommended that, even with uncertainties about
potential future regulatory actions that might relate to chlorate, water systems might want to
consider taking steps to better understand the levels of chlorate in their drinking water. The
World Health Organization (WHO) set provisional guideline values (voluntary standards) of 0.7
mg/L each for chlorate and chlorite in 2005, using an 80 percent RSC, and requested public
comments in 2015 (WHO, 2005, 2015). China adopted the WHO guideline values as its
standards (Wang et al., 2015a).
Six-Year Review 3
Technical Support Document for Chlorate
2-7
December 2016
-------
3 Health Effects
This chapter presents a summary of chlorate health effects and derivation of the health reference
level (HRL). As noted in Chapter 1, information about potential common health effects of
chlorate and chlorite, as well as the disinfectant chlorine dioxide, is presented in the Six-Year
Review 3 Technical Support Document for Disinfectants and Disinfection Byproducts Rules
(D/DBPRs) (USEPA, 2016a).
3.1 Summary of Health Effects
Human oral data are available for chlorate from reports of poisoning incidents and clinical
studies. Doses of >100 mg/kg of sodium chlorate are generally fatal in humans (USEPA, 2006b).
The smallest dose of sodium chlorate reported to be fatal was 7,500 mg (107 mg/kg for a 70 kg
adult), with two reports noting 10,000 mg (143 mg/kg for a 70 kg adult) as a fatal dose, and one
report observing that "vigorous treatment saved one person who had ingested about 40,000 mg"
sodium chlorate (USEPA, 2006b). Toxic doses of sodium chlorate can cause gastrointestinal
irritation, hemolysis, methemoglobinemia, hemoglobinuria, disseminated intravascular
coagulation, cyanosis and renal failure (WHO, 2005; USEPA, 2006b; Lee et al., 2013).
In a controlled clinical evaluation of chlorate (Lubbers et al., 1982, 1984), subjects ingesting a
liter of water containing 0.01- 2.4 mg chlorate every third day for 16 days showed small but
statistically significant changes in group means for serum bilirubin, iron and methemoglobin,
which were within the normal range for each parameter.
Khan et al. (2005) conducted a short term study in male F344 rats that received daily doses of
sodium chlorate for seven consecutive days. There was a dose-related decrease in the thyroid
gland stores of thyroglobulin, the protein from which thyroid hormones are synthesized. The no
observed adverse effect level (NOAEL) for this effect was 2.60 mg/kg/day chlorate and the
lowest observed adverse effect level (LOAEL) was 12.3 mg/kg/day. Thyroid hormones have a
shorter half-life in humans than in rats, making the rat more sensitive (Dohler et al., 1979).
However, this study raises concerns for thyroid effects from short term chlorate exposures,
especially when there is co-exposure to perchlorate. In the Khan et al. (2005) study, a mixture of
1.2 mg/kg/day chlorate combined with 0.9 mg/kg/day perchlorate resulted in both colloid
depletion and a significant decrease in serum thyroxine. Chlorite is also associated with effects
on thyroid hormones, with NOAELs that are higher (20 mg/kg/day, 30 mg/kg/day) than those
seen for chlorate (Bercz et al., 1982; Orme et al., 1985). No mixture study of chlorate and
chlorite on thyroid hormones was identified.
The major effects from subchronic and chronic exposure to sodium chlorate in animals are on the
blood and thyroid. Subchronic studies in rats have reported decreased hemoglobin, hematocrit
and red blood cell (RBC) counts (Abdel-Rahman et al., 1984; Barrett, 1987; McCauley et al.,
1995). Severe thyroid colloid depletion, follicular cell hypertrophy and hyperplasia were
reported in rats after 90-day exposures (Hooth et al., 2001). A chronic study (NTP, 2005)
reported thyroid follicular cell hypertrophy and mineralization, as well as hyperplasia of the bone
marrow in rats and hyperplasia of the bone marrow and granulosa cell hyperplasia of the ovary in
Six-Year Review 3
Technical Support Document for Chlorate
3-1
December 2016
-------
mice. These animal studies provide clear and consistent evidence that subchronic and chronic
exposures to chlorate result in effects on the blood and thyroid.
The only long-term carcinogenicity study of chlorate in animals is a 2-year bioassay on sodium
chlorate in drinking water in rats and mice (NTP, 2005). The National Toxicology Program
(NTP, 2005) exposed male and female rats to 0, 125, 1,000, or 2,000 mg/L and male and female
mice to 0, 500, 1,000, or 2,000 mg/L sodium chlorate for 2 years and concluded that there was 1)
some evidence of carcinogenicity in male and female rats based on an increased incidence of
thyroid gland neoplasms, 2) equivocal evidence of carcinogenicity in female mice based on
marginally increased incidences of pancreatic islet neoplasms and 3) no evidence of
carcinogenicity in male mice.
The chronic NTP (2005) study was identified as the critical study for establishing a reference
dose (RfD) of 0.03 mg/kg/day for chlorate (USEPA, 2006b). The RfD was derived by using the
Benchmark Dose (BMD) method and based on a Benchmark Dose Level (BMDL) of 28 mg/L as
sodium chlorate (22 mg/L as chlorate) for increased follicular cell hypertrophy as the critical
effect. The 22 mg/L concentration corresponds to a dose of 0.9 mg/kg/day for chlorate ion
(USEPA, 2006b). A net uncertainty factor (UF) of 30 was applied when deriving the RfD. This
consisted of a UF of 10 for inter-human variability for potentially sensitive individuals in the
absence of information on the variability of response in humans and a UF of 3 for interspecies
uncertainty because there is increased activity of the thyroid-pituitary axis in rats (Dohler et al.,
1979; McClain, 1992) modulating the applicability of the thyroid effects in rats when
extrapolated to humans. A UF of 1 was assigned for LOAEL-to-NOAEL adjustment because the
BMDL approach was used to set the RfD; a UF of 1 for subchronic to chronic extrapolation
because a chronic study was used; and a UF of 1 for database uncertainties because the database
of chlorate includes subchronic, chronic, developmental and reproductive studies.
Sodium chlorate is classified as "not likely to be carcinogenic to humans at doses that do not
alter thyroid hormone homeostasis" (USEPA, 2006b). This classification is in accordance with
the EPA policy, Assessment of Thyroid Follicular Cell Tumors (USEPA, 1998), which states
that nonmutagenic pesticides that induce elevated levels of thyroid-stimulating hormone (TSH)
and follicular cell tumors in rats are classified as not likely to be carcinogenic to humans at doses
that do not alter thyroid hormone homeostasis (USEPA, 2006b). Sodium chlorate is considered
to be nonmutagenic based on negative results in most in vitro and in vivo gene mutation assays,
including gene mutation tests in bacteria and Chinese hamster lung cells, and tests of micronuclei
and chromosomal damage in mouse bone marrow (USEPA, 2006b). A quantitative cancer risk
assessment was not conducted for chlorate because sodium chlorate is classified as likely to be
carcinogenic to humans at doses that disturb thyroid homeostasis but not likely at doses that do
not. Thus, protection provided by the RfD will also be protective for cancer.
Six-Year Review 3
Technical Support Document for Chlorate
3-2
December 2016
-------
EPA also evaluated whether health information is available regarding potential effects on
children and/or other sensitive populations. There was no pre- or post-natal sensitivity or
susceptibility observed in developmental studies of sodium chlorate in rats and rabbits, including
a two-generation reproduction study in rats. However, evidence suggests that there may be a
concern for developing offspring because of the effects of inorganic chlorate on thyroid function
in rats (USEPA, 2006b). Chlorate is one of a number of inorganic ions that can interfere with
iodine uptake by the thyroid, but chlorate is not highly potent in this respect (Van Sande et al.,
2003).
Chlorate is able to cause hemolysis at doses greater than the RfD. Thus, persons with low RBC
counts, such as those with anemia, may be particularly sensitive to sodium chlorate. However, it
is not clear whether the fetus or newborn is more sensitive to the hemolytic effect of chlorate
than adults (CalEPA, 2002) because of age alone. Data from the 1994 National Health Interview
Survey (O'Day et al., 1998) indicate that there were about 5 million people in the U.S. who
suffered from some form of anemia. About 3 to 5 percent of the population may have an
inherited glucose-6-phosphate dehydrogenase (G6PD) deficiency increasing their risk for
methemoglobinemia, with males more sensitive than females (Luzzatto and Mehta, 1989).
Additionally, about 1 percent may have a form of hereditary methemoglobinemia (Jaffe and
Hultquist, 1989). Each one of these conditions is a contributor to low RBC counts within the
population, which renders them more sensitive to chlorate than the general population.
Individuals co-exposed to other ions that decrease iodine uptake by the thyroid (e.g., perchlorate)
or cause methemoglobinemia and low RBC counts (e.g., nitrate or nitrite) could be more
sensitive to chlorate exposure (Khan et al., 2005) than the general population.
3.2 Derivation of the Health Reference Level
To evaluate the systems and populations exposed to chlorate in drinking water from public water
systems (PWSs), monitoring data were compared to a concentration in drinking water that is
termed the health reference level (HRL). The HRL is a risk-derived concentration against which
to compare the occurrence data from PWSs to determine if chlorate occurs with a frequency and
at levels of public health concern. HRLs are not final determinations about the level of a
contaminant in drinking water that is necessary to protect any particular population and they are
derived prior to development of a complete exposure assessment.
EPA calculated a long-term non-cancer HRL of 210 |ig/L for chlorate, using the RfD of 0.03
mg/kg/day for a 70-kg adult ingesting 2 L of drinking water per day and a default relative source
contribution (RSC) of 20 percent (USEPA, 2014a). The agency anticipates evaluating health
effects related to short-term exposures as part of potential future regulatory actions. EPA derived
the HRL for chlorate using the RfD approach as follows:
Six-Year Review 3
Technical Support Document for Chlorate
3-3
December 2016
-------
HRL (mg/L) = [(RfD x BW)/DWI] • RSC
Where:
RfD = Reference Dose (mg/kg/day)
BW = Body Weight for an adult, assumed to be 70 kilograms (kg); for a child, assumed
to be 10 kg
DWI = Drinking Water Intake for an adult, assumed to be 2 L/day (90th percentile); for a
child, assumed to be lL/day (90th percentile)
RSC = Relative Source Contribution, or the level of exposure believed to result from
drinking water when compared to other sources (e.g., food, ambient air). In the
absence of a complete exposure assessment, a default RSC value is used in the
calculation of the HRL. Default values are based on the Exposure Decision Tree
(USEPA, 2000). 20 percent is the most conservative RSC used in the derivation
of a maximum contaminant level goal (MCLG) for drinking water.
Chlorate is introduced into the food supply when tap water containing chlorate is used for food
preparation, when crops are treated with sodium chlorate as an herbicide, and when chlorine
dioxide and/or hypochlorites are used as disinfectants by the food industry (USEPA 2006a,
2006b; WHO, 2008; Asami et al., 2013).
The RfD for chlorate is protective against acute alterations in thyroid homeostasis and, therefore,
considered to also be protective of tumorigenicity as well as other chronic and subchronic
adverse health effects discussed in the literature (Hooth et al., 2001; Khan et al., 2005; NTP,
2005).
3.2.1 Considerations of Relative Source Contribution (RSC) from Drinking Water for
Chlorate
The following data sources could be useful in deriving an RSC for chlorate following the EPA
decision tree approach (USEPA, 2000):
• The Office of Pesticide Programs Reregi strati on Eligibility Decision for inorganic
chlorates as pesticides applied to a variety of crops and in antimicrobial applications, and
related documentation (USEPA, 2006a, 2006b, 2016c);
• Monitoring data from the Third Unregulated Contaminant Monitoring Regulation
(UCMR 3) from the water treatment plants and within the distribution system;
• A well designed total diet study that analyzed the levels of chlorate in foods prepared
with distilled water and in foods prepared with tap water containing a known amount of
inorganic chlorate (Asami et al., 2013). Although the study was carried out in Japan, it is
possible to harmonize the data using food group consumption data from the U.S.,
Continuing Survey of Food Intakes by Individuals in the United States (CSFII);
Six-Year Review 3
Technical Support Document for Chlorate
3-4
December 2016
-------
• A published study on the levels of chlorate in dietary supplements and flavor enhancers
(Snyder et al., 2006).
3.3 Additional Perspective on Chlorate in DBP Mixtures from Epidemiology Studies
Righi et al. (2012) conducted a case-control study in Northern Italy to investigate the relationship
between drinking water exposure to chlorite, chlorate and trihalomethanes (THMs) and
congenital anomalies. A total of 1,917 cases of congenital anomalies (neural tube, cardiac,
diaphragm and abdominal wall, esophagus (food pipe or gullet), cleft lip and palate, respiratory,
urinary tract and chromosomal anomalies) observed in the period of 2002 to 2005 were studied.
The THM exposure levels were reported to be very low (mean 3.8 + 3.6 ng/L), and no excess
risk of anomalies were associated with THM exposures. The levels of chlorite (mean 427 + 184
Hg/L) and chlorate (mean 283 + 79 |~ig/L), however, were relatively high. The authors reported
that women exposed to chlorite at levels > 700 |ig/L were at higher risk of having newborns with
renal defects (OR: 3.30; 95 percent CI: 1.35-8.09), abdominal wall defects (OR: 6.88; 95 percent
CI: 1.67-28.33) and cleft palate (OR: 4.1; 95 percent CI: 0.98-16.8); women exposed to chlorate
at levels >200 jag/1 were at higher risk of newborns with obstructive urinary defects (OR: 2.88;
95 percent CI: 1.09-7.63), cleft palate (OR: 9.60; 95 percent CI: 1.04-88.9) and spina bifida (OR:
4.94; 95 percent CI: 1.10-22). The authors noted that this was the first study showing an excess
risk of different congenital anomalies associated with chlorite and/or chlorate exposure from
drinking water, and that further research using larger datasets was needed to confirm the
observed results.
In an earlier population-based, case-control study from the same area, Aggazzotti et al. (2004)
examined the association between chlorination byproducts and adverse pregnancy outcomes. The
chlorination byproducts investigated in this study were chlorate and chlorite and total and
individual THMs: chloroform, dichlorobromomethane, dibromochloromethane and bromoform.
A total of 1,194 subjects were evaluated in the study, consisting of 343 pre-term (<37 weeks)
births, 239 full-term small for gestational age (SGA) births (< 10th percentile of birth weight
according to standard values from the Italian Society of Pediatrics) and 612 controls (born > 37
weeks and > 10th percentile of birth weight). Exposure was assessed both by a questionnaire
completed by the mothers on their personal habits during pregnancy and by water samples
collected at the homes of the participants. The median concentrations of chlorate for pre-term
births, full-term births and controls were: 76.20, 72.0 and 76.5 (J,g/L, respectively. No association
was found between pre-term births and exposure to chlorate or to any of the other chlorination
byproducts studied. For a subgroup of 59 term-SGA cases and 113 controls having "high
exposure" to THMs (>30 (J,g/L), chlorite (> 200 (J,g/L) or chlorate (> 200 (J,g/L), a weak
association was found (OR: 1.38; 95 percent CI: 0.92-2.07). However, separate analyses for
exposure to high levels of THMs, chlorite or chlorate individually showed a relationship between
term-SGA and high chlorite exposure but not for high THM or high chlorate exposure (the
authors note that there was a small number of subjects exposed to high levels of chlorate that
could be a limitation for that analysis).
Six-Year Review 3
Technical Support Document for Chlorate
3-5
December 2016
-------
4 Analytical Methods
EPA has developed four analytical methods that are available for the analysis of chlorate in
drinking water:
• EPA Method 300.0, Revision 2.1, Determination of Inorganic Anions by Ion
Chromatography, reported a Method Detection Limit (MDL) of 3 |ig/L. Reagent water
and finished drinking water samples fortified with 0.05 to 5 |ig/L chlorate yielded
recoveries that ranged from 97 to 121 percent (USEPA, 1993);
• EPA Method 300.1, Revision 1.0, Determination of Inorganic Anions in Drinking Water
by Ion Chromatography, reported MDLs that range from 0.78 to 2.55 |ig/L. Reagent
water and finished drinking water samples fortified at 100 and 500 |ig/L chlorate yielded
recoveries that ranged from 86.1 to 106 percent, and percent Relative Standard
Deviations (percent RSDs) of 0.47 to 2.14 percent (USEPA, 1997);
• EPA Method 317.0, Revision 2.0, Determination of Inorganic Oxyhalide Disinfection By-
products in Drinking Water Using Ion Chromatography with the Addition of a
Postcolumn Reagent for Trace Br ornate Analysis, reported MDLs that range from 0.62 to
0.92 |ig/L. Reagent water and finished drinking water samples fortified at 100 and 500
|ig/L chlorate yielded recoveries that ranged from 86.1 to 106 percent, and percent RSDs
of 0.47 to 2.14 percent (USEPA, 2001). Note that the recovery and RSD data are
identical to the recovery and RSD data from EPA Method 300.1;
• EPA Method 326.0, Revision 1.0, Determination of Inorganic Oxyhalide Disinfection By-
products in Drinking Water Using Ion Chromatography Incorporating the Addition of a
Suppressor Acidified Postcolumn Reagent for Trace Bromate Analysis, reports a
Detection Limit (DL) of 1.7 |ig/L. Reagent water and finished drinking water samples
fortified at 100 and 500 |ig/L chlorate yielded recoveries that ranged from 99.0 to 111
percent, and percent RSDs of 0.66 to 2.8 percent (USEPA, 2002).
ASTM International Method D6581-08 and Standard Method (SM) 4110 D are two additional,
for-purchase voluntary consensus standard organization analytical methods. Both methods were
approved for chlorate monitoring under the third cycle of the Unregulated Contaminant
Monitoring Rule (UCMR 3). ASTM D6581-08 has an operational range for chlorate of 5-500
Hg/L (using chemically-suppressed ion chromatography) and an operational range for chlorate of
20-1,000 |j,g/L (using electrolytically suppressed ion chromatography). ASTM International
indicates that Method D6581-08 is "technically equivalent with Part B of U.S. EPA Method
300.1" (ASTM, 2008). ASTM Method D6581-08 reported an MDL for the electrolytic
suppression portion of the method of 0.32 |ig/L in reagent water. Finished drinking water
samples from eight ground water or surface water sources fortified at 20, 25, 180, 220, 400 and
450 |ig/L chlorate yielded mean recoveries that ranged from 93 to 107 percent for the electrolytic
suppression portion of the method.
Six-Year Review 3
Technical Support Document for Chlorate
4-1
December 2016
-------
SM 4110 D in the 21st edition of SM, published in 2005, was approved for monitoring chlorate
under UCMR 3 (USEPA 2012; 77 FR 26072). The data reviewed was obtained from SM 4110 D
in the 22nd edition of SM (SM, 2012). The reported MDL, fortified reagent water and fortified
finished drinking water recoveries, and percent RSDs for SM 4110 D are identical to those
published by EPA in EPA Method 300.1 (USEPA, 1997).
Although not listed in any of the EPA methods, the Minimum Reporting Level (MRL) for
chlorate was established at 20 |ig/L and served as a national benchmark for laboratories that
participated in UCMR 3 using EPA Method 300.1 (USEPA, 2012; 77 FR 26072).1
Estimated reporting levels for EPA Method 300.0, Rev. 2.1; 317.0, Rev. 2.0; and 326.0, Rev. 1.0
(calculated as five times the MDL or DL) and the MRL for EPA Method 300.1, Rev. 1.0 were
compared to the Health Reference Level (HRL) for chlorate to determine whether the available
analytical methods are capable of reliable quantitation at concentrations of estimated
toxicological concern (see Exhibit 4.1). The Method Sensitivity Ratio (MSR) is calculated from
the following equation:
MSR = HRL (|ig/L) / MRL or 5x the MDL or DL (|ig/L)
A favorable MSR is one that is greater than ten. That is, it is preferable that the HRL be at least
ten times above the concentration at which data can be reliably reported; this provides a margin
of safety for uncertainty in the HRL and/or method performance (USEPA, 2009a).
Exhibit 4.1: Method Sensitivity Ratios (MSRs) for Chlorate
Method
MDL or DL
(M9/L)
5x the MDL or
DL (ng/L)
MRL (ng/L)
HRL (ng/L)
MSR
300.0, Rev. 2.1
3
15
-
210
14
300.1, Rev. 1.01
-
-
20
210
10.5
317.0, Rev. 2.0
0.62-0.92
3.1-4.6
-
210
45.6-67.7
326.0, Rev. 1.0
1.7
8.5
-
210
24.7
1 Since ASTM D6581-08 and SM 4110 D are based on EPA Method 300.1, the MSRs for these two methods are
anticipated to be similar to the MSR calculated for EPA Method 300.1.
For all of the methods tabulated in Exhibit 4.1, the MSRs are greater than ten; hence, the
available analytical methods should be capable of reliable quantitation at least ten times below
the HRL. Note that EPA Method 317.0, Rev. 2.0 has performance options that result in higher
MDLs; however, use of the other method options also results in favorable MSRs.
1 At the time of Rule publication, then-most-recent versions of ASTM D6581-08 and SM 4110 D (21st edition) were
listed as allowed alternative methods to EPA Method 300.1 for UCMR 3 monitoring (USEPA, 2012).
Six-Year Review 3
Technical Support Document for Chlorate
4-2
December 2016
-------
Additional potential analytical methods that are not approved by EPA for the analysis of drinking
water for chlorate have been identified in the literature. In particular, two methods that are based
on liquid chromatography and tandem mass spectrometry (LC-MS/MS) have been evaluated
relative to EPA Method 300.1. Stanford et al. (2013) documents a comparison of four
laboratories performing oxyhalide analyses using three analytical methods. Two of the
laboratories utilized EPA Method 300.1, one laboratory utilized the LC-MS/MS method of Li
and George (2005) (presumably adapted for the determination of chlorate), and the fourth
laboratory utilized the LC-MS/MS method of Pisarenko, et al. (2010). As reported in Stanford et
al. (2013), the LC-MS/MS method of Li and George (2005) demonstrated somewhat lower
recoveries of chlorate than those obtained from two laboratories utilizing EPA Method 300.1.
The fourth laboratory, using the LC-MS/MS method of Pisarenko et al. (2010), obtained chlorate
recoveries that were more comparable to the results from the two laboratories that utilized EPA
Method 300.1, although the results using the method of Pisarenko et al. (2010) were sometimes
biased slightly high. The lack of a stable, isotopically-labelled chlorate standard (i.e., Cl1803")
and a resultant sensitivity to matrix interferences were cited as the reasons for the low bias
observed in the results from the method of Li and George (2005) (Stanford et al., 2013).
Six-Year Review 3
Technical Support Document for Chlorate
4-3
December 2016
-------
5 Occurrence and Exposure in Drinking Water
This section presents and discusses information about the occurrence and exposure to chlorate in
drinking water from public water systems (PWSs). The best data available for chlorate
occurrence in finished drinking water in PWSs are from the nationally representative monitoring
completed under the third round of the Unregulated Contaminant Monitoring Rule (UCMR 3).
The UCMR 3 monitoring provides nationally representative contaminant occurrence data for
chlorate and other contaminants in the United States. The UCMR 3 program took place from
2012 to 2016. Most of the monitoring was conducted between 2013 and 2015. Gathering and
reporting of some data continued in 2016. Additional sources of information about the
occurrence of chlorate include the EPA's 1996 Disinfection Byproducts Rule Information
Collection Rule (DBP ICR) (USEPA, 1996; 61 FR 24353), EPA's Community Water System
Survey (USEPA, 2009b) and the Environmental Working Group Drinking Water Database
(EWG, 2015). These additional data sources are summarized in Appendix A of this document.
As indicated in Chapter 2, chlorate may be released to the environment from commercial
production and use and other sources. National data on chlorate occurrence in ambient water are
not available. Limited chlorate ground water monitoring data from various sources are reported
in the federal government's Water Quality Portal (http://vvvvvv.vvaterciualitvdata.us/portal/).
Because these data are sparse and expected to be of mixed quality, EPA did not review them in
detail for the Third Six-Year Review effort. As presented and discussed below, chlorate
occurrence in undisinfected ground water (GW) systems could imply chlorate contamination in
the source water.
As described in Chapter 3, a Health Reference Level (HRL) of 210 |ig/L was calculated for
chlorate based on a chronic study for long-term non-carcinogenic effects, identified as the critical
study. Occurrence data in finished drinking water from the UCMR 3 presented below are
compared to the HRL, twice the HRL (420 |ig/L) and three times the HRL (630 |ig/L). As
appropriate, estimates of the population exposed at concentrations above these thresholds are
also presented.
National occurrence and exposure are estimated and discussed for both individual sampling
results and locational averages. Note that the average concentrations are more relevant than the
levels in individual samples in an evaluation of the long-term exposure effects of chlorate. Thus,
in order to characterize national occurrence of and exposure to chlorate, EPA used the UCMR 3
data to calculate the average concentrations at each location (both the entry points (EPs) and the
location of maximum residence (MR) time within the distribution system). Appendix B.2
presents the analytical results based on individual chlorate samples. The UCMR 3 data are also
analyzed in order to understand: (1) chlorate occurrence by disinfectant type and system size, (2)
spatial variation of chlorate occurrence from the EPs to the MRs and (3) changes in chlorate
occurrence over time. Appendix B.4 contains the analytical results pertaining to the disinfectants
used nationally during the UCMR 3 monitoring period. Appendix B.5 discusses the analyses of
UCMR 3 data conducted to improve the understanding of the impact of changes in disinfectant
types nationally. For information on the co-occurrence of chlorate and chlorite, refer to the Six-
Year Review 3 Technical Support Document for Disinfectants/Disinfection Byproducts Rules
(USEPA, 2016a).
Six-Year Review 3
Technical Support Document for Chlorate
5-1
December 2016
-------
5.1 UCMR 3 Monitoring Program and Dataset
The purpose of EPA's unregulated contaminant monitoring program is to collect data on the
occurrence of contaminants suspected to be present in drinking water, but that do not have
established health-based national standards under the Safe Drinking Water Act. UCMR 3
monitoring, conducted between 2013 and 2016, provides the data for the chlorate occurrence
analysis presented in this section. The latest version of this dataset is available from the
Agency's website (https://www.epa.gov/dwucmr/occurrence-data-unregulated-contaminant-
monitoring-rule).
Similar in design to UCMR 1 and UCMR 2, UCMR 3 involves multiple tiers of monitoring:
Assessment Monitoring for contaminants with commonly used analytical method technologies
(including methods for chlorate), Screening Survey monitoring for contaminants that require
specialized analytical method technologies not in wide or common use, and Pre-Screen Testing
for contaminants that require analysis with methods that use new or specialized technology.
Chlorate was part of the Assessment Monitoring and underwent monitoring using EPA Method
300.1. The minimum reporting level (MRL) used for chlorate in the UCMR 3 survey was 20
Hg/L.
For UCMR 3 Assessment Monitoring, all large (serving between 10,001 and 100,000 people)
and very large (serving more than 100,000 people) community water systems (CWSs) and non-
transient non-community water systems (NTNCWSs), plus a statistically representative national
sample of 800 small PWSs (serving 10,000 people or fewer), were required to participate.
