v»EPA
United States
Environmental Protection
Agency
Online Water Quality Monitoring in
Distribution Systems
For Water Quality Surveillance and Response Systems
Office of Water (MC 140)
EPA 817-B-18-001
April 2018
-------
Disclaimer
The Water Security Division of the Office of Ground Water and Drinking Water of the EPA has reviewed
and approved this document for publication. This document does not impose legally binding requirements
on any party. The information in this document is intended solely to recommend or suggest and does not
imply any requirements. Neither the United States Government nor any of its employees, contractors, or
their employees makes any warranty, expressed or implied, or assumes any legal liability or responsibility
for any third party's use of any information, product, or process discussed in this document, or represents
that its use by such party would not infringe on privately owned rights. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
Questions concerning this document should be addressed to WO SRS@,epa.gov or the following
contacts:
Matt Umberg
EPA Water Security Division
26 West Martin Luther King Drive
Mail Code 140
Cincinnati, OH 45268
(513) 569-7357
Umberg. Matt@epa. gov
or
Steve Allgeier
EPA Water Security Division
26 West Martin Luther King Drive
Mail Code 140
Cincinnati, OH 45268
(513) 569-7131
Allgeier. Steve (ale pa. gov
-------
Acknowledgem ents
The document was developed by the EPA Water Security Division, with additional support provided
under EPA contract EP-C-15-012. The following individuals contributed to the development of this
document:
• Steve Allgeier, EPA, Water Security Division
• Jennifer Liggett, CH2M
• Christopher Macintosh, CH2M
• Ralph Rogers, Philadelphia Water Department
• Kenneth Thompson, CH2M
• Katie Umberg, CSRA
• Matt Umberg, EPA, Water Security Division
Peer review of this document was provided by the following individuals:
• Matthew Alexander, EPA, Standards and Risk Management Division
• Michael Finn, EPA, Drinking Water Protection Division
• John Hall, EPA, National Homeland Security Research Center
• Jonathan Leung, Los Angeles Department of Water and Power
• David Nelson, City of Fort Worth Water Department
• Christine Owen, Hazen and Sawyer (Tampa Bay Water at time of review)
• Deborah Vacs Renwick, EPA, Standards and Risk Management Division
• Steve Rhode, Massachusetts Water Resources Authority
• Kenneth Rotert, EPA, Standards and Risk Management Division
ii
-------
Table of Contents
List of Figures iv
List of Tables v
Abbreviations vi
Section 1: Introduction 1
1.1 Overview of Online Water Quality Monitoring in Distribution Systems 3
1.2 Purpose and Overview of this Document 3
Section2: Framework for Designing Online Monitoring Systems 5
2.1 Establish Design Goals 5
2.1.1 Monitor for Contamination Incidents 5
2.1.2 Optimize Distribution System Water Quality 6
2.2 Establish Performance Objectives 6
2.3 Review Distribution System Resources 7
2.4 Design the Online Monitoring System 9
Section 3: Monitoring Locations 11
3.1 Common Monitoring Locations 11
3.2 Tools to Support Selection of Monitoring Locations 12
3.3 Installation Sites 13
Section4: Water Quality Parameters 15
Section 5: Monitoring Stations 19
5.1 Instrumentation 20
5.2 Computing Element 21
5.3 Communications 21
5.4 Power Supply and Distribution 21
5.5 Accessories 21
5.6 Station Structure 22
Section 6: Information Management and Analysis 23
6.1 Data Validation 23
6.2 Anomaly Detection 24
6.3 Information Management System Architecture 26
6.4 Information Management System Requirements 31
Section 7: Alert Investigation Procedure 34
7.1 Developing an Effective Alert Investigation Procedure 34
7.2 Developing Investigation Tools 39
7.3 Preparing for Real-time Alert Investigations 41
Section 8: Preliminary Design 44
Section 9: Example Applications 46
9.1 Monitoring for Contamination Incidents 46
9.2 Monitoring for Red Water and Particulate Matter Incidents 48
9.3 Chlorine Residual Management 51
9.4 Verify Effectiveness of Nitrification Control 53
9.5 Verify Effectiveness of Corrosion Control 55
9.6 Source Tracking 58
9.7 Summary of Online Monitoring System Applications 60
Section 10: CaseStudies 62
10.1 Philadelphia Water Department 62
10.2 City of Dayton Water Department 64
10.3 Mohawk Valley Water Authority 65
Section 11: Lessons Learned 67
Resources 69
References 73
Glossary 75
iii
-------
List of Figures
Figure 1-1. Incorporation of Online Water Quality Monitoring into a Water Quality Surveillance
and Response System 2
Figure2-1 . Online Monitoring System Implementation Framework 5
Figure2-2. Online Monitoring System Design Elements 9
Figure 5-1. FunctionalBlock Diagram of a Monitoring Station 20
Figure 6-1. Example of Threshold Analysis 25
Figure 6-2. Example of Information Management as an Extension to an Existing SCADA System
Architecture 27
Figure 6-3. Example of a Dedicated Information Management System 29
Figure 6-4. Example of a Dashboard Display 30
Figure 7-1. Example Alert Investigation Process Diagram 36
Figure 7-2. Example Spreadsheet for Documenting Alert Investigations 41
Figure 9-1. Data Following Addition of Contaminants into Drinking Water 48
Figure 9-2. Data During a Particulate Matter Incident 50
Figure 9-3. SpectralFingerprints Before and During a Particulate Matter Incident 51
Figure 9-4. Data During an Occurrence of Chlorine ResidualDeca y in a Reservoir 53
Figure 9-5. Data During a Nitrification Incident 55
Figure9-6. Data Affecting Corrosion Control Treatment 58
Figure9-7. Data During Source Water Changes 59
Figure 10-1. Philadelphia Water Department Rapid Deployment Station 63
Figure 10-2. Philadelphia Water Department SRS Dashboard 63
Figure 10-3. City of Dayton Water Department Monitoring Station Installation 64
Figure 10-4. Data on a Mohawk Valley Water Authority SCADA Screen 66
iv
-------
List of Tables
Table 4-1. Overview ofCore Water Quality Parameters 15
Table4-2. Overview of Additional Water Quality Parameters 16
Table 6-1. SupplementalDatastreams that can beUsedto Validate Data 24
Table 6-2. Examples of Information Management System Functional Requirements 32
Table 6-3. Examples of Information Management System Technical Requirements 33
Table 7-1. Common High-level Causes of Alerts 34
Table 7-2. Additional Information Sources forUse in an Investigation 37
Table 7-3. Example of Generic Roles and Responsibilities for Alert Investigations 39
Table 7-4. Example of Alert Categories 40
Table 9-1. Water Quality Parameters for Broad-Spectrum Monitoring for Contamination 47
Table 9-2. Water Quality Parameters for Red Water and Particulate Matter Incident Monitoring.. 49
Table9-3. Water Quality Parameters for Disinfectant Residual Management 52
Table9-4. Water Quality Parameters for Verifying the Effectiveness of Nitrification Control 54
Table 9-5. Water Quality Parameters for Verifying Effectiveness of Corrosion Control 57
Table 9-6. Water Quality Parameters for Source Tracking 59
Table 9-7. Summary of Monitoring Locations and Water Quality Parameters to Support Example
Applications 60
v
-------
Abbreviations
ADS
Anomaly Detection System
APHA
American Public Health Association
AWWA
American Water Works Association
CCT
Corrosion Control Treatment
CH2M
CH2M HILL, Inc.
CM
Consequence Management
DBP
Disinfection Byproduct
DO
Dissolved Oxygen
DOC
Dissolved Organic Carbon
EPA
United States Environmental Protection Agency
GIS
Geographic Information System
HOC1
Hypochlorous solution
IT
Information Technology
LIMS
Laboratory Information Management System
LSI
Langelier Saturation Index
mg/L
Milligrams per Liter
MGD
Million Gallons per Day
MVWA
Mohawk Valley Water Authority
NEMA
National Electrical Manufacturers Association
NH3
Ammonia
NO 2
Nitrite
NO 3
Nitrate
ORP
Oxidation-Reduction Potential
OWQM
Online Water Quality Monitoring
OWQM-DS
Online Water Quality Monitoring in Distribution Systems
PWD
Philadelphia Water Department
RTCR
Revised T otal Coliform Rule
S&A
Sampling and Analysis
SCADA
Supervisory Control and Data Acquisition
SRS
Water Quality Surveillance and Response System
SUVA
Specific Ultraviolet Absorbance
SWTR
Surface Water Treatment Rule
TEVA-SPOT
Threat Ensemble Vulnerability Assessment and Sensor Placement Optimization Tool
TOC
Total Organic Carbon
uv
Ultraviolet
UV-254
UV light absorbance at 254 nanometers
-------
Online Water Quality Monitoring for Distribution Systems
Section 1: Introduction
A drinking water distribution system1 includes all infrastructure needed to convey treated, or potable,
water to service connections throughout a service area. Online Water Quality Monitoring in Distribution
Systems (OWQM-DS), as defined in this document, involves the use of online water quality instruments
for real-time measurement of water quality atone or more locations in a distribution system. OWQM-DS
enables drinking water utilities to more efficiently manage distribution system operations by detecting
changes in water quality as they occur, facilitating a timely and effective response.
OWQM-DS can be implemented as a standalone monitoring program, or it can be incorporated into a
Water Quality Surveillance and Response System (SRS). An SRS is a framework developed by the
United States Environmental Protection Agency (EPA) to support monitoring and management of water
quality from source to tap. The system consists of one or more components that provide information to
guide drinking water utility operations and enhance a utility's ability to quickly detect and respond to
water quality changes. The SRS Primer provides an overview of SRSs. The guidance provided in this
document treats OWQM-DS as an implementation of the Online Water Quality Monitoring (OWQM)
component within an SRS as described in the OWQM Primer. Figure 1-1 illustrates the way in which
OWQM can be integrated into an SRS.
-|
Words in bold italic font are terms defined in the Glossary at the end of this document.
1
-------
Online Water Quality Monitoring for Distribution Systems
SURVEILLANCE
RESPONSE
©
If unusual
water quality is
detected or
reported,
an alert is
generated and
investigated
rr
¦
-
Can distribution
NO
system
contamination
[
be ruled out?
r i
YES
Take corrective
action if
necessary,
then resume
routine
surveillance
T:)
H J ~
Customer
Complaint
Surveillance
Online Water
Dual it/
Monitoring
Enhanced
Security
Monitoring
Public Health
Surveillance
Data Access, Visualization, and Analysis
Sampling &
Analysis
-
Consequence
Management
Figure 1-1, Incorporation of Online Water Quality Monitoring into a Water Quality Surveillance and
Response System
The design of an SRS is flexible and can include any combination of components shown in Figure 1-1.
However, it is recommended that all SRS designs include at least one surveillance component and basic
capabilities for Sampling and Analysis (S&A) and Consequence Management (CM). S&A is important
because the surv eillance components of an SRS typically provide only a general indication of a potential
water quality problem; S&A establishes the capability to confirm or rule out specific contaminants or
contaminant classes. CM establishes relationships with response partners and procedures for responding
to serious water quality problems such as contamination.
2
-------
Online Water Quality Monitoring for Distribution Systems
1.1 Overview of Online Water Quality Monitoring in Distribution
Systems
Once drinking water has left a treatment plant, the water quality changes in the distribution system due to
a variety of factors:
• Changes in, and mixing of, water that is supplied to a monitoring location by multiple water
sources or storage facilities at a given time
• Decay of disinfectant residual as water flows through the system
• Chemical reactions and biological processes (e.g., corrosion, nitrification, regrowth)
• Introduction of foreign substances through cross-connections, backflow, infiltration, or
contamination
• Hydraulic upsets such as pressure surges or pressure transients
OWQM-DS provides information that can be used to detect these changes in water quality and determine
their causes. This information can help utilities achieve the design goals described in Section 2.1.
OWQM-DS has become more prevalent in recent years due to the improved capabilities and reduced cost
of required technologies, including sensors, power supplies, and communication options. The resulting
data that is produced is more accurate, more timely, and more affordable.
1.2 Purpose and Overview of this Document
This document provides guidance on the design of an OWQM-DS system that is based on best practices
and lessons learned from existing systems. It introduces key concepts, provides examples, and references
additional resources for guidance on specific technical elements of OWQM-DS.
This document is primarily intended for use by water sector
professionals involved with managing water quality and
operations in drinking water distribution systems.
The remaining sections of this document cover the following
topics:
• Section 2 describes a framework for designing an
OWQM-DS system and introduces high-level design
goals for these systems.
• Section 3 provides guidance on selecting monitoring
locations for OWQM-DS.
• Section 4 provides guidance on selecting water quality parameters for OWQM-DS.
• Section 5 provides guidance on the design of monitoring stations for OWQM-DS.
• Section 6 provides guidance on the development of an information management system and
analysis techniques to support OWQM-DS.
• Section 7 provides guidance on developing an alert investigation procedure to support
OWQM-DS.
• Section 8 describes a process for developing a preliminary design for an OWQM-DS system.
• Section 9 presents example OWQM-DS applications, including a summary and explanation of
relevant monitoring locations and water quality parameters.
• Section 10 provides case studies from utilities that have implemented OWQM-DS.
Applicability of Guidance
The methodology presented in
this document can be used to
design OWQM-DS systems that
vary widely in complexity—from
a simple system monitoring a
limited number of parameters at
a single location to a system that
monitors multiple parameters at
numerous locations.
3
-------
Online Water Quality Monitoring for Distribution Systems
• Section 11 presents lessons learned from utilities that have implemented OWQM-DS.
• Resources presents a comprehensive list of documents, tools, and other sources cited in this
document that are useful for implementing activities described in this document.
• References presents a complete list of supporting documents and sources cited in this document.
• Glossary provides definitions of terms used in this document, which are indicated by bold, italic
font at their first use in the body of this document.
4
-------
Online Water Quality Monitoring for Distribution Systems
Section 2: Framework for Designing
Onl ne Monitoring Systems
The OWQM-DS design process follows the principles of project management and master planning that
are described in Sections 2 and 3 of Guidance for Developing. Integrated Water Quality Surveillance and
Response Systems. This section presents a framework for designing an OWQM-DS system, as shown in
Figure 2-1 While depicted as a linear process, in practice it is iterative. Decisions or findings in
downstream steps may require that earlier steps be revisited.
Establish
Design Goals
J
Monitor for
contamination
incidents
Optimize distribution
system water quality
Establish
Performance
Objectives
Operational reliability
information reliability
Sustains bility
Review Distribution
System Resources
Review distribution
system hydraulic
resources
Review distribution
system water quality
resources
Design the
Monitoring System
Select monitoring
locations
Select water quality
parameters
Design monitoring
stations
Develop information
management and
analysis tools
Develop an alert
investigation
procedure
Figure 2-1, Online Monitoring System Implementation Framework
2.1 Establish Design Goals
A utility planning to implement OWQM-DS should first establish the overall purpose of such a system
and the decisions the data it produces is intended to support. The purpose and intent will inform the
development of high-level design goals, and more specific applications, to guide OWQM-DS
implementation. Design goals are the benefits a utility expects to achieve by implementing OWQM-DS.
The establishment of design goals is critical to ensuring that an OWQM-DS system will provide
information that is useful to a utility.
Examples of common, high-level design goals that cover most OWQM-DS applications are to (1) monitor
for contamination incidents and (2) optimize distribution system water quality.
2.1.1 Monitor for Contamination Incidents
The presence of a contaminant in a drinking water distribution system has the potential to cause harm to
the community and utility infrastructure. Contamination incidents may be unintentional (e.g., treatment
process failure and contaminant pass through, backflow incidents) or intentional (e.g.. purposeful
contamination of a storage tank). OWQM-DS information can be used to detect contamination incidents,
enabling a utility to isolate affected areas of its system and implement corrective actions, as needed.
Due to the uncertainty regarding the occurrence and nature of contamination incidents, design of an
OWQM-DS system to achieve this design goal faces the following challenges:
5
-------
Online Water Quality Monitoring for Distribution Systems
• It is very unlikely that the contaminant involved in a specific contamination incident can be
predicted with certainty. Thus, for an OWQM-DS system to be effective, it must be capable of
detecting a wide range of contaminants. For this reason, water quality parameters that are
responsive to a variety of contaminants are most often used to achieve this design goal.
• A contamination incident could occur anywhere in a distribution system; therefore, an
OWQM-DS system should cover as much of a system as feasible.
• Contamination incidents are typically transient and occur over a short time-period; thus, rapid
detection is important.
• Contamination incidents of high consequences are rare, but their impact could be significant
(i.e., low probability of occurrence, but high impact); thus, alerts must be reliable.
2.1.2 Optimize Distribution System Water Quality
Optimization of distribution system water quality involves operating a treatment plant and distribution
system in a manner that meets selected water quality objectives. To achieve this design goal, a utility
must:
• Understand how treatment plant and distribution system operations impact water quality
throughout a distribution system.
• Use OWQM-DS information to inform treatment and system operations to:
o Support water quality goals such as chlorine residual management and corrosion control,
o Prevent water quality problems such as nitrification, regrowth, and disinfection byproduct
(DBP) formation.
2.2 Establish Performance Objectives
Performance objectives and their associated metrics are measurable indicators of how well an
OWQM-DS system is operating. Throughout design, implementation, and operation of a system, a utility
can use performance objectives to determine whether the system is operating within acceptable
tolerances. While specific performance objectives should be developed by each utility in the context of its
unique design goals, common performance objectives include operational reliability, information
reliability, and sustainability.
Operational Reliability
Operational reliability is the degree to which an OWQM-DS system is performing at a level capable of
achieving selected design goals. It depends on proper maintenance of equipment and information
management systems necessary to operate a system. Considerations that can impact operational reliability
include accessibility of monitoring stations for maintenance, suitability of water quality sensors to the
chemistry and quality of distribution system water (e.g., impact of pH or dissolved iron on instrument
performance), environmental impact on monitoring stations (e.g., humidity, ambient temperatures), and
availability of suitable training for personnel responsible for maintaining OWQM-DS equipment.
Example metrics used to monitor operational reliability include:
• Percentage of time that an OWQM-DS system is fully operational
• Average response time to correct equipment problems
6
-------
Online Water Quality Monitoring for Distribution Systems
Information Reliability
Information reliability is the degree to which information produced by an OWQM-DS system is of
sufficient quality to support decision-making. Specifically, utility personnel must be able to interpret the
difference between typical water quality variability and changes indicative of a water quality issue
requiring a response action. Considerations for information reliability include the representativeness of
the water monitored at each monitoring location, compatibility of sensors with water chemistry
(e.g., water matrix effects that interfere with instrument
measurements), sensor capabilities (e.g., detection limits),
maintenance of sensors, operation of sensors within their defined
capabilities, and data analysis methods.
Information reliability can be characterized through data quality
objectives, which are metrics or criteria that establish the quality
and quantity of data needed to support decisions. Examples of
data quality objectives that might be considered for OWQM-DS
include:
• Data accuracy
• Data completeness
• Number of false alerts per month
• False negatives
Data quality objectives are important to build confidence in data collected for any environmental
monitoring program.
Sustainabilitv
Sustainability is the degree to which benefits derived from an OWQM-DS system justify the cost and
level of effort required for its continued operation. Benefits are largely determined by the design goals
that OWQM-DS information supports. For example, an annual reduction in chemical usage
(e.g., chlorine) can be achieved due to more efficient chemical dosing or improved water turnover in a
storage tank, which can be guided by OWQM-DS data. Other benefits may be difficult to quantify, such
as increased confidence of utility managers and operators in their ability to detect water quality problems;
however, these benefits should still be captured and described, as they are important to gauging the
sustainability of an OWQM-DS system. Costs include the capital and ongoing expenditures required to
operate OWQM-DS equipment, as well as the effort required to analyze OWQM-DS data and investigate
alerts. Example metrics for sustainability include:
• Consequences avoided through early detection of, and response to, contamination incidents
• Value of non-monetary benefits gained from the operation of an OWQM-DS system, including a
reduction in customer complaints
• Cost to maintain an OWQM-DS system
2.3 Review Distribution System Resources
Prior to designing an OWQM-DS system, it can be helpful to compile and assess distribution system
information resources to support the design process. Reviewing these resources prior to and during design
can help to ensure that the resulting system addresses selected design goals. Examples of utility resources
that can be reviewed during an assessment are described below.
f *\
" Identification of False 1
Negatives
False negatives represent a
change in water quality that is
not identified as such. This
could be tested by using
alternative sources of data that
identify changes in water
quality, such as customer
complaints, and comparing the
results of their investigations
with OWQM-DS data collected
during the same period. ,
7
-------
Online Water Quality Monitoring for Distribution Systems
Distribution System Hydraulic Resources
• Hydraulic and water quality models. A hydraulic model is a mathematical representation of
hydraulic conditions present in a distribution system under a certain set of conditions. Likewise,
water quality models are mathematical representations of the water quality present in a
distribution system under a certain set of conditions. Models can be used to understand flow paths
and hydraulic connectivity throughout a distribution system and determine the impact of
operations on flow paths and connectivity. If a model includes water quality modeling
capabilities, such as chlorine residual decay models, it can also be used to estimate the impact of
distribution system operations on water quality as water travels through a system. Models can
also be used to identify areas with low flow and high water age that may be subject to regrowth
and nitrification.
