\
Occurrence Assessment for the
Final Stage 2 Disinfectants and
Disinfection Byproducts Rule
-------
Office of Water (4606-M) EPA 815-R-05-011 December 2005 www.epa.gov/safewater
-------
Contents
Chapter 1. Introduction
1. Introduction
1.1 Purpose of the Occurrence Document
1.2 Regulatory Background
1.2.1 Statutory Authority for Promulgating the Rule
1.2.2 1979 Total Trihalomethane Rule
1.2.3 1989 Total Coliform Rule
1.2.4 1989 Surface Water Treatment Rule
1.2.5 1996 Information Collection Rule
1.2.6 1998 Interim Enhanced Surface Water Treatment Rule
1.2.7 1998 Stage 1 Disinfectants and Disinfection Byproducts Rule
1.2.8 2000 Proposed Ground Water Rule
1.2.9 2001 Arsenic Rule
1.2.10 2001 Filter Backwash Recycling Rule
1.2.11 2002 Long Term 1 Enhanced Surface Water Treatment Rule
1.2.12 2005 Long Term 2 Enhanced Surface Water Treatment Rule
-1
-2
-3
-3
-3
-4
-4
-5
-5
-6
-6
-7
-7
-7
-7
1.3 Factors Affecting DBP Formation
1.3.1 Impact of Disinfection Method on Organic DBP Formation
1.3.2 Disinfectant Dose 1-10
1.3.3 Contact Time and DBP Formation 1-10
1.3.4 Concentration and Characteristics of Precursors 1-10
1.3.5 Water Temperature 1-11
1.3.6 WaterpH 1-11
1.4 The Primary Data Source: Information Collection Rule 1-11
1.4.1 Description of the ICR Data Set 1-12
1.4.2 ICR Implementation Activities 1-17
1.4.3 ICR Sampling Plans 1-17
1.4.4 Data Management Activities 1-17
1.4.5 Quality Assurance Activities 1-18
1.4.6 Development of Auxiliary Databases 1-19
1.4.7 Representativeness of ICR Data 1-20
1.4.8 Methods and Assumptions for Analyzing ICR Results 1-22
1.4.9 Documentation of ICR Data Analyses 1-23
1.5 Other Data Sources 1-24
1.5.1 ICR Supplemental Survey 1-26
1.5.2 National Rural Water Association Survey 1-26
1.5.3 The Water Industry Database (WATER:\STATS) 1-27
1.5.4 Ground Water Supply Survey 1-27
1.5.5 State Data 1-28
1.6 Document Organization 1-29
Chapter 2. Use of Disinfectants in the United States
2. Use of Disinfectants in the United States 2-1
2.1 Overview of Disinfection Processes 2-1
2.2 Disinfection Byproducts 2-2
Occurrence Assessment for the Final Stage 2 DBPR i December 2005
-------
2.3 Inventory of Disinfecting Water Systems and Population Served 2-3
2.4 Disinfectant Types 2-5
2.5 Chlorine 2-7
2.5.1 Description of Chemistry 2-8
2.5.2 Use and Distribution 2-9
2.5.3 Pros and Cons 2-9
2.5.4 Dose Ranges and Points of Application 2-10
2.5.5 Byproducts 2-14
2.6 Chloramines 2-15
2.6.1 Description of Chemistry 2-15
2.6.2 Use and Distribution 2-15
2.6.3 Pros and Cons 2-16
2.6.4 Dose Ranges and Points of Application 2-16
2.6.5 Byproducts 2-17
2.7 Chlorine Dioxide 2-18
2.7.1 Description of Chemistry 2-18
2.7.2 Use and Distribution 2-19
2.7.3 Advantages and Disadvantages 2-20
2.7.4 Dose Ranges 2-20
2.7.5 Byproducts 2-21
2.8 Ozonation 2-22
2.8.1 Description of Chemistry 2-23
2.8.2 Use and Distribution 2-23
2.8.3 Advantages and Disadvantages 2-24
2.8.4 Dose Ranges 2-25
2.8.5 Byproducts 2-26
Chapter 3. National DBF Occurrence: Pre-Stage 1 Baselines
3. National DBP Occurrence: Pre-Stage 1 Baselines 3-1
3.1 ICRData 3-2
3.1.1 DBP Precursors 3-2
3.1.2 Disinfectant Residuals 3-10
3.1.3 DBFs 3-11
.3.1 All Measured Halogenated DBFs 3-12
.3.2 TTHM 3-13
.3.3 HAA5 3-22
.3.4 Bromate 3-30
.3.5 Chlorite 3-31
3.2 Medium and Small Systems 3-31
3.2.1 Overview of Available Data for Medium and Small Systems 3-32
3.2.2 Surface Water Systems 3-33
3.2.2.1 Medium Surface Water Systems 3-35
3.2.2.2 Small Surface Water Systems 3-41
3.2.3 Ground Water Systems 3-55
3.2.3.1 Medium Ground Water Systems 3-55
3.2.3.2 Small Ground Water Systems 3-57
3.3 Analysis of Co-Occurrence 3-59
3.3.1 Total Organic Carbon Concentration and Alkalinity 3-59
3.3.2 TOC, Bromide, and TTHM 3-60
3.3.3 TTHM and HAA5 3-65
Occurrence Assessment for the Final Stage 2 DBPR ii December 2005
-------
3.4 Analysis of Regional Trends 3-72
3.4.1 Occurrence of TOC 3-72
3.4.2 Occurrence of Bromide 3-74
Chapter 4. National DBF Occurrence: Predicted Post-Stage 1 Baselines
4. National DBF Occurrence: Predicted Post-Stage 1 Baselines 4-1
4.1 Summary of Methodology for Predicting Post-Stage 1 DBP Occurrence for Large Plants .... 4-1
4.2 Predicted Post-Stage 1 TTHM and HAAS Occurrence 4-2
4.2.1 Large Surface Water Plants 4-2
4.2.2 Large Ground Water Plants 4-5
4.2.3 Summary of Post-Stage 1 Occurrence 4-7
4.3 Variability of TTHM and HAAS Occurrence, Post Stage 1 Conditions 4-9
4.3.1 Spatial Variability of TTHM and HAAS 4-9
4.3.2 Temporal Variability of TTHM and HAAS 4-15
4.3.3 Occurrence of Yearly Averages Above the MCL at Specific Locations 4-17
4.3.4 Occurrence of Peak DBFs at Locations Other Than the DS Maximum 4-19
Chapter 5. References
Occurrence Assessment for the Final Stage 2 DBPR iii December 2005
-------
Appendices
Appendix A: TTHM and HAAS Speciation Occurrence Data
Appendix B: ICR Data Queries
Appendix C: Assessment of Data Quality Objectives
Occurrence Assessment for the Final Stage 2 DBPR iv December 2005
-------
Exhibits
Chapter 1. Introduction
Exhibit 1.1 ICR Plant Monitoring Requirements 1-14
Exhibit 1.2a Percentage of Surface Water Systems (by State) Sampled for the ICR 1-21
Exhibit 1.2b Percentage of Ground Water Systems (by State) Sampled for the ICR 1-21
Exhibit 1.3 Summary of Non-ICR Occurrence Survey Data 1-25
Chapter 2. Use of Disinfectants in the United States
Exhibit 2.1 List of Disinfection Byproducts Measured During the ICR 2-3
Exhibit 2.2 Number (and Percent) of Disinfecting CWSs and NTNCWSs 2-4
Exhibit 2.3 Population Total (and Percent) Served by Disinfecting CWSs 2-4
Exhibit 2.4 Percentage of Surface Water Plants Applying Specific Disinfectant Types for Combined
Plant/Distribution System 2-6
Exhibit 2.5 Percentage of Ground Water Plants Applying Specific Disinfectant Types for Individual and
Combined Plant/Distribution System 2-7
Exhibit 2.6 Cumulative Distributions of Mean Total Chlorine Dose for Surface Water Plants 2-11
Exhibit 2.7 Cumulative Distributions of Mean Total Chlorine Dose for Ground Water Plants 2-12
Exhibit 2.8 Number of Disinfection Points in Plants Using Only Free Chlorine by Plant Type 2-13
Exhibit 2.9 Number of Chlorine Application Locations in Conventional Plants Using Only Free
Chlorine 2-14
Exhibit 2.10 C12:NH3-N Weight Ratios in Surface Water CL2_CLM and CLM Plants 2-17
Exhibit 2.11 Chlorine Dioxide Doses (Plant Minimum, Mean, and Maximum) 2-21
Exhibit 2.12 Ozone Doses (Plant Minimum, Mean, and Maximum) 2-25
Chapter 3. National DBF Occurrence: Pre-Stage 1 Baselines
Exhibit 3.1 Summary of Influent Water Quality Parameter ICR Data for All Large Plants 3-3
Exhibit 3.2 Cumulative Distribution of Plant-Mean TOC Concentrations of Influent Samples Based on
ICR Data for Large Surface and Ground Water Plants (mg/L as C) 3-4
Exhibit 3.3 Cumulative Distribution of Plant-Mean Water Temperature of Influent Samples Based on
ICR Data for Large Surface and Ground Water Plants (degrees Celsius) 3-5
Exhibit 3.4 Cumulative Distribution of Plant-Mean Bromide Concentrations of Influent Samples Based
on ICR Data for Large Surface and Ground Water Plants (mg/L) 3-6
Exhibit 3.5 Cumulative Distribution of Plant-Mean UV254 Absorbance of Influent Samples Based on ICR
Data for Large Surface and Ground Water Plants (cm"1) 3-7
Exhibit 3.6 Cumulative Distribution of Differences Between Highest and Lowest Monthly Parameter
Values for Influent Water Sample Location Based on ICR Data for All Large Plants 3-9
Exhibit 3.7 Summary of Disinfectant Residual ICR Data for All Large Plants 3-10
Exhibit 3.8 Summary of Halogenated DBP Data Measured During the ICR, Single Highest (Parameter
Occurrence Values in (ig/L) for All Large Plants 3-13
Exhibit 3.9 Summary of TTHM (|^g/L) ICR Data for All Large Plants 3-14
Exhibit 3.10 Cumulative Distribution of Plant-Mean DS Average (RAA) for ICR TTHM Occurrence
Data for All Large Plants 3-15
Exhibit 3.11 Cumulative Distribution of Single Highest ICR TTHM Occurrence Data for All Large
Surface and Ground Water Plants 3-16
Exhibit 3.12 Cumulative Distribution of Highest LRAA for ICR TTHM Occurrence Data for Large
Occurrence Assessment for the Final Stage 2 DBPR v December 2005
-------
Surface and Ground Water Plants 3-17
Exhibit 3.13 Location of Highest TTHM LRAA for ICR Occurrence Data for Large Surface and Ground
Water Plants 3-19
Exhibit 3.14 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant Disinfectant Type for
Large Surface Water Plants 3-20
Exhibit 3.15 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant Disinfectant Type for
Large Ground Water Plants 3-21
Exhibit 3.16 Summary of HAAS ICR Data for All Large Plants (|ig/L) 3-22
Exhibit 3.17 Cumulative Distribution of Plant-Mean DS Average (RAA) for ICR HAAS Occurrence
Data for Large Surface and Ground Water Plants 3-23
Exhibit 3.18 Cumulative Distribution of Single Highest ICR HAAS Occurrence Data for Large Surface
and Ground Water Plants 3-24
Exhibit 3.19 Cumulative Distribution of Highest LRAA ICR HAAS Occurrence Data for Large Surface
and Ground Water Plants 3-25
Exhibit 3.20 Location of Highest HAAS LRAA for ICR Occurrence Data for Large Surface and Ground
Water Plants 3-27
Exhibit 3.21 Location of Highest HAAS LRAA for ICR Occurrence Data by Plant Disinfectant Type for
Large Surface Water Plants 3-28
Exhibit 3.22 Location of Highest HAAS LRAA for ICR Occurrence Data by Plant Disinfectant Type for
Large Ground Water Plants 3-29
Exhibit 3.23 Summary of Bromate in Finished Water, Plant-Mean ICR Data for All Large Plants
(ng/L) 3-30
Exhibit 3.24 Summary of Chlorite ICR Data ((ig/L) for Large Surface Water Plants 3-31
Exhibit 3.25 Summary of Non-ICR DBP Precursor Data for Medium and Small Surface and Ground
Water Plants, Plant-Means 3-34
Exhibit 3.26 Summary of Non-ICR DBP Precursor Data for Medium and
Small Surface and Ground Water Plants, Individual Observations 3-35
Exhibit 3.27 Percentages of Medium and Large Surface Water Systems Using Different Source Water
Types 3-36
Exhibit 3.28 Comparison of Source Water TOC for Medium and Large Surface Water Systems .... 3-37
Exhibit 3.29 Comparison of Source Water TOC for Small, Medium, and Large Surface Water
Systems 3-37
Exhibit 3.30 Comparison of Source Water Turbidity For Medium and Large Surface Water
Systems 3-38
Exhibit 3.31 Comparison of Source Water Alkalinity for Medium and Large Surface Water
Systems 3-38
Exhibit 3.32 Comparison of Treatment-In-Place for Medium and Large Surface Water Systems .... 3-39
Exhibit 3.33 Comparison of Physical Unit Processes for Medium and Large Surface Water Systems 3-39
Exhibit 3.34 Comparison of Disinfectant Type for Medium and Large Surface Water Systems Using
Conventional Filtration 3-40
Exhibit 3.35 Comparison of Finished Water Annual Average TTHM for Medium and Large Surface
Water Systems 3-40
Exhibit 3.36 Comparison of Distribution System TTHM Data for Medium and Large Surface Water
Systems 3-41
Exhibit 3.37 Plant Influent TOC Data for Small Surface Water Plants 3-42
Exhibit 3.38 Plant Influent Bromide Data for Small Surface Water Plants 3-43
Exhibit 3.39 Plant Influent Alkalinity for Small Surface Water Systems 3-43
Exhibit 3.40 Plant Influent Temperature for Small Surface Water Systems 3-44
Exhibit 3.41 Distribution of Time Operated per Day Among Small Surface Water Plants 3-44
Exhibit 3.42 Treatment Objectives Among Small Surface Water Plants 3-45
Exhibit 3.43 Comparison of Disinfectants Used by Small and Large Surface Water Plants 3-45
Occurrence Assessment for the Final Stage 2 DBPR vi December 2005
-------
Exhibit 3.44 Comparison of Total Chlorine Doses in Large and Small Surface Water Plants Using Only
Chlorination (C12/C12) 3-46
Exhibit 3.45 Summary of NRWA DBF Occurrence Data by Plant 3-47
Exhibit 3.46 Summary of NRWA DBF Individual Observations 3-47
Exhibit 3.47 Distribution of TTHM Occurrence in Plant Finished Water 3-48
Exhibit 3.48 Distribution of TTHM Occurrence at the Point of Average Residence Time in the
Distribution System 3-48
Exhibit 3.49 Distribution of TTHM Occurrence at the Point of Maximum Residence Time in the
Distribution System 3-49
Exhibit 3.50 Distribution of HAA5 Occurrence in Plant Finished Water 3-49
Exhibit 3.51 Distribution of HAA5 Occurrence at the Point of Average Residence Time in the
Distribution System 3-50
Exhibit 3.52 Distribution of HAA5 Occurrence at the Point of Maximum Residence Time in the
Distribution System 3-50
Exhibit 3.53 Cumulative Distribution of Mean TTHM Occurrence in Distribution Systems for Small and
Large Surface Water Plants 3-52
Exhibit 3.54 RAA TTHM vs. RAA HAA5 for 96 Small Surface Water Plants 3-53
Exhibit 3.55 Percentage of DS Maximum Observations for TTHM and HAA5 by Sampling
Location 3-54
Exhibit 3.56 Frequency at Which Highest TTHM or HAA5 LRAAs Occurred at the Same Location for
All NRWA Plants 3-54
Exhibit 3.57 Annual Average TOC in Influent Water TOC for Ground Water Systems 3-55
Exhibit 3.58 Treatment Summary for Ground Water Systems (Chlorinating and Non-Chlorinating) . 3-56
Exhibit 3.59 Annual Average Finished Water TTHM for Ground Water Systems 3-56
Exhibit 3.60 Comparison of Effluent TOC for Chlorinating Small, Medium, and Large Ground Water
Systems 3-58
Exhibit 3.61 Cumulative Distribution of TTHM Occurrence as Distribution System Average for Small
and Large Ground Water Plants 3-58
Exhibit 3.62 Percent TOC Removal Requirements for Systems Employing
Enhanced Coagulation 3-59
Exhibit 3.63 Distribution of Monthly Influent TOC (mg/L) and Monthly Influent Alkalinity (mg/L)
Samples Based on ICR Data for All Large Plants 3-60
Exhibit 3.64 Count of Plants by Influent TOC and Bromide Concentrations Based on ICR Data for All
Large Plants 3-61
Exhibit 3.65 Finished Water TTHM Concentrations (Mean of Plant-Means) by Influent TOC and
Bromide Concentrations Based on ICR Data for All Large Plants 3-62
Exhibit 3.66 Finished Water TTHM Concentrations (90th Percentile of Plant-Means) by Influent TOC
and Bromide Concentrations Based on ICR Data for All Large Plants 3-63
Exhibit 3.67 Finished Water HAA5 Concentrations (Mean of Plant-Means) by Influent TOC and
Bromide Concentrations Based on ICR Data for All Large Plants 3-64
Exhibit 3.68 Finished Water HAA5 Concentrations (90th Percentile of Plant-Means) by Influent TOC and
Bromide Concentrations Based on ICR Data for All Large Plants 3-65
Exhibit 3.69 RAA of TTHM Occurrence versus RAA of HAA5 Occurrence for Large Surface Water
Plants Based on ICR Data (N = 213) 3-66
Exhibit 3.70 RAA of TTHM Occurrence versus RAA of HAA5 Occurrence for Large Ground Water
Plants Based on ICR Data (N = 82) 3-67
Exhibit 3.71 Highest LRAA TTHM versus Highest LRAA HAA5 for Large Surface Water Plants Based
on ICR Data (N = 213) 3-68
Exhibit 3.72 Highest LRAA TTHM versus Highest LRAA HAA5 for Large Ground Water Plants Based
on ICR Data (N = 82) 3-69
Exhibit 3.73 Single Highest TTHM versus Single Highest HAA5 for Large Surface Water Plants Based
on ICR Data (N = 213) 3-70
Occurrence Assessment for the Final Stage 2 DBPR vii December 2005
-------
Exhibit 3.74 Single Highest TTHM versus Single Highest HAAS Based on ICR Data for Large Ground
Water Plants (N = 82) 3-71
Exhibit 3.75a Influent Water TOC Occurrence Distribution for Large ICR Surface Water Systems . 3-72
Exhibit 3.75b Influent Water TOC Occurrence Distribution for Large ICR Ground Water Systems . 3-73
Exhibit 3.75c Influent Water TOC Occurrence Distribution for Ground Water Systems, Derived from the
GWSS 3-73
Exhibit 3.76a Mean Influent Bromide Concentrations, Large ICR Surface Water Plants 3-74
Exhibit 3.76b Mean Influent Bromide Concentrations, Large ICR Ground Water Plants 3-75
Chapter 4. National DBF Occurrence: Predicted Post-Stage 1 Baselines
Exhibit 4.1 ICR Matrix Method for Surface Water Plants for the Stage 1 DBPR (80/60 RAA), 20 Percent
Safety Margin 4-4
Exhibit 4.2 TTHM and HAAS Levels for Stage 2-Compliant Plants Using Chloramines and/or an
Advanced Technology 4-5
Exhibit 4.3 ICR Matrix Method for Ground Water Plants for the Stage 1 DBPR (80/60 RAA), 20 Percent
Safety Margin 4-6
Exhibit 4.4 TTHM and HAAS Levels for Stage 2-Compliant Ground Water Plants Using Chloramines
and/or an Advanced Technology 4-7
Exhibit 4.5 Summary of Post-Stage 1 TTHM Occurrence for ICR Plants, Stage 1 DBPR Safety Margin
of 20% 4-8
Exhibit 4.6 Summary of Post-Stage 1 HAAS Occurrence for ICR Plants,Stage 1 MCL Safety Margin of
20% 4-9
Exhibit 4.7a Analysis of Variability for Stage 2 Non-Compliant Surface Water Plants 4-10
Exhibit 4.7b Analysis of Variability for Stage 2 Non-Compliant Ground Water Plants 4-11
Exhibit 4.8a Cumulative Distribution of ICR LRAAmax - ICR LRAA2ndHl TTHM Screened Data, Surface
Water Plants 4-12
Exhibit 4.8b Cumulative Distribution of ICR LRAAmax - ICR LRAA2ndHl HAAS Screened Data, Surface
Water Plants 4-13
Exhibit 4.8c Cumulative Distribution of ICR LRAAmax - ICR LRAA2ndHl TTHM Screened Data, Ground
Water Plants 4-14
Exhibit 4.8d Cumulative Distribution of ICR LRAAmax - ICR LRAA2ndHl HAAS Screened Data, Ground
Water Plants 4-15
Exhibit 4.9a Temporal Variability in TTHM, Range in Individual Observations by Quarter 4-17
Exhibit 4.9b Temporal Variability in HAAS, Range in Individual Observations by Quarter 4-18
Exhibit 4.10 Cumulative Percentage of TTHM LRAAs, All Plants in Compliance with 64/48 RAA
(Stage 1 MCL with Safety Margin) 4-19
Exhibit 4.11 Cumulative Percentage of HAAS LRAAs, All Plants in Compliance with 64/48 RAA (Stage
1 MCL with Safety Margin) 4-20
Exhibit 4.12 Frequency at Which Highest TTHM LRAA Occurred at Each Sample Location for All
Screened ICR Plants 4-22
Exhibit 4.13 Location of Highest TTHM LRAA for Screened ICR Surface Water Plants by Disinfectant
Type 4-23
Exhibit 4.14 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant for Screened ICR
Ground Water Plants by Disinfectant Type 4-24
Exhibit 4.15 Frequency at Which Highest HAAS LRAA Occurred at Each Sample Location for All
Screened ICR Plants 4-25
Exhibit 4.16 Location of Highest HAAS LRAA for Screened ICR Surface Water Plants by Disinfectant
Type 4-26
Occurrence Assessment for the Final Stage 2 DBPR viii December 2005
-------
Exhibit 4.17 Location of Highest HAAS LRAA for Screened ICR Ground Water Plants by Disinfectant
Type 4-27
Exhibit 4.18 Frequency at Which Highest TTHM or HAAS LRAAs Occurred at the Same Location,
Plants in Compliance with 64/48 RAA (Stage 1 MCL with Safety Margin) 4-28
Occurrence Assessment for the Final Stage 2 DBPR ix December 2005
-------
Acronyms
AUXl
AUX2
AUX3
AUX4
AUX5
AUX6
AUX8
AVG1
AVG2
BAT
BCAA
BCAN
BDCAA
BDCM
BDL
C
CaCO3
CDBAA
CH
CHBr3
CHC13
Cl
C12
C12:NH3-N
C13
C1O2
C1O2
C1O3
CNC1
CP
CT
CWS
CWSS
DBAA
DBAN
DBCM
DBF
DBPR
DCAA
DCAN
DCP
DS
DSE
EPA
ESWTR
FACA
FBRR
FR
Auxiliary Database 1
Auxiliary Database 2
Auxiliary Database 3
Auxiliary Database 4
Auxiliary Database 5
Auxiliary Database 6
Auxiliary Database 8
Average 1 Distribution System Sampling Location
Average 2 Distribution System Sampling Location
Best Available Technology
Bromochloracetic Acid
Bromochloroacetonitrile
Bromodichloroacetic Acid
Bromodichloromethane
Below Detection Limit
Carbon
Calcium Carbonate
Chlorodibromoacetic Acid
Chloral Hydrate
Bromoform
Chloroform
Chloride
Chlorine
Chlorine to Ammonia Nitrogen ratio
Trichloride
Chlorine Dioxide
Chlorite
Chlorate
Cyanogen Chloride
Chloropicrin
Disinfectant Residual Concentration* Contact Time
Community Water System
Community Water System Survey
Dibromoacetic Acid
Dibromoacetonitrile
Dibromochloormethane
Disinfection Byproducts
Disinfectants/Disinfection Byproducts Rule
Dichloroacetic Acid
Dichloroacetonitrile
Dichloropropanone
Distribution System
Distribution System Equivalent
U.S. Environmental Protection Agency
Enhanced Surface Water Treatment Rule
Federal Advisory Committee Act
Filter Backwash Recycling Rule
Federal Register
Occurrence Assessment for the Final Stage 2 DBPR x
December 2005
-------
GWSS Ground Water Supply Survey
GWUDI Ground Water Under the Direct Influence of Surface Water
HAA Haloacetic Acid
HAAS Haloacetic Acid-Five
HAA6 Haloacetic Acid-Six
HAA9 Haloacetic Acid-Nine
HAN Haloacetonitrile
HAN4 Haloacetonitriles-Four
HC1 Hydrochloric Acid
HOC1 Hypochlorous Acid
ICR Information Collection Rule
ICRFED Information Collection Rule Federal Database System
ICRSS Information Collection Rule Supplemental Surveys
IDSE Initial Distribution System Evaluation
IESWTR Interim Enhanced Surface Water Treatment Rule
kg Kilogram
L Liter
LRAA Locational Running Annual Average
LTIESWTR Long Term 1 Enhanced Surface Water Treatment Rule
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
M-DBP Microbial and Disinfection Byproduct
MBAA Monobromoacetic Acid
MCAA Monochloroacetic Acid
MCLG Maximum Contaminant Level Goal
MCL Maximum Contaminant Level
MDL Method Detection Limit
mg Milligram
(ig Microgram
MRDL Maximum Residual Disinfectant Level
MRDLG Maximum Residual Disinfectant Level Goal
MRL Minimum Reporting Level
N Nitrogen
Na Sodium
NH3 Ammonia
NODA Notice of Data Availability
NOM Natural Organic Matter
NPDWR National Primary Drinking Water Regulation
NRWA National Rural Water Association
NTNCWS Nontransient Noncommunity Water System
NTU Nephelometric Turbidity Units
O2 Oxygen
O3(aq) Aqueous Ozone
OCl" Hypochlorite Ion
OGWDW Office of Ground Water and Drinking Water
PAC Powdered Activated Carbon
ppb Parts per billion
ppm Parts per million
PWS Public Water System
RPD Relative Percent Difference
QA/QC Quality Assurance and Quality Control
RAA Running Annual Average
SDS Simulated Distribution System
Occurrence Assessment for the Final Stage 2 DBPR xi
December 2005
-------
SDWA
SQL
SUVA
SWAT
SWTR
TBAA
TCAA
TCAN
TCR
TCP
THM
TNCWS
TOC
TOX
TTHM
TWO
uv
voc
WTP
Safe Drinking Water Act
Structured Query Language
Specific UV Absorbance
Surface Water Analytical Tool
Surface Water Treatment Rule
Tribromoacetic Acid
Trichloroacetic Acid
Trichloroacetonitrile
Total Coliform Rule
Trichloropropanone
Trihalomethane s
Transient Noncommunity Water System
Total Organic Carbon
Total Organic Halides
Total Trihalomethanes
Technical Working Group
Ultraviolet Radiation
Volatile Organic Compound
Water Treatment Plant
Occurrence Assessment for the Final Stage 2 DBPR xii
December 2005
-------
1. Introduction
The United States Environmental Protection Agency (EPA) Office of Ground Water and Drinking
Water (OGWDW) is developing interrelated drinking water regulations to control microbial pathogens,
residual disinfectants, and disinfection byproducts in drinking water. These rules are required by the Safe
Drinking Water Act (SDWA) Amendments of 1996 and are collectively known as the microbial and
disinfection byproducts (M-DBP) rules.
The Stage 1 Disinfectants and Disinfection Byproducts Rule (Stage 1 DBPR) and the Interim
Enhanced Surface Water Treatment Rule (IESWTR), the first set of M-DBP rules under the SDWA
Amendments, were promulgated in December 1998. The Stage 1 DBPR and the IESWTR were the
culmination of a 6-year rule development process that included regulatory negotiations with representatives
of the water industry, environmental and public health groups, and local, State1, and Federal government
agencies.
To support rule development, EPA expanded its microbial and disinfection byproduct (DBP)
research program and entered into collaborative efforts with other agencies and the water industry to collect
data. This data collection effort included the Information Collection Rule (ICR) and the ICR Supplemental
Survey (ICRSS). In addition, under a joint effort between EPA and the National Rural Water Association
(NRWA), NRWA State chapters conducted a survey of disinfection byproduct and treatment information
at small public water systems (PWSs)2.
EPA has worked with stakeholders under the Federal Advisory Committee Act (FACA) to develop
the proposed Stage 2 Disinfection Byproducts Rule (Stage 2 DBPR) and Long Term 2 Enhanced Surface
Water Treatment Rule (LT2ESWTR). These rules have been developed concurrently, using occurrence
data from the ICR and other available sources to ensure that microbial protection is maintained or
enhanced while exposure to DBFs is reduced.
This occurrence assessment supports the Stage 2 DBPR. The document has been revised since the
proposal to reflect both public and peer review comments, as well as to maintain consistency with the Stage
2 Economic Analysis (USEPA 2005a). The remainder of this chapter is organized as follows:
Section 1.1 summarizes the purpose of this document.
Section 1.2 describes the history of drinking water regulations leading up to the Stage 2
DBPR.
• Section 1.3 provides a brief synopsis of the factors affecting DBP formation.
• Section 1.4 describes the main data source, the ICR.
'For the purposes of this document, "States" are defined as States or territories with primacy or other primacy agencies.
PWSs are systems that provide water for human consumption through pipes or other constructed conveyances and that
have at least 15 service connections or regularly serve an average of at least 25 individuals per day for at least 60 days per year.
Occurrence Assessment for the Final Stage 2 DBPR 1-1 December 2005
-------
• Section 1.5 describes other sources used.
• Section 1.6 describes the rest of the chapters and appendices that make up this document.
1.1 Purpose of the Occurrence Document
This document serves two main purposes. First, it presents new data and analyses as an addendum
to the Occurrence Assessment for Disinfectants/Disinfection Byproducts in Public Drinking Water
Supplies document (USEPA 1998c), which supported the Stage 1 DBPR. In order to update the 1998
document to support the current rulemaking, EPA conducted additional information searches to identify
articles and studies from the scientific literature and from recent conferences in the relevant subject areas.
EPA also used the results of the ICR data collection effort, which was the primary source of the new DBP
occurrence data. Analyses of this new DBP information and ICR data have been incorporated into the
document.
The second purpose of the document is to evaluate DBP occurrence to characterize the post-Stage
1 baseline occurrence conditions. Because the compliance deadline for the Stage 1 DBPR was relatively
recent (January 2002 for medium and large surface water systems and January 2004 for ground water and
small surface water systems), all observed data in this document represent pre-Stage 1 conditions (i.e.,
conditions before the implementation of the Stage 1 DBPR). Chapter 4 of this occurrence document
provides one possible analysis of DBP formation and occurrence for post-Stage 1 DBPR conditions.
To provide support to the Stage 2 DBPR rulemaking, this document focuses on analyses of the
following data:
• Disinfectant use and residual concentrations.
• DBP precursors and other water quality parameters affecting DBP formation.
• Occurrence of regulated DBFs
Total Trihalomethanes (TTHM)
Haloacetic Acid-Five (HAAS)
Bromate
Chlorite
Analyses of TTHM and HAAS occurrence focus on the distribution system. Spatial and temporal
variability of TTHM and HAAS occurrence in the distribution system is evaluated for the post-Stage 1
DBPR baseline in Chapter 4. Speciation of TTHM and HAAS are contained in Appendix A. The ICR
also contains other, non-regulated DBP data that is briefly summarized in section 3.1.3.1. Alternative and
additional analyses are presented in the Information Collection Rule Data Analysis document (McGuire et
al. 2002) (this includes some analyses of water quality data collected under the ICR that are not relevant to
the Stage 2 DBPR rulemaking).
Occurrence Assessment for the Final Stage 2 DBPR 1-2 December 2005
-------
1.2 Regulatory Background
1.2.1 Statutory Authority for Promulgating the Rule
The primary responsibility for regulating the quality of drinking water lies with EPA. The SDWA
establishes this responsibility and defines the mechanisms at the Agency's disposal to protect public health.
EPA sets standards by identifying which contaminants should be regulated and by establishing the
maximum levels of the contaminants allowed in drinking water.
Section 1412(b)(l) of the 1996 SDWA reauthorization mandated new drinking water requirements.
EPA's general authority to set Maximum Contaminant Level Goals (MCLGs) and develop the National
Primary Drinking Water Regulations (NPDWRs) was modified to apply to contaminants that "may have
an adverse effect on the health of persons," are "known to occur or there is a substantial likelihood that the
contaminant will occur in public water systems with a frequency and at levels of public health concern,"
and for which, "in the sole judgment of the Administrator, regulation of such contaminant presents a
meaningful opportunity for health risk reductions for persons served by public water systems" (SDWA
To regulate a contaminant, EPA sets an MCLG at a level at which no known or anticipated
adverse health effects occur. MCLGs are established solely on the basis of protecting public health and are
not enforceable. EPA simultaneously sets an enforceable Maximum Contaminant Level (MCL) as close as
technologically feasible to the MCLG, while taking costs into consideration. If it is not feasible to measure
the contaminant at levels presumed to have impacts on health, a treatment technique can be specified in
place of an MCL. For water systems, compliance with a drinking water regulation means either not
exceeding the MCL or meeting treatment technology requirements.
Additionally, EPA identifies maximum concentrations of residual disinfectants that can occur in
water without harming human health and sets Maximum Residual Disinfectant Level Goals (MRDLGs)
and Maximum Residual Disinfectant Levels (MRDLs). PWSs maintain residual levels of disinfectants in
the distribution system, following treatment, to ensure consumer protection from microbial contaminants.
Like MCLGs, MRDLGs are not enforceable, while MRDLs are.
In addition to the general authorities cited above, SDWA 1412(b)(2)(C) requires specifically that
EPA promulgate the Stage 2 DBPR.
The Administrator shall promulgate an Interim Enhanced Surface Water Treatment Rule,
a Final Enhanced Surface Water Treatment Rule, a Stage 1 Disinfectants and Disinfection
Byproducts Rule, and a Stage 2 Disinfectants and Disinfection Byproducts Rule in
accordance with the schedule published in Volume 29, Federal Register, Page 6361
(February 10, 1994), in Table III. 13 of the proposed Information Collection Rule.
(SDWA 1412(b)(2)(C))
The following sections summarize the development of relevant NPDWRs over the past 20 years.
1.2.2 1979 Total Trihalomethane Rule
Under the Total Trihalomethane Rule (44 Federal Register (FR) 68624, November 29, 1979), EPA
set an MCL for TTHM (the sum of the concentrations of chloroform, bromoform, bromodichloro-methane,
Occurrence Assessment for the Final Stage 2 DBPR 1-3 December 2005
-------
and dibromochloromethane) of 0.10 milligrams per liter (mg/L) as a running annual average (RAA) of
quarterly measurements. This standard applied to CWSs using surface or ground water that served at least
10,000 people and that added a disinfectant to the drinking water during any part of the treatment process.
This 1979 rule was superseded by the 1998 Stage 1 DBPR (section 1.2.9) with which all CWSs and
NTNCWSs must have complied by January 2004.
1.2.3 1989 Total Coliform Rule
The Total Coliform Rule (TCR) (54 FR 27544, June 29, 1989) applies to all PWSs. Because
monitoring PWSs for every possible pathogenic organism is not feasible, coliform organisms are used as
indicators of possible contamination. Coliforms are easily detected in water and are used to indicate a
system's vulnerability to pathogens. In the TCR, EPA set an MCLG of zero for total coliforms. EPA also
set a monthly MCL for total coliforms and required testing of total-coliform-positive cultures for the
presence of E. coll or fecal coliforms. E.coli and fecal coliforms indicate more immediate health risks
from sewage or fecal contamination and are used as the indicator of an acute MCL violation. Coliform
monitoring frequency is determined by population served, the type of system (community or
noncommunity) and the type of source water (surface water or ground water). In addition, the TCR
required sanitary surveys every 5 years (or 10 years for noncommunity systems using disinfected ground
water) for systems that collect fewer than 5 routine total coliform samples per month (typically systems
serving fewer than 4,100 people).
1.2.4 1989 Surface Water Treatment Rule
Under the Surface Water Treatment Rule (SWTR) (54 FR 27486, June 29, 1989), EPA set
MCLGs of zero for Giardia lamblia, viruses, and Legionella and established requirements for all PWSs
using surface water or GWUDI as a source. The SWTR includes treatment technique requirements for
filtered and unfiltered systems that are intended to protect against the adverse health effects associated with
Giardia lamblia, viruses, and Legionella, as well as many other pathogenic organisms. These
requirements include:
• Maintenance of a disinfectant residual in water entering and within the distribution system.
• Removal or inactivation of at least 99.9 percent (3 logs) of Giardia and 99.99 percent (4 logs)
of viruses.
For filtered systems, meeting a turbidity performance standard for the combined filter effluent
of 5 nephelometric turbidity units (NTUs) as a maximum and 0.5 NTU in 95 percent of
monthly measurements, based on 4-hour monitoring for treatment plants using conventional
treatment or direct filtration (with separate standards for other filtration technologies). These
requirements were enhanced by the 1998 Interim Enhanced Surface Water Treatment Rule
(IESWTR) and the 2002 Long Term 1 Enhanced Surface Water Treatment Rule
(LT1ESWTR).
Watershed control programs and other requirements for unfiltered systems.
Occurrence Assessment for the Final Stage 2 DBPR 1-4 December 2005
-------
1.2.5 1996 Information Collection Rule
The Information Collection Rule (ICR) (61 FR 24354, May 14, 1996) applied to PWSs serving
more than 100,000 people. A more limited set of ICR requirements covered ground water systems serving
50,000 to 100,000 people.
The ICR authorized EPA to collect occurrence and treatment information from water treatment
plants to help evaluate the possible need for changes to microbial requirements and microbial treatment
practices and to help evaluate the need for future regulation of disinfectants and DBFs. The ICR provided
EPA with information on the national occurrence of (1) chemical byproducts that form when disinfectants
used for microbial control react with naturally occurring compounds and ions present in source water; and
(2) disease-causing microorganisms including Cryptosporidium, Giardia, viruses, and coliform bacteria.
The ICR also mandated the collection of data on how water systems currently treat for contaminants. The
ICR monthly sampling data provided 18 months of information on the quality of the influent and treated
water, including pH, alkalinity, turbidity, temperature, calcium, total hardness, total organic carbon,
ultraviolet254 (UV) absorbency, bromide, ammonia, and disinfectant residual. These data provide some
indication of the "treatability" of the water, the occurrence of contaminants, and the potential for DBF
formation. The data collected under the ICR are being analyzed to help develop the LT2ESWTR and
Stage 2 DBPR. A detailed description of the ICR is provided in Section 1.4.
1.2.6 1998 Interim Enhanced Surface Water Treatment Rule
The IESWTR (63 FR 69478, December 16, 1998) enhances the 1989 SWTR It applies to PWSs
serving at least 10,000 people and using surface water or GWUDI as a source. These systems began
compliance with the IESWTR in January 2002. The purpose of the IESWTR is to improve control of the
protozoan Cryptosporidium and to address tradeoffs between the risks of microbial pathogens and those of
DBFs. The requirements and guidelines include:
• An MCLG of zero for Cryptosporidium.
• Removal of 99 percent (2 logs) of Cryptosporidium for systems that use filters.
For filtered systems, a turbidity performance standard for the combined filter effluent of 1
NTU as a maximum and 0.3 NTU as a minimum in 95 percent of monthly measurements,
based on 4-hour monitoring for treatment plants using conventional treatment or direct
filtration.
• Continuous monitoring of individual filter effluent in conventional and direct filtration plants
and recording turbidity readings every 15 minutes when these filters are on-line.
A disinfection benchmark to assess the level of microbial protection provided before facilities
change their disinfection practices to meet the requirements of the Stage 1 DBPR.
• Inclusion of Cryptosporidium in the definition of GWUDI and in the watershed control
requirements for unfiltered PWSs.
Covers for all new finished water storage facilities.
Occurrence Assessment for the Final Stage 2 DBPR 1-5 December 2005
-------
• A primacy provision that requires States to conduct sanitary surveys for all surface water
systems, including those serving fewer than 10,000 people.
The IESWTR was promulgated concurrently with the Stage 1 DBPR so that systems could
coordinate their response to the risks posed by DBFs and microbial pathogens.
1.2.7 1998 Stage 1 Disinfectants and Disinfection Byproducts Rule
The Stage 1 DBPR (63 FR 69390, December 16, 1998) applies to all CWSs and NTNCWSs that
add a chemical disinfectant to their water. Certain requirements designed to provide protection against
acute health effects from chlorine dioxide also apply to transient noncommunity water systems (TNCWSs).
Compliance for surface water and GWUDI systems serving at least 10,000 people began in January 2002.
Surface water and GWUDI systems serving fewer than 10,000 people and all ground water systems were
required to comply by January 2004.
The Stage 1 DBPR sets MRDLGs for chlorine (4 mg/L as chlorine (C12)), chloramines (4.0 mg/L
as C12), and chlorine dioxide (0.8 mg/L as C1O2); and MCLGs for bromodichloromethane (0 mg/L),
bromoform (0 mg/L), dibromochloromethane (0.06 mg/L), dichloroacetic acid (0 mg/L), trichloroacetic
acid (0.3 mg/L), bromate (0 mg/L), and chlorite (0.8 mg/L). The rule sets MRDLs for chlorine (4.0 mg/L
as C12), chloramines (4.0 mg/L as C12), and chlorine dioxide (0.8 mg/L as C1O2); and MCLs for TTHM
(0.080 mg/L), HAAS (0.060 mg/L), bromate (0.010 mg/L), and chlorite (1.0 mg/L). The MRDLs and
MCLs, except those for chlorite and chlorine dioxide, are calculated as RAAs. For conventional surface
water and GWUDI systems, a treatment technique—enhanced coagulation/softening—is specified for the
removal of DBP precursors.
As noted in section 1.2.8, the Stage 1 DBPR was promulgated concurrently with the IESWTR to
coordinate the control of DBFs and microbial contaminants.
1.2.8 2000 Proposed Ground Water Rule
The proposed Ground Water Rule (65 FR 30194, May 10, 2000) addresses fecal contamination in
ground water systems. It also builds on the TCR through provisions based on further evaluation of E. coll
monitoring results measured under the TCR. Key components of the approach for protection of ground
water included in the proposed rule are:
Sanitary surveys for all ground water systems.
Hydrogeologic sensitivity assessments to identify ground water wells that are susceptible to
fecal contamination.
• Triggered source water monitoring for an indicator of fecal contamination for all systems that
do not achieve 4-log treatment, and in addition, routine source water monitoring for an
indicator of fecal contamination that have been determined to draw from sensitive ground
water sources.
Occurrence Assessment for the Final Stage 2 DBPR 1-6 December 2005
-------
Correction of significant deficiencies and fecal contamination by eliminating the source of
contamination, correcting the deficiency, providing an alternative source of water, or providing
inactivation and/or removal of 99.99 percent (4 logs) of viruses.
Compliance monitoring to ensure that disinfection treatment is reliably operated when it is
used.
1.2.9 2001 Arsenic Rule
The Arsenic Rule (66 FR 6976, January 22, 2001) increases the level of public health protection
against exposure to arsenic in drinking water. The rule revises the MCL for arsenic in drinking water from
0.05 mg/L to 0.010 mg/L and sets an MCLG of 0 mg/L for all CWSs and NTNCWSs. Clarification on
how compliance is demonstrated for many inorganic and organic contaminants in drinking water is also
given. All existing CWSs and NTNCWSs must comply with the Arsenic Rule by January 23, 2006.
1.2.10 2001 Filter Backwash Recycling Rule
The Filter Backwash Recycling Rule (FBRR) (66 FR 31086, June 8, 2001) regulates systems that
return filter backwash to the treatment process. The rule applies to surface water and GWUDI systems
that use direct or conventional filtration and recycle spent filter backwash water, sludge thickener
supernatant, or liquids from dewatering processes. The rule requires that these recycled liquids be returned
to a location such that all steps of a system's conventional or direct filtration are employed. The rule also
requires systems to notify the State that they practice recycling. Finally, systems must collect and maintain
information for review by the State.
1.2.11 2002 Long Term 1 Enhanced Surface Water Treatment Rule
The LT1ESWTR (67 FR 1812, January 14, 2002) enhances the 1989 SWTR requirements for
small systems. LT1ESWTR enhances control of Cryptosporidium and other disease-causing microbes for
surface water and GWUDI systems that serve fewer than 10,000 people. Key provisions in the
LT1ESWTR are very similar to those for the IESWTR, but provide additional flexibility for small systems.
1.2.12 2005 Long Term 2 Enhanced Surface Water Treatment Rule
Promulgated in concert with the Stage 2 DBPR, the LT2ESWTR strengthens control of
Cryptosporidium, and applies to all PWSs that use surface water or GWUDI as a source. It incorporates
system-specific treatment requirements based on a "Microbial Framework" approach that targets high-risk
systems. This approach involves assigning systems to different categories (or "bins") based on the levels of
Cryptosporidium found in the source water. Additional treatment requirements, if any, are linked to the
level of Cryptosporidium. A system will choose technologies and management practices from a "toolbox"
of options appropriate to its bin.
Medium and large systems (those serving at least 10,000 people) that filter will be required to
conduct Cryptosporidium source water monitoring for 24 months to determine their bin classification.
Small systems (those serving fewer than 10,000 people) that filter will monitor E. coli bacteria in their
Occurrence Assessment for the Final Stage 2 DBPR 1-7 December 2005
-------
source water biweekly for 12 months. Based on their E. coll results, they may be required to monitor
Cryptosporidium as well.
In addition to requirements for filtered systems, the LT2ESWTR will require unfiltered systems to
continue to meet the filtration avoidance criteria under the 1989 SWTR and provide inactivation at 4 logs
(99.99 percent) for virus, 3 logs (99.9 percent) for Giardia, and 2 to 3 logs (99 to 99.9 percent) for
Cryptosporidium (depending on results of Cryptosporidium monitoring of source water). Building on the
SWTR requirements, inactivation requirements for unfiltered systems subject to the LT2ESWTR must be
met using a minimum of two disinfectants.
Also, the LT2ESWTR will require systems with uncovered finished water reservoirs to cover the
reservoirs or treat reservoir discharge to the distribution system to achieve 4-log virus inactivation, 3-log
Giardia inactivation, and 2-log Cryptosporidium inactivation.
1.3 Factors Affecting DBF Formation
Organic DBFs (and oxidation byproducts) are formed by the reaction between organic substances
and oxidizing agents that are added to water during treatment. In most water sources, natural organic
matter (NOM) is the major constituent of organic substances and DBF precursors. Organic substances and
DBF precursors in water also come from a variety of other sources, including stormwater and wastewater.
NOM is typically measured as TOC and as such the two terms are used interchangeably in much of the
discussion presented here. Major factors affecting the type and amount of DBFs formed include:
• Type, dose, and residual concentration of disinfectant.
• Contact time and mixing conditions between disinfectant (oxidant) and precursors.
• Concentration and characteristics of precursors, such as TOC.
• Water temperature.
• Water chemistry (including pH, bromide ion concentration, organic nitrogen concentration, and
presence of other reducing agents such as iron and manganese).
A description of these factors follows.
1.3.1 Impact of Disinfection Method on Organic DBF Formation
Organic DBFs can be subdivided into halogenated and non-halogenated byproducts. Halogenated
organic disinfection byproducts are formed when organic compounds in water react with free chlorine, free
bromine, or free iodine. The formation reactions may take place in the treatment plant or the distribution
system. Free chlorine can be introduced to water directly as a primary or secondary disinfectant, or as a
byproduct in the manufacture of chlorine dioxide. In general, primary disinfection kills or inactivates
pathogens in water at the water treatment plant. Secondary disinfection provides a residual as the water
leaves the treatment plant to control microbiological growth in the distribution system. Reactions between
NOM and chlorine lead to the formation of a variety of halogenated DBFs including THMs and HAAs.
Occurrence Assessment for the Final Stage 2 DBPR 1-8 December 2005
-------
Free chlorine and ozone oxidize bromide ion to hypobromite ion/hypobromous acid, which in turn
can react with NOM to form brominated DBFs (e.g., bromoform). The presence of bromide affects both
the formation rate and yield of DBFs. As the ratio of bromide to NOM increases, the percentage of
brominated DBFs increases. For example, Krasner (1999) reported the rate of THM formation is higher in
waters with increased concentrations of bromide. Brominated DBFs can also form by bromine substitution
in the chlorinated byproducts, with hypobromous acid an effective substituting agent (Krasner 1999). The
presence of OBr- in chlorine feedstocks may also contribute to formation of brominated DBFs.
Non-halogenated DBFs may form when precursors (NOMs) react with strong oxidants. For
example, the reaction of organics with ozone and hydrogen peroxide results in the formation of aldehydes,
aldo- and keto-acids, and organic acids (Singer 1999). Chlorine can also trigger the formation of some
non-halogenated DBFs (Singer and Harrington 1993). Many of the non-halogenated DBFs, such as
aldehydes, are biodegradable.
Studies have documented that chloramines produce significantly lower DBF levels than free
chlorine, and there is no clear evidence that the reaction of NOM and chloramine leads to the formation of
THMs (Singer and Reckhow 1999; USEPA 1999a). An empirical DBF formation model calibrated using
ICR data predicted THM and HAA formation in full-scale plants and distribution systems under
chloraminated conditions at fractions of the amount that would be expected based on observations of DBF
formation under free chlorine conditions. The amount of DBF formation under chloraminated conditions
varied from 5 percent to 35 percent of that calculated for free chlorine, depending on the individual DBF
species (Swanson et al. 2001).
It is possible that DBFs might form during the mixing of chlorine and ammonia, when free chlorine
might react with NOM before the complete formation of chloramines. In addition, monochloramine slowly
hydrolyzes to release free chlorine in water. This free chlorine may contribute to the formation of small
amounts of additional DBFs in the distribution system. Low DBF formation due to chloramines can be an
acceptable alternative, especially at the farthest locations of the distribution system where high DBF
formations could potentially occur in the presence of free chlorine residuals.