Surface water (and ground water under the direct influence of surface water (GWUDI)) sampling
locations were monitored four times during the applicable year of monitoring, and ground water
sample locations were monitored twice during the applicable year of monitoring. Monitoring for
chlorate was conducted at two types of sampling locations: the EP and MR locations. UCMR 3
also required that the participating systems indicate which MR location(s) were associated with
each EP location. (Note that in some cases, multiple EP locations could be associated with a
single MR location.) Furthermore, the UCMR 3 required PWSs to report the type of disinfectant
in use at the time of sampling for each EP and MR location and for each sampling event. See the
Federal Register (USEPA, 2012; 77 FR 26072) for more information on the UCMR 3 study
design.
The design of UCMR 3 enables estimates of national occurrence. The UCMR 3 monitoring
collects occurrence data from the survey of small systems that can be used to extrapolate national
occurrence. To calculate national extrapolations, the percent of systems (or population served)
estimated to exceed a specified threshold can be multiplied by the total number of systems (or
population served) in the nation. In UCMR 3 analysis, the extrapolation methodology is applied
only to small systems. Because all large and very large systems were required to participate in
the UCMR 3 Assessment Monitoring, the data collected by systems in these size categories
represent a census of systems and therefore directly represent national occurrence. (They do not
require national extrapolation.) Total national occurrence is then estimated by adding the
extrapolated national values of small systems to the census values of the large and very large
system size categories.
Six-Year Review 3
Technical Support Document for Chlorate
5-2
December 2016
-------
The UCMR 3 occurrence analyses presented in this report for chlorate are based on data
collected through May 2016 and released in July 2016 (USEPA, 2016e). EPA expects a
relatively small amount of data reporting to continue after July 2016. The UCMR 3 dataset will
not be considered "final" until early 2017. The final numbers will be presented and analyzed in a
future report by EPA. EPA does not anticipate that there will be any substantial difference
between findings based on the July 2016 data set and findings based on the final data set.
Exhibit 5.1 presents a comparison of the number of UCMR 3 systems expected to submit data
with the number of systems that have submitted UCMR 3 chlorate data as of July 2016. Through
July 2016, data had been received from 98 percent of systems that are expected to submit UCMR
3 chlorate data. About 94 percent of the 4,908 systems with chlorate data had submitted 100
percent of expected data as of July 2016. (Expected data include two samples per ground water
sample location and four samples per surface water sample location during a 12-month period.)
Exhibit 5.1. Comparison of the Number of Systems in the UCMR 3 Sample Design
with the Number of Systems with Chlorate Data in UCMR 3 Data
System Size
Expected Number
of UCMR 3
Systems
Systems That Have
Submitted Chlorate
Data (as of July 2016)
Percent of
Expected Systems
Small Systems (<10,000)
800
799
99.9%
Large Systems (10,001-100,000)
3,780
3,701
97.9%
Very Large Systems (>100,000)
411
408
99.3%
Total
4,991
4,908
98.3%
Source: USEPA, 2016e
It is also important to note that both CWSs and NTNCWSs monitored for chlorate under UCMR
3 and results from both system types are included in the analyses below. With both CWSs and
NTNCWSs included in the analysis, there can be concern about over-estimating exposure by
double-counting individuals who consume drinking water at home (via a CWS) and at
workplaces or schools (via NTNCWSs). Exhibit 5.2 presents a breakdown of the count of
records for these two system-type categories. A very small percentage of the chlorate data were
submitted by NTNCWSs (specifically, less than one percent of the number of samples and
approximately 2 percent of systems serving only 0.3 percent of the overall population served by
participating systems). Thus, it is not expected that the occurrence analyses below will
significantly overestimate potential exposure estimates with the inclusion of both CWS and
NTNCWS data. Additional UCMR 3 analyses based on CWS data only and NTNCWS data only
are included in Appendix B.l (i.e., Exhibit B.l through Exhibit B.4).
Six-Year Review 3
Technical Support Document for Chlorate
5-3
December 2016
-------
Exhibit 5.2. UCMR 3 Chlorate Data, By System Type
System Type1
Number
of
Samples
% of Total
Samples
Number
of
Systems
% of Total
Systems
Population
Served by
Systems
% of Total
Population
Served by
Systems
Community Water Systems
61,871
99.1%
4,805
97.9%
240,108,699
99.7%
Non-Transient Non-
Community Water Systems
543
0.9%
103
2.1%
685,183
0.3%
Total
62,414
100%
4,908
100%
240,793,882
100%
Source: USEPA, 2016e.
1 System type information was not included in the download of the UCMR 3 data in July 2016. The UCMR 3 data
were linked with information from the Safe Drinking Water Information System / Federal version (SDWIS/Fed)
database (December 2010) to identify the system type of each water system.
Note also that the July 2016 version of the UCMR 3 data (USEPA, 2016e) did not include the
source water type or the population served by each public water system identification number
(PWSID) with data. For the various occurrence analyses presented in this report, the data set
used by EPA (provided by EPA's Technical Service Center (TSC)) included the source water
type and population served information for each PWSID in the UCMR 3 sample design
inventory. This file also made it possible for EPA to assess the completeness of the July 2016
version of the UCMR 3 data included in Exhibit 5.1.
5.2 Summary of Analytical Results with UCMR 3 Data
Extrapolation of the analytical results on locational averages based on the UCMR 3 data suggests
that an estimated 25,000 sampling locations (17 percent of sampling locations nationally) serving
almost 52 million people nationally would have average chlorate concentrations above the HRL,
an estimated 10,000 sampling locations (7 percent) serving 15 million people nationally would
have average concentrations above twice the HRL and an estimated 5,100 sampling locations (3
percent) serving almost 6 million people nationally would have average concentrations above
three times the HRL.
UCMR 3 findings show that some disinfection techniques are associated with relatively high
chlorate occurrence. The disinfection techniques using bulk hypochlorite solution and on-site
generated (OSG) hypochlorite were associated with more chlorate detections (87.6 percent and
78.4 percent, respectively) than gaseous chlorine (16.3 percent). Chlorine dioxide is also
associated with elevated chlorate concentrations: 90.1 percent of samples where chlorine dioxide
was used had detectable levels of chlorate. These observations are consistent with the formation
information presented and discussed in Chapter 6 of this document. Note that chlorate was
detected in 22.3 percent of samples where no disinfectant was in use (mostly these samples were
at GW systems), thus implying chlorate contamination of source waters. Pesticides are a possible
source of source water contamination (see Chapter 2).
For a given disinfectant type (such as chlorine dioxide or hypochlorite), small systems in general
have higher chlorate levels than large systems. There are several factors that could contribute to
this trend, including higher doses of disinfectants or longer storage times in small systems.
Six-Year Review 3
Technical Support Document for Chlorate
5-4
December 2016
-------
A comparison of paired chlorate results (i.e., samples that were collected at an EP sampling
location and at a corresponding MR sampling location) shows that chlorate concentrations at MR
sampling locations tend to be slightly higher than concentrations at EP sampling locations. This
is true when all paired results are considered, and also when the subsets of paired results
associated with chlorine dioxide or with hypochlorite, respectively, are considered. More than
half of all paired samples had a higher chlorate concentration at the MR location than at the EP
location. There is wide variability in the differences between MR and EP concentrations among
the UCMR 3 systems.
The data collected between 2013 and 2016 under the UCMR 3 show approximately twice the
percentage of samples with chlorate concentrations greater than the HRL compared to samples
based on the data from the 1998 DBP ICR among common systems using chlorine dioxide or
hypochlorite.
5.3 Occurrence and Exposure Based on UCMR 3 Locational Average Concentrations
As described earlier, EPA conducted an analysis based on annual average concentrations at each
sampling location (including results from both EP and MR sampling locations) to relate the long-
term exposure in context of different chlorate thresholds (including HRL, 2xHRL, and 3xHRL).
These annual averages, based on up to one year of data per sampling location, are somewhat
similar to the locational running annual averages (LRAAs) that are the basis for compliance
monitoring for certain disinfection byproducts (DBPs) under the Microbial and Disinfection
Byproduct (MDBP) rules. However, since UCMR 3 requires participating systems to gather only
two or four samples per sampling location over the course of one year (two in the case of ground
water sampling locations, four in the case of surface water sampling locations), a "running
annual average" was not calculated. Averages were calculated for all individual sampling
locations included in the UCMR 3 dataset, regardless of whether or not sampling was complete
(e.g., ground water sampling locations with fewer than two samples or surface water sampling
locations with fewer than four samples). Non-detections were assigned a value of zero.
For the purpose of calculating population exposure, each system's population was assumed to be
equally distributed among its several sampling locations. With this assumption, the population
served by a sampling location with an average concentration of chlorate above a given threshold
is calculated by multiplying the system's total population served by the fraction of sampling
locations with an average concentration of chlorate above that threshold.
National estimates based on this LRAA analysis (Exhibit 5.3) suggests that approximately
25,000 sampling locations, serving 52 million people, may have average chlorate concentrations
above the HRL; an estimated 10,000 sampling locations, serving 15 million people, may have
average concentrations above twice the HRL; and an estimated 5,100 sampling locations, serving
6 million people, may have average chlorate concentrations above three times the HRL.
Additional analyses on locational averages based on the UCMR 3 data are included in Appendix
B.3.
Six-Year Review 3
Technical Support Document for Chlorate
5-5
December 2016
-------
Exhibit 5.3: National Estimates of Sample Locations and Associated Population
Served with Locational Average Chlorate Concentrations Greater than Threshold
Values (Based on UCMR 3 Data)
Threshold
Concentration
National Estimate of Number of
Sample Locations with Locational
Average Concentration > Threshold
(Percent1)
National Estimate (in million)
Population Served by Sample
Locations with Locational Average
Concentration > Threshold
(Percent1)
> HRL (210 |jg/L)
24,868 (16.59%)
52 (17.43%)
> 2xHRL (420 pg/L)
10,168 (6.78%)
15 (5.06%)
> 3xHRL (630 pg/L)
5,124 (3.42%)
6 (2.00%)
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
1 The estimated percentages of the national population exceeding the thresholds shown in this table are slightly
different from the percentages of the UCMR 3 sample population exceeding the thresholds (which are shown in
Exhibit B. 16 through Exhibit B.18), reflecting the fact that the small systems are only a sample whereas the larger
systems are taken as a census. These percentages are calculated by dividing the nationally estimated/extrapolated
count of systems/sampling locations/population served with threshold exceedances by the national inventory number
of systems/sampling locations/population served.
5.4 Occurrence by Disinfectant Type
As discussed in Section 2.2.2 (above) and Chapter 6 of this document, chlorate occurrence is
expected to vary among PWSs depending on disinfectant types employed. The UCMR 3
database provides a comprehensive set of information available about disinfectant use in the U.S.
Under the UCMR 3, disinfectant type is identified for specific monitoring locations (EP or MR)
rather than at a system-level. The disinfectant type for a given monitoring period was specific to
that monitoring location rather than to the system as a whole. As such, inferences about system-
level disinfectant use for a given year may tend to overestimate use of a type of disinfectant in
situations where that disinfectant was only used for a portion of the UCMR 3 monitoring
program.
Disinfectant type codes in the UCMR 3 database enable EPA to evaluate patterns in chlorate
occurrence. The following eleven disinfectant designation codes are used in the UCMR 3
database. If more than one disinfectant is used simultaneously, a sample may be associated with
more than one disinfection designation. Disinfection designations may also vary from one
sample to another at the same sampling location if, for example, disinfection practices change
over time.
• CLGA (gaseous chlorine),
• CLOF (off-site generated hypochlorite stored as liquid),
• CLON (on-site generated hypochlorite with no storage),
• CAGC (chloramine formed from gaseous chlorine),
• CAOF (chloramine formed from off-site hypochlorite),
Six-Year Review 3
Technical Support Document for Chlorate
5-6
December 2016
-------
• CAON (chloramine formed from on-site hypochlorite),
• CLDO (chlorine dioxide),
• OZON (ozone),
• ULVL (ultraviolet light),
• OTHD (all other types of disinfectant),
• NODU (no disinfection).
Off-site generated hypochlorite liquid is frequently referred to as "bulk" hypochlorite solution.
The abbreviation "OSG" is sometimes used to refer to on-site generated hypochlorite. Those
conventions will be followed in this document.
Note that the disinfection data in the database are "as reported" and did not undergo independent
verification or quality assurance review. There are numerous blank disinfectant designation
fields: approximately 18 percent of chlorate samples in the UCMR 3 database (July 2016
version) are associated with no disinfectant designation. In addition, more than 250 surface water
chlorate samples were reported as using no disinfection (NODU); this is presumably due to
incorrect self-reporting, as disinfection is required for all surface water systems. These NODU
records came from 52 different water systems.
Exhibit 5.4 through Exhibit 5.8 present results (including reported detections, exceedances of the
HRL (210 ng/L), exceedances of twice the HRL (420 ng/L) and exceedances of three times the
HRL (630 ng/L)) in individual chlorate samples under various disinfection scenarios. Exhibit 5.4
compares results under three exclusive chlorination scenarios (gaseous chlorine, OSG
hypochlorite and bulk hypochlorite solution). Exhibit 5.5 presents a comparison between results
under the three chlorination scenarios and results under three parallel chloramination scenarios.
Exhibit 5.6 presents results for all samples associated with the chlorine dioxide code, as well as
chlorine dioxide in combination with each of the three forms of chlorination. In a similar format,
Exhibit 5.7 and Exhibit 5.8 present results for ozonation and ultraviolet light, respectively.
Exhibit 5.8 also presents results associated with any "other disinfectant" and results clearly
designated as associated with no disinfectant.
To generate the counts of EPs and MRs in Exhibit 5.4 through Exhibit 5.9 (as well as Exhibit
B. 19 through Exhibit B.23), two data tables from the UCMR 3 data set ("UCMR3 A11" and
"UCMR3 DRT") were linked. The "UCMR3 A11" table contains all of the UCMR 3 data (i.e.,
sample analytical results as of July 2016); the "UCMR3 DRT" table contains the disinfectant
residual type information for all of the UCMR 3 results (as of July 2016). These two tables were
linked on the public water system identification code, facility identification code, sample point
identification code, sample event code, and sample collection date (i.e., PWSID, FacilitylD,
SamplePointID, SampleEventCode, and CollectionDate) to identify the disinfection residual
types at each sampling location. Note that there were situations where no disinfectant residual
type was specified for a particular water system. There were also situations where more than one
disinfectant residual type was specified for a particular sampling location / sample event. Such
Six-Year Review 3
Technical Support Document for Chlorate
5-7
December 2016
-------
cases were consolidated so that there were multiple disinfection codes corresponding to a unique
sampling location / sample event. Thus, multiple disinfection codes in Exhibit 5.4 through
Exhibit 5.9 (as well as Exhibit B.19 through Exhibit B.23), could reflect simultaneous use of
multiple disinfectants or changes in disinfection practices over time.
In the exhibits below, a graphical system is used in the column headings to clarify which
combinations of disinfectant codes correspond to each category of interest. In a given category,
each of the eleven disinfectant codes either must be associated with sampling locations in the
category, or may (optionally) be associated, or must not be associated. (See descriptions at the
bottom of Exhibit 5.4.) Note that in some of the tables, the three codes for chlorination (i.e.,
CLGA, CLOF and CLON) are treated as a unit, as are the three codes for chloramination (i.e.,
CAGC, CAOF and CAON).
UCMR 3 findings presented in Exhibit 5.4 through Exhibit 5.8 confirm that some disinfection
techniques are associated with greater chlorate occurrence than others. Use of bulk hypochlorite
solution or OSG hypochlorite was associated with significantly more chlorate detections (87.6
percent and 78.4 percent, respectively with bulk hypochlorite solution and OSG hypochlorite)
than gaseous chlorine (16.3 percent). Chlorine dioxide is also associated with high chlorate
concentrations: 90.1 percent of samples where chlorine dioxide was in use had detectable levels
of chlorate. In addition, chlorate was detected in 22.3 percent of samples where no disinfectant
was in use mostly among GW systems (See Appendix B.4), implying chlorate contamination in
source waters.
Six-Year Review 3
Technical Support Document for Chlorate
5-8
December 2016
-------
Exhibit 5.4: Chlorate Occurrence by Form of Chlorine
Disinfectant Type
Gaseous
chlorine only
OSG
hypochlorite
only
Bulk
hypochlorite
solution only
Disinfectant Type Code
¦
L
¦
¦
r
#Measurements
15,148
2,941
16,608
#Detections (%)
2,462 (16.3%)
2,306 (78.4%)
14,544 (87.6%)
%Detections >210
12.2%
16.6%
27.5%
%Detections > 420
4.9%
3.9%
9.2%
%Detections > 630
2.6%
1.8%
4.1%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
Note: Counts include samples from both EPs and MRs. OSG = on-site generated
The disinfection codes used to categorize each sampling location are provided graphically in the table
header above each column. The legend to the right indicates what code or set of codes corresponds
to each cell. Fully shaded cells show codes that must be present for a sampling location to be
assigned to a category, and striped cells show codes that may be present. Blank cells show codes
that must not be present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-9
December 2016
-------
Exhibit 5.5: Chlorate Occurrence, Chlorination versus Chloramination
Disinfectant Type
Gaseous
chlorine only
Chloramination
from gaseous
chlorine only
OSG
hypochlorite
only
Chloramination
from OSG
hypochlorite
only
Bulk
hypochlorite
solution only
Chloramination
from bulk
hypochlorite
solution only
¦
L
Disinfectant Type Code
¦
I
¦
r
#Measurements
15,148
4,277
2,941
1,372
16,608
2,315
#Detections (%)
2,462 (16.3%)
1,435 (33.6%)
2,306 (78.4%)
1,204 (87.8%)
14,544 (87.6%)
2,111 (91.2%)
%Detections >210
12.2%
26.6%
16.6%
52.7%
27.5%
37.2%
%Detections > 420
4.9%
9.6%
3.9%
22.2%
9.2%
12.6%
%Detections > 630
2.6%
3.6%
1.8%
11.2%
4.1%
6.2%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
Note: Counts include samples from both EPs and MRs. OSG = on-site generated
The disinfection codes used to categorize each sampling location are provided graphically in the table
header above each column. The legend to the right indicates what code or set of codes corresponds to each
cell. Fully shaded cells show codes that must be present for a sampling location to be assigned to a
category, and striped cells show codes that may be present. Blank cells show codes that must not be
present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-10
December 2016
-------
Exhibit 5.6: Chlorate Occurrence when Chlorine Dioxide is in Use
Disinfectant Type
Chlorine dioxide
alone and in any
disinfectant
combination
Chlorine dioxide
in combination
with gaseous
chlorine only
Chlorine dioxide
in combination
with OSG
hypochlorite only
Chlorine dioxide
in combination
with bulk
hypochlorite
solution only
Disinfectant Type Code
#Measurements
1,884
649
54
175
#Detections (%)
1,697 (90.1%)
580 (89.4%)
52 (96.3%)
172 (98.3%)
%Detections >210
50.5%
38.4%
84.6%
87.2%
%Detections > 420
21.0%
12.6%
44.2%
60.5%
%Detections > 630
i.8%
4.0%
25.0%
23.3%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
Note: Counts include samples from both EPs and MRs. OSG = on-site generated
The disinfection codes used to categorize each sampling location are provided graphically in the table
header above each column. The legend to the right indicates what code or set of codes corresponds to
each cell. Fully shaded cells show codes that must be present for a sampling location to be assigned to a
category, and striped cells show codes that may be present. Blank cells show codes that must not be
present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-11
December 2016
-------
Exhibit 5.7: Chlorate Occurrence when Ozone is in Use
Disinfectant Type
Ozonation alone
and in any
disinfectant
combination
Ozonation in
combination with
gaseous chlorine
only
Ozonation in
combination with
OSG hypochlorite
only
Ozonation in
combination with
bulk hypochlorite
solution only
Disinfectant Type Code
H
n
u
¦
L
¦
IS
H~
p
Iff
¦
#Measurements
1,767
210
35
192
#Detections (%)
1,295 (73.3%)
26 (12.4%)
35 (100%)
187 (97.4%)
%Detections >210
25.7%
11.5%
2.9%
35.8%
%Detections > 420
9.0%
7.7%
0.0%
5.9%
%Detections > 630
4.5%
7.7%
0.0%
0.0%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
Note: Counts include samples from both EPs and MRs. OSG = on-site generated
The disinfection codes used to categorize each sampling location are provided graphically in the table
header above each column. The legend to the right indicates what code or set of codes corresponds to
each cell. Fully shaded cells show codes that must be present for a sampling location to be assigned to a
category, and striped cells show codes that must be present. Blank cells show codes that may not be
present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-12
December 2016
-------
Exhibit 5.8: Chlorate Occurrence when Other or No Disinfectants are in Use
Disinfectant Type
UV light alone and
in any disinfectant
combination
UV light in
combination with
gaseous chlorine
only
UV light in
combination with
OSG hypochlorite
only
UV light in
combination with
bulk hypochlorite
solution only
"Other
disinfectant," alone
and in any
disinfectant
combination
No disinfectant
used (at least one
NODU code, and no
other codes)
Disinfectant Type Code
SJ
1
=i
#Measurements
1,002
97
M
180
571
2,300
#Detections (%)
777 (77.5%)
16 (16.5%)
15 (83.3%)
175 (97.2%)
336 (58.8%)
512 (22.3%)
%Detections >210
21.4%
12.5%
20.0%
29.7%
17.6%
3.8%
%Detections > 420
4.5%
6.3%
0.0%
5.7%
6.3%
2.3%
%Detections > 630
2.3%
0.0%
0.0%
3.4%
3.3%
1.6%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
Note: Counts include samples from both EPs and MRs. OSG = on-site generated
The disinfection codes used to categorize each sampling location are provided graphically in the table header
above each column. The legend to the right indicates what code or set of codes corresponds to each cell. Fully
shaded cells show codes that must be present for a sampling location to be assigned to a category, and striped
cells show codes that may be present. Blank cells show codes that must not be present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-13
December 2016
-------
Since hypochlorination and chlorine dioxide are the two disinfection techniques most commonly
associated with chlorate formation, only systems using those techniques were required to gather
chlorate data under the DBP ICR. (For more information on the DBP ICR, refer to Appendix A.)
The more recent and more comprehensive UCMR 3 survey, however, shows that chlorate is also
present in other types of systems. As shown above, chlorate concentrations exceed the HRL in
finished water at 12.2 percent of gaseous chlorine systems (Exhibit 5.4) and in 8.8 percent of
systems using no disinfectant at all (Exhibit 5.8).
One possible explanation for the finding of chlorate occurrence with no disinfectant use is that
chlorate may be present in source water. Bolyard et al. (1992, 1993) found that in a handful of
cases, gas chlorinating systems had detectable levels of chlorate in both source and finished
water. Investigating further, Bolyard et al. (1993) surveyed a number of source waters and found
that chlorate was present at detectable levels at 7 out of 22 stream sites and 3 out of 8 ground
waters. (No chlorate was detected in nine reservoirs or in three mixed surface and ground water
sources.) Concentrations ranged from the reporting level (10 |ig/L) to 81 |ig/L.
Data on chlorate occurrence in ambient water could corroborate these findings. As indicated at
the beginning of Chapter 5, data on chlorate occurrence in ambient water are extremely scarce. If
chlorate is indeed present in significant quantities in source water, possible sources include
agricultural application of sodium chlorate, industrial effluent and treated wastewater (see
Section 2.2). Gorzalski and Spiesman (2015) proposed that de facto waste water reuse could be
responsible for at least some of the chlorate contamination observed in systems that monitored
under UCMR 3 and were chlorinating with gaseous chlorine. To test their theory, they examined
a cluster of systems taking water from the Tennessee River using gaseous chlorine that had high
chlorate levels. They compared the chlorate levels of systems using gaseous chlorine on the
Tennessee River to other nearby systems also using gaseous chlorine, but drawing from surface
water sources other than the Tennessee River. They found that 11 of 23 samples from the
Tennessee River had chlorate greater than 100 |j,g/L while all 14 samples from sources other than
the Tennessee River were non-detect. This led the authors to conclude that the Tennessee River
likely had elevated concentrations of chlorate.
The DBP ICR database includes influent values for chlorate which presumably represent source
water concentrations. This includes a total of 749 samples at 75 systems. Of the 749 influent
chlorate results, 80 were above the minimum reporting level of 20 ng/L. The average among
detected concentrations was 110 |_ig/L.2 The median, 90th percentile and 99th percentile of
influent reported positive chlorate results were equal to 37 ng/L, 120 ng/L and 1,442 ng/L,
respectively. These results provide further confirmation that chlorate can be present in some
source waters.
Observing the wide variability in chlorate concentrations at various points in the distribution
system in gas chlorinating and other systems (see Section 5.6), Gorzalski and Spiesman (2015)
proposed that there may be wide temporal variability of chlorate concentrations in source water.
2 Within the DBP ICR database, all the chlorate samples marked as "influent" had a sequence number of 0,
indicating that the samples were truly source water samples.
Six-Year Review 3
Technical Support Document for Chlorate
5-14
December 2016
-------
Limited data gathered by Bolyard et al. (1993) (repeated samples at a single surface water site
over the course of 450 days) do show considerable variation, repeatedly dipping below the
reporting level (10 |ig/L) and then back up to levels in excess of 20 |ig/L or 40 |ig/L.
5.5 Occurrence of Chlorate by System Size
Exhibit 5.9 summarizes UCMR 3 results for all system sizes in the two disinfection categories
most associated with chlorite formation: chlorine dioxide and hypochlorite. For comparison, a
third column is included that summarizes results for all other UCMR 3 systems (those using
neither chlorine dioxide nor hypochlorite). As the exhibit shows, very large systems have the
lowest rates of HRL, 2xHRL and 3xHRL exceedance in all three categories. In the hypochlorite
category in particular, there is a notable trend toward higher rates of exceedance at the smallest
systems.
There are several factors that could explain higher rates of chlorate occurrence at smaller systems
than at larger systems in the chlorine dioxide and hypochlorite categories. First, if smaller
systems use chlorine dioxide, there is a risk that sub-optimal operation of chlorine dioxide
generators by staff lacking specialized expertise could generate excessive chlorate. Second, if
smaller systems use hypochlorite, lower flow rates and longer storage times could lead to higher
concentrations of chlorate than observed in larger systems (since there are minimum sizes in
which hypochlorite can be purchased). Third, smaller systems might also use more concentrated
stock solutions due to relative lack of knowledge or personnel; i.e., to handle frequent dilutions
of hypochlorite which could also lead to increased formation of chlorate. Fourth, many states
require a minimum supply of chemicals be kept on hand, usually a 30-day supply (e.g., Great
Lakes Upper Mississippi River Board of State Public Health and Environmental Managers,
1997). This may lead to extended storage times for hypochlorite solutions and higher chlorate
concentrations. (See Chapter 6 below for more details.)