• Distribution system maps and storage facility specifications. Distribution system maps and
storage facility specifications provide details that can be used in the absence of, or to supplement,
hydraulic models to show connectivity and storage capacity.
• Existing flow and pressure meter records. Existing flow and pressure meters in a distribution
system may have been installed to monitor discrete areas, wholesale customers, or high-demand
customers. Records of flow and pressure can be used to validate portions of a hydraulic model.
• Tracer study results. Tracer studies monitor the concentration profile of known chemicals as
they pass through a distribution system. Study results provide measured details of hydraulic
connectivity that can be used to validate or calibrate hydraulic models. In the absence of a
hydraulic model, these results can be used to understand flow paths and hydraulic connectivity
throughout the area evaluated during a study.
Distribution System Water Quality Resources
• Records of previous water quality problems. Records of water quality problems
(e.g., regrowth, nitrification, total coliform hits, DBP occurrences) that have occurred include
details of the nature and location of the problem, impact on water quality, date of occurrence,
other conditions at the time (e.g., temperature, pH, total organic carbon), and known or potential
causes of the problem.
• Customer complaint records. Records of customer complaints contain details of specific water
quality problems (e.g., red water, dirty water, taste and odor) that have previously occurred in a
distribution system.
• Regulatory compliance data. Records of compliance data related to federal regulations
(e.g., Surface Water Treatment Rule [SWTR]) and state regulations (e.g., monitoring disinfectant
residuals for ground water) can provide details of problem areas of a distribution system.
• Initial Distribution System Evaluation Results for Stage 2 Disinfectant and Disinfection
Byproducts. Results from this one-time requirement identify areas of a distribution system with
the potential for high DBP concentrations. These results can be used to learn about distribution
system hydraulics and areas in a system with low flow and high water age.
• Methodology for identifying Revised Total Coliform Rule (RTCR) sample siting plans.
RT CR sample siting plans specify where in a distribution system RT CR compliance samples are
collected. The intent of the plan is to select sample sites that are representative of water quality in
the system.
• Surveys or special studies. Results from surveys or special studies, such as sanitary surveys,
chlorine residual surveys, iron occurrence surveys, corrosion control surveys, and condition
assessments, contain details of analyses focused on particular aspects of a distribution system.
These results may provide insight into the location and frequency of water quality problems in a
system.
8
-------
Online Water Quality Monitoring for Distribution Systems
• Cross-connection control program. Information from a cross-connection control program
provides details of known sites within a distribution system where backflow is a concern.
• American Water Works Association (AWWA) Partnership for Safe Water Assessment
Results. The Distribution System Optimization Program, as described on the AWWA Partnership
for Safe Water Website, has identified three system integrity performance indicators: water
quality preservation (chlorine residual), hydraulic reliability (pressure), and physical security
(main break frequency). These indicators form the basis for a self-assessment to help utilities
identify performance-limiting factors and develop improvement plans.
2.4 Design the Online Monitoring System
The major design elements associated with OWQM-DS are summarized in Figure 2-2 and briefly
described in this section. Detailed guidance on each design element is presented in Sections 3 through 7.
Once the design elements have been developed, project details should be captured in a design document
as discussed in Section 8.
Investigation
& Response
Procedures
Monitoring Information
Stations Management &
Analysis
Monitoring
Locations
Water Quality
Parameters
Figure 2-2. Online Monitoring System Design Elements
Select Monitoring Locations
Monitoring locations should be selected based on design goals selected for OWQM-DS as well as
information collected during a distribution system assessment. The final selection of locations is often a
compromise between ideal locations determined to achieve a particular design goal, the potential for
locations to support multiple design goals, and installation considerations such as accessibility, security,
and environmental conditions. Guidance on the selection of monitoring locations is discussed in detail in
Section 3.
Select Water Quality Parameters
The selection of water quality parameters is based on design goals selected for OWQM-DS as well as
information collected during a distribution system assessment. Specific design goals can only be achieved
if parameters relevant to those goals are monitored. Guidance on the selection of water quality parameters
is discussed in detail in Section 4.
Design Monitoring Stations
The design of monitoring stations is based on the monitoring locations and water quality parameters
selected for OWQM-DS. It includes selection of water quality instruments and ancillary equipment
necessary to bring sensors into contact with a water sample and transmit data. The station design can
dramatically impact capital and operating costs for an OWQM-DS system, as well as data accuracy and
completeness. Guidance on the design of monitoring stations is discussed in detail in Section 5.
9
-------
Online Water Quality Monitoring for Distribution Systems
Develop an Information Management and Analysis System
Information management systems receive, process, analyze, store, and present data generated by
monitoring stations. An information management system may also be capable of generating alerts and
sending notifications to designated personnel when water quality anomalies are detected. Information
management and analysis are discussed in detail in Section 6.
Develop an Alert Investigation Procedure
Once a water quality anomaly has been detected, an investigation should be undertaken to determine the
cause of the anomaly. The process of developing an alert investigation procedure is described in
Section 7.
10
-------
Online Water Quality Monitoring for Distribution Systems
Section 3: Monitoring Locations
A monitoring location is the site in a distribution system where water is sampled to measure water quality
in real time. Selection of these locations should be guided by chosen design goals and information from a
distribution system assessment.
Target Capability
Monitoring is performed at all distribution system entry points and additional monitoring locations that are
sufficient to meet selected design goals.
This section is divided into subsections that cover the following topics:
• Common monitoring locations
• Tools to support selection of monitoring locations
• Installation sites
3.1 Common Monitoring Locations
It would be ideal for an OWQM-DS system to produce data that is representative of an entire distribution
system, but budgetary constraints often limit the number of monitoring stations that can be installed.
Therefore, monitoring locations should be strategically selected to maximize the extent to which design
goals are realized. Common monitoring locations are explained in the following subsections.
Distribution System Entry Points
Distribution system entry points include the locations where water from treatment plants, wholesale
interconnects (where treated water enters a system), or the output of one or more wells directly feeds into
a system. As such, entry points are vital locations that should be monitored for all OWQM-DS systems.
Monitoring at entry points is an important aspect of the design goals mentioned in Section 2.1.
OWQM-DS data at the entry points provides a useful benchmark for optimization of distribution system
water quality. It also provides a baseline that can be used to help confirm or rule out a possible
contamination incident detected at a downstream monitoring location in a distribution system. This data
can also be used in water quality models (e.g., chlorine residual decay, DBP formation) to predict
distribution system water quality. It can also guide treatment plant maintenance planning
(e.g., replacement of adsorptive media based on data rather than on time in service).
Operational Control Points
Operational control points include storage facilities, chlorine residual booster stations, and pump
stations. Monitoring at, or downstream of, operational control points provides an operator with
information that can be used to adjust operations that impact distribution system water quality.
OWQM-DS data from locations downstream of a control point may provide the most useful information
to guide system operations, as water quality at a downstream location can show the impact of an
operational change at a control point. However, it is often easier and less expensive to install monitoring
stations at utility facilities associated with operational control points because these facilities frequently
satisfy many of the installation requirements identified in Section 5.
Prioritization of operational control points for monitoring should be informed by a distribution system
assessment, discussed in Section 2.3, and may include the following:
11
-------
Online Water Quality Monitoring for Distribution Systems
• The population downstream of a control point. Average flow rate exiting the control point can
be used as a surrogate for the population downstream of
the point and can be obtained from Supervisory Control
and Data Acquisition (SCADA) system records,
hydraulic models, existing flow meters, or tracer study
results.
• Hydraulic travel time from a treatment plant or
upstream operational control point and hydraulic
connectivity to other stations. This information can
often be obtained from distribution system asset design
drawings and specifications, hydraulic models,
distribution system maps, or operator knowledge.
• History of water quality or compliance issues at, or downstream of, an operational control
point. This information may be available from federal or state regulatory sampling results,
customer complaints, water quality problems, other water quality data, and an Initial Distribution
System Evaluation.
Additional Monitoring Locations
Additional monitoring locations beyond entry points and operational control points can provide
information to better meet design goals. Examples of these locations include the following:
• Critical customers. Stations can be installed upstream of critical customers or areas (e.g.,
hospitals, stadiums, universities, entertainment districts, manufacturers that use large volumes of
water) to protect large and/or vulnerable populations in the event of a contamination incident.
Stations should be located a sufficient distance upstream of an asset to allow time to detect and
respond to an incident. Critical customers can be identified using details from distribution system
maps, hydraulic models, and geographic information system (GIS) resources.
• Interconnects. Interconnects with downstream customers can be monitored to determine the
quality of water transferred between systems. Interconnects can be identified using distribution
system maps.
• Areas of concern. Areas that have a history of water quality problems (e.g., nitrification
incidents, undetectable chlorine residual levels, lead pipes) areas that experience low flow and
pressure, and far reaches of the distribution system can be monitored to identify the onset of
future problems. Areas of concern can be identified through a review of federal or state regulatory
sampling results, customer complaint records, records of water quality problems, special studies,
and Initial Distribution System Evaluation results.
• Mixing zones. For distribution systems supplied by multiple sources with different water quality
(e.g., surface water, ground water), areas where water from different sources meet and mix can be
monitored to characterize the frequency and duration of mixing, and possibly the boundaries of
the mixing zones. Mixing zones can be identified using utility personnel knowledge of a
distribution system or hydraulic models.
3.2 Tools to Support Selection of Monitoring Locations
A number of tools can be used to provide information to support the selection of monitoring locations.
Three general classes of such tools are described in the following subsections:
• Hydraulic models
• Water quality models
• Sensor placement optimization tools
Water Variability
Prior to confirming a monitoring
location, one or more extensive
sampling events should be
conducted at the location to
determine whether the water
quality is sufficiently stable to
allow for detection of changes that
support selected design goals.
12
-------
Online Water Quality Monitoring for Distribution Systems
Methodologies that use these tools for single objective and multi-objective placement of monitoring
stations are discussed in Water Science & Technology: Water Supply (Rathi, et al., 2015).
Hydraulic Models
Hydraulic models are mathematical representations of distribution system hydraulics under various
conditions. EPANET is a common, open-source hydraulic modeling platform that is frequently used in
the industry. Hydraulic models can be used to:
• Understand hydraulic connectivity between potential monitoring locations and other distribution
system elements, such as operational control points.
• Determine travel time between locations in a distribution system.
Water Quality Models
Water quality modules can be incorporated into most distribution system hydraulic modeling software to
estimate water quality as water travels through a system (e.g., EPANET). Water quality modeling can
provide additional information to:
• Understand how water quality parameters change as water travels through a system.
• Predict water quality parameter values downstream of operational control points.
Station Placement Optimization Tools
Station placement optimization software, such as the Threat Ensemble Vulnerability Assessment and
Sensor Placement Optimization Tool (TEVA-SPOT), uses complex algorithms to identify and prioritize
monitoring locations for a specific objective, such as minimizing the time to detect an incident or
minimizing consequences over a large ensemble of possible contamination scenarios. Examples of how
TEVA-SPOT has been used to locate monitoring stations are provided in guidance developed by the
Philadelphia Water Department and CH2M (PWD, 2013), a summary of SRS pilot projects that employed
TEVA-SPOT (EPA, 2015), and a presentation on sensor network design and performance (Janke, et al.,
2009).
Use of optimization software may be constrained by "fixing" one or more monitoring locations that are
priorities for monitoring. For example, utilities may want to fix locations at distribution system entry
points, operational control points, and critical facilities at which the consequences of water contamination
could be severe (e.g., hospitals, universities, government buildings, entertainment venues). The
optimization software can then be used to identify additional locations while considering the detection
capabilities provided by the fixed locations.
Other station placement optimization tools are available, and not all require a distribution system model.
One such method identifies a monitoring location intended to minimize the time to detect a contamination
incident (Schal, et al., 2014). This is a simplified method that may be appropriate for relatively small and
simple distribution systems, if the above approaches are not feasible.
3.3 Installation Sites
The physical installation of a monitoring station should be
as close as feasible to the monitoring location (e.g., water
main, tank) to minimize the time between sample collection
and analysis. Selection of a monitoring station installation
site is often influenced by a variety of site-specific
considerations such as those identified in the OWQM site
evaluation checklist that can be accessed by clicking on the
box to the right. If a suitable installation site near the
A
Checklist for Assessing
Potential Monitoring
Locations (Microsoft Word)
*Note that the document that is
currently open may need to be
downloaded and opened offline
to access this checklist.
13
-------
Online Water Quality Monitoring for Distribution Systems
monitoring location cannot be found, alternate locations will need to be considered. Completed checklists
should be incorporated with the preliminary OWQM-DS system design template found in Section 8.
14
-------
Online Water Quality Monitoring for Distribution Systems
Section 4: Water Quality Parameters
This section describes water quality parameters that may be useful for OWQM-DS. Information about the
online instruments used to measure these parameters is available in Guidance for Selecting Online Water
Quality Monitoring Instruments for Source Water and Distribution System Monitoring. The scope of this
document is focused on water quality parameters; therefore, operational parameters (e.g., pressure, flow)
are not covered in this section.
Target Capability
Core water quality parameters (chlorine residual, pH, specific conductance, and temperature), and additional
parameters that are sufficient for achieving selected design goals, are measured at each monitoring location.
Within the context of OWQM-DS, there is a wide range of water quality parameters that can be
monitored to contribute to selected design goals. These parameters can be grouped into core parameters
that should be monitored as part of every OWQM-DS system and additional parameters that can be
monitored to achieve utility-specific design goals.
Core Water Quality Parameters
Table 4-1 provides an overview of a core group of water quality parameters as defined in this document.
These parameters are fundamental to understanding water chemistry and are useful in identifying a broad
spectrum of water quality changes in a distribution system. Furthermore, the interdependencies among
these core parameters make them useful during the investigation of a water quality change. For example,
the rate of chlorine residual decay typically increases as temperature increases. A situation in which
chlorine residual decay is greater than anticipated at a given temperature would warrant further
investigation.
Table 4-1. Overview of Core Water Quality Parameters
Parameter
Parameter Description
Chlorine residual
• Defined as the concentration of either free chlorine, total chlorine (free chlorine plus
chloramines), or monochloramine.
• Concentration must be maintained within lower and upper bounds, as required by
federal or state regulations.
• Decreases in chlorine residual can signal the potential for regrowth of biological
organisms and biological processes in a distribution system, including those that
cause nitrification.
• Many chemical contaminants that could enter a system will react with chlorine
residual, causing a decrease in residual concentration.
PH
• Defined as the negative logarithm of the concentration of hydrogen ions in an
aqueous solution.
• Is necessary to understand water chemistry (e.g., chemical speciation, reaction
rates).
• Can be used along with other parameters to determine the corrosion potential of
water in a distribution system.
• Chemical contaminants with acidic or basic functional groups that could enter a
system can change the pH of water; however, the magnitude of a change in pH will
be inversely related to the buffering capacity of the water.
15
-------
Online Water Quality Monitoring for Distribution Systems
Parameter
Parameter Description
Specific conductance
• Defined as the measure of the ionic strength of a solution.
• Commonly used as a surrogate for total dissolved solids.
• Can be used to track different water supplies (e.g., ground water, surface water)
feeding into a distribution system (assuming those supplies have measurably
different values for specific conductance), thus providing a better understanding of
system hydraulics.
• Some chemical contaminants that could enter a system have charged functional
groups that can dissociate and form ionic species when dissolved in water, thus
increasing the specific conductance of the water; however, a high contaminant
concentration may be necessary to produce a measurable change in specific
conductance.
Temperature
• Defined as the measure of the thermal energy in water.
• Influences chemical equilibrium and kinetics, which may impact water quality in a
distribution system and, thus, is an important parameter to monitor at any
monitoring location.
• Integrated into water quality sensors that measure temperature-dependent
parameters (e.g., pH, specific conductance) to enable temperature compensation
to those parameter measurements.
• Can be used to track different water supplies (e.g., ground water, surface water)
feeding into a distribution system (assuming those supplies have measurably
different values for temperature), thus providing a better understanding of system
hydraulics.
• A rapid change in temperature can indicate a large inflow of a foreign fluid
(e.g., plumbing high-flow cross-connection) into a system.
Additional Water Qualitv Parameters
Table 4-2 provides an overview of additional water quality parameters that may be useful for an
OWQM-DS system to meet utility-specific, and in some cases site-specific, design goals. The parameters
listed below are supplementary to the core parameters and can identify specific types of water quality
changes in a distribution system.
Table 4-2. Overview of Additional Water Quality Parameters
Parameter
Parameter Description
Alkalinity
• Defined as the measure of a water's buffering capacity (i.e., its ability to resist a
change in pH when an acid or base is added), typically measured in carbonate
equivalents.
• Influences the stability of distribution system water pH and, thus, impacts corrosion
control.
• Can be used to calculate the likelihood of calcium carbonate pipe scale formation
due to calcium carbonate saturation in water.
Ammonia, free (NH3)
• Defined as the concentration of dissolved ammonia (NH3) in solution.
• Can be added during treatment to form chloramines.
• Can exert a chlorine demand.
• As monochloramine decays it releases ammonia. Nitrifying bacteria
(i.e., nitrification) consumes ammonia and converts it to nitrite and nitrate. High
levels of ammonia should be avoided to prevent the likelihood of nitrification.
Apparent color
• Defined as the color of an unfiltered water sample due to both dissolved and
suspended material.
• An increase in apparent color can signal a potential hydraulic upset that could
impact water quality (e.g., iron release).
16
-------
Online Water Quality Monitoring for Distribution Systems
Parameter
Parameter Description
Dissolved oxygen (DO)
• Defined as the concentration of dissolved oxygen in solution.
• Is an oxidant that can affect metal solubility and release.
Dissolved organic carbon
(DOC)
Total organic carbon
(TOC)
• Defined as the concentration of organic carbon (compounds that contain carbon
and hydrogen) in solution.
• TOC includes suspended and dissolved organic carbon.
• DOC is the fraction of organic carbon that passes through a filter with a
0.45 micrometer pore size.
• Presence of DOC/TOC during chlorination can result in DBP formation.
• Assimilable organic carbon portion of DOC can support biological regrowth in
distribution systems.
• An increase in DOC and TOC can indicate the presence of an organic chemical
(e.g., pesticides, biotoxins, petroleum products) in a system.
• An increase in DOC and TOC can exert a chlorine demand, reduce the chlorine
residual concentration, and create an opportunity for chlorine-sensitive pathogens
(e.g., Vibrio cholerae) and biotoxins (e.g., botulinum toxin) to survive.
Disinfection byproducts*
• Defined as the concentration of total trihalomethanes or 5 haloacetic acids
• Concentration must be maintained below maximum contaminant levels specified by
the Disinfectants and Disinfection Byproducts Rules.
Hydrocarbons
• Defined as the concentration of long-chain, organic compounds that include
hydrogen and carbon in solution (online instrumentation commonly measures
unsaturated organic compounds).
• Can enter a system during intentional or unintentional backflow incidents and
low-pressure incidents in pipes buried in contaminated soil.
• Can impart an objectionable odor to water, and can be difficult to remove from
distribution systems and household plumbing materials.
Nitrite and nitrate
• Defined as the concentration of nitrite (NO2) and nitrate (NO3) in solution.
• Are regulated contaminants.
• Measurable concentration of nitrite, or increases in nitrate from baseline levels, can
signal the onset of nitrification in chloraminated distribution systems.
Ortho-phosphates
• Defined as the concentration of inorganic compounds, consisting of phosphorus
and oxygen in solution.
• Can be used to monitor the efficacy of corrosion control treatment if a phosphate-
based inhibitor is used during treatment.
Oxidation-reduction
potential (ORP)
• Defined as the measure of the potential flow of electrons between reducers and
oxidizers, which characterizes the oxidizing or reducing potential of a solution;
positive values indicate an oxidizing environment and negative values indicate a
reducing environment.
• Can be used for understanding corrosion control and metal solubility.
• Closely related to chlorine residual and typically responds linearly to chlorine
residual changes.
• A decrease in ORP can indicate the presence of chemical contaminants that exert
an oxidant demand.
17
-------
Online Water Quality Monitoring for Distribution Systems
Parameter
Parameter Description
Spectral absorbance
Defined as the measure of wavelength absorption across the ultraviolet (UV)/visible
spectrum.