The use of chlorine dioxide as a disinfectant does not produce significant amounts of organic
halogenated DBFs. Only small amounts of total organic halides (TOXs, the class of DBFs made up of
halogenated organic by-products that include THMs and HAAs) are formed. However, sometimes excess
chlorine is added to water to ensure complete reaction with sodium chlorite during the production of
chlorine dioxide. This excess chlorine, in the presence of NOM, can cause THMs and HAAs to form.
To date, there is no evidence to suggest that use of UV as a disinfectant results in the formation of
any disinfection byproducts; however, little research has been performed in this area. Most of the research
regarding application of UV and DBF formation has focused on chlorinated DBF formation as a result of
UV application prior to the addition of chlorine or chloramines. The evidence suggests UV does not
promote chlorinated DBF formation.
Ozone does not produce chlorinated DBFs; however, ozone can alter the reactions between chlorine
and NOM, and affect the speciation of chlorinated DBFs if chlorine is subsequently added downstream. In
waters with sufficient bromide concentrations, ozonation can lead to the formation of bromate and other
brominated DBFs. Bromate, like THMs and HAAs, is a regulated DBF. Ozonation of natural waters also
produces aldehydes, haloketones, ketoacids, carboxylic acids, and other types of biodegradable organic
material. The biodegradable fraction of organic material can serve as a nutrient source for
microorganisms, and should be removed to prevent microbial regrowth in the distribution system. This is
generally accomplished by allowing filters to run biologically, i.e., without a disinfectant residual present.
Occurrence Assessment for the Final Stage 2 DBPR 1-9 December 2005
-------
1.3.2 Disinfectant Dose
The concentration of disinfectant can affect the formation of DBFs. In general, changes in the
disinfectant dose have a great impact on DBF formation during primary disinfection because the
disinfectant is typically the limiting reactant in DBF formation reactions. The effect of disinfection dose on
DBF formation is typically more significant in primary disinfection than in secondary disinfection During
secondary disinfection, DBF formation reactions may be precursor-limited since an excess of disinfectant is
added to the water. In the distribution system, DBF formation reactions become disinfectant-limited when
the free chlorine residual drops to low levels. (Singer and Reckhow [1999] suggested a free chlorine
residual concentration of 0.3 mg/L as a rule of thumb.)
In many systems, booster disinfection is applied to raise disinfectant residual concentration,
especially in remote areas of the distribution system or near storage tanks where water age may be high and
disinfectant residuals can be low. The additional chlorine dose applied to the water at these booster
facilities can result in increased formation of THMs and HAAs. Further, booster chlorination (increasing
chlorine concentrations in the distribution system) can maintain high HAA concentrations because the
increased disinfection residuals can prevent the biodegradation of HAAs.
1.3.3 Contact Time and DBF Formation
DBFs continue to form in drinking water as long as disinfectant residuals and reactive DBF
precursors are present. Generally, the longer the contact time between disinfectant/oxidant and NOM, the
greater the amount of DBFs that can be formed. In the presence of a continuing significant disinfectant
residual, both THMs and HAAs have generally high chemical stabilities and will persist after formation
(Singer and Reckhow 1999).
However, there are some chemical stability differences between THMs and HAAs which can result
in differences in long term accumulations. High TTHM values usually occur at points in the
distribution system with the longest total residence time (the "oldest" water age). In contrast, high HAAs
values cannot be consistently related to water age because HAAs are known to biodegrade over time when
the disinfectant residual is low. This might result in relatively low HAA concentrations in areas of the
distribution system where disinfectant residuals are depleted.
In contrast to these chlorination byproducts, ozonation byproducts form more rapidly. However,
because residual ozone dissipates rapidly in water, there is a much shorter period during which ozone
byproduct formation can occur compared to chlorination byproducts (Singer and Reckhow 1999). The rate
of formation for both THMs and HAAs is relatively slow-on the order of days for ultimate formation.
Bromate formation, however, is considerably faster-on the order of seconds.
1.3.4 Concentration and Characteristics of Precursors
In addition to disinfectant dose and contact time, the formation of halogenated DBFs is related to
the concentration of NOM at the point of chlorination. Higher amounts of DBFs are formed in waters with
higher concentrations of precursors. Studies conducted with different fractions of NOM have indicated the
reaction between chlorine and NOM with high aromatic content tends to form higher DBF levels than
NOM with low aromatic content (Singer and Reckhow 1999). For this reason, UV absorbance (typically
indicated by UV absorbance at 254 nm [UV254]), which is generally related to the aromatic and unsaturated
Occurrence Assessment for the Final Stage 2 DBPR 1-10 December 2005
-------
components of NOM, is considered a good predictor of the tendency of a source water to form THMs and
HAAs (Owen et al. 1998; Singer and Reckhow 1999). The UV254 measurement immediately upstream of
the point(s) of chlorination within a treatment plant is, therefore, most directly related to THM and HAA
formation potential. It should be noted that the more highly aromatic precursors, characterized by high
UV254 in source waters, are more easily removed by coagulation.
1.3.5 Water Temperature
The rate of formation of THMs increases with increasing temperature. HAA formation rates may
also increase with temperature, though the effects are less pronounced. Consequentially, the highest THM
and HAA levels may occur in the warm summer months. However, water demands are often higher in
warmer months, resulting in lower water age within the distribution system and helping to control DBF
formation. Furthermore, high temperature conditions in the distribution system promote the accelerated
depletion of residual chlorine, which can mitigate DBF formation and promote biodegradation of HAAs
(unless chlorine dosages are increased to maintain high residuals) (Singer and Reckhow 1999). For these
reasons, depending on system-specific conditions, the highest THM and HAA levels may be observed
during months which are warm, but not necessarily the warmest.
Seasonal trends affect where high THM and HAA concentrations might be found. For example,
when water is colder, microbial activity is typically lower and DBF formation kinetics are slower. Under
these conditions, the highest THM and HAA concentrations might appear coincident with the oldest water
in the system. In warmer water, the highest HAA concentrations might appear in fresher (younger) water,
which is likely to contain higher disinfectant residuals that can prevent the biodegradation of HAAs.
1.3.6 Water pH
In the presence of NOM and chlorine, THM formation increases with increasing pH, whereas the
formation of HAAs and other DBFs increase with decreasing pH. The increase of THMs at higher pH
values is likely due to base catalyzed reactions that lead to THM formation. The HAA formation pathway
can be altered at high pH since their precursors can hydrolyze (Singer and Reckhow 1999).
The major byproducts of ozonation are not affected by base hydrolysis. However, the rate of
decomposition of ozone to hydroxyl radical is accelerated as pH increases. The increase in pH has been
shown to result in a decrease in aldehydes, though there may be circumstances where the increased pH will
lead to the formation of some carbonyl byproducts. More typically, low pH in ozone treated water has
been shown to increase the formation of brominated DBF species. This is due to two factors. First,
hypobromous acid and hypobromite are formed when ozone reacts with bromide water. Second, low pH
shifts the equilibrium to hypobromous acid, which reacts with NOM to form brominated DBFs such as
bromoform and dibromoacetic acid, but lowers the formation of hypobromite and subsequently bromate
(Singer and Reckhow 1999).
1.4 The Primary Data Source: Information Collection Rule
The main source of occurrence data for large PWSs is the ICR. ICR monitoring requirements
applied to surface and ground water CWSs serving at least 100,000 people, which included a total of 296
systems comprising 512 plants (which includes 11 plants with blended source water). The ICR generated
data sets that characterize the water quality in each plant's source water, in several steps in the treatment
Occurrence Assessment for the Final Stage 2 DBPR 1-11 December 2005
-------
process, and at several points in the distribution system (reflecting finished water). The monitoring began
in July 1997 and ended in December 1998.
This section describes the ICR data elements and other data requirements, ICR implementation
activities, ICR sampling plans, data management activities, quality assurance activities, and the
development of the auxiliary databases. The last two subsections (1.4.8 and 1.4.9) describe the methods
used to analyze ICR data and how the data analyses are documented. Appendix C summarizes the uses of
the ICR data and documents how data quality objectives are met. For more detailed information on the
ICR data collection methodology and results, refer to the following publications:
• ICR Sampling Manual (USEPA 1996b)
• Information Collection Rule Data Analysis (McGuire et al. 2002)
1.4.1 Description of the ICR Data Set
ICR monitoring requirements depended on the system size and type. Surface water systems
serving more than 100,000 people were required to monitor for DBFs and related parameters (e.g., DBF
precursors, disinfectants), conduct microbial monitoring, collect treatment plant design and operating
information, and monitor for treatment study applicability (which determined if a treatment study was
required). Ground water systems serving more than 100,000 people were required to monitor for DBFs
and related parameters and for treatment study applicability. Ground water systems serving more than
50,000 but fewer than 100,000 people were required to monitor only for treatment study applicability.
The following subsections describe ICR analytical requirements, sample locations, monitoring
frequency, and minimum reporting levels. A summary of all requirements follows the discussion (Exhibit
1.1). For more detailed information, such as specific treatment sampling locations, refer to the ICR
Sampling Manual (USEPA 1996b).
Analytical Requirements
Samples were analyzed for the following:
Water quality parameters, including DBF precursors (temperature, pH, alkalinity, total
organic carbon, etc.)
Disinfectants (free chlorine residual, chloramine residual, etc.)
• DBFs (TTHM and individual THM species such as chloroform; HAAS, HAA6, HAA9, and
individual HAA species; chlorite; bromate, etc.)
Since DBF formation depends on the type of disinfectant used, monitoring for each DBF did not occur at
every plant. For example, chlorite is a byproduct primarily related to disinfection with chlorine dioxide and
was therefore monitored only by plants that use chlorine dioxide.
Microbial analyses were also performed on some samples. Microbial results are not covered in
this document; see the LT2ESWTR Occurrence Document (USEPA 2005b) for microbial regulations and
results from the ICR.
Occurrence Assessment for the Final Stage 2 DBPR 1-12 December 2005
-------
Sample Locations
ICR samples were generally collected at the treatment plant influent, sites throughout the
treatment plant (e.g., before and after filtration, before and after each point of disinfection), a finished
water location (typically the same as the entry point sample except for plants that blend finished water from
multiple treatment plants), and sites within the distribution system. A total of four distribution system
monitoring locations were required for the ICR:
• Average Residence Time in the Distribution System (AVG1 and AVG2): two sample
points in the distribution system, each representing an approximate average residence time, as
designated by the water system.
• Maximum Residence Time in the Distribution System (DS Maximum): a sample from the
point in the distribution system that has the longest residence time, as designated by the water
system.
Distribution System Equivalent (DSE): a sample from the point in the distribution system
that has a well characterized detention time equivalent to a simulated distribution sample
(SDS).
Plant characteristics, including source water and disinfectant type, determined the specific sampling
location and frequency for certain parameters (see Exhibit 1.1).
Monitoring Frequency
General water quality parameters, DBF precursors, and disinfectant concentrations were monitored
monthly, while most DBFs were monitored quarterly. Targeted DBF monitoring for bromate and chlorite
was conducted monthly. Monthly samples were supposed to reflect typical operating conditions at the
plant and each set was required to be collected within a 72-hour period. A minimum of 14 days was
required between monthly sampling periods. A minimum of two months was required between quarterly
sampling periods (USEPA 1996b).
Minimum Reporting Levels
The method detection limit (MDL) is defined as the minimum concentration of a substance that can
be measured and reported with 99 percent confidence that the measured analyte concentration is greater
than zero. Usually, measurements below the MDL concentration are considered qualitative, not
quantitative, because they are not adequately precise to meet the needs of the data user(s). MDLs vary
from laboratory to laboratory based on the method used, equipment, etc.
Because MDLs vary from method to method and from laboratory to laboratory, EPA established
Minimum Reporting Levels (MRLs) for the ICR. MRLs were based on (1) a review of available
occurrence data to confirm that most are above the MRL; (2) whether the concentration at the MRL could
be measured taking use, burden, and status of analytical methods into consideration; and (3)
recommendations of an expert panel. Although EPA recognizes that some laboratories could provide
reliable data at concentrations below the MRL, a concentration measured below the MRL was not required
to be reported; instead, "below the MRL" was reported. Exhibit 1.1 presents the ICR MRLs for water
quality parameters, disinfectants, and DBFs. Section 1.4.8 explains how results that are "below the MRL"
were handled in the ICR data analysis.
Occurrence Assessment for the Final Stage 2 DBPR 1-13 December 2005
-------
Exhibit 1.1 ICR Plant Monitoring Requirements
Analyte
Plant Types
Required to
Monitor
Sampling Locations (# of
Sampling Locations)1
Water Quality Parameters
Total Organic Carbon
(TOC)
PH
Alkalinity2
Total Hardness 2
Turbidity 2
Temperature 2
Bromide
UV254 absorbance
All
All
All
Hypochlorite
All
All
All
All
All
All
All
All
Hypochlorite
All
All
Influent, Treatment, Finished
Influent, Treatment, Finished
Distribution System (4)
Disinfectant Stock Solution
Influent, Treatment, Finished
Distribution System (4)
Influent, Treatment, Finished
Distribution System (4)
Influent, Treatment, Finished
Distribution System (4)
Influent, Treatment, Finished
Distribution System (4)
Disinfectant Stock Solution
Influent, Treatment
Influent, Treatment, Finished
Disinfectants
Free Chlorine Residual 2
Total Chlorine Residual 2
Chlorine Dioxide (CIO2)2
Ozone 2
Free chlorine
as residual
disinfectant
Hypochlorite
All
Chlorine
Dioxide
Ozone
Treatment, Finished
Distribution System (4)
Disinfectant Stock Solution
Treatment, Finished
Distribution System (4)
Treatment, Finished, and
Distribution System (3)
Treatment
DBFs
Total Trihalomethanes
(TTHM)2
Chloroform
Bromodichloromethane
(BDCM)
Dibromochloromethane
(DBCM)
Bromoform
All
All
All
All
All
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Frequency
Monthly
Monthly
Quarterly
Quarterly
Monthly
Quarterly
Monthly
Quarterly
Monthly
Quarterly
Monthly
Quarterly
Quarterly
Monthly
Monthly
Monthly
Quarterly
Quarterly
Monthly
Quarterly
Monthly
Monthly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Minimum
Reporting
Level (MRL)
0.7 mg/L as
C
-
—
—
—
-
0.02 mg/L
0.009 crrr1
-
-
-
-
1 .0 ug/L
1 .0 ug/L
1 .0 ug/L
1 .0 ug/L
Occurrence Assessment for the Final Stage 2 DBPR
1-14
December 2005
-------
Analyte
Haloacetic Acid-Five
(HAA5)2
Haloacetic Acid-Six
(HAA6)2
Haloacetic Acid-Nine
(HAA9)2
Monochloroacetic Acid
(MCAA)
Dichloroacetic Acid
(DCAA)
Trichloroacetic Acid
(TCAA)
Monobromoacetic Acid
(MBAA)
Dibromoacetic Acid
(DBAA)
Bromochloracetic acid
(BCAA)
Bromodichloroacetic acid
(BDCAA)
Chlorodibromoacetic acid
(CDBAA)
Tribromoacetic acid
(TBAA)
Bromate (low-level)
(Method 300. 1)3
Bromate (Method 300.0)3
Chlorite (CIO2 )
Chlorate (CIO3 )
Haloacetonitriles-Four
(HAN4)2
Dichloroacetonitrile
(DCAN)
Trichloroacetonitrile
(TCAN)
Bromochloroacetonitrile
(BCAN)
Dibromoacetonitrile
(DBAN)
Plant Types
Required to
Monitor
All
All
All
Encouraged5
All
All
All
All
All
All
All
Encouraged5
All
Encouraged5
All
Encouraged5
Chlorine
Dioxide
Ozone
Ozone
Chlorine
Dioxide
Chlorine
Dioxide
Hypochlorite
All
All
All
All
All
Sampling Locations (# of
Sampling Locations)1
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished, and
Distribution System (3)
Treatment, Finished, and
Distribution System (3)
Influent, Disinfectant Stock
Solution, and Finished
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Frequency
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Monthly
Monthly
Monthly
Monthly
Monthly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Minimum
Reporting
Level (MRL)
-
-
-
2.0 ug/L
1.0 ug/L
1.0 ug/L
1.0 ug/L
1.0 ug/L
1.0 ug/L
1.0 ug/L
2.0 ug/L
4.0 ug/L
0.2 ug/L
5.0 ug/L
20 ug/L
20 ug/L
-
0.5 ug/L
0.5 ug/L
0.5 ug/L
0.5 ug/L
Occurrence Assessment for the Final Stage 2 DBPR
1-15
December 2005
-------
Analyte
Cyanogen chloride
(CNCI)
Chloral Hydrate (CH)
Chloropicrin (CP)
Trichloropropanone
(TCP)
Dichloropropanone (DCP)
Formaldehyde4
Acetaldehyde4
Butanal
Glyoxal
Methyl Glyoxal
Pentanal
Propanal
Total Organic Halides
(TOX)
Plant Types
Required to
Monitor
Chloramine
All
All
All
All
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
Ozone
Chlorine
Dioxide
All
Sampling Locations (# of
Sampling Locations)1
Finished and Distribution
System (Max)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished, and
Distribution System (4)
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Treatment, Finished
Influent, Treatment, Finished,
and Distribution System (4)
Frequency
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Minimum
Reporting
Level (MRL)
0.5 ug/L
0.5 ug/L
0.5 ug/L
0.5 ug/L
0.5 ug/L
5.0 ug/L
5.0 ug/L
5.0 ug/L
5.0 ug/L
5.0 ug/L
5.0 ug/L
5.0 ug/L
50 ug/L as
ci-
Notes:
1 "Influent" refers to the point where the water enters the plant. "Treatment" may include one or multiple sample
locations along the treatment process train, depending on the water quality parameter, DBP, and type of
plant/disinfectant used. "Finished" refers to the point of exit from the plant. "Distribution System (4)" refers to
the four points within the distribution system where samples were taken: the distribution system equivalent, two
points with average residence time, and one point with maximum residence time. "Distribution System (3)"
refers to three points in the distribution system where samples were taken: a location near the first customer,
one point of average residence, and one point with maximum residence time. Plants that purchase water were
also required to sample most DBFs at their plant influent. Plants that blend water sources within the treatment
plant also monitored water quality parameters prior to blending.
2 No MRLs were set for alkalinity, total hardness, turbidity, temperature, free chlorine residual, total chlorine
residual, chlorine dioxide, and ozone. Also, there are no MRLs for analyte summations (e.g., TTHM, HAAS)
because they are determined by adding or averaging several individual concentrations, or for pH, which simply
a direct measure.
Occurrence Assessment for the Final Stage 2 DBPR 1-16
December 2005
-------
3 Low levels of bromate may not be determined by EPA Method 300.0. Method 300.1 was written primarily to
identify specific parameters (column, eluent, and injection volume) which could be employed to allow the
quantitation of lower concentrations of bromate in drinking water. All 300.0 analytes are included in EPA
Method 300.1; however, comparable low level measurements of bromate are not possible using the Method
300.0 standard operating conditions.
4 Aldehyde samples were also analyzed for optional aldehydes (benzaldehyde, decanal, hexanal, heptanal,
nonanal, octanal).
5 Plants were encouraged, but not required, to monitor for these analytes.
Source: USEPA 1996c.
1.4.2 ICR Implementation Activities
EPA, along with other agencies, such as the American Water Works Association (AWWA) and the
Association of Metropolitan Water Agencies (AMWA), provided technical assistance before and during the
ICR. ICR reference manuals and videos were created to provide guidance on various technical aspects of
the rule, such as microbial sample collection and database use. See Chapter 1 of the Information
Collection Rule Data Analysis (McGuire et al. 2002) for a full listing of these materials. E-mail and
telephone support was provided for questions during the early stages of the ICR implementation.
During ICR implementation, laboratories analyzing ICR samples for DBFs, DBF surrogates, and
other water quality parameters were required to apply for ICR approval to ensure data quality. Over 400
commercial, utility, State, university, and Federal laboratories applied, of which 380 received approval.
Initial approval was based on criteria developed by EPA and a panel of experts and was given based on
method and analyte (see Chapter 1 of the Information Collection Rule Data Analysis [McGuire et al.
2002] for details on the approval criteria). To maintain approval status, laboratories had to successfully
conduct six quarterly ICR performance evaluations (PE) studies. On-site audits were performed as an
additional mechanism to maintain data quality.
1.4.3 ICR Sampling Plans
To ensure that ICR requirements were correctly applied, EPA required each system to submit an
Initial Sampling Plan (ISP) for approval. The ISPs included Initial Sampling Schematics (ISSs) which
were used to specify system-specific requirements. As reported in the Information Collection Rule Data
Analysis (McGuire et al. 2002), after an initial review by EPA, nearly 80 percent of the ISPs required
modifications such as correcting chemical additions or changing sampling locations. After a second
review, nearly 90 percent of the ISPs were approved.
1.4.4 Data Management Activities
The ICR data were reported and tracked through the ICR Data Management System. The ICR
Data Management System consists of three subsystems:
ICR Water Utility Database System, used by PWSs to report data
ICR Laboratory Quality Control (QC) Database System, used by independent laboratories to
report information on sample quality control
Occurrence Assessment for the Final Stage 2 DBPR 1-17 December 2005
-------
• ICR Federal Database System (ICR FED), used to upload and maintain data from systems and
laboratories in a central database
The data reported each month by ICR systems on diskettes included treatment unit process data for
each ICR plant, and collection information and analytical results for each ICR sample. Once data were
validated by EPA, they entered the ICR FED. The ICR FED is an Oracle™ database available to the
public.
1.4.5 Quality Assurance Activities
The ICR data were collected, analyzed, reported, and stored based on the Quality Assurance
Project Plan for the Implementation of the Information Collection Rule that was finalized in July 1997
(USEPA 1997c). This Plan included data objectives and measurement criteria, training requirements, and
instructions for records and documentation. It specified use of the ICR Sampling Manual (EPA 814-B-96-
001), the DBP/ICR Analytical Methods Manual (EPA 814-B-96-002), and the ICR Manual for Bench-
and Pilot-Scale Treatment Studies (EPA 814-B-96-003). The Plan also specified oversight activities
needed to ensure data quality and the data system to be used for collecting and reporting the measurement
data. The ICR DBP occurrence and water quality parameter data were critical to the decision-making
process. EPA provided oversight to ensure that procedures specified in the Plan were being followed:
On a monthly basis, EPA received data on diskettes from systems and labs. The QC
requirements established by the ICR were more extensive than those included in the laboratory
analytical methods. Laboratories and systems were required to report most of the QC data to
EPA along with the monitoring data.
Once the data were uploaded into the Water Utility Database, the data were processed using
validation algorithms. These algorithms tested whether the procedures were being followed by
verifying, for example, that the laboratory was approved to perform the analysis and the
sample was analyzed using an approved method. Information about the samples was used to
cross-check data submitted by laboratories and water systems. Examples of validation failures
include exceeding a sample holding time or failing the calibration standard. See Chapter 2 of
the Information Collection Rule Data Analysis (McGuire et al. 2002) for a list of QC data
used to validate ICR monitoring data. Laboratories and plants received reports containing
validation failures and monthly results generated by ICR FED. They were given an
opportunity to correct errors and resubmit data for the data validation process. Only data that
met the QC criteria were maintained in ICR FED.
• As noted above, laboratories had to successfully conduct six quarterly ICR performance
evaluations studies and on-site audits were also performed to maintain data quality.
Not only were the data collected with rigorous QA procedures in place, but the data and methods
were also subject to extensive technical review. This level of attention to assessing the technical quality of
the ICR data was used for several reasons. Most important was the recognition that these data were
critical to the development of a future rule, and that importance was underscored by the data collection
effort being specified in a separate federal rule with the force of law. It entailed substantial effort and
expense on the part of water systems and the government. The data collection period was 18 months.
Technical review occurred while data were being collected and enabled feedback, additional training, or
other corrections to be made if particular water systems or laboratories appeared to have consistent data
Occurrence Assessment for the Final Stage 2 DBPR 1-18 December 2005
-------
problems. This technical review, conducted in conjunction with data collection, was more extensive than
what would be possible during a peer review process after the monitoring was completed.
Some of the technical review steps undertaken were as follows:
EPA set high QC criteria and monitored the failure rate for each data element and the reasons
for failures so that EPA could adjust the validation process to accept some data that passed
lower QC criteria or, if appropriate, could work with utilities and laboratories to fix the
problems. (See Chapter 2 of the Information Collection Rule Data Analysis [McGuire et al.
2002] for a discussion of these and related data quality issues discussed below.)
As noted above, EPA received monthly updates of data from systems and laboratories. EPA
had automated some review of the technical quality or reasonableness of the data, such as
whether the analysis was sensitive enough to meet the reporting requirements. Further, the
process had built in the requirement for many fortified and duplicate samples so that precision
could be verified and those data elements that did not pass could be removed. Overall, 92
percent of the analytical data met the ICR QC requirements (McGuire et al. 2002).
• After review by EPA, the data that had passed the QC criteria were released for analysis. The
data were judged to be of sufficient quality based on the combination of using only reviewed
and approved laboratories to conduct the analyses, using specific analytical methods, requiring
all analyses to be performed within a specified amount of time, and continuously reviewing
data throughout the 18 months of data collection. McGuire et al. (2002) stated that
"[e] valuations of the QC data for DBFs and DBP surrogates indicate the national database...
contain high-quality data suitable to support regulation development."
• In addition to the internal EPA process of reviewing, accepting, and releasing ICR data, these
data were also carefully scrutinized upon release by EPA analysts, EPA contractors, and
members of the M-DBP FACA Technical Work Group (TWG). The data were released for
use in three 6-month blocks. Hundreds of statistical summaries and graphs were made
available to the TWG through a web site that was used extensively. These reviewers were
users of the information and many had extensive knowledge of the participating systems. The
essentially unlimited availability of the data to interested experts ensured additional technical
review of the data and ensured high data quality because any identified problems were
discussed and brought to EPA's attention for explanation or correction.
1.4.6 Development of Auxiliary Databases
The M-DBP FACA Technical Work Group determined that data in ICR FED needed to be
available in a more user-friendly format. To this end, EPA created seven auxiliary databases from ICR
FED. The Auxiliary Database 1 (AUX1) (USEPA 2000d) is the primary Microsoft Access™ database,
containing all system- and plant-level data extracted from ICR FED, such as sampling and treatment
operation data. The AUX1 database is structured "vertically," i.e., it is designed to facilitate analyses of a
single parameter across all plants or subsets of plants.
Occurrence Assessment for the Final Stage 2 DBPR 1-19 December 2005
-------
Six other auxiliary databases were derived from AUX1 that focused on "horizontal" data, i.e., data
that represented source water, plant, finished water, and in some cases, distribution data for an individual
plant for specific analytes:
• Auxiliary Database 2 (AUX2): Contact Time (CT) and Disinfection
• Auxiliary Database 3 (AUX3): Enhanced Coagulation
• Auxiliary Database 4 (AUX4): Sludge Production
Auxiliary Database 5 (AUX5): Washwater Return
Auxiliary Database 6 (AUX6): Disinfection Byproducts
• Auxiliary Database 8 (AUX8): Input and output data for modeling
For the data in AUX1 to be presented in an user-friendly format, fields were added as
necessary and some data manipulation occurred. Additional information on data transformation can be
obtained from the dictionary and documentation of AUX1 (USEPA 2000m). The same data manipulation
criteria were used when extracting all the auxiliary databases from AUX1. AUX7 was initially planned to
assess source water quality issues, but was never created due to overlap with the other auxiliary databases.
1.4.7 Representativeness of ICR Data
It's important to characterize the geographic distribution of the ICR data as well as the time period
over which the ICR data was collected. The geographic distribution is important to ensure the data set is
covering most, if not all, of the major watersheds and hydrology conditions in the United States. The
representativeness of the time period can be assessed by considering several factors including climate and
source water quality of the systems included in the ICR. See Chapter 3 of the Information Collection Rule
Data Analysis (McGuire et al. 2002) for further discussion of these topics. The national representativeness
of ICR data is discussed in more detail in both the ICR handbook (McGuire et al. 2002) and the Stage 2
DBPR EA (USEPA 2005a).
Geographic Distribution
The geographic distribution of the surface and ground water systems represented in the ICR data is
shown in Exhibits 1.2a and 1.2b. As mentioned previously, ICR monitoring was conducted by a census of
disinfecting systems serving more than 100,000 people. Therefore, geographic coverage is generally quite
broad though ICR systems were most concentrated in five States with large populations served by large
systems (CA, NY, TX, FL, and PA). Four States (VT, MT, ND, and WY) had neither surface nor ground
water ICR systems. Note that the majority of Florida's systems are served by ground water, though it
ranks second in the total number of ICR systems per State. About half of the U.S. population served by
CWSs is represented by ICR data (see Chapter 2, Exhibit 2.3 for estimate of population served by various
system sizes).
Occurrence Assessment for the Final Stage 2 DBPR 1-20 December 2005
-------
Exhibit 1.2a Percentage of Surface Water Systems (by State) Sampled for the ICR
Note: Maps are not drawn to scale.
Exhibit 1.2b Percentage of Ground Water Systems (by State) Sampled for the ICR
Note: Maps are not drawn to scale.
Occurrence Assessment for the Final Stage 2 DBPR 1-21
December 2005
-------
Climate and Source Water Quality
In Chapter 3 of the Information Collection Rule Data Analysis document (McGuire et al. 2002),
the authors evaluated ICR data representativeness by comparing rainfall data and other weather patterns
that occurred during the ICR monitoring period to historical weather patterns using data from the National
Weather Service. The calendar year 1998 was the warmest and fifth wettest year on record in the United
States since 1895. Rainfall amounts varied, however, across the country (approximately 22 percent of the
country was much wetter than normal; however about 2 percent was much drier than normal). Due to this
regional variations in temperature and rainfall, national trends and generalizations on the representativeness
of the ICR data are not possible (McGuire et al. 2002).
1.4.8 Methods and Assumptions for Analyzing ICR Results
Description of Data Set Evaluated
The majority of ICR data described in this chapter are derived from the ICR AUX1 Database, CD
version 5.0 (USEPA 2000d), representing monitoring results from 296 systems comprising 512 plants,
including 11 plants with blended source water. Some analyses in Chapter 2 are based on data from AUX2
(USEPA 2000i) (results are from AUX1 unless otherwise noted). Data analyses in Chapters 2, 3, 4, and
Appendix A of the document represent ICR results from the last 12 months, or last four quarters, of ICR
collection period (January through December 1998).
Plant Source Water Type
Plants reported the following source water types for each month of the ICR: "surface water;"
"ground water;" "mixed (or blended);" or "purchased water." For the purposes of this document, plant
source water type is based on data from the last 12 months of the ICR. Because there are plants that
reported more than one source water type during this period, designation of plant source water type was
done using a hierarchical approach. Specifically, the plant source water type was designated by the
following hierarchy: surface water, mixed, ground water, and purchased. In other words, if an ICR plant
treated surface water for any month, it was classified as a surface water plant. Refer to Appendix B for the
query language for source categorization (see the query "Plant Source Type, Last 12 Months").
ICR data analyses in the Information Collection Rule Data Analysis document (McGuire et al.
2002) and in other documents may have designated plant water source types differently. Because there are
a small number of plants that reported more than one source type, the differences in methodology should
not create a large discrepancy in results.
Plants With Multiple Quarterly Samples
Some of the plants in ICR data set reported multiple sampling results for a single quarter for a
given distribution system location. In these circumstances, the data were averaged for the given sampling
location and quarter. The averaged results were used in the plant screening analysis (see next section) and
in all calculations. A total of 15 plants have at least one quarter that was analyzed in this manner,
including all eleven blended plants. This does not apply to samples collected on a monthly basis.
Screening of Plants
For DBP analyses in Chapters 3 and 4 of this document, ICR plants are screened to include only
those with at least three of the last four quarters of the ICR having at least three of four distribution system
Occurrence Assessment for the Final Stage 2 DBPR 1-22 December 2005
-------
sampling locations with both TTHM and HAAS data. This screening was done to reduce the seasonal bias
that could occur if TTHM and HAAS data represented only one or two of the four quarters (e.g., if TTHM
data represent the summer only, yearly average results would most likely be skewed high), or only one or
two of the distribution system sampling locations (e.g., if data from the max location was missing, TTHM
average results for the distribution system could be skewed low).
For all other analytes (e.g., TOC, temperature, etc.), plants were screened to include only those
with at least nine of the last 12 months or three of the last four quarters of the ICR to reduce potential
seasonal biases.
See Appendix B, Section B.3 for the query language for plant screening.
Assumptions for Data Below the MRL
Any analytical results below the MRL (a non-detect) for a particular water quality parameter,
DBF, or disinfectant was assigned a value of zero. Because these levels below the MRL were assigned
values of zero, the means for each water quality parameter, DBF, and disinfectant are probably slightly
lower than they would be if the actual values were known and used in the calculations. In addition, median
concentrations that appear to be zero are not necessarily zero but are below the MRL. There is no MRL
for analyte summations (e.g., TTHM, HAAS) and DS Average values because they are determined by
adding or averaging several individual concentrations, rather than by measuring directly. Therefore, if each
THM, HAA, HAN4, or haloketone concentration is below its MRL, the resulting value for the
corresponding TTHM, HAAS, HAA6, HAA9, HAN4, and haloketones is zero.
1.4.9 Documentation of ICR Data Analyses
ICR data are available to the public through the EPA Web site. Because there are different
methods that could potentially be used to analyze ICR data, EPA has included several features in this
document to ensure transparency and reproducibility of all ICR-based results:
• Section 1.4.8 described the overall methods and assumptions used by EPA to evaluate the ICR
data.
All tables and graphs in Chapters 2 through 4 show the number of observations (or N-count)
used to generate results. The N-count will be the first data column in all tables and will be in
either the title, axis heading, or legend of each chart.
All Microsoft Access™ queries (in Structured Query Language [SQL] code) used to extract
ICR data from AUX1 are provided in Appendix B. Queries are organized alphabetically by
query name.
Query names, corresponding to queries in Appendix B, are included at the bottom of each
applicable table and chart in Chapters 2 through 4.
• An Excel reference file is also provided for those analyses that were conducted in Microsoft
Excel. All Excel files are included in the docket.
Occurrence Assessment for the Final Stage 2 DBPR 1-23 December 2005
-------
1.5 Other Data Sources
Occurrence data for medium and small water treatment plants were not included in the ICR data
collection. Data were obtained from the ICRSS, NRWA Survey, the Water Utility Database
(WATER:\STATS) (AWWA 2000), the Ground Water Supply Survey (GWSS) (USEPA 1983), and
several States in order to examine the occurrence patterns of medium and small water treatment plants.
Exhibit 1.3 briefly outlines these data sources, while the subsections that follow describe the sources in
greater detail, including the level of quality assurance. Appendix C summarizes how each data source is
used in this document and shows how the data quality objectives are met.
Occurrence Assessment for the Final Stage 2 DBPR 1-24 December 2005
-------
Exhibit 1.3 Summary of Non-ICR Occurrence Survey Data
Data Source
ICR Supplemental
Survey (ICRSS)
(Surface Water
Only)
NRWA Survey
(Surface Water
Only)
WATER:\STATS
(Surface and
Ground Water)
Ground Water
Supply Survey
State Data -
Ground Water
State Data -
Surface Water
Number of
Systems (or
Plants) by Size
(Population
Served)
• 47 plants serving
100,000 or more1
• 40 plants serving
10,000-99,999
• 40 plants serving
fewer than
10,000
117 systems serving
fewer than 10,000
• 21 9 systems
serving 100,000
or more
• 623 systems
serving 10,000-
99,999
30 systems
serving fewer
than 10,000
945 systems total
(466 random, 479
nonrandom)
562 systems serving
fewer than 10,000
2,336 systems
serving fewer than
10,000
Data Collected
Raw source water - (Large
Systems) TOC
Raw source water - (Small &
Medium Systems) TOC,
UV254, bromide, turbidity,
pH, & temperature
Population served and flows
Raw source water -
temperatures, turbidity,
pH,and source water type,
bromide, TOC, UV254,
alkalinity, calcium and total
hardness
Finished water-residence
time estimate, total and
individual THMs, individual
HAAs and HAA5, HAA6,
HAA9, TOC, UV254, bromide,
temperature, pH, free and
total chlorine residual levels
Treatment-unit processes,
disinfectant used
Population served and flows
Raw source water - Water
Quality Parameters (WQPs),
Source water type
Finished water -
WQPs,TTHM,HAAs
Treatment-unit processes,
disinfectant used
TOC and TTHM (one sample for
each parameter at the entry
point to distribution system)
Distribution system TTHM
occurrence data
Distribution system TTHM
occurrence data
Time
Frame
March
1999-
February
2000
November
1999-
March
2000
1996
December
1980-
December
1981
Varies
Varies
Geographic
Representation
Random national
distribution by
system size and
surface water
source type
Random national
distribution
Random national
distribution
Combination of
random national
sample and
nonrandom sample
AK, CA, FL, IL,
NC, TX, WA2
AK, CA, IL, MN,
MS, NC, TX, WA2
1 Source type designations include flowing stream and lake/reservoir (except for seven large plants pre-selected).
2 Over 50 percent of each State's systems are represented. In total there are approximately 20 percent of the
nation's small systems included in these data. EPA believes that the data reasonably represent a full range of
source water quality in small systems at the national level.
Occurrence Assessment for the Final Stage 2 DBPR 1-25
December 2005
-------
1.5.1 ICR Supplemental Survey
EPA conducted the ICRSS to supplement ICR information on microbial and byproduct occurrence.
The ICRSS was conducted at 120 randomly selected plants, 40 of which were classified as small surface
water plants serving fewer than 10,000 people, 40 as medium surface water plants serving 10,001 to
100,000 people, and 40 as large surface water plants serving more than 100,000 people. Seven very large
systems (> 1 million people served) were also included in the survey effort. Monitoring was conducted for
12 consecutive months beginning in March 1999. Large systems collected protozoa and limited precursor
data (i.e., TOC), while medium and small systems monitored water quality parameters (i.e., temperature,
pH and alkalinity) and DBF precursors (i.e., TOC, UV254 and bromide). EPA used these data to compare
relative treatability among different system size categories for achieving compliance with the Stage 2
DBPR regulatory alternatives. A discussion of the protozoa data is included in the Draft Occurrence and
Exposure Assessment for the LT2ESWTR (USEPA 2005b).
These measurement data were generated based on the Quality Assurance Project Plan for the
Implementation of the Information Collection Rule Supplemental Surveys, finalized in March 1999
(USEPA 1999c). The Plan employed a QA process similar to that used for ICR data, and covered
measurement and data acquisition, assessment and oversight, and data validation and usability. Also
similar to the review of ICR data, a technical review more extensive and rigorous than possible in a typical
peer review process was implemented.
1.5.2 National Rural Water Association Survey
The National Rural Water Association (NRWA) Survey was conducted to provide information on
DBFs and their precursors in small surface water systems. Results have been published in the document,
Summary Report: NRWA Small System Study ofD/DBP (Trax and Kramer, 2003).
The NRWA, in conjunction with EPA and NRWA State chapters, conducted a survey of 117
randomly selected small PWSs. A minimum number of 112 systems were targeted to ensure that the
results from the survey (in particular, the 90 percent confidence intervals) would be statistically
representative of the universe of small water systems. Also, because water temperature and other factors
can affect DBP formation, the survey collected detailed treatment process information, source water quality
data, and DBP samples for both a cold-weather period in 1999-2000 and a warm-weather period in 2000.
NRWA data are presented in Chapter 3 and support the analyses of small systems.
The NRWA conducted this survey with the assistance of EPA (EPA did not direct this effort).
EPA helped train those collecting the samples and provided QA review of the data compiled and presented
by the NRWA. EPA-approved laboratories were used for the analysis of samples. The analytical data
were generated based on the procedures used for the ICR. An extensive quality assurance protocol was
followed to ensure high quality monitoring, management, and documentation of the analytical data. These
included:
• Samples from ten percent of the 112 sites were replicated.
• Fifteen percent of the samples for each analyte were randomly replicated.
• The laboratory was provided with a THM "blank" sample. It was analyzed when results
appeared erroneous or deviated greatly from expected values.
Occurrence Assessment for the Final Stage 2 DBPR 1-26 December 2005
-------
All analytical methods and QA/QC procedures used in the survey were consistent with the
requirements listed in the DBP/ICR Analytical methods Manual (EPA 814-B-96-002).
For HAA9, the contractor had the flexibility to use Standard Method 625 IB.
The selection of the laboratory was based on the demonstration of historical ability to meet
or exceed the QA/QC requirements.
1.5.3 The Water Industry Database (WATERASTATS)
Published by the American Water Works Association (AWWA), WATER:\STATS is derived
from the AWWA Water Industry Database based on information from the 1996 survey of approximately
900 water utilities. Most of these utilities are large water systems serving at least 10,000 people. The
1996 survey collected a range of financial and operational information on these utilities, including data on
the occurrence of DBFs in finished water (however, many systems did not respond to all survey questions).
WATERASTATS does not contain individual sample results; rather, it contains summary statistics such as
minimum, maximum, and average values that were reported by each system. The WATERASTATS data
used here are those that characterize relevant treatment and byproduct information for medium surface
water plants (those serving between 10,000 and 100,000 people).
The survey was progressively improved upon since 1989, when WATERASTATS was first
developed by AWWA. Over the years, it has been technically reviewed by the AWWA and AWWARF
Advisory Committees and by the Technical and Education Committees of the AWWA. These reviews have
served to modify and improve the survey over time. Prior to sending the 1996 survey questionnaire out to
all participants, it was field tested with 25 utilities and adjusted accordingly.
Responses to the survey questions were screened by a team of experts from AWWA to ensure they
were applicable and pertinent. The AWWA staff also reviewed data for magnitude and units related issues.
Standard procedures were adopted to identify apparent "outlier" data, with involved utilities contacted to
determine if the outlier was legitimate or not (and was included or excluded accordingly).
1.5.4 Ground Water Supply Survey
There are few national studies of the occurrence of contaminants in ground water. Although
two decades old, the GWSS, conducted by EPA from December 1980 through December 1981, remains
one of the most extensive and useful studies of ground water. The GWSS sampled and analyzed levels of
volatile organic compounds (VOCs) in ground water. Results are presented in the report, The Ground
Water Supply Survey: Summary of Volatile Organic Contaminant Occurrence Data (USEPA 1983). The
data from the GWSS are presented in Chapter 3 and support the analyses of medium and small systems.
The data were collected from 945 systems, approximately half selected randomly and half selected
nonrandomly. Random selection was intended to provide a broad national perspective on the incidence of
VOC contamination; the nonrandom selection allowed States to identify sites that were presumed to have
high levels of VOCs for further investigation. Included in the sampling parameters were levels of finished
water TOC and TTHM.
The random sample included 186 systems from a random list of systems serving a population of
greater than 10,000, and 280 systems from a random list of systems serving a population of less than or
equal to 10,000. The nonrandom sample consisted of 479 systems that were selected by State agencies.
States were encouraged to choose systems for which no prior VOC data were available and believed to
Occurrence Assessment for the Final Stage 2 DBPR 1-27 December 2005
-------
have a high probability of contamination by VOCs, based on their knowledge of local conditions (e.g.,
proximity to landfills, industrial activity, etc.).
An extensive quality assurance protocol was followed to ensure high quality monitoring,
management, and documentation of the analytical data. These protocols included:
Laboratory analysis of EPA reference samples - The reference samples contained known
concentrations of compounds including four common trihalomethanes (THMs) and nine
frequently detected VOCs. They were analyzed by the laboratory once a week for each
instrument to determine whether the precision and accuracy of the instruments were within
acceptable limits per the quality protocol.
Analysis of duplicate samples by the laboratory - Duplicate analyses were performed on at
least 10 percent of the samples. The duplicate analyses were to agree within 40 percent
for compounds present below 5 (ig/L, and within 20 percent for compounds present above
5 (ig/L, in order to comply with the quality protocol.
• Confirmatory analysis - All samples found or suspected to contain purgeable aromatic and
halocarbon compounds other than THMs were re-analyzed using different
chromatographic columns that elute compounds in different orders. Samples containing
chloroform at concentrations greater than 40 (ig/L were re-analyzed using the confirmatory
column since chloroform concentrations at this level (i.e., equal to or greater than 40 (ig/L)
could potentially mask the presence of 1,2-dichloroethane. Additionally, 5 percent of all
samples were re-analyzed by gas chromatography/mass spectrometry (GC/MS) to identify
or confirm unknown/tentatively known compounds.
Blind samples - EPA used five blind samples during the initial phase of the survey to
evaluate the laboratory's ability to identify and measure specific compounds. The samples
consisted of five different mixtures of compounds, spiked into organic-free distilled water.
The contractor correctly identified the spiked compounds in every case.
• Analysis of duplicate samples by EPA - Duplicate samples were collected in separate
bottles and stored at EPA's laboratories. They were analyzed as an additional check on
the contractor laboratory's results.
1.5.5 State Data
A number of State agencies have collected data on influent water quality and DBP occurrence for
small surface water plants. As part of the data synthesis effort for small and medium systems, some of
these States provided data sets to EPA. The Agency reviewed them for applicability to EPA's national
analysis of DBP occurrence. The following States collected sufficient DBP occurrence data to include in
further surface water analyses: Alaska, California, Illinois, Minnesota, Missouri, North Carolina, Texas,
and Washington. In addition, seven States' data sets were used to analyze small ground water systems
(Alaska, California, Florida, Illinois, North Carolina, Texas, and Washington). The data for both surface
and ground water systems met several initial quality criteria:
• For each State's data set, the small surface water systems sampled by the State were
representative of at least 50 percent of the total number of small surface water systems in the
State.
Occurrence Assessment for the Final Stage 2 DBPR 1-28 December 2005
-------
TTHM data were collected and reported in a manner that approximated a typical monitoring
approach, and in some cases included individual species of THMs.
The data available from each State are not exactly comparable; some States reported individual sample
data, while others reported only plant averages. Some of the data appear to be from distribution system
locations, while other samples are from locations in the plant or from raw water. Samples in some States
were collected quarterly, while in others the sample frequency ranged from one every two months to less
than one per year.
These State occurrence findings are from existing sources and are typically summaries of data and
summary statistics rather than the raw analytical data. Although a QA plan did not exist during data
collection and data were not peer reviewed, it is assumed that the data were reviewed by States before
submission to EPA. The usage of States data for characterizing national DBF occurrence for small surface
water systems is discussed in the Stage 2 DBPR EA (USEPA 2005a). The States data are presented in
Chapter 3 and support the analyses of medium and small systems.
1.6 Document Organization
The remainder of this document is organized into the following four chapters (with Chapter 5, the
reference section) and three appendices.
Chapter 2 - Use of Disinfectants in the United States. The universe of systems using
disinfectants and their population-served are presented by system size and source water type
category. This chapter also presents information on disinfection use. An overview of
disinfection processes is provided, followed by information on the four most commonly used
disinfectants: free chlorine, chloramine (combined chlorine), chlorine dioxide, and ozone. Each
disinfectant is briefly described, including its method of application, use and distribution,
advantages and disadvantages, dosage requirements, and potential byproducts.
Chapter 3 - National Occurrence Data: Pre-Stage 1 Baselines. This chapter presents data
related to DBF occurrence in public drinking water supplies. Graphical presentations of
source water quality parameters, disinfectant residuals, and DBFs are included. Data are from
the ICR data set and other sources.
• Chapter 4 - National DBF Occurrence: Predicted Post-Stage 1 Baselines. This chapter
describes predicted post-Stage 1 occurrence for TTHM and HAA based on ICR data.
Chapter 5 - References.
Appendix A - TTHM and HAAS Speciation Occurrence Data. This appendix supplements
ICR analyses in Chapters 2 and 3 by showing results for TTHM and HAAS species over the
last 12 months of the ICR collection period.
Appendix B - ICR Data Queries. This appendix provides the Access™ Queries (in SQL
code) used in the data presentations in Chapters 2, 3, 4, and Appendix A. Queries are
organized alphabetically by query name and include a one to two sentence description of their
function.
• Appendix C - Assessment of Data Quality Objectives. This appendix summarizes how each
data source is used in this document and shows how the data quality objectives are met.
Occurrence Assessment for the Final Stage 2 DBPR 1-29 December 2005
-------
2. Use of Disinfectants in the United States
Many water treatment processes can remove or inactivate microorganisms, including pathogens
that can cause waterborne diseases. Treatment is especially important for systems that use surface water
sources or ground water under the direct influence of surface water (GWUDI). In many cases, surface
water supplies receive discharges from upstream wastewater treatment plants, industrial facilities,
stormwater runoff, or animal feed lots. In the treatment plant, certain treatment processes, such as
sedimentation and filtration, remove most of the microorganisms that cause waterborne diseases.
However, there is a need to inactivate the pathogens that pass through the filters, grow in the distribution
system (e.g., biofilm growth), or breach the distribution system (e.g., entering through cross-connections
or negative pressure). Surface water systems have relied primarily upon filtration supplemented by
disinfection to control pathogens. Disinfection has been associated with major improvements in public
health since it was widely adopted in the early 1900's.
This chapter describes the disinfection processes used in the treatment of drinking water and their
effects on finished water quality. Sections 2.1 and 2.2 provide background information on the
disinfection process and the resulting formation of DBFs, respectively. Section 2.3 provides the
inventory of disinfecting community water systems (CWSs) with respect to source water type and
population served. Section 2.4 shows data on the proportion of plants using various types of disinfection.
Sections 2.5 through 2.8 describe the four main disinfectants used by drinking water systems: chlorine,
chloramines, chlorine dioxide, and ozone. Included for each disinfectant is a description of chemistry and
method of application, use and distribution, typical dosages, and potential byproducts. Chapter 3, Section
3.1.2, builds on this chapter by examining disinfectant residual concentrations in treated water. It is
should be noted that UV disinfection is not included in this document because it is still a relatively new
disinfection technique and the amount of available field data is limited. EPA specifically publishes an UV
Disinfection Guidance Manual supporting the Long Term 2 Enhanced Surface Water Treatment Rule
(USEPA 2005c).
