vEPA
United
Environmental Protection
Ag*ncy
WaterSentinel Online Water Quality
Monitoring as an Indicator of Drinking
Water Contamination
Draff, Version 1.0
December 12, 2005
-------
-------
U.S. Environmental Protection Agency
Water Security Division
Ariel Rios Building, Mail Code 4601M
1200 Pennsylvania Avenue, N. W.
Washington, DC 20460
EPA817-D-05-002
-------
-------
WS Online Water Quality
Disclaimer
The Water Security Division, of the Office of Ground Water and Drinking Water, has reviewed and
approved this draft document for publication. This document does not impose legally binding
requirements on any party. The word "should" as used in this Guide is intended solely to recommend or
suggest and does not connote a requirement. Neither the United States Government nor any of its
employees, contractors, or their employees make any warranty, expressed or implied, or assumes any
legal liability or responsibility for any third party's use of or the results of such use of any information,
apparatus, product, or process discussed in this report, 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 or its application should be addressed to:
Elizabeth Hedrick
U.S. EPA Water Security Division
Threat Analysis, Prevention, and Preparedness Branch
26 West Martin Luther King Drive
Cincinnati, OH 45268-1320
(513)569-7296
Hedrick.Elizabeth(S),epa.gov
DRAFT-121205
-------
WS Online Water Quality
Acknowledgements
John Hall and Jeff Szabo, of the EPA's National Homeland Security Research Center (NHSRC), co-
authored this document and performed the experiments provided in the Case Study. The authors would
like to thank Steve Allgeier and Irwin Silverstein of the EPA's Water Security Division for their valuable
input in the writing of this paper, and the staff of Computer Sciences Corporation (CSC) for editorial
support.
DRAFT-121205
-------
WS Online Water Quality
Executive Summary
WaterSentinel (WS) serves as a demonstration project, or pilot, for designing and implementing an
effective contamination warning system (CWS) in a drinking water distribution system. As part of WS, a
CWS should be developed that encompasses monitoring technologies and detection strategies, combined
with enhanced public health surveillance to collect, integrate, analyze, and communicate information to
provide a timely warning of potential water contamination incidents and initiate response actions to
minimize public health and economic impacts. The success of a CWS, and hence WS, depends on the
ability to effectively integrate these components and analyze the resulting information in a timely manner
to inform response actions that can substantially reduce the potential consequences of a contamination
incident.
Current means of monitoring water quality parameters, such as periodic grab sampling for disinfectant
residual and infrequent sampling and analysis for a small number of specific contaminants, may be of
limited scope and usefulness in the WS-CWS. Therefore, the WS-CWS aims to detect contaminants by
utilizing a network of online water quality sensors, deployed throughout a drinking water distribution
system, that are responsive to many contaminants. In many cases, information from online water quality
sensors should provide the first indication of possible contamination, and should set into motion response
actions to either corroborate or rule out contamination.
Establishing which water quality parameters provide the broadest coverage and most reliable indication of
contamination is of critical importance to the success of this component of the WS-CWS. Research,
conducted by online water quality sensor manufacturers and the EPA provided information about the
potential of various contaminants of concern to produce detectable changes in specific water quality
parameters. While there are numerous parameters that respond to contamination, a literature review
found that the most effective parameters for detecting the 33 WS Baseline contaminants are free chlorine,
total organic carbon (TOC), conductivity and pH. Oxidation/reduction potential corroborates chlorine
sensor results. Other parameters such as chloride, nitrate and ammonia have been observed to change in
the presence of contaminants but mostly due to interference of concomitant ions. Turbidity, which can be
highly variable, is not a good primary indicator of contamination.
Utilizing online water quality monitoring as an indicator of drinking water contamination should be an
integral part of the WS system architecture. Online monitoring should enable water utilities to detect
potential contamination quickly and launch an appropriate response. Future work on online monitoring of
radionuclides, a broader range of chemical and biological agents, and the use of event detection software
should improve detection as the WS-CWS evolves.
DRAFT-121205 Hi
-------
WS Online Water Quality
Table of Contents
Executive Summary Hi
Section 1.0: Introduction 1
Section 2.0: Background 3
Section 3.0: Literature Review 7
Section 4.0: Case Study 11
4.1 Experimental 11
4.2 Results 12
Section 5.0: Summary and Conclusions 17
Section 6.0: References 19
Appendix A: Acronym List 21
List of Tables
Table 2-1. Partial Listing of Manufacturers of Online Water Quality Monitors 4
Table 2-2. Typical Water Quality Parameters 5
Table 3-1. Select Contaminants that Trigger the Hach Event Monitor Trigger System 7
Table 3-2. Multi-Parameter Probe Technologies Tested by TTEP 8
Table 3-3. Multi-Parameter Probes Tested by the ETV Program for Contaminant Detection 9
Table 3-4. Summary of Five ETV Studies Using Multi-Parameter Probes to Detect Contamination 9
Table 4-1. Typical Water Quality Parameter Values with Daily Variability 11
Table 4-2. Online Monitors Tested in Recirculating Pipe-loop Mode 12
Table 4-3. Online Monitor Responses to Select Contaminants in Recirculating Pipe-Loop Mode 13
Table 4-4. Online Monitor Responses to Select Contaminants in Single-Pass Pipe Mode at 80 Feet
from the Point of Injection 14
Table 4-5. Online Monitor Responses to Biological Growth Media in Single-Pass Pipe Mode at
1,100 feet from the Point of Injection 16
List of Figures
Figure 4-1. Change in Free Chlorine to Nicotine at 80 and 1,100 ft from Injection 15
Figure 4-2. Free Chlorine Response with Increasing Initial Nicotine Concentration 15
DRAFT-121205
-------
WS Online Water Quality
Section 1.0: Introduction
A successful contamination warning system (CWS) should include components that yield timely and
reliable warning of a potential contamination incident. Drinking water utilities routinely monitor a
number of water quality parameters (i.e., pH, conductivity, free chlorine, oxidation-reduction potential
(ORP) and total organic carbon (TOC) in drinking water treatment and distribution. Under routine
operations, however, these parameters are monitored only periodically through grab sampling programs,
and at a frequency that would not provide early warning of a potential contamination incident. To employ
conventional water quality monitors in a CWS, WaterSentinel (WS) proposes the application of multiple
online water quality probes and sensors configured into sensor stations, placed throughout the distribution
system, transmitting usable information in real-time. Questions that should be addressed for successful
implementation of online water quality monitoring as a component of the WS pilot include:
Do water quality parameters change in response to contaminants that could be introduced into the
distribution system and, if so, which parameters are the best indicators of contamination?