Can provide derived measurements for other water quality parameters
(e.g., DOC/TOC, nitrite, nitrate).
The DOC/TOC ratio can be used to indicate a change in the organic composition of
water, potentially indicating the presence of an organic contaminant.
Spectral absorption profiles of distribution system water can provide a baseline
spectral fingerprint; deviations from this baseline can be used to detect anomalous
water quality and may indicate the presence of a contaminant.
Some inorganic and most organic chemicals absorb in the UV/visible spectrum; as
such, a change in spectral absorption may indicate the presence of a chemical
contaminant introduced into a distribution system.
Differential spectral analysis is a rapid method to compare the significance of a
change between a baseline water quality and a potential anomaly.
Turbidity
• Defined as the measure of the cloudiness of water due to dissolved or suspended
particles.
• An increase in turbidity can indicate hydraulic upsets or intrusions in a distribution
system.
UV-254
Defined as the light absorption at the ultraviolet 254nm wavelength.
Can provide a surrogate measure of DOC/TOC.
UV-254 and DOC measurements can be used to calculate Specific Ultraviolet
Absorbance (SUVA).
A change in SUVA can indicate a change in the organic composition of water,
potentially indicating the presence of an organic contaminant.
*Note that DBPs are measured using online Gas Chromatographs
r
The water quality parameters selected from T ables 4-1 and 4-2 should be those that are most useful for
achieving selected design goals. When selecting parameters,
consider that some provide innate benefits, while others may
complement different monitored parameters and provide more
useful information when measured together. For example,
nitrification and regrowth is often exacerbated when high
temperatures accelerate chloramine decay and the growth rate of
nitrifying bacteria, producing free ammonia; thus, monitoring
chlorine residual, free ammonia, and temperature together can
provide a more reliable indication of nitrification. Additional
examples of parameter combinations that can be selected for a
number of OWQM-DS applications are presented in Section 9.
Contamination Detection
The ability to detect distribution
system contamination by
continuous monitoring of water
quality parameters is a novel
design goal of OWQM-DS. The
parameters most valuable for
monitoring for contamination
incidents (chlorine residual,
ORP, pH, specific conductance,
temperature, DOC/TOC, and
spectral absorbance) are
identified in Tables 4-1 and 4-2.
A more detailed discussion of
this design goal and other
OWQM-DS applications can be
found in Section 9.
Online instruments that measure additional parameters of interest,
such as radionuclides (e.g., alpha, beta, gamma levels), toxicity,
and refractive index, have not yet been used in OWQM-DS
systems on a wide scale. Instruments that measure radionuclides
typically have a minimum detection limit that is higher than the
maximum contaminant levels established by EPA. Instruments that measure toxicity and refractive index
effectively are commercially available, but have not been used widely in OWQM-DS systems. Guidance
for Selecting Online Water Quality Monitoring Instruments for Source Water and Distribution System
Monitoring provides more information on these technologies.
18
-------
Online Water Quality Monitoring for Distribution Systems
Section 5: Monitoring Stations
Once monitoring locations and parameters have been selected, monitoring stations can be designed. Each
station should consist of the water quality instruments and other equipment necessary to measure and
process a sample, and then collect and transmit data to a utility's control center. The design of a station
will depend on:
• The monitoring location
• Water quality parameters to be monitored at the location
• Practical considerations for installation and maintenance of the station
Target Capability
Monitoring stations are equipped with basic features necessary to support operation of water quality instruments
and transmit data to a utility's control center.
A basic functional block diagram of a monitoring station is shown in Figure 5-1, which delineates the
station functions as follows:
• Instrumentation. Provides the means to measure selected water quality parameters.
• Computing element. Facilitates the transfer of OWQM-DS data and other datastreams to the
communications function, enables remote control of monitoring stations, and provides processing
capabilities at stations.
• Communications. Provide a means to transfer data collected by a monitoring station to a control
center and instructions from a control center to a station.
• Power supply and distribution. Supplies sufficient power to energize equipment in a monitoring
station.
• Accessories. Perform other functions not defined above.
• Station structure. Provides a means to mount and protect instrumentation and ancillary
equipment both from the environment and potential tampering.
The following sections describe each of the functions identified in Figure 5-1.
19
-------
Online Water Quality Monitoring for Distribution Systems
Figure 5-1. Functional Block Diagram of a Monitoring Station
5.1 Instrumentation
In many cases, multiple sensor technologies are available to measure a given water quality parameter, and
specific instruments will need to be selected for a monitoring station. Several factors warrant
consideration when selecting an instrument, including instrument performance, sampling approach,
sampling and analysis interval, environment at an installation site,
lifecycle cost, and vendor support. Guidance for Selecting Online
Water Quality Monitoring Instruments for Source Water and
Distribution System Monitoring provides an overview of water
quality parameters and related sensor technologies, as well as
factors that should be considered during the selection process.
The most common sampling approach for OWQM-DS involves
connecting a sample side stream from a distribution main or tank to
sensors inserted into a flow-cell contained inside a monitoring
station. This method requires installation of piping or tubing to
move the sample to the flow-cell. Some instruments designed for
use in a flow-cell are equipped with wipers, brushes, or compressed
air to control sensor fouling.
If a monitoring station uses a flow-cell, it will produce a waste
stream that requires disposal. If a station houses a water quality
instrument that adds reagents to sample water, additional
requirements may dictate the method of waste stream disposal. The
Tiered Stations
Because some instrumentation
can be expensive, installing
different tiers of monitoring
stations can be a cost-effective
option. For example, a limited
number of monitoring stations
that monitor both core and
additional water quality
parameters could be installed at
critical monitoring locations,
while additional stations monitor
only core parameters. This
approach can help to minimize
the cost of an OWQM-DS
system while also providing
robust monitoring at critical
locations.
20
-------
Online Water Quality Monitoring for Distribution Systems
most convenient method of waste disposal is to direct waste streams into a sanitary sewer, if available and
permissible. At installations where a sewer is not available, a dry pit is typically used, which may require
additional approvals (note that dry pits must be designed to accommodate the flow of the waste stream).
5.2 Computing Element
Each monitoring station should include a local computing element that is typically a proprietary
instrument controller (e.g., Hach SC1000, s::can con::cube, YSIIQ SensorNet). Alternatively, an
industrial computer can be used to provide more complex and flexible processing capabilities.
A local computing element provides functions that may include the following:
• Controlling instrumentation (e.g., setting of timing intervals between measurements)
• Monitoring instrumentation and transferring analysis results to the communications function
• Monitoring instrumentation and communications for failures or error flags
• Managing and monitoring accessory functions (e.g., detection sensors, cameras)
• Controlling local functions (e.g., enabling the collection of a grab sample based on local or
remote triggers)
• Providing more complex software algorithms for data validation and anomaly detection
5.3 Communications
The selection of a communications solution to transmit data from a monitoring station to a control center
is often influenced by a station's location and proximity to existing communications solutions available to
a utility (e.g., city or county network). Communications solutions can include wired and wireless
technologies. Guidance for Designing Communications Systems for Water Quality Surveillance and
Response Systems provides further details for common communications options as well as a set of
evaluation criteria to support the selection process.
5.4 Power Supply and Distribution
The choice of power supply for a monitoring station is often limited by the location where the station is
installed and power requirements for station equipment. Where readily available, grid power is often the
simplest and least expensive power supply. However, if grid power is not available nearby, extending it to
a station may be equally or more expensive than using an alternative supply (e.g., solar supported by
batteries). When using grid power, it is suggested that stations have a dedicated circuit on the main
breaker panel or a line conditioner to avoid erratic voltage or circuit breaker trips. To ensure continued
operation of a station during minor power outages, an uninterruptible power supply should also be
installed. Guidance for Building Online Water Quality Monitoring Stations provides additional guidance
on power distribution.
5.5 Accessories
Additional features and functions may be provided as part of a monitoring station, such as:
• Autosamplers, consisting of bottles and valving, that facilitate the automatic collection of water
samples at a station immediately after a water quality anomaly is detected
• Lighting and other accessories to assist with maintenance activities
• Sensors to detect leakage from plumbing or flooding by other means
• Cameras and door switches for security
• Additional communications equipment (e.g., Ethernet switches)
21
-------
Online Water Quality Monitoring for Distribution Systems
• Operator interface terminal to interact with a local computing element to support calibration and
troubleshooting
Guidance for Building Online Water Quality Monitoring Stations provides further details related to
accessories and how they can be incorporated into a monitoring station.
5.6 Station Structure
The station structure for a monitoring station includes the materials and devices used to mount or house
OWQM-DS equipment. Where a flow-cell is used, the flow-cell is part of a station. In-pipe sensors
typically require a connection to a separate controller and communications equipment installed outside of
a pipe. To achieve selected design goals and performance objectives, stations may need to be installed
inside existing buildings, near other equipment, or inside a
structure built specifically for the station. The nature of a specific
installation site will influence the station structure. Stations are
typically constructed using one of four primary design types:
• Wall-mounted racks. Assembled by securing instruments
and related equipment to a mounting panel that is attached
to a wall.
• Free-standing racks. Constructed by securing instruments
and related equipment to a mounting panel that is attached
to an open, structural frame that provides access to both
sides of the panel.
• Enclosed stations. Designed to house instruments and
related equipment inside a custom, prefabricated, or National
Association (NEMA) rated enclosure.
• Compact stations. Smaller versions of enclosed stations that can be designed around one or two
reagent-based instruments or a flow-cell with multiple reagentless instruments to measure
multiple parameters.
Guidance for Building Online Water Quality Monitoring Stations provides details for each of these
monitoring station designs.
Insertion Sensors
At the time of document
publication, flow and pressure
sensors are the most common
types of sensors that can be
inserted directly into a pipe.
Temperature, chlorine residual,
and spectral absorbance
insertion sensors are also
available, but they are not as
commonly used.
Electrical Manufacturers
22
-------
Online Water Quality Monitoring for Distribution Systems
Section 6: Information Management and Analysis
The data generated by monitoring stations must be converted into actionable information to achieve
selected design goals and provide a utility with the maximum value for its investment. Actionable
information is produced by analyzing OWQM-DS data, along with supporting information, and
presenting relevant results to an end user in a manner that is easy to understand. To achieve these
objectives, an OWQM-DS information management system must be a combination of hardware,
software, tools, and processes to collectively support an SRS and provide users with information needed
to monitor real-time system conditions.
Target Capability
An information management system is in place that is capable of storing, processing, and displaying OWQM-DS
data.
The development process discussed in this section is consistent with the general principles of information
management system design presented in Section 4 of the Guidance for Developing Integrated Water
Quality Surveillance and Response Systems, with additional considerations that are specific to an
OWQM-DS information management system. This section is divided into subsections that cover the
following topics:
• Data validation
• Anomaly detection
• Information management system architecture
• Information management system requirements
6.1 Data Validation
Accurate data is needed to achieve OWQM-DS design goals,
maximize the benefit and sustainability of an OWQM-DS system,
and build confidence in information generated by the system.
However, even when effective data quality objectives have been
established and are being achieved overall, invalid data values are
inevitable due to issues such as instrument malfunction, flow or
pressure irregularities, and improper maintenance.
Information Utilization
During a forum with chief
information officers (ClOs)from
50 major utilities across the United
States, the ClOs estimated that
only 10 to 15 percent of the
information gathered by their
organizations is properly
evaluated. Automated analysis
and effective visualization of data
can help to address this
underutilization of collected data.
Data validation involves the identification of data that is
inaccurate so that it can be handled in a manner that does not
negatively impact the intended use and further analysis of the
data. The most straightforward method for identifying invalid
data is to consider the data values themselves. For example,
values that are outside of an instrument's measurement range, as well as null data (missing or zero) or
non-numeric values (e.g., ""N/A"). are considered invalid. Data validation can also utilize patterns in data
values. Data is typically invalid during data "flatlines," in which data values remain constant for an
extended period, and during periods of extreme variability, in which the frequency and magnitude of
changes in data values are physically improbable.
Common methods of managing invalid data include flagging or removing invalid values. Some data
validation technologies contain logic for "cleaning" the data, in which invalid data values are replaced
with values deemed more likely to better represent true water quality. This practice provides a complete
dataset for analysis, but there cannot be complete confidence that the replacement values are accurate.
23
-------
Online Water Quality Monitoring for Distribution Systems
Regardless of the approach used for OWQM-DS data validation, it is best practice to retain the original
data values to maintain data integrity. For data to be used in an OWQM-DS system, data validation must
be automated and occur before anomaly detection takes place.
Supplemental information can also be used for data validation. Table 6-1 lists information types that can
indicate when data being produced by a water quality instrument is likely inaccurate.
Table 6-1. Supplemental Datastreams that can be Used to Validate Data
Information Type
Information Source
Relevance
Water quality instrument
fault indicator
Station water quality
instruments
Signals that an instrument is functioning in a manner that
could produce inaccurate data
Communications system
status indicator
Information
management system
Signals a station communications system malfunction,
which can cause flatlined or null data
Station flow and pressure
data
Station flow meters or
pressure sensors
Can be reviewed to determine whether values are outside
of water quality instrument manufacturer specifications,
which can cause a malfunction or produce inaccurate data
Maintenance mode indicator
Switch located at a
station (e.g., manual
switch, door interlock)
Signals that utility personnel are working on the station
(e.g., calibrating a water quality instrument, replacing a
sensor), which can produce inaccurate or null data
Station environment
information (e.g., temperature
and humidity data,
surveillance video)
Station accessories
(e.g., sensors,
cameras)
Can be reviewed to determine if conditions at a station
could have caused instruments to malfunction
In addition to the automated data validation methods above, an information management system can be
designed to allow utility personnel to manually flag data as
inaccurate during data review or alert investigations. Staff can
identify invalid data that is not detectable by automated
techniques using both their knowledge of typical water quality
at monitoring locations and additional resources (e.g., system
operations information, equipment maintenance records,
grab sample results).
In some cases, it may be unnecessary to implement a separate
data validation method because the capability to manage invalid
data is built into the information management system. For
example, a sophisticated data analysis tool may have processes that ignore data flagged by water quality
instruments. However, if data validation is not incorporated into an information management system,
there may be ways to implement data validation algorithms at certain points in the system (e.g., by a
monitoring station local computing element).
6.2 Anomaly Detection
Unlike data validation, in which inaccurate values are identified, this section describes techniques for
analyzing data to identify anomalies. Anomalies are changes in OWQM-DS data values that require
attention from utility personnel and may prompt response actions.
While operators can visually review data daily, automated analysis techniques are also available. An
Anomaly Detection System (ADS) is a data analysis tool designed to detect anomalies or deviations from
an established baseline. The complexity of ADSs can vary widely, as well as the time and technical
expertise required to implement them. Three categories of automated analysis techniques are described
below.
Fault Codes
Some complex instruments may
provide "fault codes" that not only
indicate that a fault has occurred,
but also include details of the fault.
Understanding the meaning of
these fault codes is necessary to
understand the implication of the
code with respect to data quality.
24
-------
Online Water Quality Monitoring for Distribution Systems
Threshold Analysis
Threshold analysis is a process in which an alert is automatically generated when a parameter's value
surpasses a pre-defined threshold. Threshold values should be based on the normal range of values for a
given parameter at a given monitoring location so that threshold exceedances are indicative of a water
quality anomaly, as shown in Figure 6-1. A comparison of basic threshold analysis to more sophisticated
techniques for anomaly detection was presented in Journal AWWA (Umberg and Allgeier, 2016).
Pt a ^
J J r?
©)
3
2.75
2.5
2.25
2
1.75
IS
Z) 'V1 «<~~~~!
Figure 6-1. Example of Threshold Analysis
In general, threshold analysis is easy to implement and is a standard feature in most information
management systems (including SCADA systems). Many utilities already use thresholds for treatment
plant process control; similar types of thresholds can be used to guide distribution system operations. For
example, thresholds can be established at monitoring locations and used to adjust booster chlorination
doses or monitor the efficacy of flushing activities. Therefore, utilities may already have a process that
can be used to establish thresholds for anomaly detection.
Complex Analysis
While thresholds can be useful for anomaly detection, in some cases, multiple parameters could be within
established threshold limits, but the relationship between them may indicate an anomaly. These types of
water quality anomalies cannot be detected by thresholds, but may be detected using statistical techniques
that can simultaneously analyze multiple parameter values at a given monitoring location.
Common techniques consider the rate of change in a single parameter at a single monitoring location or
the relationship between values for a given parameter across multiple locations. For example, one
location's chlorine values could be compared to the range of values seen at upstream locations, including
the distribution system entry point supplying the location, during a time-period consistent with the
hydraulic travel time between the locations. It would be expected that the downstream chlorine values
would be similar to, or lower than, the upstream values by some predictable amount. If this is not the
case, an alert could be automatically generated to indicate an anomaly.
25
-------
Online Water Quality Monitoring for Distribution Systems
Complex analysis can also combine water quality data with operational information. For example,
different threshold values for chorine residual could be used to generate an alert based on whether a
storage facility located near a given monitoring station is in fill or drain mode.
ADS Software
Anomaly detection can also be performed using software
developed specifically for analysis of time-series data. These
products use a variety of statistical and computer science
techniques to analyze data. The ADS software available at the
time of writing generally fall into two categories:
• ADS software integrated into station hardware.
Some instrument vendors offer ADS software integrated
with their hardware (and, thus, the software is installed
at each monitoring station). Examples include Hach's
Event Monitor and s::can's ana::tool.
• ADS software that is independent of sensor vendor. This software is often installed at a central
location and operates independently of any station hardware or information management system.
This ADS software may be proprietary (examples include OptiWater's OptiEDS and s::can's
ana::tool) or open source (such as CANARY, developed by EPA and Sandia National
Laboratories).
Further details about ADS software can be found in the report for the Water Quality Event Detection
System Challenge: Methodology and Findings (EPA, 2013) undertaken as part of the EPA's SRS
program.
Most information management systems include standard statistical analysis functions that can be
implemented to perform simple real-time analysis and basic anomaly detection. The values that trigger an
alert for these analyses can be determined by calculating typical values in representative historic
data. Exploratory Analysis of Time-series Data to Prepare for Real-time Online Water Quality
Monitorins provides additional guidance on techniques for establishing a baseline using representative
historical data.
Prior to selecting an ADS, multiple options should be evaluated using representative historical data to
determine which option can most reliably differentiate between true water quality anomalies and typical
water quality variability at each monitoring location.
6.3 Information Management System Architecture
The design of an information management system should be captured in a system architecture, which is
the set of hardware, software, processes, and services associated with the system. Information
management functions can be supported by a variety of system architectures, but they will most likely be
centralized (e.g., in a server at a utility's operations center). In this case, data is transmitted from
monitoring locations to this centralized system.
The architecture of an information management system may involve interaction with multiple source
systems. For example, supplemental data may be stored in separate systems, or external software may be
used for data analysis. In some cases, these systems can be integrated, though significant effort may be
needed to interface with legacy systems.
ADS Configuration
Proper ADS configuration is
necessary to reliably detect water
quality anomalies without
excessive invalid alerts. Note that
some complex ADSs may require
significant time and/or effort for
configuration.
26
-------
Online Water Quality Monitoring for Distribution Systems
This section includes three examples of information management system architectures that can be used
for OWQM-DS:
• Existing control system
• Dedicated information management system
• Hosted solutions (including cloud-based sendees)
These examples are intended to illustrate system architecture approaches that may be taken depending on
a utility's existing Information Technology (IT) infrastructure and system requirements. All examples
assume that each monitoring location has a communication device that transmits data to a central location.
Existing Control System
In some cases, information management requirements for OWQM-DS can be met entirely through an
existing control system. For example, data generated by monitoring stations can be added to an existing
SCADA system used for process monitoring and control. This arrangement would likely use existing
SCADA elements, such as a historian for data storage and a human-machine interface for visualization of
OWQM-DS data. Figure 6-2 shows an example of this type of architecture
MONITORING STATION
EXISTING SCADA SYSTEM
Monitoring
Instrumentation
PLC
A
I1
PLC 1
* f
PLC 2
A
IS
PLC 3
SCADA Server
SCADA HMI SCADA Historian
PLC: Programmable Logic Controller HMI: Human Machine interface SCADA: Supervisory Control and Data Acquisition
Figure 6-2. Example of Information Management as an Extension to an Existing SCADA System
Architecture
Dedicated Information Management System
In cases where an existing control system does not satisfy information management requirements for
OWQM-DS, a dedicated information management system may be appropriate. The following are
examples of when this may occur:
27
-------
Online Water Quality Monitoring for Distribution Systems
• An OWQM-DS system produces data that is difficult to store in a SCADA historian. For
example, instruments that measure spectral absorbance over multiple wavelengths can generate a
spectral profile as an array of 256 data points for each sample. The design of some SCADA
historians is not optimal for storing such arrays, but alternate database structures are available to
store these complex datastreams more efficiently.