2.1 Overview of Disinfection Processes
The 1995 Community Water Systems Survey (CWSS) (USEPA 1997b) reports that 99 percent of
surface water systems in the United States provide some level of water treatment before distribution to
customers, and 99 percent of these treatment systems use disinfection in the treatment process
(disinfection is required for all surface water systems). Disinfection can be accomplished in several ways.
The most common method used to achieve disinfection is to add a chemical disinfectant to the water.
Disinfectants can be applied in the plant (this is referred to as primary disinfection) and/or after treatment
(secondary disinfection), like filtration or sedimentation. Secondary disinfection ensures the presence of
a disinfectant residual after treated water leaves the plant and enters the distribution system. Some
systems use booster chlorination, the adding of chlorine or chloramines at a point within the distribution
system, to raise the disinfectant residuals to required levels. The most commonly used disinfectants, in
both plants and distribution systems, are chlorine and chloramines. Chlorine dioxide, ozone, and
ultraviolet light (UV) (a non-chemical disinfection process) are also used on a limited basis to meet
disinfection goals. Chlorine or chloramines are the most common disinfectants used to achieve secondary
disinfection.
Occurrence Assessment for the Final Stage 2 DBPR 2-1 December 2005
-------
Chlorine, chloramines, ozone, and chlorine dioxide are oxidants, and, in addition to inactivating
pathogens, are used to treat drinking water for the following purposes:
• Controlling Asiatic clams and zebra mussels (except for chloramines)
• Oxidizing inorganic material such as iron, manganese, and sulfides (except for chloramines)
• Preventing microbial regrowth in the distribution system and maintaining biological stability
(except for ozone)
Removing undesirable tastes and odors through chemical oxidation (except for chloramines)
Improving coagulation and filtration efficiency
Preventing algal growth in sedimentation basins and filters
Oxidizing organic micropollutants such as pesticides and volatile organic compounds (except
for chloramines)
The effectiveness of disinfection depends on the contact time (the amount of time a disinfectant is
in contact with the water) and the residual disinfectant concentration. The efficacy of disinfection also
depends on other factors, including pH, temperature, and the type of disinfectant used.
2.2 Disinfection Byproducts
Disinfectants react with naturally occurring organic matter (NOM) and inorganic matter to form
DBFs. Three main types of DBFs are discussed in this document.
Halogenated organic byproducts
Organic oxidation byproducts
• Inorganic DBFs
Halogenated organic byproducts form during reactions with free chlorine or free bromine. Although
bromine is not used as a disinfectant, bromide ions can be naturally present in water and, when oxidized,
form free bromine. Organic oxidation byproducts, such as acetaldehyde, form during oxidation reactions
with the disinfectants. Inorganic DBFs are usually formed during reactions with chlorine dioxide and
ozone.
Temperature, pH, alkalinity, total hardness, turbidity, disinfectant type, bromide concentration,
and the amount and composition of NOM (usually measured as total organic carbon, or TOC) affect the
types and rates of DBF formation. It should be noted that many DBFs have been identified, but only a
select subset (shown in Exhibit 2.1) were monitored during the ICR and are the focus of this analysis.
Occurrence Assessment for the Final Stage 2 DBPR 2-2 December 2005
-------
Exhibit 2.1 List of Disinfection Byproducts Measured During the ICR
Halogenated Organic Byproducts1
Trihalomethanes
Chloroform (CHCI3) Dibromochloromethane (DBCM)
Bromodichloromethane (BDCM) Bromoform (CHBrS)
Haloacetic Acids-Five (HAA5) Five
Monochloroacetic acid (MCAA) Dichloroacetic acid (DCAA)
Trichloroacetic acid (TCAA) Monobromoacetic acid (MBAA)
Dibromoacetic acid (DBAA)
HAA6
HAAS Bromochloroacetic acid (BCAA)
HAA9
HAA6 Bromodichloroacetic acid (BDCAA)
Chlorodibromoacetic acid (CDBAA) Tribromoacetic acid (TBAA)
Haloacetonitriles (HAN4)
Dichloroacetonitrile (DCAN) Bromochloroacetonitrile (BCAN)
Dibromoacetonitrile (DBAN) Trichloroacetonitrile (TCAN)
Haloketones
1,1 -Dichloropropanone 1,1,1 -Trichloropropanone
Others
Chloropicrin (CP) Chloral Hydrate (CH)
Cyanogen Chloride (CNCI) Total Organic Halides (TOX)
Organic Oxidation Byproducts
Aldehydes
Formaldehyde Acetaldehyde
Propanal Butanal
Pentanal Glyoxal
Methyl Glyoxal
Inorganic Byproducts
Chlorate Ion Chlorite Ion
Bromate Ion
1 Not all individual organic halides could be measured during the ICR. TOX is used to estimate the total quantity of
dissolved halogenated organic material in water.
Source: USEPA 1996c.
2.3 Inventory of Disinfecting Water Systems and Population Served
Both CWSs and nontransient noncommunity water systems (NTNCWSs) that disinfect their
water supplies, as well as transient noncommunity water systems that use chlorine dioxide, will be
regulated under the Stage 2 DBPR. All surface water systems are required to disinfect, but only an
estimated 68 percent of ground water CWSs and 37 percent of ground water NTNCWSs disinfect
(USEPA 2005a). Exhibit 2.2 shows the combined CWS and NTNCWS estimated system size distribution
of disinfecting systems, classified by source water type (systems using GWUDI are included in the
surface water category) and by population served. Exhibit 2.2 shows that approximately 76 percent of
disinfecting CWSs and NTNCWSs are ground water systems.
Occurrence Assessment for the Final Stage 2 DBPR 2-3 December 2005
-------
Exhibit 2.2 Number (and Percent) of Disinfecting CWSs and NTNCWSs
System Size
(Population Served)
Small (< 10,001)
Medium (10,001 - 100,000)
Large (>1 00,000)
Total
Ground Water
(Percent of Total)
37,980 (73.6%)
1,365 (2.6%)
63 (0.1%)
39,408 (76.3%)
Surface Water
(Percent of Total)
9,921 (19.2%)
2,013 (3.9%)
290 (0.6%)
12,224 (23.7%)
Total
(Percent of Total)
47,901 (92.8%)
3,378 (6.5%)
353 (0.7%)
51,632 (100.0%)
Detail may not add due to independent rounding.
Notes: The "surface water" designation includes GWUDI systems.
Percent disinfecting based on 1995 CWSS, as summarized in the Drinking Water Baseline Handbook, and
adjusted for potential impacts of the Ground Water Rule. See Chapter 3 of the Stage 2 DBPR Economic
Analysis (USEPA 2005a) for further details.
Source: Derived from Chapter 3, Exhibit 3.2 of the Stage 2 DBPR Economic Analysis (USEPA 2005a).
Exhibit 2.3 shows other size distribution findings. While the number of systems are primarily
small ground water systems, the majority of people are served by large surface water systems: 37 percent
of the population is served by ground water systems, and 63 percent of the population is served by surface
water. This is because most large systems serving more than 100,000 people are surface water systems.
Exhibit 2.3 Population Total (and Percent) Served by Disinfecting CWSs
System Size
(Population Served)
Small (< 10,001)
Medium (10,001 - 100,000)
Large (>1 00,000)
Total
Ground Water
(Percent of Total)
29,413,975 (11.6%)
37,986,723 (14.9%)
26,392,250 (10.4%)
93,792,948 (36.9%)
Surface Water
(Percent of Total)
8,197,640 (3.2%)
38,616,140 (15.2%)
113,871,860 (44.7%)
160,685,640 (63.1%)
Total
(Percent of Total)
37,611,615 (14.8%)
76,602,863 (30.1%)
140,264,110 (55.1%)
254,478,588 (100.0%)
Detail may not add due to independent rounding.
Notes: The "surface water" designation includes GWUDI systems.
Percent disinfecting based on 1995 CWSS, as summarized in the Drinking Water Baseline Handbook, and
adjusted for potential impacts of the Ground Water Rule. See Chapter 3 of the Stage 2 DBPR Economic
Analysis (USEPA 2005a) for further details. NTNCWSs are typically schools, restaurants, etc., and their
population is already counted in the CWS population, thus only population served by CWSs is shown in this
exhibit.
Source: Derived from Chapter 3, Exhibit 3.3 of the Stage 2 DBPR Economic Analysis (USEPA 2005a).
Exhibits 2.2 and 2.3 also show the distribution of systems and population between different size
categories. Approximately 93 percent of all disinfecting CWSs and NTNCWSs are small systems serving
less than 10,000 people, while fewer than one percent are large systems serving more than 100,000
people. Although there are many more small systems than large in the United States, an estimated 140
million people (55 percent) are served by large disinfecting systems, 77 million (30 percent) by medium
disinfecting systems, and only 38 million (15 percent) by small disinfecting systems.
Occurrence Assessment for the Final Stage 2 DBPR
2-4
December 2005
-------
2.4 Disinfectant Types
This section presents data on chemical disinfection practices among large surface and ground
water plants as derived from the Information Collection Rule Auxiliary Database 1 (ICR AUX1) (USEPA
2000d). It also presents disinfectant dose data for chlorine, ozone, and chlorine dioxide from the
Information Collection Rule Auxiliary Database 2 (ICR AUX2) (USEPA 2000i). See Chapter 1, section
1.4.6 for a description of these databases.
In the ICR databases, disinfectant types associated with a treatment plant are classified based on
disinfectant usage within the plant and distribution system. At the plant level, five disinfectant types are
defined in the ICR AUX1 and AUX2 databases as:
• CL2—Free chlorine (C12) when only C12 is used as a disinfectant.
• CLM—Chloramine (CLM) when C12 and ammonia (NH3) are added simultaneously into a
unit process in the plant where no earlier point of chlorination exists.
CL2_CLM—C12 followed by CLM when NH3 is added after free chlorine has previously
been applied in one or more preceding unit processes.
• CLX—Chlorine dioxide (C1O2) if C1O2 is used anywhere in the plant.
O3—Ozone (O3) if O3 is used anywhere in the plant.
At the distribution system level, two disinfectant types are defined (i.e., CL2 and CLM) according to the
disinfectant type applied at the last disinfectant application point before the entry point to the distribution
system.
In order to characterize disinfection practices of surface water plants, the ICR AUX2 database
was used to derive information for plants that reported both plant- and distribution system-level
disinfectant use for the last 12 months of the ICR collection period. Because some plants switched
disinfectants during the 12-month period, the analysis was done for each plant-month individually.
Exhibit 2.4 shows the findings of the analysis for surface water plants for each combination of plant and
distribution system disinfectant for 3,927 plant-months. In Exhibit 2.4, the letters before the V represent
the primary disinfectant, while the letters after the '/' represent the secondary disinfectant. For example,
CL2/CLM would mean a plant that uses chlorine for primary disinfection and chloramine for secondary
disinfection. Chapter 15, Table 15.1, in the Information Collection Rule Data Analysis document
(McGuire et al. 2002) provides additional information on the type of disinfectants used for different
treatment plant types (e.g., conventional softening, unfiltered).
For ground water plants, the ICR AUX2 database was also used to characterize disinfection
practices. Information was extracted for ground water plants that reported both plant- and distribution
system-level disinfectant use for the last 12 months of the ICR collection period. Exhibit 2.5 shows the
results of the ground water plant analysis for chlorine only, chloramines only, and each combination of
plant and distribution system disinfectant for 647 plant-months. The "CL2 only" category includes plants
that reported chlorine use for both the plant and distribution system category, and includes plants that
reported chlorine use only in the distribution system.
Occurrence Assessment for the Final Stage 2 DBPR 2-5 December 2005
-------
Exhibit 2.4 Percentage of Surface Water Plants Applying Specific Disinfectant
Types for Combined Plant/Distribution System
60% -r
53.7%
50%
-------
Exhibit 2.5 Percentage of Ground Water Plants Applying Specific Disinfectant
Types for Individual and Combined Plant/Distribution System
yuu/o -
S b(J/0 "
z_
w
.c
o
_re
"o
1
u
0_
n% -
84.9%
13.4%
0.0% °-8% 0.0% 0.0% 0.0% °-8%
CL2 Only CLM Only CL2_CLM/CL2 CL2/CLM CLX/CL2 CLX/CLM
Plant/Distribution System Disinfectant Combination
O3/CL2
O3/CLM
Source: ICR AUX1 Database (USEPA 2000d).
Query: Screened Plant Disinfectant Type. See Appendix B for details.
Among medium surface water plants, chlorine is the most common disinfectant (see Exhibit 3.34,
which compares disinfectant use in medium and large plants). The National Rural Water Association
(NRWA) Survey indicates that almost all small surface water systems use free chlorine (see Exhibit 3.43).
2.5 Chlorine
Chlorine is the most commonly used disinfectant in public water systems in the United States.
Through filtration and chlorination, waterborne diseases, including typhoid and cholera, have been
virtually eliminated in this country. For example, in only four years (between 1911 and 1915), the
number of typhoid cases in Niagara Falls, New York dropped from 185 deaths for every 100,000 people
to nearly zero following the introduction of filtration and chlorination (White 1986).
Occurrence Assessment for the Final Stage 2 DBPR
2-7
December 2005
-------
Disinfection with chlorine is simple, economical, efficient, measurable, and practical. Several
forms of chlorine are available for use as a disinfectant.
Chlorine gas (C12)
Sodium hypochlorite (liquid) (NaOCl)
Calcium hypochlorite (tablet, granular, or powdered) (Ca(OCl)2)
2.5.1 Description of Chemistry
The chemistry is similar for all forms of chlorine; they all react with water to form disinfecting
agents. Chlorine hydrolyzes in water to form hypochlorous acid (HOC1). HOC1 is a weak acid and
ionizes to yield hypochlorite ion, or OC1". Free residual chlorine is the sum of HOC1 and OC1"
concentrations; the relative quantity of each depends on pH. Both hypochlorous acid and hypochlorite
inactivate or kill pathogens, but hypochlorous acid is more effective.
Upon addition to water, free chlorine chemically reacts with constituents in the water by various
mechanisms. Chlorine oxidizes soluble iron, manganese, and sulfides typically found in drinking water
sources. Once oxidized, the resulting products precipitate and are primarily removed by clarification and
filtration processes. When chlorine reacts with natural organic matter in the water, it reacts with electron-
rich sites to form halogenated organic byproducts (e.g., trihalomethanes and chlorophenols), some of
which have been shown to be possible human carcinogens (Weisel et al. 1999). Chlorine also oxidizes
organic matter to form compounds that do not contain a halogen, such as aldehydes, carboxylic acids,
ketones, and alcohols (Richardson 1998). The occurrence of halogenated byproducts has been studied the
most because halogenated DBFs are easily detected.
The three forms of chlorine that are typically used at water treatment plants (chlorine gas, sodium
hypochorite, and calcium hypochlorite) are described below.
Chlorine gas is often referred to as elemental chlorine. Chlorine is produced, collected, purified,
compressed, cooled, packaged, and shipped as a liquefied gas under pressure. Systems then inject
chlorine gas into the water stream, where hydrolysis and ionization (as described above) produce
the disinfecting agents.
Sodium hypochlorite is produced by reacting chlorine with sodium hydroxide. Sodium
hypochlorite solutions are also referred to as liquid bleach or Javelle water. Generally,
commercial or industrial grade solutions have hypochlorite strengths of 10 to 16 percent. Low
concentrations (i.e., 5.25 percent or less) are sold as common household bleach. The stability of a
sodium hypochlorite solution depends on the hypochlorite concentration, storage temperature,
time in storage, impurities, pH, and exposure to light. Decomposition of hypochlorite solution
over time can affect the feed rate and dosage, as well as produce undesirable byproducts such as
chlorite or chlorate ions (Gordon et al. 1995). Because of these storage problems, many systems
are investigating onsite generation of sodium hypochlorite in lieu of purchasing hypochlorite
stock supplies from a manufacturer or vendor.
Calcium hypochlorite is a crystal and can be produced by combining equivalent amounts of
sodium hypochlorite and calcium chloride (known as the Perchloron process). A slurry of lime
and caustic soda is chlorinated and cooled so that crystals are formed. These crystals are
centrifuged, then added to a chlorinated lime slurry; when warmed, the calcium hypochlorite
precipitates. Generally, the final product contains up to 70 percent available chlorine and less
than 3 percent lime (White 1992). Storage of calcium hypochlorite is a safety consideration. It
Occurrence Assessment for the Final Stage 2 DBPR 2-8 December 2005
-------
should never be stored where it is subject to heat, allowed to contact any easily oxidized organic
material, or become wet (moisture can trigger a reaction, creating heat and possibly fire).
Based on analysis of the ICR AUX2 data, gaseous chlorine is by far the most common form of
chlorine used for water system disinfection (chlorine was used in its gaseous form 91 percent of the time).
See the Information Collection Rule Data Analysis document (McGuire et al. 2002) for more details.
However, small systems are more likely to use sodium hypochlorite.
2.5.2 Use and Distribution
Chlorine gas feeders used to treat drinking water can be either direct feed or solution feed. Direct
gas feeders deliver chlorine gas under pressure directly to the point of application. Because direct feeders
are less safe than solution feed chlorinators, solution feed is typically used. In a solution feeder, chlorine
gas is metered under vacuum conditions and mixed with water in an injector to produce a chlorine
solution, which is injected at the appropriate application point(s). With this type of system, the flow of
chlorine gas automatically shuts off if there is a loss of vacuum, stoppage of the solution discharge line,
or loss of operating solution water pressure. This safety mechanism is important because chlorine gas
released into the atmosphere can cause acute health problems or even death if inhaled.
Sodium hypochlorite is normally fed directly with a motor-driven positive displacement-type
chemical metering pump(s) to the appropriate application point(s). Although unusual, feeding sodium
hypochlorite using a hydraulic injector or simple gravity flow is possible.
When calcium hypochlorite is used as a treatment process for continuous disinfection, it is often
mixed with water to form a dilute hypochlorite solution and is typically fed in the same manner as sodium
hypochlorite. For spot disinfection in a basin or pipe, calcium hypochlorite tablets are deposited in the
appropriate location, water is added, and the tablets allowed to dissolve to form a liquid hypochlorite
solution.
According to the 1995 Community Water Systems Survey (USEPA 1997b), most surface water
and ground water systems that have primary disinfection use chlorine. Exhibits 2.4 and 2.5 (displayed
previously) show that 54 percent of large ICR surface water and 85 percent of large ICR ground water
systems use chlorine. Additionally, these exhibits show that chlorine is also the most widely used
secondary disinfectant.
2.5.3 Pros and Cons
The following list presents selected advantages and disadvantages of using chlorine to disinfect
drinking water (Masschelein 1992; DeMers and Renner 1992).
Advantages
- Chlorine is an effective biocide.
- Chlorine oxidizes soluble iron, manganese, and sulfides.
- Chlorine enhances color removal.
- Chlorine controls taste and odor.
- The use of chlorine is the easiest and least expensive disinfection method, regardless of
system size.
Occurrence Assessment for the Final Stage 2 DBPR 2-9 December 2005
-------
- Chlorine is the most widely used disinfection method and, therefore, the most well
known.
- Chlorine is available as calcium and sodium hypochlorite, which are more advantageous
for smaller systems than chlorine gas because they are easier to use, safer, and need less
equipment compared to chlorine gas.
- Chlorine provides a residual, making it a good secondary disinfectant.
• Disadvantages
- Chlorine forms both halogenated and non-halogenated organic byproducts (some of
which pose health concerns).
- Chlorine gas is a hazardous and corrosive gas, and special leak containment and scrubber
facilities could be required to ensure safety.
- Sodium hypochlorite degrades over time and with exposure to light (which diminishes its
treatment effectiveness).
- Sodium hypochlorite is a corrosive chemical.
- Calcium hypochlorite requires proper storage. It must be stored in a cool, dry place to
reduce potential reactions. Also, an antiscalant chemical may be needed since impurities
may cause a precipitate to form.
- Higher concentrations of hypochlorite solutions are unstable and will produce chlorate as
a decomposition byproduct.
- Hypochlorite can contain bromate as a contaminant, resulting in an inadvertent
introduction of low concentrations of bromate to drinking water.
- Chlorine is less effective in water with higher pH.
- Chlorine forms biodegradable oxygenated byproducts that can lead to regrowth of
biological material in the distribution system.
Because of the variety of forms and dosage of chlorine and their different usage depending on system size
and water quality, not all of these advantages and disadvantages apply to all systems.
2.5.4 Dose Ranges and Points of Application
Two key operational parameters that affect DBF formation are disinfectant dose and point of
application. The chlorine dose range guideline used in reviewing public water systems' (PWSs) ICR
initial sampling plans was based on dosages provided in engineering design manuals and published
articles (see the Occurrence Assessment for Disinfectants/Disinfection Byproducts in Public Drinking
Water Supplies [USEPA 1998c] for further information). The combination of chlorination at the
treatment plant and strategic locations in the distribution system may be more effective at maintaining
residuals than an equivalent single dosage at the treatment plant (Tryby et al. 1999). Public health
benefits of booster chlorination can include decreased DBP formation (since lower residuals can be
achieved in the finished water) and better control of biological regrowth and biofilm formation in the
distribution system.
Exhibit 2.6 illustrates the difference in total chlorine dose in large surface water plants using only
free chlorine versus those using chloramines (median of 2.7 mg/L as C12 for chlorine-only plants versus a
median of 5.0 mg/L as C12 for chloramine plants). Chapter 15 in the Information Collection Rule Data
Analysis document (McGuire et al. 2002) provides additional information on chlorine dose for different
plant types (e.g., direct filtration, softening) and different influent TOC concentrations. Higher chlorine
doses applied in surface water plants using chloramines may be necessary to achieve adequate
disinfection due to higher levels of organic material. Also, higher chloramine residuals can be maintained
Occurrence Assessment for the Final Stage 2 DBPR 2-10 December 2005
-------
without the subsequent taste and odor complaints or the same increase in DBF formation in comparison to
free chlorine. Finally, higher doses are need to meet CT requirements for surface water systems.
Exhibit 2.7 shows the difference in total chlorine dose in ground water plants using only free
chlorine versus those using chloramines (median of 1.6 mg/L as C12 for chlorine-only plants versus a
median of 5.0 for chloramine plants). Higher chlorine doses applied in ground water plants using
chloramines may be a result of lower levels of organic material allowing for greater chlorine doses
without the subsequent increase in chlorinated DBF levels. Also, higher chloramine residuals can be
maintained without the subsequent taste and odor complaints or the same increase in DBF formation in
comparison to free chlorine. There are no CT requirements for ground water systems.
Exhibit 2.6 Cumulative Distributions of Mean Total Chlorine Dose for
Surface Water Plants
100%
80%
60%
OCL2 Only Plants (N=166)
ACL2_CLM or CLM Only Plants (N=85)
40%
20% -
0%
4 6 8 10
Plant-Mean Total CI2 Doses (mg CI2/L)
12
Source: ICR AUX2 Database (USEPA 2000I).
Query: Screened SWPlant-Mean CL2 Doses (w AUX2). See Appendix B for details.
Occurrence Assessment for the Final Stage 2 DBPR 2-11
December 2005
-------
Exhibit 2.7 Cumulative Distributions of Mean Total Chlorine Dose for
Ground Water Plants
100%
80%
60%
o
o
O A
O
O
O A
A
A
Q. 40%
20%
0%
O
o
o CL2 Plants (N=28)
ACLM Plants (N=14)
o
o
o
o
o
o
o
o
O A
o
10
12
14
Plant-Mean Total CI2 Dose (mg CI2/L)
Source: ICR AUX2 Database (USEPA 2000I).
Query: Screened SWPlant-Mean CL2 Doses (w AUX2). See Appendix B for details.
Occurrence Assessment for the Final Stage 2 DBPR 2-12
December 2005
-------
Exhibit 2.8, taken from Table 15.3 in the Information Collection Rule Data Analysis document
(McGuire et al. 2002), shows the number of disinfectant application points for various types of surface
water plants using free chlorine only. The data indicate that most plant types used either one or two
application points. Approximately 10 percent of direct filtration, in-line filtration plants, and
conventional plants used three C12 application points. Plants with multiple application points were most
likely using C12 to address multiple treatment objectives (e.g., disinfection, preoxidation, taste and odor
control, and prevention of microbial growth within unit processes).
Exhibit 2.8 Number of Disinfection Points in Plants Using Only Free Chlorine by
Plant Type
riant-montns (u/o)
Type of plant
Conventional (AH 696)
Softening (M=74)
Direct/in-line filtration (N=206)
Unfiltered(A&45)
Otherf (N=27)
1 location
37
51
41
47
100
2 locations
53
49
49
53
0
3 locations
10
0
10
0
0
>3 locations
0
0
0
0
0
Source: Information Collection Rule Data Analysis (McGuire et al. 2002).
Notes: Based on data collected during the last 12 months of ICR monitoring.
f Includes one membrane, one slow sand filtration, and two other plants.
Exhibit 2.9, taken from Figure 15.3 in the Information Collection Rule Data Analysis document
(McGuire et al. 2002), shows the proportion of plants using specific chlorine application locations (e.g.,
rapid mix). Results are shown separately for subsets of plants with one, two, or three points of
application. Results for the three subsets were consistent: plants apply disinfection primarily at the
clearwell, at a point just prior to filtration, or at the rapid mix stage. For plants with one application point,
rapid mix is the application point 36 percent of the time, the point just prior to filtration 28 percent, and
the clearwell is used 27 percent of the time. For plants with two application points, the rapid mix is used
53 percent, the point just prior to filtration 32 percent, and the clearwell is used 80 percent of the time.
Given the prevalence of the rapid mix stage and the point just prior to filtration as the application points,
chlorine-only systems may have some flexibility to move the point of chlorination to reduce the formation
of DBFs. Moving the point of chlorination until after the sedimentation, flocculation, and filtration
allows chlorine to be added after the majority of DBF precursors have been removed from the water,
thereby minimizing DBF formation. However, this may hinder overall disinfection, taste and odor
control, coagulation, and particulate removal (in the case of biofiltration removing assimilable oxygenated
byproducts).
Occurrence Assessment for the Final Stage 2 DBPR 2-13 December 2005
-------
Exhibit 2.9 Number of Chlorine Application Locations in Conventional Plants
Using Only Free Chlorine
1UU
M vO
fl 80
o
S 7fl
-------
2.6 Chloramines
Chloramines were first considered for use in disinfection after scientists observed that
disinfection still occurred when ammonia was present, even though free available chlorine had dissipated.
This lingering disinfection was caused by inorganic Chloramines.
Chloramines were used regularly for disinfection during the 1930s and 1940s to provide a
residual disinfectant and to control taste and odor. Because of an ammonia shortage during World War II,
however, the popularity of chloramination declined. In recent years, choramines were recognized as
being more stable than free chlorine in the distribution system and, consequently, were found to be
effective in controlling bacterial regrowth in the distribution system (LeChevallier et al. 1996). The
concern over halogenated organic byproduct (THM and HAA) formation in water treatment and
distribution systems has increased interest in chloramines because they react differently with NOM than
chlorine, generally producing lower concentrations of DBFs (Symons et al. 1998). Currently,
monochloramine is used to disinfect drinking water in approximately 25 percent of U.S. municipalities
(Kooletal. 1999).
2.6.1 Description of Chemistry
Chloramines are formed by the reaction of ammonia with aqueous chlorine. In aqueous solutions,
hypochlorous acid from the chlorine reacts with ammonia to form inorganic chloramines in a series of
competing reactions. In these reactions, monochloramine (NH2C1), dichloramine (NHC12), or nitrogen
trichloride (NC13), also referred to as trichloramine, are formed. These competing reactions depend
primarily on pH and are controlled to a large extent by the chlorine:ammonia nitrogen ratio (C12:NH3-N).
Temperature and contact time also regulate this reaction. Monochloramine is formed primarily when the
applied C12:NH3-N ratio is less than 5:1 by weight. When certain ratios of chlorine and ammonia nitrogen
are present, chloramines may not form, and ammonia and chlorine may be converted to other molecules
that do not act as disinfectants and are not detected when chlorine residual is measured. For instance, as
the applied C12:NH3-N ratio increases from 5:1 to 7.6:1, a "breakpoint" reaction occurs, reducing the
residual chloramine and ammonia nitrogen level to a minimum. Breakpoint chlorination results in the
formation of nitrogen gas or nitrate and hydrochloric acid. At C12:NH3-N ratios above 7.6:1, free chlorine
and nitrogen trichloride are present; being quite volatile, the latter usually dissipates. To avoid breakpoint
reactions, utilities normally maintain a C12:NH3-N ratio of between 3:1 and 5:1 by weight. A ratio of 6:1
is actually optimum for disinfection because dichloramine predominates (dichloramine is a stronger
disinfectant than monochloramine), but maintaining a stable operation at that point on the breakthrough
curve is difficult. Therefore, as noted above, a C12:NH3-N ratio of 3:1 to 5:1 is typically accepted as
optimal for chloramination.
2.6.2 Use and Distribution
Monochloramine is used primarily as a secondary disinfectant for maintaining a residual in the
distribution system. Monochloramine can be formed by adding ammonia first and then chlorine, by
adding chlorine first and then ammonia, or by concurrently adding both reactants. Ammonia is added
first when the formation of objectionable taste and odor compounds caused by the reaction of chlorine
and organic matter are a concern. Inactivation for bacteria is better when it is mixed in water to be treated
rather than using preformed chloramines because you get the inactivation properties of chlorine, while
combining free chlorine with ammonia instead of organics. Currently, most drinking water systems add
chlorine first and then ammonia, in order to meet the EPA Surface Water Treatment Rule (SWTR)
Occurrence Assessment for the Final Stage 2 DBPR 2-15 December 2005
-------
disinfection requirements. The point of ammonia addition typically is selected to "quench" the free
chlorine residual after the optimal contact time has been achieved.
2.6.3 Pros and Cons
The following list highlights selected advantages and disadvantages of using chloramines as a
method of disinfecting drinking water (Masschelein 1992).
Advantages
- The monochloramine residual is more stable and lasts longer than free chlorine or
chlorine dioxide—thereby providing better bacterial regrowth protection in the
distribution system.
- Chloramines can be very effective in addressing taste and odor problems.
- Chloramines are inexpensive and easy to produce.
- Production of chlorinated DBFs is minimized if C12:NH3 ratio is maintained at 3:1 to 5:1.
Disadvantages
- The disinfecting properties of chloramine are not as strong as other disinfectants, such as
chlorine, ozone, and chlorine dioxide.
- Chloramines cannot oxidize iron, manganese, or sulfides.
- When using chloramine as the secondary disinfectant, it may be necessary to periodically
convert to free chlorine for control of nitrification, which can be caused by excess
ammonia in the distribution system.
- Dichloramines can pose problems for taste and odor if the C12:NH3-N ratio isn't
maintained between 3:1 and 5:1.
- Two separate storage, feed, and control systems (one for chlorine, one for ammonia) must
be used.
- As C12:NH3 ratio approaches the breakpoint, the greater the potential for DBF formation.
Not all of these advantages and disadvantages apply to all systems, depending on the dosages of
chloramine and water quality.
2.6.4 Dose Ranges and Points of Application
The normal primary disinfection dose range for monochloramine is 1.0 to 4.0 mg/L. The
minimum dosage of monochloramine in the distribution system is typically 0.5 mg/L (Texas Natural
Resource Conservation Commission 1996). Exhibit 2.6 showed the cumulative distributions of mean
chloramine dose for surface water plants. Another way to characterize total chlorine dose is by C12:NH3-
N weight ratios. These weight ratios determine the species of chloramines formed (e.g., monochloramine,
dichloramine, trichloramine). Exhibit 2.10 shows the distribution of C12:NH3-N weight ratios in plants
using chloramines. The distributions indicate that approximately 95 percent of plants using chloramines
had a plant average C12:NH3-N weight ratio above 2.5 (McGuire et al. 2002).
Occurrence Assessment for the Final Stage 2 DBPR 2-16 December 2005
-------
Exhibit 2.10 CI2:NH3-N Weight Ratios in Surface Water CL2_CLM and CLM Plants
10
(Plant
15 20 25 30 35 40 45
C12:N Weight Ratios
Averages [indicated by o], Minimums, and Maximums)
50
Source: Information Collection Rule Data Analysis (McGuire et al. 2002).
Approximately 92 percent of ICR plants using CL2_CLM, free chlorine followed by the
application of ammonia, have one point of application of ammonia and the remaining 8 percent have two
points of application of ammonia. For CLM plants, 78 percent have one point of application and 22
percent have two points of application. Figures 15.4A and B in the Information Collection Rule Data
Analysis document (McGuire et al. 2002) show the point of application of ammonia for CL2_CLM and
CLM plants. For CL2_CLM plants with one point of application, 53 percent applied ammonia at the
clearwell, while approximately 19 percent applied ammonia at the rapid mix stage (the remaining plants
added ammonia during flocculation, sedimentation, or prior to filtration). For plants with two points of
application, most plants applied ammonia at the rapid mix and clearwell points. For CLM plants, the
rapid mix point was the predominant point of application for plants with both one and two points of
application. The predominance of the rapid mix point indicates that ammonia is being added early in the
treatment train to minimize DBF formation or to increase the contact time of chloramines throughout the
treatment plant.
2.6.5 Byproducts
The effectiveness of chloramines in controlling DBF production depends upon a variety of
factors, notably the chlorine to ammonia ratio, the point of addition of ammonia relative to that of
chlorine, the effectiveness of mixing, and pH levels in the water.
Direct reactions between monochloramine and organic matter in water produce very few
halogenated organic compounds. However, some dichloroacetic acid can be formed, and cyanogen
chloride formation is greater than with free chlorine (Jacangelo et al. 1989; Smith et al. 1993; Cowman
and Singer 1994; Symons et al. 1998). If chlorine and NH3 are added separately to water (not pre-
formed), then some free chlorine is available to react with organic matter. Another potential source of
free chlorine is monochloramine, which slowly hydrolyzes to free chlorine in an aqueous solution.
Therefore, halogen-substitution reactions occur even when pre-formed monochloramine is used (Rice and
Occurrence Assessment for the Final Stage 2 DBPR 2-17
December 2005
-------
Gomez-Taylor 1986). The closer the C12:NH3-N ratio is to the breakpoint, the greater the formation of
DBFs (Speed et al. 1987).
The application of chloramines results in the formation of total organic halide (TOX), which
includes unidentified organic byproducts. However, TOX formation occurs to a much lesser degree than
it would given an equivalent dose of free chlorine. Little is known about the nature of these byproducts,
except that they are more hydrophilic and larger in molecular size than the organic halides produced from
free chlorine (Jensen et al. 1985; Singer 1992; Symons et al. 1998).
2.7 Chlorine Dioxide
Chlorine dioxide is a powerful oxidant originally used by industries as a bleaching agent and
disinfectant. Chlorine dioxide was first used for drinking water treatment in 1944 at the Niagara Falls,
New York Water Treatment Plant. Currently, the major use of chlorine dioxide is as a pre-oxidant to
control tastes and odors and to reduce THM formation in finished water (DeMers and Renner 1992).
2.7.1 Description of Chemistry
Chlorine dioxide (C1O2) is a neutral compound of chlorine in the +IV oxidation state. C1O2 is a
yellow to red colored gas at temperatures above 11-12°C. Because C1O2 does not hydrolyze in water, it
exists as a dissolved gas as long as the pH of the water ranges from 2 to 10. In strongly alkaline solutions
(pH greater than 9 or 10), however, formation rates of DBFs increase with increasing concentrations of
chlorine dioxide. Chlorine dioxide is a volatile free radical that functions as an oxidant by way of a one-
electron transfer mechanism in which it is reduced to chlorite (C1O2 ) (Hoehn et al. 1996; Noack and
Doerr 1978). During drinking water treatment, chlorite is the predominant reaction byproduct, with 50 to
70 percent of the reacted chlorine dioxide being converted to chlorite and 30 percent to chlorate (C1O3 )
or chloride (Cl~) depending on the secondary disinfectant.
Although chlorine dioxide can be produced from sodium chlorate (NaClO3), for most potable
water applications, chlorine dioxide is generated from sodium chlorite (NaClO2). The proportion of
chlorine dioxide relative to impurities, including chlorite, chlorate, or free chlorine, is important when
chlorine dioxide is applied to drinking water (Aieta and Berg 1986). Although a significant amount of
chlorite ion can appear in drinking water from the application and subsequent reduction of chlorine
dioxide, both precursor chlorite and chlorate ions can be constituent contaminants in generated solutions.
EPA recommends that systems limit the formation of chlorite and chlorate by maintaining high generator
purity (i.e., more than 95-percent efficiency) and limiting excess chlorine to no more than 5 percent of the
applied dose of chlorine dioxide. Two feed chemical combinations that generate chlorine dioxide yield in
excess of 95 percent are chlorine-sodium chlorite and acid-sodium hypochlorite-sodium chlorite.
Several feed chemical combinations that are used in the water industry are described below.
Acid-Chlorite Solution. Chlorine dioxide can be generated by acidification of a sodium chlorite
solution, usually with hydrochloric acid, and several stoichiometric reactions have been reported
for such processes (Gordon et al. 1972). When catalyzed by the presence of chloride ions, acid
activation of sodium chlorite has a maximum possible yield of 80 percent of the quantity of
chlorine dioxide that could be produced from a reaction of the same amount of sodium chlorite
with chlorine (Petochelli 1995). The reaction is relatively slow, and production rates using this
Occurrence Assessment for the Final Stage 2 DBPR 2-18 December 2005
-------
method are practically limited to about 25 to 30 pounds per day, due to the exothermic nature of
the reactions.
Chlorine Solution-Chlorite Solution. Chlorite ions (from dissolved sodium chlorite) will react
in aqueous solutions with chlorine or hypochlorous acid to form chlorine dioxide. Two moles of
chlorite ions will theoretically react with one mole of chlorine to produce two moles of chlorine
dioxide. To fully utilize the sodium chlorite solution, excess chlorine is often used, reducing the
pH and driving the reaction further toward completion. The reaction is faster than the
acid-chlorite solution method, but much slower than the other commercial methods described in
the following discussion. Chlorine dioxide production by this method is limited to about 1,000
pounds per day.
Chlorine Gas-Chlorite Solution. Sodium chlorite solution can be "vaporized" and reacted in a
vacuum with molecular gaseous chlorine. This process uses concentrated reactants and is much
more rapid than chlorine solution-chlorite solution methods (Petochelli 1995). If the chlorine
and chlorite ions react stoichiometrically, the resulting pH is close to 7. Production rates are
virtually unlimited, and some systems have reported producing more than 60,000 pounds per day.
Chlorine Gas-Solid Chlorite. This process reacts dilute, humidified chlorine gas with specially
processed solid sodium chlorite contained in sealed reactor cartridges. The reaction is rapid and
produces high-purity chlorine dioxide gas inherently free of chlorine and chlorate ions because
these ions do not carry into the gas phase. Using multiple cartridges in series ensures an excess
of sodium chlorite; thus, all chlorine is reacted and the chlorine dioxide produced is chlorine-free.
Because the chlorine dioxide production rate is solely a function of the chlorine gas feed rate,
generators that use chlorine gas-solid chlorite technology are capable of infinite turndown (i.e.,
the chlorine dioxide production rate can be adjusted without requiring recalibration between
settings) (Petochelli 1995; Hoehn and Rosenblatt 1996). Chlorine gas-solid chlorite solution
method production capacities are limited to 2,000 pounds per day.
In addition to the commercial processes discussed previously, other potential methods for
generating chlorine dioxide include electrolysis of a sodium chlorite solution (with or without the use of
membranes to purify the chlorine dioxide product), irradiation of dilute sodium chlorite solution with UV
light, and reduction of sodium chlorate with concentrated sulphuric acid and 50 percent hydrogen
peroxide.
2.7.2 Use and Distribution
Chlorine dioxide is almost never used commercially as a gas because it cannot be safely
compressed and shipped. For potable water treatment process, it is predominantly generated in aqueous
solutions. Because of the volatile nature of the gas, chlorine dioxide works extremely well in plug flow
reactors, such as pipe lines. It can be easily removed from dilute aqueous solution by aerated turbulence,
such as in a rapid mix tank or aerated cascade. For post-disinfection, chlorine dioxide can be added
before clearwells or transfer pipelines.
An estimated 700 to 900 U.S. drinking water systems use chlorine dioxide, largely to oxidize
iron and manganese, control taste and odor, and reduce THM levels (Hoehn et al. 1992). Some systems
are looking to the higher disinfection efficacy of chlorine dioxide to decrease contact time needed for
Occurrence Assessment for the Final Stage 2 DBPR 2-19 December 2005
-------
Cryptosporidium control. Nineteen plants (3.7 percent) that participated in the ICR reported using
chlorine dioxide for at least nine of the last 12 months of the ICR collection period (USEPA 2000i).
2.7.3 Advantages and Disadvantages
The following list highlights selected advantages and disadvantages of using chlorine dioxide to
disinfect drinking water (Masschelein 1992; DeMers and Renner 1992; Gallagher et al. 1994).
Advantages
- More effective than chlorine and chloramines for inactivation of viruses (with longer
contact times), Cryptosporidium, and Giardia (with shorter contact times).
Oxidizes iron, manganese, and sulfides.
- Provides taste, odor, and color control.
- Under proper generation conditions (i.e., no excess chlorine), TTHM is not formed.
- Biocidal properties are not influenced by pH.
• Disadvantages
- Incomplete generation of chlorine dioxide leaves unreacted chlorite and chlorate.
Generator inefficiency and optimization difficulty can result in excess chlorine feed at
the application point, leading to formation of halogenated organic DBFs.
Costs are a concern: training, sampling, and laboratory testing for chlorite and chlorate
are expensive; in many cases equipment must be rented; and the cost of the sodium
chlorite is high.
- Measuring a chlorine dioxide residual for determining disinfection credit is difficult.
Chlorine dioxide gas is explosive, so it must be generated on-site and requires careful
handling.
Chlorine dioxide decomposes in sunlight.
- Chlorine dioxide can lead to production of noxious odors for customers in some systems
if chlorine dioxide is present at the tap.
- It is difficult to maintain a chlorine dioxide residual in the distribution system.
Because of the wide variation in system size, water quality, and resulting dosages of chlorine dioxide
applied, not all of these advantages and disadvantages apply to all systems.
2.7.4 Dose Ranges
Before chlorine dioxide, or any disinfectant, is selected as a primary disinfectant, an oxidant
demand study must be completed. Ideally, this study should consider the seasonal variations in water
quality, temperature, and application points. EPA recommends that the combined concentrations of
chlorine dioxide, chlorate, and chlorite not exceed 1.0 mg/L in finished water. This means that if the
desired oxidant dosage is greater than about 1.4 mg/L, the chlorite/chlorate byproduct concentrations
would already be at the maximum level (based on a roughly 70 percent conversion rate); therefore,
chlorine dioxide would not be acceptable as a disinfectant. Higher doses are possible if the plant uses
ferrous ion treatment to remove chlorite or reduced sulfur compounds (Singer and Reckhow 1999). The
range of doses includes both primary and secondary disinfection, although chlorine dioxide typically is
Occurrence Assessment for the Final Stage 2 DBPR 2-20 December 2005
-------
used for primary disinfection. Note that these ranges represent the extremes; normal doses fall within
these ranges.
Exhibit 2.11 presents chlorine dioxide doses (average, minimum, and maximum of all plant-
months where data was reported for nine of the last 12 months of the ICR collection period) for surface
water plants using chlorine dioxide. A number of surface water plants using chlorine dioxide (58 percent)
had average chlorine dioxide doses between 1.0 and 1.5 mg/L as C1O2.
Exhibit 2.11 Chlorine Dioxide Doses
(Plant Minimum, Mean, and Maximum)
100% -
80% -
60% -
01
fe 40%
Q_
20% -
0%
i e 1
i e 1
e
i e-
i-e—
0.5
i e 1
CD
1 1.5 2 2.5
Chlorine Dioxide Dose (mg CIO2/L)
3.5
Note: Open circles represent plant means and lines represent minimum and maximum values.
Source: ICR AUX2 Database (USEPA 2000I).
Query: Screened SWPlant-Mean CLX Doses (wAUX2). See Appendix B for details.
2.7.5 Byproducts
Small amounts of chlorine are often present when chlorine dioxide is used, so halogenated
organic DBFs are often detected in small quantities. However, the application of chlorine-free chlorine
dioxide does not form THMs and produces only a small amount of TOX (Werdehoff and Singer 1987)
and other halogenated-substituted compounds at very low concentrations (Richardson 1998). Primarily,
however, the application of chlorine dioxide to water results in oxidation/reduction reactions that form
two inorganic DBFs: chlorite and chlorate (Rav-Acha et al. 1984; Werdehoff and Singer 1987). Chlorite
and chlorate frequently are found as contaminants in chlorine dioxide feed streams, and chlorite is formed
as a byproduct from disinfection with chlorine dioxide (Griese et al. 1991). However, chlorine dioxide
Occurrence Assessment for the Final Stage 2 DBPR
2-21
December 2005
-------
does not generate bromine-substituted byproducts to the same extent as ozone in bromide-containing
waters. Chlorite in drinking water results from two parts of the chlorine dioxide disinfection process:
• Unreacted chlorite from the chlorine dioxide generation process
• Reduction of chlorine dioxide when it reacts with organic matter in water
Incomplete reaction or non-stoichiometric addition of the sodium chlorite and chlorine reactants
can result in unreacted chlorite or, more likely, chlorate in the chlorine dioxide feed stream. Upon
application to water, chlorine dioxide is fairly unstable and rapidly dissociates into chlorite and chlorate at
pHs above 10. This occurs only to a limited extent where residuals of chlorine dioxide are greater than 1
percent. Chlorite ions are the primary product of chlorine dioxide reduction, but the percentage of
chlorite and chlorate present is influenced by pH and sunlight, as well as the efficiency of the chlorine
dioxide generator.
The quantity of chlorate produced during chlorine dioxide generation increases with excess
chlorine addition. Similarly, low pH can increase the quantity of chlorate during chlorine dioxide
generation. The predominant source of chlorate ions in finished water, however, results from the
oxidation of chlorite (from the applied chlorine dioxide) by free available chlorine used as a final
distribution system disinfectant (Gallagher et al. 1994). Consequently, chlorate concentrations are
expected to increase with increasing contact time in water containing chlorite and chlorine. Once formed,
chlorate is stable in finished drinking water.
2.8 Ozonation
Ozone (O3) is used in water treatment for disinfection and oxidation. Early application of ozone
was primarily for non-disinfection purposes, such as color removal or taste and odor control. Since
implementation of the SWTR, Stage 1 DBPR, and IESWTR, ozone usage for disinfection has increased.
Ozone is a powerful oxidant capable of oxidizing many organic and inorganic compounds in water.
Ozone was first used for drinking water treatment in 1893 in the Netherlands. While used
frequently in Europe to disinfect drinking water, ozonation technology was slow to transfer to the United
States. In 1991, approximately 40 water treatment plants serving more than 10,000 people in the United
States used ozone (Langlais et al. 1991). This number had grown to 201 by 1997 (Rice and Dimitrou
1997). Most of these facilities are small: 90 plants treat fewer than 1 million gallons per day (mgd) and
only six exceeded 100 mgd as of May 1997. ICR data show that 14 large surface water plants reported
using ozone for at least nine of the last 12 months of the ICR collection period (USEPA 2000d). Another
source cites that as of January 2000, 275 plants were using ozone, with another 16 plants expected to
come on line in the next year (Rice 2000). Many of these plants are using ozone for purposes besides
disinfection. Rice also estimates that many very small systems in California, many of which may be non-
community systems, use ozone as a disinfectant in their storage tanks.
Occurrence Assessment for the Final Stage 2 DBPR 2-22 December 2005
-------
2.8.1 Description of Chemistry
A gas at room temperature, ozone is highly corrosive and toxic. The gas is colorless with a
pungent odor readily detectable at concentrations as low as 0.01 to 0.05 parts per million (ppm), which is
below concentrations that would cause a health problem.
Ozone decomposes spontaneously in water by a complex reaction involving the generation of
oxygen and hydroxyl free radicals. Hydroxyl radicals are among the most reactive oxidizing agents in
water due to their unpaired electrons (Hoigne and Bader 1983a; Hoigne and Bader 1983b; Glaze et al.
1987). Ozone reacts in two modes in aqueous solutions: direct oxidation of compounds by aqueous
ozone (O3(aq)) and oxidation of compounds by hydroxyl radicals (OH") produced during the spontaneous
decomposition of ozone (Hoigne and Bader 1977).
2.8.2 Use and Distribution
Ozone is used in drinking water treatment for various purposes.
• Disinfection to control Giardia, Cryptosporidium, and other microbes
• Inorganic pollutant oxidation, including iron, manganese, and sulfide
• Organic micropollutant oxidation, including taste and odor compounds, phenolic pollutants,
and pesticides
Organic macropollutant oxidation, including color removal, increasing the biodegradability
of organic compounds, THM and TOX precursor control, and destruction of chlorine demand
• Improvement of coagulation and filtration
Ozone is unstable, so it must be generated at the point of application. It is generally formed by combining
an oxygen atom with an oxygen molecule (O2). This reaction is endothermic and requires considerable
energy. Ozone can be produced several ways, including by irradiating an oxygen-containing gas with
electrolytic reactions, ultraviolet light, or high-energy radiation. These are all processes that produce free
oxygen radicals from electron or photon energy input.
One method, corona discharge, predominates in the water industry. Corona discharge, also
known as silent electrical discharge, consists of passing an oxygen-containing gas through two electrodes
separated by a dielectric and an air gap. A voltage is applied to the electrodes, causing an electron flow
across the air gap. These electrons provide the energy to dissociate the oxygen molecules, leading to the
formation of ozone in the gas stream. Then, the ozone is transferred into the water, with any non-
transferred ozone being converted to oxygen before being released into the atmosphere. Therefore, no
chemical inputs are needed.