How can true contamination events be detected from the normal background variability of these
parameters when deployed for extended periods of time?
For a given distribution system, what are the optimal number of, and locations for, these sensor
stations that would provide the best coverage and protection of the population served (sensor
network design)?
The information provided in this document addresses the first question; do water quality parameters
change in response to contaminants that could be intentionally or accidentally introduced into the
distribution system and, if so, which parameters are the best indicators of contamination? The second and
third questions relating to online water quality monitoring are addressed in Overview of Event Detection
Systems for WaterSentinel (USEPA, 2005a) and WaterSentinel System Architecture (USEPA, 2005b),
respectively.
This document describes the state-of-the-science of real-time, online water quality monitoring using
conventional water quality parameters. The results from controlled studies have demonstrated that certain
water quality parameters respond rapidly and predictably to contaminants of interest to water security, at
concentrations well below the LD50 (a dose that results in death in 50% of the population exposed to that
dose), and that, when configured to be used online, these online monitors promise to be an important
component of the WS contamination warning system (WS-CWS).
The remaining sections of this document describe the following aspects of online water quality as an
indicator of drinking water contamination:
Section 2.0: Background. Presents an overview of manufacturers of online water quality
monitors, the benefit of multiple sensors, and how various water quality parameters are measured.
Section 3.0: Literature Review. Provides a description of research using online water quality
monitors to detect contamination events using vendor and American Water Works Association
Research Foundation (AwwaRF) studies as well as EPA studies.
Section 4.0: Case Study. Describes the testing of two different modes of distribution pipe
simulation in a pilot scale system.
Section 5.0: Conclusions. Summarizes and concludes the use of online water quality as an
indicator of drinking water contamination.
DRAFT-121205
-------
WS Online Water Quality
Section 6.0: References. This section provides a bibliography of the references cited in this
document.
Appendix A: Acronyms.
A complete glossary of terms related to online water quality and the WS program is available in
WaterSentinel System Architecture (USEPA, 2005a).
DRAFT-121205
-------
WS Online Water Quality
Section 2.0: Background
Ideally, the monitoring devices used in a water distribution CWS would detect any and all of the possible
agents that could be encountered. With the vast array of chemical, biological and radiological agents
available for possible use in an intentional attack (USEPA, 2005c), deployment of contaminant-specific
monitors that could specifically detect all of the chemical, biological, and radiological contaminants of
concern would be impractical and cost prohibitive. In-depth reviews of contaminant-specific, emerging
sensor technologies, and their roles in early warning systems, can be found elsewhere (ICF, 2005; States,
et al., 2004). Instead of using contaminant-specific monitors throughout the distribution system, the
design of the WS-CWS proposes using commercially available, rugged and cost-effective water quality
monitors that are sensitive and responsive to a wide array of contaminants. Should a contaminant be
introduced that results in a measurable change in water quality that is 'detected' as an anomaly by an
event detection system (EDS), then site specific field testing and/or sampling with laboratory-based
analyses would be performed to determine the specific contaminant that was introduced.
Online monitoring of water quality parameters, transmitting data in real-time to the utility's Supervisory
Control and Data Acquisition (SCADA) system can serve the dual purpose of early detection of
contamination and can save utility operators hours of sampling and testing time that would otherwise be
required to collect even a fraction of the same data (Schlegal, 2004). Online monitors differ from hand-
held or stand alone monitors in that the data are continuously collected (at specified and frequent
intervals), and can be transmitted in real or near-real time (results within minutes) to the SCADA system,
or be processed by event detection algorithms prior to transmission to the SCADA. A partial listing of
manufacturers of online water quality monitors is presented in Table 2-1. Hach Company (Loveland,
Colorado) and YSI (Yellow Springs, Ohio) are two companies that make online monitors with data
transmission capabilities. The Hach sensor stations can be equipped with Hach proprietary event
detection system software that uses chemometrics (mathematical modeling of chemical data) to detect and
characterize changes in water quality parameters (Kroll, 2005). Hach has compiled libraries of
contaminant response profiles to achieve a level of contaminant specificity when the event detection
system is triggered. Other companies, such as PureSense (Moffett Field, CA) and Source Sentinel LLC
(Syracuse, New York), do not make water quality monitors but use commercially available monitors and
sensors and their own algorithms to detect and transform the data from many monitors and sensors into a
usable information stream that is transmitted to the SCADA system. Event detection systems (EDSs) are
discussed in more detail in the Overview of Event Detection Systems for WaterSentinel (USEPA, 2005a).
Use of multiple water quality sensors provides corroborating information that may lead to more
appropriate response decisions when a change in water quality is observed. For example, if one water
quality parameter changes, an appropriate response may be to perform remote diagnostic testing on the
sensor and review additional data and baseline trends to determine if the change is due to a harmless
cause. However, if multiple water quality parameters change, a more appropriate response may be to
dispatch site characterization teams to perform field tests and collect water samples from that area of the
distribution system for further laboratory-based analyses. Automatic collection devices in the distribution
system permit the capture of the water sample that triggers the response. Such a tiered approach for
responding to anomalous water quality readings can be one means of managing the effort required to
respond to triggers while still giving each trigger due consideration as a possible contamination threat
(Hasan, et al., 2004; Alai, et al., 2005).