Six-Year Review 3
Technical Support Document for Chlorate
5-15
December 2016
-------
Exhibit 5.9: UCMR 3 Chlorate Occurrence at Systems Using Chlorine Dioxide and
Hypochlorination, by System Size
System Size
Measurements
IP
if
1
Pi
Chlorine Dioxide:
Number of
Samples
(% of Total)
Hypochlorite:
Number of
Samples
(% of Total)
All Other
Disinfectant Types:
Number of Samples
(% of Total)
Small Systems
Total Number of Samples
182
2,011
3,782
Small Systems
Detections >210 |jg/L
86 (47.3%)
655 (32.6%)
315 (8.3%)
Small Systems
Detections > 420 |jg/L
35 (19.2%)
303 (15.1%)
120 (3.2%)
Small Systems
Detections > 630 |jg/L
13 (7.1%)
153 (7.6%)
64 (1.7%)
Large Systems
Total Number of Samples
1,255
18,018
26,743
Large Systems
Detections >210 |jg/L
608 (48.4%)
4,659 (25.9%)
1,924 (7.2%)
Large Systems
Detections > 420 |jg/L
270 (21.5%)
1,601 (8.9%)
680 (2.5%)
Large Systems
Detections > 630 |jg/L
111 (8.8%)
737 (4.1%)
283 (1.1%)
Very Large Systems
Total Number of Samples
447
6,478
3,912
Very Large Systems
Detections >210 |jg/L
163 (36.5%)
1,472 (22.7%)
182 (4.7%)
Very Large Systems
Detections > 420 |jg/L
51 (11.4%)
453 (7.0%)
67 (1.7%)
Very Large Systems
Detections > 630 |jg/L
25 (5.6%)
192 (3.0%)
31 (0.8%)
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). Note: Counts include samples from both EPs and MRs.
Note: UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at all large (serving 10,001 to 100,000 people) and very large
systems (serving >100,000 people) systems in the nation.
The disinfection codes used to categorize each sampling location are provided graphically in the
table header above each column. The legend to the right indicates what code or set of codes
corresponds to each cell. Fully shaded cells show codes that must be present for a sampling location
to be assigned to a category, and striped cells show codes that may be present. Blank cells show
codes that must not be present.
Layout Key
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
5-16
December 2016
-------
5.6 Variation of Occurrence from EPs to MRs
As noted above, UCMR 3 monitoring is conducted at two sampling locations for chlorate (i.e.,
EP and MR locations). EPA investigated variations in occurrence to see whether concentrations
rose or fell in the distribution system. Paired EP and MR concentrations were studied. In most
cases only one sample was taken at each EP or MR on a particular day; if there were multiple
values for a given EP or MR location they were averaged. Non-detections were assigned a value
of zero.
Exhibit 5.10 is a scatterplot of paired EP and MR chlorate concentrations (restricted to
measurements ^ 500 |ig/L for a better resolution of the graph). This plot shows a substantial
variability between the chlorate concentrations at EP and MR locations among systems.
Exhibit 5.10: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at All UCMR 3 Sampling Locations
Chlorate Levels in MR vs. EP Locations
Based on Averages Per Date and Sampling Point)
All Disinfectant Types
Measurements < 500 |jg/L (N = 15,113)
Average EP Result (|jg/L)
Note: Based on UCMR 3 chlorate data available as of July 2016. Non-detections are assigned a value of zero. Paired
non-detections are excluded from the analysis. The blue diagonal line is not a regression line; it is a line that
represents all points with equal chlorate concentrations at the MR and the EP.
Six-Year Review 3
Technical Support Document for Chlorate
5-17
December 2016
-------
The scatterplot of EP and MR chlorate concentrations in sampling locations where chlorine
dioxide was reported to be in use (Exhibit 5.11) shows that much of the paired chlorate
concentrations lie above the blue EP = MR chlorate concentration line (i.e., they have an MR
chlorate concentration greater than the EP chlorate concentration). Note the scatterplot is
restricted to samples where the EP and MR chlorate concentrations were both less than or equal
to 500 |ig/L for a better resolution of the graph.
Exhibit 5.11: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at UCMR 3 Sampling Locations Where Chlorine Dioxide was
Reported to Be in Use
Chlorate Levels in MR vs. EP Locations
Based on Averages Per Date and Sampling Point)
Using Chlorine Dioxide
Measurements < 500 |jg/L (N = 615)
(V
q:
01
-------
Exhibit 5.12: Distribution of Paired MR and EP Locational Average Chlorate
Concentrations at UCMR 3 Sampling Locations Where Hypochlorination was
Reported to Be in Use
450
Chlorate Levels in MR vs. EP Locations
Based on Averages Per Date and Sampling Point)
Using Hypochlorite (No Chlorine Dioxide)
Measurements < 500 |jg/L (N = 9160)
• • ¦ _
• . •••; •~ **j/T
••• • • f . jfi P« •
• • •* •
D
(/)
CD
(H
CXl
Q)
CD
(0
q3
•
• •
•
• • •
•
•
•• • ¦
• •
• • ••
• m mm
• • •
m
%
V
• •
• vs
• : ••••
•: W
9
• • •• • *9
1 • •
• • • •
•
• _ € • •
I •
• •• •
1 m—m —
vjvri
• • ••
• •• • • #-
/J« ./• 4
iA is-
in •• •
„
i«li •••«¦*•• •• ^ t
' •* ••• r » • • _
•f - Ji < • •• •
¦*^*1*1 * •• •• • *
i. ii:H •".v"* •- •
!%•• • • ••• ••
rfM,« • ' t
100
200
300
•••• ••••«
400 500
Average EP Result (|jg/L)
Note: Based on UCMR 3 chlorate data available as of July 2016. Non-detections are assigned a value of zero. Paired
non-detections are excluded from the analysis. The blue diagonal line is not a regression line; it is line that represents
all points with equal chlorate concentrations at the MR and the EP.
Possible explanations for increasing chlorate concentrations in the distribution system might
include use of hypochlorite booster disinfection in the distribution system, or continuous
formation of chlorate in the presence of chlorite and free chlorine residual throughout the
distribution system (see Chapter 6 of this document for more discussion). Among systems with
multiple entry points, mixing waters with different chlorate levels can result in lower or higher
chlorate levels in the mixed water in the distribution system. Another possible mechanism for
lower chlorate levels in the distribution system could be microbial degradation. Chlorate
reducing bacteria are widely distributed in the natural environment (Logan, 1998); it is not
known whether or to what extent they may inhabit pipe biofilms.
Six-Year Review 3
Technical Support Document for Chlorate
5-19
December 2016
-------
5.7 Comparing UCMR 3 and DBP ICR Data for Occurrence
The DBP ICR (USEPA, 1996; 61 FR 24353) required all PWSs serving at least 100,000 people
to monitor and collect data on DBPs over an 18-month period from July 1997 to December 1998;
a total of 296 water systems reported data. (For more details on DBP ICR, including additional
occurrence analyses, see Appendix A.)
EPA performed a comparison of the DBP ICR (1998 results) to those from UCMR 3. Since the
DBP ICR was limited to the nation's largest water systems (as a census of systems serving over
100,000 customers), the common systems included in the analysis conducted here are all very
large systems. Only systems using chlorine dioxide or hypochlorite were included in the
analysis. System identification numbers (PWSIDs) were used to link the databases. A total of
199 systems were found to have participated and reported disinfection data in both surveys. In
the DBP ICR, 262 individual surface water plants reported data from these 199 systems. These
results were compared with data from the 342 EP locations and 238 MR location associated with
surface water source codes at the 199 systems in the UCMR 3 data set (as of July 2016). Surface
water facilities were the focus of the analysis, because they were more likely to have greater
chlorate occurrence than ground water facilities.
Of the 199 common systems, 47 reported chlorate data in both surveys. Exhibit 5.13 displays the
results of a comparison of chlorate findings from the two surveys for common systems using
chlorine dioxide or hypochlorite. The data collected under UCMR 3 between 2013 and 2016
show approximately twice the percentage of the samples with chlorate concentrations greater
than the HRL compared to the samples based on the data from the 1998 DBP ICR among
common systems. At the 2xHRL threshold, the higher percentage from the UCMR 3 survey is
even more pronounced.
There could be a number of explanations for why chlorate occurrence has increased in the time
since the DBP ICR with systems using the same disinfectant type. These may include changes in
disinfectant dose, longer solution storage times or changes in disinfection practices such as
methods of generating chlorine dioxide, methods for lowering chlorite, or shifts to hypochlorite
over chlorine gas as a secondary disinfectant with chlorine dioxide being used as a primary
disinfectant. When comparing the results of the two programs, it should be recognized that they
were conducted nearly two decades apart, and that sampling schedules also differed. In the DBP
ICR, systems that used chlorine dioxide were required to monitor for chlorate monthly, while
those using bulk hypochlorite solution monitored quarterly for chlorate and those using other
disinfectants did not monitor at all. Under UCMR 3, monitoring was conducted on a quarterly
basis at surface water sampling locations, and twice per year at ground water sampling locations.
Certain changes in disinfection practices over time, described in the following paragraphs, may
help explain the higher rates of chlorate occurrence observed at common systems in the UCMR 3
survey compared with the DBP ICR survey.
Six-Year Review 3
Technical Support Document for Chlorate
5-20
December 2016
-------
Exhibit 5.13: DBP ICR and UCMR 3 Comparison - Chlorate Occurrence in
Common Systems Using Chlorine Dioxide and Hypochlorite
Among 47
Common
Systems
Measurements
Chlorine
Dioxide:
Number of
Samples1
Chlorine
Dioxide:
Percent of
Total
Hypochlorite:
Number of
Samples2
Hypochlorite:
Percent of
Total
DBP ICR
Total Number of Samples
581
-
114
-
DBP ICR
Detections >210 |jg/L
112
19.3%
14
12.3%
DBP ICR
Detections > 420 |jg/L
13
2.2%
2
1.8%
DBP ICR
Detections > 630 |jg/L
1
0.2%
1
0.9%
UCMR 3
Total Number of Samples
168
-
203
-
UCMR 3
Detections >210 |jg/L
67
39.9%
55
27.1%
UCMR 3
Detections > 420 |jg/L
20
11.9%
13
6.4%
UCMR 3
Detections > 630 |jg/L
8
4.8%
2
1.0%
Notes: For DBP ICR counts, only used chlorate data collected between 1/1998 and 12/1998 (i.e., periods 7 through
18) and with the following "event codes" (representing distribution system samples): AVG, AVG1, AVG2, FINISH and
MAX. For UCMR 3 counts, used chlorate data collected between 1/2013 and 5/2016.
1. Chlorine dioxide counts were identified as follows: in DBP ICR, all records with M_Source_Cat = "SW" and
disinfectant type = "CLX"; in UCMR 3, all records with FacilityWaterType = "SW" and disinfectant type = "CLDO" (with
or without other disinfectants).
2. Hypochlorite counts were identified as follows: in DBP ICR, all records with M_Source_Cat = "SW" and disinfectant
type = "CL2"; in UCMR 3, all records with FacilityWaterType = "SW" and disinfectant type = "CAOF" or "CLOF"
(without other disinfectants).
Six-Year Review 3
Technical Support Document for Chlorate
5-21
December 2016
-------
6 Formation in Drinking Water
As shown and discussed in Chapter 5, the majority of chlorate detections take place among
disinfecting systems, suggesting that disinfection practices are the most common source of
chlorate in drinking water; occurrence of chlorate (as a DBP) is a function of disinfectant types
used. This chapter discusses possible pathways by which chlorate contamination can occur for
different disinfectant types. While this chapter provides supplemental information in the context
of national chlorate occurrence (as discussed in Chapter 5 of this document), it offers a scientific
basis for potential control strategies, which are discussed in Chapter 7 of this document. This
chapter is divided into the following sections:
• Chlorine dioxide,
• Bulk hypochlorite solution,
• On-site generated (OSG) hypochlorite,
• Ozone,
• Gaseous chlorine along with other disinfectant types.
6.1 Chlorine Dioxide
As described in the Agency's Alternative Disinfectants and Oxidants Guidance Manual (USEPA,
1999), chlorine dioxide has been used as a pre-oxidant and as an alternative to chlorine as a
disinfectant in drinking water treatment. Because of its incompressible, volatile nature and
explosive characteristics, chlorine dioxide cannot be shipped or stored in large quantities and is
generated in situ for drinking water treatment (USEPA, 1999; Bergmann and Koparal, 2005;
Gates et al., 2009). While use of chlorine dioxide has been effective at preventing or reducing the
formation of chlorination DBPs (including trihalomethanes and haloacetic acids) and
simultaneously achieving inactivation of pathogens, formation of chlorate (along with chlorite) is
a well-established unintended consequence. As shown by the UCMR 3 data presented in Section
5.4, systems using chlorine dioxide alone or in combination with other disinfectants show a
higher frequency of occurrence of chlorate (particularly at higher concentrations) in the finished
water than systems using other forms of disinfection (see, for example, Exhibit 5.5 and Exhibit
5.8). Similar findings on the occurrence of chlorate associated with chlorine dioxide usage are
also provided by the Disinfection Byproducts Information Collection Rule (DBP ICR) (see
Appendix A).
There are several mechanisms that result in the formation/occurrence of chlorate in drinking
water when chlorine dioxide is used. These are discussed below:
Chlorate Occurrence/Formation during Chlorine Dioxide Generation:
Gates et al. (2009) provide a comprehensive overview of generation technologies for chlorine
dioxide. There are several generation methods available. Each of these involves using sodium
chlorite or sodium chlorate along with other reactants such as hypochlorite, chlorine gas or acid.
Six-Year Review 3
Technical Support Document for Chlorate
6-1
December 2016
-------
During chlorine dioxide generation with any of these methods, chlorate occurrence/formation
can arise from impurities in the chemicals used and/or improper operational conditions
(including not using a precise chemical dose or ratio). For instance, with the generation method
based on the reaction between chlorite solution and chlorine gas, the intermediate complex of
CI2O2 can decay to chlorate at low pH; chlorate may also result from the presence of excess
hypochlorous acid (at low pH) in the mixture (Gates, 2009).
Chlorate Occurrence/Formation from Decomposition of Chlorine Dioxide:
Richardson et al. (2009) provide a comprehensive review of chlorine dioxide chemistry. Under
neutral pH conditions, chlorine dioxide solution is stable. However, in strongly acidic or basic
conditions (i.e., pH < 2 or > 11), chlorine dioxide can disproportionate to chlorate. Also, chlorine
dioxide may undergo photodecomposition to chlorate (Bolyard et al., 1993; Gallagher et al.,
1994; Bergmann and Koparal, 2005) or be present as an impurity in the chlorine dioxide
(USEPA, 2006a; Gates et al., 2009). In addition, Liu et al. (2013) found that metal oxide
particles CuO, CU2O and NiO catalyzed the disproportionation of chlorine dioxide into chlorate
and chlorite, while a-FeOOH had minimal effect. Gordon et al. (1995), however, found that
despite earlier research indicating transition metal catalysis of decomposition above pH 9, no
accelerated chlorate formation occurred in the presence of metals such as Fe2+, Ni2+, Cu2+ and
Mn2+.
Chlorate Occurrence/Formation from Conversion of Chlorine Dioxide or Chlorite:
A literature review conducted by Richardson et al. (2009) indicates that dilute solutions of
chlorine dioxide are stable under low- or zero-oxidant-demand conditions, but when chlorine
dioxide is in contact with organic or inorganic matter, chlorine dioxide rapidly degrades to
chlorate along with chlorite and chloride. Overall, chlorate levels can be up to 20 percent of the
original chlorine dioxide dose and chlorite levels can vary between 30 percent and 70 percent,
depending on conditions under which chlorine dioxide is applied during water treatment.
Chlorate levels can also be elevated from the oxidation of chlorite in the presence of free
chlorine, as observed in a finished water storage tank in a full-scale treatment plant in Barcelona,
Spain (Conio et al., 2009), and in distribution systems (Gallagher et al., 1994). In addition,
adding sulfur dioxide and the sulfite ion to reduce the concentrations of chlorine dioxide and
chlorite residuals when using chlorine dioxide (Gordon et al., 1990) can result in the formation
of chlorate (Griese et al., 1991). Sulfur dioxide and/or sulfite (as S(IV)) are thought to accelerate
the disproportionation of chlorine dioxide to produce chlorate, although other reactions are
prevalent and result in the formation of chlorite (Griese et al., 1991).
Korn et al. (2002) developed empirical equations to model the disappearance of chlorine dioxide
and the formation of chlorite and chlorate. The models were validated against measurements of
these species from water systems. The authors indicate that, on average, 68 percent by mass of
chlorine dioxide consumed becomes chlorite, while 9 percent by mass of chlorine dioxide
consumed becomes chlorate. The most important parameters cited for the formation of chlorite
and chlorate were chlorine dioxide concentration, non-purgeable organic carbon concentration
and ultraviolet (UV) absorbance at 254 nm of the aqueous solution. Such model predictions
could underestimate chlorate occurrence in the distribution systems if chlorine dioxide is used in
treatment plants as an oxidant/disinfectant and free chlorine is used in distribution systems as a
Six-Year Review 3
Technical Support Document for Chlorate
6-2
December 2016
-------
disinfectant residual, primarily due to the reaction between chlorite and free chlorine as
discussed earlier. By studying byproduct residuals in water treated by chlorine dioxide generated
with sodium chlorite and free chlorine, Gallagher et al. (1994) estimated that between 60 and 85
percent of chlorate occurrence in the distribution system was attributable to the reaction between
free chlorine residual and chlorite, and between 15 and 40 percent was associated with the
generator.
6.2 Bulk Hypochlorite Solution
As discussed earlier, commercial hypochlorite solutions (or bulk hypochlorite solution) have
become a more-commonly used form of chlorine for water treatment. One of the unintended
consequences associated with use of hypochlorite solutions is occurrence/formation of chlorate.
Nieminski et al. (1993) surveyed six water treatment plants using hypochlorite and found
chlorate concentrations up to 700 ng/L in finished water. They also observed no detection of
chlorate when gaseous chlorine was used.
Bolyard et al. (1992) examined bulk hypochlorite solutions from 14 drinking water utilities in the
United States. At a reporting limit of 10.0 |ig/L, chlorate was detected in each of the 14
hypochlorite solutions, with concentrations ranging from 180,000 to 42,000,000 |ig/L. A second
set of analyses was performed on the same solutions approximately four months later using
smaller dilution ratios, which allowed for quantitation in a range of high precision and for lower
detection/reporting levels. Chlorate was again detected in all 14 solutions, with concentrations
ranging from 190,000 to 50,000,000 |ig/L.
The same study found that the chlorate in the bulk solutions carried over into the drinking water
at all 14 utilities. Concentrations in finished water ranged from 11 |ig/L to 660 |ig/L. Three
samples exceeded the Health Reference Level (HRL) of 210 ng/L, with concentrations of 320,
600 and 660 |ig/L. The two highest concentrations also exceeded twice the HRL. Chlorate was
also detected in 2 (14 percent) of 14 source water samples (with concentrations of 20 |ig/L and
22 |ig/L) collected at the 14 drinking water utilities. In both cases, finished water samples had
higher chlorate concentrations (Bolyard et al., 1992).
A follow-up study (Bolyard et al., 1993) presented additional details and reported the results for
one additional water system using hypochlorite solution for disinfection. Chlorate was detected
in the treated water at this system at a concentration of 49 |ig/L, but was not detected in the
system's source water (at a reporting limit of 10.0 |ig/L). Overall, Bolyard et al. (1992, 1993)
detected chlorate in 15 (100 percent) of 15 treated drinking water samples disinfected with
hypochlorite solution.
For comparison, Bolyard et al. (1993) also conducted sampling at utilities that use gaseous
chlorine for disinfection rather than sodium hypochlorite. Chlorate was detected in three (11
percent) of 28 treated drinking water samples (with concentrations ranging from 17 |ig/L to 43
|ig/L) and in 2 (13 percent) of 16 source water samples (at concentrations of 17 |ig/L and 81
|ig/L), All detected concentrations of chlorate were below the HRL. The study authors
interpreted their data to indicate that use of hypochlorite solution led to chlorate formation and
the use of gaseous chlorine did not.
Six-Year Review 3
Technical Support Document for Chlorate
6-3
December 2016
-------
Gordon et al. (1993) confirmed the presence of chlorate at concentrations of 330,000 |ig/L to
15,600,000 |ig/L in hypochlorite feedstocks used at 16 utilities. (Hypochlorite concentrations in
the feedstocks ranged from 1,680,000 |ig/L to 95,900,000 |ig/L.) Chlorate was found in finished
drinking water samples from 16 (100 percent) of 16 utilities at concentrations ranging from 30
|ig/L to 300 |ig/L. Reporting levels were not identified in this study; for perspective, the
minimum detected concentration (30 |ig/L) is three times that of the reporting limit in the
Bolyard et al. (1992, 1993) studies. One of the 16 utilities had chlorate concentrations that
exceeded the HRL (and another had a concentration equal to the HRL); none of the sample
results exceeded twice the HRL. In a follow up study, Gordon et al. (1995) indicated that
chlorate concentrations in finished water were higher at facilities that used hypochlorite solutions
for disinfection (mean chlorate concentration of 0.49 mg/L) than at facilities that used chlorine
dioxide (mean chlorate concentrations of 0.25-0.29 mg/L).
Stanford et al. (2011) examined hypochlorite solutions from six utilities using either bulk
hypochlorite solution or on-site hypochlorite generation. They found chlorate concentrations
ranging from 760,000 to 19,000,000 |ig/L in the bulk hypochlorite solutions. Chlorate
concentrations in the finished water ranged from 19 to 1,500 |ig/L. Three of the seven finished
water samples had chlorate concentrations greater than the HRL. Based on the observations of
bulk hypochlorite solutions they developed a kinetic model of the decomposition of
hypochlorite. They found that hypochlorite initially decomposes to form chlorate and chloride.
The formed chlorate can then react with another hypochlorite molecule to form perchlorate. They
found the initial reaction was second order in hypochlorite concentration, and the second reaction
was first order in hypochlorite and chlorate concentration. Both reactions were dependent on
temperature and ionic strength. Their model, which was only valid for stock solutions at pH
between 11 and 13, showed good agreement with the actual decomposition (Snyder et al., 2009;
Stanford et al., 2011).
Gordon et al. (1993) showed that at the initial hypochlorite concentration of 11.5 percent,
chlorate levels in the solution appeared relatively steady (i.e., between approximately 2 and 6
g/L) at 10 degrees C over the course of 30 days of storage. However, at higher temperatures the
chlorate levels increased: e.g., to over 18 g/L over the course of 30 days at 25 degrees C. The
study also demonstrated that at a given temperature, the increasing trend was less pronounced
with lower initial hypochlorite concentrations. Gordon et al. (1995) and Stanford et al. (2011)
confirmed that the decomposition of hypochlorite solutions during storage was a probable source
of chlorate formation under the conditions of relatively high temperatures and/or initial
hypochlorite concentrations. In this formation mechanism, hypochlorite disproportionates in
basic solutions to chlorite and finally to chlorate (Bolyard et al., 1992; Bolyard et al., 1993). The
decomposition can be accelerated by sunlight (Gallagher et al., 1994). Upon a literature review
and analysis of field data, Snyder et al. (2009) and Stanford et al. (2011) identified and assessed
the key factors (besides sunlight) affecting chlorate (along with perchlorate and bromate)
occurrence/formation during the storage of hypochlorite solutions. These factors include the
dilution rates of the solution on delivery, temperature, pH, levels of transition metal ions and
storage time. Upon these reports, the American Water Works Association (AWW A, 2012)
developed a web-based predictive modeling tool (available exclusively to AWWA members)
that provides guidance on the expected levels of chlorate (along with perchlorate) in stored bulk
hypochlorite solutions. Further discussion of these individual factors and the AWWA predictive
tool is presented in Chapter 7 of this document in context of chlorate control strategies when
Six-Year Review 3
Technical Support Document for Chlorate
6-4
December 2016
-------
hypochlorite solutions are used. Overall, chlorate levels increase linearly with storage time.
Compounding this problem is the concomitant decrease of available chlorine in the hypochlorite
solutions as they age, which causes the need for the addition of larger volumes of hypochlorite to
achieve the required residual of available chlorine, thereby introducing even more chlorate into
the treated drinking water (WHO, 2005).
6.3 On-Site Generated Hypochlorite
In addition to delivery of bulk hypochlorite solution, hypochlorite can be produced on-site via
the electrolysis of brine. AWWA (2015) developed a Manual of Water Supply Practices—M65
for on-site generation (OSG) of hypochlorite. In 2009, Snyder et al. estimated that 8 percent of
utilities used OSG to produce hypochlorite for disinfection (Snyder et al., 2009). More recent
UCMR 3 data indicated that 15 percent of UCMR 3 systems that reported disinfectant types used
OSG hypochlorite. Thus, it appears that there is an increasing trend toward use of site-generated
hypochlorite in lieu of chlorine gas. Snyder et al. (2009) also indicated that OSG could result in
the formation of oxyhalides, including chlorate and perchlorate. As discussed in Chapter 5, the
UCMR 3 confirms that elevated chlorate levels occur much more frequently when OSG
hypochlorite is used, as compared to chlorine gas.
According to the M65 Manual (AWWA, 2015), commercially available OSG systems typically
consist of electrolytic cells that initiate an electro-oxidation process. This process converts
chloride (typically fed as a sodium chloride brine) to hypochlorite. In general, the OSG systems
can be grouped into low- and high-strength systems. Low strength-generators typically generate
0.4 percent to 0.8 percent hypochlorite from electrolysis of brine. The electrolyzed brine is then
fed directly into a day tank for short-term (48 hours or less) storage and/or is then pumped into
the water stream for disinfection. High strength units generate chlorine gas, which is then mixed
with water to produce hypochlorite at concentrations of 12 percent to 15 percent (roughly
equivalent to the concentration in bulk hypochlorite solution) and can be stored or used
immediately.
Stanford et al. (2013) conducted a comprehensive study for understanding oxyhalide formation
during hypochlorite OSG, and Stanford and Rivera (2015) later summarized this study in
Chapter 4 of the M65 Manual (AWWA, 2015). This study involved collecting chlorate
occurrence data through 54 samples from 26 low-strength and 3 high-strength OSG systems. The
formation of chlorate was widely variable among different individual OSGs. Of six OSG systems
studied, the formation of chlorate exhibited no apparent bias based on age, size or manufacturer
of the units. Measured chlorate concentrations ranged from 5 to 90 |Lxg of chlorate per mg/L of
hypochlorite generated (Stanford et al., 2013). Overall, high-strength OSGs produced less
chlorate on a per-mg free chlorine concentration basis than low-strength OSGs, which could be
attributable to the difference in the process used to generate the final hypochlorite solution.
Chlorate generated during the electrolysis process with a low-strength OSG (with a direct
application of generated hypochlorite) could be carried into the water being treated but would not
move into treated water in the case of a high-strength OSG (where hypochlorite is applied by
introducing generated chlorine gas into a caustic solution). However, these observations may be
limited due to the small sample size.
Six-Year Review 3
Technical Support Document for Chlorate
6-5
December 2016
-------
While limited information is available from systems at full scale, a number of laboratory studies
have been conducted to evaluate the formation of byproducts such as chlorate and to determine
formation mechanisms and means of optimization, as discussed below. While conditions in these
laboratory studies likely differ from conditions in full-scale systems, the types of reactions and
the products formed in the laboratory are likely similar to those in full-scale systems.