• An OWQM-DS system requires access to data on networks (e.g., a utility's business network)
that cannot be accessed by a SCADA system due to security policies.
• Remote access to OWQM-DS data is required, and security policies preclude remote access to the
SCADA system.
The use of a dedicated information management system for OWQM-DS provides greater flexibility for
achieving the required functionality. It also allows for connection with other utility information
management systems, such as a laboratory information management system (LIMS), that contain
analytical results from grab samples collected from a distribution system. Figure 6-3 illustrates a
conceptual architecture of a dedicated information management system with connections to a customer
information system, LIMS, and work order system. SCADA data could also be utilized with such an
architecture if implemented in a manner that complies with utility information security policies.
28
-------
Online Water Quality Monitoring for Distribution Systems
I |V mm— i'tfn » «» -1 VHlfil Priva'C'
c L \ *' (VPN)I
u — Internet i connector> I
1 s L n ^ p— —r Js&>S\
t 1 - - f- ':*->. . Dashboard
% ¦—?'•- Database
5 i P ¦ •• ••* -•—^-^81 Rrte
3 ^ ~ Dashboard Notification^-
sj Local User * Server ^
1 r i * l
Ajarf
Operational Analytics
« .^Data Store I J*™*™
m & y
13 \
< * !=££=,
z
CO
uj * Ifc #*$
!!»•»»• D'
_ OWQM-DS LIMS Customer Work Order
"5 Information System
o System
OWQM-DS: Online Water Quality Monitoring in Distribution Systems LIMS: Laboratory Info'maticr Management System
CIS: Geographic Information System
Figure 6-3. Example of a Dedicated Information Management System
This type of architecture can also incorporate more powerful analytics and visualization approaches, such
as a dashboard, which is a visually oriented user interface that integrates and displays data from multiple
sources spatially and graphically. An example of a GIS-based dashboard designed to display data from
monitoring stations is shown in Figure 6-4. Additional information resources that can support the
interpretation of OWQM-DS data, such as LIMS and customer complaint information, can be
incorporated into a dashboard design. Presenting information from a variety of resources in a spatial
context can be valuable during the investigation of a water quality anomaly, as discussed in Section
7. Dashboard Design Guidance for Water Quality Surveillance and Response Systems provides additional
information about the features and design of dashboards.
29
-------
Online Water Quality Monitoring for Distribution Systems
Surwillanc® and ftespoits* S/ttem
jur/towin, USA
Icon
Description
Display OWQM-DS
Layer
Display Customs'
Comofaint Information
Layer
Display S&A Layer
Display Alert
Management Too!
Control Layers
Monitoring Station
Operating Normally
Monitoring Station
with Alert
S&A Sampling
Location wth'Chlorne
Values in Range
S&A Sampling
Location win High
Chlorine Values
Customer Complaint
Information Call
Displayed
Icon
tr
O
0
O
o
*v
Figure 6-4. Example of a Dashboard Display
Hosted Solutions
Hosted solutions can also be used for information management systems. The decision to use a hosted
solution should be based on whether there is a preference to procure and manage an information system
within a utility or contract information management capabilities as a third-party service. A hosted solution
generally has low capital costs and does not require the time and technical expertise necessary to
implement and maintain a new system. A hosted solution may also be used temporarily if responsibilities
for managing the information management system will change. Under this scenario, a hosted solution
may be used initially to reduce capital costs and internal staffing requirements. Once a utility is ready to
take over responsibility for the system, it can be migrated to an internally managed system.
A hosted solution can vary in complexity (e.g., if supplemental data is integrated) and can be used to meet
all or some requirements of an information management system. Potential functions of a hosted solution
can include data storage, data analysis, and data access. If a hosted solution is used for data storage,
OWQM-DS data (as well as relevant supplemental data, if desired) is transferred directly from each
station's local computing element to the external system, often via the internet. For external data analysis,
relevant data is transmitted to an external system, and the output is returned to the utility and displayed
through visualization tools. Hosted solutions generally contain a user interface for data access and
analysis, as well as alert notification and tracking. Users can often access the password-protected system
from any device connected to the internet, including work or home computers and mobile devices.
Some challenges with this approach can include security concerns during data transmission or storage,
potential difficulty using the information stored in the hosted system outside of its provided functions,
and loss of ownership of the data. However, data security concerns are currently being addressed by
technology providers that are developing large data portals for smart city programs (e.g., Microsoft
Trusted Data Platform, Amazon Web Services, Cisco CDP, Mtuity Atlantis).
Many vendors now collect, store, and process data and provide a user interface to the data using a
proprietary cloud. However, diis approach can present a potential issue or concern when data in the
proprietary cloud requires integration with other data that resides within other utility information
management systems. Where a vendor offers data in a proprietary cloud, there should be an option to
30
-------
Online Water Quality Monitoring for Distribution Systems
automatically pass this data to a utility's information management system for further processing and
storage. This strategy allows for data from all sources to be stored in a normalized database for processing
and the information generated to be utilized by a dashboard.
A hosted solution can be used for a SCADA system architecture or a dedicated information management
system architecture. In both cases, the analytics and data storage can be provided by cloud services. For a
SCADA-hosted architecture, the functions provided by the SCADA server and SCADA historian shown
in Figure 6-2 would be in the cloud. For a dedicated information management system, the hosted
architecture would provide the functions shown in the analytical infrastructure layer in Figure 6-3.
6.4 Information Management System Requirements
An effective information management system provides users with the information they need when they
need it and in a usable, easily consumable format. Information management systems for OWQM-DS are
unique for every utility due in part to differences in existing systems and capabilities, expertise of utility
personnel responsible for developing and using the systems, expected uses of the system, and resources
available to develop the system.
Section 4 of the Guidance for Developing Integrated Water Quality
Surveillance and Response Systems describes a methodical process
for selecting and implementing an information management
system. An important first step in this process is developing
requirements for a system in the following two categories.
• Functional requirements define key features and
attributes of a system that are visible to end users.
Examples of functional requirements include the way data
can be accessed, the types of tables and plots that can be
produced through a user interface, the means by which
alerts are transmitted to utility personnel, and the ability to
generate custom reports. Functional requirements are
generally defined by end users.
• Technical requirements are system attributes and design
features that are often not readily apparent to end users but
are essential to meeting functional requirements and other design constraints. Examples of
technical requirements include system availability, information security and privacy, backup and
recovery, data storage needs, and integration requirements. Technical requirements are generally
developed by IT personnel or derived from IT standards.
The Information Management Requirements Development Tool is a software tool designed to help users
define, prioritize, and document requirements for an information management system. This tool is
populated with functional and technical requirements commonly used to support OWQM-DS operations,
ft also allows users to generate a consolidated list of functional and technical requirements that can be
used to develop design and/or bid documents.
Functional Requirements
Before developing functional requirements, expected uses of an information management system for
OWQM-DS should be defined. Expected uses are simply the ways users expect to interact with a system.
For example, users may want to review recent distribution system water quality data to guide system
operations, be notified of anomalous water quality conditions, or access a variety of information resources
to investigate the cause of a water quality anomaly. Section 4.2 of Guidance for Developing Integrated
Helpful Hint
Throughout the selection and
implementation process,
information technology (IT)
personnel and staff responsible
for OWQM-DS implementation
should compile requirements
across users and prioritize them
in light of design goals,
performance objectives, and
available resources.
Coordination between the
design team and end users is
needed to precisely describe
and refine necessary
functionality.
31
-------
Online Water Quality Monitoring for Distribution Systems
Water Quality Surveillance and Response Systems provides guidance on identifying expected uses of a
system, as weli as their relationships to functional requirements.
The expected uses of an information management system should guide the development of detailed
functional requirements. Table 6-2 provides examples of functional requirements.
Table 6-2. Examples of Information Management System Functional Requirements
Requirements
Description
Robust time-series plotting
The system should allow users to create time-series plots that display multiple
parameters on the same plot. Users should be able to customize the plots,
including specifying the time-period and parameters to display.
Threshold analysis and
alerting
The system should be able to produce automated alerts if data values fall outside
of pre-defined thresholds. These thresholds should be configurable and specific to
each parameter.
Automated report generation
The system should provide automated reports that can be customized by individual
users. These reports should provide analysis output, time-series plots, statistical
summaries of validated data, and a summary of flagged data.
Data export
Users should be able to export data from the system in a format that can be
imported into analytical software, such as a spreadsheet program.
Incident reporting
Users should be able to obtain a list of water quality incidents based on user-
defined filter criteria such as type of incident, date range, or cause of incident.
Automatic data validation
The system should perform real-time data validation and flag data points using
system and/or user-defined logic.
Alert notifications
The system should contain flexible and robust alert transmission options. These
can include:
• The ability to receive and view alert details on mobile devices, including
smartphones
• The ability to identify the staff member who should receive alert notifications
based on the time of day and day of the week
• The ability to resend notifications if acknowledgement is not received in a pre-
defined time-period; the notification can be re-sent to either the intended receiver
or sent to someone else
Investigation tracking
The system should allow users to enter and view real-time alert investigation
details through the user interface. The system should require alerts to be
acknowledged and then to be closed out within a defined time-period.
Other data sources
The information management system should provide access to the latest
information from the following resources:
• Customer complaints database
• Work order system
• LIMS
Remote access
Users should be able to access the system remotely (i.e., outside of the utility's
computer network) over a secure connection.
GIS-based presentation
of monitoring station
operating status
Colored icons should be used to identify the current operating status of each
monitoring station on the GIS display using the following attributes:
• Blue - Normal operation, all systems operating
• Yellow - Some subsystems (e.g., sensors) not functioning to specification
• Red - Station producing an ADS alert
• Gray - Station not communicating and assumed to be offline
32
-------
Online Water Quality Monitoring for Distribution Systems
Requirements
Description
Mouse over and drill down
capability
When users hover over an icon on the map, a pop-up box should appear that
displays high-level information associated with the icon (e.g., hovering over a
monitoring location should display data values, instrument status, a list of
hydraulically connected locations, and its physical location).
These pop-up boxes should contain hyperlinks that take the user to detailed
information, such as time-series plots of OWQM-DS data or guides for
investigating water quality at a location.
Display of overlays
Multiple overlays can be displayed at the same time. Overlays that may be
displayed concurrently include:
• Distribution system model showing mains, pressure zones, and utility facilities
• Map view showing major highways and community facilities
• Monitoring station location and status
• Latest work order locations
Hydraulic modeling
and forecasting
The system should be able to trace a water quality incident from a specified
location in the distribution system and predict arrival times at all downstream
monitoring locations.
Technical Requirements
Technical requirements are often dependent on functional requirements and should be developed after the
functional requirements have been defined. Generally, development of technical requirements is the
responsibility of IT personnel who consider the technical aspects of the information management system
design that are necessary to meet functional requirements. Technical requirements will also be informed
by IT policies such as security protocols and the need to adapt the system over time to incorporate new
functions, datastreams, and features. Table 6-3 provides examples of technical requirements.
Table 6-3. Examples of Information Management System Technical Requirements
Requirements
Description
Design flexibility and ability
to accommodate changing
requirements
The system should have the flexibility to change key parameters and display
features without modifying the underlying code. These changes include updates to
the user interface as well as adding, removing, or changing datastreams,
monitoring locations, and external data sources.
Encryption
All interactions with the information management system should be encrypted via
Secure Socket Layer.
Individual login accounts
Users should have individual login accounts, and administrators should be able to
track system use for individual users. Tracking may include the times a user is
logged on and actions taken.
Map service utilization
The information management system should be able to read and display map
services provided by the utility's GIS using a configurable list of map services.
Operational data store size
The operational data store should provide ready access to the last 90 days of data
for all source data systems used in the information management system.
Parameter data storage
The information management system should provide storage of datastreams for
spectral profiles (256 datapoints per sample).
Vendor-neutral platform or
open architecture
The system must be able to interface with any vendor's technology.
Minimal programming
expertise required
The system should be able to be installed and configured by staff with intermediate
skills and experience designing or implementing IT systems. Installation can
require some integration and design work, but should not require extensive coding
or engineering.
33
-------
Online Water Quality Monitoring for Distribution Systems
Section 7: Alert Investigation Procedure
An alert investigation procedure for OWQM-DS formalizes and standardizes the investigation of
distribution system water quality anomalies. Such a procedure is triggered by notification of an alert and
continues until (1) an explanation for the alert is identified, or (2) all activities are completed and water
contamination cannot be ruled out.
Target Capability
A procedure has been developed, documented, and put into practice that facilitates timely and efficient
investigation of alerts.
This section describes considerations for development of an alert investigation procedure for OWQM-DS
and covers the following topics:
• Developing an effective alert investigation procedure
• Developing tools to support investigations
• Preparing for real-time alert investigations
7.1 Developing an Effective Alert Investigation Procedure
This section describes an approach for developing an alert investigation procedure, which consists of the
following three activities:
• Define potential alert causes
• Establish an alert investigation process
• Assign roles and responsibilities
Define Potential Alert Causes
Consideration of potential alert causes provides a useful starting point when developing an alert
investigation procedure as they help define types of information to consider during the investigation.
Table 7-1 lists and describes the most common causes of both invalid alerts (not due to anomalous water
quality conditions) and valid alerts (triggered by anomalous water quality conditions) based on
experience from utilities that have implemented OWQM-DS systems (EPA, 2014).
Table 7-1. Common High-level Causes of Alerts
Alert Cause
Description
Invalid
Alerts
OWQM-DS equipment
issue
An alert triggered by inaccurate data values due to a malfunction or failure of
station hardware, data transmission, or data handling
Data analysis issue
An alert caused by an artifact of an ADS (i.e., an alert is produced even though
data is within the normal range of values and variability)
(0
Treatment or distribution
system equipment
malfunction
An alert due to a water quality change caused by a malfunction or failure of
treatment equipment (e.g., chlorine feed system) or distribution system
equipment (e.g., pump)
•c
Distribution system issue
An alert due to a water quality change caused by distribution work or a
distribution system upset, such as a main break or pressure surge
Water contamination
An alert triggered by accidental or intentional introduction of a foreign
substance into a distribution system, which may or may not be harmful
34
-------
Online Water Quality Monitoring for Distribution Systems
Establish an Alert Investigation Process
The core of an effective alert investigation procedure is a detailed, step-by-step alert investigation process
for identifying the cause of an alert. The alert investigation process is generally structured to consider the
most likely causes first, allowing contamination to be quickly ruled out for the majority of alerts. If no
cause can be identified by the end of the process, water contamination is considered possible and
designated personnel are notified.
Each step of an alert investigation process should be documented and include the following information:
• Detailed instructions for completing the step
• Roles and names of specific individual(s) responsible for completing the step
• Information resources that should be consulted during the step
• Actions that should be taken, including personnel to be notified, upon completion of the step
The Alert Investigation Procedure Template provides a
framework for documenting an alert investigation process
clearly and completely. The template, which includes a
checklist, can be opened by clicking on the box to the
right.
An alert investigation process can be visually depicted in a
diagram that shows the progression of steps through the
entire process. This simplified representation of the process allows individuals with responsibilities for
discrete steps to see how their activities support an overall investigation.
Figure 7-1 provides an example of an alert investigation process diagram. The major steps and decision
points are shown in the flow chart on the left side of the figure. Additional detail on the actions
implemented are shown to the right in the figure.
A range of estimated times for properly trained personnel to complete steps in the investigation is shown
on the left edge of Figure 7-1. Totaling the length of time required for each step yields the time required
for a full investigation. The shortest investigations are those in which an alert cause is identified in an
early step (e.g., if the data is found to be inaccurate and only the first step of the investigation process
must be completed).
Alert Investigation Procedure j
Template (Microsoft Word)
*Note that the document that is
currently open may need to be
downloaded and opened offline to
access this template.
35
-------
Online Water Quality Monitoring for Distribution Systems
ALERT INVESTIGATION PROCESS
ACTIONS IMPLEMENTED
H
Utility personnel
alerted to potential water
quality anomaly
Personnel evaluate underlying
water quality data to determine
il alert is valid
. YES f
r
Does available information
indicate that water quality
data is inaccurate?
NO
*
Is water quality data
consistent with normal
values and variability?
NO
T
Personnel consider
common causes of water
quality anomalies
Can a benign cause
for the water quality change
be Identified?
NO
t
Personnel inspect the
monitoring station far signs
of malfunction
.YES
Alert is invalid.
Close investigation,
initiate equipment
maintenance if
necessary, and return
to normal operations
Alert is invalid.
Close investigation
and return to
normal operations
YESl
Ctose investigation,
take corrective
action if necessary,
and return to
normal operations
r
Can issue with
station hardware
explain abnormal data?
A
NO
*
Water contamination
L
is possible
J
YES
Alert is jnvalkt.
Close investigation,
initiate equipment
maintenance if
necessary, and return
to normal operations
V "
Figure 7-1. Example Alert Investigation Process Diagram
Review water quality data at
alerting monitoring station for:
• Information from remote diagnostics
• Date patterns that match known
sensor malfunctions
• Inslrument or communications
system faults
• Station flow or pressure outside of
acceptable operating range
Review available inforrnalion resources
to determine whether any of the tollow-
ing caused (he water quality anomaly:
¦ Cftartges in treatment processes
» Granges in system operations
(e.g.. valving, pompirvg)
» Changes in system pressure or flow
« Distribution system work
• System upsats
(e.g., main breaks)
Personnel visit the monitoring location
to verify:
* Sufficient supply of power to
instruments and ancillary equipment
* Sufficient flew of sample water to the
station and rsiamems
* Proper functioning of instruments and
slation communications
« Normal envronmental conditions
(e.g., temperature and humidity)
OSMcmlw t03 4
36
-------
Online Water Quality Monitoring for Distribution Systems
The specific actions included in an alert investigation process depend largely on the availability of
relevant information and how it can be accessed (e.g., through an information management system or by
calling various utility departments). It is important to identify these information resources as an alert
investigation process is being developed. Table 7-2 provides examples of supplementary information that
may be useful during an investigation. Data available through existing information management systems
may impact the activities included in an alert investigation process and, conversely, the information
needed to support investigations may inform information management system requirements (see
Section 6.4).
Table 7-2. Additional Information Sources for Use in an Investigation
Information Type
Example Information
Value to Alert Investigation
Instrument performance
indicators
Instrument error codes, remote
diagnostics indicators, and data
quality indicators
Assist in determining whether data is valid;
malfunctioning instruments may generate error
codes useful for assessing data quality
Communications system
status indicators
System error codes,
communications status between
the monitoring station and the
central location
Assist in determining whether data is valid;
malfunctioning communications may result in
missing or incorrect data
Instrument maintenance
records
Maintenance records and
instrument logs
Assist in determining whether data is valid;
instrument maintenance activities and ongoing
instrument performance issues may result in
missing or inaccurate data points
Distribution system
operations information
Tank levels, pump status,
valve status, system flow, and
pressure data
Support investigation of the cause of water
quality changes and the impact of operations on
water flow paths; changes in operations often
cause changes in system flow paths and
corresponding changes in the water quality
measured at a monitoring location
Treatment process
information
Treatment process settings,
chemical doses, and treatment
process monitoring data
Support investigation of the cause of water
quality changes and identification of water
source(s) supplying each monitoring location;
treatment process settings often affect
distribution system water quality
Distribution system
work records
System maintenance records and
work orders
Support investigation of the cause of water
quality changes; system work or upsets can
impact system flow paths and water quality
Customer complaint
records
Customer complaint entries
Support investigation of the cause of water
quality changes; some water quality changes
impact water aesthetics, which customers can
detect and report
Modeling results
System flow rate and flow path
Support investigation of the cause of water
quality changes; modeling results can provide
insight into flow paths between monitoring
locations
Sampling results
Results from analysis of grab
samples, including those collected
during inspection of the alerting
monitoring station
Assist in determining whether data is accurate,
and provide water quality data from additional
monitoring locations to support investigation of
the cause of water quality changes
Calendar of regional
events
Date and time of large community
events
Support investigation of the cause of water
quality changes; large events that significantly
alter water demand in a specific area can
impact system flow paths and water quality
37
-------
Online Water Quality Monitoring for Distribution Systems
An alert investigation process may point to the need to make additional datastreams available to
investigators or improve access to existing datastreams. Desired updates to information management
systems should be noted during development of the alert investigation procedure. This information is
particularly useful for developing requirements if a new information management system will be
implemented or if existing systems will be updated.
Using Data from Hydraulically Connected Monitoring Stations
Data from hydraulically connected monitoring locations upstream and downstream of a location where an alert: is
generated can provide the following insight during an investigation:
® The presence of a similar water quality change at more than one location can increase confidence that a
water quality change is real.