For most applications, ozone is applied either to the raw water or after some type of clarification
process. Turbidity and ozone demand (the amount of ozone required to oxidize all the constituents in the
water) influence the way ozone is used in the treatment process. By moving the ozonation process further
downstream, the ozone demand and production of oxidation byproducts are reduced. The advantage of
placing ozone ahead of filtration is that biodegradable organics produced during ozonation can be
Occurrence Assessment for the Final Stage 2 DBPR 2-23 December 2005
-------
removed in the filters if they are allowed to operate biologically (i.e., with no disinfectant residual).
Bacteria living in the biofilm growing on filters can break down and feed on the oxidized NOM.
Biological filtration is often necessary for waters that have high levels of NOM.
2.8.3 Advantages and Disadvantages
The following list highlights selected advantages and disadvantages of using ozone to disinfect
drinking water (Masschelein 1992).
• Advantages
- Ozone is more effective than chlorine, chloramines, and chlorine dioxide for inactivation
of viruses, Cryptosporidium, and Giardia.
- Ozone oxidizes iron, manganese, and sulfides.
- Ozone can sometimes enhance the clarification process and turbidity removal.
- Ozone improves color, taste, and odors.
- Ozone requires a very short contact time.
- Halogenated organic DBFs are not formed by ozonation if bromide is absent.
- Enhances the biodegradability of natural and synthetic organic compounds and destroys
many organic compounds.
- Since most of the oxidant demand is satisfied by ozone, the amount of chlorine needed
for secondary disinfection is generally much lower.
• Disadvantages
- DBFs formed include bromate and bromine-substituted DBFs (when bromide is present),
as well as aldehydes and ketones (if there is incomplete oxidation of some organic
compounds and acids).
- The initial cost of ozonation equipment is high.
- The generation of ozone is energy-intensive and must be generated on site.
- Ozone is highly corrosive and toxic.
- Ozone decays rapidly at high pH and warm temperatures.
- Ozone provides no residual and therefore a secondary disinfectant such as chlorine may
be needed for the distribution system.
- Ozone plants require a higher level of maintenance and operator skill.
- Low disinfection efficiency at low water temperatures.
- Ozone forms biodegradable oxygenated byproducts that can lead to regrowth of
biological material in the distribution system.
- Storage of all the necessary oxygen feed gas generators, the ozone generation equipment,
the cooling equipment, and the off gas collection and destruction equipment, is likely to
take up more space than a typical liquid storage and feed facility
Because of the wide variation in system size, water quality, and ozonation dosages applied, not all of
these advantages and disadvantages apply to all systems.
Occurrence Assessment for the Final Stage 2 DBPR 2-24 December 2005
-------
2.8.4 Dose Ranges
Engineering design manuals and published articles were used in developing an ozone dose range
for the guideline used to review PWSs' ICR initial sampling plans. This range of doses is for primary
disinfection only. Note that these ranges represent extremes, and normal values fall between these values.
Ozone plants participating in the ICR also reported the doses of ozone they used. Exhibit 2.12 presents
ozone doses (average, minimum, and maximum of all plant-months where data was reported for nine of
the last 12 months of the ICR collection period) for surface water plants using ozone. Approximately 86
percent of surface water plants using ozone had an average ozone dose below 3.0 mg/L as O3.
Exhibit 2.12 Ozone Doses
(Plant Minimum, Mean, and Maximum)
100% -
80% -
II 60% -
oi
fe 40% -
Q.
20% -
0%
i e-
i e—i
i-e-i
e<
6 8 10 12
Ozone Dose (mg O3/L)
14
16
18
Note: Open circles represent plant means and lines represent minimum and maximum values.
Source: ICR AUX2 Database (USEPA 2000I).
Query: Screened SWPlant-Mean O3 Doses (wAUX2). See Appendix B for details.
2.8.5 Byproducts
A variety of organic and inorganic byproducts have been observed following ozonation of water.
Ozone can react with bromide naturally present in water to form bromate and bromine-substituted DBFs.
The primary factors affecting the speciation and concentrations of bromine-substituted byproducts are pH
and the ratios of ozone-to-bromide and total organic carbon-to-bromide (Singer 1992). Refer to Chapter
15 of the Information Collection Rule Data Analysis document (McGuire et al. 2002) for data on source
water bromide concentrations for plants using ozone.
Occurrence Assessment for the Final Stage 2 DBPR 2-25
December 2005
-------
The principal benefit of using ozone to control THM formation is that ozone allows free chlorine
to be applied at lower doses later in the treatment process, after some of the TTHM and HAAS precursors
have been removed, thereby reducing the potential for TTHM and HAAS formation. However,
application of a secondary disinfectant following ozonation requires special consideration for potential
interaction between disinfectants. For example, chloral hydrate formation has been observed to increase
when chlorine is used as a secondary disinfectant after ozone (McKnight and Reckhow 1992; Logsdon et
al. 1992). One byproduct of ozonation, acetaldehyde, is a known precursor of chloral hydrate.
Enhancement of chloral hydrate formation has not been observed when monochloramine is applied as the
secondary disinfectant, or if biologically active filtration is used following ozonation and prior to
chlorination (Singer 1992). Chloropicrin formation from free chlorine also appears to be enhanced by
pre-ozonation (Hoigne and Bader 1988).
Organic oxidation byproducts, including aldehydes, ketones, aldo-acids, ketoacids, carboxylic
acids, and assimilable organic carbon (AOC) can be formed upon ozonation of water containing a high
level of NOM. The primary aldehydes that have been detected are formaldehyde, acetaldehyde, glyoxal,
and methyl glyoxal (Glaze et al. 1991). The ICR data provided occurrence information on these
substances along with propanal, pentanal, and butanal. Total aldehyde concentration in drinking water
disinfected with ozone ranges from less than 5 to 300 i-ig/L, depending on the TOC concentration and the
applied ozone-to-organic-carbon ratio (Van Hoof et al. 1985; Yamada and Somiya 1989; Glaze et al.
1989; Krasner et al. 1989; Glaze et al. 1991; LeLacheur et al. 1991).
Ozonation of water containing bromide can produce hypobromous acid and hypobromite, which,
in turn, can contribute to the formation of bromine-substituted byproducts, the brominated analogues of
the chlorinated DBFs. These bromine-substituted byproducts include: bromoform; the bromine-
substituted acetic acids, acetonitriles, and aldo-acids; bromopicrin; and cyanogen bromide. Bromoform
has been found to form when water is ozonated (McGuire et al. 1990). Cyanogen bromide has been
found to form when water is ozonated (McGuire et al. 1990). An ICR plant with the median ozone dose
of 1.84 mg/L had influent bromide levels of 0.133 mg/L and finished water bromate levels of 3.1 i-ig/L.
Dibromoacetic acid levels of 3.5 |ig/L were also detected, but all other brominated DBFs were below their
minimum reporting levels.
Ozone can react with the hypobromite ion to form bromate (Siddiqui and Amy 1993; Krasner et
al. 1993; Amy et al. 1997). Bromate formation is affected by NOM, pH, bromide ion concentrations,
inorganic carbon, and ozone dose. Decreasing pH (8.5-6.5) generally decreases bromate formation
because the equilibrium is shifted to hypobromous acid, which does not form bromate. Lower pH,
however, enhances the formation of bromine-substituted DBFs formed by the reaction of hypobromous
acid and NOM. Higher bromide ion concentrations and high inorganic carbon concentrations have been
noted with increased bromate ion formations (Amy et al. 1997).
The amount of bromide incorporated into the detected DBFs accounts for only one-third of the
total source water bromide concentration. This indicates that other bromine-substituted DBFs exist that
are not yet identified (McGuire et al. 1989; AWWARF 1991).
Occurrence Assessment for the Final Stage 2 DBPR 2-26 December 2005
-------
3. National DBF Occurrence: Pre-Stage 1 Baselines
This chapter summarizes the data used to assess disinfectant residuals and disinfection byproduct
(DBF) occurrence in public drinking water supplies. The Information Collection Rule (ICR) data for
large surface water plants and other data for medium sized plants was collected prior to compliance
deadline for the Stage 1 DBPR (2002 for medium and large surface water systems and 2004 for small
surface water systems and all ground water systems). This data is used in this chapter to investigate DBF
occurrence before the implementation of Stage 1 (pre-Stage 1). Chapter 4 provides a prediction of post-
Stage 1 DBF occurrence.
The main source of DBF data for this analysis is the ICR, which authorized EPA to require the
collection of occurrence and treatment information from disinfecting water systems serving 100,000 or
more people. The ICR data described in this chapter are from the AUX1 database, CD version 5.0
(USEPA 2000d). Information about medium (10,000 to 99,999 people served) and small (fewer than
10,000 people served) systems comes from the National Rural Water Association (NRWA) Survey
(USEPA 200 Ig), ICR Supplemental Surveys, the Water Utility Database (WATERASTATS), and data
provided by several States. Because the data available for medium and small systems are not as extensive
as the ICR data, the majority of this chapter is a presentation of ICR data for individual water quality
parameters and DBFs for large systems. EPA found that there are significant similarities between large
systems and medium and small systems with regard to source water quality (affecting DBP formation)
and use of treatment technologies. Because of these similarities, EPA expects that small and medium
systems would find DBP distribution system levels similar to those found in large systems following
compliance with the Stage 1 DBPR requirements.
The organization of the remainder of this chapter is as follows:
Section 3.1 presents large system occurrence data provided through the ICR for DBP
precursors and other parameters that affect DBP formation, disinfectant residuals, and DBFs.
Section 3.2 presents medium and small system occurrence data derived from sources other
than the ICR.
• Section 3.3 evaluates co-occurrence among certain ICR water quality parameters and the
relationships of these interactions.
Section 3.4 evaluates regional occurrence trends for some DBP precursors.
Data analyses in this chapter are supported by two appendices. Appendix A provides summary
information on the individual species of trihalomethanes (THMs) and five haloacetic acids (HAAs).
Appendix B provides the Microsoft Access™ query language that was used to extract data from the ICR
AUX1 database. There are uncertainties in using the ICR to characterize the pre-Stage 1 baseline. See
Chapter 3 of the Stage 2 Economic Analysis for a detailed discussion of these uncertainties.
Occurrence Assessment for the Final Stage 2 DBPR 3-1 December 2005
-------
3.1 ICR Data
Sections 3.1.1 through 3.1.3 present the ICR summary data for large systems with a description of
analyte characteristics as follows:
Section 3.1.1 DBF Precursors
Section 3.1.2 Disinfectant Residuals
Section 3.1.3 DBFs
Summary statistics in this section are generally for "plant-mean" data—that is, for each plant, the mean
concentration of an analyte was calculated using all reported data during the last 12 months of the ICR
collection period. Summary statistics were then generated based on the distribution of all plant-means.
See section 1.4.8 for a detailed description of the methodology used to generate ICR data summaries in
this section.
3.1.1 DBF Precursors
This section summarizes plant-mean data for water quality parameters that can affect the
formation of DBFs. These water quality parameters are total organic carbon (TOC), temperature,
bromide, and UV254 absorbance. Summary statistics shown in Exhibit 3.1 are calculated using the last 12
months of the ICR collection period for plants that have at least nine months of data for each parameter.
Values below the minimum reporting level (MRL) were converted to zero to calculate plant-means.
Exhibits 3.2 through 3.5 compare the cumulative distributions of plant-mean values for ground and
surface water plants. Exhibit 3.6 characterizes plant-level variability by showing the distribution of the
maximum value minus the minimum value at each plant for each water quality parameter.
All data in this section represent samples collected from the influent water sampling location.
Although the ICR required samples to be collected throughout the treatment plant, the influent water
sampling point was selected to illustrate parameters of the influent water quality matrix that ICR systems
consider during treatment. Although the water quality characteristics that directly affect DBF formation
are present at the point of disinfectant addition, this section focuses on influent characteristics with the
understanding that treatment will change the parameters. Observations regarding the data follow the
exhibits.
Occurrence Assessment for the Final Stage 2 DBPR 3-2 December 2005
-------
Exhibit 3.1 Summary of Influent Water Quality Parameter ICR Data
for All Large Plants
Source
Water
Number of
Plants
Mean of
Plant-Means
Median of
Plant-Means
90th Percentile of
Plant-Means
Range of
Plant-Means3
Total Organic Carbon (mg/L as Carbon [C])
Surface
Ground
All1
307
103
423
3.14
1.46
2.71
2.71
0.19
2.45
5.29
3.36
5.10
0.0-21.4
0.0-16.1
0.0-21.4
Temperature (degrees Celsius)
Surface
Ground
All1
334
121
473
16.0
19.9
17.1
16.1
20.1
17.0
20.7
26.3
24.5
3.7-27.7
9.5-30.5
3.7-30.5
Bromide (mg/L)2
Surface
Ground
All1
320
118
449
0.055
0.103
0.068
0.027
0.066
0.036
0.115
0.190
0.151
0.000- 1.325
0.000-1.325
0.000-1.325
UV-254 Absorbance (cm"1)
Surface
Ground
All1
306
104
424
0.098
0.062
0.091
0.079
0.009
0.069
0.176
0.266
0.180
0.000-0.880
0.000 - 0.606
0.000-0.880
Notes: 1"AN" plants include those with surface, ground, blended, mixed, or purchased source water types, so "AH"
does not equal the sum of surface and ground.
2Plant 402 was removed from the analysis for bromide. Its plant-mean bromide value of 2.36 mg/L was
calculated based on one month of bromide levels of 28 mg/L. All the other values for that plant in the last
12 months of the ICR were below 0.1 mg/L. The 28 mg/L value is most likely a reporting error as
laboratories often report bromide values in ug/L rather than mg/L, and this value may not have been
converted to mg/L.
3Values below the minimum reporting level (MRL) were converted to zero in order to calculate plant-
means.
Source: ICR AUX1 database (USEPA 2000d).
Queries: Screened TOO INF, Screened TEMP INF, Screened BROMIDE INF, and Screened UV_254 INF. See
Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-3
December 2005
-------
Exhibit 3.2 Cumulative Distribution of Plant-Mean TOC Concentrations of
Influent Samples Based on ICR Data for Large Surface and Ground Water Plants
(mg/L as C)
100%
90%
80%
» SW ICR TOC Plant-Means (N=307)
n GW ICR TOC Plant-Means (N=103)
8 10 12 14 16
Plant-Mean TOC (mg/L as C)
18
20
22
24
Source: ICR AUX1 database (USEPA 2000d).
Query: Screened TOC INF. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-4
December 2005
-------
Exhibit 3.3 Cumulative Distribution of Plant-Mean Water Temperature of
Influent Samples Based on ICR Data for Large Surface and Ground Water Plants
(degrees Celsius)
100%
90%
• SW ICR Temperature Plant-Means (N=334)
n GW ICR Temperature Plant-Means (N=121)
10 15 20 25
Plant-Mean Temperature (degrees Celsius)
Source: ICR AUX1 database (USEPA 2000d).
Query: Screened TEMP INF. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-5
December 2005
-------
Exhibit 3.4 Cumulative Distribution of Plant-Mean Bromide Concentrations of
Influent Samples Based on ICR Data for Large Surface and Ground Water Plants
(mg/L)
100%
90%
80%
• SW ICR Bromide Plant-Means (N=320)
a GW ICR Bromide Plant-Means (N=118)
0.2
0.4
0.6 0.8
Plant-Mean Bromide (mg/L)
1.2
1.4
Note: Plant 402 was removed from the analysis for bromide. Its plant mean bromide value of 2.36 mg/L was
calculated based on one month of bromide levels of 28 mg/L. All the other values for that plant in the last
12 months of the ICR were below 0.1 mg/L. The 28 mg/L value is most likely a reporting error as
laboratories often report bromide values in ug/L rather than mg/L, and this value may not have been
converted to mg/L.
Source: ICR AUX1 database (USEPA 2000d).
Query: Screened BROMIDE INF. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-6
December 2005
-------
Exhibit 3.5 Cumulative Distribution of Plant-Mean UV254 Absorbance of
Influent Samples Based on ICR Data for Large Surface and Ground Water Plants
(cm'1)
0%
• SW ICR UV-254 Plant-Means (N=306)
n GW ICR UV-254 Plant-Means (N=104)
0.6
Plant-Mean UV-254 (cm'1
Source: ICR AUX1 database (USEPA 2000d).
Query: Screened UV_254 INF. See Appendix B for query language.
0.7
0.8
0.9
Occurrence Assessment for the Final Stage 2 DBPR 3-7
December 2005
-------
TOC is a measure of the organic carbon content of water. This organic content contributes to the
formation of DBFs. In unpolluted source water, humic and fulvic acids from the decay of vegetation are
the major constituents of TOC; in polluted water, pesticides and other manmade chemicals may be
constituents of TOC as well (Amirtharajah and O'Melia 1990). Researchers have found that TOC can be
a good indicator of the amount of THMs and other DBFs that may form as a result of chemical
disinfection (Singer and Chang 1989). Correlations between TOC and DBFs are presented in section 3.3.
Mean TOC concentrations for ICR influent samples at surface water plants are more than double
the mean influent TOC concentrations in ground water plants. Approximately 42 percent of ground water
plants have mean TOC concentrations less than or equal to 0.1 mg/L as C, whereas less than 1 percent of
surface water plants have mean TOC concentrations less than 0.1 mg/L as C. However, as shown in the
cumulative distribution of TOC concentrations in Exhibit 3.2, TOC concentrations in the upper 95th
percentile are higher for ground water than surface water plants.
Temperature can affect many aspects of water chemistry and treatment. Generally, as
temperature increases so do chemical reaction rates which increase the amount of DBFs formed
(specifically trihalomethanes) and the efficiency of chlorine disinfection. Temperature also affects the
solubility of different substances in water (including calcium carbonate, which can change pH, alkalinity,
and hardness).
Temperature fluctuates much more in surface water than in ground water. Plant-mean
temperature can be lower in surface water than in ground water, since the surface water is directly
exposed to the air and ground water sources are insulated by the ground. The mean of plant-mean
temperature level for surface water plants is 16.0° C, while the mean of plant-mean temperature levels for
ground water plants is 19.9° C.
Bromide can be present as a result of salt water intrusion into an aquifer, human activities such as
pesticide and road salt application, and dissolution of minerals in geologic formations (Siddiqui et al.
1995). The presence of bromide in source water can affect the type and amount of DBFs formed, shifting
the distribution of DBFs generated to the more brominated species (Krasner et al. 1989). In addition,
bromide can react with strong oxidants, such as ozone or chlorine dioxide, to form bromate, another
byproduct of concern.
Bromide concentrations are typically higher in ground water than in surface water sources, in part
because ground water has long contact time with geologic formations that can be sources of bromide.
This is reflected by the ICR results—mean bromide concentration was 0.103 mg/L for ground water
plants compared to 0.055 mg/L for surface water plants. Peak values, however, were identical for surface
and ground water plants, calculated at 1.33 mg/L. Bromide levels can be impacted by seasonal climate
conditions. Bromide levels tend to be higher during drought periods because of the concentration of ions
in a smaller volume of water. Bromide occurrence also varies regionally. Section 3.4 shows analysis of
regional trends for influent bromide. See Chapter 14 of the Information Collection Rule Data Analysis
document (McGuire et al. 2002) for additional information regarding bromide occurrence.
The absorbance of UV radiation at a wavelength of 254 nanometers correlates with the amount of
unsaturated organic compounds, particularly dissolved matter such as humic substances, in the water
(USEPA 1999a). UV254 absorbance can be used as an alternative to measuring TOC or dissolved organic
carbon (DOC) as an indicator of DBP precursors in raw water. Exhibits 3.2 and 3.5 show that the
distributions of TOC and UV254 absorbance for ground and surface water plants follow very similar
trends, with surface water plants generally showing higher plant-mean values (except in the upper 5th
percentile).
Occurrence Assessment for the Final Stage 2 DBPR 3-8 December 2005
-------
Exhibit 3.6 provides statistics for the difference between the highest and lowest monthly values
on a plant-by-plant basis (as reported during the last 12 months of the ICR) for TOC, temperature,
bromide, and UV254. On average, the difference between the highest and lowest values for TOC,
temperature, and UV254 is roughly three to four times greater in plants using surface waters than in plants
using ground waters. The difference between the highest and lowest bromide value is also greater, on
average, for surface water plants compared to ground water plants, but not by as much. These general
trends are consistent across the range of percentiles. These findings are consistent with general
observations that ground water varies less over a year than surface water.
Exhibit 3.6 Cumulative Distribution of Differences Between Highest and Lowest
Monthly Parameter Values for Influent Water Sample Location Based on ICR Data
for All Large Plants
Source
Water
Number of
Plants
Mean
25th
Percentile
50th
Percentile
75th
Percentile
90th
Percentile
95th
Percentile
Total Organic Carbon (mg/L as Carbon [C])
Surface
Ground
All1
307
103
423
2.26
1.01
1.93
0.95
0.00
0.80
1.55
0.80
1.30
2.70
1.30
2.30
4.10
1.95
3.75
5.60
2.75
5.35
Temperature (degrees Celsius)
Surface
Ground
All1
334
121
473
17.0
4.3
13.5
13.2
2.0
6.9
18.0
3.8
14.3
21.0
5.2
19.8
23.0
8.2
22.0
24.5
12.0
24.0
Bromide (mg/L)2
Surface
Ground
All1
320
118
449
0.078
0.060
0.073
0.023
0.024
0.024
0.040
0.043
0.041
0.091
0.082
0.090
0.160
0.110
0.150
0.260
0.210
0.260
UV-254 Absorbance (cm"1)
Surface
Ground
All1
306
104
424
0.121
0.033
0.106
0.034
0.000
0.023
0.070
0.015
0.049
0.134
0.036
0.120
0.302
0.085
0.280
0.448
0.127
0.412
Notes: 1"AN" plants include those with surface, ground, blended, mixed, or purchased source water types, so "AH"
does not equal the sum of surface and ground.
2Plant 402 was removed from the analysis for bromide. Its plant mean bromide value of 2.36 mg/L was
calculated based on one month of bromide levels of 28 mg/L. All the other values for that plant in the last
12 months of the ICR were below 0.1 mg/L. The 28 mg/L value is most likely a reporting error as
laboratories often report bromide values in ug/L rather than mg/L, and this value may not have been
converted to mg/L.
Source: ICR AUX1 database (USEPA 2000d).
Queries: Screened TOC INF, Screened TEMP INF, Screened BROMIDE INF, and Screened UV_254 INF. See
Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-9
December 2005
-------
3.1.2 Disinfectant Residuals
This section summarizes residual concentrations for chlorine, chloramine, and chlorine dioxide in
finished water (taken from the end of the treatment plant, before water enters the distribution system) and
ozone residuals after the last contact chamber (see Chapter 2 for a discussion of disinfectant use and
doses). Disinfectants were monitored monthly, although there are no ground water data for some of the
disinfectant types. Summary statistics, calculated using the last 12 months of the ICR collection period
for plants that have at least nine months of data for each parameter, are shown in Exhibit 3.7. General
observations regarding the data follow Exhibit 3.7.
Exhibit 3.7 Summary of Disinfectant Residual ICR Data for All Large Plants
Source
Data Sample
Location
Number of
Plants
Mean of
Plant-Means
Median of
Plant-Means
90th Percentile of
Plant-Means
Range of
Plant-Means
Free Chlorine Residual (mg/L)1
Surface
Ground
All3
Finished
Finished
Finished
183
33
224
1.23
1.13
1.22
1.14
1.04
1.13
1.98
2.06
1.98
0.00-4.37
0.17-2.61
0.00-4.37
Total Chlorine Residual (mg/L)
Surface -CL21
Ground -CL21
All3 - CL21
Surface -CLM2
Ground -CLM2
All3 - CLM2
Finished
Finished
Finished
Finished
Finished
Finished
187
37
232
89
14
105
1.56
1.45
1.55
2.51
3.07
2.58
1.33
1.17
1.33
2.35
3.23
2.46
2.53
3.25
2.58
3.58
4.57
3.66
0.33-4.58
0.17-3.73
0.17-4.58
1.07-5.19
1.33-4.62
1.07-5.19
Chlorine Dioxide Residual (mg/L)
Surface
Finished
20
0.61
0.21
2.13
0.00-2.74
Ozone Residual (mg/L)
Surface
After Last Contact
Chamber
13
0.08
0.06
0.13
0.01-0.21
Notes:
1 For plants using chlorine only.
2 For plants using chlorine and chloramines or chloramines only.
3 "AN" plants include those with surface, ground, blended, mixed, or purchased source water types.
Source: ICR AUX1 database (USEPA 2000d).
Queries: Screened EXFCLRES FIN, Screened EXTCLRES FIN, Screened EXCLXRES FIN, and Screened
EXO3RES. See Appendix B for query language.
Excel File: ICR Disinfectant Residuals.xls
In water, chlorine exists as hypochlorous acid (HOC1) and hypochlorite (OC1~). Free chlorine is
defined as the sum of the concentrations of HOC1 and OC1" measured as C12. The Surface Water
Treatment Rule (SWTR) sets minimum requirements for residual disinfectant concentration. For
instance, the residual disinfectant concentration at the point of entry to the distribution system may not
drop below 0.2 mg/L for more than four hours, although the regulations do not state that the disinfection
concentration must be measured as free chlorine residual. The rule also sets CT (the product of contact
time and disinfectant concentration) requirements for surface water systems (USEPA 1989a). For CT
Occurrence Assessment for the Final Stage 2 DBPR 3-10
December 2005
-------
calculation, plants must take all contact chambers into account. CT is used to achieve the inactivation of
microorganisms, as required by the SWTR.
The mean of all plant-mean free chlorine residual concentrations were similar for surface water
(1.23 mg/L) and ground water plants (1.13 mg/L), with the mean for all plants of approximately 1.22
mg/L. Although surface water plants exhibit a higher upper range value (4.37 mg/L), ground water plants
have a slightly higher 90th percentile of plant means (2.06 mg/L).
Total chlorine is defined as the sum of free chlorine and combined chlorine (chloramine)
concentrations, and is expressed in mg/L as C12. Total chlorine residuals for surface and ground water
plants that use chloramines are higher, and the difference is statistically significant, than total chlorine
residuals for surface and ground water plants that use only free chlorine. Higher total chlorine residual
concentrations in chloramine systems may be due to organic material and DBF precursors and thus,
higher chlorine demand in those systems. Also, the higher chlorine residuals in the chloraminated
systems are due to the slower decay rate of chloramine compared to free chlorine. Higher chloramine
levels can be maintained with a lower rate of total trihalomethane (TTHM) and HAAS formation.
Only twenty surface water plants reported using chlorine dioxide for at least nine of the last 12
months of the ICR collection period. Although not in effect at the time of the ICR, the Stage 1 DBPR sets
a daily maximum residual disinfectant level (MRDL) of 0.8 mg/L for chlorine dioxide based on sampling
at the entry point to the distribution system (which can be interpreted in most cases to mean finished
water). Average chlorine dioxide residuals range from 0 to 2.74, with four plants (or 20 percent) having
mean residual concentrations greater than 0.8 mg/L.
Ozone (O3) is a colorless gas that is unstable and decomposes rapidly, reacting with hydroxide
ions (OH") to form hydroxyl radicals and organic radicals. (Radicals are unstable molecules with
unpaired electrons.) As part of ICR sampling, plants measured ozone residuals of the effluent of each
ozone contact chamber. For this current analysis, only the ozone residual at the last contact chamber is
presented to show the small potential for DBF formation outside the contact chambers. The plant-mean
ozone residual concentrations in the last ozone contact chamber are very low, with an average of 0.08
mg/L. Averages ranged from 0.01 to 0.21 mg/L for the 13 surface water plants that submitted ICR data.
3.1.3 DBFs
Halogenated organic DBFs form as a result of reactions of free chlorine, bromide, or chloramines
with naturally occurring organic matter. Studies show that some of these DBFs can cause adverse
reproductive and development health effects and some forms of cancer (USEPA 2005a). Inorganic DBFs
are also of concern, and are usually formed during reactions of chlorine dioxide with water and ozone
with bromide.
A description of how distribution system DBF data are aggregated is provided below. Next, this
section summarizes results for all halogenated DBFs measured during the ICR (see Exhibit 1.4 for a full
list of DBFs measured during the ICR). The remainder of the section focuses on regulated DBFs (TTHM,
HAAS, bromate, and chlorite). See section 1.4.8 for a detailed description of the methodology used to
generate DBF results using ICR data.
Chapter 4 builds on this section by providing additional analyses of TTHM and HAAS
occurrence (e.g., spatial and temporal variation in the distribution system (DS)) for only those plants in
compliance with the Stage 1 DBPR.
Occurrence Assessment for the Final Stage 2 DBPR 3-11 December 2005
-------
Aggregation of DBF Data
As explained in section 1.4.1, each ICR plant collected samples from a single finished water
location and from four distribution system sample locations (DSE, AVG1, AVG2, DS Maximum) for the
ICR. DBF data in this section and in Chapter 4 are aggregated into the following data types for analyses:
• Finished Water means a sample taken from the end of the treatment plant, before water enters
the distribution system. The plant-mean finished water concentration is the average of the
last four quarters of finished water data for that plant.
• DS Average (or RAA) is the calculated average of four distribution system samples (DSE,
AVG 1, AVG 2, and DS Maximum). The plant-mean DS Average concentration is the
average of the last four quarters of calculated DS Average data for that plant. The plant-mean
DS Average concentration is equivalent to the running annual average (RAA) concentration
for that year.
• Single Highest is the highest concentration of the four distribution system samples collected
by a plant in the last four quarters of the ICR (16 possible values). This value may represent
any of the four distribution system locations—DSE, AVG1, AVG2, or DS Maximum.
• Locational Running Annual Average (LRAA) is the average of four quarters of data from a
single distribution system location (DSE, AVG1, AVG2, and DS Maximum). For example,
the LRAA for the DSE location would be the average of the last four quarters of data
collected from that location. The highest LRAA is the maximum of the four (DSE, AVG1,
AVG2, and DS Maximum) calculated LRAAs. Since the LRAA covers one year's worth of
data, it already represents a plant-mean value.
Max-Min is the highest concentration of the four distribution system samples collected by a
plant during the last four quarters of the ICR (16 possible values) minus the lowest
concentration of the four distribution system samples reported during the last four quarters of
the ICR. In other words, Max-Min is a single value that represents the difference between the
maximum and minimum concentrations from all four distribution system samples collected
during the last four quarters of the ICR.
3.1.3.1 All Measured Halogenated DBFs
Exhibit 3.8 summarizes DS Average results for all halogenated DBFs measured under the ICR.
As can be seen from the measured concentrations for all plants, TTHMs and HAAS comprise
approximately 50 percent of the measured total organic halides (TOX), whereas the other measured
organic halides (HAN4, CH, CP, DCP, and TCP1) represent approximately 7 percent of the TOX
concentration.
HAN4 stands for Haloacetonitriles (the sum of dichloroacetonitrile, trichloroacetonitrile, bromochloroacetonitrile, and
dibromoacetonitrile). CH stands for chloral hydrate. CP stands for chloropicrin. DCP stands for dichloropropanone. TCP
stands for trichloropropanone.
Occurrence Assessment for the Final Stage 2 DBPR 3-12 December 2005
-------
Exhibit 3.8 Summary of Halogenated DBF Data Measured During the ICR, Single
Highest (Parameter Occurrence Values in ug/L) for All Large Plants
Source
Surface
Ground
All
Parameter
TTHM
HAAS
HAN4
CH
CP
DCP
TCP
TOX
TTHM
HAAS
HAN4
CH
CP
DCP
TCP
TOX
TTHM
HAAS
HAN4
CH
CP
DCP
TCP
TOX
Number of
Plants
213
213
209
208
208
209
209
213
82
82
80
80
79
80
80
81
304
304
305
304
303
305
305
310
Mean of Plant
Means
68.68
24.60
3.79
4.82
0.34
0.54
1.50
144.19
15.36
6.35
2.22
0.59
0.03
0.19
0.07
54.40
25.78
19.40
3.37
3.63
0.25
0.44
1.09
119.40
Median of Plant-
Means
63.90
20.85
3.20
4.36
0.19
0.37
1.23
138.16
6.79
0.33
0.75
0.03
0.00
0.00
0.00
7.88
23.13
16.20
2.69
2.76
0.07
0.20
0.53
115.86
90th Percentile
of Plant -Means
118.70
45.78
7.62
9.91
0.94
1.35
0.00
241.25
36.95
18.83
6.01
2.18
0.13
0.91
0.00
160.00
53.83
41.45
7.32
8.56
0.74
1.24
3.02
237.19
Range of
Plant -Means
0-177
0-104
0-17.5
0-18.7
0-2.4
0-2.8
0-6.4
0-305
0-123
0-97
0-14.8
0-5.5
0-0.6
0-2.0
0-1.2
0-482
0-119
0-104
0-17.5
0-18.7
0-2.4
0-2.8
0-6.4
0-482
Source: ICR AUX1 database (USEPA 2000d).
Queries: Plants min 3x3, RAA - Other DBFs and Plants min 3x3, RAA and Max LRAA - TTHM & HAAS. See
Appendix B for query language.
3.1.3.2
TTHM
Aggregate Data
TTHM measurements are the sum of concentrations of chloroform (CHC13),
bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform (CHBr3). Exhibit 3.9
presents summary statistics of plant-mean (except for the single highest observation) TTHM data
collected under the ICR by source water and data type. (See Chapter 6 of the Information Collection Rule
Data Analysis document [McGuire et al. 2002] and Appendix A for occurrence data on individual TTHM
constituents.) Exhibit 3.10 shows the cumulative distribution of the plant-mean DS Average (or RAA)
TTHM data for ICR surface and ground water plants. Exhibits 3.11 and 3.12 show the cumulative
distributions of plant-mean TTHM single highest and plant-mean highest TTHM LRAA, respectively.
Discussions of the findings follow the exhibits.
Occurrence Assessment for the Final Stage 2 DBPR 3-13
December 2005
-------
Exhibit 3.9 Summary of TTHM (|jg/L) ICR Data for All Large Plants
Source
Surface
Ground
All2
Data Type1
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max- Min3
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max- Min3
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max - Min3
Number of
Plants
213
213
213
213
213
82
82
82
82
82
304
311
311
311
311
Mean of
Plant-Means
31.60
42.28
68.68
49.29
50.01
9.69
15.36
32.32
20.21
26.53
25.78
34.98
58.48
41.38
43.15
Median of
Plant-Means
28.75
40.36
63.90
45.80
44.30
1.48
6.79
18.50
11.80
15.40
23.13
33.16
54.00
39.50
38.40
90th Percentile of
Plant-Means
55.53
69.82
118.70
80.67
91.90
24.75
36.95
74.40
52.63
60.00
53.83
65.88
113.80
78.20
85.20
Range of Plant
Means
0-97
0-117
0-177
0-124
0-129
0-119
0-123
0-300
0-127
0-300
0-119
0-123
0-300
0-127
0-300
Notes: 1 For a description of the data types, see "Aggregation of DBP Data" at the beginning of the subsection.
2 The "AH" plants include those with surface, ground, blended, mixed, or purchased source water types.
Finished water data were not available for blended plants.
3Max-Min is the highest concentration of the four distribution system samples collected by a plant during
the last four quarters of the ICR (16 possible values) minus the lowest concentration of the four
distribution system samples reported during the last four quarters of the ICR.
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, average by finish location - TTHM & HAA5, Plants min 3x3, RAA & Max LRAA - TTHM &
HAAS and Plants min 3x3, Single High - TTHM & HAAS, Plants min 3x3, Max-Min - TTHM & HAAS. See
Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-14
December 2005
-------
Exhibit 3.10 Cumulative Distribution of Plant-Mean DS Average (RAA)
for ICR TTHM Occurrence Data for All Large Plants
100%
80%
60%
m
0.
HI
>
1
3
E
40%
20%
• Surface Water TTHM (N=213)
n Ground Water TTHM (N=82)
100
120
140
Plant-Mean TTHM (ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-15
December 2005
-------
Exhibit 3.11 Cumulative Distribution of Single Highest ICR TTHM Occurrence
Data for All Large Surface and Ground Water Plants
100%
80%
60%
40%
20%
• Surface Water TTHM (N=213)
n Ground Water TTHM (N=82)
100
150 200
Single Highest TTHM (ug/L)
250
300
350
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-16
December 2005
-------
Exhibit 3.12 Cumulative Distribution of Highest LRAA for ICR TTHM Occurrence
Data for Large Surface and Ground Water Plants
100%
80%
= 60%
• Surface Water TTHM (N=213)
D Ground Water TTHM (N=82)
40%
20%
100
120
140
Plant-Mean TTHM (ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
TTHM has been regulated by EPA since the interim TTHM Rule, promulgated in 19792. The
TTHM Rule established an MCL of 100 (ig/L, calculated as an RAA of distribution system TTHM data
measured at four locations, collected quarterly. As explained at the beginning of this section, an RAA is
the average of the most recent four quarters of data; when data from a new quarter is obtained it replaces
the data from the oldest quarter in the four quarter averaging. The Stage 1 DBPR sets an MCL for TTHM
of 80 (ig/L, calculated as an RAA. However, compliance with the Stage 1 DBPR was not required until
2002 for large surface water systems and 2004 for ground water and small surface water systems (USEPA
1998a). Thus, TTHM ICR data, collected in 1997 and 1998 and presented in this section, represents pre-
Stage 1 DBPR conditions.
Plant-mean DS Average (or RAA) TTHM data can be used to estimate the percentage of plants
that may have exceeded the Stage 1 MCLs at the time of the ICR. Although the 90th percentile RAA
See Chapter 4 of the Information Collection Rule Data Analysis document (McGuire et al. 2002) for a
thorough analysis of historical TTHM occurrence in large systems since the mid-to late- 1970's.
Occurrence Assessment for the Final Stage 2 DBPR 3-17 December 2005
-------
concentrations for ground and surface plants are less than the Stage 1 DBPR MCL of 80 i-ig/L, the
maximum plant-mean RAA concentrations are higher than 80 i-ig/L for both plant types. From the
cumulative distributions of TTHM RAA data (Exhibit 3.10), the following information can be derived:
For surface water plants, approximately 4 percent (8 plants) had TTHM RAA levels above 80
I-ig/L; however, 13 percent (28 plants) had TTHM RAA levels above 64 i-ig/L (20 percent less
than the Stage 1 DBPR MCL). The 64 i-ig/L level represents the safety margin occurrence
level that utilities may try to achieve to avoid noncompliance.
• For ground water plants, approximately 2 percent (2 plants) had TTHM RAA levels above 80
I-ig/L; however, 4 percent (3 plants) had TTHM RAA levels greater than 64 i-ig/L.
It is important to note that ICR sampling locations may not be the locations that will be used for
compliance with the Stage 1 DBPR nor were they the locations used for compliance with the 1979 TTHM
rule. Also, because compliance is based on a RAA for the water system rather than the plant, it is
possible for a plant to report TTHM data that is above the Stage 1 DBPR MCLs but for the system to still
be in compliance with this regulation.
The Single Highest and Highest LRAA TTHM values in Exhibit 3.9 indicate that concentrations
at some locations in the distribution system are much higher than DS Average concentrations. Many of
these high values may not be reduced through compliance with the Stage 1 DBPR as high values are
averaged over the whole system. The Stage 2 EA provides additional analyses of peak TTHM data as it
relates to the Stage 1 DBPR and proposed Stage 2 DBPR.
Spatial and Temporal Variability
DBP sampling occurs throughout the distribution system, with particular attention on the finished
water entry point (where the water enters the distribution system), average residence time points, and the
maximum residence time point (where the water is typically the oldest). The location with the highest
DBP levels can move throughout the distribution system due to distribution system hydraulics (e.g.,
changes in flow during peak hours, dead ends stagnating water). In Exhibit 3.13, the ICR data was
evaluated to see with what frequency this is the case.
Occurrence Assessment for the Final Stage 2 DBPR 3-18 December 2005
-------
Exhibit 3.13 Location of Highest TTHM LRAA for ICR Occurrence Data for Large
Surface and Ground Water Plants
60%
50%
40% --
30%
m
20%
10%
DSE
AVG1
Distribution System Location
AVG2
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
While the location of the highest TTHM LRAA occurs most often at the location with the
maximum residence time, it is only that case in roughly 50 percent for surface water systems and roughly
40 percent for ground water systems. The difference between surface water and ground water plants is
due to the more consistent water quality in ground water systems, and possibly the difference in treatment
technologies employed at the different plants.
Exhibit 3.14 compares the location of the highest TTHM levels for chlorine and chloramine for
surface water plants, and Exhibit 3.15 compares the location of the highest TTHM levels for chlorine and
chloramine for ground water plants. For surface water plants, high TTHM values are more likely to occur
at the MAX location for plants using free chlorine than plants using chloramination. Since chloramines
are more stable throughout the distribution systems, their highest locations are more likely to change.
The difference is lower for ground water plants, most likely due more stable influent water quality.
Occurrence Assessment for the Final Stage 2 DBPR 3-19
December 2005
-------
Exhibit 3.14 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant
Disinfectant Type for Large Surface Water Plants
70% i
60%
D Chlorine (N=133)
• Chloramine (N=80)
FIN
DSE AVG1 AVG2
Distribution System Location
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-20
December 2005
-------
Exhibit 3.15 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant
Disinfectant Type for Large Ground Water Plants
50% i
45%
40%
35%
30%
25%
20%
15%
10%
FIN
D Chlorine (N=65)
• Chloramine (N=16)
DSE AVG1 AVG2
Distribution System Location
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-21
December 2005
-------
3.1.3.3 HAAS
HAAS measurements represent the sum of concentrations of monochloroacetic acid (MCAA),
dichloroacetic acid (DCAA), trichloroacetic acid (TCAA), monobromoacetic acid (MBAA), and
dibromoacetic acid (DBAA). Exhibit 3.16 presents summary statistics of plant-mean (except for the
single highest observation) HAAS data collected under the ICR by source water and data type. (See
Chapter 6 of the Information Collection Rule Data Analysis document [McGuire et al. 2002] and
Appendix A for occurrence data on individual HAAS constituents.) Exhibit 3.17 shows the cumulative
distribution of the plant-mean DS Average (or RAA) for HAAS data for ICR surface and ground water
plants. Exhibits 3.18 and 3.19 show the cumulative distributions of plant-mean HAAS single highest and
plant-mean highest HAAS LRAA, respectively. Observations follow the exhibits.
Exhibit 3.16 Summary of HAAS ICR Data for All Large Plants (ug/L)
Source
Surface
Ground
All2
Data Type1
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max - Min3
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max - Min3
Finished Water
DS Average (or RAA)
Single Highest
Highest LRAA
Max - Min3
Number of
Plants
213
213
213
213
213
82
82
82
82
82
304
311
311
311
311
Mean of
Plant-Means
24.60
29.07
47.77
33.66
34.85
6.35
8.45
17.79
11.13
14.68
19.40
22.98
38.66
26.93
28.48
Median of
Plant-Means
20.85
24.38
40.00
28.30
28.20
0.33
2.24
6.30
3.80
6.20
16.20
19.11
31.40
22.53
23.30
90th Percentile of
Plant-Means
45.78
52.31
86.00
58.37
68.00
18.83
21.53
46.30
30.43
43.60
41.45
47.14
75.30
55.73
59.00
Range of Plant
Means
0-104
0-116
0-189
0-124
0-150
0-97
0-71
0-124
0-93
0-94
0-104
0-116
0-189
0-124
0-150
Notes: 1 For a description of the data types, see "Aggregation of DBP Data" at the beginning of the subsection.
2 The "AH" plants include those with surface, ground, blended, mixed, or purchased source water types.
Finished water data were not available for blended plants.
3Max-Min is the highest concentration of the four distribution system samples collected by a plant during
the last four quarters of the ICR (16 possible values) minus the lowest concentration of the four
distribution system samples reported during the last four quarters of the ICR.
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, average by finish location - TTHM & HAAS, Plants min 3x3, RAA & Max LRAA - TTHM
& HAAS and Plants min 3x3, Single High - TTHM & HAAS, Plants min 3x3, Max-Min - TTHM & HAAS.
See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-22
December 2005
-------
Exhibit 3.17 Cumulative Distribution of Plant-Mean DS Average (RAA)
for ICR HAAS Occurrence Data for Large Surface and Ground Water Plants
100%
80%
»«" •
m
0.
HI
>
1
3
60%
40%
20%
• Surface Water HAAS (N=213)
n Ground Water HAAS (N=82)
20
40
60 80
Plant-Mean HAAS (ug/L)
100
120
140
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-23
December 2005
-------
Exhibit 3.18 Cumulative Distribution of Single Highest ICR HAAS Occurrence
Data for Large Surface and Ground Water Plants
100%
80%
60%
* **
40%
20%
• Surface Water HAAS (N=213)
n Ground Water HAAS (N=82)
40
60
80 100 120
Single Highest HAAS (ug/L)
140 160 180 200
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-24
December 2005
-------
Exhibit 3.19 Cumulative Distribution of Highest LRAA ICR HAAS Occurrence Data
for Large Surface and Ground Water Plants
100%
80%
60%
m
0.
HI
>
1
3
Q « « «
/"
» Surface Water HAAS (N=213)
n Ground Water HAAS (N=82)
40%
20%
20
40
60 80
Plant-Mean HAAS (ug/L)
100
120
140
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
The Stage 1 DBPR sets an HAAS MCL of 60 (ig/L, with the HAAS regulated value calculated as
the RAA of distribution system data measured at four locations per plant, collected quarterly. As noted
previously, compliance with the Stage 1 DBPR was not required until 2002 for large surface water
systems and 2004 for ground water and small surface water systems. Thus, HAAS ICR data in this
section represent pre-Stage 1 conditions. Further, unlike the case of TTHMs, HAAs were not regulated at
all prior to the Stage 1 DBPR.
Plant-mean DS Average (or RAA) HAAS data can be used to estimate the percent of plants that
may have exceeded the Stage 1 MCLs at the time of the ICR. Although the 90th percentile RAA
concentrations for ground and surface water plants are less than the Stage 1 DBPR MCL of 60 |ig/L, the
maximum plant-mean RAA concentrations are higher than 60 |ig/L for both plant types. From the
cumulative distributions of HAAS RAA data (Exhibit 3.17), the following information can be derived:
• For surface water plants, approximately 6 percent (12 plants) had HAAS RAA levels above
60 j-ig/L; however, 12 percent (26 plants) had HAAS RAA levels greater than 48 |ig/L (20
percent less than the Stage 1 DBPR MCL). The 48 |ig/L level represents the safety margin
occurrence level that utilities may try to achieve to avoid noncompliance.
For ground water plants, approximately 2 percent (2 plants) had HAAS RAA levels above 60
l-ig/L; however, 4 percent (3 plants) had HAAS RAA levels greater than 48 |ig/L.
Occurrence Assessment for the Final Stage 2 DBPR 3-25
December 2005
-------
It is important to note that ICR sampling locations may not be the locations used for compliance with the
Stage 1 DBPR. Also, because compliance is based on a RAA for the water system rather than the plant, it
is possible for a plant to report HAAS data that is above the Stage 1 DBPR MCLs but still be in
compliance with current regulations.
The Single Highest and Highest LRAA HAAS values in Exhibit 3.16 indicate that concentrations
at some locations in the distribution system are much higher than DS Average concentrations. Many of
these high values may not be reduced by the Stage 1 DBPR. Chapter 4 provides additional analysis of
peak HAAS data as it relates to the Stage 1 DBPR and proposed Stage 2 DBPR.
Spatial and Temporal Variability
DBP sampling occurs throughout the distribution system, with particular attention on the finished
water entry point (where the water enters the distribution system), average residence time points, and the
maximum residence time point (where the water is typically the oldest). The location with the highest
DBP levels can move throughout the distribution system due to distribution system hydraulics (i.e.,
changes in flow during peak hours, dead ends stagnating water). In Exhibit 3.20, the ICR data was
evaluated to see with what frequency this is the case.
Occurrence Assessment for the Final Stage 2 DBPR 3-26 December 2005
-------
Exhibit 3.20 Location of Highest HAAS LRAA for ICR Occurrence Data for Large
Surface and Ground Water Plants
45%
40%
35%
30%
0)
>25%
20%
15%
10%
0%
FIN
DSE AVG1
Distribution System Location
AVG2
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
While the location of the highest HAAS LRAA occurs most often at the location with the
maximum residence time, it occurs at just over 40 percent for surface water systems and just over 25
percent for ground water systems. The difference between surface water and ground water plants is due
to the more consistent water quality in ground water systems, and possibly the difference in treatment
technologies employed at the different plants. Unlike TTHM, HAAS are less likely to form within the
distribution system and degrade more rapidly, so the highest HAAS points are more likely to be found at
locations that are not the maximum residence time.
Exhibit 3.21 compares the location of the highest HAAS levels for chlorine and chloramine for
surface water plants, and Exhibit 3.22 compares the location of the highest HAAS levels for chlorine and
chloramine for ground water plants. For surface water plants, high HAAS values are more likely to occur
at the MAX location for plants using free chlorine than plants using chloramination. Since chloramines
are more stable throughout the distribution systems, their highest locations are more likely to change.
The difference is much larger for surface water plants than for ground water plants.
Occurrence Assessment for the Final Stage 2 DBPR 3-27
December 2005
-------
Exhibit 3.21 Location of Highest HAAS LRAA for ICR Occurrence Data by Plant
Disinfectant Type for Large Surface Water Plants
50%
45%
40%
35%
0%
D Chlorine (N=133)
• Chloramine (N=80)
DSE
AVG1
Distribution System Location
AVG2
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-28
December 2005
-------
Exhibit 3.22 Location of Highest HAAS LRAA for ICR Occurrence Data by Plant
Disinfectant Type for Large Ground Water Plants
35%
30%
D Chlorine (N=65)
• Chloramine (N=16)
DSE
AVG1
Distribution System Location
AVG2
MAX
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-29
December 2005
-------
3.1.3.4 Bromate
Prior to Stage 1, bromate was not a regulated DBF. Therefore, little to no data exists for bromate
occurrence. ICR requirements for bromate monitoring pertained to plants that use oxygenated
disinfectants—ozone or chlorine dioxide. Bromate forms when these disinfectants react with bromide,
which is commonly found in many source waters (see Exhibit 3.1 for source water bromide
concentrations). Bromate can also occur as an impurity in hypochlorite solutions. Because bromide
reacts immediately with ozone and bromate formation does not increase with residence time in the
absence of a residual, monthly monitoring was required at the finished water sampling point but not in the
distribution system. However, bromate formation does increase with contact time if there is a residual
present.