Bergmann and Koparal (2005) indicate that chlorine dioxide is a primary byproduct of the
electrolysis of chloride-containing water and that this process can also lead to the formation of
chlorite and chlorate, with chlorite being predominant. However, Vacca et al. (2013) found that
chlorate was the only chlorine byproduct formed when using boron-doped diamond electrodes in
a brine solution containing 20 g/L of chloride. They found up to 7 g/L of chlorate were produced
by a current density of 5 milliamperes per square centimeter (mA/cm2) with a residence time of 4
minutes.
Conversely, Li and Ni (2012) found that chloroform, chlorate and perchlorate were formed at
various time intervals when using a boron-doped diamond electrode in a brine solution
containing 10 millimolar (mM) of chloride (355 mg/L as chloride ion; converted) following the
application of a current density of 20 mA/cm2 over the course of 25 hours. Chloroform and
chlorate were formed initially as free available chlorine production increased. A peak chlorate
concentration of 1.67 mM (139 mg/L as chlorate ion, converted) was observed after 4 hours. As
free available chlorine levels decreased, chlorate was oxidized to perchlorate, which was the sole
target DBP remaining at 25 hours, peaking at 3.84 mM (382 mg/L as perchlorate ion, converted)
(Li and Ni, 2012).
Anodic oxidation of sodium chloride solutions can result in oxidation of hypochlorite to produce
chlorate. This reaction can be viewed either as the reaction of hypochlorous acid or hypochlorite
with water to produce chlorate, chloride, protons, oxygen and electrons (Czarnetzki and Janssen,
1992). It has also been suggested that anodic reactions can directly oxidize chloride to chlorate
(Tasaka and Tojo, 1985). Vacca et al. (2013) have also suggested that hydroxyl radical formation
from hydrolysis of water at the anode may contribute to chlorate formation. Bergmann et al.
(2014) report that radical-generating electrodes had a strong tendency to produce chlorate in a
laboratory-scale electrochemical reactor. Yoon et al. (2013) found that using 100 mg/L of
chloride in the feed solution resulted in the production of between 1.7 and 4.8 mg/L of chlorate
depending on the electrode type.
6.4 Ozone
Ozone is also used as a drinking water oxidant/disinfectant. Because ozone is highly reactive and
only moderately soluble in water, it does not maintain a residual in the drinking water
distribution system. As a result, EPA has suggested that when ozone is used, it is done in
conjunction with a secondary chlorine-based disinfectant such as chlorine, chlorine dioxide or
chloramine (USEPA, 1999).
Siddiqui (1996) indicates that when ozone is used in the presence of hypochlorous acid in
equilibrium with hypochlorite ion, oxidation of hypochlorite, first to chlorite and then to
chlorate, takes place via two primary mechanisms. The first mechanism involves chloroxyl and
hydroxyl radicals, while the second, slower mechanism involves direct attack of ozone on
Six-Year Review 3
Technical Support Document for Chlorate
6-6
December 2016
-------
hypochlorite and chlorite. The kinetics of the conversion of hypochlorite to chlorate are
complicated by the presence of organic matter and alkalinity, which may scavenge free radicals.
Siddqui (1996) further studied the effect of solution pH during ozonation and found that the
formation of chlorate was reduced by more than 85 percent when the pH was reduced from 8.0 to
6.0. The generation of hydroxyl radical was observed to be reduced by 85 percent as a result of
this pH adjustment.
Another study (von Gunten, 2003) confirms these two mechanisms of chlorate formation;
however, the author indicates that the reaction involving hydroxyl radicals is much less
important than the more rapid reaction of chlorite directly with ozone. The paper also indicates
that the formation of oxychlorine species, including chlorate, only occurs during ozonation if
pretreatment with chlorine or chlorine dioxide is performed. Rakness (2015) presents a full-scale
case study of pre-oxidation with chlorine dioxide used to reduce bromate formation in an
ozonation plant. He finds that although pre-oxidation with chlorine dioxide forms chlorite, and
chlorite creates ozone demand for the post-ozone contactor, the total ozone dose is reduced.
More importantly, such an operation can result in the nearly complete oxidation of chlorite to
chlorate (i.e., the plant effluent chlorite concentration is below the detection limit) while
substantially reducing bromate formation. In addition, the Water Research Foundation (WRF)
identifies application of ozone as one option to reduce chlorite levels in its fact sheet Strategies
to Control Disinfection By-Products (WRF, 2012).
6.5 Gaseous Chlorine along with Other Disinfectant Types
As discussed earlier, an application of gaseous chlorine alone as a disinfectant will generally not
lead to occurrence of chlorate in treated water. However, free chlorine residuals in water can be
oxidized by ozone (a stronger oxidant than chlorine) into chlorate. In addition, free chlorine itself
can be an oxidant in the presence of chlorite and can oxidize chlorite to chlorate. Furthermore, a
recent study with UV light by Wang et al. (2015b) shows that UV can partially convert chlorine
to chlorate. They find that 2 to 17 percent of the chlorine entering the UV reactor is converted to
chlorate, with UV doses of 1800 millijoules/cm2 and chlorine doses varying from 2 to 10 mg/L.
Similar conversion rates (5-11 percent) are also found at 13 to 20 kilojoules per square meter by
Cimetiere and De Laat (2014).
Six-Year Review 3
Technical Support Document for Chlorate
6-7
December 2016
-------
7 Treatment
Building upon the discussion of chlorate occurrence in Chapter 5 and formation in Chapter 6,
this chapter discusses treatment and other strategies that have been used for reducing
concentrations of chlorate in drinking water. The discussion in this chapter is divided into the
following three categories:
• Reduction of disinfectant demand,
• Modification of disinfection practices,
• Removal of chlorate
7.1 Reduction of Disinfectant Demand
As discussed earlier in this document, the occurrence of chlorate at elevated levels in most cases
is associated with the use of disinfectants or oxidants. Chlorate levels generally increase with an
increased dose of disinfectants (including with use of chlorine dioxide or hypochlorite solution).
Therefore, one potential strategy for reducing the concentration of chlorate in drinking water is
to reduce the doses of the disinfectants that lead to chlorate occurrence. For instance, Sorlini et
al. (2014) found that pre-oxidation with hypochlorite could reduce the amount of chlorate formed
by chlorine dioxide disinfection of a groundwater source. They also found the reduction in
chlorate formation was achieved by using powdered activated carbon (PAC) adsorption prior to
the disinfection. They were able to lower chlorate formation by 20 to 30 percent using PAC prior
to addition of chlorine dioxide.
Reduction of disinfectant dose (while maintaining the disinfection credits for compliance with
the microbial rules) can be accomplished by practicing better source water management to
reduce disinfectant demand (source water management may include bank filtration, as discussed
in the Six-Year Review 3 Technical Support Document for Disinfectants and Disinfection
Byproducts Rules (D/DBPRs) (USEPA, 2016a)). Also, use of alternative oxidants for pre-
oxidation (including permanganate and hydrogen peroxide, as described in the Agency's
Alternative Disinfectants and Oxidants Guidance Manual (USEPA, 1999)) can be considered,
especially for source water contaminated by inorganic reductants such as sulfide, iron and
manganese. In addition, removal of organic matter during treatment (for instance, via enhanced
coagulation, biofiltration or granular activated carbon (GAC)) prior to application of
disinfectants is also generally helpful for reducing the dose of disinfectant needed (USEPA,
2016a). Furthermore, improving the biostability of treated water and management of distribution
systems can reduce the need for additional application of disinfectants in distribution systems,
thus reducing elevated levels of chlorate throughout distribution systems (Hammes et al., 2010).
Overall, these potential strategies can be evaluated in the context of the water system's site-
specific conditions for their applicability, effectiveness and unintended consequences for
chlorate control. For more detailed discussion, see the Six-Year Review 3 Technical Support
Document for Disinfectants/Disinfection Byproducts Rules (USEPA, 2016a), Alternative
Disinfectants and Oxidants Guidance Manual (USEPA, 1999), and the Simultaneous Compliance
Guidance Manual for the Long Term 2 and Stage 2 DBP Rules (USEPA, 2007).
Six-Year Review 3
Technical Support Document for Chlorate
7-1
December 2016
-------
7.2 Modification of Disinfection Practices
This section is organized by the same categories as those used for discussion of chlorate
formation in Chapter 6:
• Chlorine dioxide,
• Bulk hypochlorite solution,
• On-site generated (OSG) hypochlorite,
• Ozone,
• Gaseous chlorine along with other disinfectant types.
7.2.1 Chlorine Dioxide
As mentioned earlier, chlorine dioxide is usually generated on site for drinking water treatment,
and chlorate can occur as a byproduct during generation of chlorine dioxide, during storage, or
after application of chlorine dioxide.
Storage of chlorine dioxide as a solution is a rare practice, and this section does not cover this
practice. When chlorine dioxide is generated and introduced into the treatment train as a gas, co-
generated chlorate (as an ion) will generally not be introduced into the water. However, when
chlorine dioxide is generated and introduced to the treatment train in solution, excess dissolved
chlorine gas in the solution could form hypochlorite and react with chlorite to form chlorate.
Therefore, maintaining optimal reagent ratios is critical to minimizing chlorate formation in
treated water from such a pathway (Gates et al., 2009). Developments in feedstock delivery
systems (e.g., automatically monitored systems that help to maintain proper reagent ratios) have
helped in the optimization of chloride dioxide generation (Gallagher et al., 1994). The following
methods may help to reduce chlorate formation during chlorine dioxide generation (USEPA,
1999; Gates et al., 2009):
• Avoid very low (<3) and very high (>11) pH;
• Avoid large excess of free chlorine relative to chlorite ion and high free chlorine
concentrations at low pH; and
• Avoid excess hypochlorous acid, which can oxidize chlorite to chlorate rather than to
chlorine dioxide.
One strategy for chlorate control is to reduce chlorine dioxide demand prior to applying chlorine
dioxide because less chlorate is generally produced with a lower chlorine dioxide dose. See the
previous section for more discussion on this strategy. As discussed in Chapter 6, chlorite and
chlorine dioxide can be converted to chlorate when they co-exist with ozone or free chlorine
residuals in water. Thus, a second strategy to lower chlorate levels may be to minimize use of
ozone or free chlorine during water treatment and in distribution systems. For instance, in
situations where chlorine dioxide is a preferred primary disinfection process, systems could
Six-Year Review 3
Technical Support Document for Chlorate
7-2
December 2016
-------
consider using chloramines rather than free chlorine as a disinfectant residual. Reducing chlorite
levels using thiosulfate ion or ferrous ion in lieu of ozone may also be effective in lowering
chlorate.
As stated in Chapter 6, the use of sulfur dioxide or sulfite ion as a reducing agent to control
chlorine dioxide and chlorite residuals in treated drinking water can result in the formation of
chlorate. In contrast, the use of thiosulfate ion or ferrous ion as alternatives to sulfur
dioxide/sulfite ion did not result in appreciable formation of chlorate (Griese et al., 1991).
Subsequent research by Griese and co-workers indicated that the use of ferrous ion at pH 5.0-5.6
could result in increased chlorate concentrations, and this increase could be prevented by
maintaining the pH at 7.0-7.5 with lime (Griese et al., 1992).
7.2.2 Bulk Hypochlorite Solution
Predictive models have been developed to determine chlorate levels during storage of
hypochlorite solutions. For example, the American Water Works Association (AWW A) has
developed a web-based predictive model for assessing the factors that contribute to the formation
of chlorate along with perchlorate and other contaminants related to hypochlorite solutions
(Snyder et al., 2009).3 The model predicts concentrations of chlorate and perchlorate in
hypochlorite solutions based on user-specified, site-specific input data (or default inputs if
specific data are not known). Input data include initial hypochlorite concentration; specific
gravity; pH; specific conductance; initial chlorate, perchlorate and chloride levels; and storage
temperatures and durations. The model output includes graphs of predicted changes in
hypochlorite, chlorate and perchlorate concentrations over time; the calculated half-life of
hypochlorite at the specified temperature(s); and tabulated data of the modeled changes in
hypochlorite, chlorate, perchlorate and dissolved oxygen. With the considerations of the quality
of initial hypochlorite stock solution and doses that need to be applied, this tool can help utilities
determine a maximum storage time and a series of storage conditions needed to meet a targeted
level of chlorate in treated water.
Overall, to minimize the decomposition of hypochlorite solutions (to chlorate) stored at drinking
water facilities, the following have been recommended (Gordon et al., 1995; Stanford et al.,
2011):
• Assessing the initial quality of hypochlorite stock upon receipt by at least measuring
hypochlorite, chlorate concentrations and pH,
• Dilution of hypochlorite solutions immediately upon receipt,
• Maintaining hypochlorite solutions at low temperature,
3 The model is posted on the AWWA website at the following URL:
http://www.awwa.org/resourcestools/waterandwastewaterutilitvmanagement/hvpochloriteassessmentmodel.aspx.
AWWA member credentials are required to access the model.
Six-Year Review 3
Technical Support Document for Chlorate
7-3
December 2016
-------
• Avoiding exposure of hypochlorite solution to sunlight,
• Measuring pH periodically, keeping the pH in the range of 11-13,
• Avoiding extended storage times for hypochlorite solutions, and
• Considering the alternative of using solid calcium hypochlorite, which decomposes
much more slowly.
7.2.3 On-Site Generation of Hypochlorite
As discussed in Chapter 6, AWWA has published a comprehensive manual regarding
hypochlorite OSG (AWWA, 2015). The manual summarizes the principles of disinfection;
regulations and standards; the principles of electrolytic cell operations; the available OSG
systems; design, installation and operational considerations; economics of OSG; and DBP
formation and control. As described in Chapter 6, the manual cites Stanford et al. (2013), who
explain that high-strength OSGs generally produce less chlorate on a per-milligram free chlorine
concentration basis than low-strength OSGs, because chlorate generated with a low-strength
OSG can be carried into the water being treated while that is not the case for high-strength OSG.
Thus, high-strength OSGs may be preferred for controlling chlorate occurrence in treated water.
In laboratory studies, Yoon et al. (2013) found that the type of electrode used affected the
production of chlorate. They tested platinum, iridium and ruthenium electrodes in a laboratory
set-up. They found that with 100 mg/L of chloride in the feed water, platinum electrodes
produced 5 mg/L of chlorine and 4.8 mg/L of chlorate, iridium electrodes produced 48 mg/L of
chlorine and 2.5 mg/L chlorate, and ruthenium electrodes produced 96 mg/L of chlorine and 1.7
mg/L chlorate, respectively. As noted in Chapter 6, Vacca et al. (2013) hypothesize that the
formation of chlorate may be due to reaction of chloride with hydroxyl radicals formed from
hydrolysis of water. They suggest that reaction of chloride to form chlorine and the reaction of
water to form hydroxyl radicals are in competition with each other. They suggest electrodes with
higher overpotentials for hydrolysis of water, such as boron-doped diamond, to reduce chlorate
formation. They found that lower current densities, shorter residence times and higher chloride
concentrations led to lower chlorate production. While the reactions taking place during
electrochemical disinfection are still being studied, it appears that altering electrode materials
and operating parameters can help reduce chlorate formation.
7.2.4 Ozone
As discussed in Chapter 6, ozone can react with chlorine (or hypochlorous acid), chlorine
dioxide, or chlorite to form chlorate. Based on this understanding of chlorate chemistry, it is
important to minimize the encounter of ozone with chlorine species for chlorate control when
ozone is used as a disinfectant or oxidant. For instance, if ozone is used as a primary disinfectant
and chlorine is used as a secondary disinfectant, monitoring of ozone residual can be conducted
prior to application of chlorine to ensure that there is essentially no or very low ozone residual.
Also, thiosulfate ion or ferrous ion in lieu of ozone can be used for reducing chlorite levels
(Griese etal., 1991, 1992).
Six-Year Review 3
Technical Support Document for Chlorate
7-4
December 2016
-------
7.2.5 Gaseous Chlorine along with Other Disinfectant Types
As discussed in Chapter 6, an application of gaseous chlorine alone as a disinfectant will
generally not lead to occurrence of chlorate in treated water. However, chlorine can react with
chlorite/chlorine dioxide or ozone to produce chlorate. In addition, ultraviolet light (UV) can
convert the chlorine residual to chlorate. Therefore, the chemistry of chlorate formation suggests
that if chlorine is being applied prior to the application of chlorine dioxide or ozone in the
treatment train (for instance, as a pre-oxidant to reduce formation of bromate from ozone in the
presence of relatively high levels of bromide), water systems may consider reducing or
eliminating chlorine residuals before applying chlorine dioxide or ozone. If chlorine is being
applied after the application of chlorine dioxide or ozone (for instance, where chlorine dioxide or
ozone is the primary disinfectant and chlorine is the secondary disinfectant), water systems may
consider reducing chlorite levels (and chlorine dioxide residual levels) or ozone residual levels
before chlorine is applied. In addition, water systems could consider improving distribution
system management practices to reduce chlorine doses (while still maintaining adequate chlorine
residual levels for disinfection) and reduce the elevated levels of chlorate that result from the
continuous reaction between chlorite and chlorine throughout distribution systems. In the case of
UV, it may be beneficial to minimize chlorine residual levels in the influent to an UV unit.
7.2.6 Removal of Chlorate
A few studies have examined methods for removal of chlorate, although none appear to have
demonstrated effectiveness at full scale at a water treatment system (Gonce and Voudrias, 1994;
Westerhoff and Johnson, 2001; Westerhoff, 2003; Srinivasan and Sorial, 2009; Kishimoto et al.,
2015; Sivasubramanian, 2015). An early demonstration with a GAC column from Gonce and
Voudrias (1994) indicates that all influent chlorite was reduced to chloride by GAC, and chlorate
was not reduced but was only physically and reversibly sorbed by GAC. A much higher sorption
capacity of chlorate was observed at pH 5 (4.9 mg/g) than pH 7 (0.5 mg/G).
Westerhoff and Johnson (2001) describe a study of a system designed to treat contaminated
groundwater with zero valent iron. They reported the complete reduction of chlorate by zero
valent iron, but the reaction was slow, with a half-life of over 200 minutes, even at 28°C. A zero
valent iron reactor might remove 30 percent of the chlorate with a 20-minute contact time. A
further bench-scale study conducted by Westerhoff (2003) suggests that higher solid (Fe°)-to-
liquid ratios increase chlorate reduction, and water system operators scaling up must consider the
solid-liquid ratios. Srinivasan and Sorial (2009) found that the chlorate reduction obtained with
zero-valent iron filings was better than that obtained with electrochemical reduction, even in the
presence of catalysts and with thin film-coated electrodes.
Sivasubramanian (2015) evaluated the effectiveness of combinations of four reducing agents
(sulfite, dithionite, sulfide and ferrous iron) and three UV light sources (UV-L, UV-M and UV-
B) for chlorate removal. This author found that dithionite irradiated by broad-band UV-B lamp,
which has output between 280 nm and 320 nm with peak energy at 312 nm, showed the highest
chlorate removal. Kishimoto et al. (2015) demonstrated that vacuum ultraviolet photolysis at 172
nm was more effective for chlorate removal than conventional UV photolysis at 254 nm.
However, vacuum ultraviolet photolysis may not be feasible in water with a high concentration
of nitrate due to a strong inhibitory effect from nitrate. The literature reviewed by Richardson et
Six-Year Review 3
Technical Support Document for Chlorate
7-5
December 2016
-------
al. (2009) show that ferrous iron can be effective for removing chlorite by reducing it to chloride,
but not effective for removing chlorate. Overall, it appears that there is not a practical process yet
for removing or reducing chlorate in drinking water supplies. Therefore, the prevention of
chlorate contamination or formation is the most practical method for lowering chlorate exposure.
Six-Year Review 3
Technical Support Document for Chlorate
7-6
December 2016
-------
8 References
Abdel-Rahman, M.S., D. Couri, and R.J. Bull. 1984. Toxicity of chlorine dioxide in drinking
water. International Journal of Toxicology. 3(4): 277-284.
Aggazzotti, G., E. Righi, G. Fantuzzi, B. Biasotta, G. Ravera, S. Kanitz, and L. Fabiani. 2004.
Chlorination by-products (CBPs) in drinking water and adverse pregnancy outcomes in Italy.
Journal of Water and Health. 2(4): 233-47 (as cited in USEPA, 2006b).
American Water Works Association (AWW A). 2000a. Committee report: Disinfection at small
systems. Journal of the American Waterworks Association. 92(5): 24-31.
AWWA. 2000b. Committee report: Disinfection at large and medium-size systems. Journal of
the American Waterworks Association. 92(5): 32-43.
AWWA. 2008a. Committee Report: Disinfection Survey, Part 1-Recent changes, current
practices, and water quality. Journal of the American Water Works Association. 100(10):76-90.
AWWA. 2008b. Committee Report: Disinfection Survey, Part 2-Alternatives, experiences, and
future plans. Journal of the American Water Works Association. 100(11): 110-124.
AWWA. 2012. Hypochlorite Assessment Model. Available online at:
http://www.awwa.org/resources-tools/water-and-wastewater-utility-management/hypochlorite-
assessment-model.aspx.
AWWA. 2014. The Potential regulatory implications of chlorate. Available online at:
http://www.awwa.Org/portals/0/files/legreg/documents/2014awwachloratebriefingpaper.pdf.
AWWA. 2015. On-Site Generation of Hypochlorite, Manual of Water Supply Practices M65,
First Edition. Denver, CO: AWWA.
Asami, M., N. Yoshida, K. Kosaka, K. Ohno, and Y. Matsui. 2013. Contribution of tap water to
chlorate and perchlorate intake: A market basket study. Science of the Total Environment. 463-
464: 190-208.
ASTM International. 2008. ASTM D6581-08, Standard Test Methods for Bromate, Bromide,
Chlorate, and Chlorite in Drinking Water by Suppressed Ion Chromatography, ASTM
International, West Conshohocken, PA. Description and option for purchase available online at:
http://www.astm.org/DATABASE.CART/HISTORICAL/D6581-08.htm
Barrett, D. 1987. A subchronic (3 month) oral toxicity study of sodium chlorate in the rat
gavage: Final Report: Project No. 86-3112. Unpublished study prepared by Bio/dynamics, Inc.
464 p. MRID: 40444801 (as cited in USEPA, 2006b).
Bercz, J.P., L.L. Jones, and L. Garner. 1982. Subchronic toxicity of chlorine dioxide and related
compounds in drinking water in the nonhuman primate. Environmental Health Perspectives 46
:47-55.
Six-Year Review 3
Technical Support Document for Chlorate
8-1
December 2016
-------
Bergmann, H.E. and S. Koparal. 2005. The formation of chlorine dioxide in the electrochemical
treatment of drinking water for disinfection. Electrochimica Acta. 50(25): 5218-5228.
Bergmann, H.E., A.S. Koparal, and T. Iourtchouk. 2014. Electrochemical advanced oxidation
processes, formation of halogenate and perhalogenate species: a critical review. Critical Reviews
in Environmental Science and Technology. 44(4): 348-390.
Bolyard, M., P. Snyder Fair, and D.P. Hautman. 1992. Occurrence of chlorate in hypochlorite
solutions used for drinking water disinfection. Environmental Science & Technology. 26(8):
1663-1665.
Bolyard, M., P. Snyder Fair, and D.P. Hautman. 1993. Sources of chlorate ion in US drinking
water. Journal of the American Waterworks Association. 85(9): 81-88.
Bommaraju, T.V. and T.F. O'Brien. 2015. Brine electrolysis. Electrochemistry Encyclopedia.
Available online at: http://knowledge.electrochem.org/encycl/art-b01-brine.htm.
California Department of Public Health (CDPH). 2007. Drinking Water Notification Levels and
Response Levels: An Overview. Available online at:
http://www.waterboards.ca.gov/water_issues/programs/tmdl/records/region_5/2008/ref2659.pdf.
California Environmental Protection Agency (CalEPA). 2002. Proposed Action Level for
Chlorate. Office of Environmental Health Hazard Assessment, California EPA. Memorandum
from Robert A. Howd to David P. Spath. January 7, 2002. Available online at:
http://www.oehha.ca.gov/water/pals/chlorate.html.
California Environmental Protection Agency (CalEPA). 2015. Drinking Water Notification
Levels and Response Levels: An Overview. Available online at:
http://www.swrcb.ca.gov/drinking_water/certlic/drinkingwater/documents/notificationlevels/noti
ficationlevels.pdf.
ChemlDPlus. 2009. Available online at: http://chem.sis.nlm.nih.gov/chemidplus/. Accessed
August 10, 2009.
Cimetiere, N. and J. De Laat. 2014. Effects of UV-dechloramination of swimming pool water on
the formation of disinfection by-products: A lab-scale study. Microchemical Journal. 112: 34-41.
Conio, O., F. De Novellis, R. Hofmann, and N. Paradis. 2009. Application strategies in drinking
water. In: State of the Science of Chlorine Dioxide in Drinking Water, eds. Gates, D., G. Ziglio,
andK. Ozekin, Chapter 4, pp. 212-157.
Czarnetzki, L.R. and J.J. Janssen. 1992. Formation of hypochlorite, chlorate, and oxygen during
NaCl electrolysis from alkaline solutions at an Ru02/Ti02 anode. Journal of Applied
Electrochemistry. 22(4): 315-324.
Six-Year Review 3
Technical Support Document for Chlorate
8-2
December 2016
-------
Dohler, K.D., C.C. Wong, and A. von zur Miihlen. 1979. The rat as a model for the study of drug
effects on thyroid function: consideration of methodological problems. Pharmacology &
Therapeutics. 5(1-3): 305-318 (as cited inUSEPA, 1998).
Environmental Working Group (EWG). 2015. National Drinking Water Database. Available
online at: http://www.ewg.org/tap-water/home. Accessed September 14, 2015.
Gallagher, D.L, R.C. Hoehn and A.M. Dietrich. 1994. Sources, occurrence, and control of
chlorine dioxide by-product residuals in drinking water. Denver, CO: AwwaRF and AWWA.
Gates, D., G. Ziglio, and K. Ozekin. 2009. State of the science of chlorine dioxide in drinking
water. Water Research Foundation/Fondazione AMGA.
Gonce, N. and E.A. Voudrias. 1994. Removal of chlorite and chlorate ions from water
usinggranular activated carbon. Water Research. 28(5): 1059-1069.
Gordon, G., B. Slootmaekers, S. Tachiyashiki and D.W. Wood III. 1990. Minimizing Chlorite
ion and chlorate ion in water treated with chlorine dioxide. Journal of the American Water
Works Association. 82(4): 160-165.
Gordon, G., L.C. Adam, B.P. Bubnis, B. Hoyt, A.J. Gillette, and A. Wilczak. 1993. Controlling
the formation of chlorate ion in liquid hypochlorite feedstocks. Journal of the American Water
Works Association. 85(9): 89-97.
Gordon, G., L.C. Adam, B.P. Bubnis. 1995. Minimizing Chlorate Ion Formation in Drinking
Water when Hypochlorite Ion is The Chlorinating Agent. Denver, CO: AwwaRF and AWWA.
Gorzalski, A.S. and A.L. Spiesman. 2015. Insights on chlorate occurrence, intra-system
variability, and source water concentrations. The Journal of American Water Works Association.
107: 11.
Great Lakes Upper Mississippi River Board of State Public Health and Environmental Managers.