DRAFT-121205
-------
WS Online Water Quality
Table 2-1. Partial Listing of Manufacturers of Online Water Quality Monitors
Vendor
Dascore, Inc., (Jacksonville, FL)
Hach Company, (Loveland, CO)
ManTech Associates, Inc., (Ontario, Canada)
Rosemount Analytical, Inc., (Irvine, CA)
YSI, Inc.,(Yellow Springs, OH)
Applied Microsystems Ltd. (Sidney, BC, Canada)
Analytical Technology, Inc., (Collegeville, PA)
Isco, Inc., (Lincoln, NE)
Clarion Systems, Inc., (Indianapolis, IN)
GE/Sievers Ionics, Inc., (Boulder, CO)
Wallace and Tiernan Products (Vineland, NJ)
An important consideration in using online water quality monitors is the design and placement of sensor
stations in the distribution system, and the capital and operating costs associated with installation and
operation of an online water quality monitoring network. These topics are discussed in Sections 5.1 and
4.1, respectively, of WaterSentinel System Architecture (USEPA, 2005b). The remainder of this paper
will summarize EPA and vendor studies that have been conducted to test the response of various water
quality parameters to specific contaminants of concern, including some studies evaluating sensor
performance during periods of extended operation. All the contaminant data presented in this paper are
for concentrations well below the literature LD50 values (WaterSentinel Contaminant Fact Sheets,
USEPA, 2005d). Research is underway by the EPA's NHSRC to determine the health risks associated
with consuming sub-acute concentrations of WS priority contaminants. As that information becomes
available it will likely be used in the design of future studies.
Table 2-2 presents a brief description of typical water quality parameters, how each is measured, and how
utilities use the information to assess water quality and refine treatment, if necessary (Shaw
Environmental Inc., 2004; American Society of Civil Engineers (ASCE), 2004). The parameters listed in
Table 2-2 have been tested for their ability to respond to contaminants of concern; however, a smaller
subset of parameters should form the core capability for the online water quality monitoring component
of the WS-CWS. The parameters of pH, conductivity, total or free chlorine and TOC have proved to be
the most sensitive and class specific responders to contaminants of concern.
DRAFT-121205
-------
WS Online Water Quality
Table 2-2. Typical Water Quality Parameters
Parameter
PH
Conductivity
Free chlorine
Total chlorine
ORP
TOC
Dissolved oxygen
(DO)
Chloride
Ammonia
Nitrate
Turbidity
Modes of Online Detection
Proton selective glass bulb
electrode, proton selective metal
oxide
Annular ring electrode, nickel
electrode, titanium or noble metal
electrode
Polarographic membrane, 3-
electrode voltametric, colorimetric
Colorimetric
Potentiometric, platinum or noble
metal electrode
UV-persulfate digestion with near
infrared detection or membrane
conductometric detection of CC>2.
Membrane electrode, 3-electrode
voltametric
Ion selective electrode
Ion selective electrode
Ion selective electrode
Nephelometric (light scattering)
method
How Information is Routinely Used
pH is controlled for disinfection and corrosion
control in distributions systems. Also, the
formation of some disinfection byproduct (DBP)
is pH dependant.
Ability of water to carry an electrical current. A
strong indicator of the concentration of dissolved
solids.
Critical to demonstrate that disinfection
requirements are met in the plant. In the
distribution system, chlorine residual monitoring
is important to ensure that, 1) detectable residual
levels are maintained at all points in the system
as required by the SWTR; and 2) that maximum
residual disinfectant levels are not exceeded as
required by the Stage 1 D/DBP Rule. Also
important for controlling biofilms, regrowth,
nitrification, and other water quality problems.
ORP values above 700 millivolts (mV) kill
chlorine-sensitive organisms in drinking water. A
groundwater incursion may lower ORP by
increasing chlorine demand. Chlorination of
drinking water produces an ORP background of
-700 mV.
Dissolved and particulate organic carbon.
Concentrations in finished water can range from
less than 1.0 mg/Lto more than 10 mg/L. In any
case, TOC concentrations are typically stable in
distributed water from a single source. TOC is
used as an indicator of DBP formation potential.
Usually stable and near saturation for surface
waters. A decrease may be an indication of
chemical and biochemical activity in water.
Indicator of salinity.
Naturally occurring form of nitrogen in the
nitrogen cycle. May be added during treatment
to form a combined chorine residual, which
greatly reduces DBP formation. Excess
ammonia can result in distribution system water
quality problems such as nitrification.
Essential nutrient for plants and animals. Nitrate
is the most soluble form of nitrogen. For plants
that have a combined distribution system
residual, can be a result of nitrification. Drinking-
water maximum contaminant level (MCL) is 10
mg/L.
Indicator of suspended matter and microscopic
organisms. Used as a process control tool in the
plant to ensure that regulatory mandated
treatment techniques for removal of pathogens
are met. Limited application in distribution
systems, but could be indicative of corrosion
problems or other degradation in the quality of
distributed water.
DRAFT-121205
-------
WS Online Water Quality
Section 3.0: Literature Review
Prior to September 11, 2001, the vulnerability of the U.S. water supply was recognized but very little
research had been done to improve monitoring of the distribution system (Clark and Deininger, 2000).
Indeed, monitoring of the distribution system typically included a minimal set of water quality parameters
for which data were collected at a low frequency. Models and physical testing show that water quality
readings can change rapidly, and over a short timeframe in a backflow or injection attack (King and Kroll,
2005a). After September 11, 2001, research accelerated in an effort to determine if conventional water
quality monitors could be deployed for extended periods of time and if the water quality parameters
would respond to chemical and biological contamination. Research using online water quality monitors
to detect contamination events has been performed sponsored primarily by manufacturers of water quality
monitors (Table 2-1), EPA programs, the Edgewood Chemical Biological Center (ECBC) at Department
of Defense, and the AwwaRF in collaboration with utilities and private companies.