• If a water quality change can be seen at an upstream location, that information can be used to identify
potential sources of a change. Conversely, if a change was not present upstream, it can be concluded that
the cause of a water quality anomaly occurred between the locations or that the alert is due to a station-
specific issue.
® Water quality at downstream locations can be reviewed to see whether the anomalous water quality arrives
there in a time-period consistent with the hydraulic travel time between locations.
An example of how OWQM-DS data has been used to support investigations is displayed in the graph below.
This graph shows how online turbidity data generated at hydraulically-connected monitoring stations can be used
to detect and track the impact of a transmission main break in an upstream area of the system.
40
30
20
10
o
4
3
2
1
Monday 12:00 AM Monday 12:00 PM Tuesday 12:00 AM Tuesday 12:00 PM Wednesday 12:00 PM
Date/Time
Assign Roles and Responsibilities
Once an alert investigation process is defined, responsibility for every activity must be assigned to one or
more individuals. Roles for alert investigations should align with existing job functions. Leveraging
existing expertise in this manner can reduce the amount of new training required and can result in
increased acceptance of new responsibilities for investigating alerts.
Arrangements should be made for providing constant coverage of alert investigation responsibilities
Approaches to ensuring around-the-clock coverage include:
38
-------
Online Water Quality Monitoring for Distribution Systems
• Training personnel from all shifts on the alert investigation procedure
• Assigning backup personnel for each activity for cases when the primary investigator is
unavailable
• Cross-training investigators on multiple roles
• Assigning personnel to be on-call for critical investigative functions, particularly those requiring
a decision about the validity of an alert
Table 7-3 provides example of generic roles and responsibilities for investigating an alert.
Table 7-3. Example of Generic Roles and Responsibilities for Alert Investigations
Role
Alert Investigation Responsibilities
Water quality manager
• Receives alerts
• Manages investigation of alerts
• Facilitates communication among investigators
• Decides whether an alert is valid and indicative of possible contamination
Water quality specialist
• Leads or assists with the investigation of alerts using knowledge of the distribution
system and historical water quality
System operator
• Provides information on plant or system operations as needed
• Collaborates with alert investigators about potential causes for changes in water quality
Distribution system
maintenance staff
• Provides information about current distribution system operations and maintenance
activities, as well as any system upsets
Sensor technician
• Provides information about recent sensor issues and equipment maintenance
• Assists in the investigation of alerts by inspecting OWQM-DS equipment to determine
whether it is operating properly
7.2 Developing Investigation Tools
This section describes tools that can be developed from this documentation to assist investigators
efficiently carrying out their responsibilities. The investigation tools that will be discussed in this
include:
• Checklists
• Records of previous alert investigations
• Quick reference guides
• Other information sources
Checklists
Alert investigation checklists guide personnel through their investigative responsibilities and allow them
to document activities and findings. The checklists can ensure consistency among investigators, verify
that all activities are completed, and reduce the time required to conduct alert investigations. Checklists
generally list the activities assigned to a specific individual, and thus multiple checklists may be
developed to support an alert investigation procedure.
A checklist should be streamlined, concise, and intuitive for personnel trained on the corresponding
procedure. It should guide personnel through the steps of an investigation and provide space for them to
record important information (e.g., an explanation of the cause of the alert, the explicit data used) for each
activity completed. In some cases, it may be sufficient to simply check a box indicating completion of an
activity. In others, an investigator may need to record a time or provide more details on activities or
conclusions.
in
section
39
-------
Online Water Quality Monitoring for Distribution Systems
Record of Alert Investigations
It is important to formally document each alert investigation, including the steps implemented,
information used, and the likely cause of an alert. In addition, it can be valuable to retain resources used
during an investigation, such as screen shots of the water quality changes that triggered an alert. These
records can be used to monitor the frequency of alerts by cause categories and can serve as a resource
during investigation of future alerts. For example, if an investigator cannot readily identify the cause of a
water quality change, the records can be filtered by location or parameter impacted to see whether a
similar change triggered an alert in the past and, if so, whether a cause was identified.
Table 7-4 provides examples of alert categories that can be used to populate a record of alert
investigations. While these examples require users to select a main category and a subcategory, they
could be adapted to use only one level of classification.
Table 7-4. Example of Alert Categories
Main Category
Subcategory
Invalid Alert
Inaccurate data: sensor issue
Station power or flow loss
Data collection failure
No significant deviation from normal water quality
Other
Valid Alert, Cause Identified
Verified non-standard system operation
Treatment plant change or upset
Distribution system work
Main break
System pressure or flow anomaly
Other
Valid Alert, Cause Not Identified
Possible contamination
There are a variety of ways to document alert investigations. Simple solutions include keeping a
spreadsheet on a shared drive or a paper record in a central location. Figure 7-2 provides an example of a
spreadsheet record in which investigators can select from the pre-defined alert categories shown in
Table 7-4. This example shows key fields that are recommended for inclusion in a record of alert
investigations. A record could be expanded to list the water quality parameters that triggered an alert,
personnel who supported the main investigator, and any other information deemed useful to the record.
40
-------
Online Water Quality Monitoring for Distribution Systems
A
B C
D
t
f
G
I
Alert Date/Time Alt a Location Investigator
investigation
Sun Dj isi1 Time
Investigation
End Date/Time
Alert Cause Category
Notes
2
5/4/2017 2; 15
Park SI
Stof age Tank John Webber
5/4/2017 2 20
Invalid atert Inaccurate
5/4/2017 3,15 data sensor tssue
Chlorine sensor obvtously
malfunctioning
i
5/14/2017 13 22
South Pump
Station John Webber
5/ 14/2017 13 35
Valid alert cause
identified: Verified non-
standard system
5/14/2017 14 00 operation
The unusual change in
water quaMy can be
attributed to opening of
Valve Ma
4
6/1/2017 U 66
University
Hospflai Andre Brown
6/1/2017 1610
6/1/2017 15 30
Invalid alert No significant
deviation from normal
water duality
The water nudity change is
consistent with what is seen
when T ank A begins
• Saung
In/valid Alert. IruccurAte £Sa1 a smwoi
rtt, iTiraimrmtrzrm
irwairf iflert Oal® coDcdlon
Figure 7-2. Example Spreadsheet for Documenting Alert Investigations
If a dashboard will be used to support an SRS. electronic alert investigation tracking may be incorporated
into its design. For example, electronic checklists can be developed that automatically enter investigation
records into a data management system, facilitating further analy sis and use of die records.
See Dashboard Design Guidance for a Water Quality Surveillance and Response System for more
information.
Quick Reference Guides
While many alert investigation activities become second nature to investigators, additional tools may be
useful for guiding investigators through complex or less frequently implemented tasks. Development of
quick reference guides, in which key information is concisely summarized in an easily accessible form,
ensures investigators can quickly and easily get the information they need.
Examples of quick reference guides that can be useful for OWQM-DS include:
• A list of contact information for all individuals who investigators may need to contact during alert
investigations
• Monitoring location-specific guidelines for investigating water quality anomalies that include
details such as a summary of system facilities that can impact the location's water quality
• A summary of OWQM-DS equipment in use, including instrument faults produced and common
issues encountered
• A distribution system map or a summary of the connectivity between monitoring locations
• A list of water quality changes that occur under different conditions (e.g., nitrification, pressure
transients, chemical contamination, biological contamination).
7.3 Preparing for Real-time Alert Investigations
This section describes a suggested process for putting an alert investigation procedure into practice.
Effective implementation is crucial, as the benefits of OWQM-DS can be fully realized only if alerts are
investigated and responded to appropriately. The following topics are covered in this section:
• Training
• Preliminary operation
• Real-time operation
41
-------
Online Water Quality Monitoring for Distribution Systems
Training
Proper training ensures that all utility personnel with a role in the investigation of an alert are aware of
their responsibilities and have the knowledge and expertise needed to execute those responsibilities. It is
suggested that training on an alert investigation procedure include the following:
• An overview of the purpose and design of OWQM-DS
• A detailed description of an alert investigation procedure and the role of each participant
• A review of checklists, quick reference guides, user interfaces, and other tools available to
support alert investigations
• Instructions for both entering new records of alert investigations and retrieving previous records
to support new alert investigations
Section 6 of Guidance for Developing Integrated Water Quality Surveillance and Response Systems
provides general guidance on implementing a training and exercise program. In general, classroom
training is used first to orient personnel to a procedure and their responsibilities during alert
investigations. Once personnel are comfortable with a procedure, drills and exercises give them the
opportunity to practice performing their responsibilities in a controlled environment. The SRS Exercise
Development Toolbox is an interactive software program designed to assist utilities in the design and
execution of exercises.
Preliminary Operation
Following initial training, a period of preliminary operation allows personnel to practice their
responsibilities in test mode before transitioning to real-time operation. For example, staff can be asked to
investigate alerts (either all alerts or a subset) in batches as they have time—not necessarily as alerts are
generated. At this stage, investigators may or may not receive alert notifications such as emails or text
messages.
During preliminary operation, it may be useful to hold regular meetings with all investigators to discuss
recent data and alerts. It is generally most effective if participants are asked to perform specific analyses
or alert investigations before each meeting, and then discuss
conclusions, observations, insights, and challenges as a group.
These meetings can be held frequently initially, but become
less frequent as proficiency increases and issues are resolved.
Meeting once or twice per month during the period of
preliminary operation would be appropriate and sufficient for
most OWQM-DS design goals.
Preliminary operation provides excellent opportunities to
refine an alert investigation procedure and investigation tools.
Based on feedback from investigators, responsibilities can be
clarified, unnecessary steps can be eliminated, existing tools can be refined, new tools can be developed if
needs are identified, and roles can be better integrated into existing job functions.
Real-time Operation
During real-time operation, alerts are investigated as they are generated, and CM is activated if a
contamination incident is considered possible. The transition from preliminary operation to real-time
operation should be clearly communicated to all utility personnel with a role in alert investigations. This
includes establishing a date for the transition to real-time operation and providing expectations for how
alert investigations will be performed and documented.
Value of
Preliminary Operation
Do not rush preliminary operation. It
provides an opportunity for
personnel to practice their
responsibilities and learn the data
used during investigations, thus
improving the efficiency of alert
investigations.
42
-------
Online Water Quality Monitoring for Distribution Systems
After transitioning to real-time operation, it is important to continue to oversee and support investigators.
Documentation of alert investigations should be regularly reviewed to ensure that all personnel are
accurately and thoroughly earn ing out their responsibilities, and individual instruction should be
provided to individuals who are not doing so. Ongoing drills, exercises, and training are important to
ensure that staff remain familiar with their responsibilities and to address any changes such as updates to
a procedure or investigation tools. Finally, it is important to thoroughly train new staff on their
responsibilities and the analysis of OWQM-DS data.
Regularly Review and Update the Alert Investigation Procedure
Routine updates to the alert investigation procedure and investigation tools are necessary to maintain their
usefulness. Recommendations for procedure maintenance include these:
• Designate one or more individuals with responsibility for maintaining alert investigation materials.
• Establish a review schedule (annual reviews should suffice in most cases).
• Review the record of alert investigations, conduct tabletop exercises, and solicit feedback from investigators
to identify necessary updates.
• Establish a protocol for submitting and tracking change requests.
43
-------
Online Water Quality Monitoring for Distribution Systems
Section 8: Preliminary Design
The information presented in the previous sections of this document can guide development of a
preliminary design of an OWQM-DS system that supports a utility's design goals and performance
objectives. If OWQM-DS will be a component of a multi-component SRS, the design of the integrated
system will likely be guided by a project management team. In this case, guidelines for design of the
individual components should be provided to the personnel implementing the components and should
include the following:
• Overarching design goals and performance objectives for the SRS
• Existing resources that could be leveraged to implement the SRS components, including
personnel, procedures, equipment, and information management systems
• Project constraints, such as budget ceilings, schedule milestones, and policy restrictions
• Instructions or specific guidelines for the development of preliminary component designs
If an OWQM-DS system will be part of a larger SRS, it should be incorporated into a master plan, as
described in Section 3 of Guidance for Developing an Integrated Water Quality Surveillance and
Response System. Master planning for an SRS involves the development of a complete SRS design, which
can be implemented in phases based on available resources.
Regardless of whether OWQM-DS will be developed as a stand-alone component or as part of a
multi-component SRS, the preliminary design should be documented in sufficient detail to assess whether
it can achieve the selected design goals within project
constraints. The Preliminary Design Template can be
opened and edited by clicking on the box to the right.
Utilities can update and expand on this template throughout
the design process until the final design is completed. A
complete design for an OWQM-DS system may be
captured in a number of technical documents and
specifications that support the overarching design
document.
This template covers the following aspects of the design of an OWQM-DS system:
• Component implementation team. Identify personnel from the various departments of a utility
who will have a role in the design, implementation, operation, and maintenance of an OWQM-DS
system. Document the role, responsibilities, and estimated time commitment of each team
member.
• Design goals and performance objectives. Use the overarching SRS design goals and
performance objectives to develop design goals and performance objectives that will guide the
design of an OWQM-DS system.
• Monitoring locations. Identify preliminary monitoring locations and briefly describe the
rationale for location selection. If necessary, prioritize the locations and include potential backup
locations should a preferred location prove infeasible.
• Water quality parameter selection. List the parameters to be monitored and briefly describe the
rationale for parameter selection.
• Monitoring station design. Summarize the key attributes of the design of the monitoring station
that will be installed at each monitoring location.
Preliminary Design Template
(Microsoft Word)
*Note that the document that is
currently open may need to be
downloaded and opened offline to
access this template.
44
-------
Online Water Quality Monitoring for Distribution Systems
Information management requirements. Summarize the preliminary functional and technical
requirements for an information management system designed to support operation of a utility's
OWQM-DS system.
Training Plan. Describe the training that will be provided to utility personnel to support
OWQM-DS system operations.
Budget. Provide an order-of-magnitude budget for OWQM-DS system implementation.
Schedule. Provide a preliminary schedule for implementation of an OWQM-DS system.
An OWQM-DS system can be implemented in phases, which can allow a utility to incorporate lessons
learned from early phases into the final system, accommodate
budgetary constraints, provide adequate time for training, and
allow personnel to gradually acclimate to the new system.
If multiple designs emerge during the design process, an
evaluation of alternatives should be conducted to consider the cost
and benefits associated with each. For example, some alternatives
may offer tradeoffs between the number of parameters monitored
and the number of monitoring locations. Each of these alternatives
will likely have different capabilities and a different cost for
procurement, operation, and maintenance throughout the life of
the system. Framework for Comparing Alternative Water Quality
Surveillance and Response Systems provides a systematic process for comparing alternative designs that
considers both the capabilities and cost of each design.
Helpful Hint
It can be useful to develop a
preliminary alert investigation
procedure in parallel with
developing a preliminary design
of an OWQM-DS system.
Information in this procedure
can inform various aspects of
the design, such as information
management requirements.
Funding Opportunities
Both financial and personnel resources are required to implement an OWQM-DS system. There are a variety of
methods to fund such a project. For information on current federal, state, local, private, and other sources of
funding that your utility may be able to use to implement a system, visit EPA's Water Finance Clearinghouse.
45
-------
Online Water Quality Monitoring for Distribution Systems
Section 9: Example Applications
This section provides examples of OWQM-DS applications that align with the design goals discussed in
Section 2.1. Each example includes a summary of a given application and describes types of monitoring
locations and water quality parameters that can facilitate that application.
9.1 Monitoring for Contamination Incidents
Contamination incidents in distribution systems can include both intentional incidents (e.g., malicious
insertion of contaminants into a system) and unintentional incidents (e.g., cross-connections). Research
has shown that the most useful water quality parameters for detecting contamination are chlorine residual,
pH, specific conductance, temperature, DOC/TOC (or a surrogate), ORP,and spectral absorbance (EPA,
2005a; EPA, 2005b; EPA, 2005c; EPA, 2009; and Allgeier, et al, 2010).
Although contamination incidents are rare, the consequences can be extreme. OWQM-DS data can be
used to detect contamination incidents in sufficient time to implement response actions that limit the
spread of a contaminant in a distribution system and protect public health (i.e., limit fatalities and
illnesses).
Monitoring Locations
The uncertainty of the location and extent of a potential contamination incident can make it a challenge to
identify effective monitoring locations. The approach commonly used for such a problem is to optimize
locations for a specific objective, such as minimizing the time to detection. The objective most often used
in reported studies and system designs is to optimize locations to reduce overall consequences from a
large ensemble of simulated contamination incidents. In other words, locations are selected to provide
rapid detection of simulated contamination incidents that produce the most severe consequences.
Monitoring at the following types of monitoring locations can provide information that can be used to
detect contamination incidents:
• Entry points to distribution system. Provide a baseline for distribution system water quality.
• Locations identified using station placement optimization tools. Maximize the effectiveness of
an OWQM-DS system to detect contamination with respect to a specific objective, such as
minimizing the time to detection or minimizing consequences from a large ensemble of simulated
contamination incidents.
• Locations identified by distribution system models or operator experience. Can be used to
detect contamination incidents that have the potential to impact a large population (e.g., incidents
at, or directly downstream of, large storage reservoirs or pump stations).
46
-------
Online Water Quality Monitoring for Distribution Systems
Water Quality Parameters
Table 9-1 presents the water quality parameters that can be monitored to detect contamination incidents.
Table 9-1. Water Quality Parameters for Broad-Spectrum Monitoring for Contamination
Role in Application
Parameter
Purpose of Monitoring Parameter
Sufficient for detecting a wide
range of contaminants
Chlorine
residual
• Many chemical and microbial contaminants will react with
chlorine residual, thus decreasing the residual
concentration.
pH+
• Chemical contaminants with acidic or basic functional
groups can change the pH; however, the magnitude of a
change in pH will be inversely related to the buffering
capacity of the water.
Specific
conductance*
• Some chemical contaminants have charged functional
groups that can dissociate and form ionic species when
dissolved in water, thus increasing the specific
conductance of the water.
• A measurable change in specific conductance may only
occur when contaminant concentrations are relatively high.
Temperature*
• A rapid change in temperature can indicate a large inflow
of a foreign fluid (e.g., cross-connection and backflow from
an industrial customer).
Spectral
absorbance
• Some inorganic and most organic chemicals absorb in the
UV-visible spectrum. As such, a change in spectral
absorption may indicate the presence of a chemical
contaminant.
• Some spectral instruments provide a spectral fingerprint. A
change in the spectral fingerprint from an established
baseline can indicate the presence of a contaminant.
Increase the number of
contaminants that can be
detected and the degree of
confidence in contaminant
detection
DOC/TOC
• Many contaminants of concern are organic chemicals, and
the presence of these contaminants can increase
DOC/TOC concentrations.
• An increase in DOC/TOC can exert a chlorine demand,
reduce the chlorine residual concentration, and create an
opportunity for the survival of chlorine-sensitive pathogens
(e.g., E. coli) and biotoxins (e.g., microcystins).
ORP
• A change in ORP can indicate the presence of a
contaminant with oxidizing or reducing potential.
• Can be used to confirm changes in chlorine residual
concentrations.
UV-254
• UV-254 and DOC measurements can be used to calculate
SUVA (specific ultraviolet absorbance).
• A change in SUVA can indicate a change in the organic
composition of the water, potentially indicating the
presence of an organic contaminant.
t Core parameters
Figure 9-1 shows time-series plots of OWQM-DS data following the addition of aldicarb, glyphosate,
secondary wastewater effluent, and microbial growth media (e.g., terrific broth) into drinking water. It is
important to note these examples are for illustrative purposes and summarize the best available data to
represent contamination incidents in a distribution system:
• Aldicarb was selected to represent the carbamate class of pesticides, while glyphosate was
selected to represent organophosphorus herbicides and insecticides. Following the addition of
47
-------
Online Water Quality Monitoring for Distribution Systems
both aldicarb (top left) and glyphosate (top right), free chlorine and ORP levels decreased and
TOC increased
• Secondary wastewater effluent was selected to represent contamination of drinking water with
untreated wastewater (health and safety concerns precluded the use of raw wastewater in the
study). Following the addition of secondary wastewater effluent (bottom left), free chlorine
decreased while T OC and specific conductance increased.
• Terrific broth was selected to represent contamination of drinking water with bacteria in growth
media. Because most bacteria exist as vegetative cells that are highly susceptible to inactivation
by chlorine, addition of a co-contaminant, such as terrific broth, is necessary to quench the
chlorine residual and maintain viability of the bacteria. Following the addition of terrific broth
(bottom right), free chlorine decreased and T OC concentration increased.