1997. Recommended Standards for Water Works. Albany, NY: Health Education Services.
Griese, M.H., K. Hauser, M. Berkemeier, and G. Gordon. 1991. Using reducing agents to
eliminate chlorine dioxide and chlorite ion residuals in drinking water. Journal of the American
Waterworks Association. 83(5): 56-61.
Griese, M.H., J.J. Kaczur and G. Gordon. 1992. Combining methods for the reduction of
oxychloriner Residuals in drinking water. Journal of the American Water Works Association.
84(11): 69-77.
Hammes, F., C. Berger, O. Koster, and T. Egli. 2010. Assessing biological stability of drinking
water without disinfectant residuals in a full-scale water supply system. Journal of Water Supply:
Research and Technology. 59(1): 31-40
Hazardous Substances Data Bank (HSDB). 2015. Available online at:
http://toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen7HSDB. Accessed October 26, 2015.
Six-Year Review 3
Technical Support Document for Chlorate
8-3
December 2016
-------
Health Canada, 2008. Guidelines for Canadian Drinking Water Quality: Guideline Technical
Document Chlorite and Chlorate. Available online at:
http://healthycanadians.gc.ca/publications/healthy-living-vie-saine/water-chlorite-chlorate-
eau/index-eng.php.
Hooth, M.J., A.B. DeAngelo, M.H. George, E.T. Gaillard, G.S. Travlos, G.A. Boorman, and
D.C. Wolf. 2001. Subchronic sodium chlorate exposure in drinking water results in a
concentration-dependent increase in rat thyroid follicular cell hyperplasia. Toxicologic
Pathology. 29(2): 250-259 (as cited in USEPA, 2006b).
ICIS. 2006. Chemical Profile: Sodium Hypochlorite, last revision 2/27/06. Available online at:
http://www.icis.eom/resources/news/2006/02/28/2012753/chemical-profile-sodium-
hypochlorite/. Accessed November 27, 2015.
Jaffe, E.R., and D.E. Hultquist. 1989. Cytochrome bs reductase deficiency and enzymopenic
hereditary methemoglobinemia. In. Metabolic Basis of Inherited Disease, eds. Scriver, C., A.L.
Beaudet, W.S. Sly, and D. Valle, pp.2267-2280. New York: McGrawHill Information Services
Co.
Khan, M.A., S.E. Fenton, A.E. Swank, S.D. Hester, A. Williams, and D.C. Wolf. 2005. A
mixture of ammonium perchlorate and sodium chlorate enhances alterations of the pituitary-
thyroid axis caused by the individual chemicals in adult male F344 rats. Toxicologic Pathology.
33(7): 776-783.
Kishimoto, N, Y. Yamamoto, and S. Nishimura. 2015. Efficacy of vacuum ultraviolet photolysis
for bromate and chlorate removal. Water Science and Technology: Water Supply. 15(4): 810-
816.
Korn, C., R.C. Andrews, and M.D. Escobar. 2002. Development of chlorine dioxide-related by-
product models for drinking water treatment. Water Research. 36(1): 330-342.
Lee, E., D.H. Phua, B.L. Lim, and H.K. Goh. 2013. Severe chlorate poisoning successfully
treated with methylene blue. Journal of Emergency Medicine. 44(2): 381-384.
Li, Y. and E.J. George. 2005. Analysis of perchlorate in water by reversed-phase LC/ESI-
MS/MS using an internal standard technique. Analytical Chemistry. 77(14): 4453-4458.
Li, H. and J. Ni. 2012. Electrogeneration of disinfection byproducts at a boron-doped diamond
anode with resorcinol as a model substance. Electrochimica Acta. 69: 268-274.
Lide, D. 1984. Handbook of Chemistry and Physics. 65th edition. Boca Raton, FL: CRC Press.
Liu, C., U. von Gunten, and J-P. Croue. 2013. Enhanced chlorine dioxide decay in the presence
of metal oxides: relevance to drinking water distribution systems. Environmental Science &
Technology. 47: 8365-8372.
Logan, B.E. 1998. A Review of chlorate- and perchlorate-respiring microorganisms.
Bioremediation Journal. 2(2): 69-79.
Six-Year Review 3
Technical Support Document for Chlorate
8-4
December 2016
-------
Lubbers, J.R., S. Chauhan, and J.R. Bianchine. 1982. Controlled clinical evaluations of chlorine
dioxide, chlorite and chlorate in Man. Environmental Health Perspectives. 46: 57-62.
Lubbers, J.R., S. Chauhan, J.K. Miller, and J.R. Bianchine. 1984. The effects of chronic
administration of chlorine dioxide, chlorite and chlorate to normal healthy adult male volunteers.
Journal of Environmental Pathology, Toxicology and Oncology: official organ of the
International Society for Environmental Toxicology and Cancer. (4-5): 229-238 (as cited in
USEPA, 2006b; CalEPA, 2002).
Luzzatto, L. and A. Mehta. 1989. Glucose-6-phosphate dehydrogenase deficiency. In: The
Metabolic Basis of Inherited Disease, eds. Scriver, C., A.L. Beaudet, W.S. Sly, andD. Valle,
pp.2237-2239. New York: McGrawHill Information Services Co.
McCauley, P.T., M. Robinson, F.B. Daniel, and G.R. Olsen. 1995. The effects of subchronic
chlorate exposure in Sprague-Dawley rats. Drug and Chemical Toxicology. 18(2-3): 185-199.
McClain, R.M. 1992. Thyroid gland neoplasia: non-genotoxic mechanisms. Toxicology Letters.
64: 397-408 (as cited in USEPA, 1998).
McGuire, M.J., J.L. McLain, and A. Obolensky. 2002. Information Collection Rule Data
Analysis. Sponsored by the Microbial/Disinfection By-Products Research Council. Jointly
funded by AwwaRF and USEPA. Denver, CO: AwwaRF and AWWA.
National Center for Food and Agricultural Policy (NCFAP). 2009. National Pesticide Use
Database. Available online at: http://www.ncfap.org/pesticideuse.html. Accessed June 24, 2009.
National Toxicology Program (NTP). 2005. NTP Technical Report on the Toxicology and
Carcinogenesis Studies of Sodium Chlorate (CAS No. 7775-09-9) in F344/N Rats and B6C3F1
Mice (Drinking Water Studies).
Nieminski E.C., S. Chaudhuri, and T. Lamoreaux. 1993. The occurrence of DBPs in Utah
drinking waters. Journal of the American Water Works Association. 85(9): 98-105.
O'Day, R., J. Rench, R. Oen, and A. Castro. 1998. Demographic distribution of sensitive
populations groups. USEPA, Office of Water, Office of Science and Technology, Health and
Ecological Criteria Division. Contract No. 68-C6-0036. SRA 557-05/14. February 24.
Orme, J., Taylor, D.H., and Laurie, R.D. 1985. Effects of chlorine dioxide on thyroid function in
neonatal rats. Journal of Toxicology and Environmental Health, Part A Current Issues. 15: 315-
322.
Pisarenko, A.N, B.D. Stanford, O. Quinones, G.E. Pacey, G. Gordon, and S.A Snyder. 2010.
Rapid analysis of perchlorate, chlorate, and bromate ions in concentrated hypochlorite solutions.
Analytica Chimica Acta 659(1): 216-223.
Rakness, K.L. (2015). Ozone in Drinking Water Treatment: Process Design, Operation, and
Optimization. Denver, CO: AwwaRF and AWWA.
Six-Year Review 3
Technical Support Document for Chlorate
8-5
December 2016
-------
Richardson, S.D., C. Rav-Acha, and G.D. Simpson. 2009. Chlorine dioxide chemistry, reactions,
and disinfection by-products. In State of the Science of Chlorine Dioxide in Drinking Water, eds.
Gates, D., G. Ziglio, and K. Ozekin, Chapter 2, pp. 21-49.
Righi E., Bechtold, D. Tortorici, P. Lauriola, E. Calzolari, G. Astolfi, M.J. Nieuwenhuijsen, G.
Fantuzzi, and G. Aggazzotti. 2012. Trihalomethanes, chlorite, chlorate in drinking water and risk
of congenital anomalies: A population-based case-control study in Northern Italy. Environmental
Research. 116: 66-73.
Siddiqui, M.S. 1996. Chlorine-ozone interactions: formation of chlorate. Water Research. 30(8):
2160-2170.
Sivasubramanian, R. 2015. Chlorate reduction in water using advanced reduction processes.
Thesis. Available online at: http://oaktrust.library.tamu.edu/handle/1969. l/155465?show=full.
Snyder, S.A., R.C. Pleus, and B.J. Vanderford. 2006. Perchlorate and chlorate in dietary
supplements and flavor or enhancing ingredients. Analytica Chimica Acta. 567: 26-32.
Snyder, S.A., B.D. Stanford, A.N. Pisarenko, G. Gordon, and M. Asami. 2009. Hypochlorite -
an assessment of factors that influence the formation of perchlorate and other contaminants.
Jointly funded by American Water Works Association and Water Research Foundation.
Available online at:
http://www.awwa.0rg/Portals/O/f1les/legreg/documents/HypochloriteAssess.pdf.
Sorlini, S., F. Gialdini, M. Biasibetti, and C. Collivignarelli. 2014. Influence of drinking water
treatments on chlorine dioxide consumption and chlorite/chlorate formation. Water Research. 54:
44-52.
Srinivasan, R. and G. Sorial. 2009. Removal of perchlorate and chlorate in aquatic systems using
integrated technologies. Environmental Engineering Science. 26(11): 1661-1671.
SM. 2012. Standard Methods for the Examination of Water and Wastewater, 22nd Edition.
American Public Health Association (APHA), American Water Works Association (AWW A),
and Water Environment Federation (WEF).
Stanford, B.D. and Rivera, S. 2015. Inorganic DBP Formation in Hypochlorite Solution
Generation and Storage. In the American Water Works Association M65: On-site generation of
hypochlorite, First Edition, Chapter 4, pp 53-61.
Stanford, B.D., A.N. Pisarenko, S.A. Snyder, and G. Gordon. 2011. Perchlorate, bromate, and
chlorate in hypochlorite solutions: guidelines for utilities. Journal of the American Water Works
Association. 103(6): 71-83.
Stanford, B.D., A.N. Pisarenko, D.J. Dryer, J.C. Ziegler-Holady, S. Gamage, O. Quinones,B. J.
Vanderford, and E.R.V. Dickenson. 2013. Chlorate, perchlorate, and bromate in onsite-generated
hypochlorite systems. Journal of the American Water Works Association. 105(3): E93-E102.
Six-Year Review 3
Technical Support Document for Chlorate
8-6
December 2016
-------
Stark, B. 2013. Hypochlorite Treatment Chemicals Specifications in NSF/ANSI Standard 60.
Available online at:
http://www.nsf.org/newsroom_pdfAVATER_Hypochlorite_Treatment_Chemicals_Specifications
_NSFANSI_60.pdf.
Tasaka, A. and T. Tojo. 1985. Anodic oxidation mechanism of hypochlorite ion on platinum
electrode in alkaline solution. Journal of the Electrochemical Society 132(8): 1855-1859.
United States Environmental Protection Agency (USEPA). 1993. Method 300.0. Determination
of Inorganic Anions by Ion Chromatography. Revision 2.1. Environmental Monitoring Systems
Laboratory, Office of Research and Development. EPA 600-R-93-100. August 1993.
USEPA. 1996. National Primary Drinking Water Regulations: Monitoring Requirements for
Public Drinking Water Supplies; Cryptosporidium, Giardia, Viruses, Disinfection Byproducts,
Water Treatment Plant Data and Other Information Requirements; Final Rule. 61 FR 24354.
May 14, 1996.
USEPA. 1997. Method 300.1. Determination of Inorganic Anions in Drinking Water by Ion
Chromatography. Revision 1.0. National Exposure Research Laboratory, Office of Research and
Development. EPA 600-R-98-118.
USEPA. 1998. Assessment of Thyroid Follicular Cell Tumors. Risk Assessment Forum.
EPA/630/R-97-002. Available online at: https://www.epa.gov/osa/assessment-thyroid-follicular-
cell-tumors.
USEPA. 1999. Alternative Disinfectants and Oxidants Guidance Manual. Office of Water. EPA
815-R-99-014. Available online at:
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=2000229L.TXT.
USEPA. 2000. Methodology for Deriving Ambient Water Quality Criteria for the Protection of
Human Health. EPA 822-B-00-004. October 2000.
USEPA. 2001. Method 317.0. Determination of Inorganic Oxyhalide Disinfection By-Products
in Drinking Water Using Ion Chromatography with the Addition of a Postcolumn Reagent for
Trace Bromate Analysis. Revision 2.0. Technical Support Center, Office of Ground Water and
Drinking Water. EPA 815-B-01-001. July 2001.
USEPA. 2002. Method 326.0. Determination of Inorganic Oxyhalide Disinfection By-Products
in Drinking Water Using Ion Chromatography Incorporating the Addition of a Suppressor
Acidified Postcolumn Reagent for Trace Bromate Analysis. Revision 1.0. Technical Support
Center, Office of Ground Water and Drinking Water. EPA 815-R-03-007. June 2002.
USEPA. 2006a. Reregi strati on Eligibility Decision (RED) for Inorganic Chlorates. Office of
Prevention, Pesticides and Toxic Substances. EPA 738-R-06-014. July 2006.
Six-Year Review 3
Technical Support Document for Chlorate
8-7
December 2016
-------
USEPA. 2006b. Revised Inorganic Chlorates. HED Chapter of the Reregi strati on Eligibility
Decision Document (RED). EPA-HQ-OPP-2005-0507-0004. January 2006. Available online at:
https://archive.epa.gov/pesticides/reregistration/web/pdf/inorganicchlorates_red.pdf.
USEPA. 2007. Simultaneous Compliance Guidance Manual for the Long Term 2 and Stage 2
DBP Rules. EPA. 815-R-07-017. March 2007.
USEPA. 2009a. Final Contaminant Candidate List 3 Chemicals: Classification of the PCCL to
CCL. Office of Water. EPA 815-R-09-008. August 2009.
USEPA. 2009b. Community Water System Survey 2006, Draft 3.5. EPA 815-R-09-001.
USEPA. 2012. Revisions to the Unregulated Contaminant Monitoring Regulation (UCMR 3) for
Public Water Systems. 77 FR 26072. May 2, 2012.
USEPA. 2014a. Announcement of Preliminary Regulatory Determinations for Contaminants on
the Third Drinking Water Contaminate Candidate List; Proposed Rule. 79 FR 62715. October
20, 2014.
USEPA. 2014b. Chemical Data Reporting. Fact Sheet: Basic Information. EPA 740-K-13-001.
June 2014.
USEPA. 2015a. Chemical Data Reporting under the Toxic Substances Control Act. Available
online at: https://www.epa.gov/chemical-data-reporting.
USEPA. 2015b. Regulatory Determinations 3 Support Document. EPA 815-R-15-014.
USEPA. 2015c. Occurrence Data from the Second Unregulated Contaminant Monitoring
Regulation (UCMR 2). Including Appendices A-C. EPA 815-R-15-013. December 2015.
USEPA. 2016a. Six-Year Review 3 Technical Support Document for Disinfectants/Disinfection
Byproducts Rules. EPA-810-R-16-012. December 2016.
USEPA. 2016b. Toxic Release Inventory (TRI)-Listed Chemicals. Available online at:
https://www.epa.gov/toxics-release-inventory-tri-program/tri-listed-chemicals.
USEPA. 2016c. Inorganic Chlorates Preliminary Work Plan. Registration Review: Initial
Docket, Case Number 4049. March 2016.
USEPA. 2016d. Inorganic Chlorates (Sodium Chlorate) Problem Formulation for Registration
Review. March 10, 2016 Memorandum from Stephen Carey and colleagues to Britany Pruitt and
colleagues.
USEPA. 2016e. Third Unregulated Contaminant Monitoring Rule Dataset. July, 2016 version.
United States Geological Survey (USGS). 2015. Pesticide National Synthesis Project, 2012
Pesticide Use Maps. Available online at:
http://water.usgs. gov/nawqa/pnsp/usage/maps/compound_listing.php?year=02.
Six-Year Review 3
Technical Support Document for Chlorate
8-8
December 2016
-------
Vacca, A., M. Mascia, S. Palmas, L. Mais, and S. Rizzardini. 2013. On the formation of bromate
and chlorate ions during electrolysis with boron doped diamond anode for seawater treatment.
Journal of Chemical Technology and Biotechnology 88(12): 2244-2251.
Van Sande, J., C. Massart, R. Beauwens, A. Schoutens, S. Costagliola, J.E. Dumont, and J.
Wolff. 2003. Anion selectivity by the sodium iodide symporter. Endocrinology. 144(1): 247-
252.
von Gunten, U. 2003. Ozonation of drinking water: Part II. Disinfection and by-product
formation in presence of bromide, iodide or chlorine. Water Research. 37(7): 1469-1487.
Wang, X., Y. Mao, S. Tang, H. Yang, and Y.F. Xie. 2015a. Disinfection byproducts in drinking
water and regulatory compliance: a critical review. Frontiers of Environmental Science and
Engineering. 9(1): 3-15.
Wang, D., J.R. Bolton, S.A. Andrews, and R. Hofmann. 2015b. Formation of disinfection by-
products in the ultraviolet/chlorine advanced oxidation process. Science of the Total
Environment. 518: 49-57.
Water Research Foundation (WRF). 2012. Fact Sheet: Strategies to Control Disinfection By-
products. Available online at: http://www.waterrf.org/knowledge/dbps/FactSheets/DBP-
ControlStrategies-FactSheet.pdf.
Westerhoff, P. 2003. Reduction of nitrate, bromate, and chlorate by zero valent iron (Fe 0).
Journal of Environmental Engineering. 129(1): 10-16.
Westerhoff, P. and P. Johnson. 2001. A Zero-Valent Iron (Fe°) Packed-Bed Treatment Process.
Denver, CO: AwwaRF and AWWA.
World Health Organization (WHO). 2005. Chlorite and Chlorate in Drinking-water, Background
Document for Development of WHO Guidelines for Drinking-water Quality,
WHO/SDE/WSH/05.08/86. Available online at:
http://www.who.int/water_sanitation_health/dwq/chemicals/chlorateandchlorite0505.pdf.
WHO. 2008. Safety Evaluation of Certain Food Additives and Contaminants. World Health
Organization Food Additive Series: 59. Geneva, Switzerland: WHO.
WHO. 2015. Chlorite and Chlorate in Drinking-water, Draft Revised Background Document for
Development of WHO Guidelines for Drinking-water Quality, WHO/SDE/WSH/XXX.
Available online at: http://www.who.int/water_sanitation_health/water-
quality/guidelines/chemicals/chlorine-dioxide-chlorite-chlorate-draft-background-document-
22dec2015.pdf.
Yoon, Y., Y. Jung, M. Kwon, E. Cho, and J-W. Kang. 2013. Alternative electrode materials and
ceramic filter minimize disinfection byproducts in point-of-use electrochemical water treatment.
Environmental Engineering Science. 30(12): 742-749.
Six-Year Review 3
Technical Support Document for Chlorate
8-9
December 2016
-------
Appendix A Supplemental Data Sources
Occurrence data for chlorate are available from several sources in addition to UCMR 3. Those
sources include the Disinfection Byproducts Information Collection Rule (DBP ICR), the 2006
Community Water System Survey (CWSS) and a dataset compiled by the Environmental
Working Group (EWG). These data are described below.
A.l Disinfection Byproducts Information Collection Rule (DBP ICR), 1997-1998
The DBP ICR (USEPA, 1996; 61 FR 24353) required all public water systems (PWSs) serving at
least 100,000 people ("very large systems") to monitor and collect data on DBPs over an 18-
month period from July 1997 to December 1998. The DBP data were reported from 296 water
systems, most serving a population of over 100,000. (In the end, one participating ground water
system and two surface water systems reported serving populations less than 100,000.) Details
on the data collection process for the DBP ICR, along with an independent analysis of the data,
can be found in a report sponsored by the Microbial/Disinfection Products Council (MDPC)
(McGuire et al., 2002). Some relevant aspects of the sampling protocol are described in Chapter
2 of thq Regulatory Determinations 3 Support Document (USEPA, 2015b). The dataset used for
analysis here was generated with the Auxl database (version 5.0) of information collected under
the DBP ICR.
Of the 296 water systems participating in the survey, 82 reported chlorate results. According to
the survey design, only those that used chlorine dioxide or hypochlorite solutions for disinfection
were required to monitor for chlorate (USEPA, 1996; 61 FR 24353). Hypochlorite use triggered
quarterly sampling for chlorate at the entry point to the distribution system (EP). Chlorine
dioxide use triggered monthly sampling for chlorate at the EP and at three points within the
distribution system: near the first customer, in the middle of the distribution system, and at a
point representative of maximum residence time in the distribution system (MR). (In practice, it
appears that the EP was generally also counted as the sampling point "near the first customer.")
Systems were not required to indicate whether monitoring was triggered by use of chlorine
dioxide or sodium hypochlorite, but a review of sampling locations suggests that monitoring at
31 plants, belonging to 22 systems, may have been triggered by chlorine dioxide use.
Hypochlorite use may have triggered monitoring at the other plants and systems. However, it is
possible that both chlorine dioxide and hypochlorite may have been in use at some plants, and it
cannot be ruled out that some chlorate samples were collected and reported from systems that
used neither disinfectant and misunderstood the reporting requirements.
An inventory of chlorate results by system is presented in Exhibit A. 1. Results are broken out by
source water type and primary disinfectant type. Source water type and primary disinfectant type
were determined based on information reported in the DBP ICR database. Bins for results where
chlorine or chloramines were the primary disinfectant are labeled "sodium hypochlorite," on the
assumption that sodium hypochlorite was used to generate the primary disinfectant in most or all
of those cases. Note, however, that there may have been exceptions. Some chlorinating systems,
for example, gathered chlorate results on a schedule and at sampling locations characteristic of
monitoring triggered by chlorine dioxide use.
Six-Year Review 3
Technical Support Document for Chlorate
A-l
December 2016
-------
In cases where the primary disinfectant field in the DBP ICR database was blank, results were
binned based on the presumed trigger for chlorate monitoring. Those with sampling locations
characteristic of chlorine dioxide monitoring were binned with chlorine dioxide systems, and the
rest were binned with chlorinating/chloraminating ("sodium hypochlorite") systems. While
sampling locations provide a reasonable basis for inferring that chlorine dioxide or sodium
hypochlorite was used at these systems, the assumption that chlorine dioxide or hypochlorite-
generated chlorination/chloramination was the primary disinfectant used at these systems is an
assumption made for convenience in this report.
Ozone was listed as the primary disinfectant at nine systems in the database. Based on sampling
locations, one ozone system appears to have had chlorate monitoring triggered by chlorine
dioxide use and the others do not. Since chlorine dioxide is normally used in the treatment plant
and not the distribution system, an assumption was made that at the first system both chlorine
dioxide and ozone should be considered primary disinfectants. In the remaining eight ozone
systems, which all used chlorine and/or chloramines in the distribution system, an assumption
was made that monitoring was triggered by use of sodium hypochlorite to generate chlorine for
secondary disinfection.
All ground water systems in the database were included in the "hypochlorite" bin.
It will be seen in Exhibit A.l that some systems are listed in more than one category. This
happened when systems had multiple plants with different primary disinfectants. It also
happened when a plant changed disinfectant type during the sampling period. In such cases,
samples are distributed appropriately among the two categories. For example, if a system had a
single plant that used chlorine dioxide for 12 months and hypochlorite for 6 months then it is
categorized as both a chlorine dioxide system with 12 months' worth of samples and a
hypochlorite system with 6 months' worth of samples. Because of double-counting, the values in
the "Systems" and "Population Served" columns do not add up to the totals at the bottom of the
table.
Six-Year Review 3
Technical Support Document for Chlorate
A-2
December 2016
-------
Exhibit A.1: Inventory of DBP ICR Systems Reporting Chlorate Occurrence Data
Source
Water Type
Primary
Disinfectant Type
Total Number of
Samples
Total
Number of
Systems
Total Population
Served by
Systems
SW
Hypochlorite
420
49
30,972,523
sw
Chlorine Dioxide and
Ozone
50
1
223,411
SW
Chlorine Dioxide
1,063
21
5,947,815
sw
Ozone
53
8
5,156,001
GW
Hypochlorite
133
17
6,146,635
Total
1,719
82
41,584,457
Note: Systems that had multiple plants with different primary disinfectants, or that had changes in disinfection practices, are counted
in multiple rows. Therefore, the numbers in the "Systems" and "Population Served" columns do not add up to the values in the "total"
row.
A.l.l Summary Analysis
An analysis of chlorate occurrence data from the DBP ICR is presented in Exhibit A.2 through
Exhibit A.6. With a minimum reporting level (MRL) of 20 |ig/L, chlorate was detected in 1,490
(86.7 percent) of 1,719 samples collected at the 82 systems. Of the 1,719 samples, 332 (19.3
percent) reported concentrations greater than the Health Reference Level (HRL) (210 |ig/L) and
101 (5.9 percent) reported concentrations greater than twice the HRL (420 |ig/L). Thirty-four
systems (41.5 percent) reported concentrations of at least one sample exceeding the HRL, and 11
systems (13.4 percent) reported concentrations of at least one sample exceeding twice the HRL.
The maximum detected concentration was 2,234 |ig/L, and the median detected concentration
was 120 |ig/L. The mean concentration among samples with detections (not shown in the tables)
was 172 |ig/L.
Exhibit A.4 shows that only five systems had no detections at all; these were all surface water
systems disinfecting with hypochlorite. Hypochlorite systems, both surface water and ground
water, also had the lowest rates of HRL and 2xHRL exceedance at the system level (Exhibit A. 5
and Exhibit A.6). On the other hand, surface water and ground water systems using hypochlorite
had the highest chlorate concentrations reported in the DBP ICR database (Exhibit A.2).
Although ozone systems and chlorine dioxide systems had higher rates of HRL and 2xHRL
exceedance than surface water hypochlorite systems at the system level (Exhibit A.5 and Exhibit
A.6), the reverse is true at the sample level (Exhibit A.3). Very high rates of occurrence in the
"chlorine dioxide plus ozone" category (e.g., in Exhibit A.3) may reflect sampling at a single
system.