Split samples for bromate were collected during the ICR: one set was analyzed by plant
laboratory personnel or by EPA-certified contract laboratories, and one was analyzed by EPA. EPA's
laboratory used a different laboratory analytical method and was able to detect bromate at much lower
levels than most utility laboratories. The MRL for the utility method is 5.0 (ig/L, while the MRL for the
EPA method is 0.20 (ig/L. Plant-mean bromate data are summarized in Exhibit 3.23. Ground water plant
data were not included in this analysis—no ground water plants used chlorine dioxide, and only one used
ozone.
For surface water plants using chlorine dioxide disinfection, approximately 47 percent of plant-
mean finished water bromate results were less than the MRL based on the EPA method, and 88 percent
were less than the MRL based on the utility method. Bromate concentrations for plants using ozone are
much higher than for plants using chlorine dioxide. Plant-mean finished water concentrations were as
high as 7.2 (ig/L based on the EPA method and 6.4 (ig/L based on the plant laboratory method. It is
difficult to compare values obtained by the EPA and plant laboratory methods. Because the MRL for the
utility method is so high, most individual values were below the MRL of 5.0 (ig/L and thus were assigned
a value of zero, affecting the calculation of the medians and means. For plants treating with chlorine
dioxide, the median of the EPA method data was 0.02 (ig/L, while the median of the plant laboratory data
was 0 (ig/L. For plants using ozone, the mean, median, and 90th percentile plant-mean bromate
concentrations were higher based on the EPA method versus the plant laboratory method.
Exhibit 3.23 Summary of Bromate in Finished Water, Plant-Mean ICR Data for All
Large Plants (|jg/L)
Data Type
Number of
Plants
Mean of
Plant-Means
Median of
Plant-Means
90th Percentile
of Plant-Means
Range of
Plant-Means
Chlorine Dioxide Plants
EPA Analytical Method
Plant Laboratory Analytical Method
19
16
0.06
0.09
0.02
0.0
0.10
0.64
0-0.7
0-0.8
Ozone Plants
EPA Analytical Method
Plant Laboratory Analytical Method
16
14
2.42
1.75
2.2
0.0
5.64
5.09
0-7.2
0-6.4
Note: EPA laboratory analytical method has an MRL of 0.02 ug/L and the plant laboratory analytical method
has an MRL of 5.0 ug/L These different MRLs greatly affect plant-mean bromate calculations.
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Screened BROMATE EPA FIN and Screened BROMATE UTIL FIN. See Appendix B for query
language.
Occurrence Assessment for the Final Stage 2 DBPR 3-30
December 2005
-------
3.1.3.5 Chlorite
ICR requirements for chlorite monitoring pertained only to plants that use chlorine dioxide for
disinfection. Monthly monitoring was required at the finished water sampling location and at three rather
than four locations in the distribution system. The three required monitoring locations were: (1) a
location near the first customer; (2) a location with average residence time (AVG1), and; (3) the location
with maximum residence time (DS Max).
Exhibit 3.24 summarizes plant-mean chlorite data. All data is for surface water systems (there
were no ICR ground water systems that used chlorine dioxide). Different plant-mean data types are
displayed to reflect the Stage 1 DBPR compliance calculations: (1) Plant-mean finished water chlorite
concentrations reported by a plant; (2) Maximum of monthly finished water chlorite concentrations
reported by a plant; (3) Plant-mean DS Average concentration (DS Average for chlorite is the average of
data from the three distribution system sample locations described above) for a plant; (4) Maximum of
monthly calculated DS Average concentration for a plant, and; (5) Single Highest concentration reported
in one year in the distribution system for a plant.
The Stage 1 DBPR requires daily monitoring for chlorite at the finished water location and
monthly monitoring at three locations in the distribution system. Under that rule, if a single daily sample
at the finished water location exceeds 1,000 ug/L, additional monitoring (outside the monthly monitoring
requirement) at the three distribution system locations is then required. The MCL for chlorite is 1.0 mg/L
(1,000 ug/L), based on the average of the three distribution system locations. The maximum of monthly
finished water chlorite concentrations ranged from 0 to 1,719 ug/L. Approximately 78 percent of
maximum finished water samples are below 1,000 ug/L.
Exhibit 3.24 Summary of Chlorite ICR Data (ug/L) for Large Surface Water Plants
Data Type
Finished VAfeter, Plant-Mean
Finished VAfeter, Maximum Plant Month
DS Average, Plant-Mean
DS Average, Maximum Plant Month
Single Highest
Number of
Plants
18
18
16
16
16
Mean of
Plant-Means
432
720
345
572
645
Median of
Plant-Means
461
690
409
653
700
90th Percentile of
Plant-Means
768
1,300
645
871
886
Range of
Plant-Means
2-1,105
20-1,719
5-650
20-1,100
41-1,200
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Screened Chlorite DSAVG, Screened CHLORITE FIN, and Screened Chlorite Single High. See
Appendix B for query language.
3.2 Medium and Small Systems
As discussed in Chapter 2, an estimated 12,224 surface water systems and 39,408 ground water
systems (51,632 systems total) use disinfection and are subject to the Stage 1 and Stage 2 DBPRs.
Although less than 1 percent of disinfecting systems fall into the large system size category (serving more
than 100,000 people), they serve 55.1 percent of the total population served by disinfecting systems (see
Exhibit 2.3). This is one reason that the ICR data collection effort focused on large plants. However,
because roughly 45 percent of the population served by disinfecting systems obtains water from small and
medium systems, it is also important to characterize DBP occurrence in drinking water provided by these
systems.
Occurrence Assessment for the Final Stage 2 DBPR 3-31
December 2005
-------
There is no extensive, focused database similar to the ICR that provides information on DBF
occurrence in small and medium systems. Consequently, it is necessary to use more limited and disparate
sets of occurrence data, together with inferences drawn from the ICR data on large plants, to characterize
DBF occurrence in medium and small systems.
This section presents available information on DBF occurrence, the occurrence of DBF
precursors (e.g., TOC, bromide), and operational characteristics for small and medium systems in order to
compare them to large system ICR data. One important factor to note when considering the possible
similarities and differences in DBF levels among small, medium, and large systems is that the 1979
interim standards for TTHMs do not apply to systems serving fewer than 10,000 people. Some States do
have DBF standards in place for small systems, but it is expected that nationally, a larger percentage of
small systems will have higher DBF levels than large systems, due to the absence of that regulatory
"driver." Similarly, it is expected that DBF levels in medium systems (serving 10,000 to 100,000 people)
will be closer to those in large systems than the levels in small systems will, because these systems are
currently regulated under the 1979 TTHM Rule.
3.2.1 Overview of Available Data for Medium and Small Systems
In addition to the ICR data on large plants, which can be used to draw inferences about small and
medium systems, several data sets provide information specifically useful for evaluating small and
medium systems. Chapter 1, section 1.5 describes each data set in full. A summary of each is provided
below.
• ICR Supplemental Survey (ICRSS). The ICRSS, conducted by EPA from March 1999
through February 2000, was designed to provide information to supplement the ICR data
collection effort for microbiological and byproduct occurrence data. The ICRSS involved 40
randomly selected surface water systems in each of the small, medium, and large system size
categories, as well as seven very large systems. The ICRSS did not collect DBF occurrence
data, but did collect information on byproduct precursors in influent source waters, notably
TOC and bromide levels.
The National Rural Water Association (NRWA) Survey. Developed in cooperation with
EPA, the NRWA Survey was designed to obtain relevant treatment, influent water quality,
and byproduct occurrence information for a random sample of 117 small surface water
systems (serving fewer than 10,000 people). The survey collected water quality and
byproduct data during a cold weather period (November 1999 to March 2000) and a warm
weather period (July 2000 to November 2000).
DBF samples were collected at a finished water location, a distribution system site with
average residence time, and a distribution system site with maximum residence time. For
small system DBF analyses presented in Section 3.2.2.2, samples at the average residence
time location are given a weight three times that of data at the maximum residence location to
produce a "DS Weighted Avg" result. The weighted average was used to make NRWA data
comparable to ICR DS Average (or RAA) data, which is calculated by averaging data at four
locations approximating the average and maximum residence time locations.
• Water Utility Database (WATERASTATSV Published by the American Water Works
Association (AWWA), WATERV STATS is derived from the AWWA Water Industry
Database resulting from a 1996 survey of approximately 900 water utilities, mostly entities
Occurrence Assessment for the Final Stage 2 DBPR 3-32 December 2005
-------
serving at least 10,000 people. The WATER:\STATS data used here are aimed mainly at
characterizing relevant treatment and byproduct information for medium surface water plants.
Although 900 systems participated in the 1996 survey, the relevant table in WATER:\STATS
contains data only from those systems that chose to respond to the section on water quality.
WATER:\STATS does not contain data on individual samples; it contains averages, minima,
and maxima for each parameter for each plant.
The Ground Water Supply Survey (GWSS). This survey, conducted by EPA in 1981-82,
was designed to collect treatment, influent water quality, and finished water contaminant
occurrence information on 979 small, medium, and large ground water systems from across
the United States. Although TTHM data from this survey are available, they are probably not
representative of current TTHM levels for large and medium systems because they were
collected more than 20 years ago, prior to the implementation of the 1979 TTHM standard.
Due to the rolling implementation schedule of the TTHM Rule, systems may or may not have
been in compliance with the rule in 1981 and 1982. In addition, the TTHM data were
collected only at the entry point to the distribution system, not from the distribution system
itself.
State Data. Data from several States were used to gain insights into the occurrence of DBFs
and DBF precursors in surface water and ground water. For surface water, the data were
available from eight States: Alaska, California, Illinois, Minnesota, Missouri, North
Carolina, Texas, and Washington. The data from these States represent 562 small surface
water systems. While the systems in these data sets were not randomly selected, they include
at least 50 percent of the small systems in each State. Also, all the small surface water
systems in these eight States together account for approximately one-third of all small non-
purchased surface water systems in the United States, which is a significant sample. There
were also some ground water data on DBFs available from seven States: Alaska, California,
Florida, Illinois, North Carolina, Texas, and Washington.
The data available from each State are not exactly comparable; some States reported
individual sample data, while others reported only plant averages. Some of the data appear to
be from distribution system locations, while other samples are from the plant or from raw
water. Samples in some States were collected quarterly, while in others, the time between
samples at some plants was anywhere from two months to more than a year.
3.2.2 Surface Water Systems
DBF precursor occurrence data for medium and small surface water systems from the sources
described in Section 3.2.1 are summarized in Exhibits 3.25 and 3.26. Exhibit 3.25 shows plant-mean
data, while Exhibit 3.26 shows individual observations for the plants included in Exhibit 3.25. NRWA
data were included only if both summer and winter data were available for a plant. ICRSS data were
included only for plants that had data for at least three-fourths of the total possible number of samples.
Detailed discussion of medium and small system data are provided in the next two subsections.
Occurrence Assessment for the Final Stage 2 DBPR 3-33 December 2005
-------
Exhibit 3.25 Summary
Surface
of Non-ICR DBF Precursor Data for Medium and Small
and Ground Water Plants, Plant-Means
System Size
&Type
Source of Data
Number of
Plants
Mean of
Plant-Means
Median of
Plant-Means
90th Percent! le
of Plant-Means
Range of
Plant-Means
Source Water TOC (rrg/L as C)
Small
Surface Water
Medium
Surface Water
Medium
Ground Water
NRWA
ICRSS
ICRSS
WATERASTATS
WATERASTATS
96
38
40
102
51
3.0
2.4
3.6
5.6
2.3
2.6
2.1
3.7
3.2
0.8
5.4
4.5
5.5
6.4
7.0
0.3-9.0
0.1-7.1
0.2-7.9
0.0-200
0.0-25
Source Water Bromide (mg/L)
Small
Surface Water
Medium
Surface Water
NRWA
ICRSS
ICRSS
95
38
40
0.063
0.020
0.050
0.021
0.000
0.016
0.108
0.044
0.092
0-1.724
0.000-0.274
0.000-0.534
Source Water UV-254 (cm1)
Small
Surface Water
Medium
Surface Water
NRWA
ICRSS
ICRSS
96
38
40
0.082
0.074
0.093
0.074
0.051
0.083
0.127
0.113
0.171
0.012-0.228
0.016-0.444
0.029-0.208
Notes: Small systems are those that serve fewer than 10,000 people; medium systems serve between 10,000 and
100,000 people. See text in Section 3.2.1 fora description of "Source of Data."
Sources: USEPA2001g; USEPA 2000k; AWWA2000.
Occurrence Assessment for the Final Stage 2 DBPR 3-34
December 2005
-------
Exhibit 3.26 Summary of Non-ICR DBF Precursor Data for Medium and
Small Surface and Ground Water Plants, Individual Observations
System Size
&Type
Source of Data
Number of
Observations
Mean
Median
90th
Percentile
Range
Source Water TOC (mg/L as C)
Small
Surface Water
Medium
Surface Water
NRWA
ICRSS
ICRSS
192
384
478
3.0
2.4
3.6
2.6
1.8
3.2
5.5
5.7
7.0
0.3-9.9
0.0-17.0
0.0-21.6
Source Water Bromide (mg/L)
Small
Surface Water
Medium
Surface Water
NRWA
ICRSS
ICRSS
190
384
473
0.063
0.020
0.050
0.019
0.000
0.014
0.114
0.056
0.116
0-1.862
0-0.355
0-0.865
Source Water UV-254 (cm'1)
Small
Surface Water
Medium
Surface Water
NRWA
ICRSS
ICRSS
192
380
467
0.082
0.074
0.1
0.070
0.053
0.1
0.150
0.118
0.2
0-0.350
0.004-0.676
0-0.805
Notes: Small systems are those that serve fewer than 10,000 people; medium systems serve between 10,000
and 100,000 people. See text in Section 3.2.1 fora description of "Source of Data."
Sources: USEPA 2001 g; USEPA 2000k.
3.2.2.1 Medium Surface Water Systems
The main purpose of this section is to evaluate medium surface water system DBF occurrence and
water quality data and determine if these parameters in medium surface water systems are similar to those
in large surface water systems. The data in the WATERASTATS (AWWA 2000) and ICRSS (USEPA
2000k) data were primarily used for this purpose. All WATERASTATS data in this section represent
plant-average values.
WATERASTATS occurrence data shows that source water types and quality in medium and large
surface water systems are similar on a national level. Exhibit 3.27 indicates that medium and large
surface water systems use very similar types of water sources. Exhibits 3.28 and 3.29 compare TOC data
for different system sizes using WATERASTATS and ICRSS data, respectively. These graphs show
similar distributions of TOC occurrence in large and medium surface water systems. TOC occurrence can
also be assessed by comparing medium system TOC data in Exhibit 3.25 to large system TOC data in
Exhibit 3.1. WATERASTATS and ICRSS values are similar to ICR TOC data, with median values of 3.2
mg/L, 3.7 mg/L, and 2.7 mg/L, respectively. ICRSS data on bromide and UV254 levels, shown in Exhibit
3.25, are quite close to ICR plant levels (see Exhibit 3.1). Exhibits 3.30 and 3.31 show that medium and
large systems have similar distributions of other parameters affecting treatability and, indirectly, DBF
formation, such as turbidity and alkalinity.
The type of treatment technologies used by medium surface water systems is also similar to those
used by large systems. As shown in Exhibits 3.32 through 3.34, medium and large systems are similar
with respect to major categories of treatment (conventional vs. others), the use of key physical unit
processes, and the use of specific disinfection methods among conventional plants. One reason that
medium and large plants are similar is that both have historically been subject to the same regulatory
requirements.
Occurrence Assessment for the Final Stage 2 DBPR 3-35
December 2005
-------
Exhibits 3.35 and 3.36 compare cumulative distributions of annual average TTHM levels in
finished water and in distribution system water from WATER:\STATS for medium and large surface
water systems, and confirm that the distributions are similar. Also, these cumulative distributions are
consistent with TTHM values reported for large ICR plants earlier in this chapter for DS Averages (where
the median plant-mean TTHM value is 41 (ig/L).
Exhibit 3.27 Percentages of Medium and Large Surface Water Systems Using
Different Source Water Types
60% i
50%
40%
5x
tn
•5 30%
-------
Exhibit 3.28 Comparison of Source Water TOC for Medium and Large Surface
Water Systems
100%
• Medium Surface Water Systems (N=102)
A Large Surface Water Systems (N=196)
10 15 20
Plant-Mean TOC (mg/L as C)
25
30
Source: WATERASTATS (AWWA2000).
Exhibit 3.29 Comparison of Source Water TOC for Small, Medium, and Large
Surface Water Systems
100% -
90% -
80% -
v 70% -
§ 60%-
1
g 50% -
ra
1 40% -
3
° 30% -
20% -
10% -
no/
• A
*»*"
i V* /
/A !'••""
t t
AA » Large Syst(
A ** • • Medium Sy
± f m ASmallSystf
A £ •
b //
i ,. •: /
A » •
;ms (N=47)
stems (N=40)
>ms (N=38)
At^"
3456
Plant-Mean TOC (mg/L as C)
Source: ICRSS (USEPA 2000k).
Occurrence Assessment for the Final Stage 2 DBPR 3-37
December 2005
-------
Exhibit 3.30 Comparison of Source Water Turbidity For Medium and Large
Surface Water Systems
• Medium Surface Water Systems (N = 243)
A Large Surface Water Systems (N = 240)
100 150 200
Plant-Mean Turbidity (NTU)
250
300
Source: WATERASTATS (AWWA2000).
Exhibit 3.31 Comparison of Source Water Alkalinity for Medium and Large
Surface Water Systems
100% -f
90%
80%
• Medium Surface Water Systems (N=224)
A Large Surface Water Systems (N=234)
50
100 150 200 250
Plant-Mean Alkalinity (mg CaCO3/L)
300
350
Source: WATERASTATS (AWWA2000).
Occurrence Assessment for the Final Stage 2 DBPR 3-38
December 2005
-------
Exhibit 3.32 Comparison of Treatment-ln-Place for Medium and Large Surface
Water Systems
D Medium Surface Water Systems (N = 265)
D Large Surface Water Systems (N = 255)
Conventional Softening
Source: WATERASTATS (AWWA2000).
Direct Filtration
Unfiltered
Other
Exhibit 3.33 Comparison of Physical Unit Processes for Medium and Large
Surface Water Systems
100%
90%
80%
70%
i
•2 60% H
OT
•5 50% -
| 40% -
-------
Exhibit 3.34 Comparison of Disinfectant Type for Medium and Large Surface
Water Systems Using Conventional Filtration
100%
90%
80%
70%
S 60%
• Medium Surface Water Systems (N = 266)
D Large Surface Water Systems (N = 249)
PreCI2/ Post CI2 Only PreCI2/
PostCI2 PostNH2CI
W/O3
Disinfectant Type
Source: WATERASTATS (AWWA2000).
W/CIO2
Other
Exhibit 3.35 Comparison of Finished Water Annual Average TTHM for Medium
and Large Surface Water Systems
• Medium Surface Water Systems (N=211)
A Large Surface Water Systems (N=210)
0%
60
TTHM (ug/L)
Source: WATERASTATS (AWWA2000).
Occurrence Assessment for the Final Stage 2 DBPR 3-40
December 2005
-------
Exhibit 3.36 Comparison of Distribution System TTHM Data for Medium and
Large Surface Water Systems
» Medium Surface Water Systems (N=207)
A Large Surface Water Systems (N=131)
40
60
TTHM (ug/L)
80
100
120
Source: WATERASTATS (AWWA2000).
3.2.2.2 Small Surface Water Systems
The key sources of information for small surface water systems are the ICRSS, the NRWA
Survey, and some State data. DBF Precursor data from these sources are summarized in Exhibits 3.25
and 3.26.
Exhibits 3.25 and 3.29 show that the distribution of TOC levels for small surface water plants
differs somewhat from that for medium and large plants. Generally, small plants appear to have lower
TOC levels (i.e., lower levels of byproduct precursors) than do the medium and large plants. However, at
the upper end of the TOC distributions, above approximately the 90th percentile, small plant TOC levels
are very similar to those of medium and large plants (see Exhibit 3.29).
Exhibits 3.37 through 3.40 provide the cumulative distributions of influent TOC, bromide,
alkalinity, and temperature measurements from the NRWA Survey for winter and summer monitoring
periods. Seasonal variability in TOC, bromide, and alkalinity appear low, although temperature was
markedly different between winter and summer months (as expected). The TOC distribution in Exhibit
3.37 is similar to but slightly higher than that for small systems in the ICRSS (see Exhibit 3.25 for
summary statistics on TOC). However, the ICRSS data set is more comprehensive than the NRWA data
set. The ICRSS data reflect mean values for 12 months of sampling, whereas the NRWA data reflect a
single sample for each site during two sampling periods. The bromide levels in Exhibit 3.38 are higher
than the small surface waters in the ICRSS, but similar to medium ICRSS plants, as well as the ICR
plants. The alkalinity levels in Exhibit 3.39 are higher than for ICRSS data (where the median for small
systems is 50 mg/L as CaCO3), but are similar to those observed for medium and large surface water
plants in the ICR and WATER:\STATS. The temperature levels in Exhibit 3.40 are similar to those found
in the ICR.
Occurrence Assessment for the Final Stage 2 DBPR 3-41
December 2005
-------
Exhibits 3.41 to 3.44 illustrate some operational characteristics of small surface water plants that
may correlate with the DBF levels observed in such plants. For example, Exhibit 3.41 shows that almost
50 percent of plants in the NRWA survey are in operation 12 hours a day or less. Some small plants are
designed for a peak flow that may be seasonal, and the rest of the time they may operate at reduced flow.
At these rates they may meet their production needs in less than 24 hours. Because they do not operate all
of the time, small water systems may have water with higher residence times within their plant, depending
on the size of their distribution system. This may increase DBF formation in cases where water stays for
a long period of time in a clearwell or finished water storage facility after chlorination. In addition, many
small systems have a smaller number of connections spread over areas as large, or larger, than medium
and large water systems. Therefore, the water may be retained longer in pipes, thereby allowing for
greater DBF formation.
Exhibit 3.42 indicates that only 15 percent of NRWA survey plants listed DBF control as a
treatment objective. This is understandable due to the fact that small systems are not subject to the 1979
TTHM Rule. As shown in Exhibit 3.43, almost all NRWA plants use chlorine as a disinfectant, whereas
40 percent of ICR plants use chloramines, chlorine dioxide, and ozone, which are thought to contribute
less to DBF formation than chlorine. With respect to disinfectant dose, small plants reported larger
chlorine doses than the large ICR plants (Exhibit 3.44).
Exhibit 3.37 Plant Influent TOC Data for Small Surface Water Plants
» Winter Data (N=96)
A Summer Data (N=96)
0%
TOC (mg/L as C)
Source: NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-42
December 2005
-------
Exhibit 3.38 Plant Influent Bromide Data for Small Surface Water Plants
100%
ntile
oi
a.
01
>
1
3
E
3
o
10%
0% ¥-
0
* Winter Data (N=95)
A Summer Data (N=95)
500
1000 1500
Bromide (ug/L)
2000
2500
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.39 Plant Influent Alkalinity for Small Surface Water Systems
» Winter Data (N=95)
A Summer Data (N=95)
100 150 200 250
Alkalinity (mg/L as CaCO3)
300
350
Source: NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-43
December 2005
-------
Exhibit 3.40 Plant Influent Temperature for Small Surface Water Systems
100%
20%
» Winter Data (N=73)
A Summer Data (N=73)
..*
10 15 20 25
Temperature (as degrees Celsius)
30
35
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.41 Distribution of Time Operated per Day Among Small Surface Water
Plants
100% -
80%
-------
Exhibit 3.42 Treatment Objectives Among Small Surface Water Plants
100%
Treatment Objective
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.43 Comparison of Disinfectants Used by Small and Large Surface Water
Plants
100%
90%
80%
70%
• ICR Large SW Plants (N=292, based on month 8)
DSmall Plants in NRWA Survey (N=107)
CI2only
CI2&NH3 CIO2
Plant Disinfectant
Sources: ICRAUX1 (USEPA 2000d); NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-45
December 2005
-------
Exhibit 3.44 Comparison of Total Chlorine Doses in Large and Small Surface
Water Plants Using Only Chlorination (CI2/CI2)
100%
AICR Large SW Plants (N=166, ave. of months 7-9)
• Small Plants in NWRA Survey (N=94)
10%
0%
10 15 20 25 30
Total Chlorine Dose (mg CL2/L)
35
40
Sources: ICRAUX1 (USEPA2000d); NRWA Survey (USEPA 2001 g).
Although the NRWA survey, a key source of DBF data for small surface water systems,
paralleled the ICR effort, the data collection was not as extensive. In the distribution system, NRWA
samples were collected only at the location with the maximum residence time and one location with an
average residence time. Exhibits 3.45 and 3.46 summarize the combined winter and summer NRWA
results for TTHM and HAAS occurrence data. Exhibits 3.47 through 3.49 provide the summer and winter
cumulative distributions of the NRWA TTHM analyses for finished water, average residence time, and
maximum residence time locations, respectively. Similar data are provided for HAAS in Exhibits 3.50
through 3.52.
Despite the fact that small systems generally have lower DBF precursor concentrations than
medium and large systems, NRWA results for small surface water systems show higher byproduct levels
than in medium and large systems. This is understandable, given that small systems have not been
subject to the requirements of the 1979 TTHM standards, which resulted in some medium and large
systems making treatment changes (e.g., increased precursor removal or lower chlorination rates) to limit
byproduct formation. Therefore, it is possible that small systems are applying greater amounts of chlorine
while treating their water. Distribution system size may also play a part. Even though small systems
serve a lower population, their distribution systems are typically as large, if not larger, than those of
medium and large systems, increasing retention times in the pipes.
Occurrence Assessment for the Final Stage 2 DBPR 3-46
December 2005
-------
Exhibit 3.45 Summary of NRWA DBP Occurrence Data by Plant
Data Type
Number of
Plants
Mean of
Plant-Means
Median of
Plant-Means
90th Percentile of
Plant-Means
Range of
Plant-Means
TTHM
Finished
Avg Res Time
DS Weighted Average
Single High
Max Res Time
96
96
96
96
96
62.78
80.22
82.80
118.40
90.54
46.20
56.75
62.06
97.20
67.15
137.00
181.35
179.05
224.80
188.30
0-326.05
0-328.85
0-328.09
0-451.40
0-325.80
HAAS
Finished
Avg Res Time
DS Weighted Average
Single High
Max Res Time
96
96
96
96
96
42.19
46.17
45.32
65.34
42.78
31.75
35.30
33.99
52.90
35.20
82.50
85.00
83.89
113.40
88.95
0-326.90
0-327.50
0-261.56
0-474.90
0-182.20
Note: DS Weighted Average is calculated by giving the average residence time result a weight three times that
of data at the maximum residence location. Refer to section 3.2.1 for a full description of NRWA data
types.
Source: USEPA2001g.
Exhibit 3.46 Summary of NRWA DBP Individual Observations
Data Type
Number of
Observations
Mean
Median
90th Percentile
Range
TTHM
Finished
Avg Res Time
Max Res Time
192
192
192
62.78
80.22
90.54
45.10
58.00
73.30
132.90
153.50
174.50
0-471.50
0-443.90
0-451.40
HAAS
Finished
Avg Res Time
Max Res Time
192
192
192
42.19
46.17
42.78
28.80
34.10
34.60
87.30
90.10
87.90
0-481.10
0-474.90
0-225.00
Note: Refer to section 3.2.1 for a description of NRWA data types.
Source: USEPA2001g.
Occurrence Assessment for the Final Stage 2 DBPR 3-47
December 2005
-------
Exhibit 3.47 Distribution of TTHM Occurrence in Plant Finished Water
» Winter Data (N=96)
A Summer Data (N=96)
100
150
200 250 300
TTHM (ug/L)
350 400 450 500
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.48 Distribution of TTHM Occurrence at the Point of Average Residence
Time in the Distribution System
» Winter Data (N=96)
A Summer Data (N=96)
100
Source: NRWA Survey (USEPA 2001 g).
150 200 250 300 350 400 450 500
TTHM (ug/L)
Occurrence Assessment for the Final Stage 2 DBPR 3-48
December 2005
-------
Exhibit 3.49 Distribution of TTHM Occurrence at the Point of Maximum Residence
Time in the Distribution System
» Winter Data (N=96)
A Summer Data (N=96)
100
Source: NRWA Survey (USEPA 2001 g).
150 200 250 300
TTHM (ug/L)
350
400 450
500
Exhibit 3.50 Distribution of HAAS Occurrence in Plant Finished Water
» Winter Data (N=96)
A Summer Data (N=96)
10%
0%
100
200
300
HAAS (ug/L)
400
500
600
Source: NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-49
December 2005
-------
Exhibit 3.51 Distribution of HAAS Occurrence at the Point of Average Residence
Time in the Distribution System
» Winter Data (N=96)
A Summer Data (N=96)
0%
100
150 200 250 300 350 400
HAAS (ug/L)
450 500
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.52 Distribution of HAAS Occurrence at the Point of Maximum Residence
Time in the Distribution System
» Winter Data (N=96)
A Summer Data (N=96)
100 150
HAAS (ug/L)
200
250
Source: NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-50
December 2005
-------
Exhibit 3.53 compares cumulative distributions of ICR, NRWA, and State plant-mean TTHM
occurrence in distribution systems for small and large surface water plants. For the ICR, the running
annual average of the last four quarters of distribution system data, based on plants with at least three
sampling locations each quarter and at least three quarters of data, is plotted. The NRWA plant-means are
weighted averages of the winter and summer average and maximum residence time samples, where
average residence time samples are given weights three times those of maximum residence samples. (As
noted previously, the NRWA data were weighted to make them comparable to ICR data, for which the
DS Average is calculated for each quarter by averaging results of three samples from locations
approximating average residence time and one sample at the maximum residence location.) The State
data on small surface water systems were collected from over 500 small surface water systems. However,
not all points on the graph represent the same type of data—the points are plant "averages," but some
plants took only one sample, while others took multiple samples. The plants with single samples may
explain some of the very high TTHM plant-means at the upper end of the distribution.
The median TTHM plant-mean value was 66 (ig/L and 62 (ig/L for State and NRWA data,
respectively, while the median RAA value for the ICR was 41 (ig/L. The upper end of the NRWA
distributions for TTHM is much higher than that of the ICR distributions. For example, NRWA 90th
percentile TTHM concentrations are more than double their corresponding ICR concentrations. Only five
ICR plants (2 percent) have TTHM levels exceeding 100 (ig/L (the MCL under the 1979 rule), while 23
(24 percent) NRWA plants and 192 (34 percent) plants in the State data set do.
The distribution of the State data shows TTHM levels at the upper end of the distribution are
higher than those observed in the NRWA data. For example, the 90th percentile concentration in the
State data set is 215 (ig/L, while the NRWA value is 168 (ig/L. This probably is due to the fact that some
of the data points in the State data set were not averaged since some plants reported only one observation.
Exhibit 3.54 shows the co-occurrence of HAA5 and TTHM at NRWA plants. The TTHM and
HAA5 values are plant-means weighted as discussed above. Roughly 22 percent of NRWA (small) plants
had both TTHM and HAA5 plant-means exceeding Stage 1 DBPR limits, as compared to 1.4 percent of
ICR surface water plants (see section 3.3.3 for ICR large plant DBF data analyses).
Although results in this section show that TTHM levels from the State data set are higher than the
levels from the NRWA data set, the NRWA data may also be biased slightly high in terms of national
DBF concentrations. This is because some States with high TOC (as compared to the national average
and based on ICR data) are overrepresented in the survey, while other States with low TOC may be
underrepresented. For example, plants in Louisiana, a high-TOC State, represent 4 percent of plants in
the NRWA survey, but only 1 percent of small non-purchased surface water plants in the country,
according to the Baseline Handbook (USEPA 200 le). The sampling results from plants in over- or
underrepresented States may be skewing the distribution of TOC and DBF data.
Occurrence Assessment for the Final Stage 2 DBPR 3-51 December 2005
-------
Exhibit 3.53 Cumulative Distribution of Mean TTHM Occurrence in Distribution
Systems for Small and Large Surface Water Plants
« 70%
0)
a
0)
Q.
0)
3
o
*Data from 8 States (Small plants, 1998-1999, N=562)
A Data from ICR (Large Plant RAA of last 4-quarter values, N=213)
• NRWA Data (Weighted Small Plant Means, N=96)
80 ug/L, Stage 1 MCL
50 100 150 200 250 300 350 400 450 500
TTHM (ug/L)
Sources: USEPA2000I; USEPA2000d; USEPA2001g.
Occurrence Assessment for the Final Stage 2 DBPR 3-52
December 2005
-------
Exhibit 3.54 RAA TTHM vs. RAA HAAS for 96 Small Surface Water Plants
300
250
200
O)
_3_
1 15°
W
100
21.9%
50 100 150 200 250
TTHM RAA (ug/L)
300
350
400
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.55 shows the percent of plants that reported the maximum TTHM LRAA's at the
finished water, average residence time, and maximum residence time locations. One would expect the
locations with the highest residence time to have the highest DBF levels. However, for reasons stated in
section 3.1.3.2, this is not always the case. Similar to ICR plants, NRWA plants have the highest TTHM
LRAA concentration occurring at sites other than the maximum residence time monitoring site 33 percent
of the time. The highest HAA5 LRAA occurred at the maximum residence time monitoring site in only
48 percent of the plants.
If TTHM and HAA5 occur at the same location rather than different locations, fewer monitoring
sites would be needed to represent TTHM and HAA5 occurrence. However, this is not the case. The
NRWA data set indicates that 56 percent of their plants experienced their highest LRAA TTHM and
HAA5 concentrations at different locations in the distribution system. For plants that had their highest
TTHM and HAA5 LRAA concentrations at the same location, it was not necessarily at the maximum
residence time location. Exhibit 3.56 illustrates that for NRWA plants with the highest TTHM and
HAA5 levels occurring at the same location, the highest TTHM and HAA5 LRAA simultaneously
occurred at a location other than the maximum residence time monitoring location 36 percent of the time.
Occurrence Assessment for the Final Stage 2 DBPR 3-53
December 2005
-------
Exhibit 3.55 Percentage of DS Maximum Observations for TTHM and HAAS by
Sampling Location
80% -r
70%
Finished Water Point Average Residence Time Maximum Residence Time
Source: NRWA Survey (USEPA 2001 g).
Exhibit 3.56 Frequency at Which Highest TTHM or HAAS LRAAs Occurred at the
Same Location for All NRWA Plants
Highest LRAA
TTHM/HAA5
(N=96)
Maximum occurred at
different locations for
56.3% of plants
Maximum occurred at
same locations for
43.8% of plants
Among Plant with Highest LRAA
TTHM/HAA5 at Same Location
(N=42)
64.3% @ MAX
31%@AVG
4.8% @ FINISH
Note: MAX = Maximum Residence Time Point, AVG = Average Residence Time Point, FINISH = Finished Water
Point
Source: NRWA Survey (USEPA 2001 g).
Occurrence Assessment for the Final Stage 2 DBPR 3-54
December 2005
-------
3.2.3 Ground Water Systems
3.2.3.1 Medium Ground Water Systems
Only limited data are available on precursor and byproduct occurrence for medium disinfecting
ground water systems. The most relevant information for assessing byproduct occurrence in ground
water is that provided in the WATER:\STATS database. Exhibits 3.57 to 3.59 provide comparisons of
influent average TOC levels, treatment used, and average TTHM levels for medium and large ground
water systems in the WATER:\STATS data set.
The TOC data and the treatment process information show considerable similarity between
medium and large systems. It should also be noted that the TOC distributions derived from
WATER:\STATS for large and medium systems are similar to those observed for the large ground water
plants in the ICR (see Exhibit 3.2). Average TTHM levels in medium and large ground water systems are
also similar, as shown in Exhibit 3.59, based on WATER:\STATS data. The median average distribution
system concentration for large ground water systems was 12 (ig/L and for medium systems was 10 (ig/L.
Exhibit 3.57 Annual Average TOC in Influent Water TOC for Ground Water
Systems
Percen
I
Cumula
1 \J\J /U
90% -
80% -
70% -
60% -
50% -
>\ •** °a
A **
*f
/
/
1
f~
40% JTA
ZA
30% j
20% «
10%
no/. :
;A
A » Medium Ground Water Systems (N=51 )
5
> A Large Ground Water Systems (N=38)
10 15 20
Plant-Mean TOC (mg/L C)
25
30
Source: WATERASTATS (AWWA2000).
Occurrence Assessment for the Final Stage 2 DBPR 3-55
December 2005
-------
Exhibit 3.58 Treatment Summary for Ground Water Systems (Chlorinating and
Non-Chlorinating)
45%
40%
35% -
• Medium systems 10-100K (N=364)
DLarge systems >100K (N=110)
E O
Source: WATERASTATS (AWWA2000).
Exhibit 3.59 Annual Average Finished Water TTHM for Ground Water Systems
• Medium Ground Water Systems (N=169)
A Large Ground Water Systems (N=68)
40
60
TTHM (ug/L)
80
100
120
Source: WATERASTATS (AWWA2000).
Occurrence Assessment for the Final Stage 2 DBPR 3-56
December 2005
-------
3.2.3.2 Small Ground Water Systems
As with the small surface water systems, there is very limited information available on DBF
precursor levels in small ground water systems that disinfect and insufficient data for determining
national occurrence levels of DBFs (GWSS DBF data were not used because only distribution system
entry point data were available).
Data are not available on influent TOC levels for small ground water systems that disinfect.
However, there are some data available on effluent (finished water) TOC in small, medium, and large
disinfecting ground water systems that can provide some insight into how small system DBF precursor
levels compare with those at larger systems.
Exhibit 3.60 provides the effluent TOC data obtained in the 1982 GWSS. Though this
information is somewhat dated, it is reasonable to assume the following with respect to these data: (1) the
fraction of TOC removed in ground water systems is probably not substantial (based on comparisons of
influent and effluent ICR TOC levels, as well as the absence of TOC-removal technologies, such as
coagulation and filtration, from the majority of ground water plants), so these effluent TOC levels are
reasonable indicators of influent TOC; (2) the levels of TOC in influent ground waters probably have not
changed much since these data were collected (support for this is provided by comparing the effluent data
for the large systems in the GWSS data set to the observed influent TOC levels for large systems in the
ICR); and (3) the comparison across system sizes indicates that, on a national scale, TOC levels in small
disinfecting ground water systems are similar to those of medium and large systems.
For TTHMs themselves, there are some data available for small ground water systems. These
data were collected by seven States during 1998 and/or 1999, as described in the beginning of the chapter.
Exhibit 3.61 shows annual average TTHM levels for more than 2,300 observations and compares them
with ground water ICR running annual averages from the last four quarters of data collection. As with the
surface water data, the State data are inconsistent. A few systems took only one sample per year; the
average of such a value cannot easily be compared to that of a system taking 20 samples a year. This may
explain some of the very high ground water values (e.g., the maximum value is 655 (ig/L). Overall,
however, the State ground water data compare favorably with WATER:\STATS TTHM data for medium
and large plants, with a median value of 3 (ig/L, much less than the median distribution system values for
the other size categories (see Exhibit 3.25). The mean concentration, 17 (ig/L, is slightly below the mean
of 19 (ig/L for medium WATER:\STATS plants.
Occurrence Assessment for the Final Stage 2 DBPR 3-57 December 2005
-------
Exhibit 3.60 Comparison of Effluent TOC for Chlorinating Small, Medium, and
Large Ground Water Systems
E
3
o
20%
Small Systems (<10,000)
Medium Systems (10,000 - 100,000)
- - - Large Systems (>100,000)
6 8 10
Effluent TOC (mg/L)
Source: GWSS (USEPA 1983).
Exhibit 3.61 Cumulative Distribution of TTHM Occurrence as Distribution System
Average for Small and Large Ground Water Plants
• Data from 7 States (1998-1999, N=2336)
A Data from ICR (RAA, N=82)
100 120
TTHM(ug/L)
140
160
180
200
Sources: USEPA 2000I; USEPA 2000d.
Occurrence Assessment for the Final Stage 2 DBPR 3-58
December 2005
-------
3.3 Analysis of Co-Occurrence
Due to the extensive data collection effort of the ICR, many analyses of source water quality
parameters, treatment characteristics, and the resulting finished water quality are possible. This section
presents the results of select analyses of the relationships between several source water quality
parameters, disinfectants, and DBFs.
3.3.1 Total Organic Carbon Concentration and Alkalinity
Organic DBF formation occurs when disinfectants react with organic matter in water. The Stage
1 DBPR requires water systems to remove a certain percentage of TOC based on the TOC and alkalinity
levels of the influent water. Exhibit 3.62 shows the percentage removals required by the Stage 1 DBPR
in the 3x3 matrix for conventional plants. The last column of the matrix also applies to enhanced
softening plants. There are various exceptions and alternative compliance criteria, which are explained in
detail in the Stage 1 DBPR (USEPA 1998a).
Exhibit 3.62 Percent TOC Removal Requirements for Systems Employing
Enhanced Coagulation
Source Water TOC (mg/L)
>2.0-4.0
>4. 0-8.0
>8.0
0-60
35%
45%
50%
Source Water Alkalinity (mg/L as
>60-120
25%
35%
40%
CaC03)
>120
15%
25%
30%
Source: The Stage 1 Disinfectants/Disinfection Byproducts Rule (USEPA 1998a).
Exhibit 3.63 shows the percent of monthly samples in each TOC removal category over the last
12 months (January to December 1998) of the ICR monitoring period. Due to seasonal variation and
other factors affecting source water, the percentage removal requirements for each plant may change from
month to month as the influent TOC and alkalinity vary. Of the three alkalinity groups, the 60-120 mg/L
category had the fewest samples. There were fewer samples with TOC concentrations greater than 4.0
mg/L (20 percent) than samples with TOC concentrations less than 4.0 mg/L (80 percent) over all three
alkalinity ranges. In the 4.0-8.0 mg/L TOC range there is virtually no difference in the number of
samples across the alkalinity groups. Many samples are close to the limits for a percentage removal
group, indicating that the treatment requirements of a plant can easily change.
Occurrence Assessment for the Final Stage 2 DBPR 3-59 December 2005
-------
Exhibit 3.63 Distribution of Monthly Influent TOC (mg/L) and Monthly Influent
Alkalinity (mg/L) Samples Based on ICR Data for All Large Plants
Source Water TOC
Range (mg/L)
<2.0
2.0-4.0
4.0-8.0
>8.0
Total
Percentage
Alkalinity (mg
<60
14%
14%
5%
1%
34%
60-120
10%
14%
5%
0%
29%
/L)
> 120
16%
13%
6%
2%
37%
Total
39%
41%
16%
4%
100%
Source: ICR AUX1 database (USEPA 2000d).
Query: Screened INF TOC and ALK. See Appendix B for query language.
3.3.2 TOC, Bromide, and TTHM
TOC and bromide in raw water influence the formation of DBFs. Although the concentration of
DBFs in the finished water is affected by the treatment applied, higher concentrations of TOC in the
source water are expected to cause a greater occurrence of DBFs if not well controlled. Increases in the
concentration of influent bromide are expected to shift the types of DBFs formed more to brominated
species and raise the concentration by weight of DBFs, because bromide is heavier than chlorine. DBF
formation and speciation, however, depend on many factors other than TOC and bromide, and include the
type of disinfectant, pH, temperature, inorganic demand, and disinfectant residual. This section examines
the relationship between influent TOC and bromide; this relationship is an indicator of the treatability of
the water. A comparison of TOC and bromide source water occurrence is presented. Additional analyses
were performed relating TOC and bromide occurrence in source water to TTHM and HAAS levels in
finished water.
Exhibit 3.64 contains the number of plants by each TOC and bromide category for Exhibits 3.65
through 3.68. There were 286 plants used for this analysis, which is fewer than the 311 presented
previously for TTHM and HAAS analyses of ICR data. The lower number is due to some plants not
having enough bromide or TOC data, or both. There is one category, bromide "MRL-30" and TOC "3-
4", which contains no plants meeting this criteria, and is reflected in the subsequent graphs as zero.
Exhibits 3.65 and 3.66 contain three-dimensional graphs comparing influent bromide, influent
TOC, and finished water TTHM concentrations for surface and ground water plants. Exhibits 3.67 and
3.68 contain the same graphs for finished water HAAS concentrations. The graphs were prepared by first
categorizing each ICR plant by its mean influent water TOC and bromide concentration based on the last
12 months of the ICR collection period (TOC and bromide plant-means were based on monthly data for
only those months that had corresponding TTHM or HAAS data). This resulted in a 5 by 5 matrix
according to the following bromide and TOC concentrations:
• TOC (mg/L): < 0 - 1; 1 - 2; 2 - 3; 3-4; and > 4
• Bromide (i-ig/L): < 0; 0 - 30; 30 - 50; 50 - 100; > 100
For each of the 25 TOC/bromide categories, the mean and 90th percentile of all plant-mean TTHM and
HAA5 concentrations were calculated using data from all of the plants in that category. Like influent
Occurrence Assessment for the Final Stage 2 DBPR 3-60 December 2005
-------
TOC and bromide, TTHM and HAAS plant-mean data is based on the last 12 months of the ICR
collection period. The highest level of TTHM, approximately 50 i-ig/L, is indicated by a light-colored bar
that identifies corresponding values of TOC of > 4 mg/L and of bromide of 30-50 i-ig/L.
These comparisons have some uncertainty because TOC and bromide levels are from raw water,
and TTHM and HAA5 are from finished water. It is therefore not known how treatment (other than
disinfection) might have affected the TOC and bromide concentrations. If not controlled, higher influent
TOC and bromide levels result in higher concentrations of DBFs. However, the pattern is not clear in this
data set because the different treatment processes of most plants reduce DBF formation by removing TOC
at varying levels. Also, bromide forms many other brominated acids that are not included in the
measurements of TTHM or HAA5, making a direct correlation between TTHM, TOC, and bromide
unlikely.
The general trend in all graphs is that TTHM formation increases as TOC increases, but there
seems to be no simple correlation with bromide. These analyses do not account for the effect of
alternative disinfectants, which may have been used in plants that had difficulty treating water with high
TOC and particularly high bromide concentrations. In addition, because all TOC, bromide, and DBF
concentrations are calculated as averages or 90th percentiles for each plant, the exhibits may not capture
relationships between individual observations in one quarter.
The formation of HAA5 related to TOC and bromide in finished water is shown in Exhibit 3.67
and 3.68. The mean and 90th percentile graphs of all sampling points show that HAA5 formation
increases as TOC increases and bromide decreases. Increasing bromide concentrations are expected to
shift the speciation of HAAs to the more bromine-substituted species, which are not included in HAA5.
Exhibit 3.64 Count of Plants by Influent TOC and Bromide Concentrations Based
on ICR Data for All Large Plants
4
Total
100
17
5
6
8
19
55
Total
67
54
69
37
59
286
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, FW TTHM & HAAS by Inf Bromide & TOC. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-61
December 2005
-------
Exhibit 3.65 Finished Water TTHM Concentrations (Mean of Plant-Means) by
Influent TOC and Bromide Concentrations Based on ICR Data for All Large Plants
(ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, FW TTHM & HAA5 by Inf Bromide & TOC. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-62
December 2005
-------
Exhibit 3.66 Finished Water TTHM Concentrations (90th Percentile of Plant-
Means) by Influent TOC and Bromide Concentrations Based on ICR Data for All
Large Plants
>100
Bromide (Ug/L)
50-
100 3°-5° MRL-
30
-------
Exhibit 3.67 Finished Water HAAS Concentrations (Mean of Plant-Means) by
Influent TOC and Bromide Concentrations Based on ICR Data for All Large Plants
>100
50-
,nn 30-50
100 MRL-
Bromide (un/i \ 30
-------
Exhibit 3.68 Finished Water HAAS Concentrations (90th Percentile of Plant-Means)
by Influent TOC and Bromide Concentrations Based on ICR Data for All Large
Plants
>100
50-
100 3°-5° MRL-
30
1 (ug/L)
-------
Note that HAAS does not represent all the HAAs, particularly the more bromine-substituted
HAAs. Hence, for high bromide waters, HAAS may not be as representative of brominated DBF
formation as TTHM.
Exhibit 3.69 RAA of TTHM Occurrence versus RAA of HAAS Occurrence for
Large Surface Water Plants Based on ICR Data (N = 213)
140
40
60 80
RAATTHM(ug/L)
100
120
140
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-66
December 2005
-------
Exhibit 3.70 RAA of TTHM Occurrence versus RAA of HAAS Occurrence for
Large Ground Water Plants Based on ICR Data (N = 82)
120 -
100 -
j"
|> 80
X
2
40 -
20 -
n t
<
tol
1.2%
o
0°
96.3%
o o o
o
O <± O o
L^ t :
1.2%
1.2%
O
RAA TTHM (ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-67
December 2005
-------
Exhibit 3.71 Highest LRAA TTHM versus Highest LRAA HAAS for Large Surface
Water Plants Based on ICR Data (N = 213)
140
120 -
100 -
80 -
60
40 -
20 -
20
O
o o
O
v
-$
o
O
0$
o
o
o o
o
/s
*>o o
^> «
o
o
40
60 80
LRAA TTHM (ug/L)
100
120
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
140
Occurrence Assessment for the Final Stage 2 DBPR 3-t
December 2005
-------
Exhibit 3.72 Highest LRAA TTHM versus Highest LRAA HAAS for Large Ground
Water Plants Based on ICR Data (N = 82)
140
120 -
100 -
i" 80
60
o o
O
0 0
40 -
20 -
«
x><> o
yv NX
o«
o <>
to
o
<®>
20
40
60 80
LRAA TTHM (ug/L)
100
120
140
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-69
December 2005
-------
Exhibit 3.73 Single Highest TTHM versus Single Highest HAAS for Large Surface
Water Plants Based on ICR Data (N = 213)
300 -
OT
15.0%
17.4%
100 150 200
Single High TTHM (ug/L)
250
300
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-70
December 2005
-------
Exhibit 3.74 Single Highest TTHM versus Single Highest HAAS Based on ICR
Data for Large Ground Water Plants (N = 82)
300 -
250 -
200 -
•, 150
O)
c
(55
100 -
50 -
6.1%
O
O
O
0 0
86.6% & <$>
, o V °o <%, o
**m> ««*
1 .2%
O
0 6.1%
0 *
50
100 150 200
Single High TTHM (ug/L)
250
300
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 3-71
December 2005
-------
3.4 Analysis of Regional Trends
3.4.1 Occurrence of TOC
EPA evaluated ICR surface and ground water system data to determine if there were differences
in influent water quality among regions. Exhibits 3.75a and 3.75b show average TOC concentrations by
State for surface and ground water systems, respectively, using ICR data. Exhibit 3.75c shows average
TOC concentrations by State for ground water systems using GWSS data. Surface water systems did not
exhibit any notable regional trends; however, ICR and GWSS data show that Florida has very high TOC
concentrations compared to other States. Florida also has the largest proportion of large ground water
systems of all the States.