Vendor and AwwaRF Studies - Hach Homeland Security Technologies (Hach HST), manufacturer of
TOC analyzers and multi-parameter probes, is one of the few companies to have performed extensive
testing of the ability of their equipment to respond to contaminants of interest to water security. They
also have developed proprietary algorithms for event detection. An interesting add-on feature with the
Hach HST system is a fully searchable library of contaminant signatures with the capability to add new
contaminant profiles available to the end-user. To date, Hach HST has tested more than 80 chemicals
(Kroll, 2004 and Kroll and King, 2005b) that include 12 of the chemicals on the WS contaminant list and
20 of the contaminants from the priority list (Table 3-1). This priority list was the starting point for the
WS contaminant selection process through which 33 contaminants were identified for consideration in
implementation in the initial WS pilot contaminants as described in WaterSentinel System Architecture
(USEPA, 2005b). The results of the Hach tests have revealed that all the agents listed in Table 3-1.
(subset of the 80 that have been tested) cause a distinct change in one or more of the measured parameters
pH, turbidity, free chlorine, conductivity and TOC, and trigger Hach's Event Monitor Trigger System
(EMTS) when tested at contaminant concentrations less than the LD50 values in simulated distribution
system studies (Kroll, 2004 and Kroll, 2005b).
Table 3-1. Select Contaminants that Trigger the Hach Event Monitor Trigger System
Aflatoxin B-i (biotoxin)
Aldicarb (insecticide)
Arsenic trioxide (toxic metal)
Carbofuran (insecticide)
Colchcine (anti-inflammatory)
Dicamba (pesticide)
Dichlorvos (insecticide)
Diesel
£. co// (bacterial agent)
Ethoprophos (nerve agent surrogate)
Ferricyanide (cyanide surrogate)
Gasoline (hydrocarbon)
Glyphosate (herbicide)
Lead nitrate (toxic metal)
Malathion (pesticide)
Mercuric chloride (toxic metal)
Methanol (industrial solvent)
Methomyl (agricultural insecticide)
Nicotine (insecticide, used free base)
Oxamyl (pesticide)
Paraquat (herbicide)
Phorate (insecticide)
Sodium cyanide (toxic agent)
Sodium fluoroacetate (rodenticide)
Strychnine nitrate (pesticide)
Thallium (metal)
To address chemical warfare agents, Hach HST has entered into a Cooperative Research and
Development Agreement (CRADA) with ECBC and the U.S. Army Corps of Engineers to assess which
water quality parameters change in response to Sarin, Soman, VX, Ricin and Anthrax (Kroll, 2005b).
ECBC is one of the few facilities in the country where testing that involves chemical or biological warfare
agents can be conducted.
DRAFT-121205
-------
WS Online Water Quality
There are two AwwaRF studies in progress that will utilize online water quality monitors in a drinking
water distribution system. In one of those studies, the contaminants sodium fluoroacetate, aldicarb,
sodium arsenate and sodium cyanide have been tested, and detected, by online water quality monitors
when injected into a recirculating pipe loop (Cook, et al., 2005). In the second AwwaRF study, event
detection algorithms are anticipated to be tested using the data from online water quality monitors placed
in the distribution systems of four different cities (AwwaRF, 2005).
EPA Studies - The EPA's Technology Testing and Evaluation Program (TTEP) provides unbiased
third-party evaluation of commercially available homeland security related technologies. TTEP
rigorously tests technologies against a wide range of performance requirements and specifications and
posts summary reports and verification statements on their website (NHSRC, 2004). In the TTEP, the
EPA is free to test, evaluate and compare performance of different vendor products. To date three
different multi-parameter water quality probes have been tested by Battelle under the auspices of TTEP
(NHSRC, 2005). The multi-parameter probes that were tested in the TTEP are listed in Table 3-2.
Table 3-2. Multi-Parameter Probe Technologies Tested by TTEP
Company
YSI, Inc. (Yellow Springs, OH)
General Oceanics, Inc. (Miami, FL)
AANDERAA Instruments, Inc.
(Norway)
Instrument Model
6600 Extended
Deployment System
Ocean Seven 316
RCM Mk II with Optode
3830
Water Quality Parameters
DO, conductivity, temperature, pH,
turbidity and chlorophyll
DO, conductivity, temperature, pH
and turbidity
DO, temperature and turbidity
The three technologies were evaluated in field studies with respect to accuracy, relative bias, inter-unit
reproducibility and precision in three waters over a period of extended deployment (2.5 months). In
general, precision and accuracy was comparable for the common parameters (DO, conductivity, pH and
temperature). Turbidity was found to be highly variable. In fact, turbidity is not a good primary indicator
of contamination due to its high variability. The TTEP testing did not involve contaminant detection but
does validate the precision, accuracy and bias of water quality probes in extended deployment
environments such as would exist in the distribution system.
The Environmental Technology Verification (ETV) Program, established in 1995, uses voluntary vendor
participation with stakeholder oversight to test and evaluate innovative technologies for use in
environmental applications. In the ETV program, test plans are prepared with developers of new
technologies and the tests conducted by an independent third party. After the results of those tests have
been compiled and evaluated, verification reports and verification statements are posted on the EPA's
ETV website (ETV, 2005). In 2005, five different multi-parameter online water quality probes were
tested to independently verify the capability of the water quality monitors to respond and detect
contamination events when deployed in a simulated distribution system. The reports of these studies are
available in draft form but should appear on the EPA's ETV website when the reports are finalized
(USEPA, 2005e,f,g,h,i). Two identical monitors from each vendor were tested before and after extended
deployment for 52 days in a recirculating pipe loop distribution system simulator (DSS). The vendors,
models and parameters tested are listed in Table 3-3.