Six-Year Review 3
Technical Support Document for Chlorate
A-3
December 2016
-------
Exhibit A.2: Chlorate DBP ICR Occurrence Data from Systems Required to
Monitor - Summary of Detected Concentrations
Source
Water
Type
Primary
Disinfectant
Type
Minimum of
Detected
Concentrations
(ng/L)
Median of
Detected
Concentrations
(ng/L)
90th Percentile
of Detected
Concentrations
(ng/L)
99th Percentile
of Detected
Concentrations
(ng/L)
Maximum
of Detected
Concentrations
(ng/L)
SW
Hypochlorite
20
99
484
1,787
2,234
sw
Chlorine Dioxide
and Ozone
100
620
846
995
1,000
SW
Chlorine Dioxide
20
129
260
511
880
sw
Ozone
20
114
284
868
990
GW
Hypochlorite
20
59
184
962
1,200
Total
20
120
320
991
2,234
Exhibit A.3: Chlorate DBP ICR Occurrence Data from Systems Required to
Monitor - Summary of Samples
Source
Water
Type
Primary
Disinfectant
Type
Total
Number
of
Samples
Number of
Detections
Percent
Detections
Number of
Detections
> HRL
(210 |jg/L)
Percent of
Detections
> HRL
(210 |jg/L)
Number of
Detections
> 2xHRL
(420 |jg/L)
Percent of
Detections
> 2xHRL
(420 |jg/L)
SW
Hypochlorite
420
336
80.0%
87
20.7%
39
9.3%
SW
Chlorine Dioxide
and Ozone
50
47
94.0%
45
90.0%
36
72.0%
sw
Chlorine Dioxide
1,063
974
91.6%
184
17.3%
20
1.9%
SW
Ozone
53
37
69.8%
9
17.0%
2
3.8%
GW
Hypochlorite
133
96
72.2%
7
5.3%
4
3.0%
Total
1,719
1,490
86.7%
332
19.3%
101
5.9%
Six-Year Review 3
Technical Support Document for Chlorate
A-4
December 2016
-------
Exhibit A.4: Chlorate DBP ICR Occurrence Data - Summary of System and
Population Served Data from Systems Required to Monitor - All Detections
Source
Water
Type
Primary
Disinfectant
Type
Total
Number
of
Systems
Total
Population
Served by
Systems
Number of
Systems with
Detections
Population
Served by
Systems
with
Detections
Percent of
Systems
with
Detections
Percent of
Population
Served by
Systems with
Detections
SW
Hypochlorite
49
30,972,523
44
25,871,147
89.8%
83.5%
sw
Chlorine Dioxide
and Ozone
1
223,411
1
223,411
100%
100%
SW
Chlorine Dioxide
21
5,947,815
21
5,947,815
100%
100%
sw
Ozone
8
5,156,001
8
5,156,001
100%
100%
GW
Hypochlorite
17
6,146,635
17
6,146,635
100%
100%
Total
82
41,584,457
80
37,828,450
97.6%
91.0%
Note: Systems that had changes in disinfection practices are counted in multiple rows. Therefore the numbers in the table do not
add up to the value in the "total" row.
Exhibit A.5: Chlorate DBP ICR Occurrence Data - Summary of System and
Population Served Data from Systems Required to Monitor - Detections > HRL
Source
Water
Type
Primary
Disinfectant
Type
Total
Number
of
Systems
Total
Population
Served by
Systems
Number of
Systems with
Detections
> HRL
(210 |jg/L)
Population
Served by
Systems
with
Detections
> HRL
(210 |jg/L)
Percent of
Systems
with
Detections
> HRL
(210 |jg/L)
Percent of
Population
Served by
Systems with
Detections
> HRL
(210 |jg/L)
SW
Hypochlorite
49
30,972,523
18
7,335,757
36.7%
23.7%
SW
Chlorine Dioxide
and Ozone
1
223,411
1
223,411
100%
100%
sw
Chlorine Dioxide
21
5,947,815
14
3,773,354
66.7%
63.4%
SW
Ozone
8
5,156,001
4
3,749,231
50.0%
72.7%
GW
Hypochlorite
17
6,146,635
3
481,489
17.6%
7.8%
Total
82
41,584,457
34
11,824,489
41.5%
28.4%
Note: Systems that had changes in disinfection practices are counted in multiple rows. Therefore, the numbers in the table do not
add up to the value in the "total" row.
Six-Year Review 3
Technical Support Document for Chlorate
AS
December 2016
-------
Exhibit A.6: Chlorate DBP ICR Occurrence Data - Summary of System and
Population Served Data from Systems Required to Monitor - Detections > 2xHRL
Source
Water
Type
Primary
Disinfectant
Type
Total
Number
of
Systems
Total
Population
Served by
Systems
Number of
Systems
with
Detections
> 2xHRL
(420 |jg/L)
Population
Served by
Systems
with
Detections
> 2xHRL
(420 |jg/L)
Percent of
Systems
with
Detections
> 2xHRL
(420 |jg/L)
Percent of
Population
Served by
Systems
with
Detections
> 2xHRL
(420 ua/L)
SW
Hypochlorite
49
30,972,523
5
2,016,629
10.2%
6.5%
sw
Chlorine Dioxide
and Ozone
1
223,411
1
223,411
100%
100%
SW
Chlorine Dioxide
21
5,947,815
6
1,656,548
28.6%
27.9%
sw
Ozone
8
5,156,001
1
1,200,000
12.5%
23.3%
GW
Hypochlorite
17
6,146,635
1
156,450
5.9%
2.5%
Total
82
41,584,457
11
3,594,305
13.4%
8.6%
Note: Systems that had changes in disinfection practices are counted in multiple rows. Therefore, the numbers in the table do not
add up to the value in the "total" row.
A. 1.2 Limitations of DBP ICR Data
The DBP ICR was designed to gather chlorate monitoring data from very large systems (those
serving over 100,000 customers) using hypochlorite or chlorine dioxide, two disinfectants known
to be capable of generating chlorate as a DBP, in 1997-1998. Subsequent monitoring (see
discussion of UCMR 3 in Chapter 5) has demonstrated that chlorate also sometimes occurs at
systems using other disinfectants (or no disinfection at all). Therefore, DBP ICR results should
not be interpreted as capturing chlorate occurrence at all very large systems. DBP ICR does not
provide an indication of occurrence at systems serving 100,000 or fewer customers.
Interpretation of DBP ICR results is complicated by a lack of documented responsiveness to
chlorate monitoring requirements. As noted above, it cannot be ruled out that some chlorate
samples were collected and reported from systems that used neither hypochlorite nor chlorine
dioxide and misunderstood the reporting requirements. Nor has EPA verified that all very large
systems required to monitor for chlorate did monitor.
Since chlorate was only measured quarterly in systems using hypochlorite, it is also possible that
fluctuations in concentrations were missed. Also, maximum residence times may not correspond
well with locations of chlorine boosting stations, which are more likely to use hypochlorite due
to logistical considerations.
A.2 Community Water System Survey (CWSS), 2006
The 2006 CWSS (USEPA, 2009b) gathered data on the financial and operating characteristics of
a random sample of community water systems (CWSs) nationwide. All systems serving more
than 500,000 people (94 systems in 2006) received the survey, and systems in that size category
Six-Year Review 3
Technical Support Document for Chlorate
A-6
December 2016
-------
were asked questions about concentrations of unregulated contaminants in their raw and finished
water. Of the 94 systems asked about unregulated contaminants, 62 percent (58 systems)
responded to the survey, though not all of these systems answered every question. EPA
supplemented the data set by gathering additional information about contaminant occurrence at
the 94 systems from publicly available sources (e.g., Consumer Confidence Reports (CCRs)).
In the 2006 CWSS, three of the 94 systems serving more than 500,000 people reported
monitoring data for chlorate. A total of four chlorate samples were reported from these systems.
Chlorate was detected in two (50 percent) of the four samples, at concentrations of 34 |ig/L and
51 |ig/L, which were both below the HRL. Reporting levels were not specified in this survey.
A.3 Environmental Working Group (EWG) Drinking Water Database, 2004-2009
In December 2009, EWG released its "National Drinking Water Database," which includes data
on occurrence of drinking water contaminants (regulated and unregulated) collected between
2004 and 2009 from 45 states and the District of Columbia (EWG, 2015). The EWG database
includes results from approximately 48,000 water systems; data were obtained primarily from
state water offices.
EWG reviewed the data for inconsistencies, potential outliers and other errors. EWG also invited
the American Waterworks Association (AWW A) and the Association of Metropolitan Water
Agencies (AMWA) to review the data and/or have their constituents, the PWSs, review the data.
Some edits and corrections may have been made since the public release of the data in
December, 2009. EPA's analysis is based solely on the data made publicly available by EWG on
its Web site as of September, 2015. Reporting levels are not specified in EWG's database.
A total of 58 systems reported having tested for chlorate between 2004 and 2009. These data
showed detections of chlorate in 44 (75.86 percent) of the PWSs, serving 11.7 million people in
six states (Exhibit A.7). Of the ten PWSs with the highest study-wide average concentrations of
chlorate, all ten had at least one daily average concentrations of chlorate that exceeded the HRL
of 210 |ig/L. (Daily averages include both detections and non-detections; non-detections are
assumed equal to 0.) Five systems had at least one daily average concentration in excess of 420
|ig/L. The highest reported daily average concentration was 1,062 |ig/L.
The EWG data base has several limitations. It is a selection of drinking water data and not a
random sample, so it might not be representative. The use of daily average concentrations
obscures some variability in the data. Furthermore, there is no information about sampling
locations in the distribution system or disinfection practices associated with data in the database.
Six-Year Review 3
Technical Support Document for Chlorate
A-7
December 2016
-------
Exhibit A.7: Summary of EWG Chlorate Data, 2004-2009
State
Number of PWSs with
Detections of Chlorate
Population served by
PWSs with Detections
of Chlorate
California
31
10,798,220
Virginia
9
536,066
Alabama
1
237,390
New York
1
54,269
Iowa
1
41,795
Minnesota
1
10,778
Total
44
11,678,518
Six-Year Review 3
Technical Support Document for Chlorate
A-8
December 2016
-------
Appendix B Additional UCMR 3 Occurrence Analyses
This appendix presents additional analyses on the occurrence of chlorate based on the UCMR 3
data available as of July 2016. These analyses include a presentation of chlorate occurrence
estimates for community water systems (CWSs) as compared to non-transient non-community
water systems (NTNCWSs). In addition, estimates of the number of systems and sampling
locations with at least one detection greater than four thresholds are included (Minimum
Reporting Level (MRL), Health Reference Level (HRL), twice the HRL (2xHRL) and three
times the HRL (3xHRL)), as well as estimates of the number of sampling locations with a
locational running annual average (LRAA) greater than the same four thresholds. Furthermore,
this appendix also contains analytical results for characterizing the national use of disinfectant
types based on the UCMR 3 data.
B.l Additional Analyses on Occurrence by System Type
As was discussed earlier, the majority of UCMR 3 results came from CWSs rather than
NTNCWSs. Exhibit B. 1 and Exhibit B.2 present a summary of chlorate results from CWSs.
Exhibit B.3 and Exhibit B.4 present a summary of chlorate results from NTNCWSs. A very
small percentage of the chlorate data were submitted by NTNCWSs (specifically, less than one
percent of the number of samples and approximately two percent of systems serving only 0.3
percent of the overall population served by participating systems).
Six-Year Review 3
Technical Support Document for Chlorate
B-l
December 2016
-------
Exhibit B.1: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of Samples with Detections
Greater than Threshold Values (Community Water Systems)
System Size and
Source Water Type
Total
Number of
Samples
from CWSs
Number of
Samples
with
Detections
> MRL
(20 ng/L)
Percent of
Samples
with
Detections
> MRL
(20 ng/L)
Number of
Samples
with
Detections
> HRL
(210 ng/L)
Percent of
Samples
with
Detections
> HRL
(210 ng/L)
Number of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Percent of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Number of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Percent of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Small Ground Water
Systems
2,930
1,185
40.44%
440
15.02%
207
7.06%
104
3.55%
Small Surface Water
Systems
2,563
1,347
52.56%
488
19.04%
183
7.14%
83
3.24%
All Small Systems
5,493
2,532
46.10%
928
16.89%
390
7.10%
187
3.40%
Large Ground Water
Systems
18,639
8,270
44.37%
2,287
12.27%
848
4.55%
411
2.21%
Large Surface Water
Systems
26,989
15,630
57.91%
4,640
17.19%
1,546
5.73%
648
2.40%
All Large Systems
45,628
23,900
52.38%
6,927
15.18%
2,394
5.25%
1,059
2.32%
Very Large Ground
Water Systems
2,849
1,874
65.78%
321
11.27%
95
3.33%
31
1.09%
Very Large Surface
Water Systems
7,901
5,638
71.36%
1,446
18.30%
446
5.64%
204
2.58%
All Very Large
Systems
10,750
7,512
69.88%
1,767
16.44%
541
5.03%
235
2.19%
All Water Systems
61,871
33,944
54.86%
9,622
15.55%
3,325
5.37%
1,481
2.39%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-2
December 2016
-------
Exhibit B.2: Chlorate National Occurrence Measures Based on UCMR 3 Assessment Monitoring Data - Summary
of System and Population Served Data - Detections in CWSs
System Size and Source Water
Type
Number of
UCMR 3
CWSs
Population
Served by
UCMR 3
CWSs
Number of
UCMR 3
CWSs With At
Least One
Detection
> MRL
Population
Served by
UCMR 3 CWSs
With At Least
One Detection
> MRL
Percent of
UCMR 3
CWSs With At
Least One
Detection
> MRL
Percent of
Population Served
by UCMR 3 CWSs
With At Least One
Detection
> MRL
Small Ground Water Systems
454
1,416,623
248
734,560
54.6%
51.9%
Small Surface Water Systems
256
1,197,079
177
807,221
69.1%
67.4%
All Small Systems
710
2,613,702
425
1,541,781
59.9%
59.0%
Large Ground Water Systems
1,440
36,890,862
877
24,317,497
60.9%
65.9%
Large Surface Water Systems
2,248
69,292,924
1,673
52,320,637
74.4%
75.5%
All Large Systems
3,688
106,183,786
2,550
76,638,134
69.1%
72.2%
Very Large Ground Water Systems
68
16,355,951
58
14,508,549
85.3%
88.7%
Very Large Surface Water Systems
339
114,955,260
293
98,329,514
86.4%
85.5%
All Very Large Systems
407
131,311,211
351
112,838,063
86.2%
85.9%
All Water Systems
4,805
240,108,699
3,326
191,017,978
69.2%
79.6%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving
<10,000 people) and at all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-3
December 2016
-------
Exhibit B.3: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of Samples with Detections
Greater than Threshold Values (Non-Transient Non-Community Water Systems)
System Size and
Source Water Type
Total
Number of
Samples
from
NTNCWSs
Number of
Samples
with
Detections
> MRL
(20 ng/L)
Percent of
Samples
with
Detections
> MRL
(20 ng/L)
Number of
Samples
with
Detections
> HRL
(210 ng/L)
Percent of
Samples
with
Detections
> HRL
(210 ng/L)
Number of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Percent of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Number of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Percent of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Small Ground Water
Systems
308
107
34.74%
53
17.21%
27
8.77%
17
5.52%
Small Surface Water
Systems
135
122
90.37%
46
34.07%
25
18.52%
19
14.07%
All Small Systems
443
229
51.69%
99
22.35%
52
11.74%
36
8.13%
Large Ground Water
Systems
58
41
70.69%
10
17.24%
3
5.17%
2
3.45%
Large Surface Water
Systems
30
22
73.33%
10
33.33%
3
10.00%
1
3.33%
All Large Systems
88
63
71.59%
20
22.73%
6
6.82%
3
3.41%
Very Large Ground
Water Systems
0
0
0.00%
0
0.00%
0
0.00%
0
0.00%
Very Large Surface
Water Systems
12
3
25.00%
0
0.00%
0
0.00%
0
0.00%
All Very Large
Systems
12
3
25.00%
0
0.00%
0
0.00%
0
0.00%
All Water Systems
543
295
54.33%
119
21.92%
58
10.68%
39
7.18%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-4
December 2016
-------
Exhibit B.4: Chlorate National Occurrence Measures Based on UCMR 3 Assessment Monitoring Data - Summary
of System and Population Served Data - Detections in NTNCWSs
System Size and Source Water
Type
Number of
UCMR 3
NTNCWSs
Population
Served by
UCMR 3
NTNCWSs
Number of
UCMR 3
NTNCWSs
With At Least
One
Detection
> MRL
Population
Served by
UCMR 3
NTNCWSs
With At Least
One Detection
> MRL
Percent of
UCMR 3
NTNCWSs
With At Least
One
Detection
> MRL
Percent of
Population Served
by UCMR 3
NTNCWSs With At
Least One Detection
> MRL
Small Ground Water Systems
73
82,222
28
54,024
38.4%
65.7%
Small Surface Water Systems
16
53,136
16
53,136
100.0%
100.0%
All Small Systems
89
135,358
44
107,160
49.4%
79.2%
Large Ground Water Systems
9
185,929
7
163,029
77.8%
87.7%
Large Surface Water Systems
4
160,896
3
111,413
75.0%
69.2%
All Large Systems
13
346,825
10
274,442
76.9%
79.1%
Very Large Ground Water Systems
0
0
0
0
0.0%
0.0%
Very Large Surface Water Systems
1
203,000
1
203,000
100.0%
100.0%
All Very Large Systems
1
203,000
1
203,000
100.0%
100.0%
All Water Systems
103
685,183
55
584,602
53.4%
85.3%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving
<10,000 people) and at all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-5
December 2016
-------
B.2 Analyses on Samples with Detections
This appendix shows analyses of chlorate concentrations at EP and MR locations using the
UCMR 3 data. Exhibit B.5 shows a statistical summary of reported chlorate concentrations by
system size and source water type (including the minimum, median, maximum, 90th percentile
and 99th percentile). The remaining exhibits in this section show chlorate concentrations
expressed as a percentage of systems, samples, or sampling locations with at least one detection
greater than a given threshold. A sample-level summary of the results relative to the MRL,
HRL, twice the HRL (2xHRL) and three times the HRL (3xHRL) is presented in Exhibit B.6.
Simple detections are evaluated on a "greater than or equal to" basis (> the MRL), while health-
based thresholds are evaluated in terms of exceedances (> HRL, > 2xHRL, and > 3xHRL).
Exhibit B.7 through Exhibit B. 10 show more detailed system-level results, including national
extrapolations for small systems, at the same four thresholds. These tables summarize the
number of systems and associated population served with at least one detection greater than each
threshold. Figures for large and very large systems represent a census of systems in those
categories. No extrapolation was necessary in these categories, as it was for the small systems, to
derive national estimates of occurrence in these exhibits. National estimates of occurrence are
reported separately in each system size and source water category, and also in aggregate. Exhibit
B. 11 through Exhibit B. 14 show the equivalent results for monitoring locations rather than
systems (making an assumption that each system's population is equally distributed among its
several sampling locations, as described in more detail below). Exhibit B. 15 presents a summary
of additional nationally extrapolated results at the system level and sample location level at the
four thresholds. It includes system-level national population estimates.
Reported chlorate concentrations range from 20 |ig/L (the MRL) to 22,000 |ig/L (Exhibit B.5).
As of July 2016, a total of 62,414 chlorate samples had been collected from 4,908 systems. As
shown in Exhibit B.6, of these samples, 9,741 (15.6 percent) reported at least one detection
exceeding the HRL of 210 |ig/L, 3,383 (5.4 percent) reported at least one detection exceeding
twice the HRL (420 |ig/L) and 1,520 (2.4 percent) reported at least one detection exceeding three
times the HRL (630 |ig/L).
Additional details of system-level findings at each threshold are presented in Exhibit B.7 through
Exhibit B. 10. These tables show that 1,887 (38.4 percent of UCMR 3 systems, serving 48.2
percent of the PWS-served population) reported at least one detection greater than the HRL of
210 |ig/L, 982 (20.0 percent of UCMR 3 systems, serving 24.9 percent of the PWS-served
population) reported at least one detection greater than twice the HRL (420 |ig/L), and 558 (11.4
percent of UCMR 3 systems, serving 12.8 percent of the PWS-served population) reported at
least one detection greater than three times the HRL (630 |ig/L). As summarized in Exhibit B. 15,
an estimated 22,497 PWSs serving between 79 and 134 million people nationally have at least
one chlorate detection greater than the HRL, an estimated 11,757 PWSs serving between 32 and
68 million people nationally have at least one chlorate detection greater than twice the HRL and
an estimated 6,730 PWSs serving between 14 and 35 million people nationally have at least one
chlorate detection greater than three times the HRL. The derivation of these population ranges is
described later in this section.
Six-Year Review 3
Technical Support Document for Chlorate
B-6
December 2016
-------
While the analyses presented in this appendix differentiate by system size, it is important to note
that disinfectant type is a significant factor affecting chlorate occurrence and different system
sizes can have different distributions of disinfectant types. See Section 5.5 for a breakdown of
detections by system size and disinfection type.
Exhibit B.5: Chlorate Occurrence Data from UCMR 3 - Summary of Detected
Concentrations
System Size and
Source Water Type
Minimum of
Detected
Concentrations
(ng/L)
Median of
Detected
Concentrations
(ng/L)
90th Percentile
of Detected
Concentrations
(ng/L)
99th Percentile
of Detected
Concentrations
(ng/L)
Maximum of
Detected
Concentrations
(ng/L)
Small Ground Water
Systems
20
150
598
2,635
7,208
Small Surface Water
Systems
20
150
520
1,500
3,472
All Small Systems
20
150
559
1,840
7,208
Large Ground Water
Systems
20
110
430
1,339
22,000
Large Surface Water
Systems
20
130
420
1,100
13,600
All Large Systems
20
120
421
1,190
22,000
Very Large Ground
Water Systems
20
74
300
700
1,658
Very Large Surface
Water Systems
20
120
370
992
3,000
All Very Large
Systems
20
110
360
970
3,000
All Water Systems
20
120
420
1,200
22,000
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a
representative sample of small systems (serving <10,000 people) and at all large (serving 10,001 to 100,000 people)
and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-7
December 2016
-------
Exhibit B.6: National Occurrence of Chlorate Based on UCMR 3 Data - Summary of Samples with Detections
Greater than Threshold Values
System Size and
Source Water Type
Total
Number of
Samples
Number of
Samples
with
Detections
> MRL
(20 ng/L)
Percent of
Samples
with
Detections
> MRL
(20 ng/L)
Number of
Samples
with
Detections
> HRL
(210 ng/L)
Percent of
Samples
with
Detections
> HRL
(210 ng/L)
Number of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Percent of
Samples
with
Detections
> 2xHRL
(420 ng/L)
Number of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Percent of
Samples
with
Detections
> 3xHRL
(630 ng/L)
Small Ground Water
Systems
3,238
1,292
39.90%
493
15.23%
234
7.23%
121
3.74%
Small Surface Water
Systems
2,698
1,469
54.45%
534
19.79%
208
7.71%
102
3.78%
All Small Systems
5,936
2,761
46.51%
1,027
17.30%
442
7.45%
223
3.76%
Large Ground Water
Systems
18,697
8,311
44.45%
2,297
12.29%
851
4.55%
413
2.21%
Large Surface Water
Systems
27,019
15,652
57.93%
4,650
17.21%
1,549
5.73%
649
2.40%
All Large Systems
45,716
23,963
52.42%
6,947
15.20%
2,400
5.25%
1,062
2.32%
Very Large Ground
Water Systems
2,849
1,874
65.78%
321
11.27%
95
3.33%
31
1.09%
Very Large Surface
Water Systems
7,913
5,641
71.29%
1,446
18.27%
446
5.64%
204
2.58%
All Very Large
Systems
10,762
7,515
69.83%
1,767
16.42%
541
5.03%
235
2.18%
All Water Systems
62,414
34,239
54.86%
9,741
15.61%
3,383
5.42%
1,520
2.44%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving
<10,000 people) and at all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
Six-Year Review 3
Technical Support Document for Chlorate
B-8
December 2016
-------
Exhibit B.7: National Occurrence and Exposure Based on UCMR 3 Data - Summary of System and Population
Served Data - Detections of Chlorate
System Size and Source
Water Type
Number of
UCMR 3
Systems
Population
Served by
UCMR 3
Systems
Number of
UCMR 3
Systems
With At
Least One
Detection
> MRL
Population
Served by
UCMR 3
Systems With
At Least One
Detection
> MRL
Percent of
UCMR 3
Systems
With At
Least One
Detection
> MRL
Percent of
Population
Served by
UCMR 3
Systems
With At
Least One
Detection
> MRL
National
Inventory
of
Systems1
National
Inventory of
Population
Served by
System1
National
Estimate
of
Systems
With At
Least One
Detection
> MRL2
National
Estimate of
Population
Served by
Systems
With At
Least One
Detection
> MRL2
Small Ground Water
Systems
527
1,498,845
276
788,584
52.4%
52.6%
55,700
38,730,597
29,171
20,377,243
Small Surface Water
Systems
272
1,250,215
193
860,357
71.0%
68.8%
9,728
20,007,917
6,903
13,768,793
All Small Systems
799
2,749,060
469
1,648,941
58.7%
60.0%
65,428
58,738,514
36,074
34,146,036
Large Ground Water
Systems
1,449
37,076,791
884
24,480,526
61.0%
66.0%
1,470
37,540,614
884
24,480,526
Large Surface Water
Systems
2,252
69,453,820
1,676
52,432,050
74.4%
75.5%
2,310
70,791,005
1,676
52,432,050
All Large Systems
3,701
106,530,611
2,560
76,912,576
69.2%
72.2%
3,780
108,331,619
2,560
76,912,576
Very Large Ground Water
Systems
68
16,355,951
58
14,508,549
85.3%
88.7%
68
16,355,951
58
14,508,549
Very Large Surface Water
Systems
340
115,158,260
294
98,532,514
86.5%
85.6%
343
120,785,622
294
98,532,514
All Very Large Systems
408
131,514,211
352
113,041,063
86.3%
86.0%
411
137,141,573
352
113,041,063
All Water Systems
4,908
240,793,882
3,381
191,602,580
68.9%
79.6%
69,619
304,211,706
38,986
224,099,675
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1 The small system national inventory numbers for systems and population served by systems were derived from a freeze of the December 2010 SDWIS/Fed data. These counts are
based on all community and non-transient non-community water systems that served 10,000 people or fewer. All large and very large systems were required to conduct UCMR 3
Assessment Monitoring; thus, the national inventory numbers for the large and very large systems are based on the number of systems expected to complete UCMR 3 monitoring.
2 National estimates for the small systems are generated by multiplying the UCMR 3 national statistical sample findings of system/population percentages and national
system/population inventory numbers for PWSs. National estimates for the large and very large systems are based directly on the UCMR 3 results, since this was a census.