Exhibit 3.75a Influent Water TOC Occurrence Distribution for Large ICR Surface
Water Systems
No Data
AJ TOC < 1 to mg/L
D TOC >= 1 to 2 mg/L
= 2 to 3 mg/L
TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: ICR AUX1 Database (USEPA 2000d); mean of all plant-means for each State.
Occurrence Assessment for the Final Stage 2 DBPR 3-72
December 2005
-------
Exhibit 3.75b Influent Water TOC Occurrence Distribution for Large ICR Ground
Water Systems
No Data
AJ TOC < 1 to mg/L
I] TOC >= 1 to 2 mg/L
TOC >= 2 to 3 mg/L
D] TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: ICR AUX1 Database (USEPA 2000d); mean of all plant-means for each State.
Exhibit 3.75c Influent Water TOC Occurrence Distribution for Ground Water
Systems, Derived from the GWSS
No Data
A] TOC < 1 to mg/L
B] TOC >= 1 to 2 mg/L
c] TOC >= 2 to 3 mg/L
D] TOC >= 3 to 4 mg/L
TOC >= 4 mg/L
Source: GWSS (USEPA 1983); mean of all finished water TOC samples in the State.
Occurrence Assessment for the Final Stage 2 DBPR 3-73
December 2005
-------
3.4.2 Occurrence of Bromide
Regional trends in occurrence of bromide in source water were evaluated in the Information
Collection Rule Data Analysis document (McGuire et al. 2002). Exhibits 3.76a and 3.76b shows the State
by State average bromide levels in surface water and ground systems for each State. For surface water
systems, Texas and Florida exhibit the highest influent bromide concentrations, both over 100 (ig/L. The
Midwest region of the country exhibits high influent bromide concentrations overall whereas the
Northeast water contains very little influent bromide. For ground water systems, there are far more States
with very high (>100 (ig/1) bromide levels, primarily the southern half of the continental US.
Overall levels for ground water plants are typically higher than for surface water plants. This is
due, in part because ground water has long contact time with geologic formations that can be sources of
bromide. In addition, methyl bromide, used as a pesticide in agricultural operations, can contribute to
high levels of bromide in water and may explain why the levels are much higher in the Midwest, South,
and California - large agricultural centers.
Exhibit 3.76a Mean Influent Bromide Concentrations, Large ICR Surface Water
Plants
V
>AC
i,
3
Mean Bromide (ug/L)
1 1 No data
H < MRL (20)
IjLl > MRL -30
E > 30 -50
01 > 50-1 00
• >100
Source: Chapter 14 Information Collection Rule Data Analysis document (McGuire et al. 2002).
Occurrence Assessment for the Final Stage 2 DBPR 3-74
December 2005
-------
Exhibit 3.76b Mean Influent Bromide Concentrations, Large ICR Ground Water
Plants
Mean Bromide (ng/L)
H No data
] < MRL (20)
] > MRL-30
30 -50
• > 50-100
• >100
Source: ICR AUX1 Database (USEPA 2000d); mean of all plant-means for each State.
Occurrence Assessment for the Final Stage 2 DBPR 3-75
December 2005
-------
4. National DBF Occurrence: Predicted Post-Stage 1 Baselines
Analyses of disinfection byproducts (DBF) occurrence data to support the development of the
Stage 2 Disinfectants and Disinfection Byproducts Rule (DBPR) is complicated by the fact that existing
national occurrence data were collected before systems had to comply with the Stage 1 DBPR1. Although
some plants might have begun making changes prior to the ICR in anticipation of the Stage 1 DBPR, EPA
believes that DBP occurrence data in this document (specifically, total trihalomethanes [TTHMs] and
haloacetic acids [HAA5]) can be used to represent pre-Stage 1 DBPR baseline conditions.
The purpose of this chapter is to support the development of the Stage 2 DBPR by predicting
TTHM and HAAS levels that could exist after the Stage 1 DBPR is implemented. Section 4.1 provides an
overview of the methodology used to predict post-Stage 1 occurrence using the ICR data. Section 4.2
shows detailed derivation of post-Stage 1 occurrence for large ground and surface water plants. Section
4.3 focuses on the spatial and temporal variability in distribution system TTHM and HAAS occurrence,
including a separate analysis of chlorine (CL2) and chloramine (CLM) plants. EPA uses the methodology
derived herein as one approach to characterize the post-Stage 1 occurrence for conducting the Stage 2
DBPR Economic Analysis (EA). EPA also uses an alternative method for estimating the post-Stage 1
occurrence (known as the Surface Water Analytical Tool [SWAT]) to support the Stage 2 DBPR EA. See
the Stage 2 DBPR EA (USEPA 2005a) for further elaboration on the description of these two methods.
There was no suitable database available to conduct similar analyses for medium and small water systems.
The Stage 2 DBPR EA (USEPA 2005a) presents a detailed discussion of the DBP occurrence for these
systems following the Stage 1 DBPR.
4.1 Summary of Methodology for Predicting Post-Stage 1 DBP Occurrence for Large
Plants
In order to develop a post-Stage 1 DBPR baseline, EPA's analysis must (1) identify which plants
need to make treatment changes to meet the Stage 1 DBPR and (2) quantify the changes in TTHM and
HAAS occurrence resulting from those treatment changes. To this end, EPA developed a method called
"The ICR Matrix Method" that manipulates occurrence data for non-compliant plants to generate a post-
Stage 1 DBPR baseline2. This method uses ICR data and is thus limited to large surface water and
groundwater plants.
The method has three main steps. First, ICR plants are screened to ensure that there are enough
TTHM and HAAS distribution system data so as not to skew the analysis (See chapter 3 for a discussion of
the screening process, including a discussion of data representativeness). Next plants are placed into
compliant and non-compliant "bins" based on their calculated running annual average (RAA) and
locational running annual average (LRAA) TTHM and HAAS concentrations. Compliance is generally
Information Collection Rule (ICR) data were collected in 1997 and 1998. Other occurrence data for
medium and small systems represent similar time frames. Surface water systems serving 10,000 or more people
were required to comply with the Stage 1 DBPR by January 2002 and surface water systems serving fewer than
10,000 people and all ground water systems were required to comply with the Stage 1 DBPR by January 2004.
2Note that the Surface Water Analytical Tool (SWAT) was also used to predict changes in average DBP
occurrence for Stage 1 and Stage 2. SWAT is discussed in detail in the Economic Analysis for the Stage 2 DBPR
(USEPA 2005a).
Occurrence Assessment for the Final Stage 2 DBPR 4-1 December 2005
-------
based on the TTHM/HAA5 Maximum Contaminant Levels assuming a 20 percent safety margin (e.g.,
64/48 RAA for the Stage 1 DBPR) to be consistent with recommendations made during the Microbial-
Disinfection Byproducts Committee (M-DBP) Federal Advisory Committee Act (FACA) Meetings.
The third step in the ICR Matrix Method predicts the post-Stage 1 occurrence (after treatment
changes are made to meet the Stage 1 DBPR) for those plants that are originally in the Stage 1 DBPR non-
compliant bin. To do this, EPA relies on the assumption that some plants using chloramines and/or
advanced technologies at the time of the ICR data collection (1997-1998) had installed those technologies
in anticipation of the Stage 1 DBPR. Thus, the occurrence data for the subset of stage 1-compliant plants
already using chloramines and/or an advanced technology can be used as indicators of occurrence data for
those plants changing technology to meet the Stage 1 (and Stage 2) DBPR. EPA goes a step further and
assumes that plants making treatment changes for the Stage 1 DBPR will also achieve compliance with the
Stage 2 DBPR (TTHM/HAA5 LRAA of 80/60 ug/L with a 20 percent safety margin). The rationale for
this assumption is summarized below (additional discussion can be found in the compliance forecast
analysis of the Stage 2 EA):
• The Stage 2 DBPR is a required rule in the Safe Drinking Water Act (SDWA) Amendments of
1996. Details of the Stage 2 DBPR were published in the Agreement in Principle, which
includes the Stage 2 MCLs, in December 2000, which is well before the Stage 1 compliance
deadlines. It is less costly and, therefore, in a water system's best interest to develop a
comprehensive treatment strategy to achieve simultaneous compliance with both Stage 1 and
Stage 2.
A large portion of systems use chloramines to achieve compliance with the Stage 1 DBPR.
Chloramines generally result in lower spatial and temporal variability of TTHM and HAAS
concentrations in distribution systems compared to chlorine (this will be discussed further in
Section 4.3). Therefore, systems that have switched to chloramines to comply with Stage 1
DBPR will likely have LRAA values already below 80 (ig/L for TTHM and 60 (ig/L for
HAAS and will not need to make a second treatment change to comply with the Stage 2 DBPR.
Analysis of screened plants revealed that 64 of the 172 Stage 2-compliant surface water plants
(assuming a 20 percent safety margin) use chloramines and/or an advanced technology. For screened
ground water plants, 12 of the 72 Stage 2-compliant plants (assuming a 20 percent safety margin) use
chloramines and/or an advanced technology. EPA recognizes that there is uncertainty in using TTHM and
HAAS occurrence data for this subset of plants to represent the occurrence of TTHM and HAAS once all
plants have made changes to meet the Stage 1 DBPR. Plants may have installed advanced technologies for
reasons other than the Stage 2 DBPR. Also, the occurrence of TTHM and HAAS after a treatment change
is dependent on plant-specific conditions. The Stage 2 EA provides some quantification of this uncertainty
by using a second method to assess changes in average TTHM and HAAS concentrations from the pre-
Stage 1 to post-Stage 1 baselines.
4.2 Predicted Post-Stage 1 TTHM and HAAS Occurrence
4.2.1 Large Surface Water Plants
This section provides the detailed derivation of the post-Stage 1 DBPR baseline for ICR surface
water plants using the ICR Matrix Method. Exhibit 4.1 shows how the ICR Matrix Method can be used to
estimate changes in average TTHM and HAAS concentrations for large surface water systems, assuming a
20 percent safety margin for Stage 1 and Stage 2 MCLs. The left side of the exhibit contains two tables or
Occurrence Assessment for the Final Stage 2 DBPR 4-2 December 2005
-------
matrices that are divided into different "bins." The bins are cells defined by ranges of RAA values for
TTHM and HAAS across the top, and maximum LRAA values for TTHM and HAAS down the left-hand
side. The method works by moving plants from the non-compliant bin (Bin B2) into the compliant bin (Bin
Al) in the second table, representing their actions to comply with Stage 1.
The number and percent of plants in each bin under pre-Stage 1 conditions is shown in the tables
on the right-hand side of Exhibit 4.1. Plants are assigned to a bin based on their RAA and LRAA
observations as calculated from the ICR data. Note that a plant is considered in one of the non-compliant
bins if it exceeds either the TTHM or HAAS MCL.
The analysis of TTHM and HAAS levels for Stage 2-compliant plants that use advanced
technologies and/or chloramines during the ICR is summarized in Exhibit 4.2. TTHM and HAAS data for
Stage 2-compliant plants in Exhibit 4.2 is assumed to represent TTHM and HAAS occurrence for those
plants that change technology to meet the Stage 1 rule (plants in Bin B2). This change in TTHM and
HAAS occurrence is reflected on the right hand side of Exhibit 4.1, Post-Stage 1 data (the row for "B2" is
shaded for emphasis). The resulting change in the national average TTHM and HAAS concentration is
calculated as the weighted average for the Stage 1 / Stage 2 compliant plants and the non-compliant
changers (see Exhibit 4.1, right-hand side, post-Stage 1 occurrence data, all plants).
Occurrence Assessment for the Final Stage 2 DBPR 4-3 December 2005
-------
Exhibit 4.1 ICR Matrix Method for Surface Water Plants for the Stage 1 DBPR
(80/60 RAA), 20 Percent Safety Margin
P re-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
136
36
41
213
Percent of
Plants
64%
17%
19%
100%
Average of Plant Averages
(ug/L)
TTHM
31.64
51.64
69.34
42.28
HAAS
20.67
33.12
53.36
29.07
Post-Stage 1
Max
LRAA
<64/48
>= 64/48
(S2 non-
compliant)
RAA
<64/48 |>=64/48(S1
non-compliant)
A1+B2
A2
Bin
A1
A2
B2
All Plants
Number
of Plants
136
36
41
213
Percent of
Plants
64%
17%
19%
100%
Average of P
(Ufi
TTHM
31.64
51.64
31.48
34.99
ant Averages
/L)
HAAS
20.67
33.12
19.14
22.48
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for each
bin under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the averages
of all plant-averages are shown in the first table on the right. A total of 213 ICR plants were evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and Maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64
ug/L and its HAAS RAA level below 48 ug/L to be placed into the bin). The maximum TTHM or HAAS
result determines a plant's bin placement.
3) The crossed-out bin represents plants that have moved from out of compliance bins to in compliance
bins.
4) The gray bin on the right-hand side represents plants that have moved into compliance with Stage 1.
The TTHM and HAAS concentrations for these plants is the average of the values for those ICR plants that
are compliant with Stage 1 and Stage 2 and that use either an advanced technology, chloramines, or both
(64 plants) from Exhibit 4.2.
Source: ICR Aux 1 database (USEPA, 2000h), analysis of ICR screened data (213 surface water plants).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS and Plants min 3x3. See Appendix B for query
language.
Occurrence Assessment for the Final Stage 2 DBPR 4-4
December 2005
-------
Exhibit 4.2 TTHM and HAAS Levels for Stage 2-Compliant Plants Using
Chloramines and/or an Advanced Technology
Subset of Stage 2
Compliant Plants
CLM only
ADV tech only
CLM & Adv. tech
Total
Preferred Regulatory Alternative (20
Percent Safety Margin)
Number of
Plants
47
5
12
64
Mean TTHM
(ug/L)
34.50
32.20
19.33
31.48
Mean HAAS
(ug/L)
20.24
23.19
13.14
19.14
Notes:
Source:
Queries:
All TTHM and HAAS values represent the mean of plant-means
CLM = chloramine
ICR Aux 1 database (USEPA, 2000h), analysis of ICR screened data (213 surface water plants).
Plants min 3x3, RAA & Max LRAA - TTHM & HAAS and Plants min 3x3. See Appendix B for query
language.
4.2.2 Large Ground Water Plants
EPA used the ICR Matrix Method to predict changes in average TTHM and HAAS levels for large
ground water systems following the Stage 1 and Stage 2 rules. A detailed description of the method can be
found in the previous section.
Exhibit 4.3 shows the results of the ICR matrix method for the Stage 1 DBPR. The analysis of
Stage 2-compliant, screened ground water plants using chloramines and/or an advanced technology at the
time of the ICR data collection is shown in Exhibit 4.4. Average TTHM and HAAS concentrations for
those plants are used as indicators of TTHM and HAAS concentrations after plants make treatment
changes to meet rule requirements. The number of ICR GW plants that use chloramines and/or advanced
disinfectants and comply with the Stage 2 DBPR is low: 12 plants (considering a 20 percent safety margin
on compliance). This is roughly 15 percent of the total number of screened ground water plants. EPA
compared TOC levels for the Stage 2-compliant ground water plants that use chloramines and/or an
advanced technology to levels for the Stage 2 non-compliant plants and found them to be similar.
Occurrence Assessment for the Final Stage 2 DBPR 4-5
December 2005
-------
Exhibit 4.3 ICR Matrix Method for Ground Water Plants for the Stage 1 DBPR
(80/60 RAA), 20 Percent Safety Margin
Pre-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
75
2
5
82
Percent of
Plants
91%
2%
6%
100%
Average of Plant Averages
(ug/L)
TTHM
11.62
35.29
63.54
15.36
HAAS
5.50
31.71
43.37
8.45
Post-Stage 1
Bin
A1
A2
B2
All Plants
Number
of Plants
75
2
5
82
Percent of
Plants
91%
2%
6%
100%
Average of Plant Averages
(ug/L)
TTHM
11.62
35.29
27.50
13.16
HAAS
5.50
31.71
18.95
6.96
Notes: 1) In the first table on the left, A1 through B2 are the number of ICR plants that meet the criteria for
each bin under pre-Stage 1 conditions. Their calculated average TTHM and HAAS values based on the
averages of all plant-averages are shown in the first table on the right. A total of 82 ICR plants were
evaluated.
2) Each cell (bin) represents a range of the TTHM and HAAS RAA concentrations and Maximum LRAA
concentrations in |jg/L (i.e., RAA <64/48 means the plant needs to have its TTHM RAA level below 64
pg/L and its HAAS RAA level below 48 pg/L to be placed into the bin). The maximum TTHM or HAAS
result determines the bin placement.
3) The crossed-out bin represents plants that have moved from out of compliance bins to in
compliance bins.
4) The gray bin on the right-hand side represents plants that have moved into compliance with Stage
1. The TTHM and HAAS concentrations for these plants is the average of the values for those ICR
plants that are compliant with Stage 1 and Stage 2 and that use either an advanced technology,
chloramines, or both (12 plants) from Exhibit 4.4.
Source: ICR Aux 1 database (USEPA, 2000h), analysis of ICR screened data (82 ground water plants).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-6
December 2005
-------
Exhibit 4.4 TTHM and HAAS Levels for Stage 2-Compliant Ground Water Plants
Using Chloramines and/or an Advanced Technology
Subset of Stage 2
Compliant Plants
CLM only
ADV tech only
CLM & Adv. tech
Total
Preferred Regulatory Alternative (20
Percent Safety Margin)
Number of
Plants
10
0
2
12
Mean TTHM
(ug/L)
29.0
0.0
19.9
27.5
Mean HAAS
(ug/L)
19.4
0.0
16.5
18.9
Notes: All TTHM and HAAS values represent the mean of plant-means.
Source: ICR Aux 1 database (USEPA, 2000h), analysis of ICR screened data (82 ground water plants).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
4.2.3 Summary of Post-Stage 1 Occurrence
Exhibits 4.5 and 4.6 summarize post-Stage 1 DBPR TTHM and HAAS data, respectively. The
data are presented separately for different source water types (surface, ground, and all) and data types
(plant-mean finished water, distribution system average, locational running annual average (LRAA), and
single high). See section 3.1.3 for a description of the different datatypes. Plant-compliance was
determined by using the Stage 1 DBPR MCLs with a 20 percent safety margin.
TTHM and HAAS values decreased from pre-Stage 1 to post-Stage 1 for both surface water and
ground water. For surface water plants, average values decreased by approximately 17 percent for TTHM
(42.28 to 34.99 |ig/L) and 23 percent for HAAS (29.07 to 22.47 |ig/L). For ground water plants, the
decrease in DBP values was less substantial due to a lower percentage of plants changing their treatment
technology. Still, average concentrations decreased approximately 14 percent for TTHM (15.36 to 13.16
|ig/L) and 18 percent for HAAS (8.45 to 6.96 |ig/L)
For surface water systems results show that the highest LRAA for some Stage 1-compliant plants
is significantly above the Stage 2 DBPR MCL of 80 |ig/L for TTHM and 60 |ig/L for HAAS. Single
highest values are also still very high after Stage 1 compliance (maximum of 124 and 115 |ig/L for TTHM
and HAAS, respectively, based on compliance with a safety margin). Single high values for ground water
system are less, but LRAA values are still above the Stage 2 DBPR MCLs for some plants. However,
LRAA values for HAAS do not exceed the Stage 2 MCL without the safety margin. Section 4.3.1 provides
more detailed analyses of the occurrence of individual TTHM and HAAS peak measurements.
Occurrence Assessment for the Final Stage 2 DBPR 4-7
December 2005
-------
Exhibit 4.5 Summary of Post-Stage 1 TTHM Occurrence for ICR Plants,
Stage 1 DBPR Safety Margin of 20%
Source
Surface
Ground
All2
Data Type1
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Number of
Plants
213
213
213
213
82
82
82
82
302
308
308
308
Mean of
Plant-Means
26.74
34.99
56.38
40.95
8.64
13.16
29.54
18.46
21.91
28.91
48.82
34.75
Median of
Plant-Means
26.57
34.23
54.00
39.68
1.48
6.79
18.50
11.80
20.63
29.78
50.30
33.65
90th Percentile of
Plant-Means
46.20
55.61
91.20
65.73
24.75
35.76
65.50
52.63
45.90
52.89
87.30
61.70
Range of
Plant-Means
0-75
0-64
0-124
0-98
0-58
0-55
0-300
0-99
0-87
0-64
0-300
0-99
Notes: 1 Fora description of the datatypes. See "Aggregation of DBP Data" in section 3.1.3.
2 The "All" plants include those with surface, ground, blended, mixed, or purchased source water types.
Finished water data were not available for blended plants.
Source: ICR AUX1 Database (USEPA 2000d).
Derivation for DS Average shown in Exhibits 4.1 and 4.2 for surface water plants and in Exhibits 4.3
and 4.4 for ground water plants. Derivation for other data types follows the same methodology.
Queries: Plants min 3x3, average by finish location - TTHM & HAAS, Plants min 3x3, RAA & Max LRAA - TTHM
& HAAS and Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Exhibit 4.6 Summary of Post-Stage 1 HAAS Occurrence for ICR Plants,
Stage 1 MCL Safety Margin of 20%
Source
Surface
Ground
All2
Data Type1
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Finished, Plant Mean
DS Average
Single Highest
Highest LRAA
Number of
Plants
213
213
213
213
82
82
82
82
302
308
308
308
Mean of
Plant-Means
19.66
22.48
36.38
26.13
5.03
6.96
14.79
9.25
15.59
18.09
30.13
21.32
Median of
Plant-Means
17.68
21.34
33.20
24.90
0.33
2.24
6.30
4.00
14.75
17.32
27.70
20.57
90th Percentile of
Plant-Means
37.33
39.22
62.50
45.65
17.53
18.48
45.00
25.60
33.70
35.18
57.10
41.45
Range of
Plant-Means
0-53
0-48
0-115
0-60
0-34
0-46
0-84
0-58
0-53
0-48
0-115
0-60
Notes: 1 Fora description of the datatypes. See "Aggregation of DBP Data" in section 3.1.3.
2 The "AN" plants include those with surface, ground, blended, mixed, or purchased source water types.
Finished water data were not available for blended plants.
Source: ICR AUX1 Database (USEPA 2000d).
Derivation for DS Average shown in Exhibits 4.1 and 4.2 for surface water plants and in Exhibits 4.3
and 4.4 for ground water plants. Derivation for other data types follows the same methodology.
Queries: Plants min 3x3, average by finish location - TTHM & HAAS, Plants min 3x3, RAA & Max LRAA - TTHM
& HAAS and Plants min 3x3, Single High - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR
December 2005
-------
4.3 Variability of TTHM and HAAS Occurrence, Post Stage 1 Conditions
This section supports the development of the Stage 2 DBPR by evaluating spatial and temporal
variability of TTHM and HAAS occurrence. Section 4.3.1 evaluates the spatial variability TTHM and
HAAS occurrence for large ground and surface water plants using ICR data. Section 4.3.2 evaluates the
temporal variability in occurrence of individual TTHM and HAAS peaks. Sections 4.3.3 and 4.3.4
characterize the occurrence of yearly average TTHM and HAAS data at different locations.
4.3.1 Spatial Variability of TTHM and HAAS
For the purposes of this document, spatial variability in TTHM and HAAS levels is defined as
differences in TTHM and HAAS concentrations at different locations in the distribution system. Spatial
variability is influenced by system configuration (including storage facilities), residual disinfectant type
(free chlorine versus chloramines), water age, and other factors that affect water quality, such as
heterotrophic bacterial growth.
The extent of spatial variability in TTHM and HAAS values can be an indicator of the impact of
the Initial Distribution System Evaluation (IDSE). For most systems, compliance monitoring for the Stage
2 DBPR is proceeded by an IDSE. The goal of the IDSE is to identify sites that represent high TTHM and
HAAS concentrations in distribution systems. EPA expects there will some increase in high TTHM and
HAAS levels found in distribution systems from the ICR data collection to Stage 2 compliance monitoring
due to the IDSE.
ICR screened data, consisting of 213 surface water plants and 82 ground water plants, were used
to assess spatial variability of DBFs in distribution systems of large and medium surface water systems3.
See section 3.1.3 for a description of the ICR data set and the screening method (only those plants with 3 of
4 quarters of data have TTHM and HAAS data for at least 3 of 4 distribution system locations are
considered in the analysis). EPA examined the spatial variability in the ICR data by examining the
difference between the maximum LRAA value as reported for the last four quarters of the ICR (ICR
LRAAmax) and the second highest LRAA as reported for the last four quarters of the ICR (ICR LRAA2ndHl).
Exhibit 4.7a characterizes the difference between the ICR LRAAmax and ICR LRAA2ndHl for
surface water plants, and Exhibit 4.7b for ground water plants. Note that the average difference between
the ICR LRAAmax and ICR LRAA2ndHl is 5.99 |ig/L for TTHM and 3.20 |ig/L for HAAS. The cumulative
distribution for ICR LRAAmax - LRAA2ndHl for TTHM and HAAS are shown in Exhibits 4.8a and 4.8b for
surface water plants, and Exhibits 4.8c and 4.8d for ground water plants.
A large portion of systems are expected to use chloramines to achieve compliance with the Stage 1
DBPR. EPA believes that systems using chloramines as a secondary disinfectant can operate with a lower
safety margin since chloramines generally result in lower spatial and temporal variability in distribution
systems compared to chlorine. For example, Exhibit 4.7a below shows the average difference in DBP
concentrations between the maximum and average residence time sites in ICR data separately for surface
water plants using chlorine and chloramine. It shows that the average increase in concentration for chlorine
plants is 8.13 |ig/L and the average increase for chloramine plants is 2.44 |ig/L. The increase in HAAS
concentrations is also higher for chlorine plants. At first gland, the opposite is true for ground water
3In the Stage 2 DBPR EA, EPA assess spatial variability for the subset of Stage 2 non-compliant plants to
characterize the impacts of the IDSE. For this document, however, EPA examined all screened plants.
Occurrence Assessment for the Final Stage 2 DBPR 4-9 December 2005
-------
systems. However, the average LRAAmax are much different, with CLM plants more than double the CL2
numbers, which is roughly the same ratio as the average differences.
Exhibit 4.7a Analysis of Variability for Stage 2 Non-Compliant
Surface Water Plants
Number of Screened Plants
Average of LRAAMAX
Average of LRAA2ndHi
Average of (LRAAMAX - LRAA2ndHi)
Max of (LRAAMAX - LRAA2ndHI)
TTHM
CL2
133
53.17
45.04
8.13
38.37
CLM
80
42.84
40.40
2.44
17.70
All
213
49.29
43.30
5.99
38.37
HAAS
CL2
133
36.17
32.39
3.78
56.23
CLM
80
29.49
27.24
2.25
16.33
All
213
33.66
30.46
3.20
56.23
Note: Represents all screened ICR SW plants with LRAAmax + (LRAAmax - LRAA2ndHi) > either 64 TTHM
LRAA or 48 HAAS LRAA that are in compliance with Stage 1.
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Exhibit 4.7b Analysis of Variability for Stage 2 Non-Compliant
Ground Water Plants
Number of Screened Plants
Average of LRAAMAX
Average of LRAA2ndm
Average of (LRAAMAX - LRAA2ndHi)
Max of (LRAAMAX - LRAA2ndHi)
TTHM
CL2
65
14.83
11.18
3.65
37.43
CLM
16
35.45
28.16
7.30
59.08
None
1
126.50
124.50
2.00
2.00
All
82
20.21
15.87
4.34
59.08
HAAS
CL2
65
5.27
3.41
1.86
23.73
CLM
16
31.37
28.67
2.70
13.98
None
1
68.33
65.60
2.73
2.73
All
82
11.13
9.09
2.04
23.73
Note: Represents all screened ICR GW plants with LRAAmax + (LRAAmax - LRAA2ndHi) > either 64 TTHM
LRAA or 48 HAAS LRAA that are in compliance with Stage 1.
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-10
December 2005
-------
Exhibit 4.8a Cumulative Distribution of ICR LRAAmax - ICR LRAA
TTHM Screened Data, Surface Water Plants
2ndHi
100% --
o
12 16 20 24 28 32 36 40 44 48
Difference in TTHM LRAA (ug/L) [LRAAmax - LRAA2ndHi]
52
56
60
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-11
December 2005
-------
Exhibit 4.8b Cumulative Distribution of ICR LRAAmax - ICR LRAA
HAAS Screened Data, Surface Water Plants
2ndHi
CM
II
01
O)
ro
'c
Ol
O
I
E
3
O
Source:
Queries:
3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51
Difference in HAAS LRAA (ug/L) [LRAAmax - LRAA2ndHi]
ICRAUX1 Database (USEPA 2000d).
Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-12
December 2005
-------
Exhibit 4.8c Cumulative Distribution of ICR LRAAmax - ICR LRAA
TTHM Screened Data, Ground Water Plants
2ndHi
100% --
90%
80%
sr 70%
oo
ii
g, 60%
S
§
S 50%
Q.
I
™ 40%
30%
20%
10%
0%
12 16 20 24 28 32 36 40 44
Difference in TTHM LRAA (ug/L) [LRAAmax - LRAA2ndHi]
48
52
56
60
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-13
December 2005
-------
Exhibit 4.8d Cumulative Distribution of ICR LRAAmax - ICR LRAA
HAAS Screened Data, Ground Water Plants
2ndHi
100%
90%
80%
w 70%
60%
'c
01
5 50%
Q_
Ol
I 40%
3
O 30%
20%
10%
0%
12 15 18 21 24 27 30 33 36
Difference in HAAS LRAA (ug/L) [LRAAmax - LRAA2ndHi]
39 42 45 48
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-14
December 2005
-------
4.3.2 Temporal Variability of TTHM and HAAS
Temporal, or seasonal, variability in TTHM and HAAS levels is related to increased temperature
and changes in source water quality. Temporal variability is typically much more of a factor for surface
water systems compared to ground water systems. Even though TTHM and HAAS RAA levels may be
within the MCL, there are still observations that exceed the MCL and may pose health risks.
Exhibit 4.9a shows the delta of RAA and the individual quarterly averages for each plant. There is
greater mean variation in surface water systems than in ground water system for both TTHM and HAAS.
This is most likely a result of the greater temporal stability of water temperatures in ground water systems
than in surface water systems, as described in section 3.1.1. Exhibits 4.9b and 4.9c display the quarterly
TTHM deltas for surface and ground water plants, respectively. Exhibits 4.9d and 4.9e display the
quarterly HAAS deltas for surface and ground water plants, respectively. Similar to Exhibit 4.9a, Exhibits
4.9b and 4.9d show the greater temporal variability in surface waters when compared to Exhibits 4.9c and
4.9e.
Exhibit 4.9a Summary Statistics for Quarterly Average Minus RAA
Surface
Water
Ground
Water
Quarter Three
Quarter Four
Quarter Five
Quarter Six
Quarter Three
Quarter Four
Quarter Five
Quarter Six
Number
of Plants
198
208
195
202
78
80
75
75
Mean
-8.54
2.26
8.96
-2.60
0.41
-1.42
0.88
0.22
Median
-7.39
1.56
5.61
-1.54
-0.09
-0.01
0.00
0.00
90th
Percent! le
5.21
16.02
29.64
10.89
5.09
2.48
7.34
7.26
Range
-50.58-44.70
-40.40-62.52
-30.01 -51.44
-40.96 - 36.60
-16.19-45.86
-28.78-18.73
-27.26 - 33.57
-21.86-29.38
Mean
-2.90
4.38
2.10
-3.70
-0.50
-0.06
0.57
0.02
Median
-1.39
2.32
0.29
-3.07
0.00
0.00
0.00
0.00
90th
Percent! le
7.47
14.94
13.17
3.37
2.05
3.18
3.41
4.03
Range
-44.83-41.85
-40.97 - 82.07
-24.58 - 57.28
-62.56-26.06
-20.89 - 32.30
-26.65-6.53
-7.40 - 29.63
-6.38 - 9.75
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Quarterly Ave - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-15
December 2005
-------
Exhibit 4.9b TTHM Quarterly Average Minus TTHM RAA for Surface Water Plants
TTHM Q3 Delta (N=198)
TTHM Q4 Delta (N=208)
TTHM Q5 Delta (N=195)
TTHM Q6 Delta (N=202)
-20 0 20 40
Quarterly Average TTHM Minus TTHM RAA(ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Quarterly Ave - TTHM & HAAS. See Appendix B for query language.
Exhibit 4.9c TTHM Quarterly Average Minus TTHM RAA for Ground Water Plants
TTHM Q3 Delta (N= 78)
TTHM Q4 Delta (N=80)
TTHM Q5 Delta (N= 75)
TTHM Q6 Delta (N=75)
-40 -30
-20 -10 0 10 20 30
Quarterly Average TTHM Minus TTHM RAA(ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Quarterly Ave - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-16
December 2005
-------
Exhibit 4.9d HAAS Quarterly Average Minus HAAS RAA for Surface Water Plants
HAA5Q3 Delta (N= 198)
HAA5Q4 Delta (N= 208)
HAA5Q5 Delta (N= 195)
HAA5Q6 Delta (N= 202)
-20 0 20 40 60
Quarterly Average HAAS Minus HAAS RAA(ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Quarterly Ave - TTHM & HAAS. See Appendix B for query language.
Exhibit 4.9e HAAS Quarterly Average Minus HAAS RAA for Ground Water Plants
HAAS Q3 Delta (N=78)
HAA5Q4 Delta (N=80)
HAAS Q5 Delta (N=75)
HAA5Q6 Delta (N=75)
-10 0 10 20
Quarterly Average HAAS Minus HAAS RAA(ug/L)
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Quarterly Ave - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-17
December 2005
-------
4.3.3 Occurrence of Yearly Averages Above the MCL at Specific Locations
Cumulative distributions of LRAA values for each distribution system location (AVG1, AVG2,
MAX, and DSE) for the subset of Stage 1-compliant plants are shown in Exhibit 4.10. Results indicate
that there are still locations that regularly receive water over the MCLs, even after systems comply with
the Stage 1 DBPR. From Exhibit 4.10, approximately 1.9 percent of plants (six out of 308) in compliance
with Stage 1 MCLs of 64/48 RAA had one or more locations that, on average, exceeded 80 i-ig/L as a
TTHM LRAA for that same year. Customers served at these locations regularly received water with
TTHM concentrations higher than the MCL.
Exhibit 4.11 shows similar results for HAAS. From Exhibit 4.11, shows that no plants in
compliance with Stage 1 MCLS of 64/48 RAA exceeded 60 ^ig/L as an LRAA. Evaluating TTHM and
HAAS results together, six plants have a maximum TTHM LRAA of 80 i-ig/L or greater, or a maximum
HAAS LRAA of 60 |ig/L or greater.
Exhibit 4.10 Cumulative Percentage of TTHM LRAAs, All Plants in Compliance
with 64/48 RAA (Stage 1 MCL with Safety Margin)
Cumulative Percentage of Plants
100% -
95% -
90% -
t
.
80% -
70% -
J**
•*ff J *
*j
<
a — ° A >$r* cr^
!>*
OTTHM LRAA @ DSE (N=302)
nTTHM LRAA@AVG1 (N=301)
A TTHM LRAA @ AVG2 (N=302)
XTTHM LRAA @ MAX (N=308)
50
60
70
90
100
110
TTHM LRAA, ug/L
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Each LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-18
December 2005
-------
Exhibit 4.11 Cumulative Percentage of HAAS LRAAs, All Plants in Compliance
with 64/48 RAA (Stage 1 MCL with Safety Margin)
100%
95% -
90% -
E
3
o
85%
80%
75% -
70%
O HAAS LRAA DSE (N=302)
D HAA5 LRAA AVG1 (N=301)
A HAAS LRAA AVG2 (N=302)
X HAAS LRAA MAX (N=308)
40
50
60
70 80
HAAS LRAA, ug/L
90
100
110
120
Source: ICR AUX1 Database (USEPA 2000d).
Queries: Plants min 3x3, RAA & Each LRAA - TTHM & HAAS. See Appendix B for query language.
4.3.4 Occurrence of Peak DBFs at Locations Other Than the DS Maximum
The 1979 TTHM rule and Stage 1 DBPR monitoring locations must include a site reflecting
maximum residence time in the distribution system with the intent of capturing the highest DBF levels in
the distribution system. As described in Section 1.4.8, this location is referred to as the "DS Maximum"
for the ICR data set. Analysis of the ICR data in this section show two important results: 1) the monitoring
locations identified as the maximum residence time locations often did not represent those locations with the
highest DBF levels and 2) the highest TTHM and HAAS level often occurred at different points in the
distribution system.
Exhibit 4.12 shows the frequency at which the maximum TTHM LRAA occurred at each
distribution sampling location for screened surface water and ground water plants (see chapter 3 for a
description of data screening). For surface water plants, more than half have the highest TTHM LRAA
concentration occurring at sites other than the maximum residence time monitoring site. For ground water,
more than 60 percent of ICR plants have the highest TTHM LRAA concentration occurring at sites other
than the maximum residence time monitoring site.
Occurrence Assessment for the Final Stage 2 DBPR 4-19
December 2005
-------
For small surface water plants, the frequency at which the highest LRAA occurred at different
locations can be analyzed using National Rural Water Association (NRWA) data. The analysis was only
done for pre-Stage 1 conditions and was presented in Chapter 3 (Section 3.2).
Exhibit 4.13 compares the location of the highest TTHM levels for surface water plants using
chlorine and surface water plants using chloramine. Exhibit 4.14 compares the location of the highest
TTHM levels for ground water plants using chlorine and ground water plants using chloramine. For both
surface water and ground water plants, high TTHM values are more likely to occur at the MAX location
for chlorine plants than chloramine plants.
Exhibit 4.15 shows the frequency at which the maximum HAAS LRAA occurred at each
distribution sampling location for all plants that are in compliance with the Stage 1 DBPR. For surface
water plants, more than 50 percent have the highest HAAS LRAA concentration occurring at sites other
than the MAX location, which is lower than the pre-Stage 1 value of 40.9 percent. For ground water, more
than 70 percent ICR plants have the highest HAAS LRAA concentration occurring at sites other than the
MAX location. Exhibit 4.16 compares the location of the highest HAAS levels for surface water plants
using chlorine and surface water plants using chloramine. Exhibit 4.17 compares similar data for ground
water plants. For surface water, trends in HAAS data are similar to TTHM - high HAAS values are more
likely to occur at the MAX location for chlorine plants than for chloramine plants. For ground water,
however, the frequency that high HAAS levels occurred at the MAX location was lower for chlorine than
for chloramine plants.
Occurrence Assessment for the Final Stage 2 DBPR 4-20 December 2005
-------
Exhibit 4.12 Frequency at Which Highest TTHM LRAA Occurred at Each Sample
Location for All Screened ICR Plants
60% i
50%
40% --
30%
20%
10%
FIN
DSE AVG1
Distribution System Location
AVG2
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-21
December 2005
-------
Exhibit 4.13 Location of Highest TTHM LRAA for Screened ICR Surface Water
Plants by Disinfectant Type
70%
60%
D Chlorine (N=109)
• Chloramine(N=104)
FIN
DSE AVG1
Distribution System Location
AVG2
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-22
December 2005
-------
Exhibit 4.14 Location of Highest TTHM LRAA for ICR Occurrence Data by Plant
for Screened ICR Ground Water Plants by Disinfectant Type
45%
40%
D Chlorine (N=63)
• Chloramine (N=19)
DSE
AVG1
Distribution System Location
AVG2
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-23
December 2005
-------
Exhibit 4.15 Frequency at Which Highest HAAS LRAA Occurred at Each Sample
Location for All Screened ICR Plants
45% i
40% --
35%
30% --
25%
20%
15%
10%
FIN
DSE AVG1 AVG2
Distribution System Location
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-24
December 2005
-------
Exhibit 4.16 Location of Highest HAAS LRAA for Screened ICR Surface Water
Plants by Disinfectant Type
50%
45%
D Chlorine (N=109)
• Chloramine (N=104)
DSE
AVG1
Distribution System Location
AVG2
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-25
December 2005
-------
Exhibit 4.17 Location of Highest HAAS LRAA for Screened ICR Ground Water
Plants by Disinfectant Type
40%
35% --
D Chlorine (N=63)
• Chloramine (N=19)
FIN
DSE AVG1
Distribution System Location
AVG2
MAX
Source: Screened Plants from ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA & Max LRAA - TTHM & HAAS. See Appendix B for query language.
EPA also analyzed whether the highest LRAA for TTHM and HAAS occurred at the same
location. If the highest TTHM and HAAS values occur at the same location rather than different locations,
fewer monitoring sites would be needed to represent TTHM and HAAS occurrence. However, this is not
the case. Only 47.7 percent of plants in compliance with Stage 1 DBPR experienced their highest LRAA
TTHM and HAAS concentrations at different locations in the distribution system. For plants that did have
their highest LRAA TTHM and HAAS concentrations at the same location, it was not necessarily the
MAX location. Exhibit 4.13 illustrates that for Stage 1-compliant ICR plants with the highest TTHM and
HAAS levels occurring at the same location, the highest TTHM and HAAS LRAA simultaneously occurred
at the MAX location in 50.9 percent of the cases.
Occurrence Assessment for the Final Stage 2 DBPR 4-26
December 2005
-------
Exhibit 4.18 Frequency at Which Highest TTHM or HAAS LRAAs Occurred at the
Same Location, Plants in Compliance with 64/48 RAA (Stage 1 MCL with Safety
Margin)
47.7% of
plants
occurred at
different
locations
52.3% of
plants
occurred at
same location
Highest LRAA
TTHM/HAA5
8.7% @ FINISH
9.3% @ DSE
50.9% @ MAX
Among Plants with
Highest LRAA
TTHM/HAA5 at Same
Location
Source:
Queries:
ICRAUX1 Database (USEPA2000d).
Plants min 3x3, RAA & Each LRAA - TTHM & HAAS. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR 4-27
December 2005
-------
5. References
Aieta, E.M.; Berg, J.D. 1986. A review of chlorine dioxide in drinking water treatment. Journal of
American Waterworks Association 78(6):64-72.
American Public Health Association, AWWA, and Water Environment Federation. 1998. Standard
Methods for the Examination of Water and Wastewater, 20th Edition. Clesceri LS, Greenberg AE,
Eaton AD, Editors. American Public Health Association, Washington, DC.
Amirtharajah, A., and C.R. O'Melia. 1990. Coagulation Processes: Destabilization, Mixing, and
Flocculation. In: Water Quality and Treatment: Fourth Edition. AWWA. F.W. Pontius, Technical
Ed. McGraw-Hill, Inc., Washington, DC. 269-365.
Amy, G. L.; Minear, R.A.; Westeroff, R.; Sony, R. 1997. Formation and Control of Brominated ozone
Byproducts. CAT No. 90714. American Water Works Association. Denver, CO. 72.
AWWA. 2000. WATER:\STATS. Database containing 1996 AWWA survey of water systems. American
Waterworks Association: Denver, CO.
AWWARF. 1991. Pilot scale evaluation of ozone and peroxone. American Waterworks Association.
Denver, CO. Prepared by Metropolitan Water District of Southern California and James M.
Montgomery, Consulting Engineers, Inc. AWWARF Order No. 90591.
Backer, L.C., D.L. Ashley, M.A. Bonin, F.L. Cardinali, S.M Kieszak, and J.V Wooten. 2000. Household
exposures to drinking water disinfection by-products: whole blood trihalomethane levels. J
Exposure Analysis and Environmental Epidemiology. 10(4):321-326.
Bissonette, E., R. Grover, J. Trax and E. Kramer. 2000. A National Characterization of Small Surface
Water System Disinfection By-Product and Precursor Levels. Presented at the Annual
Conference of the Association of State Drinking Water Administrators, October, 2000, Portland,
Oregon.
Bissonette, E., R. Grover, J. Trax and E. Kramer. 2001. Effect of Source Water Quality on Disinfection
Byproduct Formation in Small Systems. Presented at the Fourth Biennial CECIA-UIPR
Symposium on Potable Water Issues , February 15-16, 2001, San Juan, PR.
Bolyard, M.; Fair, P.S.; Hautman, D.P. 1993. Sources of chlorate ion in U.S. drinking water. Journal of
AWWA (September):81-88.
Bryant, E.A.; Fulton, G.P.; Budd, G.C. 1992. Disinfection Alternatives for Safe Drinking Water. Hazen
and Sawyer. Van Norstrand Reinhold, New York.
Carlson M.; Hardy D. 1998. Controlling DBFs with Monochloramine. Journal of the American Water
Works Association 90(2): 95-106.
Chen W.J.; Weisel C.P. 1998. Halogenated DBP Concentrations in a Distribution System. American
Waterworks 90(4): 151-163.
Clark R.M.; Sivaganesan M. 1998. Predicting Chlorine Residuals and Formation of TTHMs in Drinking
Water. Journal of Environmental Engineering 124(12): 1203-1210.
Occurrence Assessment for the Final Stage 2 DBPR 5-1 December 2005
-------
Cohn, P, M. Cox, and P. Berger. 1999. Health and Aesthetic Aspects of Water Quality. In: Water
Quality and Treatment: Fifth Edition. AWWA. Raymond D. Letterman, Technical Ed. McGraw-
Hill, Inc., Washington, DC. 254-266.
Corley, R.A., S.M. Gordon, L.A. Wallace. 2000. Physiologically based pharmacokinetic modeling of the
temperature-dependent dermal absorption of chloroform by humans following bath water
exposures. Toxicol Sci 53(1): 13-23.
Cowman, G.A.; Singer, P.C. 1994. Effect of bromide ion on haloacetic acid speciation resulting from
chlorination and chloramination of humic extracts. Proceedings, American Water Works
Association Annual Conference, American Waterworks Association.
DeMers, L.O. and Renner, R. 1992. Alternative Disinfectant Technologies for Small Drinking Water
Systems. AWWA Research Foundation.
Gallagher, D.L., Hoehn, R.C., and Dietrich, A.M. 1994. Sources, Occurrence, and Control of Chlorine
Dioxide By-Product Residuals in Drinking Water. American Water Works Research Foundation,
Denver, CO.
Glaze, W. H.; et al. 1987. The chemistry of water treatment processes involving ozone, hydrogen
peroxide, and ultraviolet radiation. Ozone: Science & Engineering 9(4):335.
Glaze, W.H.; Koga, M.; Cancilla, D. 1989. Ozonation byproducts. 2. Improvement of an aqueous-phase
derivatization method for the detection of formaldehyde and other carbonyl compounds formed
by the ozonation of drinking water. Environmental Science and Technology 23(7):838.
Glaze, W.H., Weinberg, H.S., Krasner, S.W., Sclimenti, M.J. 1991. Trends in Aldehyde Formation and
Removal Through Plants Using Ozonation and Biological Active Filters. In: Proceedings,
AWWAAC-Water Quality for the New Decade, pp. 913-943, Philadelphia, PA, June 22-27.
Gordon, G.; Kieffer, R.G.; Rosenblatt, D.H. 1972. The chemistry of chlorine dioxide. S.J. Lippard (Ed.)
Progress of Inorganic Chemistry, Vol XV, J. Wiley & Sons, New York. pp. 202-286.
Gordon, G.; Adam, L.; Bubnis, B.; Hoyt, B.; Gillette, S.; Wilczak, A. 1995. Minimizing the Chlorate Ion
Formation in Drinking Water when Hypochlorite Ion is the Chlorinating Agent. American Water
Works Research Foundation, Denver, Colorado.
Gordon, S.M., L.A. Wallace, P.J. Callahan, D.V. Kenny, M.C. Brinkman. 1998. Effect of water
temperature on dermal exposure to chloroform. Environmental Health Perspectives 106(6): 337-
345.
Griese, M.H.; Hauser, K.; Berkemeier, M.; Gordon, G. 1991. Using reducing agents to eliminate chlorine
dioxide and chlorite ion residuals in drinking water. Journal of American Waterworks
Association 83(5):56.
Hoehn, R.C., Dietrich, A.M.; Farmer, W.S.; Orr, M.P.; Lee, R.G.; Aieta, E.M; Wood, D.W.; Gordon, G.
1990. Household odors associated with the use of chlorine dioxide. Journal of AWWA
(June): 166-172.
Hoehn, R.C.; Rosenblatt, A.A.; and Gates, D.J. Considerations for Chlorine Dioxide Treatment of
Drinking Water. Proceedings of the American Water Works Association, Water Quality
Technology Conference. November 17-21, 1996. Boston, MA.
Occurrence Assessment for the Final Stage 2 DBPR 5-2 December 2005
-------
Hoehn, R.C.; Rosenblatt, A.A. 1996. Full-scale, high-performance chlorine dioxide gas generator. Poster
presented at: American Waterworks Association, Engineering and Construction Conference.
Denver, CO.
Hoigne, J. and Bader, H. 1977. The role of hydroxyl radical reactions in ozonation processes in aqueous
solutions. Water Research 10:377-386.
Hoigne, J. and Bader, H. 1983a. Rate constants of reaction of ozone with organic and inorganic
compounds in water B I. Non-dissociating organic compounds. Water Research 17:173-183.
Hoigne, J. and Bader, H. 1983b. Rate constants of reaction of ozone with organic and inorganic
compounds in water B II. Dissociating organic compounds. Water Research 17:185-194.