DRAFT-121205
-------
WS Online Water Quality
Table 3-3. Multi-Parameter Probes Tested by the ETV Program for Contaminant Detection
Manufacturer
Hach Company, (Loveland,
CO)
Rosemount Analytical, Inc.
(Irvine, CA)
Clarion Systems, Inc.
(Indianapolis, IN)
Analytical Technology, Inc.
(Collegeville, PA)
ManTech Associates, Inc.
(Ontario, Canada)
Model
Water Distribution Monitoring Panel
(WDMP)
WQS Unit
Sentinal 500
Q45WQ
TitraSip SA
Water Quality Parameters
total chlorine, turbidity,
temperature, conductivity, pH,
and TOC
free chlorine, turbidity,
temperature, conductivity, and
ORP
free chlorine, temperature,
conductivity, pH, and ORP
free chlorine, turbidity,
temperature, conductivity, pH,
and ORP
total chlorine, turbidity,
temperature, conductivity, pH,
and total alkalinity
The ETV tests were conducted by dissolving contaminant (nicotine, arsenic trioxide, aldicarb, and E. coll
in growth medium) in five gallons of dechlorinated water and injecting the five gallons into a
recirculating pipe loop of Cincinnati tap water. Tests were performed at the EPA's Test and Evaluation
(T&E) Facility in Cincinnati, Ohio. The selected contaminants are on the WS contaminant list and
represent organic, inorganic and biological contaminants. The final concentration of contaminant
recirculating in the loop was 10 mg/L, well below the LD50 values for these contaminants. To increase
the solubility of the arsenic trioxide, the pH of the water used to dissolve the arsenic trioxide was adjusted
to pH 12. Prior to, and at 3, 15 and 30 minutes after contaminant injection, water samples were collected
and the water quality parameters were measured off-line using reference methods. A result was
considered confirmed if a test probe detected a change in water quality as a result of contaminant
injection that was also detected by the reference method. All contaminant injections were performed in
duplicate. After 52 days of deployment in the recirculating pipe loop, additional experiments were
performed to evaluate the ability of the probes to detect contaminant injections of aldicarb and E. coll in
growth medium (E. coll represents the class of chlorine-sensitive contaminants on the WS contaminant
list). A summary of the results from all five ETV tests are presented in Table 3-4. Table 3-4 indicates
only if a change was detected; however, the directional change in water quality parameter can be found in
the full ETV reports.
Table 3-4. Summary of Five ETV Studies Using Multi-Parameter Probes to Detect Contamination
Contaminant
Nicotine
Arsenic
Trioxide
Aldicarb
Company
Hach
Rosemount
Clarion
ATI
ManTech
Hach
Rosemount
Clarion
ATI
ManTech
Hach
Rosemount
Clarion
ATI
ManTech
Total
CI2
Free
CI2
Turbidity
Temperature
-
Conductivity
-
PH
-
TOC
ORP
DRAFT-121205
-------
WS Online Water Quality
Contaminant
Company
Total
CI2
Free
CI2
Turbidity
Temperature
Conductivity
PH
TOC
ORP
After Extended Deployment
EColi
Aldicarb
Hach
Rosemount
Clarion
ATI
ManTech
Hach
Rosemount
Clarion
ATI
ManTech
-
= detectable change
- = no observed change
Gray cells = parameter not monitored
Regardless of probe manufacturer or contaminant type, changes in total or free chlorine were able to
detect the presence of contaminants, prior to, and after extended deployment. Since contaminant reaction
with chlorine is an oxidation/reduction reaction, ORP tracked changes in free or total chlorine very well.
The data would indicate that contaminant injections were detected by turbidity; however, there is less
certainty with these data due to the large variability observed by the reference method. It is hypothesized
that the turbidity changes may have been due to small air bubbles that could have been introduced during
contaminant injection. Turbidity is thus not a good primary indicator of contaminant injection.
Conductivity and pH changed with arsenic trioxide injection perhaps due to the pH adjustment required to
get the arsenic trioxide in solution. Aldicarb and nicotine did not cause any changes in pH or
conductivity in the pre-extended deployment experiments. In the extended deployment experiments,
however, aldicarb appeared to elicit a change in pH in 4 of the 5 probes tested. The reference methods
corroborated the changes detected by the probes yet it is not known why the pH did not change in
previous experiments with aldicarb. TOC was able to detect the organic compounds aldicarb and
nicotine, whereas, arsenic trioxide did not elicit a change in TOC.
Numerous studies involving contaminant detection using water quality monitors have been performed at
the EPA's T&E Center in Cincinnati, Ohio (USEPA, 2005J); Water Information Sharing and Analysis
Center (WaterlSAC), 2005; Hall, et al., 2005). The specific program in which the research has been
performed is the Water Assessment Technology Evaluation Research and Security (WATERS) Center.
Within the WATERS Center there are multiple DSSs used to evaluate and understand the dynamics that
influence water quality within the distribution system infrastructures typical in the U.S. Previously
unpublished research performed by the EPA at the WATERS Center is presented as a Case Study below.
Similar to the findings of other studies, total and free chlorine were the most sensitive water quality
parameters, showing significant changes from baseline for the majority of chemicals and biological media
tested. Changes in TOC were significant for organic contaminants and biological media. Conductivity
was found to be a highly stable water quality parameter with response to contaminants tending to be
detectable, but low. Turbidity response appeared to be sensitive but was highly erratic. The summary
data presented in the Case Study show the percent changes in water quality parameters. The significance
associated with that percent change is an estimation based upon the variability of the parameter typically
observed in daily operations at the T&E Facility.