Six-Year Review 3
Technical Support Document for Chlorate
B-9
December 2016
-------
Exhibit B.8: National Occurrence and Exposure Based on UCMR 3 Data - Summary of System and Population
Served Data - Detections of Chlorate > HRL (210 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Systems
Population
Served by
UCMR 3
Systems
Number of
UCMR 3
Systems
With At
Least One
Detection
> HRL
Population
Served by
UCMR 3
Systems With
At Least One
Detection
> HRL
Percent of
UCMR 3
Systems
With At
Least One
Detection
> HRL
Percent of
Population
Served by
UCMR 3
Systems
With At
Least One
Detection
> HRL
National
Inventory
of
Systems1
National
Inventory of
Population
Served by
System1
National
Estimate
of
Systems
With At
Least One
Detection
> HRL2
National
Estimate of
Population
Served by
Systems
With At
Least One
Detection
> HRL2
Small Ground Water
Systems
527
1,498,845
160
399,068
30.4%
26.6%
55,700
38,730,597
16,911
10,312,035
Small Surface Water
Systems
272
1,250,215
111
504,760
40.8%
40.4%
9,728
20,007,917
3,970
8,077,968
All Small Systems
799
2,749,060
271
903,828
33.9%
32.9%
65,428
58,738,514
20,881
18,390,002
Large Ground Water
Systems
1,449
37,076,791
447
13,236,245
30.8%
35.7%
1,470
37,540,614
447
13,236,245
Large Surface Water
Systems
2,252
69,453,820
935
29,762,039
41.5%
42.9%
2,310
70,791,005
935
29,762,039
All Large Systems
3,701
106,530,611
1,382
42,998,284
37.3%
40.4%
3,780
108,331,619
1,382
42,998,284
Very Large Ground Water
Systems
68
16,355,951
40
10,734,989
58.8%
65.6%
68
16,355,951
40
10,734,989
Very Large Surface Water
Systems
340
115,158,260
194
61,434,961
57.1%
53.3%
343
120,785,622
194
61,434,961
All Very Large Systems
408
131,514,211
234
72,169,950
57.4%
54.9%
411
137,141,573
234
72,169,950
All Water Systems
4,908
240,793,882
1,887
116,072,062
38.4%
48.2%
69,619
304,211,706
22,497
133,558,236
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1 The small system national inventory numbers for systems and population served by systems were derived from a freeze of the December 2010 SDWIS/Fed data. These counts are
based on all community and non-transient non-community water systems that served 10,000 people or fewer. All large and very large systems were required to conduct UCMR 3
Assessment Monitoring; thus, the national inventory numbers for the large and very large systems are based on the number of systems expected to complete UCMR 3 monitoring.
2 National estimates for the small systems are generated by multiplying the UCMR 3 national statistical sample findings of system/population percentages and national
system/population inventory numbers for PWSs. National estimates for the large and very large systems are based directly on the UCMR 3 results, since this was a census.
Six-Year Review 3
Technical Support Document for Chlorate
B-10
December 2016
-------
Exhibit B.9: National Occurrence and Exposure Based on UCMR 3 Data - Summary of System and Population
Served Data - Detections of Chlorate > 2xHRL (420 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Systems
Population
Served by
UCMR 3
Systems
Number of
UCMR 3
Systems
With At
Least One
Detection
> 2xHRL
Population
Served by
UCMR 3
Systems With
At Least One
Detection
> 2xHRL
Percent of
UCMR 3
Systems
With At
Least One
Detection
> 2xHRL
Percent of
Population
Served by
UCMR 3
Systems
With At
Least One
Detection
> 2xHRL
National
Inventory
of
Systems1
National
Inventory of
Population
Served by
System1
National
Estimate
of
Systems
With At
Least One
Detection
> 2xHRL2
National
Estimate of
Population
Served by
Systems
With At
Least One
Detection
> 2xHRL2
Small Ground Water
Systems
527
1,498,845
83
184,230
15.7%
12.3%
55,700
38,730,597
8,772
4,760,558
Small Surface Water
Systems
272
1,250,215
60
254,260
22.1%
20.3%
9,728
20,007,917
2,146
4,069,071
All Small Systems
799
2,749,060
143
438,490
17.9%
16.0%
65,428
58,738,514
10,918
8,829,628
Large Ground Water
Systems
1,449
37,076,791
256
7,813,505
17.7%
21.1%
1,470
37,540,614
256
7,813,505
Large Surface Water
Systems
2,252
69,453,820
460
14,077,954
20.4%
20.3%
2,310
70,791,005
460
14,077,954
All Large Systems
3,701
106,530,611
716
21,891,459
19.3%
20.5%
3,780
108,331,619
716
21,891,459
Very Large Ground Water
Systems
68
16,355,951
24
6,838,332
35.3%
41.8%
68
16,355,951
24
6,838,332
Very Large Surface Water
Systems
340
115,158,260
99
30,829,946
29.1%
26.8%
343
120,785,622
99
30,829,946
All Very Large Systems
408
131,514,211
123
37,668,278
30.1%
28.6%
411
137,141,573
123
37,668,278
All Water Systems
4,908
240,793,882
982
59,998,227
20.0%
24.9%
69,619
304,211,706
11,757
68,389,365
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1 The small system national inventory numbers for systems and population served by systems were derived from a freeze of the December 2010 SDWIS/Fed data. These counts are
based on all community and non-transient non-community water systems that served 10,000 people or fewer. All large and very large systems were required to conduct UCMR 3
Assessment Monitoring; thus, the national inventory numbers for the large and very large systems are based on the number of systems expected to complete UCMR 3 monitoring.
2 National estimates for the small systems are generated by multiplying the UCMR 3 national statistical sample findings of system/population percentages and national
system/population inventory numbers for PWSs. National estimates for the large and very large systems are based directly on the UCMR 3 results, since this was a census.
Six-Year Review 3
Technical Support Document for Chlorate
B-ll
December 2016
-------
Exhibit B.10: National Occurrence and Exposure Based on UCMR 3 Data - Summary of System and Population
Served Data - Detections of Chlorate > 3xHRL (630 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Systems
Population
Served by
UCMR 3
Systems
Number of
UCMR 3
Systems
With At
Least One
Detection
> 3xHRL
Population
Served by
UCMR 3
Systems With
At Least One
Detection
> 3xHRL
Percent of
UCMR 3
Systems
With At
Least One
Detection
> 3xHRL
Percent of
Population
Served by
UCMR 3
Systems
With At
Least One
Detection
> 3xHRL
National
Inventory
of
Systems1
National
Inventory of
Population
Served by
System1
National
Estimate
of
Systems
With At
Least One
Detection
> 3xHRL2
National
Estimate of
Population
Served by
Systems
With At
Least One
Detection
> 3xHRL2
Small Ground Water
Systems
527
1,498,845
47
89,473
8.9%
6.0%
55,700
38,730,597
4,968
2,312,009
Small Surface Water
Systems
272
1,250,215
36
121,852
13.2%
9.7%
9,728
20,007,917
1,288
1,950,068
All Small Systems
799
2,749,060
83
211,325
10.4%
7.7%
65,428
58,738,514
6,255
4,262,077
Large Ground Water
Systems
1,449
37,076,791
161
4,950,177
11.1%
13.4%
1,470
37,540,614
161
4,950,177
Large Surface Water
Systems
2,252
69,453,820
248
7,450,699
11.0%
10.7%
2,310
70,791,005
248
7,450,699
All Large Systems
3,701
106,530,611
409
12,400,876
11.1%
11.6%
3,780
108,331,619
409
12,400,876
Very Large Ground Water
Systems
68
16,355,951
12
3,062,099
17.6%
18.7%
68
16,355,951
12
3,062,099
Very Large Surface Water
Systems
340
115,158,260
54
15,193,581
15.9%
13.2%
343
120,785,622
54
15,193,581
All Very Large Systems
408
131,514,211
66
18,255,680
16.2%
13.9%
411
137,141,573
66
18,255,680
All Water Systems
4,908
240,793,882
558
30,867,881
11.4%
12.8%
69,619
304,211,706
6,730
34,918,633
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1 The small system national inventory numbers for systems and population served by systems were derived from a freeze of the December 2010 SDWIS/Fed data. These counts are
based on all community and non-transient non-community water systems that served 10,000 people or fewer. All large and very large systems were required to conduct UCMR 3
Assessment Monitoring; thus, the national inventory numbers for the large and very large systems are based on the number of systems expected to complete UCMR 3 monitoring.
2 National estimates for the small systems are generated by multiplying the UCMR 3 national statistical sample findings of system/population percentages and national
system/population inventory numbers for PWSs. National estimates for the large and very large systems are based directly on the UCMR 3 results, since this was a census.
Six-Year Review 3
Technical Support Document for Chlorate
B-12
December 2016
-------
Exhibit B.11: National Occurrence and Exposure Based on UCMR 3 Data - Summary of Sampling Locations and
Proportional Population Served Data - Detections of Chlorate
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With At
Least One
Detection
Population
Served by
UCMR 3
Sample
Locations With
At Least One
Detection
Percent of
UCMR 3
Sample
Locations
With At
Least One
Detection
Percent of
Population
Served by
UCMR 3
Sample
Locations
With At
Least One
Detection
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
With At
Least One
Detection
National
Estimate of
Population
Served by
Sample
Locations
With At
Least One
Detection
Small Ground Water
Systems
1,632
1,498,845
741
693,639
45.4%
46.3%
110,647
38,730,597
50,238
17,923,829
Small Surface Water
Systems
762
1,250,215
476
757,170
62.5%
60.6%
16,759
20,007,917
10,469
12,117,428
All Small Systems
2,394
2,749,060
1,217
1,450,808
50.8%
52.8%
127,405
58,738,514
60,707
30,041,257
Large Ground Water
Systems
9,584
37,076,791
4,705
19,576,322
49.1%
52.8%
9,584
37,076,791
4,705
19,576,322
Large Surface Water
Systems
8,595
69,453,820
5,604
45,539,999
65.2%
65.6%
8,595
69,453,820
5,604
45,539,999
All Large Systems
18,179
106,530,611
10,309
65,116,321
56.7%
61.1%
18,179
106,530,611
10,309
65,116,321
Very Large Ground Water
Systems
1,446
16,355,951
1,032
11,224,224
71.4%
68.6%
1,446
16,355,951
1,032
11,224,224
Very Large Surface Water
Systems
2,866
115,158,260
2,189
78,034,002
76.4%
67.8%
2,866
115,158,260
2,189
78,034,002
All Very Large Systems
4,312
131,514,211
3,221
89,258,226
74.7%
67.9%
4,312
131,514,211
3,221
89,258,226
All Water Systems
24,885
240,793,882
14,747
155,825,356
59.3%
64.7%
149,896
296,783,336
74,237
184,415,804
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that had provided UCMR 3 data as of July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-13
December 2016
-------
Exhibit B.12: National Occurrence and Exposure Based on UCMR 3 Data - Summary of Sampling Locations and
Proportional Population Served Data - Detections of Chlorate > HRL (210 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With At
Least One
Detection
> HRL
Population
Served by
UCMR 3
Sample
Locations With
At Least One
Detection
> HRL
Percent of
UCMR 3
Sample
Locations
With At
Least One
Detection
> HRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With At
Least One
Detection
> HRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
With At
Least One
Detection
> HRL
National
Estimate of
Population
Served by
Sample
Locations
With At
Least One
Detection
> HRL
Small Ground Water
Systems
1,632
1,498,845
356
292,785
21.8%
19.5%
110,647
38,730,597
24,136
7,565,661
Small Surface Water
Systems
762
1,250,215
237
401,436
31.1%
32.1%
16,759
20,007,917
5,212
6,424,411
All Small Systems
2,394
2,749,060
593
694,221
24.8%
25.3%
127,405
58,738,514
29,348
13,990,072
Large Ground Water
Systems
9,584
37,076,791
1,634
8,569,088
17.0%
23.1%
9,584
37,076,791
1,634
8,569,088
Large Surface Water
Systems
8,595
69,453,820
2,334
20,173,198
27.2%
29.0%
8,595
69,453,820
2,334
20,173,198
All Large Systems
18,179
106,530,611
3,968
28,742,285
21.8%
27.0%
18,179
106,530,611
3,968
28,742,285
Very Large Ground Water
Systems
1,446
16,355,951
226
5,595,636
15.6%
34.2%
1,446
16,355,951
226
5,595,636
Very Large Surface Water
Systems
2,866
115,158,260
811
30,225,347
28.3%
26.2%
2,866
115,158,260
811
30,225,347
All Very Large Systems
4,312
131,514,211
1,037
35,820,984
24.0%
27.2%
4,312
131,514,211
1,037
35,820,984
All Water Systems
24,885
240,793,882
5,598
65,257,490
22.5%
27.1%
149,896
296,783,336
34,353
78,553,340
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that had provided UCMR 3 data as of July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-14
December 2016
-------
Exhibit B.13: National Occurrence and Exposure Based on UCMR 3 Data - Summary of Sampling Locations and
Proportional Population Served Data - Detections of Chlorate > 2xHRL (420 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With At
Least One
Detection
> 2xHRL
Population
Served by
UCMR 3
Sample
Locations With
At Least One
Detection
> 2xHRL
Percent of
UCMR 3
Sample
Locations
With At
Least One
Detection
> 2xHRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With At
Least One
Detection
> 2xHRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
With At
Least One
Detection
> 2xHRL
National
Estimate of
Population
Served by
Sample
Locations
With At
Least One
Detection
> 2xHRL
Small Ground Water
Systems
1,632
1,498,845
173
116,958
10.6%
7.8%
110,647
38,730,597
11,729
3,022,219
Small Surface Water
Systems
762
1,250,215
119
175,139
15.6%
14.0%
16,759
20,007,917
2,617
2,802,858
All Small Systems
2,394
2,749,060
292
292,097
12.2%
10.6%
127,405
58,738,514
14,346
5,825,077
Large Ground Water
Systems
9,584
37,076,791
716
4,074,423
7.5%
11.0%
9,584
37,076,791
716
4,074,423
Large Surface Water
Systems
8,595
69,453,820
946
8,243,509
11.0%
11.9%
8,595
69,453,820
946
8,243,509
All Large Systems
18,179
106,530,611
1,662
12,317,932
9.1%
11.6%
18,179
106,530,611
1,662
12,317,932
Very Large Ground Water
Systems
1,446
16,355,951
82
1,953,161
5.7%
11.9%
1,446
16,355,951
82
1,953,161
Very Large Surface Water
Systems
2,866
115,158,260
291
11,606,516
10.2%
10.1%
2,866
115,158,260
291
11,606,516
All Very Large Systems
4,312
131,514,211
373
13,559,677
8.7%
10.3%
4,312
131,514,211
373
13,559,677
All Water Systems
24,885
240,793,882
2,327
26,169,706
9.4%
10.9%
149,896
296,783,336
16,381
31,702,686
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that had provided UCMR 3 data as of July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-15
December 2016
-------
Exhibit B.14: National Occurrence and Exposure Based on UCMR 3 Data - Summary of Sampling Locations and
Proportional Population Served Data - Detections of Chlorate > 3xHRL (630 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With At
Least One
Detection
> 3xHRL
Population
Served by
UCMR 3
Sample
Locations With
At Least One
Detection
> 3xHRL
Percent of
UCMR 3
Sample
Locations
With At
Least One
Detection
> 3xHRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With At
Least One
Detection
> 3xHRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
With At
Least One
Detection
> 3xHRL
National
Estimate of
Population
Served by
Sample
Locations
With At
Least One
Detection
> 3xHRL
Small Ground Water
Systems
1,632
1,498,845
99
56,782
6.1%
3.8%
110,647
38,730,597
6,712
1,467,266
Small Surface Water
Systems
762
1,250,215
65
73,804
8.5%
5.9%
16,759
20,007,917
1,430
1,181,126
All Small Systems
2,394
2,749,060
164
130,586
6.9%
4.8%
127,405
58,738,514
8,142
2,648,391
Large Ground Water
Systems
9,584
37,076,791
372
2,241,580
3.9%
6.0%
9,584
37,076,791
372
2,241,580
Large Surface Water
Systems
8,595
69,453,820
431
3,604,117
5.0%
5.2%
8,595
69,453,820
431
3,604,117
All Large Systems
18,179
106,530,611
803
5,845,697
4.4%
5.5%
18,179
106,530,611
803
5,845,697
Very Large Ground Water
Systems
1,446
16,355,951
29
677,853
2.0%
4.1%
1,446
16,355,951
29
677,853
Very Large Surface Water
Systems
2,866
115,158,260
143
5,058,507
5.0%
4.4%
2,866
115,158,260
143
5,058,507
All Very Large Systems
4,312
131,514,211
172
5,736,360
4.0%
4.4%
4,312
131,514,211
172
5,736,360
All Water Systems
24,885
240,793,882
1,139
11,712,643
4.6%
4.9%
149,896
296,783,336
9,117
14,230,448
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that had provided UCMR 3 data as of July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-16
December 2016
-------
Exhibit B. 15 presents a summary of the national estimates of the number of PWSs, sampling
locations and the population served by PWSs that had at least one detection of chlorate at or
above the MRL and greater than the HRL, 2xHRL and 3xHRL. The third and fourth columns
present a range of estimates of the population served by PWSs that had at least one detection of
chlorate greater than the threshold. The two values in each population range represent a central
value estimate and a high end estimate. The high end estimate of the population served was
derived by adding the entire system population of all PWSs with at least one detection of
chlorate above the threshold (see Exhibit B.7 through Exhibit B.10). EPA considers this a high
end estimate because it is based on the assumption that the entire system population is served
water from the sampling location that had the highest reported chlorate concentration. However,
for the PWSs with multiple sampling locations, it is unlikely that the entire population served by
the system would receive water from the one sampling location with the highest single
concentration. Therefore, EPA also provides a central value estimate of the population served
water with chlorate above a threshold (see Exhibit B. 11 through Exhibit B. 14). This central value
estimate was developed by assuming that each system's population is equally distributed among
its several sampling locations. With this assumption, the population served by a sampling
location with at least one detection of chlorate above a given threshold is calculated by
multiplying the system's total population served by the fraction of sampling locations with at
least one detection of chlorate above that threshold.
Exhibit B.15: National Estimates of Systems and Population Served by Systems
with At Least One Detection of Chlorate Greater than Threshold Values (Based on
UCMR 3 Data)
Threshold
Concentration
National Estimate
of Number of
PWSs with At
Least One
Detection
> Threshold
(Percent1)
National Estimate of
Number of Sample
Locations with At
Least One
Detection
> Threshold
(Percent1)
Estimated Range of
Population Served by
PWSs with
At Least One
Detection
> Threshold
(in millions)
Estimated Range
of Percent
Population Served
by PWSs with
At Least One
Detection
> Threshold1
> MRL (20 |jg/L)
38,986 (56.0%)
74,237 (49.5%)
184-224
62.1%-73.7%
> HRL (210 |jg/L)
22,497 (32.3%)
34,353 (22.9%)
79-134
26.5%-43.9%
> 2xHRL (420 pg/L)
11,757 (16.9%)
16,381 (10.9%)
32-68
10.7%-22.5%
> 3xHRL (630 pg/L)
6,730 (9.7%)
9,117 (6.1%)
14-35
4.8%-11.5%
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
1 The estimated percentages of the national population exceeding the thresholds shown in this table are slightly
different from the percentages of the UCMR 3 sample population exceeding the thresholds (which are shown in
Exhibit B.7 through Exhibit B.14), reflecting the fact that the small systems are only a sample whereas the larger
systems are taken as a census. These percentages are calculated by dividing the national estimate of
systems/sampling locations/population served with threshold exceedances, shown in Exhibit B.7 through Exhibit
B.14, by the national inventory of number of systems/sampling locations/population served, also shown in those
exhibits.
Six-Year Review 3
Technical Support Document for Chlorate
B-17
December 2016
-------
Since participating systems reported data up to four times per year, UCMR 3 data can be used to
perform an analysis of seasonal patterns in contaminant occurrence. EPA has not yet performed
such an analysis for chlorate, but the U.S. Army Corps of Engineers (USACE) performed a
seasonality analysis for chlorate using the partial UCMR 3 dataset posted in January 2015
(Gorzalski and Spiesman, 2015). USACE concluded that chlorate concentrations varied over the
course of the year and were highest in the summer months. This seasonal pattern was more
pronounced in systems with higher concentrations (at least one detection of chlorate in excess of
the HRL of 210 |_ig/L) than in those with lower concentrations. This was true for gaseous
chlorine systems, on-site generation (OSG) hypochlorite systems, bulk hypochlorite solution
systems and chlorine dioxide systems. Observing that the seasonal trend was less pronounced in
chlorine dioxide systems, the study authors speculated that one reason for this may be that
sensitivity to maximum regulatory concentration levels for chlorine dioxide and chlorite may
limit the dose of chlorine dioxide used during the summer months. If systems lower chlorine
dioxide doses to avoid higher chlorite levels, then the chlorate produced will also be lower
during a time when higher temperatures would tend to increase concentrations.
B.3 Additional Analyses on Locational Averages
Exhibit B. 16 through Exhibit B. 18 show similar results as above at the four thresholds, but at the
sample point level (i.e., the number of sample points and associated population served with
detections greater than each threshold). As is discussed earlier in this report, for the purpose of
calculating population exposure, each system's population was assumed to be equally distributed
among its several sampling locations. With this assumption, the population served by a sampling
location with an average concentration of chlorate above a given threshold is calculated by
multiplying the system's total population served by the fraction of sampling locations with an
average concentration of chlorate above that threshold.
Six-Year Review 3
Technical Support Document for Chlorate
B-18
December 2016
-------
Exhibit B.16: National Occurrence Based on UCMR 3 Data - Summary of Sample Locations - Locational
Average Chlorate Concentration > HRL (210 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> HRL
Population
Served by
UCMR 3
Sample
Locations With
Locational
Ave. Cone.
> HRL
Percent of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> HRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With
Locational
Ave. Cone.
> HRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
with
Locational
Ave.
Cone.
> HRL
National
Estimate of
Population
Served by
Sample
Locations
with
Locational
Ave. Cone.
> HRL
Small Ground Water
Systems
1,632
1,498,845
265
209,929
16.2%
14.0%
110,647
38,730,597
17,966
5,424,626
Small Surface Water
Systems
762
1,250,215
157
267,768
20.6%
21.4%
16,759
20,007,917
3,453
4,285,245
All Small Systems
2,394
2,749,060
422
477,697
17.6%
17.4%
127,405
58,738,514
21,419
9,709,871
Large Ground Water
Systems
9,584
37,076,791
1,254
7,006,621
13.1%
18.9%
9,584
37,076,791
1,254
7,006,621
Large Surface Water
Systems
8,595
69,453,820
1,509
12,610,558
17.6%
18.2%
8,595
69,453,820
1,509
12,610,558
All Large Systems
18,179
106,530,611
2,763
19,617,179
15.2%
18.4%
18,179
106,530,611
2,763
19,617,179
Very Large Ground Water
Systems
1,446
16,355,951
162
3,880,974
11.2%
23.7%
1,446
16,355,951
162
3,880,974
Very Large Surface Water
Systems
2,866
115,158,260
524
18,511,111
18.3%
16.1%
2,866
115,158,260
524
18,511,111
All Very Large Systems
4,312
131,514,211
686
22,392,085
15.9%
17.0%
4,312
131,514,211
686
22,392,085
All Water Systems
24,885
240,793,882
3,871
42,486,961
15.6%
17.6%
149,896
296,783,336
24,868
51,719,135
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that have provided UCMR 3 data through July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-19
December 2016
-------
Exhibit B.17: National Occurrence Based on UCMR 3 - Summary of Sample Locations - Locational Average
Chlorate Concentration > 2xHRL (420 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> 2xHRL
Population
Served by
UCMR 3
Sample
Locations With
Locational
Ave. Cone.
> 2xHRL
Percent of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> 2xHRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With
Locational
Ave. Cone.
> 2xHRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
with
Locational
Ave.
Cone.
> 2xHRL
National
Estimate of
Population
Served by
Sample
Locations
with
Locational
Ave. Cone.
> 2xHRL
Small Ground Water
Systems
1,632
1,498,845
116
67,553
7.1%
4.5%
110,647
38,730,597
7,865
1,745,600
Small Surface Water
Systems
762
1,250,215
57
69,826
7.5%
5.6%
16,759
20,007,917
1,254
1,117,461
All Small Systems
2,394
2,749,060
173
137,379
7.2%
5.0%
127,405
58,738,514
9,118
2,863,061
Large Ground Water
Systems
9,584
37,076,791
450
2,721,316
4.7%
7.3%
9,584
37,076,791
450
2,721,316
Large Surface Water
Systems
8,595
69,453,820
411
3,332,774
4.8%
4.8%
8,595
69,453,820
411
3,332,774
All Large Systems
18,179
106,530,611
861
6,054,090
4.7%
5.7%
18,179
106,530,611
861
6,054,090
Very Large Ground Water
Systems
1,446
16,355,951
36
944,185
2.5%
5.8%
1,446
16,355,951
36
944,185
Very Large Surface Water
Systems
2,866
115,158,260
153
5,161,422
5.3%
4.5%
2,866
115,158,260
153
5,161,422
All Very Large Systems
4,312
131,514,211
189
6,105,607
4.4%
4.6%
4,312
131,514,211
189
6,105,607
All Water Systems
24,885
240,793,882
1,223
12,297,077
4.9%
5.1%
149,896
296,783,336
10,168
15,022,758
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that have provided UCMR 3 data through July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-20
December 2016
-------
Exhibit B.18: National Occurrence Based on UCMR 3 - Summary of Sample Locations - Locational Average
Chlorate Concentration > 3xHRL (630 |jg/L)
System Size and Source
Water Type
Number of
UCMR 3
Sample
Locations
Population
Served by
UCMR 3
Sample
Locations
Number of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> 3xHRL
Population
Served by
UCMR 3
Sample
Locations With
Locational
Ave. Cone.
> 3xHRL
Percent of
UCMR 3
Sample
Locations
With
Locational
Ave.
Cone.
> 3xHRL
Percent of
Population
Served by
UCMR 3
Sample
Locations
With
Locational
Ave. Cone.
> 3xHRL
National
Inventory
of Sample
Locations1
National
Inventory of
Population
Served by
Sample
Locations1
National
Estimate
of Sample
Locations
with
Locational
Ave.
Cone.
> 3xHRL
National
Estimate of
Population
Served by
Sample
Locations
with
Locational
Ave. Cone.
> 3xHRL
Small Ground Water
Systems
1,632
1,498,845
60
36,190
3.7%
2.4%
110,647
38,730,597
4,068
935,164
Small Surface Water
Systems
762
1,250,215
29
25,209
3.8%
2.0%
16,759
20,007,917
638
403,441
All Small Systems
2,394
2,749,060
89
61,400
3.7%
2.2%
127,405
58,738,514
4,706
1,338,605
Large Ground Water
Systems
9,584
37,076,791
214
1,233,262
2.2%
3.3%
9,584
37,076,791
214
1,233,262
Large Surface Water
Systems
8,595
69,453,820
134
1,021,164
1.6%
1.5%
8,595
69,453,820
134
1,021,164
All Large Systems
18,179
106,530,611
348
2,254,426
1.9%
2.1%
18,179
106,530,611
348
2,254,426
Very Large Ground Water
Systems
1,446
16,355,951
10
150,738
0.7%
0.9%
1,446
16,355,951
10
150,738
Very Large Surface Water
Systems
2,866
115,158,260
60
2,192,785
2.1%
1.9%
2,866
115,158,260
60
2,192,785
All Very Large Systems
4,312
131,514,211
70
2,343,523
1.6%
1.8%
4,312
131,514,211
70
2,343,523
All Water Systems
24,885
240,793,882
507
4,659,349
2.0%
1.9%
149,896
296,783,336
5,124
5,936,555
Source: UCMR 3 chlorate data available in July 2016 (USEPA, 2016e). UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at
all large (serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
11t was assumed that participating UCMR 3 systems were representative of the nation's systems (in each source water type / system size category) in terms of the number of sample
locations per PWS. Thus, for the small systems, the national inventory of sample locations was derived by multiplying the national inventory of small systems by the average number of
sample locations per system observed in the small system category in the UCMR 3 data. For the large and very large system categories, the national inventory of sample locations
was set equal to the number of sample locations observed in the UCMR 3 data in each category. The national numbers of sample locations (and population served by sample
locations) listed in this table currently for the large and very large systems are based on the systems that have provided UCMR 3 data through July 2016.