INJECTION
INJECTION-
Free Criloone
TOC
Free Chlorine
TOC
ORP
INJECTION-
—Free Chlorine
. —TOC
ORP
900
800 o
x
700 |
o
600 ^
500 =
400 °
¦o
o
300 ®
200 -
3
100 S
0.0
10:00
Aldicarb
11:00 12:00 13:00
400
Glyphosate
10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
13:00
Terrific Broth (E. COH surrogate)
00
10:00
12:00
13:00
Secondary Wastewater Effluent
—Free Chlorine
—TOC
INJECTION Specific Conductance
14:00 15:00 16:00 17:00 18:00
Figure 9-1. Data Following Addition of Contaminants into Drinking Water
Monitoring stations designed for comprehensive monitoring for contamination incidents can be expensive
due to the addition of water quality parameters such as spectral absorbance or DOC/TOC. By selecting
locations that can be used to collect data for other applications, the overall benefit to a utility can be
maximized.
9.2 Monitoring for Red Water and Particulate Matter Incidents
Red water incidents are typically associated with elevated levels of iron release and often caused by water
chemistry changes or hydraulic scouring of iron pipe walls. In some cases, red water can also be caused
by a failure of treatment plants to adequately remove dissolved or particulate iron from source or process
48
-------
Online Water Quality Monitoring for Distribution Systems
water. Particulate matter incidents are typically associated with turbid or cloudy water and can be caused
by hydraulic upsets due to distribution system flushing, main breaks, and pressure surges. For more
information on red water and particulate matter incidents, see the AWWA manual M58 Internal
Corrosion Control in Water Distribution Systems (Hill and Cantor, 2011).
Red water and particulate matter incidents can cause problems, such as discoloration of porcelain
plumbing fixtures in homes, stained clothing, unpleasant odors, and the release of contaminants
accumulated in pipe scales (e.g., lead, arsenic). Customers can experience these problems for days after a
large main break, both in areas where an upset occurs and in more distant areas of a system. OWQM-DS
data can be used to provide timely detection of red water and particulate matter incidents, which can
enable utilities to take corrective actions to contain and flush affected areas.
Monitoring Locations
Monitoring at the following types of monitoring locations can provide information that can be used to
detect red water and particulate matter incidents:
• Entry points to a distribution system. Provide a baseline for distribution system water quality.
• Areas where older, unlined iron pipes are in use. Indicate areas in the system that may be
susceptible to these types of incidents.
• Areas with historically high volumes of customer complaints. Indicate areas in the system that
have experienced similar types of incidents in the past that have not been addressed.
Water Quality Parameters
Table 9-2 presents water quality parameters that can be monitored to detect red water and particulate
matter incidents.
Table 9-2. Water Quality Parameters for Red Water and Particulate Matter Incident Monitoring
Role in Application
Parameters
Purpose of Monitoring Parameter
Necessary and sufficient to
monitor for red water and
particulate matter incidents
Chlorine
residual
• A reduction in chlorine residual is expected during red
water/particulate matter incidents as chlorine can react with
dissolved metal species or suspended particles.
pH+
• A reduction in pH can indicate suitable conditions for red
water/particulate matter incidents.
Specific
conductance*
• If a distribution system is supplied by multiple sources, a
change in specific conductance might indicate a change in
source water that can disrupt pipe scales, causing iron and
particulate release from pipe walls.
Temperaturet
• An increase in temperature can increase metal solubility,
increasing the potential for red water incidents from iron
release.
Turbidity
• An increase in turbidity can indicate that a red water/
particulate matter incident is occurring.
Achieve more reliable and
specific detection of red water
and particulate matter events
Apparent color
• Cloudiness or colored water (red to reddish-brown, yellow, or
black) can indicate that an incident is occurring.
DOC/TOC
• An increase in TOC with relatively stable DOC concentrations
can indicate the release of organic particulate matter from
biofilms.
t Core parameters
Figure 9-2 shows a time-series plot of monochloramine, turbidity, DOC, and TOC data during a
particulate matter incident at a utility facility. At the onset of the incident, the monochloramine
49
-------
Online Water Quality Monitoring for Distribution Systems
concentration decreased, turbidity and TOC levels increased, and the DOC concentration remained stable.
These changes were consistent with the expected parameter responses identified in Table 9-2. Following
an investigation of the water quality changes, it was determined that a particulate matter incident occurred
because of a major pipeline operation that caused a flow reversal, stirring up sediment and biofilm in the
pipeline. Utility personnel implemented response actions, and the OWQM-DS data began to return to
normal operating ranges.
Turbiditv Mowxlik* amine TOC — — DOC
Figure 9-2. Data During a Particulate Matter Incident
Figure 9-3 shows spectral fingerprints generated before and during the particulate matter incident. The
data collected during the incident, when compared to data collected before the incident, showed an
increase in spectral absorption. The "delta fingerprint" datastream indicated the difference in absorbance
between the two datasets across the UV-visible spectrum. This data was used to support the investigation
of the particulate matter incident described above.
50
-------
Online Water Quality Monitoring for Distribution Systems
22-03-201S
Figure 9-3. Spectral Fingerprints Before and During a Particulate Matter Incident
9.3 Chlorine Residual Management
Free chlorine and chloramines are residual disinfectants that are commonly used to provide continuous
control of microbial regrowth in drinking water distribution systems. However, the chlorine residual
decreases as water ages in a system. As such, the SWTR specifies that grab samples collected at a
system's entry points cannot have a chlorine residual of less than 0.2 mg/L for more than four hours and
no more than 5% of samples collected from within a system can have an undetectable concentration for
any two consecutive months. The AWWA Partnership for Safe Water's Distribution System Optimization
Program has set residual goals for free chlorine (0.2 -4.0 mg/L) and chloramines (0.5 - 4.0 mg/L), for
95% of monthly routine grab samples collected in a system (Lauer, 2010).
Utilities strive to ensure that water with a chlorine residual within desired operating limits is delivered to
all customers. Lower limits can be established based on the federal or state regulations (e.g., SWTR) and
AWWA goals mentioned above. Upper limits can be based on customer acceptance and must be below
the maximum disinfectant residual level of 4.0 mg/L established by the Stage 1 Disinfectants and
Disinfection Byproducts Rule. OWQM-DS data can be used to guide residual disinfectant dosing at
treatment plants and booster stations. It can also guide the operation of storage facilities to better manage
water age and be used to evaluate the effectiveness of flushing programs for maintaining chlorine
residuals within a target range. For more information on chlorine residual management, refer to AWWA
manual M20 Water Chlorination/Chloramination Practices and Principles (AWWA, 2006) and the
Water Research Foundation Impact of Distribution System Water Quality on Disinfection Efficacy
(Baribeau, et.al, 2005).
Monitoring Locations
Monitoring at the following types of monitoring locations can provide information that can be used to
manage chlorine residual concentrations:
51
-------
Online Water Quality Monitoring for Distribution Systems
• Entry points to a distribution system. Provide a baseline for the chlorine residual in water
entering the system.
• Storage tanks and reservoirs. Provide information that can be used to adjust the operation of
tanks and reservoirs (e.g., cycling water more frequently to reduce residence times, boosting
chlorine levels) to maintain acceptable chlorine residual concentrations. If it is not feasible to
monitor water quality on an outflow line or if it is desirable to place a station in an alternate
location to achieve multiple design goals, a station can be located in an area of the distribution
system that receives water from the storage facility.
• Outflow from disinfectant booster stations. Informs the dosing of chlorine added to water,
which can be adjusted by an operator or automated process to maintain a sufficient residual in
areas of a system that experience chronically low residual levels.
• Point of entry to critical customer facilities. Indicates whether water of an acceptable quality is
delivered to critical facilities, such as hospitals, that may have requirements for residual levels for
prevention ofregrowth of bacteria or specific pathogens.
• Areas with historically low residual. Provide information that can be used to determine the
efficacy of actions taken to maintain chlorine residual concentration within a target range in
these areas.
Water Quality Parameters
Table 9-3 presents water quality parameters that can be monitored to manage chlorine residual
concentrations. (Note: the purpose of a given monitoring parameter is dependent on whether free chlorine
or chloramines are used within a system, as shown in the table).
Table 9-3. Water Quality Parameters for Disinfectant Residual Management
Role in Application
Parameters
Purpose of Monitoring Parameter
Necessary and sufficient
for managing chlorine
residual in all distribution
systems
Chlorine
residual
• Provides a direct measure of the chlorine residual.
• For chlorinated systems, free chlorine should be monitored
• For chloraminated systems, total chlorine or monochloramine
should be monitored
pH+
• Can affect chlorine speciation (HOCI is the stronger disinfectant
and is the dominant chlorine species below pH 7.5).
• Can affect the rate of chloramine formation due to changes in
reaction rates (lower pH increases the concentration of HOCI and
decreases the concentration ofNhb; likewise, higher pH
decreases the concentration of HOCI and increases the
concentration of NH3).
Specific
conductance*
• If a distribution system is supplied by multiple sources, a change
in specific conductance can help to identify a source change as
the cause of an abrupt change in chlorine residual concentration.
Temperature*
• An increase in temperature can create conditions suitable for
chlorine residual decay and microbial regrowth.
Necessary for managing
chlorine residual in
chloraminated systems
Ammonia, free
• Specific chlorine to ammonia ratios are required for optimal
monochloramine formation.
• Measuring free ammonia can provide information on
monochloramine formation and available ammonia for
downstream chlorine residual boosting.
• In a distribution system, excessive free ammonia can indicate the
start of chloramine degradation.
52
-------
Online Water Quality Monitoring for Distribution Systems
Role in Application
Parameters
Purpose of Monitoring Parameter
Can be used to verify
changes in chlorine
residual levels
ORP
• Responds linearly to changes in chlorine residual concentrations.
• Can be important when switching from free chlorine to chloramine,
as ORP values are higher for free chlorine; therefore, a decrease
in ORP is expected when switching from free chlorine to
chloramine and an increase is expected when switching from
chloramine to free chlorine.
t Core parameters
Figure 9-4 shows a time-series plot of chlorine residual, specific conductance, and temperature in a
reservoir experiencing chlorine residual decay. Note that the monitoring station at this reservoir is
equipped with two chlorine residual instruments that produce primary and secondary chlorine residual
data (shown as "Redundant Chlorine Residual" in the figure). At the onset of this incident, chlorine
residual concentrations from both instruments decreased. However, specific conductance levels remained
relatively constant over this time, which led investigators to rule out the possibility that a change in the
source supplying water to this monitoring location caused the chlorine decay. Further investigation of this
incident determined that the decrease in chlorine residual was due to an increase in temperature and high
water age in the reservoir.
Fivnary CNonre FevkuJ
Fedcrdirt Ctonre Rcwdual
Twpprafijm
Finfayl2MAU SdUcb'J l200«rf Suvtyl?00iiM Kandm'SMAV
Dstr.Tme
Figure 9-4. Data During an Occurrence of Chlorine Residual Decay in a Reservoir
9.4 Verify Effectiveness of Nitrification Control
Nitrification is a microbial process that is caused by the presence of free ammonia and ammonia-
oxidizing bacteria that require ammonia for energy. During this process, ammonia is sequentially
oxidized to nitrite and then nitrate through biological and chemical processes. Nitrification in a drinking
water distribution system creates water quality problems such as increased nitrite and nitrate levels;
reduced alkalinity, pH, dissolved oxygen, and chloramine residual levels; and increased potential for
bacterial growth. Storage tanks and reservoirs can be especially prone to nitrification due to the potential
for long residence times and high temperatures, both of which exacerbate chloramine decay. For more
information on nitrification, refer to the EPA distribution system issue paper "Nitrification" (EPA.
2002)" and theAWWA manual M56 Fundamentals and Control of Nitrification in Chloraminated
Drinking Water Distribution Systems (AWWA, 2013).
53
-------
Online Water Quality Monitoring for Distribution Systems
Nitrification can occur even in the most carefully operated and maintained systems. OWQM-DS data can
be used to detect conditions that indicate the onset of nitrification (enabling utilities to take response
actions to prevent or limit nitrite and nitrate formation), reduce the cost of mitigation (e.g., the cost of
draining and cleaning a storage facility), and maintain regulatory compliance.
Monitoring Locations
Monitoring at the following types of monitoring locations can provide information that can be used to
guide nitrification control:
• Entry points to a distribution system: Provide a baseline for the ChNEb-N ratio as well as the
chloramine and free ammonia concentrations in water entering the system.
• Storage tanks and reservoirs: Provide information that can be used to detect and manage
nitrification. If it is not feasible to monitor an outflow line or if it is desirable to place a station in
an alternate location to achieve multiple design goals, a station can be located in a downstream
area of the distribution system that receives water from the storage facility.
Water Quality Parameters
Table 9-4 presents water quality parameters that can be monitored to inform actions to control
nitrification. This table is based on recommendations given in the AWWA manual M56Nitrification
Prevention and Control in Drinking Water (AWWA, 2013); however, some recommendations have been
modified to apply to online water quality monitoring.
Table 9-4. Water Quality Parameters
for Verifying the Effectiveness of Nitrification Control
Role in Application
Parameters
Purpose of Monitoring Parameter
Necessary and
sufficient for detecting
nitrification incidents
Chlorine
residual
• Total chlorine or monochloramine should be monitored.
• A reduction in chlorine residual is expected during nitrification due
to the degradation of chloramines.
• Low chlorine residual concentrations can increase the potential for
bacterial growth, including nitrifying bacteria that can exacerbate
nitrification.
pH+
• A reduction in pH can indicate suitable conditions for nitrification.
Specific
conductance*
• If a distribution system is supplied by multiple sources, specific
conductance can be used to determine whether a monitoring
location is receiving water from a source that is more likely to
promote nitrification.
Temperature*
• An increase in temperature can create suitable conditions for
bacterial growth and nitrification.
Ammonia, free
• An initial increase followed by a decrease in free ammonia can
signal the initial onset of nitrification. As chloramines decay ammonia
is released, nitrifying bacteria will oxidize excess ammonia to nitrite
and then nitrate.
Achieve more timely
confirmation of
nitrification incidents
Nitrate
• An increase in nitrate concentration indicates a nitrification incident.
Nitrite
• A measurable concentration of nitrite that coincides with an increase
in nitrate and a decrease in ammonia can signal the oxidation of
ammonia to form nitrite and nitrate.
t Core parameters
Figure 9-5 shows a time-series plot of monochloramine, nitrate (NO3-N), and tank level data during a
nitrification incident in a storage tank. Nitrification had already been occurring in the tank prior to the
time-period shown in the plot, but it had not been detected due to stratification in the tank. However,
when the tank level began to decrease due to emergency water use for firefighting, a decrease in
54
-------
Online Water Quality Monitoring for Distribution Systems
monochloramine and an increase in nitrate were detected soon after, indicating that nitrification was
occurring. When fresh water began to refill the tank, the water quality returned to normal levels.
440
'cS
E
z
c
E
S
0.5
Morvochlpraminc Nitrate N03 level
Figure 9-5. Data During a Nitrification Incident
9.5 Verify Effectiveness of Corrosion Control
Corrosion in drinking water distribution systems can occur through an electrochemical process between
metal surfaces (e.g., pipes, fittings, faucets, valves) and water. Metals can be released from these surfaces
into drinking water if a protective barrier along the surfaces has not been established or if particulate
metals have been released, as was discussed in Section 9.2. Additionally, metals can be released when
changes to water chemistry occur, which can destabilize the protective barrier and corrosion by-products.
Once the protective barrier has been compromised, the exposed metal can be corroded and a new
protective barrier must be established for corrosion control.
Adjustment of parameters such as pH and alkalinity, as well as addition of ortho-phosphate, has been
widely used in successful Corrosion Control Treatment (CCT) programs; monitoring these parameters can
provide insight into the efficacy of CCT. Chemical saturation indices, such as the Langelier Saturation
Index (LSI), can be calculated based on pH, alkalinity, conductivity, and temperature values to provide an
indication of the calcium saturation and potential for calcium carbonate pipe scale formation. The LSI can
be calculated using Equation 9-1, which was derived from Standard Methods for the Examination of
Water and Wastewater (APHA, et al., 2012) and Chemical Equilibriain Water Treatment (Langelier,
1946). Additional tools, such as Pourbaix diagrams, can be used with OWQM-DS data to determine
changes to mineral stability that can lead to metal release. For more information on corrosion control, see
AWWA manual M58 Internal Corrosion Control in Water Distribution Systems (Hill and Cantor, 2011).
55
-------
Online Water Quality Monitoring for Distribution Systems
r
LSI = pHa - pHs
Where:
pHd= the actual pH of water
pHs - pK2- pKs+p[Ca"+] + p[HC03]+5p/m
Where:
p preceding a variable designates -logio of that variable
K2 = second dissociation constant for carbonic acid, at the water temperature
Ks = solubility product constant for CaC03 at the water temperature
[Ca2+] = calcium ion concentration, g-moles/L
[HCO3 ] = bicarbonate ion concentration, g-moles/L
fm - activity coefficient for monovalent species at the specified temperature
Understanding the potential for corrosion within a distribution system and the efficacy of protective pipe
scale formation can support a utility's ability to optimize CCT and prevent the release of lead and copper
into a system. In addition to maintaining compliance with the Lead and Copper Rule, this can prevent an
erosion of customer confidence and reduce the cost of replacing premise plumbing.
Monitoring Locations
Monitoring at the following types of monitoring locations can provide information that can be used to
verify the effectiveness of corrosion control:
• Entry points to a distribution system. Provide a baseline for distribution system water quality.
• Areas that exhibit variable water quality parameter values. Provide information from areas
that may be susceptible to corrosion. These areas may exist in mixing zones (for systems that are
supplied by multiple sources) and areas with high water age.
Water Quality Parameters
OWQM-DS systems can include water quality parameters known to promote corrosion, inhibit corrosion,
and indicate optimal CCT within a distribution system. Table 9-5 presents parameters that can be
monitored to verify the efficacy of corrosion control treatment. This table is based on recommendations
given in the AWWA manual M58 Internal Corrosion Control in Water Distribution Systems (AWWA,
2013); however, some recommendations have been modified to apply to OWQM-DS.
Equation 9-1. Langelier Saturation Index
56
-------
Online Water Quality Monitoring for Distribution Systems
Table 9-5. Water Quality Parameters for Verifying Effectiveness of Corrosion Control
Role in Application
Parameters
Purpose of Monitoring Parameter
Necessary and sufficient
to verify effectiveness
of CCT
Chlorine residual
• Corrosion can release particulate and dissolved metal species
that can react with and consume chlorine residual.
pH+
• A significant change can indicate suitable conditions for
corrosion and metal release.
Specific
conductance*
• Specific conductance can be converted to an approximate
concentration of total dissolved solids and used in Equation 1.
• If a distribution system is supplied by multiple sources, specific
conductance data can be used to determine whether a
monitoring location is in a mixing zone and thus subject to
significant shifts in water quality that could promote corrosion.
Temperaturet
• An increase in temperature can increase metal solubility
and release.
Alkalinity
• Directly impacts the stability of distribution system water pH.
• Can be used to calculate the LSI.
• Can be used with pH adjustments for corrosion control.
Necessary if a
phosphate-based
inhibitor is used for CCT
Ortho-phosphate
• Decrease in ortho-phosphate can indicate a reduction in the
efficacy of CCT.
Achieve more timely and
reliable detection of
problems with CCT
Apparent color
• Red water due to ferric iron release and yellow or black due to
ferrous iron release.
DO
• Is an electron acceptor at the cathodic side of an electro-
chemical interaction between the pipe wall and the water.
• Is an oxidant that can affect metal solubility and release.
ORP
• Can be used to analyze metal solubility and the potential for
release.
• Is particularly important for utilities switching from free chlorine
to chloramines for disinfection, as use of chloramines results in
lower ORP values compared to free chlorine, which can cause
a change in metal speciation.
Spectral
absorbance
• Can be used to monitor iron oxide concentrations by
measuring the primary spectral absorbance wavelength.
Turbidity
• Cloudiness or colored water (red to reddish-brown, yellow, or
black) can indicate that distribution system corrosion is
occurring.
t Core parameters
Figure 9-6 shows a theoretical example of a time-series plot of pH, alkalinity, and chlorine at a ground
water and surface water mixing zone. A treatment process error allowed the ground water to enter the
distribution system unchlorinated and unbuffered. This error caused the pH to decrease and the chlorine
residual to drop. The decrease in chlorine residual also decreased the ORP, thereby causing
destabilization of pipe scales and metal release. pH and alkalinity adjustments were made to correct this
error and return the water quality to normal levels.