Hoigne, J. and Bader, H. 1988. The formation of trichloronitromethane (chloropicrin) and chloroform in a
combined ozonation/chlorination treatment of drinking water. Water Resources 22(3):313.
Hoyer, O. 1998. Measures to reduce or to avoid THM formation during water treatment and distribution.
Water Supply 16(3-4): 143.
Jacangelo, J.G.; Patania, N.L.; Reagan, K.M.; Aieta, E.M.; Krasner, S.W.; McGuire, M.J. 1989.
Ozonation: Assessing Its Role In the Formation and Control of Disinfection By-products. Journal
of American Water Works Association 81(8):74.
Jensen, J.N.; Johnson, J.D.; St. Aubin, J.J; Christman, R.F. 1985. Effect of Monochloramine on Isolated
Fulvic Acid. Org. Geochem. 8(1):71.
Kool, J.L., J.C. Carpenter, and B.S. Fields. 1999. Effect of monochloramine disinfection of municipal
drinking water on risk of nosocomial Legionnaires' disease. Lancet 353(9149):272-277.
Krasner, S.W.; McGuire, M.J.; Jacangelo, J.J. 1989. The occurrence of disinfection byproducts in U.S.
drinking water. Journal of American Water Works Association 81(8):41.
Krasner, S.W.; Glaze, W.H.; Weinberg, H.S. 1993. Formation of control of bromate during ozonation of
water containing bromide. Journal of American Water Works Association 85(5):62.
Krasner S. W. 1999. "Chemistry of Disinfection By-Product Formation", in Formation and Control of
Disinfection By-Products in Drinking Water, Singer P.C. editor, American Water Works
Association, Denver CO.
Kuo, H.W., T.F. Chiang, I.I. Lo, J.S. Lai, C.C. Chan, and J.D. Wang. 1998. Estimates of cancer risk from
chloroform exposure during showering in Taiwan. The Science of the Total Environment 218(1):
1-7.
Langlais, B.; Reckhow, D.A.; Brink, D.R. (Editors). 1991. Ozone in Drinking Water Treatment:
Application and Engineering. AWWARF and Lewis Publishing, Chelsea, MI.
LeChevallier, M.W., N.J. Welch, and D.B. Smith. 1996. Full-scale studies of factors related to coliform
regrowth in drinking water. Appl. Env. Microbiol. 62(7):2201-2211.
LeLacheur, R.M.; Singer, P.C.; Charles, M.J. 1991. Disinfection byproducts in New Jersey drinking
waters. Proceedings, American Water Works Association Annual Conference, American Water
Works Association, Denver, CO, pp. 791-805.
Occurrence Assessment for the Final Stage 2 DBPR 5-3 December 2005
-------
Logsdon, G.S.; Foellmi, S.; Long, B. 1992. Filtration Studies for Greater Vancouver's Water Supply.
Proceedings, American Water Works Association Annual Conference, American Water Works
Association, Denver CO.
Masschelein, W.J. 1992. Unit Processes in Drinking Water Treatment. New York, NY: Marcel Dekker,
Inc.
McGuire, M.J.; Krasner, S.W.; Gramith, J.T. 1990. Comments on Bromide Levels in State Project Water
and Impacts on Control of Disinfection Byproducts.
McGuire, M.J.; Krasner, S.W.; Reagan, K.M; Aieta, E.M.; Jacangelo, J.G.; Patania, N.L.; Gramith, K.M.
1989. Final report for disinfection by-products in United States drinking waters, Vol. 1. U.S.
Environmental Protection Agency and California Dept. of Health Services.
McGuire, M.J.; Meadow R.G. 1988. AWWARF trihalomethane survey. Journal of AWWA 80(l):61-68.
McGuire, M. J., Mclain, J. L., and Obolensky, A., 2002. Information Collection Rule Data Analysis.
AWWA Research Foundation and AWWA, Denver. Ordering Information:
http://www.awwa.org/bookstore/product.cfm?id=90947
McKnight, A.; Reckhow, D.A. 1992. Reactions of ozonation byproducts with chlorine and chloramines.
Proceedings, American Water Works Association Annual Conference, American Water Works
Association, Denver, CO.
Noack, M. G. and Doerr, R. L. 1978. Reactions of Chlorine, Chlorine Dioxide and Mixtures thereof with
Humic Acid. Water Chlorination, Environmental Input and Health, Vol. 2. Edited by R. C. Jolley,
Ann Arbor, MI, Ann Arbor Science Publishing, Inc.
Owen et al., 1998. Removal of DBP Precursors by GAC Adsorption. AWWA Research Foundation
Report No. 90744.
Petochelli, A. 1995. Chlorine dioxide generation chemistry. Proceedings of the Third International
Symposium. Chlorine Dioxide: Drinking Water, Process Water, and Wastewater Issues.
September 14-15. New Orleans, LA.
Rav-Acha, C.; Serri, A.; Choshen, E.; Limoni, B. 1984. Disinfection of drinking water rich in bromide
with chlorine and chlorine dioxide, while minimizing the formation of undesirable byproducts.
Water Science and Technology 17:611.
Reckhow, D.A.; Singer, P.C.; Malcolm, R.L. 1990. Chlorination of humic materials: Byproduct formation
and chemical interpretations. Environmental Science and Technology 24(11): 1655.
Rice, R.G.; Dimitrou, M.A. 1997. Ozone matures in U.S. drinking water treatment: Impacts of the 1996
Safe Drinking Water Act Amendments. Presented at: Annual Conference of the International
Ozone Association (IOA), Lake Tahoe, NV, August 17-20.
Rice, R.G.; Gomez-Taylor, M. 1986. Occurrence of By-Products of Strong Oxidants Reacting with
Drinking Water Contaminants B Scope of the Problem. Environmental Health Perspectives
69:31.
Rice, R.G. 2000. Personal communication with Natalie Auer, The Cadmus Group. E-mail. August 17,
2000.
Occurrence Assessment for the Final Stage 2 DBPR 5-4 December 2005
-------
Richardson, S.D. 1998. Drinking Water Disinfection By-Products in the Encyclopedia of Environmental
Analysis and Remediation.3:1398-1421. Ed. Robert A. Meyers, John Wiley and Sons, NY.
SAIC. 1996. In development of ICR sampling plan review guidance.
Siddiqui, M.S.; Amy, G.L. 1993. Factors affecting DBF formation during ozone-bromide reactions.
Journal of American Waterworks Association 85(1):63.
Siddiqui, M.S., G. Amy, and R.G. Rice. 1995. Bromate ion formation: a critical review. Journal of
American Waterworks Association. 87(10):58-70.
Singer, P.C.; Chang, S.D. 1989. Correlations between trihalomethanes and total organic halides formed
during water treatment. Journal of AWWA (August):61-65.
Singer, P.C. and G.W. Harrington. 1993. "Coagulation of DBF Precursors: Theoretical and Practical
Considerations." Conference proceedings, AWWA Water Quality Technology Conference,
Miami, FL.
Singer, P.C. and D.A. Reckhow. 1999. "Chemical Oxidation." Water Quality and Treatment, 5th edition.
Letterman R.D. technical editor, American Water Works Association, McGraw-Hill, New York,
NY.
Singer, P.C. 1992. Formation and characterization of disinfection byproducts. Presented at: First
International Conference on the Safety of Water Disinfection: Balancing Chemical and Microbial
Risks.
Singer, P.C. (editor) 1999. Formation and Control of Disinfection Byproducts in Drinking Water.
American Waterworks Association, Denver, CO.
Smith, M.E.; Cowman, G.A.; Singer, P.C. 1993. The impact of ozonation and coagulation on DBP
formation in raw waters. Proceedings, AWWAAC-Water Research.
Speed, M.A.; A. Barnard, R.P. Arber, G.C. Budd, and F.S. Johns. 1987. Treatment alternatives for
controlling chlorinated organic contaminants in drinking water. U.S. Environmental Protection
Agency, Washington DC. EPA/600/2-87/011.
Swanson, W.J., Z. Chowdhury, R. Summers, and G. Solarik. 2001. "Predicting DBFs at Full-Scale:
Calibration and Validation of the Water Treatment Plant Model Using ICR Data" Conference
proceedings, 2001 AWWA Annual Conference and Exhibition, Washington, DC.
Symons, J.M., G.E. Speitel Jr., C. Hwang, S.W. Krasner, and S.E. Barrett. 1998. Factors Affecting
Disinfection Byproduct Formation during Chloramination. Project #803. Report 90728. American
Waterworks Association Research Foundation: Denver, CO. 316 p.
Tarquin, A.; Hansel, G.; Rittman D. 1995. Reduction of chlorite concentrations of potable water with
ferrous chloride. Water Engineering and Management (February):35-37.
Texas Natural Resource Conservation Commission. 1996. Rules and Regulations for Public Water
Supplies Systems.
The Chlorine Institute. 2000. Bromate in sodium hypochlorite-potable water treatment. 2 pages.
http://www.cl2.com/whats new/hypobrom.pdf Website visited October 18, 2000.
Occurrence Assessment for the Final Stage 2 DBPR 5-5 December 2005
-------
Trax, J.R. and Kramer, E.A. 2003. Summary Report: NRWA Small System Study of D/DBP (Winter
1999 - Summer 2003). June, 2003.
Tryby M.E.; Boccelli D.L.; Koechling M.T.; Uber J.G.; Summers R.S.; Rossman L.A. 1999. Booster
Chlorination for Managing Disinfectant Residuals. Journal of AWWA 91(1):95-109.
U.S. Court of Appeals, DC Circuit. 2000. Chlorine Chemistry Council et al. v. Environmental Protection
Agency. No. 98-1627. March 31, 2000. http://pacer.cadc.uscourts.gov/common/opinions/200003/
98-1627a.txt. Website visited August 14, 2000.
USEPA. 1979. National Interim Primary Drinking Water Regulations; Control of Trihalomethanes in
Drinking Water. Federal Register 44(231):68624. November 29, 1979.
USEPA 1981. Community Water Supply Survey - Resample 1981. Preliminary Report, USEPA, TSD,
Cincinnati, Ohio.
USEPA 1982. Community Water Supply Survey - Resample 1981. Preliminary Report, US EPA, TSD,
Cincinnati, Ohio.
USEPA. 1983. Ground Water Supply Survey: Summary of Volatile Organic Contaminant Occurrence
Data. Office of Drinking Water, January 1983.
USEPA. 1989a. National Primary Drinking Water Regulations; Filtration, Disinfection; Turbidity,
Giardia lamblia, Viruses, Legionella, and Heterotrophic Bacteria; Final Rule. Part II. Federal
Register 54(124):27486. June 29, 1989
USEPA. 1989b. National Primary Drinking Water Regulations; Total Coliforms (Including Fecal
Coliform andE. coif); Final Rule. Federal Register 54(124):27544. June 29, 1989.
USEPA. 1994. Final Draft for the Drinking Water Criteria Document on Trihalomethanes. Office of
Science and Technology, U.S. EPA, Washington, DC.
USEPA. 1996a. National Primary Drinking Water Regulations: Monitoring Requirements for Public
Drinking Water Supplies; Final Rule. Federal Register 61(94):24353-24388.
USEPA. 1996b. ICR Sampling Manual. U.S. EPA Office of Water. Washington, DC. EPA-814-B-96-
001. April 1996. Ordering Information:
http://vosemite.epa.gOv/water/owrccatalog.nsf/0/8f4f63885eadOacb85256b0600723d087QpenDo
cument
USEPA. 1996c. DBP/ICR Analytical Methods Manual. U.S. EPA Office of Water. Washington, DC.
EPA-814-B-96-002.
USEPA. 1996d. ICR Microbial Laboratory Manual. U.S. EPA Office of Research and Development.
Washington, DC. EPA/600/R-95/178. April 1996.
USEPA. 1997a. National Primary Drinking Water Regulations; Disinfectants and Disinfection
Byproducts; Notice of Data Availability; Proposed Rule. Federal Register 62(212):59388-59484.
Novembers, 1997.
USEPA. 1997b. Community Water System Survey B Volume I. Overview. U.S. EPA 815-R-97-001a,
January 1997.
Occurrence Assessment for the Final Stage 2 DBPR 5-6 December 2005
-------
USEPA. 1997c. Quality Assurance Project Plan for the Implementation of the Information Collection
Rule. Office of Ground Water and Drinking Water, July 1997.
USEPA. 1998a. National Primary Drinking Water Regulations: Disinfectants and Disinfection
Byproducts; Final Rule. Federal Register 63(241):69389-69476. December 16, 1998.
USEPA. 1998b. National Primary Drinking Water Regulations: Interim Enhanced Surface Water
Treatment; Final Rule. Federal Register 63(241):69477-69521. December 16, 1998.
USEPA. 1998c. Occurrence Assessment for Disinfectants/Disinfection Byproducts in Public Drinking
Water Supplies. Prepared by SAIC for the U.S. EPA, November 1998.
USEPA. 1998d. National Primary Drinking Water Regulations; Disinfectants and Disinfection
Byproducts Notice of Data Availability; Proposed Rule. Federal Register 63(61): 15673-15692.
USEPA. 1999a. Alternative Disinfectants and Oxidants Guidance Manual. Office of Drinking Water, U.S.
EPA 815-R-99-014, April 1999.
USEPA. 1999b. Guidance Manual for Compliance with the Interim Enhanced Surface Water Treatment
Rule: Turbidity Provisions. Office of Drinking Water, U.S. EPA 815/R-99-010, April 1999.
USEPA. 1999c. Quality Assurance Project Plan for the Implementation of the Information Collection
Rule Supplemental Surveys. Office of Ground Water and Drinking Water, March 1999.
USEPA. 2000a. Removal of the Maximum Contaminant Level Goal for Chloroform from the National
Primary Drinking Water Regulations. Fed. Reg. 65(104):34404-34405. May 30, 2000.
USEPA. 2000b. National Primary Drinking Water Regulations: Long Term 1 Enhanced Surface Water
Treatment Rule/Filter Backwash Recycling Rule; Proposed Rule. Federal Register 65(69): 19046-
19150. April 10,2000.
USEPA. 2000c. Stage 2 Microbial and Disinfection Byproducts Federal Advisory Committee Agreement
in Principle. Federal Register 65(251):83015-83024. December 29, 2000.
USEPA. 2000d. AUX1 Database CD version 5.0. EPA Office of Water. Data extracted from ICRFED
database.
USEPA. 2000e. Occurrence Assessment for the Long-Term 1 Enhanced Surface Water Treatment and
Filter Backwash Rule. Prepared for Office of Ground Water and Drinking Water by Science
Applications International Corporation. McLean, VA.
USEPA. 2000f Surface Water Analytical Tool (SWAT) Program Overview.
USEPA. 2000g. Baseline Profile Document for the Ground Water Rule. Final Draft. Prepared by The
Cadmus Group, Inc., Arlington, VA, and SAIC, McLean, VA, for EPA Office of Ground Water
and Drinking Water, Washington, DC. July 2000.
USEPA. 2000h. National Primary Drinking Water Regulations: Ground Water Rule; Proposed Rules.
Federal Register 65(91):30194-30274. May 10, 2000.
USEPA. 2000L Auxiliary 2 (AUX2)~Disinfectant CT and Decay Analyses Version 4.0. EPA Office of
Water. Data extracted from ICRFED database. February 10, 2000.
Occurrence Assessment for the Final Stage 2 DBPR 5-7 December 2005
-------
USEPA. 2000J. Surface Water Analytical Tool (SWAT) Run Outputs for the Stage 2 DBPR Economic
Analysis. EPA Office of Ground Water and Drinking Water. Prepared by The Cadmus Group,
Inc.
USEPA. 2000k. ICR Supplemental Survey Data. EPA Office of Ground Water and Drinking Water.
Prepared by Dyncorp.
USEPA. 20001. TTHM Plant-Mean Data. EPA Office of Ground Water and Drinking Water.
USEPA. 2000m. ICR Auxiliary 1 Database Version 5.0, Query Tool Version 2.0 [CD-ROM]. EPA
815-C-00-002. Cincinnati, Ohio: USEPA.
USEPA. 200Ib. National Primary Drinking Water Regulations; Filter Backwash Recycling Rule; Final
Rule. Federal Register 66(111): 31085-31105. JuneS, 2001.
USEPA. 200le. Drinking Water Baseline Handbook, Third Edition. Draft. Prepared by International
Consultants, Inc. Contract 68-C6-0039. May, 2001.
USEPA. 200If National Primary Drinking Water Regulations; Arsenic and Clarifications to Compliance
and New Source Contaminants Monitoring; Final Rule. Federal Register 66:6976-7066. January
22, 2001. Arsenic Rule.
USEPA. 2001g. National Rural Water Association (NRWA) Survey User Database. EPA OFfice of
Ground Water and Drinking Water. Prepared by The Cadmus Group, Inc.
USEPA. 2002a. National Primary Drinking Water Regulations: Long Term 1 Enhanced Surface Water
Treatment Rule; Final Rule. Federal Register 67(9): 1812-1844. January 14, 2002.
USEPA. 2003c. National Primary Drinking Water Regulations: Stage 2 Disinfectants and Disinfection
Byproducts Rule; Proposed Rule. Federal Register 68(159): 49547-49681. August 18, 2003.
USEPA. 2003d. Work Assignment 1-05 Project-Specific Supplement to the Programmatic Quality
Assurance Project Plan for EPA Contract No. 68-C-02-026. June 2003.
USEPA. 2005a. Economic Analysis for the Stage 2 Disinfectants and Disinfection Byproducts Rule.
Prepared by The Cadmus Group, Inc. Contract 68-C-99-206. EPA 815-R-05-010, December,
2005.
USEPA. 2005b. Occurrence and Exposure Assessment for the Long Term 2 Enhanced Surface Water
Treatment Rule. Prepared by The Cadmus Group, Inc. Contract 68-C-99-206. EPA 815-R-06-
002, December, 2005.
USEPA. 2005c. UV Disinfection Guidance Manual. Prepared by The Cadmus Group, Inc. Contract 68-C-
99-206. December, 2005. In Press.
Van Hoof, F.; Janssens, J.G.; van Dijck, H. 1985. Formation of mutagenic activity during surface water
preozonation and its removal in drinking water treatment. Chemosphere 14(5):501.
Weisel, C.P., H. Kim, P. Haltmeier, and J. Klotz. 1999. Exposure estimates to disinfection by-products of
chlorinated drinking water. Env. Health Perspectives. 107(2): 103-110.
Occurrence Assessment for the Final Stage 2 DBPR 5-8 December 2005
-------
Werdehoff, K.S.; Singer, P.C. 1987. Chlorine dioxide effects on THMFP, TOXFP and the formation of
inorganic by-products. Journal of American Water Works Association 79(9): 107.
White, G.C. 1986. Handbook of chlorination. Van Nostrand Reinhold Company, Inc.: New York, New
York.
White, G.C. 1992. Handbook of chlorination and alternative disinfectants. Third Edition. Van Nostrand
Reinhold Company, Inc.: New York, New York.
Yamada, H. and Somiya, I. 1989. The Determination of Carbonyl Compounds in Ozonated Water By the
PFBOA Method. Ozone Sci. Engin. 11(2): 127.
Occurrence Assessment for the Final Stage 2 DBPR 5-9 December 2005
-------
Appendix A
TTHM and HAAS Speciation Occurrence Data
-------
-------
Appendix A: TTHM and HAAS Speciation Occurrence Data
A.I Introduction
This appendix summarizes ICR data for the individual species of trihalomethanes (THM) and the
five haloacetic acids (HAAS), providing descriptive statistics (number of samples, mean, median, 90th
percentile, and range) for data collected during the Information Collection Rule (ICR) as contained in
EPA ICR AUX1 database (USEPA 2000d). TTHM and HAAS data were sampled quarterly, with the
first quarter running from July to September 1997 and the last (sixth) quarter running from October to
December 1998. The data present here presents the annual period from January 1998 to December 1998.
A.2 Trihalomethanes
The THMs sampled are chloroform (CHC13), bromodichloromethane (BDCM),
dibromochloromethane (DBCM), and bromoform (CHBr3). Total trihalomethane (TTHM) is the
summation of those four THM. If any individual THM was not reported at a sampling location, no
concentration was determined for TTHM at that location at that plant for that sampling period. All
observations reported below the MRL for the individual species were considered zero for calculations,
thus the minimum value for TTHM is zero when all four individual species are below the MRL.
Exhibit A.I presents summary statistics for plant-means of THMs during the last 12 months of
the ICR collection period for plants. Exhibit A.2 presents the mean concentrations, calculated by weight,
and the percent each THM contributes to the TTHM concentration calculated from those means. For
surface water plants, chloroform accounts for the majority of the TTHM concentrations at each location,
ranging from 63.2 percent to 67.1 percent. For ground water plants, chloroform accounts for the majority
of the TTHM concentrations at each location, ranging from 55.2 percent to 43.6 percent. The difference
between surface and ground water systems is reflected in the increased percentages of
dibromochloromethane and bromoform. This is most likely a switch to more brominated THMs as a
result of the higher bromide levels in ground water as compared to surface water (see Chapter 3 for more
information on bromide levels).
Occurrence Assessment for the Final Stage 2 DBPR A-l December 2005
-------
Exhibit A.1 Summary of Trihalomethane Data
Source
Number of
Plants
Mean or Plant
Means
Median or
Plant-Means
9utn percentile or
Plant-Means
Range or
Plant-Means
Chloroform (CHCI3)
Surface
Ground
All
213
82
311
27.89
6.86
21.97
27.10
1.52
18.44
51.01
16.94
46.25
0.00-92.25
0.00-104.03
0.00-104.03
Bromodichloromethane (BDCM)
Surface
Ground
All
213
82
311
9.39
3.23
7.77
8.35
1.36
6.70
17.97
8.87
16.79
0.00-34.13
0.00-21.26
0.00-34.13
Dibromochloromethane (DBCM)
Surface
Ground
All
213
82
311
3.97
3.14
3.88
2.30
1.27
1.99
9.71
8.25
10.63
0.00-32.75
0.00-33.25
0.00-33.25
Bromoform (CHBr3)
Surface
Ground
All
213
82
311
1.03
2.13
1.35
0.00
0.48
0.08
2.41
4.74
3.53
0.00-25.25
0.00-23.60
0.00-25.25
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA - TTHM Speciation. See Appendix B for query language.
Exhibit A.2 TTHM Speciation as a Concentration and Percent of Total
Source
Surface
Ground
Data
Type
FINISH
DSE
AVG1
AVG2
MAX
FINISH
DSE
AVG1
AVG2
MAX
Mean CHCI3
H9/L
A
19.96
26.12
26.99
27.28
31.26
5.35
6.32
7.12
6.79
7.21
% of Total
B =
A/(A+C+E+G)
63.2%
65.2%
65.7%
66.0%
67.1%
55.2%
47.2%
43.6%
44.1%
44.5%
Mean BDCM
H9/L
C
7.52
9.09
9.23
9.19
9.96
1.91
2.70
3.54
3.25
3.41
% of Total
D =
C/(A+C+E+G)
23.8%
22.7%
22.5%
22.2%
21.4%
19.7%
20.2%
21.7%
21.1%
21.0%
Mean DBCM
H9/L
E
3.27
3.87
3.87
3.86
4.23
1.45
2.56
3.42
3.17
3.36
% of Total
F =
E/(A+C+E+G)
10.4%
9.7%
9.4%
9.3%
9.1%
14.9%
19.1%
20.9%
20.6%
20.7%
Mean CHBr3
H9/L
G
0.85
1.00
0.99
0.99
1.11
0.99
1.81
2.25
2.19
2.23
% of Total
H =
G/(A+C+E+G)
2.7%
2.5%
2.4%
2.4%
2.4%
1 0.2%
13.5%
13.8%
14.2%
13.8%
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, average by location - TTHM Speciation. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR A-2
December 2005
-------
A.3 Haloacetic Acids (HAAs)
The haloacetic acids (HAAs) sampled were monochloroacetic acid (MCAA), dichloroacetic acid
(DCAA), trichloroacetic acid (TCAA), monobromoacetic acid (MBAA), dibromoacetic acid (DBAA),
bromochloroacetic acid (BCAA), bromodichloroacetic acid (BDCAA), chlorodibromoacetic acid
(CDBAA), and tribromoacetic acid (TBAA). HAAS is the sum of MCAA, DCAA, TCAA, MBAA, and,
DBAA. The MRL for most HAA species is 1.0 (ig/L, except for MCAA (2.0 (ig/L). All observations
reported below the MRL for the individual species were considered zero for calculations.
Exhibit A.3 presents summary statistics for plant-means of HAAs. Exhibit A.4 presents the mean
concentrations, calculated by weight, and the percent each HAA contributes to the HAAS concentration
calculated from those means. For surface water plants, DCAA accounts for the majority of the HAAS
concentrations at each location, ranging from 42.7 percent to 43.7 percent. For ground water plants,
DCAA accounts for the majority of the HAAS concentrations at each location, ranging from 38.2 percent
to 41.5 percent. The notable difference between surface and ground water systems is between TCAA and
DBAA. For TCAA, There is a much higher range of TCAA percentages in surface water systems (39.0
percent to 41.6 percent) than ground water systems (19.0 percent to 21.0 percent). However, the opposite
is true for DBAA, which has a much higher range of DBAA percentages in ground water systems (32.1
percent to 37.7 percent) than surface water systems (11.2 percent to 13.2 percent). This is most like a
switch to more brominated HAAs as a result of the higher bromide levels in ground water as compared to
surface water (see Chapter 3 for more information on bromide levels).
Exhibit A.3 Summary of Haloacetic Acids
Source
Number of
Plants
Mean of Plant
Means
Median of
Plant-Means
90th Percent! le of
Plant-Means
Range of
Plant-Means
Monochloroacetic Acid (MCAA)
Surface
Ground
All
213
82
311
1.12
0.55
0.93
0.50
0.00
0.30
3.15
1.95
2.89
0.00-9.99
0.00-7.84
0.00-9.99
Dichloroacetic Acid (DCAA)
Surface
Ground
All
213
82
311
13.65
4.57
10.99
12.61
0.44
9.79
24.15
12.84
22.29
0.00-62.80
0.00-42.63
0.00-62.80
Trichloroacetic Acid (TCAA)
Surface
Ground
All
213
82
311
13.06
2.29
9.86
11.19
0.12
7.32
25.90
7.32
22.71
0.00-60.08
0.00-22.71
0.00-60.08
Monobromoacetic Acid (MBAA)
Surface
Ground
All
213
82
311
0.27
0.13
0.22
0.00
0.00
0.00
1.03
0.56
0.74
0.00-10.04
0.00-1.39
0.00-10.04
Dibromoacetic Acid (DBAA)
Surface
Ground
All
213
82
311
0.96
0.91
0.97
0.09
0.09
0.13
2.80
3.03
2.96
0.00-11.77
0.00-12.85
0.00-12.85
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, RAA - HAAS Speciation. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR A-3
December 2005
-------
Exhibit A.4 HAAS Speciation as a Concentration and Percent of Total
Source
Surface
Ground
Data
Type
FINISH
DSE
AVG1
AVG2
MAX
FINISH
DSE
AVG1
AVG2
MAX
Mean MCAA
ug/L
A
0.98
1.17
1.16
1.08
1.11
0.41
0.59
0.56
0.61
0.48
% of Total
B =
A/(A+C+E+G+
I)
3.6%
3.7%
3.6%
3.4%
3.5%
4.4%
5.3%
5.1%
5.3%
4.2%
Mean DCAA
ug/L
C
11.89
13.61
13.89
13.51
13.78
3.58
4.48
4.39
4.75
4.61
% of Total
D =
C/(A+C+E+G+
I)
43.5%
43.7%
43.3%
42.7%
42.7%
38.2%
40.2%
40.2%
41.5%
41.1%
Mean TCAA
ug/L
E
10.66
12.53
13.16
13.10
13.41
1.78
2.30
2.20
2.27
2.35
% of Total
F =
E/(A+C+E+G+
I)
39.0%
40.2%
41.0%
41.4%
41.6%
19.0%
20.7%
20.1%
19.8%
21.0%
Mean MBAA
ug/L
G
0.18
0.24
0.23
0.31
0.34
0.06
0.14
0.12
0.15
0.11
% of Total
H =
G/(A+C+E+G+
I)
0.6%
0.8%
0.7%
1.0%
1.0%
0.6%
1.3%
1.1%
1.3%
1.0%
Mean DBAA
ug/L
I
3.62
3.60
3.65
3.64
3.60
3.54
3.63
3.65
3.67
3.67
% of Total
J =
I/(A+C+E+G+
I)
13.2%
11.6%
11.4%
11.5%
11.2%
37.7%
32.6%
33.4%
32.1%
32.7%
Source: ICR AUX1 Database (USEPA 2000d).
Query: Plants min 3x3, average by location - HAA5 Speciation. See Appendix B for query language.
Occurrence Assessment for the Final Stage 2 DBPR A-4
December 2005
-------
Appendix B
ICR Data Queries
-------
-------
Appendix B: ICR Data Queries
B.I Introduction
This appendix provides the Structured Query Language (SQL) code used to extract data from the
EPA Information Collection Rule (ICR) AUX1 database (USEPA 2000d). Data include water quality
parameters, disinfectants, halogenated organic disinfection byproducts (DBFs), and inorganic DBFs.
Section B.2 provides query language for data analyses, while B.3 provides all queries used for data
screening. Queries are organized alphabetically. A brief description of each query precedes the SQL
code.
B.2 Queries for Data Analysis
Last Ozone Contact Chamber
This query is used to extract the finished water total trihalomethanes (TTHM) and haloacetic acid
(HAAS). Plants screened by whetherthey meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT TUXOZCHM.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
Max(TUXOZCHM.CHMB ID) AS MaxOfCHMB ID, Count(TUXPLTMON.ICRWTPID) AS
CountOflCRWTPID FROM (TUXPLTMON INNER JOIN TUXOZCHM ON
(TUXPLTMON.SAMP PER = TUXOZCHM.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXOZCHM.ICRWTPID)) INNER JOIN [Plant Source Type, Last 12 Months] ON
TUXOZCHM.ICRWTPID = [Plant Source Type, Last 12 Months].ICRWTPID WHERE
(((TUXOZCHM.SAMP_PER)>=7) AND ((TUXPLTMON.WTP_DIS)="o3")) GROUP BY
TUXOZCHM.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type];
Plants min 3x3, average by finish location - TTHM & HAAS
This query is used to extract the finished water total trihalomethanes (TTHM) and haloacetic acid
(HAAS). Plants screened by whetherthey meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by
location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM &
HAA5].EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type], Avg([Plants min
3x3, by location & quarter - TTHM & HAA5].TTHM1) AS AvgOfTTHMl, Avg([Plants min 3x3, by
location & quarter - TTHM & HAA5].HAA51) AS AvgOfHAAS 1, Count([Plants min 3x3, by location &
quarter - TTHM & HAA5].ICRPWSID) AS CountOflCRPWSID, Sum(IIf([tthml]>=40,l,0)) AS
[Quarters > 40?], Sum(IIf([tthml]>=60,l,0)) AS [Quarters > 60?], Sum(IIf([tthml]>=75,l,0)) AS
[Quarters > 75?], Sum(IIf([tthml]>=80,l,0)) AS [Quarters > 80?], Sum(IIf([tthml]>=100,l,0)) AS
[Quarters > 100?], Sum(IIf([tthml]>=120,l,0)) AS [Quarters > 120?], Sum(IIf([haa51]>=30,l,0)) AS
[Quarters > 30?], Sum(IIf([haa51]>=45,l,0)) AS [Quarters > 45?], Sum(IIf([haa51]>=60,l,0)) AS
[Quarters > 60-HAA5?], Sum(IIf([haa51]>=75,l,0)) AS [Quarters > 75-HAA5?],
Sum(IIf([haa51]>=90,l,0)) AS [Quarters > 90?] FROM [Plants min 3x3, by location & quarter - TTHM
& HAA5] INNER JOIN [Plant Source Type, Last 12 Months] ON [Plants min 3x3, by location & quarter
- TTHM & HAA5].ICRWTPID = [Plant Source Type, Last 12 Months].ICRWTPID GROUP BY [Plants
min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by location & quarter -
TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM & HAA5].EVNTNAME,
[Plant Source Type, Last 12 Months].[Derived Source Type] HAVING ((([Plants min 3x3, by location &
Occurrence Assessment for the Final Stage 2 DBPR B-l December 2005
-------
quarter - TTHM & HAA5].EVNTNAME)="Finish")) ORDER BY [Plants min 3x3, by location & quarter
- TTHM & HAA5].ICRWTPID;
Plants min 3x3, FW TTHM & HAAS by Inf Bromide & TOC
This query is used to extract the finished water TTHM and HAAS with plants that have influent
water data for bromide and total organic carbon (TOC). Plants are screened by whether they meet the 3x3
criteria defined in sections 1.4.8 and 3.1.3.
SELECT TUXSAMPLE.ICRWTPID, Avg([Plants min 3x3, by location & quarter - TTHM &
HAA5].HAA51) AS AvgOfHAAS 1, Avg([Plants min 3x3, by location & quarter - TTHM &
HAA5].TTHM1) AS AvgOfTTHMl, Avg(IIf([tuxsample].[evntname]="Influent",IIf([bromide]=-999,
0,[bromide]))) AS [Influent Bromide], Avg(IIf([tuxsample].[evntname]="Influent",IIf([toc]=-999,
0,[toc]))) AS [Influent TOC] FROM ([Plants min 3x3, by location & quarter - TTHM & HAAS] INNER
JOIN TUXSAMPLE ON ([Plants min 3x3, by location & quarter - TTHM & HAA5].SAMP_QTR =
TUXSAMPLE.SAMP_QTR) AND ([Plants min 3x3, by location & quarter - TTHM &
HAA5].ICRWTPID = TUXSAMPLE.ICRWTPID)) LEFT JOIN TUXWQP ON
TUXSAMPLE.EVENTJD = TUXWQP.EVENT ID WHERE ((([Plants min 3x3, by location & quarter -
TTHM & HAA5].EVNTNAME)="finish")) GROUP BY TUXSAMPLE.ICRWTPID HAVING
(((TUXSAMPLE.ICRWTPID)>100)AND((Avg(IIf([tuxsample].[evntname]="Influent",IIf([bromide]=
-999,0,[bromide])))) Is Not Null) AND ((Avg(IIf([tuxsample].[evntname]="Influent",IIf([toc]=-999,0,
[toe])))) Is Not Null)) ORDER BY TUXSAMPLE.ICRWTPID;
Plants min 3x3, RAA - Other DBFs
This query is used to extract the running annual average (RAA) for total organic halide (TOX),
haloacetonitriles (HAN4), chloral hydrate (CH), chloropicrin (CP), 1,1-dichloropropanone (DCP), and
1,1,1-trichloropropanone (TCP). Plants are screened by whether they meet the 3x3 criteria defined in
sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, average by quarter - Other DBFs].ICRWTPID, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfTOXl)
AS [Plant Mean TOX], Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfHAN4_l) AS
[Plant Mean HAW], Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfCHl) AS [Plant
Mean CH], Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfCPl) AS [Plant Mean CP],
Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfDCPl) AS [Plant Mean DCP],
Avg([Plants min 3x3, average by quarter - Other DBPs].AvgOfTCPl) AS [Plant Mean TCP] FROM
[Plants min 3x3, average by quarter - Other DBFs] INNER JOIN [Plant Source Type, Last 12 Months]
ON [Plants min 3x3, average by quarter - Other DBFs] .ICRWTPID = [Plant Source Type, Last 12
Months] .ICRWTPID GROUP BY [Plants min 3x3, average by quarter - Other DBFs] .ICRWTPID, [Plant
Source Type, Last 12 Months].[Derived Source Type];
Plants min 3x3, RAA & Each LRAA - TTHM & HAA5
This query is used to extract the RAA and locational running annual average (LRAA) for FFN,
AVG1, AVG2, DSE, and MAX distribution system sampling locations for paired TTHM and HAAS
values. Plants are screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, average by location - TTHM & HAA5].ICRPWSID, [Plants min 3x3, average
by location - TTHM & HAA5].ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
Avg([Plants min 3x3, average by quarter - TTHM & HAA5].AvgOfTTHMl) AS [TTHM RAA],
Occurrence Assessment for the Final Stage 2 DBPR B-2 December 2005
-------
Avg([Plants min 3x3, average by quarter - TTHM & HAA5].AvgOfHAA51) AS [HAAS RAA],
Min(IIf([Plants min 3x3, average by location - TTHM & HAA5].[evntname]="FINISH",[Plants min 3x3,
average by location - TTHM & HAA5].[AvgOfTTHMl])) AS [TTHM FIN RAA], Min(IIf([Plants min
3x3, average by location - TTHM & HAA5].[evntname]="DSE",[Plants min 3x3, average by location -
TTHM & HAA5].[AvgOfTTHMl])) AS [TTHM DSE RAA], Min(IIf([Plants min 3x3, average by
location - TTHM & HAA5].[evntname]="AVG",[Plants min 3x3, average by location - TTHM &
HAA5].[AvgOfTTHMl])) AS [TTHM AVG RAA], Min(IIf([Plants min 3x3, average by location -
TTHM & HAA5].[evntname]="AVGl",[Plants min 3x3, average by location - TTHM &
HAA5].[AvgOfTTHMl])) AS [TTHM AVG1 RAA], Min(IIf([Plants min 3x3, average by location -
TTHM & HAA5].[evntname]="AVG2",[Plants min 3x3, average by location - TTHM &
HAA5].[AvgOfTTHMl])) AS [TTHM AVG2 RAA], Min(IIf([Plants min 3x3, average by location -
TTHM & HAA5].[evntname]="MAX",[Plants min 3x3, average by location - TTHM &
HAA5].[AvgOfTTHMl])) AS [TTHM MAX RAA], Min(IIf([Plants min 3x3, average by location -
TTHM & HAA5].[evntname]="FINISH",[Plants min 3x3, average by location - TTHM &
HAA5].[AvgOfHaa51])) AS [HAAS FIN RAA], Min(IIf([Plants min 3x3, average by location - TTHM &
HAAS]. [evntname]="DSE", [Plants min 3x3, average by location - TTHM & HAA5].[AvgOfHaa51])) AS
[HAAS DSE RAA], Min(IIf([Plants min 3x3, average by location - TTHM &
HAAS].[evntname]="avg",[Plants min 3x3, average by location - TTHM & HAA5].[AvgOfHaa51])) AS
[HAAS AVG RAA], Min(IIf([Plants min 3x3, average by location - TTHM &
HAA5].[evntname]="avgl",[Plants min 3x3, average by location - TTHM & HAA5].[AvgOfHaa51])) AS
[HAAS AVG1 RAA], Min(IIf([Plants min 3x3, average by location - TTHM &
HAA5].[evntname]="AVG2",[Plants min 3x3, average by location - TTHM & HAA5].[AvgOfHaa51]))
AS [HAAS AVG2 RAA], Min(IIf([Plants min 3x3, average by location - TTHM &
HAA5].[evntname]="MAX",[Plants min 3x3, average by location - TTHM & HAA5].[AvgOfHaa51]))
AS [HAAS MAX RAA] FROM ([Plants min 3x3, average by location - TTHM & HAAS] INNER JOIN
[Plants min 3x3, average by quarter - TTHM & HAAS] ON ([Plants min 3x3, average by location -
TTHM & HAA5].ICRWTPID = [Plants min 3x3, average by quarter - TTHM & HAA5].ICRWTPID)
AND ([Plants min 3x3, average by location - TTHM & HAA5].ICRPWSID = [Plants min 3x3, average
by quarter - TTHM & HAA5].ICRPWSID)) INNER JOIN [Plant Source Type, Last 12 Months] ON
[Plants min 3x3, average by location - TTHM & HAA5].ICRWTPID = [Plant Source Type, Last 12
Months].ICRWTPID GROUP BY [Plants min 3x3, average by location - TTHM & HAA5].ICRPWSID,
[Plants min 3x3, average by location - TTHM & HAA5].ICRWTPID, [Plant Source Type, Last 12
Months].[Derived Source Type];
Plants min 3x3, RAA & Max LRAA - TTHM & HAA5
This query is used to extract the RAA for TTHM and HAAS. Plants are screened by whether
they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, average by location - TTHM & HAA5].ICRPWSID, [Plants min 3x3, average
by location - TTHM & HAA5].ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
Avg([Plants min 3x3, average by quarter - TTHM & HAA5].AvgOfTTHMl) AS [TTHM RAA],
Avg([Plants min 3x3, average by quarter - TTHM & HAA5].AvgOfHAA51) AS [HAAS RAA],
Max([Plants min 3x3, average by location - TTHM & HAA5].AvgOfTTHMl) AS [TTHM LRAA],
Max([Plants min 3x3, average by location - TTHM & HAA5].AvgOfHAA51) AS [HAAS LRAA],
IIf([tthm raa]<=80,IIf([haa5 raa]<=60,"Yes","No"),"No") AS [Stage 1 Compliant (w/out SF)], IIf([tthm
raa]<=64,IIf([haa5 raa]<=48,"Yes","No"),"No") AS [Stage 1 Compliant (w/ SF)], IIf([tthm
lraa]<=80,IIf([haa5 lraa]<=60,"Yes","No"),"No") AS [Stage 2 Compliant (w/out SF)], IIf([tthm
lraa]<=64,IIf([haa5 lraa]<=48,"Yes","No"),"No") AS [Stage 2 Compliant (w/ SF)] FROM ([Plants min
3x3, average by location - TTHM & HAAS] INNER JOIN [Plants min 3x3, average by quarter - TTHM
& HAAS] ON ([Plants min 3x3, average by location - TTHM & HAAS].ICRWTPID = [Plants min 3x3,
Occurrence Assessment for the Final Stage 2 DBPR B-3 December 2005
-------
average by quarter - TTHM & HAA5].ICRWTPID) AND ([Plants min 3x3, average by location - TTHM
& HAA5].ICRPWSID = [Plants min 3x3, average by quarter - TTHM & HAA5].ICRPWSID)) INNER
JOIN [Plant Source Type, Last 12 Months] ON [Plants min 3x3, average by location - TTHM &
HAA5].ICRWTPID = [Plant Source Type, Last 12 Months].ICRWTPID WHERE ((([Plants min 3x3,
average by location - TTHM & HAA5].EVNTNAME)o"fmish")) GROUP BY [Plants min 3x3, average
by location - TTHM & HAA5].ICRPWSID, [Plants min 3x3, average by location - TTHM &
HAA5].ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type];
Plants min 3x3, RAA & Quarterly Ave - TTHM&HAA5
This query is used to extract the difference between the plants RAA and each quarterly average
for TTHM and HAAS. Plants are screened by whether they meet the 3x3 criteria defined in sections 1.4.8
and 3.1.3.
SELECT [Plants min 3x3, average by quarter - TTHM & HAA5].ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], Avg([Plants min 3x3, average by quarter - TTHM &
HAA5].AvgOfTTHMl) AS AvgOfAvgOfTTHMl, Avg([Plants min 3x3, average by quarter - TTHM &
HAA5].AvgOfHAA51) AS AvgOfAvgOfHAASl, Sum(IIf([samp_qtr]=3,[avgoftthml]))-
Avg([AvgOfTTHMl]) AS [TTHM Q3 Delta], Sum(IIf([samp_qtr]=4,[avgoftthml]))-
Avg([AvgOfTTHMl]) AS [TTHM Q4 Delta], Sum(IIf([samp_qtr]=5,[avgoftthml]))-
Avg([AvgOfTTHMl]) AS [TTHM Q5 Delta], Sum(IIf([samp_qtr]=6,[avgoftthml]))-
Avg([AvgOfTTHMl]) AS [TTHM Q6 Delta], Sum(IIf([samp_qtr]=3,[avgofhaa51]))-
Avg([AvgOfHaa51]) AS [HAAS Q3 Delta], Sum(IIf([samp_qtr]=4,[avgomaa51]))-Avg([AvgOfHaa51])
AS [HAAS Q4 Delta], Sum(IIf([samp_qtr]=5,[avgofhaa51]))-Avg([AvgOfHaa51]) AS [HAAS Q5 Delta],
Sum(IIf([samp_qtr]=6,[avgofhaa51]))-Avg([AvgOfHaa51]) AS [HAAS Q6 Delta] FROM [Plants min
3x3, average by quarter - TTHM & HAAS] INNER JOIN [Plant Source Type, Last 12 Months] ON
[Plants min 3x3, average by quarter - TTHM & HAAS].ICRWTPID = [Plant Source Type, Last 12
Months].ICRWTPID GROUP BY [Plants min 3x3, average by quarter - TTHM & HAAS].ICRWTPID,
[Plant Source Type, Last 12 Months].[Derived Source Type];
Plants min 3x3, Single High - TTHM & HAA5
This query is used to extract the single high value for TTHM and HAAS. Plants are screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, max by quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, max by
quarter - TTHM & HAAS].ICRWTPID, Max([Plants min 3x3, max by quarter - TTHM &
HAA5].MaxOfTTHMl) AS [Single High TTHM], Max([Plants min 3x3, max by quarter - TTHM &
HAA5].MaxOfHAA51) AS [Single High HAAS] FROM [Plants min 3x3, max by quarter - TTHM &
HAAS] GROUP BY [Plants min 3x3, max by quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3,
max by quarter - TTHM & HAAS].ICRWTPID;
Plant Source Type, Last 12 Months
This query is used to determine the source water type for each plant in the ICR. It is used to
determine source water type in the majority of queries mentioned in this appendix.
SELECT TUXPLTMON.ICRPWSID, TUXPLTMON.ICRWTPID, Count(TUXPLTMON.SAMP PER)
ASCountOfSAMP_PER,Iif(Count(IIf([msrc_cat]="SW",[msrc_cat]))>0,"SW",
Iif(Count(IIf([msrc_cat]="Mix",[msrc_cat]))>0,"Mix",IIf(Count(IIf([msrc_cat]="GW",[msrc_cat]))>0,"G
W",IIf(Count(IIf([msrc_cat]="PUR",[msrc_cat])),"PUR","NA")))) AS [Derived Source Type] FROM
Occurrence Assessment for the Final Stage 2 DBPR B-4 December 2005
-------
TUXPLTMON WHERE (((TUXPLTMON.SAMP PER)>=7)) GROUP BY TUXPLTMON.ICRPWSID,
TUXPLTMON.ICRWTPID;
Screened BROMATE EPA FIN
This query extracts all bromate values, as measured by EPA, at the finished water point from
plants with at least 9 of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], TUXSAMPLE.EVNTNAME, Avg(IIf([bro3_epa]=-999,0,[bro3_epa]))
AS [plant mean BROMATE], Count(TUXCLDIOX.BRO3_EPA) AS CountOfBRO3_EPA FROM
(TUXPLTMON INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON
TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12 Months] .ICRWTPID) ON
(TUXPLTMON.SAMP PER = TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXSAMPLE.ICRWTPID) AND (TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER
JOIN TUXCLDIOX ON TUXSAMPLE.EVENT ID = TUXCLDIOX.EVENT ID WHERE
(((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY TUXSAMPLE.ICRWTPID,
TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXSAMPLE.EVNTNAME HAVING (((TUXPLTMON.WTP DIS)="CLX" Or
(TUXPLTMON.WTP_DIS)="O3") AND ((TUXSAMPLE.EVNTNAME)="fmish") AND
((Avg(IIf([bro3_epa]=-999,0,[bro3_epa]))) Is Not Null) AND ((Count(TUXCLDIOX.BRO3_EPA))>=9))
ORDER BY TUXSAMPLE.ICRWTPID;
Screened BROMATE UTIL FIN
This query extracts all bromate values, as measured by the utility, at the finished water point from
plants with at least 9 of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], TUXSAMPLE.EVNTNAME, Avg(IIf([bro3util]=-999,0,[bro3util])) AS
[plant mean BROMATE], Count(TUXCLDIOX.BRO3UTIL) AS CountOfBRO3UTIL FROM
(TUXPLTMON INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON
TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12 Months] .ICRWTPID) ON
(TUXPLTMON.SAMP PER = TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXSAMPLE.ICRWTPID) AND (TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER
JOIN TUXCLDIOX ON TUXSAMPLE.EVENT ID = TUXCLDIOX.EVENT ID WHERE
(((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY TUXSAMPLE.ICRWTPID,
TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXSAMPLE.EVNTNAME HAVING (((TUXPLTMON.WTP DIS)="CLX" Or
(TUXPLTMON.WTP_DIS)="O3") AND ((TUXSAMPLE.EVNTNAME)="fmish") AND
((Avg(IIf([bro3util]=-999,0,[bro3util]))) Is Not Null) AND ((Count(TUXCLDIOX.BRO3UTIL))>=9))
ORDER BY TUXSAMPLE.ICRWTPID;
Screened BROMIDE INF
This query extracts all bromide values at the influent water point from plants with at least 9 of 12
months of data.