DRAFT-121205
10
-------
WS Online Water Quality
Section 4.0: Case Study
Commercially available continuous, online sensors designed to monitor pH, free chlorine, total chlorine,
ORP, DO, conductivity, turbidity, TOC, chloride, ammonia, and nitrate were tested in a pilot-scale system
at the WATERS Center within the EPA's T&E Facility, Cincinnati, Ohio.
4.1 Experimental
Two modes of distribution pipe simulation were tested: a recirculating pipe-loop and a single-pass pipe.
Table 4-1 contains typical values, with daily variation, for water quality parameters of Cincinnati tap
water that was used at the T&E facility. Daily variation was approximately 10% for a volatile parameter
like free chlorine and less than 2% for stable parameters like TOC and conductivity. Over a long time
period (i.e. one year) values can vary by 20% or more, but this is due to seasonal and operational changes.
Table 4-1. Typical Water Quality Parameter Values with Daily Variability
Water Quality Parameter*
Total Chlorine
Free Chlorine
Total Organic Carbon
Oxidation Reduction Potential
Conductivity
PH
DO
Nitrate-Nitrogen
Chloride
Turbidity
Typical Value
1.1 +0.1 mg/L
1.0 + 0.1 mg/L
0.6 + 0.01 mg/L
650 + 20 mV
375 + 5 uS/cm
8.5 + 0.1
7.0 + 0.1 mg/L
4.0 + 0.1 mg/L
30 + 2 mg/L
0.5 + 0.1 NTU
*Data collected once every minute, except TOC which was collected once every 4-8 minutes.
In the recirculating pipe-loop mode, water continuously circulated through a 150-gallon capacity pipe-
loop (75 feet by 6-inch diameter unlined, cast-iron) and a 100 gallon recirculation tank. A separate 30-
gallon feed tank added Cincinnati Water Works chlorinated tap water to the loop at a rate (0.16 gallons
per minute (gpm)) such that the entire volume of water in the loop and recirculation tank was replaced
every 24 hours. Water quality monitors for total chlorine, free chlorine, ORP, conductivity, TOC,
chloride, nitrate and ammonia were placed 70 feet downstream from the point of contaminant injection
into the pipe-loop. Under these conditions, contaminants reached the monitors approximately 75 seconds
after injection. The water quality parameters' response profiles reflected this design, with responses to
the contaminants persisting until dilution, or chemical degradation via hydrolysis or reaction with free
chlorine, resulted in a gradual return to baseline. The following contaminants were evaluated in the
recirculating pipe-loop mode: potassium ferricyanide, a malathion insecticidal formulation (the
insecticide Real Kill), a glyphosate herbicidal formulation (the weed killer Round-Up), and a
secondary wastewater effluent. Table 4-2 lists the online monitors tested in the recirculating pipe-loop
mode.
DRAFT-121205
11
-------
WS Online Water Quality
Table 4-2. Online Monitors Tested in Recirculating Pipe-loop Mode
Single Parameter Probes
ATI A15
Hach, CI-17*
Hach, 1720D*
GLI Model PHD*
GLI Model 3422*
Hach Astro TOC UV Process Analyzer
Multi-Parameter Probes
Dascore, Six-Sense Sonde
YSI, 6600 Sonde
Hach, Hydrolab Data Sonde 4a
Water Quality Parameters
Free chlorine
Free and total chlorine
Turbidity
PH
Conductivity
TOC
Water Quality Parameters
Conductivity, DO, ORP, pH, temp, free chlorine
Conductivity, DO, ORP, pH, temp, ammonia-
nitrogen, chloride, nitrate-nitrogen, turbidity
Conductivity, DO, ORP, pH, temp, ammonia-
nitrogen, chloride, nitrate-nitrogen, turbidity
*These sensors were contained within the same panel (Aquatrend).
In single-pass pipe mode, water flowed through a plastic-lined ductile iron pipe that was 1,200 feet long
by 3 inches in diameter in a single pass at 20 gpm (no recirculation). The same water quality monitors
used in the recirculating pipe-loop experiments were used in the single-pass pipe experiments (Table 4-
3). The specific water quality parameters tested were free chlorine, total chlorine, chloride, conductivity,
DO, ORP, pH and turbidity. The monitors were placed at 80 and 1,100 feet downstream from the point of
contaminant injection into the single-pass pipe. The water quality parameters' response profiles reflected
a single-pass design, with responses to the contaminants persisting for approximately 20 minutes, with a
rapid return to baseline. The following contaminants were evaluated in the single-pass pipe mode:
aldicarb, glyphosate (not a store-bought formulation), colchicine (an anti-inflammatory), dicamba (an
herbicide), dimethyl sulfoxide (DMSO) (solvent for chemical), lead nitrate, mercuric chloride, nicotine,
potassium ferricyanide, sodium thiosulfate (reducing agent), sucrose (found in most biological growth
media) and various growth media for biological contaminants.
4.2 Results
Recirculating Pipe-Loop Mode - Event detection algorithms were not used in these experiments.
Without event detection software, it was necessary to estimate the significance of changes in water quality
parameters. This was done by establishing a stable baseline in parameter response prior to injection of the
contaminant, and measuring the percent change of that response from baseline when the contaminant was
injected. Knowledge of the within day variability (Table 4-1) was used to estimate the significance
change in a parameter response from baseline. Table 4-3 shows the percent changes from baseline values
for a small set of contaminants. Cells in the table (and in subsequent tables) that are highlighted in green
indicate that the change in response was less than 10%. This was deemed to be a marginally significant
change from baseline response. Cells in the table that are highlighted in yellow indicate that the change
in response was between 10% and 50% which was deemed to be a significant change from baseline.
Cells that are highlighted in red indicate a change greater than 50% which was deemed to be highly
significant.