57
-------
Online Water Quality Monitoring for Distribution Systems
9.0
3.0
mkLJTfwrwkiwyv
2.0
3.B
5
8,7
1-0
20
o.s
0
fhkirine AlkJllnily pH
Figure 9-6. Data Affecting Corrosion Control Treatment
While OWQM-DS systems can provide useful information about the effectiveness of corrosion control,
there are limitations to this application. First, water samples are taken from the bulk water in a main rather
than at the water-pipe interface where corrosion occurs. Water quality at this interface can be significantly
different from water quality in bulk solution. Furthermore, water quality can change as it travels through
and resides in a premise plumbing system. Even with these limitations, online monitoring of bulk water
quality in a distribution main can provide useful information about the stability of water quality
parameters that are important for corrosion control. To augment this application of OWQM-DS, grab
samples for other parameters, such as aluminum and sulfate, can be collected at the tap.
9.6 Source Tracking
Many utilities use multiple source waters to meet water demands in their distribution systems.
Throughout a given day, water from different sources can serve the same area of a system. If these
sources have water qualities that are significantly different from each other, source changes can
significantly and abruptly change distribution system water quality in that area.
Significant changes in water quality that is delivered to a given area can potentially impact the aesthetics
of water (e.g., taste and odor) and compliance with federal or state regulations. OWQM-DS data can be
used to identify the water source that is supplying an area at a given time, which can guide system
operations to manage mixing and enable utilities to meet water quality requirements, respond to customer
complaints, and inform key industrial customers of water quality changes that could impact internal
processes.
58
-------
Online Water Quality Monitoring for Distribution Systems
Monitoring Locations
Monitoring at the following types of monitoring locations can provide information that can be used to
track sources:
• Entry points to a distribution system. Provide a baseline for water quality in the distribution
system.
• Mixing zones. Provide information on frequency and duration of mixing
Water Quality Parameters
Table 9-6 presents water quality parameters that can be monitored to track water sources that are feeding
a given area. (Note: this table assumes the sources providing water to a distribution system have
consistently different values for the parameters listed.)
Table 9-6. Water Quality Parameters for Source Tracking
Role in Application
Parameters
Purpose of Monitoring Parameter
Necessary and sufficient for
source tracking
Chlorine residual
• Monitor the impact of source mixing on maintenance of
chlorine residual.
pH+
• A change can be a direct indicator of a source change.
Specific
conductance*
• A change can be a direct indicator of a source change.
Temperaturet
• A change can be a direct indicator of a source change.
Achieve more reliable source
tracking
DOC/TOC
• A change can be a direct indicator of a source change.
Spectral
absorbance
• A change in the delta can be a direct indicator of a
source change.
t Core parameters
Figure 9-7 shows a time-series plot of conductivity data during a daily source water change. Specific
conductance decreased during the source water change and then remained stable until the next cycling.
TOO
890
660
560
500
12:00 AM Wadrw*day 12:00 AM Thursday 12:QO AA1 Friday 12:00 AM Saturday 12:00 AM
DaflaTime
Figure 9-7. Data During Source Water Changes
K
! 1
n
| 1
i I
i
1
i
1
J
?
1
i
i
59
-------
Online Water Quality Monitoring for Distribution Systems
9.7 Summary of Online Monitoring System Applications
A summary of the suggested monitoring locations and water quality parameters that can facilitate the
examples of OWQM-DS applications is provided in Table 9-7.
Table 9-7. Summary of Monitoring Locations and Water Quality Parameters to Support Example
Applications
Applications
Locations
Parameters
Monitoring for
contamination incidents
• Distribution system entry points
• Locations identified using optimization tools
• Locations identified by distribution system models or
operator experience
• Chlorine residual*
• pH*
• Specific conductance*
• Spectral absorbance*
• Temperature*
• DOC/TOC
• ORP
• UV-254
Monitoring for red water/
particulate matter
incidents
• Distribution system entry points
• Areas where older, unlined iron pipes are in use
• Areas with historically high volumes of customer
complaints
• Chlorine residual*
• pH*
• Specific conductance*
• Temperature*
• Turbidity*
• Apparent color
• DOC/TOC
Chlorine residual
management
• Distribution system entry points
• Storage tanks and reservoirs
• Booster station outflow
• Entry of critical facilities
• Areas with historically low residual
• Chlorine residual*
• pH*
• Specific conductance*
• Temperature*
• Ammonia, free*
• ORP
Verify effectiveness of
nitrification control
• Distribution system entry points
• Storage tanks and reservoirs
• Ammonia, free*
• Chlorine residual, total*
• pH*
• Specific conductance*
• Temperature*
• Nitrate
• Nitrite
Verify effectiveness of
corrosion control
• Distribution system entry points
• Areas that exhibit variable water quality
parameter values
• Alkalinity*
• Chlorine residual*
• pH*
• Specific conductance*
• Temperature*
• Ortho-phosphate*
• Apparent color
• DO
• ORP
• Spectral absorbance
• Turbidity
60
-------
Online Water Quality Monitoring for Distribution Systems
Applications
Locations
Parameters
Source Tracking
• Distribution system entry points
• Mixing zones
• Chlorine residual*
• pH*
• Specific conductance*
• Temperature*
• DOC/TOC
• Spectral absorbance
* Parameters are necessary and sufficient for achieving the design example. (Note: free ammonia and ortho-phosphate are only
necessary and sufficient for chloraminated systems and systems using a phosphate-based inhibitor for CCT, respectively.)
T able 9-7 summarizes locations and parameters for specific applications and shows that the incremental
addition of one or more monitoring locations or parameters can result in a system that achieves multiple
design goals. For example, if monitoring stations were initially installed for disinfection residual
management, the addition of locations identified through optimization software and addition of spectral
absorbance instruments at strategically selected locations could provide capabilities to monitor for
contamination incidents. Likewise, chlorinated systems considering switching to chloramination could
add free ammonia instruments to select locations to achieve both disinfection residual management and
verification of nitrification control.
61
-------
Online Water Quality Monitoring for Distribution Systems
Section 10: Case Studies
This section provides case studies of utilities that have implemented OWQM-DS systems. Each case
study provides high-level utility information, describes how OWQM-DS data is used, and provides details
on the OWQM-DS system design (e.g., monitoring locations, water quality parameters, monitoring
station structure, information management and analysis approach, alert investigation procedure).
10.1 Philadelphia Water Department
Philadelphia Water Department (PWD) is a combined urban utility that serves treated drinking water to
1.6 million customers in Philadelphia, Pennsylvania. Raw water is pumped from the Delaware and
Schuylkill Rivers and treated at three water treatment plants that produce an average of 250 MGD of
water. Chloramines are used as a secondary disinfectant in the water.
PWD has used OWQM-DS data to:
• Monitor for contamination incidents. The primary concerns are intentional contamination of
the distribution system and detection of water quality changes that occur during operational
events.
• Optimize distribution system water quality. Data helps to establish baseline water quality
conditions to ensure that water quality parameters are in acceptable ranges and inform
distribution system operations.
Monitoring Locations
The OWQM-DS system consists of 38 fixed stations that are located at distribution system entry points,
storage tanks, reservoirs, fire stations, a hotel, a hospital, and inside an enclosure within a public right-of-
way. Some of these locations were selected based on results from sensor placement optimization software
analyses, while others were determined based on their ability to impact system operations. PWD also has
eight mobile stations, or "rapid deployment stations," that facilitate timely installations to achieve short-
term monitoring goals at a wide range of potential locations.
Water Quality Parameters
The monitoring stations monitor a range of water quality parameters that includes total chlorine, pH,
specific conductance, temperature, ORP, turbidity, and UV-254. (Note: PWD also monitors fluoride in
real time.)
Monitoring Station Structure
Every fixed monitoring station has redundant total chlorine instruments to enable the comparison of data
that is generated by identical instruments. Most stations use a cellular network to transmit OWQM-DS
data from the stations to a central location for analysis. However, some stations at PWD facilities use
Ethernet cable to transmit data. Additionally, autosamplers are incorporated with some stations to
facilitate the automatic collection of water samples. PWD deployed stations in the form of wall-mounted
racks, enclosed stations, and compact stations. Figure 10-1 shows one of the compact stations (i.e., rapid
deployment stations).
62
-------
Online Water Quality Monitoring for Distribution Systems
Figure 10-1. Philadelphia Water Department Rapid Deployment Station
information Management and Analysis
PWD incorporates OWQM-DS data into its SCADA system. The data is also directed to a centralized
ADS that automatically generates alerts based on changes in data for individual parameters and
relationships between multiple parameters. OWQM-DS data, alert information, and information related to
other SRS components is integrated into a dashboard. Figure 10-2 shows a screenshot of the dashboard.
Personnel can also use supplementary information management software to review historical data.
Figure 10-2. Philadelphia Water Department SRS Dashboard
63
-------
Online Water Quality Monitoring for Distribution Systems
Alert Investigation Procedure
PWDhas a Consequence Management Plan that guides the investigation of, and response to, a possible
water quality incident. Personnel are available at all times to investigate alerts. Personnel have found that
using OWQM-DS data along with customer complaint information can be very effective for timely
detection and response to water quality incidents in the distribution system.
10.2 City of Dayton Water Department
The City of Dayton Water Department serves treated drinking water to over 400,000 customers in and
around Dayton, Ohio. Dayton collects water from approximately 110 production wells in the Miami and
Mad River Well Fields. Raw water is pumped from the Great Miami River Buried Valley Aquifer to two
water treatment plants that produce an average of 65 MGD of water. Free chlorine is used as a secondary
disinfectant in the water.
Dayton has used OWQM-DS data to:
• Monitor for contamination incidents. The primary concern is intentional contamination of the
distribution system.
• Optimize distribution system water quality. Data helps to ensure that monitored parameter
values remain within normal operating ranges, inform reservoir operation, and identify
unexpected system conditions (e.g., a valve being closed when it was thought to be open).
Monitoring Locations
The OWQM-DS system consists of 12 monitoring stations
located at distribution system entry points, storage tanks,
booster stations, and pump stations. Dayton selected
monitoring locations to monitor as much of the distribution
system as feasible, including major storage systems, pump
stations, and the furthest reaches of the distribution system.
Water Quality Parameters
The monitoring stations monitor a range of water quality
parameters that includes free chlorine, pFI, specific
conductance, temperature, and turbidity. Some stations
monitor all of these parameters, while other stations monitor
free chlorine only.
Monitoring Station Structure
A mix of radio and fiber optic cable is used to transmit
OWQM-DS data from monitoring stations to a central
location for analysis. Dayton deployed stations in the form o
wall-mounted racks. Figure 10-3 shows a typical installation. Figure 10-3. City of Dayton Water
Department Monitoring Station
Information Management and Analysis Installation
Dayton incorporates OWQM-DS data into its SCADA system. Threshold values have been configured in
the SCADA system to automatically generate alerts based on changes in data for individual parameters.
Operators can access SCADA screens to view current parameter values and time-series plots of data from
the past year. Other utility personnel can view this data in read-only mode via the utility's Intranet site.
Microsoft Excel files that contain water quality reports are provided to personnel daily.
64
-------
Online Water Quality Monitoring for Distribution Systems
Alert Investigation Procedure
If a water quality anomaly is detected, personnel review data from the alerting station(s) and other
stations to determine whether an on-site investigation and water quality testing are required. Response
actions are implemented following an on-site investigation, as needed.
10.3 Mohawk Valley Water Authority
The Mohawk Valley Water Authority (MVWA) serves treated drinking water to approximately 126,000
customers in and around Utica, New York. MVWA delivers raw water from the Hinckley Reservoir and
provides treatment at a single water treatment plant that produces an average of 19 MGD of water. Free
chlorine is used as a secondary disinfectant in the water.
MVWA has used OWQM-DS data to:
• Monitor for contamination incidents. The primary concerns are intentional and unintentional
contamination of the distribution system.
• Optimize distribution system water quality. Data is monitored to evaluate the effectiveness of
treatment (including CCT), inform the dosing of chlorine at booster stations, and ensure that all
measured values remain within normal operating ranges.
Monitoring Locations
The OWQM-DS system consists of 15 stations that are located at the distribution system entry point,
storage tanks, and booster stations. Locations were selected based on a ranking of critical facilities,
availability of historical water quality data collected from a location, whether a location typically
experiences high water age, and the extent to which a location satisfied station installation requirements
(e.g., sufficient space, accessibility, available communication solution).
Water Quality Parameters
The monitoring stations monitor a range of water quality parameters that includes free chlorine, pH,
specific conductance, temperature, turbidity, and UV-254. Some stations monitor all of these parameters,
while others monitor free chlorine only.
Monitoring Station Structure
All monitoring stations use a fixed radio network to transmit OWQM-DS data from the stations to a
central location for analysis. Autosamplers are incorporated with some stations to facilitate the automatic
collection of water samples. MVWA deployed stations in the form of wall-mounted racks.
Information Management and Analysis
MVWA incorporates OWQM-DS data into its SCADA system. Figure 10-4 shows a screenshot of one of
the utility's SCADA system screens. Threshold values have been configured in the SCADA system to
automatically generate alerts based on changes in data for individual parameters. Operators can access
SCADA system screens to view current parameter values and time-series plots of data. Other personnel
with appropriate clearances can also access this data via the internet on a password-protected website.
65
-------
Online Water Quality Monitoring for Distribution Systems
Log in j
Tl«t Vonqf lltnfc
WTP rlortn,7 COMM
COMM SCRJT
A 11:54:37 AM C)
ENABLED
ARMED
Water Quality Tram 1
INTERMEDIATE SYSTEP
Water Quality Train 2
LOW SYSTEM
Hach 1720
Ha
Hach 1720
Aqua Sensor Temp
Aqua Sensor Te
Hach
rbidity
n
Turbidity
Aqua Sensor eond
Aqua Sensor Cond
UV 254
O^Tf]
UV 254
0.041
Aqua Sensor PH
Sensor PH
UVA
95.225 |
UVT
UVA
90.887
UVT
8.95
r! OW YFSTFRDAY
FLOW
TOTALIZED FLOW
16" | 5.4 |j Q,0 MOD
0.0 MGD 0,000 MGD
FLOW TOTALIZED FLOW FLOW YESTERDAY
1,6 MGD
166.0 MOD
0,134 MGD
COMM
INTRUSION
LOW TEMP
SMOKE ALARM
POWER FAIL
16 HI FLOW
20 HI FLOW
24 HI FLOW
16 HI PSI
20 HI PSI
24 HI PSI
WQ TR1 ALARM
WQ TR2 ALARM
YORK WRITE
FLOW
woroi
PSI
wans
SETPOINTS
riowfsi
mmi
WQTR2
20" 151,1
FLOW TOTALIZED FLOW FLOW YESTERDAY
7.0 MGD [ 1 2773.2 [MGD] 6.44 |MGD
COMM
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
NORMAL
Figure 10-4. Data on a Mohawk Valley Water Authority SCADA Screen
Alert Investigation Procedure
If a water quality anomaly is detected, personnel review data from the alerting station(s) and other
stations to determine whether an on-site investigation is required. Response actions are implemented as
needed following an on-site investigation.
66
-------
Online Water Quality Monitoring for Distribution Systems
Section 11: Lessons Learned
An overview of lessons learned, as shown below, can be critical to designing, implementing, and
maintaining a successful OWQM-DS system. Summary of Implementation Approaches and Lessons
Learned from the Water Security Initiative Contamination Warning System Pilots (EPA, 2015) covers
additional lessons learned that can help utilities implement a more efficient, cost-effective system.
• Establish a strong business case and staffing plan with sustainability in mind. It is important
to develop both a strong business case for how OWQM-DS data will be used to support day-to-
day operations and utility goals, as well as a staffing plan that specifies how operation and
maintenance activities will be incorporated into existing or projected personnel job functions.
These materials can help build support for an OWQM-DS system among senior management and
users at all levels, which is critical to ensure that the system will be used and maintained
effectively.
• Engage all stakeholders from the beginning of OWQM-DS system design. It is important to
engage all personnel responsible for design, implementation, and maintenance of an OWQM-DS
system. Each of these stakeholders can provide a unique perspective on how a system should be
designed with respect to their area of expertise.
• Use a phased approach for OWQM-DS system implementation. It can be effective to initially
install a limited number of monitoring stations, possibly with a variety of viable technologies, to
gain practical experience and generate real-world, OWQM-DS data before final selection of
instrumentation for an entire OWQM-DS system. This approach allows utilities to assess water
quality instruments against performance objectives (e.g., data quality), determine whether the
data collected can be used to achieve selected design goals, understand requirements for
information management and analysis software, and determine how data will be managed and
used. If a selected instrument is new to a utility, this approach can also provide insight to the
training, level of effort, and funding required to operate and maintain the system. Experience and
information gained during such a demonstration period can then be used to inform future phases
of implementation.
• Consider specialized capabilities required of personnel. A review of existing staff, roles, and
responsibilities may present a need for additional training or hiring of new staff. Specific
capabilities that may be required for successful operation and maintenance of an OWQM-DS
system include these:
o Developing protocols (e.g., standard operating procedures, quality assurance project plans) to
use OWQM-DS data to inform treatment and distribution system operations
o Calibrating, maintaining, troubleshooting, and repairing instrumentation as well as analog and
digital electronic equipment
o Maintaining highly integrated systems related to water quality instrumentation,
communications, and data acquisition
o Interpreting OWQM-DS data and data analysis results, which may require an understanding
of mathematical and statistical techniques used by ADS software, to inform treatment and
distribution system operations
o Performing complex chemical analyses (e.g., of volatile and semi-volatile organics) and using
quality assurance techniques (e.g., automated drift correction) to enable the use of
sophisticated instrumentation and sampling equipment
67
-------
Online Water Quality Monitoring for Distribution Systems
Discuss OWQM-DS system design with utilities that
have implemented systems of their own. Utilities that
have implemented OWQM-DS systems can provide
valuable feedback on the performance of water quality
instruments and information management and analysis
software, as well as how to incorporate OWQM-DS-
related activities into day-to-day operations at a utility.
Additional information
For information on utility
experience with OWQM-DS
instruments and information
management and analysis
software, and/or to be put into
contact with utilities that have
implemented OWQM-DS
systems, contact
WQ SRS@epa.gov.
-------
Online Water Quality Monitoring for Distribution Systems
Resources
Introduction
Water Quality Surveillance and Response System Primer
This document provides an overview of SRSs, and serves as a foundation for the use of technical
guidance and products used to implement an SRS. EPA 817-B-15-002, May 2015.
https://www, ero. gov/sitcs/product ion/file s/2015-
06/dociimcnts/watcr quality surcvcillancc and response system primer.pdf
Online Water Quality Monitoring Primer for Water Quality Surveillance and Response Systems
This document provides an overview of OWQM, a component of an SRS. It also presents basic
information on the goals and objectives of OWQM within the context of an SRS. EPA 817-B-15-
002A, May 2015.
https://www. epa. gov/sitcs/product ion/file s/2015-
06/documcnts/onlinc water quality monitoring primer.pdf
Framework for Designing Online Monitoring Systems
Guidance for Developing Integrated Water Quality Surveillance and Response Systems
This document provides guidance for applying system engineering principles to the design and
implementation of an SRS to ensure that the system functions as an integrated whole and is
designed to effectively perform its intended function. Section 2 provides guidance on establishing
a project team and coordinating SRS implementation activities. Section 3 provides guidance on
developing a master plan for an SRS. EPA 817-B-15-006, October 2015.
https://www. epa. gov/sitcs/product ion/file s/2015-
12/documents/guidance for developing integrated wq srss 1104l5.pdf
American Water Works Association Partnership for Safe Water
A four-phase distribution system optimization program for drinking water distribution systems
that add a residual disinfectant and are interested improving performance.
Monitoring Locations
EPANET
Software that models the hydraulic and water quality behavior of water distribution piping
systems.
https ://www. epa. gov/water-research/epanet
Threat Ensemble Vulnerability Assessment and Sensor Placement Optimization Tool (TEVA-
SPOT)
Software that provides command-line interfaces to computational tools that compute impacts for
contamination incidents and optimizes monitoring locations in a water distribution system,
https ://so ftw arc .sand ia. gov /trac/sp o t/
Checklist for Assessing Potential Monitoring Locations
A checklist that can be used to assess potential monitoring locations with respect to monitoring
station installation requirements.