SELECT TUXSAMPLE.ICRWTPID, Avg(IIf([bromide]=-999,0,[bromide])) AS [plant mean Bromide],
[Plant Source Type, Last 12 Months].[Derived Source Type], Count(TUXWQP.BROMIDE) AS
CountOfBROMIDE, TUXSAMPLE.EVNTNAME
Occurrence Assessment for the Final Stage 2 DBPR B-5 December 2005
-------
FROM ([Plant Source Type, Last 12 Months] INNER JOIN TUXSAMPLE ON [Plant Source Type, Last
12 Months] .ICRWTPID = TUXSAMPLE.ICRWTPID) INNER JOIN TUXWQP ON
TUXSAMPLE.EVENTJD = TUXWQP.EVENT ID WHERE (((TUXSAMPLE.SAMP_PER)>=7))
GROUP BY TUXSAMPLE.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXSAMPLE.EVNTNAME HAVING (((Avg(IIf([bromide]=-999,0,[bromide]))) Is Not Null) AND
((Count(TUXWQP.BROMIDE))>=9)AND((TUXSAMPLE.EVNTNAME)="influent"))
ORDER BY Avg(IIf([bromide]=-999,0,[bromide]));
Screened Chlorite
This query extracts all the chlorite levels in the distribution from plants with chlorine dioxide
disinfection.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg(IIf([chlorite]=-999,0,[chlorite])) AS [plant mean CHLORITE],
Max(IIf([chlorite]=-999,0,[chlorite])) AS [Max of CHLORITE], TUXSAMPLE.SAMP PER,
Count(TUXCHLORS.CHLORITE) AS CountOfCHLORITE FROM (TUXPLTMON INNER JOIN
(TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID =
[Plant Source Type, Last 12 Months].ICRWTPID) ON (TUXPLTMON.SAMP PER =
TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID = TUXSAMPLE.ICRWTPID) AND
(TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER JOIN TUXCHLORS ON
TUXSAMPLE.EVENTJD = TUXCHLORS.EVENT ID WHERE
(((TUXSAMPLE.EVNTNAME)="nfc" Or (TUXSAMPLE.EVNTNAME)="avgl" Or
(TUXSAMPLE.EVNTNAME)="avg2" Or (TUXSAMPLE.EVNTNAME)="avg" Or
(TUXSAMPLE.EVNTNAME)="max" Or (TUXSAMPLE.EVNTNAME)="dse")) GROUP BY
TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived
Source Type], TUXSAMPLE.SAMP PER HAVING (((TUXPLTMON.WTP_DIS)="CLX") AND
((Avg(IIf([chlorite]=-999,0,[chlorite]))) Is Not Null) AND ((Max(IIf([chlorite]=-999,0,[chlorite]))) Is Not
Null) AND ((TUXSAMPLE.SAMP_PER)>=7)) ORDER BY TUXSAMPLE.ICRWTPID;
Screened Chlorite DSAVG
This query extracts all RAA of the chlorite levels in the distribution from plants with at least 9 of
12 months of data.
SELECT [Screened CHLORITE] .ICRWTPID, [Screened CHLORITE] .WTP DIS, [Screened
CHLORITE].[Derived Source Type], Avg([Screened CHLORITE].[plant mean CHLORITE]) AS
[AvgOfplant mean CHLORITE], Count([Screened CHLORITE].SAMP PER) AS CountOfSAMP_PER
FROM [Screened CHLORITE] GROUP BY [Screened CHLORITE] .ICRWTPID, [Screened
CHLORITE].WTP DIS, [Screened CHLORITE].[Derived Source Type] HAVING (((Count([Screened
CHLORITE].SAMP_PER))>=9)) ORDER BY Avg([Screened CHLORITE].[plant mean CHLORITE]);
Screened CHLORITE FIN
This query extracts all chlorite values at the finished water point from plants with at least 9 of 12
months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg(IIf([chlorite]=-999,0,[chlorite])) AS [plant mean CHLORITE],
Max(IIf([chlorite]=-999,0, [chlorite])) AS [Max of CHLORITE] FROM (TUXPLTMON INNER JOIN
(TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID =
Occurrence Assessment for the Final Stage 2 DBPR B-6 December 2005
-------
[Plant Source Type, Last 12 Months].ICRWTPID) ON (TUXPLTMON.SAMP_PER =
TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID = TUXSAMPLE.ICRWTPID) AND
(TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER JOIN TUXCHLORS ON
TUXSAMPLE.EVENT ID = TUXCHLORS.EVENT ID WHERE
(((TUXSAMPLE.EVNTNAME)="fmish") AND ((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY
TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived
Source Type] HAVING (((TUXPLTMON.WTP_DIS)="CLX") AND
((Avg(IIf([chlorite]=-999,0,[chlorite]))) Is Not Null) AND ((Max(IIf([chlorite]=-999,0,[chlorite]))) Is Not
Null) AND ((Count(TUXCHLORS.CHLORITE))>=9)) ORDER BY
Avg(IIf([chlorite]=-999,0,[chlorite]));
Screened Chlorite Single High
This query extracts the highest chlorite value in the distribution from plants with at least 9 of 12
months of data. See section 1.4.8 for a detail description of single high calculations.
SELECT [Screened CHLORITE] .ICRWTPID, [Screened CHLORITE] .WTP DIS, [Screened
CHLORITE].[Derived Source Type], Max([Screened CHLORITE].[Max of CHLORITE]) AS [Single
High Chlorite], Count([Screened CHLORITE].SAMP PER) AS CountOfSAMP_PER FROM [Screened
CHLORITE] GROUP BY [Screened CHLORITE] .ICRWTPID, [Screened CHLORITE] .WTP DIS,
[Screened CHLORITE].[Derived Source Type] HAVING (((Count([Screened
CHLORITE].SAMP_PER))>=9)) ORDER BY Max([Screened CHLORITE].[Max of CHLORITE]);
Screened EXCLXRES FIN
This query extracts all chlorine dioxide residual values at the finished water point from plants
with at least 9 of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg(IIf([EXCLXRES]=-333,0,[EXCLXRES])) AS [Plant Mean CLX
Residual], Count(TUXSAMPLE.ICRWTPID) AS CountOflCRWTPID FROM (TUXPLTMON INNER
JOIN (TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID
= [Plant Source Type, Last 12 Months] .ICRWTPID) ON (TUXPLTMON.SAMP PER =
TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID = TUXSAMPLE.ICRWTPID) AND
(TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER JOIN TUXDISFRES ON
TUXSAMPLE.EVENT ID = TUXDISFRES.EVENT ID WHERE
(((TUXSAMPLE.EVNTNAME)="Finish") AND ((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY
TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived
Source Type] HAVING (((TUXPLTMON.WTP_DIS)="CLX") AND
((Avg(IIf([EXCLXRES]=-333,0,[EXCLXRES]))) Is Not Null) AND
((Count(TUXSAMPLE.ICRWTPID))>=9));
Screened EXFCLRES FIN
This query extracts all free chlorine residual values at the finished water point from plants with at
least 9 of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg(IIf([EXFCLRES]=-333,0,[EXFCLRES])) AS [Plant Mean Free
CL2 Residual], Count(TUXSAMPLE.ICRWTPID) AS CountOflCRWTPID FROM (TUXPLTMON
INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON
Occurrence Assessment for the Final Stage 2 DBPR B-7 December 2005
-------
TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12 Months] .ICRWTPID) ON
(TUXPLTMON.SAMP PER = TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXSAMPLE.ICRWTPID) AND (TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER
JOIN TUXDISFRES ON TUXSAMPLE.EVENT ID = TUXDISFRES.EVENT ID WHERE
(((TUXSAMPLE.EVNTNAME)="Finish") AND ((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY
TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived
Source Type] HAVING (((TUXPLTMON.WTP_DIS)="CL2") AND
((Avg(IIf([EXFCLRES]=-333,0,[EXFCLRES]))) Is Not Null) AND
((Count(TUXSAMPLE.ICRWTPID))>=9));
Screened EXO3RES
This query extracts all ozone residual values at the finished water point from plants with at least 9
of 12 months of data.
SELECT [Last Ozone Contact Chamber].ICRWTPID, [Last Ozone Contact Chamber].[Derived Source
Type], Avg(IIf([EXO3RES]=-333,0,[EXO3RES])) AS [Plant Mean Ozone Residual] FROM [Last Ozone
Contact Chamber] INNER JOIN (TUXDISFRES INNER JOIN TUXOZCHM ON
TUXDISFRES.EVENTJD = TUXOZCHM.EVENTJD) ON ([Last Ozone Contact
Chamber].MaxOfCHMBJD = TUXOZCHM.CHMB ID) AND ([Last Ozone Contact
Chamber].ICRWTPID = TUXOZCHM.ICRWTPID) WHERE (((TUXOZCHM.SAMP_PER)>=7))
GROUP BY [Last Ozone Contact Chamber].ICRWTPID, [Last Ozone Contact Chamber].[Derived
Source Type] HAVING ((([Last Ozone Contact Chamber].[Derived Source Type])="sw") AND
((Count(TUXOZCHM.ICRWTPID))>=9));
Screened EXTCLRES FIN
This query extracts all total chlorine residual values at the finished water point from plants with at
least 9 of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12
Months].[Derived Source Type], Avg(IIf([EXTCLRES]=-333,0,[EXTCLRES])) AS [Plant Mean Total
CL2 Residual], Count(TUXSAMPLE.ICRWTPID) AS CountOflCRWTPID FROM (TUXPLTMON
INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON
TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12 Months] .ICRWTPID) ON
(TUXPLTMON.SAMP PER = TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXSAMPLE.ICRWTPID) AND (TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER
JOIN TUXDISFRES ON TUXSAMPLE.EVENT ID = TUXDISFRES.EVENTJD WHERE
(((TUXSAMPLE.EVNTNAME)="Finish") AND ((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY
TUXSAMPLE.ICRWTPID, TUXPLTMON.WTP_DIS, [Plant Source Type, Last 12 Months].[Derived
Source Type] HAVING (((TUXPLTMON .WTP_DIS) Is Not Null And
(TUXPLTMON.WTP_DIS)<>"O3" And (TUXPLTMON.WTP_DIS)<>"CLX") AND
((Avg(IIf([EXTCLRES]=-333,0,[EXTCLRES]))) Is Not Null) AND
((Count(TUXSAMPLE.ICRWTPID))>=9));
Screened Plant Disinfectant Type
This query extracts all plant-month treatment plant disinfectant types for plants with at least 9 of
12 months of data.
Occurrence Assessment for the Final Stage 2 DBPR B-8 December 2005
-------
TRANSFORM Count(TUXPLTMON.ICRWTPID) AS CountOflCRWTPID SELECT
TUXPLTMON.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type]
FROM TUXPLTMON INNER JOIN [Plant Source Type, Last 12 Months] ON
TUXPLTMON.ICRWTPID = [Plant Source Type, Last 12 Months].ICRWTPID WHERE
(((TUXPLTMON.ICRWTPID)>99) AND ((TUXPLTMON.SAMP_PER)>=7)) GROUP BY
TUXPLTMON.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Souce Type] PIVOT
[WTP_DIS] & "/" & [DS_DIS];
ScreenedSWPlant-Mean CL2 Doses (w AUX2)
This query extracts all chlorine dose values from plants with at least 9 of 12 months of data.
Chlorine dose values are taken from TUXCTSUM from the AUX2 database.
SELECT TUXCTSUM_AUX2.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source
Type], IIf([WTP_DIS] & "-" & [DS_DIS]="CL2-CL2","CL2 Only",IIf([WTP_DIS] Like
"*CLM*","CLM Only",[WTP_DIS] & "-" & [DS_DIS])) AS [System Disinfect Type],
Avg(TUXCTSUM_AUX2.TCL2DOSE) AS AvgOfTCL2DOSE,
Count(TUXCTSUM_AUX2.TCL2DOSE) AS CountOfTCL2DOSE FROM TUXPLTMON INNER JOIN
([Plant Source Type, Last 12 Months] INNER JOIN TUXCTSUM_AUX2 ON ([Plant Source Type, Last
12 Months].ICRWTPID = TUXCTSUM AUX2.ICRWTPID) AND ([Plant Source Type, Last 12
Months].ICRPWSID = TUXCTSUM_AUX2.ICRPWSID)) ON (TUXPLTMON.SAMP PER =
TUXCTSUM AUX2.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXCTSUM AUX2.ICRWTPID) AND (TUXPLTMON.ICRPWSID =
TUXCTSUM AUX2.ICRPWSID) WHERE (((TUXCTSUM_AUX2.SAMP_PER)>=7) AND
((TUXCTSUM_AUX2.ICRWTPID)>99)) GROUP BY TUXCTSUM_AUX2.ICRWTPID, [Plant Source
Type, Last 12 Months].[Derived Source Type], IIf([WTP_DIS] & "-" & [DS_DIS]="CL2-CL2","CL2
Only",IIf([WTP_DIS] Like "*CLM*","CLM Only",[WTP_DIS] & "-" & [DS_DIS])) HAVING
(((IIf([WTP_DIS] & "-" & [DS_DIS]="CL2-CL2","CL2 Only",IIf([WTP_DIS] Like "*CLM*","CLM
Only",[WTP_DIS] & "-" & [DS_DIS])))="CL2 Only" Or (IIf([WTP_DIS] & "-" &
[DS_DIS]="CL2-CL2","CL2 Only",IIf([WTP_DIS] Like "*CLM*","CLM Only",[WTP_DIS] & "-" &
[DS_DIS])))="CLM Only") AND ((Count(TUXCTSUM_AUX2.TCL2DOSE))>=9)) ORDER BY
IIf([WTP_DIS] & "-" & [DS_DIS]="CL2-CL2","CL2 Only",IIf([WTP_DIS] Like "*CLM*","CLM
Only",[WTP_DIS] & "-" & [DS_DIS])), Avg(TUXCTSUM_AUX2.TCL2DOSE);
Screened SW Plant-Mean CLX Doses (w AUX2)
This query extracts all chlorine dioxide dose values from plants with at least 9 of 12 months of
data. Chlorine dioxide dose values are taken from TUXCTSUM from the AUX2 database.
SELECT TUXPLTMON.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXPLTMON.WTP DIS, Avg(TUXCTSUM AUX2.TCLXDOSE) AS AvgOfTCLXDOSE,
Avg([tclxdose])-Min([tclxdose]) AS [Low Error], Max([tclxdose])-Avg([tclxdose]) AS [High Error],
Count(TUXCTSUM_AUX2.TCLXDOSE) AS CountOfTCLXDOSE FROM (TUXPLTMON INNER
JOIN TUXCTSUM AUX2 ON (TUXPLTMON.SAMP PER = TUXCTSUM AUX2.SAMP PER)
AND (TUXPLTMON.ICRWTPID = TUXCTSUM AUX2.ICRWTPID) AND
(TUXPLTMON.ICRPWSID = TUXCTSUM_AUX2.ICRPWSID)) INNER JOIN [Plant Source Type,
Last 12 Months] ON (TUXCTSUM AUX2.ICRWTPID = [Plant Source Type, Last 12
Months] .ICRWTPID) AND (TUXCTSUM AUX2.ICRPWSID = [Plant Source Type, Last 12
Months] .ICRPWSID) WHERE (((TUXPLTMON.SAMP_PER)>=7) AND
((TUXPLTMON.ICRWTPID)>99)) GROUP BY TUXPLTMON.ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], TUXPLTMON.WTP DIS HAVING ((([Plant Source Type, Last 12
Occurrence Assessment for the Final Stage 2 DBPR B-9 December 2005
-------
Months].[Derived Source Type]) Is Not Null) AND ((TUXPLTMON.WTP_DIS)="CLX") AND
((Count(TUXCTSUM_AUX2.TCLXDOSE))>=9)) ORDER BY
Avg(TUXCTSUM_AUX2 .TCLXDOSE);
ScreenedSWPlant-Mean O3 Doses (w AUX2)
This query extracts all ozone dose values from plants with at least 9 of 12 months of data. Ozone
dose values are taken from TUXCTSUM from the AUX2 database.
SELECT TUXPLTMON.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXPLTMON.WTP DIS, Avg(TUXCTSUM AUX2.TO3DOSE) AS AvgOfTOSDOSE,
Avg([to3dose])-Min([to3dose]) AS [Low Error], Max([to3dose])-Avg([to3dose]) AS [High Error],
Count(TUXCTSUM_AUX2.TO3DOSE) AS CountOfTO3DOSE FROM (TUXPLTMON INNER JOIN
TUXCTSUM AUX2 ON (TUXPLTMON.SAMP PER = TUXCTSUM AUX2.SAMP PER) AND
(TUXPLTMON.ICRWTPID = TUXCTSUM AUX2.ICRWTPID) AND (TUXPLTMON.ICRPWSID =
TUXCTSUM_AUX2.ICRPWSID)) INNER JOIN [Plant Source Type, Last 12 Months] ON
(TUXCTSUM_AUX2.ICRWTPID = [Plant Source Type, Last 12 Months].ICRWTPID) AND
(TUXCTSUM_AUX2.ICRPWSID = [Plant Source Type, Last 12 Months].ICRPWSID) WHERE
(((TUXPLTMON.SAMP_PER)>=7) AND ((TUXPLTMON.ICRWTPID)>99)) GROUP BY
TUXPLTMON.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXPLTMON.WTP_DIS HAVING ((([Plant Source Type, Last 12 Months].[Derived Source Type]) Is
Not Null) AND ((TUXPLTMON.WTP_DIS)="O3") AND
((Count(TUXCTSUM_AUX2.TO3DOSE))>=9)) ORDER BY Avg(TUXCTSUM_AUX2.TO3DOSE);
Screened TEMP INF
This query extracts all total hardness values at the influent water point from plants with at least 9
of 12 months of data.
SELECT TUXSAMPLE.ICRWTPID, Avg(IIf([temp]=-999,0,[temp])) AS [plant mean Temp], [Plant
Source Type, Last 12 Months].[Derived Source Type], Count(TUXWQP.TEMP) AS CountOfTEMP,
TUXSAMPLE.EVNTNAME FROM ([Plant Source Type, Last 12 Months] INNER JOIN TUXSAMPLE
ON [Plant Source Type, Last 12 Months] .ICRWTPID = TUXSAMPLE.ICRWTPID) INNER JOIN
TUXWQP ON TUXSAMPLE.EVENT ID = TUXWQP.EVENT ID WHERE
(((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY TUXSAMPLE.ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], TUXSAMPLE.EVNTNAME HAVING
(((Avg(IIf([temp]=-999,0,[temp]))) Is Not Null) AND ((Count(TUXWQP.TEMP))>=9) AND
((TUXSAMPLE.EVNTNAME)="influent")) ORDER BY TUXSAMPLE.ICRWTPID;
Screened INF TOC and ALK
This query extracts all paired TOC and alkalinity values at the influent water point from plants
with at least 9 of 12 months of data.
SELECT [Screened ALK INF].[Derived Source Type], TUXSAMPLE.ICRWTPID,
TUXSAMPLE.SAMP_PER, IIf([TOC]=-999,0,[TOC]) AS [INF TOC], IIf([ALK]=-999,0,[ALK]) AS
[INF ALK], TUXSAMPLE.EVNTNAME FROM ([Screened ALK INF] INNER JOIN TUXSAMPLE
ON [Screened ALK INF] .ICRWTPID = TUXSAMPLE.ICRWTPID) INNER JOIN TUXWQP ON
TUXSAMPLE.EVENT ID = TUXWQP.EVENT ID WHERE (((TUXSAMPLE.SAMP PER)>=7) AND
((IIf([TOC]=-999,0,[TOC])) Is Not Null) AND ((IIf([ALK]=-999,0,[ALK])) Is Not Null) AND
((TUXSAMPLE.EVNTNAME)="Influent"));
Occurrence Assessment for the Final Stage 2 DBPR B-10 December 2005
-------
ScreenedTOC INF
This query extracts all TOC values at the influent water point from plants with at least 9 of 12
months of data.
SELECT TUXSAMPLE.ICRWTPID, Avg(IIf([toc]=-999,0,[toc])) AS [plant mean TOC], [Plant Source
Type, Last 12 Months].[Derived Source Type], Count(TUXWQP.TOC) AS CountOfTOC,
TUXSAMPLE.EVNTNAME FROM ([Plant Source Type, Last 12 Months] INNER JOIN TUXSAMPLE
ON [Plant Source Type, Last 12 Months] .ICRWTPID = TUXSAMPLE.ICRWTPID) INNER JOIN
TUXWQP ON TUXSAMPLE.EVENT ID = TUXWQP.EVENT ID WHERE
(((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY TUXSAMPLE.ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], TUXSAMPLE.EVNTNAME HAVING
(((Avg(IIf([toc]=-999,0,[toc]))) Is Not Null) AND ((Count(TUXWQP.TOC))>=9) AND
((TUXSAMPLE.EVNTNAME)="influent")) ORDER BY TUXSAMPLE.ICRWTPID;
Screened UV-254 INF
This query extracts all UV-254 values at the influent water point from plants with at least 9 of 12
months of data.
SELECT TUXSAMPLE.ICRWTPID, Avg(IIf([uv_254]=-999,0,[uv_254])) AS [plant mean UV_254],
[Plant Source Type, Last 12 Months].[Derived Source Type], Count(TUXWQP.UV_254) AS
CountOfUV_254, TUXSAMPLE.EVNTNAME FROM ([Plant Source Type, Last 12 Months] INNER
JOIN TUXSAMPLE ON [Plant Source Type, Last 12 Months] .ICRWTPID =
TUXSAMPLE.ICRWTPID) INNER JOIN TUXWQP ON TUXSAMPLE.EVENT ID =
TUXWQP.EVENT ID WHERE (((TUXSAMPLE.SAMP_PER)>=7)) GROUP BY
TUXSAMPLE.ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type],
TUXSAMPLE.EVNTNAME HAVING (((Avg(IIf([uv_254]=-999,0,[uv_254]))) Is Not Null) AND
((Count(TUXWQP.UV_254))>=9) AND ((TUXSAMPLE.EVNTNAME)="influent")) ORDER BY
TUXSAMPLE.ICRWTPID:
B.3 Queries for DBF Plant Screening
Last 4 Quarters, average by quarter - TTHM & HAA5
This query builds off the query Last 4 Quarters, by location & quarter - TTHM &HAA5. It
creates a data set with quarterly averages for TTHM and HAAS for quarters with data for three of the four
distribution system locations. See section 1.4.8 and 3.1.3 for further details.
SELECT [Last 4 Quarters, by location & quarter - TTHM & HAA5].ICRPWSID, [Last 4 Quarters, by
location & quarter - TTHM & HAA5].ICRWTPID, [Last 4 Quarters, by location & quarter - TTHM &
HAA5].SAMP_QTR, Count([Last 4 Quarters, by location & quarter - TTHM & HAA5].TTHM1) AS
CountOfTTHMl, Avg([Last 4 Quarters, by location & quarter - TTHM & HAA5].TTHM1) AS
AvgOfTTHMl, Avg([Last 4 Quarters, by location & quarter - TTHM & HAA5].HAA51) AS
AvgOfHAASl, Sum(IIf([tthml]>=40,l,0)) AS [Quarters > 40?], Sum(IIf([tthml]>=60,l,0)) AS [Quarters
> 60?], Sum(IIf([tthml]>=75,l,0)) AS [Quarters > 75?], Sum(IIf([tthml]>=80,l,0)) AS [Quarters > 80?],
Sum(IIf([tthml]>=100,l,0)) AS [Quarters > 100?], Sum(IIf([tthml]>=120,l,0)) AS [Quarters > 120?],
Sum(IIf([haa51]>=30,l,0)) AS [Quarters > 30?], Sum(IIf([haa51]>=45,l,0)) AS [Quarters > 45?],
Sum(IIf([haa51]>=60,l,0)) AS [Quarters > 60-HAA5?], Sum(IIf([haa51]>=75,l,0)) AS [Quarters >
75-HAA5?], Sum(IIf([haa51]>=90,l,0)) AS [Quarters > 90?] FROM [Last 4 Quarters, by location &
Occurrence Assessment for the Final Stage 2 DBPR B-ll December 2005
-------
quarter - TTHM & HAAS] GROUP BY [Last 4 Quarters, by location & quarter - TTHM &
HAA5].ICRPWSID, [Last 4 Quarters, by location & quarter - TTHM & HAA5].ICRWTPID, [Last 4
Quarters, by location & quarter - TTHM & HAA5].SAMP_QTRHAVING (((Count([Last 4 Quarters, by
location & quarter - TTHM & HAA5].TTHM1))>=3)) ORDER BY [Last 4 Quarters, by location &
quarter - TTHM & HAA5].ICRWTPID;
Last 4 Quarters, by location & quarter - TTHM & HAAS
This query is the first step in screening plants with the proper amount of DBP data. It creates a
data set with only plant-months that have both TTHM and HAAS at the four distribution system
locations. See section 1.4.8 and 3.1.3 for further details.
SELECT TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, Avg(IIf([haa5]=-999,0,[haa5])) AS HAA51,
Avg(IIf([tthm]=-999,0,[tthm])) AS TTHM1 FROM (TUXPLTMON INNER JOIN TUXSAMPLE ON
(TUXPLTMON.SAMP PER = TUXSAMPLE.SAMP PER) AND (TUXPLTMON.ICRWTPID =
TUXSAMPLE.ICRWTPID) AND (TUXPLTMON.ICRPWSID = TUXSAMPLE.ICRPWSID)) INNER
JOIN TUXDBP ON TUXSAMPLE.EVENT ID = TUXDBP.EVENT ID WHERE
(((TUXSAMPLE.SAMP_QTR)>=3)) GROUP BY TUXSAMPLE.ICRPWSID,
TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR, TUXSAMPLE.EVNTNAME HAVING
(((TUXSAMPLE.EVNTNAME)="dse" Or (TUXSAMPLE.EVNTNAME)="avg" Or
(TUXSAMPLE.EVNTNAME)="avgl" Or (TUXSAMPLE.EVNTNAME)="avg2" Or
(TUXSAMPLE.EVNTNAME)="max") AND ((Avg(IIf([haa5]=-999,0,[haa5]))) Is Not Null) AND
((Avg(IIf([tthm]=-999,0,[tthm]))) Is Not Null)) ORDER BY TUXSAMPLE.ICRWTPID;
Last 4 Quarters, Plants min 3x3
This query builds off the query Last 4 Quarters, average by quarter - TTHM &HAA5. It creates
a data set with yearly averages for TTHM and HAAS for plants with three of four quarters of data. See
section 1.4.8 and 3.1.3 for further details.
SELECT [Last 4 Quarters, average by quarter - TTHM & HAA5].ICRPWSID, [Last 4 Quarters, average
by quarter - TTHM & HAA5].ICRWTPID, Count([Last 4 Quarters, average by quarter - TTHM &
HAA5].ICRWTPID) AS CountOflCRWTPID FROM [Last 4 Quarters, average by quarter - TTHM &
HAAS] GROUP BY [Last 4 Quarters, average by quarter - TTHM & HAA5].ICRPWSID, [Last 4
Quarters, average by quarter - TTHM & HAA5].ICRWTPID HAVING (((Count([Last 4 Quarters,
average by quarter - TTHM & HAA5].ICRWTPID))>=3)) ORDER BY [Last 4 Quarters, average by
quarter - TTHM & HAA5].ICRWTPID;
Plants min 3x3, average by location - TTHM & HAA5
This query is used to extract the average TTHM and HAAS data by location for plants screened
by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by
location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM &
HAA5].EVNTNAME, Avg([Plants min 3x3, by location & quarter - TTHM & HAA5].TTHM1) AS
AvgOfTTHMl, Avg([Plants min 3x3, by location & quarter - TTHM & HAA5].HAA51) AS
AvgOfHAAS 1, Count([Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID) AS
CountOflCRPWSID, Sum(IIf([tthml]>=40,l,0)) AS [Quarters > 40?], Sum(IIf([tthml]>=60,l,0)) AS
[Quarters > 60?], Sum(IIf([tthml]>=75,l,0)) AS [Quarters > 75?], Sum(IIf([tthml]>=80,l,0)) AS
Occurrence Assessment for the Final Stage 2 DBPR B-12 December 2005
-------
[Quarters > 80?], Sum(IIf([tthml]>=100,l,0)) AS [Quarters > 100?], Sum(IIf([tthml]>=120,l,0)) AS
[Quarters > 120?], Sum(IIf([haa51]>=30,l,0)) AS [Quarters > 30?], Sum(IIf([haa51]>=45,l,0)) AS
[Quarters > 45?], Sum(IIf([haa51]>=60,l,0)) AS [Quarters > 60-HAA5?], Sum(IIf([haa51]>=75,l,0)) AS
[Quarters > 75-HAA5?], Sum(IIf([haa51]>=90,l,0)) AS [Quarters > 90?] FROM [Plants min 3x3, by
location & quarter - TTHM & HAAS] GROUP BY [Plants min 3x3, by location & quarter - TTHM &
HAA5].ICRPWSID, [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRWTPID, [Plants min
3x3, by location & quarter - TTHM & HAA5].EVNTNAME ORDER BY [Plants min 3x3, by location &
quarter - TTHM & HAA5].ICRWTPID;
Plants min 3x3, average by quarter - Other DBFs
This query is used to extract the average TOX, HAN4, CH, CP, DCP, and TCP data by quarter
for plants screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - Other DBPs].ICRPWSID, [Plants min 3x3, by location
& quarter - Other DBPs].ICRWTPID, [Plants min 3x3, by location & quarter - Other
DBPs].SAMP QTR, Avg([Plants min 3x3, by location & quarter - Other DBPs].TOXl) AS
AvgOfTOXl, Avg([Plants min 3x3, by location & quarter - Other DBPs].HAN4_l) AS AvgOfHAN4_l,
Avg([Plants min 3x3, by location & quarter - Other DBPs].CHl) AS AvgOfCHl, Avg([Plants min 3x3,
by location & quarter - Other DBPs].CPl) AS AvgOfCPl, Avg([Plants min 3x3, by location & quarter -
Other DBPs].DCPl) AS AvgOfDCPl, Avg([Plants min 3x3, by location & quarter - Other DBPs].TCPl)
AS AvgOfTCPl, Count([Plants min 3x3, by location & quarter - Other DBPs].ICRPWSID) AS
CountOflCRPWSID FROM [Plants min 3x3, by location & quarter - Other DBFs] WHERE ((([Plants
min 3x3, by location & quarter - Other DBPs].EVNTNAME)o"finish")) GROUP BY [Plants min 3x3,
by location & quarter - Other DBPs].ICRPWSID, [Plants min 3x3, by location & quarter - Other
DBPs].ICRWTPID, [Plants min 3x3, by location & quarter - Other DBPs].SAMP_QTR ORDER BY
[Plants min 3x3, by location & quarter - Other DBPs].ICRWTPID;
Plants min 3x3, average by quarter - TTHM & HAA5
This query is used to extract the average TTHM and HAAS data by quarter for plants screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by
location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM &
HAA5].SAMP_QTR, Avg([Plants min 3x3, by location & quarter - TTHM & HAA5].TTHM1) AS
AvgOfTTHMl, Avg([Plants min 3x3, by location & quarter - TTHM & HAA5].HAA51) AS
AvgOfHAAS 1, Count([Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID) AS
CountOflCRPWSID, Sum(IIf([tthml]>=40,l,0)) AS [Quarters > 40?], Sum(IIf([tthml]>=60,l,0)) AS
[Quarters > 60?], Sum(IIf([tthml]>=75,l,0)) AS [Quarters > 75?], Sum(IIf([tthml]>=80,l,0)) AS
[Quarters > 80?], Sum(IIf([tthml]>=100,l,0)) AS [Quarters > 100?], Sum(IIf([tthml]>=120,l,0)) AS
[Quarters > 120?], Sum(IIf([haa51]>=30,l,0)) AS [Quarters > 30?], Sum(IIf([haa51]>=45,l,0)) AS
[Quarters > 45?], Sum(IIf([haa51]>=60,l,0)) AS [Quarters > 60-HAA5?], Sum(IIf([haa51]>=75,l,0)) AS
[Quarters > 75-HAA5?], Sum(IIf([haa51]>=90,l,0)) AS [Quarters > 90?] FROM [Plants min 3x3, by
location & quarter - TTHM & HAA5] WHERE ((([Plants min 3x3, by location & quarter - TTHM &
HAA5].EVNTNAME)o"finish")) GROUP BY [Plants min 3x3, by location & quarter - TTHM &
HAA5].ICRPWSID, [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRWTPID, [Plants min
3x3, by location & quarter - TTHM & HAA5].SAMP_QTR ORDER BY [Plants min 3x3, by location &
quarter - TTHM & HAA5].ICRWTPID;
Occurrence Assessment for the Final Stage 2 DBPR B-13 December 2005
-------
Plants min 3x3, by location & quarter - Other DBFs
This query is used to extract the plant-month TOX, HAN4, CH, CP, DCP, and TCP data by
quarter and location. Plants screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and
3.1.3.
SELECT TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, Avg(IIf([tox]=-999,0,[tox])) AS TOX1, Avg(IIf([han4]=-999,0,[han4]))
AS HAN4_1, Avg(IIf([ch]=-999,0,[ch])) AS CHI, Avg(IIf([cp]=-999,0,[cp])) AS CP1,
Avg(IIf([dcp_hk]=-999,0,[dcp_hk])) AS DCP1, Avg(IIf([tcp_hk]=-999,0,[tcp_hk])) AS TCP1, [Plant
Source Type, Last 12 Months].[Derived Source Type] FROM (([Plant Source Type, Last 12 Months]
INNER JOIN ([Last 4 Quarters, Plants min 3x3] INNER JOIN TUXSAMPLE ON [Last 4 Quarters,
Plants min 3x3].ICRWTPID = TUXSAMPLE.ICRWTPID) ON [Plant Source Type, Last 12
Months] .ICRWTPID = TUXSAMPLE.ICRWTPID) INNER JOIN TUXDBP ON
TUXSAMPLE.EVENT ID = TUXDBP.EVENT ID) INNER JOIN TUXWQP ON
TUXSAMPLE.EVENTJD = TUXWQP.EVENT ID WHERE (((TUXSAMPLE.SAMP_QTR)>=3))
GROUP BY TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type] HAVING
(((TUXSAMPLE.EVNTNAME)="finish" Or (TUXSAMPLE.EVNTNAME)="dse" Or
(TUXSAMPLE.EVNTNAME)="avg" Or (TUXSAMPLE.EVNTNAME)="avgl" Or
(TUXSAMPLE.EVNTNAME)="avg2" Or (TUXSAMPLE.EVNTNAME)="max")) ORDER BY
TUXSAMPLE.ICRWTPID;
Plants min 3x3, by location & quarter - TTHM & HAA5
This query is used to extract the plant-month TTHM and HAAS data by quarter and location.
Plants screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type],
Avg(IIf([haa5]=-999,0,[haa5])) AS HAA51, Avg(IIf([tthm]=-999,0,[tthm])) AS TTHM1
FROM ([Last 4 Quarters, Plants min 3x3] INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source
Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12
Months].ICRWTPID) ON [Last 4 Quarters, Plants min 3x3].ICRWTPID = TUXSAMPLE.ICRWTPID)
INNER JOIN TUXDBP ON TUXSAMPLE.EVENTJD = TUXDBP.EVENT ID WHERE
(((TUXSAMPLE.SAMP_QTR)>=3)) GROUP BY TUXSAMPLE.ICRPWSID,
TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR, TUXSAMPLE.EVNTNAME, [Plant Source
Type, Last 12 Months].[Derived Source Type] HAVING (((TUXSAMPLE.EVNTNAME)="finish" Or
(TUXSAMPLE.EVNTNAME)="dse" Or (TUXSAMPLE.EVNTNAME)="avg" Or
(TUXSAMPLE.EVNTNAME)="avgl" Or (TUXSAMPLE.EVNTNAME)="avg2" Or
(TUXSAMPLE.EVNTNAME)="max") AND ((Avg(IIf([haa5]=-999,0,[haa5]))) Is Not Null) AND
((Avg(IIf([tthm]=-999,0,[tthm]))) Is Not Null)) ORDER BY TUXSAMPLE.ICRWTPID;
Plants min 3x3, max by quarter - TTHM & HAAS
This query is used to extract the RAA for TTHM and HAAS. Plants are screened by whether
they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by
location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM &
HAA5].SAMP_QTR, Max([Plants min 3x3, by location & quarter - TTHM & HAA5].TTHM1) AS
Occurrence Assessment for the Final Stage 2 DBPR B-14 December 2005
-------
MaxOfTTHMl, Max([Plants min 3x3, by location & quarter - TTHM & HAA5].HAA51) AS
MaxOfHAASl, Count([Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID) AS
CountOflCRPWSID FROM [Plants min 3x3, by location & quarter - TTHM & HAAS] GROUP BY
[Plants min 3x3, by location & quarter - TTHM & HAA5].ICRPWSID, [Plants min 3x3, by location &
quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM &
HAA5].SAMP_QTR ORDER BY [Plants min 3x3, by location & quarter - TTHM &
HAA5].ICRWTPID;
Plants min 3x3, Max-Min - TTHM & HAA5
This query is used to extract the maximum minus the minimum value for TTHM and HAAS.
Plants are screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by
location & quarter - TTHM & HAAS].[Derived Source Type], Max([tthml])-Min([tthml]) AS [Max-Min
TTHM], Max([haa51])-Min([haa51]) AS [Max-Min HAAS]
FROM [Plants min 3x3, by location & quarter - TTHM & HAAS]
GROUP BY [Plants min 3x3, by location & quarter - TTHM & HAA5].ICRWTPID, [Plants min 3x3, by
location & quarter - TTHM & HAAS].[Derived Source Type];
B.4 Queries for Appendix A
Plants min 3x3, by location & quarter - HAA5 Speciation
This query is used to extract the plant-month HAAS specie data by quarter and location. Plants
screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type],
Avg(IIf([mcaa]=-999,0,[mcaa])) AS MCAA1, Avg(IIf([dcaa]=-999,0,[dcaa])) AS DCAA1,
Avg(IIf([tcaa]=-999,0,[tcaa])) AS TCAA1, Avg(IIf([mbaa]=-999,0,[mbaa])) AS MBAA1,
Avg(IIf([dbaa]=-999,0,[dbaa])) AS DBAA1 FROM ([Last 4 Quarters, Plants min 3x3] INNER JOIN
(TUXSAMPLE INNER JOIN [Plant Source Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID =
[Plant Source Type, Last 12 Months].ICRWTPID) ON [Last 4 Quarters, Plants min 3x3].ICRWTPID =
TUXSAMPLE.ICRWTPID) INNER JOIN TUXDBP ON TUXSAMPLE.EVENT ID =
TUXDBP.EVENT ID WHERE (((TUXSAMPLE.SAMP QTR)>=3)) GROUP BY
TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type] HAVING
(((TUXSAMPLE.EVNTNAME)="finish" Or (TUXSAMPLE.EVNTNAME)="dse" Or
(TUXSAMPLE.EVNTNAME)="avg" Or (TUXSAMPLE.EVNTNAME)="avgl" Or
(TUXSAMPLE.EVNTNAME)="avg2" Or (TUXSAMPLE.EVNTNAME)="max")) ORDER BY
TUXSAMPLE.ICRWTPID;
Plants min 3x3, by location & quarter - TTHM Speciation
This query is used to extract the plant-month TTHM specie data by quarter and location. Plants
screened by whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT TUXSAMPLE.ICRPWSID, TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR,
TUXSAMPLE.EVNTNAME, [Plant Source Type, Last 12 Months].[Derived Source Type],
Occurrence Assessment for the Final Stage 2 DBPR B-15 December 2005
-------
Avg(IIf([bdcm]=-999,0,[bdcm])) AS BDCM1, Avg(IIf([dbcm]=-999,0,[dbcm])) AS DBCM1,
Avg(IIf([chcl3]=-999,0,[chcl3])) AS CHCL31, Avg(IIf([chbr3]=-999,0,[chbr3])) AS CHBR31
FROM ([Last 4 Quarters, Plants min 3x3] INNER JOIN (TUXSAMPLE INNER JOIN [Plant Source
Type, Last 12 Months] ON TUXSAMPLE.ICRWTPID = [Plant Source Type, Last 12
Months].ICRWTPID) ON [Last 4 Quarters, Plants min 3x3].ICRWTPID = TUXSAMPLE.ICRWTPID)
INNER JOIN TUXDBP ON TUXSAMPLE.EVENT ID = TUXDBP.EVENT ID WHERE
(((TUXSAMPLE.SAMP_QTR)>=3)) GROUP BY TUXSAMPLE.ICRPWSID,
TUXSAMPLE.ICRWTPID, TUXSAMPLE.SAMP QTR, TUXSAMPLE.EVNTNAME, [Plant Source
Type, Last 12 Months].[Derived Source Type] HAVING (((TUXSAMPLE.EVNTNAME)="finish" Or
(TUXSAMPLE.EVNTNAME)="dse" Or (TUXSAMPLE.EVNTNAME)="avg" Or
(TUXSAMPLE.EVNTNAME)="avgl" Or (TUXSAMPLE.EVNTNAME)="avg2" Or
(TUXSAMPLE.EVNTNAME)="max") AND ((Avg(IIf([bdcm]=-999,0,[bdcm]))) Is Not Null) AND
((Avg(IIf([dbcm]=-999,0,[dbcm]))) Is Not Null) AND ((Avg(IIf([chcl3]=-999,0,[chcl3]))) Is Not Null)
AND ((Avg(IIf([chbr3]=-999,0,[chbr3]))) Is Not Null)) ORDER BY TUXSAMPLE.ICRWTPID;
Plants min 3x3, average by location - HAA5 Speciation
This query is used to extract the average HAAS specie data by location for plants screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].EVNTNAME, Avg([Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)].MCAAl) AS AvgOfMCAAl, Avg([Plants min 3x3,
by location & quarter - TTHM & HAAS (w species)].DCAAl) AS AvgOfDCAAl, Avg([Plants min 3x3,
by location & quarter - TTHM & HAAS (w species)] .TCAA1) AS AvgOfTCAAl, Avg( [Plants min 3x3,
by location & quarter - TTHM & HAAS (w species)].MBAA1) AS AvgOfMBAAl, Avg([Plants min
3x3, by location & quarter - TTHM & HAAS (w species)] .DBAA1) AS AvgOfDBAAl, Count( [Plants
min 3x3, by location & quarter - TTHM & HAAS (w species)].ICRPWSID) AS CountOflCRPWSID
FROM [Plants min 3x3, by location & quarter - TTHM & HAAS (w species)] GROUP BY [Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)] .EVNTNAME ORDER BY [Plants min 3x3, by location & quarter
- TTHM & HAAS (w species)].ICRWTPID;
Plants min 3x3, average by location - TTHM Speciation
This query is used to extract the average TTHM specie data by location for plants screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].EVNTNAME, Avg([Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)].BDCMl) AS AvgOfBDCMl, Avg([Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)].DBCMl) AS AvgOfDBCMl, Avg([Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)].CHCL31) AS AvgOfCHCL31, Avg([Plants min 3x3,
by location & quarter - TTHM & HAAS (w species)].CHBR31) AS AvgOfCHBR31, Count([Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].ICRPWSID) AS CountOflCRPWSID FROM
[Plants min 3x3, by location & quarter - TTHM & HAAS (w species)] GROUP BY [Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)] .EVNTNAME ORDER BY [Plants min 3x3, by location & quarter - TTHM
& HAAS (w species)].ICRWTPID;
Occurrence Assessment for the Final Stage 2 DBPR B-16 December 2005
-------
Plants min 3x3, average by quarter - HAA5 Speciation
This query is used to extract the average HAAS specie data by quarter for plants screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAAS (w species)].ICRPWSID, [Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].SAMP_QTR, Avg([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].MCAAl) AS AvgOfMCAAl, Avg([Plants min 3x3, by location & quarter
- TTHM & HAAS (w species)].DCAAl) AS AvgOfDCAAl, Avg([Plants min 3x3, by location & quarter
- TTHM & HAAS (w species)].TCAA1) AS AvgOfTCAAl, Avg([Plants min 3x3, by location & quarter
- TTHM & HAAS (w species)].MBAA1) AS AvgOfMBAAl, Avg([Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].DBAAl) AS AvgOfDBAAl, Count([Plants min 3x3, by location
& quarter - TTHM & HAAS (w species)].ICRPWSID) AS CountOflCRPWSID FROM [Plants min 3x3,
by location & quarter - TTHM & HAAS (w species)] WHERE ((([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].EVNTNAME)o"fmish")) GROUP BY [Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].ICRPWSID, [Plants min 3x3, by location & quarter - TTHM &
HAAS (w species)] .ICRWTPID, [Plants min 3x3, by location & quarter - TTHM & HAAS (w
species)].SAMP_QTR ORDER BY [Plants min 3x3, by location & quarter - TTHM & HAAS (w
species)].ICRWTPID;
Plants min 3x3, average by quarter - TTHM Speciation
This query is used to extract the average TTHM specie data by quarter for plants screened by
whether they meet the 3x3 criteria defined in sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, by location & quarter - TTHM & HAAS (w species)].ICRPWSID, [Plants min
3x3, by location & quarter - TTHM & HAAS (w species)].ICRWTPID, [Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].SAMP_QTR, Avg([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].BDCM1) AS AvgOfBDCMl, Avg([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].DBCM1) AS AvgOfDBCMl, Avg([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].CHCL31) AS AvgOfCHCL31, Avg([Plants min 3x3, by location & quarter
- TTHM & HAAS (w species)].CHBR31) AS AvgOfCHBR31, Count([Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].ICRPWSID) AS CountOflCRPWSID FROM [Plants min 3x3, by
location & quarter - TTHM & HAAS (w species)] WHERE ((([Plants min 3x3, by location & quarter -
TTHM & HAAS (w species)].EVNTNAME)o"finish")) GROUP BY [Plants min 3x3, by location &
quarter - TTHM & HAAS (w species)].ICRPWSID, [Plants min 3x3, by location & quarter - TTHM &
HAAS (w species)].ICRWTPID, [Plants min 3x3, by location & quarter - TTHM & HAAS (w
species)].SAMP_QTR ORDER BY [Plants min 3x3, by location & quarter - TTHM & HAAS (w
species)].ICRWTPID;
Plants min 3x3, RAA - HAA5 Speciation
This query is used to extract the RAA for AVG1, AVG2, DSE, and MAX distribution system
sampling locations by HAAS specie. Plants are screened by whether they meet the 3x3 criteria defined in
sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, average by quarter - HAAS Speciation].ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], Avg([Plants min 3x3, average by quarter - HAAS
Speciation].AvgOfMCAAl) AS AvgOfAvgOfMCAAl, Avg([Plants min 3x3, average by quarter -
HAAS Speciation].AvgOfDCAAl) AS AvgOfAvgOfDCAAl, Avg([Plants min 3x3, average by quarter -
Occurrence Assessment for the Final Stage 2 DBPR B-l 7 December 2005
-------
HAAS Speciation].AvgOfTCAAl) AS AvgOfAvgOfTCAAl, Avg([Plants min 3x3, average by quarter -
HAAS Speciation].AvgOfMBAAl) AS AvgOfAvgOfMBAAl, Avg([Plants min 3x3, average by quarter
- HAAS Speciation].AvgOfDBAAl) AS AvgOfAvgOfDBAAl FROM [Plants min 3x3, average by
quarter - HAAS Speciation] INNER JOIN [Plant Source Type, Last 12 Months] ON [Plants min 3x3,
average by quarter - HAAS Speciation] .ICRWTPID = [Plant Source Type, Last 12 Months] .ICRWTPID
GROUP BY [Plants min 3x3, average by quarter - HAAS Speciation].ICRWTPID, [Plant Source Type,
Last 12 Months].[Derived Source Type];
Plants min 3x3, RAA - TTHM Speciation
This query is used to extract the RAA for AVG1, AVG2, DSE, and MAX distribution system
sampling locations by TTHM specie. Plants are screened by whether they meet the 3x3 criteria defined in
sections 1.4.8 and 3.1.3.
SELECT [Plants min 3x3, average by quarter - TTHM Speciation] .ICRWTPID, [Plant Source Type, Last
12 Months].[Derived Source Type], Avg([Plants min 3x3, average by quarter - TTHM
Speciation].AvgOfBDCMl) AS AvgOfAvgOfBDCMl, Avg([Plants min 3x3, average by quarter -
TTHM Speciation].AvgOfDBCMl) AS AvgOfAvgOfDBCMl, Avg([Plants min 3x3, average by quarter
- TTHM Speciation].AvgOfCHCL31) AS AvgO£AvgOfCHCL31, Avg([Plants min 3x3, average by
quarter - TTHM Speciation] .AvgOfCHBR31) AS AvgO£AvgOfCHBR31 FROM [Plant Source Type,
Last 12 Months] INNER JOIN [Plants min 3x3, average by quarter - TTHM Speciation] ON [Plant
Source Type, Last 12 Months] .ICRWTPID = [Plants min 3x3, average by quarter - TTHM
Speciation].ICRWTPID GROUP BY [Plants min 3x3, average by quarter - TTHM
Speciation].ICRWTPID, [Plant Source Type, Last 12 Months].[Derived Source Type];
Occurrence Assessment for the Final Stage 2 DBPR B-18 December 2005
-------
Appendix C
Assessment of Data Quality Objectives
-------
-------
Exhibit C.1 Assessment of Data Quality Objectives for Existing Data Used in the Stage 2 DBPR Occurrence
Document
Existing Data Source
1. Information
Collection Rule
(ICR)
2. ICR Supplemental
Survey
3. National Rural
Water Association
(NRWA) Survey
4. Water Utility Survey
(WATERASTATS
database)
5. Ground Water
Supply Survey
6. State Data
Use in Stage 2 DBPR Occurrence Document
Used to characterize occurrence of disinfectants, disinfection
byproducts (DBFs), and DBP precursors (e.g., total organic
carbon[TOC]) in large surface water (SW) and ground water (GW)
systems.
Used to compare TOC occurrence in small, medium and large SW
systems.
Used to characterize operational characteristics, disinfection practices,
DBP occurrence and occurrence of DBP precursors (e.g., TOC) for
small SW systems. DBP and DBP precursor data were compared to
that of large systems. Used to assess variability in TTHM and HAAS
occurrence in distribution systems of small SW systems.
Used to compare operational characteristics, disinfection practices, DBP
occurrence, and DBP precursor occurrence of medium and large SW
systems and medium and large ground water GW systems
Used to compare TOC occurrence between small, medium, and large
GW systems
Used to compare TTHM occurrence on small GW systems to
occurrence in large GW systems.
Level 1
2
1
1
1
1
1
QA Plan? 2
Yes
Yes
Yes
Yes
Yes
No
Peer Reviewed?
Yes
Yes
No
Yes
No
No
Notes:
1 Level 1 data are those data that provide background information or context for a particular assessment or discussion, but are not deemed to be influential in
EPA's decision-making process. Level 2 data are those data that are deemed to be highly important or influential in EPA's decision-making process. Refer to the
Work Assignment 1-05 Project-Specific Supplement to the Programmatic Quality Assurance Project Plan (USEPA 2003d) for additional information on level
designations.
2 See Sections 1.4 and 1.5 for a description of QA plans and/or peer review processes for each existing data source shown.
Occurrence Assessment for the Final Stage 2 DBPR
C-l
December 2005
------- |