DRAFT-121205
12
-------
WS Online Water Quality
Table 4-3. Online Monitor Responses to Select Contaminants in Recirculating Pipe-Loop Mode**
Contaminant
Potassium
Ferricyanide
Malathion
Formulation
Glyphosate
Formulation
Secondary
Wastewater
Effluent
Amount
Added
2g*
15 g*
0.04 g*
1.0g*
1.0g*
2 gallons
Percent Change in Response From Baseline
Total CI2
10-50%
>50%
<10%
10-50%
>50%
>50%
Free CI2
10-50%
>50%
<10%
10-50%
>50%
10-50%
ORP
<10%
10-50%
<10%
<10%
10-50%
10-50%
TOC
10-50%
>50%
10-50%
10-50%
10-50%
<10%
cr
>50%
>50%
<10%
<10%
<10%
10-50%
NO3"
>50%
>50%
<10%
<10%
<10%
10-50%
NH3"
10-50%
>50%
<10%
<10%
<10%
10-50%
Green = marginally significant, Yellow = significant, Red = highly significant
"""Monitor output recorded after 15 minutes of contaminant recirculation.
*g = grams. Final concentrations of contaminants in circulation were 0.04 -15 mg/L.
No one sensor responded to all contaminants, however, total and free chlorine, ORP and TOC yielded the
most significant responses to the widest variety of contaminants and concentrations tested. At the
concentrations used in these experiments, the change in conductivity from baseline was less than 10%.
Understanding the mode of detection for these online monitors is essential since each has interferences,
fouling and maintenance issues that should be considered when interpreting the data (USEPA, 2005J).
For example, the nitrate and chloride ion selective electrodes are likely responding to interference from
the potassium ion in the potassium ferricyanide and not to nitrate or chloride ions.
Regardless of manufacturer, the online monitors were able to detect the presence of contaminants in the
recirculating pipe-loop experiments with total and free chlorine, ORP, and TOC exhibiting the most
significant changes with contaminant introduction. Further studies are planned to determine the threshold
responses of the online monitors relative to toxicity and nuisance levels for real and surrogate
contaminants.
Single-Pass Pipe Mode - Table 4-4 shows the percent change from baseline for a subset of all the
contaminants tested at three different concentrations in the single-pass pipe.
DRAFT-121205
13
-------
WS Online Water Quality
Table 4-4. Online Monitor Responses to Select Contaminants in Single-Pass Pipe Mode at 80 Feet
from the Point of Injection
Contaminant
Aldicarb
Glyphosate
DMSO
Nicotine
Potassium
Ferricyanide
Sodium
Thiosulfate
Initial Cone.
In Pipe (mg/L)
0.2
1.1
2.2
0.4
1.5
3.0
0.6
2.0
4.0
0.4
1.9
3.8
0.6
1.6
3.2
0.2
1.3
2.6
Percent Change in Response From Baseline
Total CI2
<10%
10-50%
>50%
10-50%
10-50%
>50%
<10%
<10%
10-50%
<10% <10%*
10-50% 10-50%*
10-50% 10-50%*
<10%
10-50%
10-50%
10-50%
>50%
>50%
Free CI2
<10%
10-50%
>50%
10-50%
>50%
>50%
<10%
10-50%
10-50%
<10% 10-50%
ORP
<10%
<10%
10-50%
<10%
10-50%
10-50%
<10%
<10%
<10%
<10%
10-50% >50% <10%
10-50% >50% 10-50%
<10%
<10%
<10%
10-50%
>50%
>50%
<10%
<10%
<10%
<10%
10-50%
10-50%
cr
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
10-50%
10-50%
<10%
<10%
<10%
DO
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
Green = marginally significant, Yellow = significant, Red = highly significant
*Percent change at 1,100 feet. Nicotine showed a greater change in total and free chlorine at 1,100 feet from the
point of injection due to slow reaction kinetics.
Ion selective electrodes, like chloride, did not respond to contaminants as strongly in the single pass loop
as in the recirculating loop. In general, recirculation experiments allow time for chemical reactions to
approach equilibrium. Thus, while the single pass experiments more accurately simulate the actual
hydraulics of a distribution system, the recirculation experiments better simulate reactions that proceed at
a slower rate. The single-pass pipe mode represents the worst case scenario for injection of a
contaminant: the sensors detect a contaminant close to the point of injection where reactions that may
trigger the sensors have not fully developed.
Turbidity data are not presented in Table 4-4 since they were erratic and not a good primary predictor of
contamination. Conductivity and pH were monitored and did not exhibit a change in baseline response
greater than 5% for any contaminant or concentration tested so are not presented in Table 4-4. This does
not mean that the changes were insignificant, especially for a very stable parameter such as conductivity
which only varies approximately 2% from baseline over a day. Colchicine, dicamba, lead nitrate,
mercuric chloride and sucrose were tested from 0.4-4.0 mg/L and elicited changes in baseline responses
less than 10% for all contaminants and tested concentrations for all parameters. These data are not
presented in the table for the sake of brevity.
Figure 4-1 shows the change in free chlorine in response to injection of nicotine (3.8 mg/L) at two points
in the single-pass pipe; 80 and 1,100 feet from injection. The experiments were performed in duplicate.
It is apparent that the drop in free chlorine was greater at 1,100 feet than at 80 feet. This is due to the
slower reaction kinetics of chlorine with nicotine.
DRAFT-121205
14
-------
WS Online Water Quality
0.0
-30 -20 -10
10 20 30 40 50 60 70 80
Time (min)
Figure 4-1. Change in Free Chlorine to Nicotine at 80 and 1,100 ft from Injection
Figure 4-2 shows the response of free chlorine as a function of nicotine concentration. The figure shows
the linear relationship between free chlorine loss and nicotine concentration and also shows that the rate
of loss of free chlorine was greater at 1,100 feet than at 80 feet. This indicates that the reaction between
chlorine and nicotine had not reached equilibrium in the very short travel times evaluated in these single
pass experiments.