Click this link to open the template
69
-------
Online Water Quality Monitoring for Distribution Systems
Monitoring Parameters
Guidance for Selecting Online Water Quality Monitoring Instruments for Source Water and
Distribution System Monitoring
This document provides detailed information about commonly monitored water quality
parameters and guidance on selecting appropriate parameters to monitor for a given application. It
also provides a summary of available technologies for monitoring each parameter. In press.
https://www.epa.gov/waterqualitvsurveillance/online-water-qualitv-monitoring-resources
Monitoring Stations
Guidance for Designing Communications Systems for Water Quality Surveillance and Response
Systems
This document provides guidance and information to help utilities select an appropriate
communications system to support operation of an SRS. It provides rigorous criteria for
evaluation communications system options, evaluates common technologies with respect to these
criteria, describes the process for establishing requirements for a communications system, and
provides guidance on selecting and implementing a system. EPA 817-B-16-002, July 2016.
https://www, epa. gov/sitcs/product ion/file s/2017-
04/documcnts/srs communications guidance 081016.pdf
Guidance for Building Online Water Quality Monitoring Stations
This document provides guidance for designing OWQM stations for both source water
monitoring and OWQM-DS. It describes different station designs and provides detailed design
schematics, describes basic station equipment and station accessories, and provides
considerations for fabricating and installing OWQM stations. In press.
https://www.epa.gov/wateraualitvsurveillance/online-water-qualitv-monitoring-resources
information Management and Analysis
Guidance for Developing Integrated Water Quality Surveillance and Response Systems
This document provides guidance for applying system engineering principles to the design and
implementation of an SRS to ensure that the system functions as an integrated whole and is
designed to effectively perform its intended function. Section 4 provides guidance on developing
information management system requirements, selecting an information management system, and
IT master planning. Appendix B provides an example outline for an IT operations and
maintenance plan. EPA 817-B-15-006, October 2015.
https://www, epa. gov/sitcs/product ion/file s/2015-
12/dociimcnts/guidance for developing integrated wq srss 110415.pdf
Exploratory Analysis of Time-series Data to Prepare for Real-time Online Water Quality
Monitoring
This document describes methods for analyzing time-series water quality data to establish normal
variability for water quality at unique monitoring locations. It also describes how the results of
this exploratory analysis can be used to develop tools and training to prepare utility personnel for
real-time analysis of OWQM-DS data. EPA 817-B-16-004, November 2016.
https://\vww.cpa.gov/sitcs/prodiiction/filcs/2016-1 1/documcnts/cxploratorv analysis of time-
series data for owqm.pdf
70
-------
Online Water Quality Monitoring for Distribution Systems
Dashboard Design Guidance for Water Quality Surveillance and Response Systems
This document provides information about useful features and functions that can be incorporated
into an SRS dashboard. It also provides guidance on a systematic approach that can be used by
utility managers and IT personnel to define requirements for a dashboard. EPA 817-B-15-007,
November 2015.
https://www, epa. gov/sitcs/product ion/file s/2015-
12/documents/srs dashboard guidance I I20l5.pdf
Information Management Requirements Development Tool
This tool is intended to help users develop requirements for an SRS information management
system, thereby preparing them to select and implement an information management solution.
Specifically, this tool (1) assists SRS component teams with development of component
functional requirements, (2) assists IT personnel with development of technical requirements, and
(3) allows the IT design team to efficiently consolidate and review all requirements. EPA 817-B-
15-004, October 2015.
https://www.epa.gov/waterqualitv surveillance/information-management-requirements-
development-tool
Alert Investigation Procedure
OWQM Alert Investigation Procedure Template (Word
The alert investigation procedure template includes
checklists that can be used to document the utility's
April 2018.
Click this link to open the template
Guidance for Developing Integrated Water Quality Surveillance and Response Systems
This document provides guidance for applying system engineering principles to the design and
implementation of an SRS to ensure that the system functions as an integrated whole and is
designed to effectively perform its intended function. Section 6 provides guidance on developing
a training and exercise program to support SRS operations. EPA 817-B-15-006, October 2015.
https://www. epa. gov/sitcs/product ion/file s/2015-
12/dociimcnts/guidance for developing integrated wq srss 1104l5.pdf
SRS Exercise Development Toolbox
Software that helps utilities and response partner agencies to design, conduct, and evaluate
exercises around contamination scenarios. These exercises can be used to develop and refine
investigation and response procedures, and train personnel in the proper implementation of those
procedures. The toolbox guides users through the process of developing realistic scenarios,
designing discussion-based and operations-based exercises, and creating exercise documents.
March 2016.
https://www.cpa.gov/watcrqiialitvsiirvcillancc/watcr-qiialitv-siirvcillancc-and-rcsponsc-svstcm-
exercise -developm ent-too lbox
File)
an editable flow diagram, table, and
role in an OWQM alert investigation process.
71
-------
Online Water Quality Monitoring for Distribution Systems
Preliminary Design
Preliminary OWQM Design Template (Word File)
This Word template can be used to document aspects of OWQM component design such as the
component implementation team, design goals and performance objectives, preliminary
monitoring locations, preliminary water quality parameters, preliminary monitoring station
design, preliminary information management requirements, initial training requirements, budget,
and schedule. April 2018.
Click this link to open the template
Framework for Comparing Alternatives for Water Quality Surveillance and Response Systems
This document provides guidance for selecting the most appropriate SRS design for a utility from
a set of viable alternatives. It guides the user through an objective, stepwise analysis for ranking
multiple alternatives and describes, in general terms, the types of information necessary to
compare the alternatives. EPA 817-B-15-003, June 2015.
https://www, epa. gov/sites/production/file s/2015-
07/dociimcnts/framework for comparing alternatives for water quality surveillance and resp
onse svstems.pdf
Water Finance Clearinghouse
This website provides information on current federal, state, local, private, and other sources of
funding for water related projects.
https ://ofmpub. epa. gov/apex/wfe/f ?p= 165:6:990 0315311146 ::N 0:6::
72
-------
Online Water Quality Monitoring for Distribution Systems
References
Rathi, S., Gupta, R. & Ormsbee, L., 2015. A review of sensor placement objective metrics for
contamination detection in water distribution networks. Water Science & Technology: Water Supply,
15(5), 898-917.
Philadelphia Water Department and CH2M HILL, 2013. Philadelphia Water Department Contamination
Warning System Demonstration Pilot Project: Guidance for Locating Online Water Quality
Monitoring Stations Using TEVA-SPOT. Philadelphia, PA: Author.
EPA, 2015. Summary of Implementation Approaches and Lessons Learned from the WSI Contamination
Warning System Pilots, EPA 817-R-15-002. Washington, D.C. Retrieved from
http ://www. epa. gov/sites/production/files/2015 -
12/documents/wsi pilot summary report 102715.pdf
Janke, R., Haxton, T. B., Grayman, W.„ Bahadur, R., Murray, R., Samuels, W., & Taxon, T., 2009.
Sensor network design and performance in water systems dominated by multi-story buildings.
Proceedings of the World Environmental and Water Resources Congress 2009. Kansas City, MO.
Schal, S., Bryson, L. S., & Ormsbee, L., 2014. A graphical procedure for sensor placement guidance for
small utilities. Journal A WWA, 106, 10.
Umberg, K. and Allgeier, S.,2016. Parameter set points: an effective solution for real-time data analysis.
Journal AWWA, 108, E60-66.
EPA, 2013. Water Quality Event Detection System Challenge: Methodology and Findings, EPA 817-R-
13-002. Washington, D.C. Retrieved from https://www.epa.gov/waterqualitvsurveillance/water-
qualitv-event-detection-svstem-challenge-methodologv-and-findings
EPA, 2014. Water Security Initiative: Evaluation of the Water Quality Monitoring Component of the
Cincinnati Contamination Warning System Pilot, EPA-817-R-14-001B. Washington, D.C. Retrieved
from https://www.epa. gov/sites/production/files/2015 -
06/documcnts/wsi evaluation of the water quality monitoring component of the Cincinnati con
tamination warning system pilot.pdf
EPA, 2005a. Online Water Quality Monitoring as an Indicator of Drinking Water Contamination, EPA-
817-D-05-002. Washington, D.C. Retrieved from
https://nepis.epa. gov/Exe/ZvNET. exe/P 1004B3M.TXT ?ZvActionD=Zv Document&C lient=EP A&In
dex=2000+Thru+2005&Docs=&Querv=&T ime=&EndT ime=& SearchMethod=l&T ocRestrict=n&T
oc=&TocEntrv=&QField=&QFieldYear=&QFieldMonth=&QFieldDav=&IntQFieldOp=Q&ExtQFie
ldOp=0&XmlOuerv=&File=D%3A%5Czvfiles%5CIndex%20Data%5C00thru05%5CTxt%5C000Q
0()20%5CP 1004B3M. txt&User=ANONYM OUS&Pass\vord=anonvmous&Sort Method=h%7C-
&MaximumDocuments= l&FuzzvDegree=0&ImageQualitv=r75 g8/r75 g8/xl 50yl50gl6/i425&Displ
av=hpfr&DefSeekPage=x&SearchBack=ZvActionL&Back=ZvActionS&BackDesc=Results%20pag
e&MaximumP ages= 1 &Z vEntrv=1 & Se ekP age=x&ZvP URL
EPA, 2005b. WaterSentinel Contaminant Selection. SENSITIVE. For Official Use Only.
EPA, 2005c. WaterSentinel System Architecture, EPA-817-D-05-003. Washington, D.C. Retrieved from
https ://www. hsdl. org/?abstract&did=34214
73
-------
Online Water Quality Monitoring for Distribution Systems
EPA, 2009. Distribution System Water Quality Monitoring: Sensor Technology Evaluation Methodology
and Results, EPA 600/R-09/076. Washington, D.C. Retrieved from
http://www, epa. gov/sites/production/files/2015 -
06/documents/distribution system water quality monitoring sensor technology evaluation metho
dology results.pdf
Allgeier, S.C., Hall, J., Rahman, M., and Coates, W., 2010. Selection ofWater Quality Sensors for a
Drinking Water Contamination Warning System. In Proceedings for the AWWA Water Quality
Technology Conference. Savannah, GA.
Hill C. P., & Cantor, A. F., 2011. Internal corrosion control in water distribution systems. Denver, CO:
American Water Works Association.
Lauer, B., 2010. Partnership Offers New Distribution System Optimization Program. Journal AWWA,
102(12): 32.
American Water Works Association, 2006. M20 Water Chlorination/Chloramination Practices and
Principles. Denver, CO: American Water Works Association.
Baribeau, H., Po/os, N. L., Boulos, L., & Crozes, G. F. (Eds.), 2005. Impact of distribution system water
quality on disinfection efficacy. American Water Works Association Research Foundation and EP A.
EPA, 2002. Distribution System Issue Paper: Nitrification. Retrieved from
https://www.epa. gov/sitcs/product ion/file s/2015-09/documents/nitrification 1. pdf
American Water Works Association, 2013. A (56 Fundamentals and Control of Nitrification in
Chloraminated Drinking Water Distribution Systems. Denver, CO: American Water Works
Association.
American Public Health Association, American Water Works Association, & Water Environment
Foundation, 2012. Standard Methods for the Examination ofWater and Wastewater (22nd ed.).
Washington, D.C.
Langelier, W.F., 1946. Chemical Equilibria in Water Treatment. Journal AWWA. 38(2), 169-178.
74
-------
Online Water Quality Monitoring for Distribution Systems
Glossary
accuracy. The degree to which a measured value represents the true value.
alert. An indication from an SRS surveillance component that an anomaly has been detected in a
datastream monitored by that component. Alerts may be visual or audible, and may initiate automatic
notifications such as pager, text, or email messages.
alert investigation. The process of investigating the validity and potential causes of an alert generated by
an SRS surveillance component.
alert investigation checklist. A form that lists a sequence of steps to follow when investigating an SRS
alert. This form ensures consistency with an alert investigation procedure and provides documentation of
each investigation.
alert investigation process. A documented process that guides the investigation of an SRS alert. A
typical procedure defines roles and responsibilities for alert investigations, includes an investigation
process diagram, and provides one or more checklists to guide investigators through their role in the
process.
anomaly. A deviation from an established baseline in a monitored datastream. Detection of an anomaly
by an SRS surveillance component generates an alert.
anomaly detection system (ADS). A data analysis tool designed to detect deviations from an established
baseline. An ADS may take a variety of forms, ranging from thresholds to complex computer algorithms.
application. A specific use of OWQM-DS to meet a design goal. An example would be monitoring of
chlorine residual to optimize distribution system water quality.
architecture. The fundamental organization of a system embodied in its components, their relationships
to each other, and to the environment, and the principles guiding its design and evolution. The
architecture of an information management system is conceptualized as three tiers: source data systems,
analytical infrastructure, and presentation.
asset. A piece of equipment, IT system, instrument, or other physical resource used in the implementation
of an SRS component or system.
baseline. Values for a datastream that include the variability observed during typical system conditions.
cloud service. A third-party provider of data storage or a computer application that uses the Internet as a
means of transmitting data to a client.
completeness. The percentage of data that is of sufficient quality to support its intended use.
component. One of the primary functional areas of an SRS. There are four surveillance components:
Online Water Quality Monitoring (including source water and distribution system monitoring), Enhanced
Security Monitoring, Customer Complaint Surveillance, and Public Health Surveillance. There are two
response components: Consequence Management and Sampling and Analysis.
75
-------
Online Water Quality Monitoring for Distribution Systems
concentration. In solutions, the mass, volume, or number of moles of solute present in proportion to the
amount of solvent or total solution. Common measures are molarity, normality, percent and by specific
gravity scales.
consequence. An adverse public health or economic impact resulting from a contamination incident.
Consequence Management (CM). One of the response components of an SRS. This component
encompasses actions taken to plan for and respond to possible drinking water contamination incidents to
minimize the response and recovery timeframe, and ultimately minimize consequences to a utility and the
public.
contamination incident. The presence of a contaminant in a drinking water distribution system that has
the potential to cause harm to a utility or the community served by the utility. Contamination incidents
may have natural (e.g., sloughing of pathogens from accumulated biofilm), accidental (e.g., chemicals
introduced through accidental cross-connection), or intentional (e.g., purposeful addition of a contaminant
at a fire hydrant) causes.
control center. A utility facility that houses operators who monitor and control treatment and distribution
system operation, as well as other personnel with monitoring or control responsibilities. Control centers
often receive system alerts related to operations, water quality, security, and some of the SRS surveillance
components.
dashboard. A visually oriented user interface that integrates data from multiple SRS components to
provide a holistic view of distribution system water quality. The integrated display of information in a
dashboard allows for more efficient and effective management of water quality and the timely
investigation of water quality anomalies.
data access. The process of retrieving data from an information management system for review and
analysis.
data analysis. The process of analyzing data to support routine system operation, rapid identification of
water quality anomalies, and generation of alert notifications.
data analysis tool. Any tool used to analyze data for the purpose of generating useful information.
data quality objectives. Qualitative and quantitative statements that clarify study objectives, define the
appropriate types of data, and specify the tolerable levels of potential decision errors that will be used as
the basis for establishing the quality and quantity of data needed to support decisions.
datastream. A time series of values for a unique parameter or set of parameters. Examples of SRS
datastreams include, chlorine residual values, water quality complaint counts, and number of emergency
department cases.
design goal. The specific benefits to be realized through deployment of an SRS and each of its
components.
distribution system. The infrastructure needed to convey water from a treatment plant, well,
interconnection, or other entry point to service connections throughout a city, town, or county.
76
-------
Online Water Quality Monitoring for Distribution Systems
distribution system model. A mathematical representation of a drinking water distribution system,
including pipes, junctions, valves, pumps, tanks, reservoirs, and other appurtenances. These models
predict flow and pressure of water through the system, and, in some cases, water quality.
functional requirement. A type of information management requirement that defines key features and
attributes of an information management system that are visible to the end user. Examples of functional
requirements include the manner in which data is accessed, the types of tables and plots that can be
produced through the user interface, the manner in which component alerts are transmitted to
investigators, and the ability to generate custom reports.
geographic information system (GIS). Hardware and software used to store, manage, and display
geographically referenced information. Typical information layers used by water utilities include utility
infrastructure, hydrants, service lines, streets, and hydraulic zones. GIS can also be used to display
information generated by an SRS.
hardware. Physical IT assets such as servers or user workstations.
historical data. Data that has been generated and stored, including recent data that is readily available
in an information management system as well as older data that has been stored or archived in a
historian.
hydraulic connectivity. The hydraulic relationship between locations in a distribution system. Two
locations are hydraulically connected if water flows from one to the other.
information management system. The combination of hardware, software, tools, and processes that
collectively support an SRS and provide users with information needed to monitor real-time system
conditions. The system allows users to efficiently identify, investigate, and respond to water quality
incidents.
information technology (IT). Hardware, software, and data networks that store, manage, and process
information.
interconnects. Interconnects are connections between different systems. They could be to wholesale
customers, from wholesale suppliers, or from neighboring systems
invalid alert. An alert from an OWQM-DS system that is not due to a true water quality anomaly or a
contamination incident.
lifecycle cost. The total cost of a system, component, or asset over its useful life. Lifecycle cost includes
the cost of implementation, operation and maintenance, and renewal.
monitoring station. A configuration of one or more water quality sensors and associated support
systems, such as plumbing, electric, and communications that is deployed to monitor water quality in
real time at a specific location in a drinking water distribution system.
Online Water Quality Monitoring (OWQM). One of the surveillance components of an SRS. OWQM
utilizes data collected from monitoring stations that are deployed at strategic locations in a source water
or a distribution system. Monitored parameters can include common water quality parameters (e.g., pH,
specific conductance, turbidity) and advanced parameters (e.g., total organic carbon, spectral
absorbance). Data from monitoring stations is transferred to a central location and analyzed.
77
-------
Online Water Quality Monitoring for Distribution Systems
operational control point. A location within a distribution system, such as a water storage facility,
booster station, or pump station, where operational adjustments are made to achieve design goals.
performance objectives. Measurable indicators of how well an SRS or its components meet selected
design goals.
possible. In the context of the threat level determination process, water contamination is considered
possible if the cause of an alert from one of the surveillance components cannot be identified or
determined to be benign.
preliminary operation. A period of SRS component operation during which all equipment and IT
systems are operational, but data analysis and investigations are not performed in real time. The purpose
of preliminary operations is to evaluate the performance of the SRS component, address problems, and
allow personnel to become familiar with SRS component procedures.
real-time. A mode of operation in which data describing the current state of a system is available in
sufficient time for analysis and subsequent use to support assessment, control, and decision functions
related to the monitored system.
reservoir. A structure designed to store very large volumes of finished water, which may be located
underground, in-ground, or at grade.
response action. An action taken by a utility, public health agency or another response partner to
minimize the consequences of an undesirable water quality incident. Response actions may include
issuing a public notification, changing system operations, or flushing the system.
Sampling and Analysis (S&A). One of the response components of an SRS. S&A is activated during
Consequence Management to help confirm or rule out possible water contamination through field and
laboratory analyses of water samples. In addition to laboratory analyses, S&A includes all the activities
associated with site characterization. S&A continues to be active throughout remediation and recovery if
contamination is confirmed.
spectral fingerprint. The spectral absorbance of a sample over a range of wavelengths (typically in the
visible and ultraviolet spectrum). Spectral fingerprints can be measured for specific compounds or
complex mixtures, and can be a means of identifying the presence of a specific compound or a change in
the characteristics of a complex mixture.
storage facility. A structure in a distribution system where water is temporarily held, such as a tank or
reservoir.
Supervisory Control and Data Acquisition (SCADA). A system that collects data from various sensors
at a drinking water treatment plant and locations in a distribution system, and sends this data to a central
information management system.
tank. A structure designed to store large volumes of finished water, which may be at grade or elevated.
technical requirement. A type of information management requirement that defines system attributes
and design features that are often not readily apparent to the end user, but are essential to meeting
functional requirements or other design constraints. Examples include attributes such as system
availability, information security and privacy, backup and recovery, data storage needs, and inter-system
integration requirements.
78
-------
Online Water Quality Monitoring for Distribution Systems
threshold. Minimum and/or maximum acceptable values for individual datastreams that are compared
against current or recent data to determine whether conditions are anomalous or atypical of normal
operations.
user interface. A visually oriented interface that allows a user to interact with an information
management system. A user interface typically facilitates data access and analysis.
valid alert. An alert due to water contamination, a verified water quality incident, an intrusion at a utility
facility, or a public health incident.
water quality incident. An incident that results in an undesirable change in water quality (e.g., low
residual disinfectant, rusty water, taste & odor, etc.). Contamination incidents are a subset of water quality
incidents.
water quality instrument. A unit that includes one or more sensors, electronics, internal plumbing,
displays, and software that is necessary to take a water quality measurement and generate data in a format
that can be communicated, stored, and displayed. Some instruments also include diagnostic tools.
water quality sensor. The part of a water quality instrument that performs the physical measurement of a
water quality parameter in a sample.
Water Quality Surveillance and Response System (SRS). A system that employs one or more
surveillance components to monitor and manage source water and distribution system water quality in
real time. An SRS utilizes a variety of data analysis techniques to detect water quality anomalies and
generate alerts. Procedures guide the investigation of alerts and the response to validated water quality
incidents that might impact operations, public health, or utility infrastructure.
79
------- |