Six-Year Review 3
Technical Support Document for Chlorate
B-21
December 2016
-------
B.4 Analyses on Disinfectants Used
The UCMR 3 database provides a snapshot of disinfection practices in use at PWSs between
2013 and 2016. These findings help to inform the discussion of changes in disinfection practice
over time presented in Section B.5 of this report.
The following exhibits (Exhibit B.19 through Exhibit B.23) illustrate the distribution of UCMR 3
sampling locations by disinfection technique(s) associated with samples at those sampling
locations. Exhibit B.19 through Exhibit B.23 show the breakout across major categories by
source water type, system size category and sampling location (EP and MR). Exhibit B.19 and
Exhibit B.20 cover EP sampling locations, while Exhibit B.21 and Exhibit B.22 cover MR
sampling locations. The first exhibit in each pair (i.e., Exhibit B.19 and Exhibit B.21 presents
counts and percentages of sampling locations in select disinfection categories. Note that there is
overlap between some of the categories in use in these two exhibits, reflecting the use of multiple
disinfectants by some systems. The second exhibit in each pair (i.e., Exhibit B.20 and Exhibit
B.22) presents counts and percentages of sampling locations in a set of more complex but
mutually exclusive disinfection categories.
In both EP and MR locations, more than 30 percent of very large surface water systems (serving
>100,000 people) use only chloramines or "chlorine and chloramines," while approximately 50
to 54 percent of very large surface water systems (serving >100,000 people) use chloramines
alone or with another disinfectant such as ozone.
Exhibit B.23 presents an inventory of chlorate samples associated with various disinfectant codes
and combinations of disinfectant codes. Counts in this table include samples from both EP and
MR sampling locations.
Six-Year Review 3
Technical Support Document for Chlorate
B-22
December 2016
-------
Exhibit B.19: Use of Disinfectants by Source Water Type and System Size Based on UCMR 3 Data in EPs (select
categories)
Sampling
Location
Source
Water
Type1
System
size
(population
served)
Number
of EPs
Count of EPs
Indicating
Exclusive
Use of
Chlorine
(% of Total)
IS
Count of EPs
Indicating
Exclusive Use of
Chloramines, OR
both Chlorine and
Chloramines
(% of Total)
Count of EPs
Indicating
Any Instance
of Using
Chlorine
(% of Total)
Count of EPs
Indicating
Any Instance
of Using
Chloramines
Count of EPs
Indicating
Any Instance
of Using
Ozone
(% of Total)
isi.
Count of EPs
Indicating
Any Instance
of Using
Chlorine
Dioxide
(% of Total)
Count of EPs
Indicating
Any Instance
of Using UV
Light
(% of Total)
Count of EPs
Indicating
Any Instance
of Using
"Other
Disinfectant"
(% of Total)
mwk
Count of EPs
Indicating
"No
Disinfectant
Used"
(% of Total)
ffl
GW
Small
992
690
(69.6%)
90
(9.1%)
803
(80.9%)
108
(10.9%)
1
(0.1%)
3
(0.3%)
5
(0.5%)
34
(3.4%)
127
(12.8%)
GW
Large
6,590
5,244
(79.6%)
602
(9.1%)
5,419
(82.2%)
620
(9.4%)
16
(0.2%)
37
(0.6%)
13
(0.2%)
97
(1.5%)
546
(8.3%)
GW
Very Large
2,256
1,947
(86.3%)
204
(9.0%)
2,017
(89.4%)
228
(10.1%)
28
(1.2%)
(0.4%)
2
(0.1%)
10
(0.4%)
55
(2.4%)
SW
Small
293
155
(52.9%)
75
(25.6%)
256
(87.4%)
101
(34.5%)
19
(6.5%)
33
(11.3%)
12
(4.1%)
(2.7%)
0
(0%)
SW
Large
2,257
1,240
(54.9%)
591
(26.2%)
1,594
(70.6%)
742
(32.9%)
130
(5.8%)
180
(8.0%)
92
(4.1%)
24
(1.1%)
35
(1.6%)
SW
Very Large
629
253
(40.2%)
213
(33.9%)
397
(63.1%)
317
(50.4%)
86
(13.7%)
53
(8.4%)
30
(4.8%)
5
(0.8%)
1
(0.2%)
Note: Based on EP locations with data posted from July 2016. UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at all large
(serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR 3 database) was used to develop these counts. The "SW" category includes "GU" and
"MX".
The disinfection codes used to categorize each sampling location are provided graphically in the table header above each
column. The legend to the right indicates what code or set of codes corresponds to each cell. Fully shaded cells show codes
that must be present for a sampling location to be assigned to a category, and striped cells show codes that may be present.
Blank cells show codes that must not be present. Because the categories shown in this table are neither exhaustive nor
mutually exclusive, results do not add up to totals.
Layout Key
CLGA
and/or
CLOF
and/or
CLON
CAGC
and/or
CAOF
and/or
CAON
OZON
OTHD
CLDO
NODU
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
B-23
December 2016
-------
Exhibit B.20: Use of Disinfectants by Source Water Type and System Size Based on UCMR 3 Data in EPs
(mutually exclusive categories)
Sampling
Location
Source
Water
Type1
System
size
(population
served)
Number
of EPs
Count of EPs
Indicating
Exclusive
Use of
Chlorine
(% of Total)
IP
Count of EPs
Indicating
Exclusive Use of
Chloramines, OR
both Chlorine and
Chloramines
(% of Total)
Count of EPs
Indicating Any
Instance of
Chlorine Dioxide
(Except in
Combination
with Ozone)
(% of Total)
Count of EPs
Indicating Any
Instance of
Ozone (Except
in Combination
with Chlorine
Dioxide)
(% of Total)
Count of EPs
Indicating Any
Instance of
Chlorine Dioxide
and Ozone in
Combination
(% of Total)
Count of EPs
Indicating Any
Instance of UV
Light (Except in
Combination
with Chlorine
Dioxide or
Ozone)
(% of Total)
Count of EPs
Indicating Any
Instance of Any
Other
Disinfectant or
Combination of
Disinfectants
(% of Total)
GW
Small
992
690
(69.6%)
90
(9.1%)
3
(0.3%)
1
(0.1%)
0
(0%)
5
(0.5%)
76
(7.7%)
GW
Large
6,590
5,244
(79.6%)
602
(9.1%)
35
(0.5%)
14
(0.2%)
2
(0%)
12
(0.2%)
135
(2.0%)
GW
Very Large
2,256
1,947
(86.3%)
204
(9.0%)
(0.4%)
28
(1.2%)
0
(0%)
2
(0.1%)
12
(0.5%)
SW
Small
293
155
(52.9%)
75
(25.6%)
32
(10.9%)
18
(6.1%)
1
(0.3%)
7
(2.4%)
5
(1.7%)
SW
Large
2,257
1,240
(54.9%)
591
(26.2%)
176
(7.8%)
126
(5.6%)
4
(0.2%)
49
(2.2%)
36
(1.6%)
SW
Very Large
629
253
(40.2%)
213
(33.9%)
50
(7.9%)
83
(13.2%)
3
(0.5%)
22
(3.5%)
4
(0.6%)
Note: Based on EP locations with data posted from July 2016. UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at all large
(serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR 3 database) was used to develop these counts. The "SW" category includes "GU" and
"MX".
The disinfection codes used to categorize each sampling location are provided graphically in the table header above each column.
The legend to the right indicates what code or set of codes corresponds to each cell. Fully shaded cells show codes that must be
present for a sampling location to be assigned to a category, and striped cells show codes that may be present. Blank cells show
codes that must not be present. The categories shown in this table are mutually exclusive and encompass all the data, so results add
up to totals.
Layout Key
CLGA
and/or
CLOF
and/or
CLON
CAGC
and/or
CAOF
and/or
CAON
OZON
OTHD
CLDO
NODU
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
B-24
December 2016
-------
Exhibit B.21: Use of Disinfectants by Source Water Type and System Size Based on UCMR 3 Data in MRs (select
categories)
Sampling
Location
Source
Water
Type1
System size
(population
served)
Number
of MRs
Count of
MRs
Indicating
Exclusive
Use of
Chlorine
(% of Total)
IS
Count of MRs
Indicating
Exclusive Use of
Chloramines, OR
both Chlorine and
Chloramines
(% of Total)
Count of MRs
Indicating
Any Instance
of Using
Chlorine
(% of Total)
Count of MRs
Indicating
Any Instance
of Using
Chloramines
Count of MRs
Indicating
Any Instance
of Using
Ozone
(% of Total)
Count of MRs
Indicating
Any Instance
of Using
Chlorine
Dioxide
(% of Total)
Count of
MRs
Indicating
Any
Instance of
Using UV
Light
(% of Total)
Count of MRs
Indicating
Any Instance
of Using
"Other
Disinfectant"
(% of Total)
Count of
MRs
Indicating
"No
Disinfectant
Used"
(% of Total)
a
GW
Small
710
535
(75.4%)
67
(9.4%)
596
(83.9%)
69
(9.7%)
(0.1%)
4
(0.6%)
5
(0.7%)
20
(2.8%)
75
(10.6%)
GW
Large
3,813
3,031
(79.5%)
435
(11.4%)
3,198
(83.9%)
450
(11.8%)
27
(0.7%)
34
(0.9%)
14
(0.4%)
50
(1.3%)
199
(5.2%)
GW
Very Large
697
555
(79.6%)
113
(16.2%)
596
(85.5%)
120
(17.2%)
13
(1.9%)
0
(0%)
1
(0.1%)
2
(0.3%)
7
(1.0%)
SW
Small
285
153
(53.7%)
74
(26.0%)
250
(87.7%)
99
(34.7%)
16
(5.6%)
30
(10.5%)
11
(3.9%)
9
(3.2%)
0
(0%)
SW
Large
2,176
1,163
(53.4%)
604
(27.8%)
1,513
(69.5%)
750
(34.5%)
128
(5.9%)
171
(7.9%)
91
(4.2%)
28
(1.3%)
29
(1.3%)
SW
Very Large
591
189
(32.0%)
218
(36.9%)
354
(59.9%)
322
(54.5%)
85
(14.4%)
68
(11.5%)
34
(5.8%)
5
(0.8%)
5
(0.8%)
Note: Based on MR locations with data posted from July 2016. UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at all large (serving
10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR 3 database) was used to develop these counts. The "SW" category includes "GU" and
"MX".
The disinfection codes used to categorize each sampling location are provided graphically in the table header above each column.
The legend to the right indicates what code or set of codes corresponds to each cell. Fully shaded cells show codes that must be
present for a sampling location to be assigned to a category, and striped cells show codes that may be present. Blank cells show
codes that must not be present. Because the categories shown in this table are neither exhaustive nor mutually exclusive, results do
not add up to totals.
Layout Key
CLGA
and/or
CLOF
and/or
CLON
CAGC
and/or
CAOF
and/or
CAON
OZON
OTHD
CLDO
NODU
UVLV
Color Key
Used
May be
-used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
B-25
December 2016
-------
Exhibit B.22: Use of Disinfectants by Source Water Type and System Size Based on UCMR 3 Data in MRs (mutually exclusive
categories)
Sampling
Location
Source
Water
Type1
System
size
(population
served)
Number of
MRs
Count of MRs
Indicating
Exclusive
Use of
Chlorine
(% of Total)
Count of MRs
Indicating
Exclusive Use of
Chloramines, OR
both Chlorine and
Chloramines
(% of Total)
Count of MRs
Indicating Any
Instance of
Chlorine Dioxide
(Except in
Combination
with Ozone)
(% of Total)
Count of MRs
Indicating Any
Instance of Ozone
(Except in
Combination with
Chlorine Dioxide)
(% of Total)
Count of MRs
Indicating Any
Instance of
Chlorine
Dioxide and
Ozone in
Combination
(% of Total)
Count of MRs
Indicating Any
Instance of UV
Light (Except in
Combination
with Chlorine
Dioxide or
Ozone)
(% of Total)
Count of MRs
Indicating Any
Instance of Any
Other
Disinfectant or
Combination of
Disinfectants
(% of Total)
Count of MRs
Indicating
"No
Disinfectant
Used"
(% of Total)
Iff
=B
GW
Small
710
535
(75.4%)
67
(9.4%)
4
(0.6%)
1
(0.1%)
0
(0%)
5
(0.7%)
23
(3.2%)
75
(10.6%)
GW
Large
3,813
3,031
(79.5%)
435
(11.4%)
30
(0.8%)
23
(0.6%)
4
(0.1%)
13
(0.3%)
78
(2.0%)
199
(5.2%)
GW
Very Large
697
555
(79.6%)
113
(16.2%)
0
(0%)
13
(1.9%)
0
(0%)
1
(0.1%)
(1.1%)
7
(1.0%)
SW
Small
285
153
(53.7%)
74
(26.0%)
30
(10.5%)
16
(5.6%)
0
(0%)
6
(2.1%)
6
(2.1%)
0
(0%)
SW
Large
2,176
1,163
(53.4%)
604
(27.8%)
166
(7.6%)
123
(5.7%)
5
(0.2%)
47
(2.2%)
39
(1.8%)
29
(1.3%)
SW
Very Large
591
189
(32.0%)
218
(36.9%)
65
(11.0%)
82
(13.9%)
3
(0.5%)
23
(3.9%)
6
(1.0%)
5
(0.8%)
Note: Based on MR locations with data posted from July 2016. UCMR 3 monitoring was required at a representative sample of small systems (serving <10,000 people) and at all large
(serving 10,001 to 100,000 people) and very large systems (serving >100,000 people) systems in the nation.
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR 3 database) was used to develop these counts. The "SW" category includes "GU" and
"MX".
The disinfection codes used to categorize each sampling location are provided graphically in the table header above each column.
The legend to the right indicates what code or set of codes corresponds to each cell. Fully shaded cells show codes that must be
present for a sampling location to be assigned to a category, and striped cells show codes that may be present. Blank cells show
codes that must not be present. The categories shown in this table are mutually exclusive and encompass all the data, so results add
up to totals.
Layout Key
CLGA
and/or
CLOF
and/or
CLON
CAGC
and/or
CAOF
and/or
CAON
OZON
OTHD
CLDO
NODU
UVLV
Color Key
Used
May be used
Not used
Six-Year Review 3
Technical Support Document for Chlorate
B-26
December 2016
-------
Exhibit B.23: UCMR 3 Inventory of Chlorate Samples by Disinfectant Type
Sampling
Location
Source
Water
Type1
All Chlorate
Samples
Chlorination:
Gaseous
chlorine
only
Chlorination:
OSG
hypochlorite
only
Chlorination:
Bulk
solution
hypochlorite
only
Chloramination:
from gaseous
chlorine only
Chloramination:
from OSG
hypochlorite
only
Chloramination:
from bulk
solution
hypochlorite
only
Chlorine
dioxide:
alone and in
any
disinfectant
combination
Chlorine
dioxide:
in
combination
with gaseous
chlorine only
Chlorine
dioxide:
in
combination
with OSG
hypochlorite
only
Chlorine
dioxide:
in
combination
with bulk
solution
hypochlorite
only
¦
u-
-¦
=H
IP
¦
H=
¦
H
¦
¦
SW
27,975
6,339
966
4,496
3,244
724
1,546
1,743
576
48
173
GW
34,439
8,809
1,975
12,112
1,033
648
769
141
73
6
2
All
62,414
15,148
2,941
16,608
4,277
1,372
2,315
1,884
649
54
175
Sampling
Location
Source
Water
Type1
Ozonation:
alone and in
any
disinfectant
combination
Ozonation:
in
combination
with
gaseous
chlorine
only
Ozonation:
in
combination
with OSG
hypochlorite
only
Ozonation:
in
combination
with bulk
solution
hypochlorite
only
UV light:
alone and in
any disinfectant
combination
UV light:
in combination
with gaseous
chlorine only
UV light:
in combination
with OSG
hypochlorite
only
UV light:
in
combination
with bulk
solution
hypochlorite
only
"Other
disinfectant,"
alone and in
any
disinfectant
combination
No
disinfectant
used (at
least one
NODU code,
and no
other codes)
Unknown
disinfection
status
1
J
iiml
P
¦
u-
¦
IP
¦
M
u
-
H-
¦
¦
n
¦
M
r
SW
1,616
178
35
153
933
85
16
153
207
291
5,099
GW
151
32
0
39
69
12
2
27
364
2,009
5,903
All
1,767
210
35
192
1,002
97
18
180
571
2,300
11,002
Note: Counts include samples from both EPs and MRs. Based on UCMR 3 chlorate data available in July 2016 (USEPA, 2016e).
The disinfection codes used to categorize each sampling location are provided graphically in the table header above each column.
The legend to the right indicates what code or set of codes corresponds to each cell. Fully shaded cells show codes that must be
present for a sampling location to be assigned to a category, and striped cells show codes that may be present. Blank cells show
codes that must not be present.
Because the categories shown in this table are neither exhaustive nor mutually exclusive, results do not add up to totals.
CLGA
CAGC
OZON
OTHD
CLOF
CAOF
CLDO
NODU
CLON
CAON
UVLV
Color Key
Used
—
May be used
Not used
OSG = on-site generated
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR 3 database) was used to develop these counts. The "SW" category includes "GU" and
"MX".
Six-Year Review 3
Technical Support Document for Chlorate
B-27
December 2016
-------
B.5 Changes in Disinfection Practice
Changes in occurrence of chlorate need to be considered in combination with changes in
disinfection practice. Changes in disinfection practice can be examined by comparing the
disinfectant types listed in the DBP ICR data gathered in 1997-1998 (see Appendix A. 1) and
those listed in the UCMR 3 data gathered in 2013-2016. These data represent a time spread of
nearly 20 years.
The mix of disinfection techniques used by the 199 common systems (those with occurrence data
in both the DBP ICR and the UCMR 3) changed over the course of those two decades. The data
shown in Exhibit B.24 suggest that exclusive use of gaseous chlorine diminished, while use of
chlorine dioxide, ozone and ultraviolet light increased. Rates of chloramine use as a primary
disinfectant stayed fairly stable, while chloramine use as a secondary disinfectant (to maintain a
disinfectant residual in the distribution system) increased. Comparing the UCMR 3 dataset to
findings from the previous round of the UCMR program, UCMR 2 (2008-2011), sheds
additional light on the increasing trend in chloramination in recent years (Exhibit B.25). (Note
that this exhibit presents a comparison of all chloraminating sample locations in each data set,
not just those that are common to the two data sets.) The comparison indicates that chloramine
usage has increased across the board, regardless of source water type or system size. For
background information on UCMR 2, see USEPA (2015c).
Exhibit B.26 shows another important difference between the surveys: hypochlorite use has
increased dramatically in the 199 systems between the time of the DBP ICR and the UCMR 3.
At the time of the DBP ICR, only 13.0 percent of surface water plants at the common systems
employed hypochlorite, presumably bulk hypochlorite solution. (On-site generation of
hypochlorite was not common at that time, and it was not tracked as a separate category in the
survey.) By the time of UCMR 3, nearly half (42.7 percent) of the surface water facilities at the
common systems used hypochlorite in one form or another, and 11 percent were generating
hypochlorite on-site.
The foregoing indicates that usage of chlorine dioxide and sodium hypochlorite has increased
over the last two decades. This finding is not unexpected. EPA believes the increase has been
driven by (a) the need to comply with the Stage 1 and Stage 2 Disinfectants and Disinfection
Byproducts Rules and (b) concerns over the safety of storing and transporting gaseous chlorine.
In a 2007 American Water Works Association (AWW A) survey, 70 of 233 respondents indicated
that they had switched from gaseous chlorine to another disinfectant in the past eight to ten years
(AWW A, 2008a, 2008b; see also the preceding survey by AWWA, 2000a, 2000b). Over 80
percent of the systems that had ceased using gaseous chlorine had switched to some form of
hypochlorite.
Six-Year Review 3
Technical Support Document for Chlorate
B-28
December 2016
-------
Exhibit B.24: DBP ICR and UCMR 3 Comparison - Use of Disinfectants in Surface
Water Plants (Select Categories)
Number of
Number
Number
Number of
Plants / EPs
Number of
of
of Plants /
Number of
Plants / MRs
with Any
Instance of
Chloramines
(% of Total)
Among
199
Common
Systems
Total
Plants / EPs
with Exclusive
Plants / EPs
Plants/
EPs with
Total
Number
with
Use of
with Any
EPs with
Any
Number
of SW
Exclusive
Chloramines,
Instance of
Any
Instance
of SW
Plants /
Use of
OR Both
Chlorine
Instance
of UV
Plants/
EPs
Chlorine
Chlorine and
Dioxide
of Ozone
Light
MRs
(% of Total)
Chloramines
(% of Total)
(% of Total)
(% of
Total)
(% of
Total)
DBP ICR1
262
149
(56.9%)
75
(28.6%)
24
(9.2%)
14
(5.3%)
0
(0.0%)
262
113
(43.1%)
UCMR 32
342
137
(40.15%)
101
(29.5%)
44
(12.9%)
50
(14.6%)
17
(5.0%)
238
128
(53.8%)
1. The DBP ICR timeframe was 1/1998 through 12/1998. DBP ICR counts of the number of SW plants were
generated as follows: exclusive use of chlorine => plant used no other disinfectant except chlorine (CL2);
exclusive use of chloramines, OR both chlorine and chloramines => plant used no other disinfectant
except chloramine (CLM) or chloramine & chlorine (CL2_CLM); any instance of chlorine dioxide => plant
used chlorine dioxide (and may have also used other disinfectants); any instance of ozone => plant used
ozone (and may have also used other disinfectants); any instance of UV light => plant used UV (and may
have also used other disinfectants); any instance of chloramines => distribution disinfectant type was
chloramine with or without other disinfectants. To determine the number of plants with any instance of
chloramines in MR locations in the DBP ICR, the data field for the disinfectant type in the distribution system
was consulted. Only DBP ICR plants served by surface water were included.
2. The UCMR 3 timeframe was 1/2013 through 5/2016. UCMR 3 counts of the number of SW EPs were
generated as follows: exclusive use of chlorine => EP used no other disinfectant except chlorine (CLGA,
CLOF, or CLON); exclusive use of chloramines, OR both chlorine and chloramines => EP used no
other disinfectant except chloramine (CAGC, CAOF, or CAON) or chloramine and chorine (a plant using
both chloramine and chlorine would be counted in this column); any instance of chlorine dioxide => EP
used chlorine dioxide (and may have also used other disinfectants); any instance of ozone => EP used
ozone (and may have also used other disinfectants); any instance of UV light => EP used UV (and may
have also used other disinfectants); any instance of chloramines => MR used chloramine with or without
other disinfectants. Only UCMR 3 EP and MR locations with source water designation "SW" were included in
this analysis; those served by ground water, ground water under the direct influence of surface water ("GU")
or mixed source water ("MX") were excluded.
Six-Year Review 3
Technical Support Document for Chlorate
B-29
December 2016
-------
Exhibit B.25: Comparison of Chloraminating Systems in UCMR 2 and UCMR 3
Data Source
Sampling
Location
Source Water
Type1
Number of Sample
Locations Using
Chloramines at
Systems Serving
< 100,000
(and Percent of Total)2
Number of Sample
Locations Using
Chloramines at
Systems Serving
> 100,000
(and Percent of Total)2
UCMR23
SW
165
(21%)
467
(44%)
UCMR23
GW
95
(5%)
238
(9%)
UCMR 3 4
SW
1,467
(27%)
472
(48%)
UCMR 3 4
GW
1,247
(8%)
348
(12%)
1. The source water type of the sampling location (listed as "FacilityWaterType" in the UCMR databases) was used to
develop these counts. Also, the "SW" category does not include "GU," "MX" or "unknown."
2. A sample location is defined as a unique combination of PWSID / Facility ID / Sample Point ID. Note that this
exhibit is presenting information on all chloraminating sample locations in UCMR 2 and UCMR 3, not just those
sample locations that are common to both rounds of UCMR.
3. The UCMR 2 timeframe was 2008 through 2011. Counts of the UCMR 2 chloraminating sampling locations were
derived using the facility-level disinfection type. All facilities listed as "CA only" or "CA w/ CL-OT" were included in the
counts. Under UCMR 2, "CA" stood for chloramination, "CL" stood for chlorination and "OT" stood for other
disinfectants (including ozone, chlorine dioxide and UV).
4. The UCMR 3 timeframe was 1/2013 through 5/2016. Counts of the UCMR 3 chloraminating sampling locations
were derived using the "UCMR3_DRT" table from the July 2016 version of the database. All facilities listed as having
at least one occurrence of "CAGC," "CAOF" or"CAON" were included.
Exhibit B.26: DBP ICR and UCMR 3 Comparison - Hypochlorite Use in Surface
Water Plants
Among 199
Common
Systems
#Plants / #EP
Locations
(For SW only)1
Number (and
Percent) of Plants /
EP Locations
Using Bulk
Solution
Hypochlorite (A)
Number (and
Percent) of Plants /
EP Locations Using
On-site Generated
Hypochlorite (B)
Number (and
Percent) of Plants /
EP Locations
Using A and/or B
DBP ICR2
262
34 (13.0%)
N/A
N/A
UCMR 32
342
115 (33.6%)
39 (11.4%)
146 (42.7%)
1. SW counts in this table do not include ground water under the direct influence of surface water (GU) or mixed
systems (MX).
2. The DBP ICR timeframe was 1/1998 through 12/1998. For DBP ICR, the counts represent the number of plants. It
was assumed that any plants in the TUXHYPO table were using hypochlorite and those not in the table were not
using hypochlorite. Since on-site generation was not a common practice at the time of the DBP ICR and was not
tracked as a separate category, it is assumed that all hypochlorite was bulk hypochlorite solution.
3. The UCMR 3 timeframe was 1/2013 through 5/2016. For UCMR 3, the counts represent the number of entry points
(EPs). "Bulk Hypochlorite Solution (A)" was counted as all EPs with disinfectant types "CAOF" and/or "CLOF". "On-
site Generated Hypochlorite (B)" was counted as all EPs with disinfectant types "CAON" and/or "CLON".
Six-Year Review 3
Technical Support Document for Chlorate
B-30
December 2016
-------
In summary: Disinfection techniques have changed over time. A comparison of data from DBP
ICR to UCMR 3 suggests that exclusive use of gaseous chlorine diminished, while use of
chlorine dioxide, ozone and ultraviolet light increased. Rates of chloramine use as a primary
disinfectant have stayed fairly stable, while chloramine use as a secondary disinfectant (to
maintain a disinfectant residual in the distribution system) increased. Furthermore, hypochlorite
use appeared to have increased dramatically when comparing data from systems in both the DBP
ICR and the UCMR 3 data sets. In addition, a comparison of data from UCMR 2 to UCMR 3
suggests a substantial increase of chloramine use. Implications of these observed changes on
occurrence of different groups of DBPs are further discussed in EPA's Six-Year Review 3
Technical Support Document for Disinfectants/Disinfection Byproducts Rules (USEPA, 2016a).
Six-Year Review 3
Technical Support Document for Chlorate
B-31
December 2016
------- |