0.00
S" -0.10 -
D)
£ -0.20 -
11 n ir
c -U.oL
| -0.40 -
" -0.50 -
£ -0.60 -
~ -0.70 -
2 -0.80 -
-0.90
0.0
y = -0.0978x-0.0297
R2= 0.9989
80ft
y = -0.1852x-0.0531
R2 = 0.9999
1.0 2.0 3.0
Initial Nicotine Concentration (mg/L)
4.0
Figure 4-2. Free Chlorine Response with Increasing Initial Nicotine Concentration
The LD50 for nicotine is 50 mg/kg (oral, rat). This would correspond to a concentration in drinking water
of 1,750 mg/L for a 70 kg person consuming 2 liters of water per day. The concentrations of nicotine
tested in these experiments were between 0.4 and 3.8 mg/L, well below the drinking water equivalent
DRAFT-121205
15
-------
WS Online Water Quality
level of 1,750 mg/L. Similar sensitivity testing for other contaminants based on their toxicity should be
performed in future research.
In separate experiments, common biological growth media at various concentrations were injected into
the single-pass pipe and water quality parameters of total chlorine, free chlorine, chloride, conductivity,
DO, ORP, pH, and turbidity were measured at 80 and 1,100 feet from the point of injection. Growth
media were tested because in an intentional contamination event, a biological agent would most likely be
injected into a pipe in growth media. Table 4-5 contains only the results from measurements taken at
1,100 feet, because, similar to nicotine, more significant results were observed at 1,100 feet versus 80 feet
from the point of injection due to reaction kinetics.
Table 4-5. Online Monitor Responses to Biological Growth Media in Single-Pass Pipe Mode at
1,100 Feet from the Point of Injection
Biological
Growth
Media
Nutrient Broth
Trypticase
Soy Broth
Terrific Broth
£. co// in
Terrific Broth
Initial Cone.
In Pipe
(mg/L)*
0.12
0.48
0.96
0.12
0.48
0.96
0.12
0.47
0.97
0.12
0.47
0.97
Percent Change in Response From Baseline
Total CI2
<10%
<10%
<10%
<10%
10-50%
10-50%
<10%
<10%
<10%
10-50%
10-50%
10-50%
Free CI2
<10%
<10%
10-50%
<10%
10-50%
10-50%
<10%
10-50%
10-50%
10-50%
>50%
>50%
ORP
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
cr
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
10-50%
DO
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
Green = marginally significant, Yellow = significant, Red = highly significant
*Concentration of growth media.
It is apparent from Table 4-5 that free and total chlorine are the water quality parameters most sensitive to
biological contaminant incursion.
Total and free chlorine were the most sensitive water quality parameters, showing significant changes
from baseline in both the recirculating and single-pass pipe experiments. Changes in TOC were
significant for organic contaminants in recirculating pipe-loop mode. Conductivity and pH did not
display changes in response greater than 5% for the contaminants and concentrations tested; however, the
concentrations used in most experiments were well below lethal concentrations. Again, turbidity was
erratic and not a good primary predictor of contamination. In the single-pass pipe mode, ORP showed
marginally significant to significant change in baseline response.
To distinguish the normal variability of water quality parameters from real anomalies that could indicate
contamination, it is necessary to use an EDS (USEPA, 2005a). It should be noted that results from the
EPA's T&E experiments were reported as an observed percent change from an estimated baseline and
that the use of an event detection system may have resulted in lesser or greater sensitivity to these
contaminants. Future work should utilize event detection software and test a wider variety of chemical
and biological contaminants at concentrations of interest. Online monitoring of radiological
contamination has unique considerations that should also be addressed in future research.
DRAFT-121205
16
-------
WS Online Water Quality
Section 5.0: Summary and Conclusions
In summary, the work done by manufacturers of online water quality monitors and EPA demonstrate that
online water quality monitors are sensitive and responsive to contaminant incursion (Kroll, 2005a; Shaw
Environmental Inc., 2004; Kroll and King, 2005a; Cook, et al., 2005; USEPA, 2005J; WaterlSAC, 2005;
Hall, et al., 2005; Kroll and King 2005c). The water quality parameters most sensitive to contaminants
of concern to water security are free or total chlorine, total organic carbon, conductivity and pH.
Oxidation/reduction potential corroborates the chlorine sensor results. Other parameters such as chloride,
nitrate and ammonia have been observed to change in the presence of contaminants but mostly due to
interferences of concomitant ions. Turbidity, which can be highly variable, is not a good primary
indicator of contamination.
The WS contaminant selection process identified 33 contaminants for consideration in implementation of
the WS pilot. Based on the means by which contaminants could be detected in the proposed WS
contaminant warning system, EPA classified contaminants into 12 categories (WaterSentinel
Contaminant Selection, 2005k). Online water quality monitoring data from distribution system
simulation studies were used to inform the process of contaminant classification. Contaminants that
impact two or more water quality parameters have a high potential for detection by online monitoring. If
the presence of the contaminant only has an impact on a single water quality parameter, online monitoring
provides a moderate potential for detection.
Online water quality monitoring is just one component in the WS System Architecture (USEPA, 2005b).
The purpose of online water quality monitoring in the WS design is to serve as a means for a utility to
detect potential contamination and initiate an appropriate response. The subsequent role of field and
laboratory testing is to determine if a contamination event occurred and, if so, what contaminant was
introduced.
DRAFT-121205 17
-------
WS Online Water Quality
Section 6.0: References
Alai, M., L. Glascoe, A. Love, M. Johnson and W. Einfeld. Sensor Acquisition for Water Utilities:
Survey, Down Selection Process and Technology List. Report by the Lawrence Livermore National
Laboratory, Department of Energy Contract No. W-7405-Eng-48. June 30, 2005.
American Society of Civil Engineers. Interim Voluntary Guidelines for Designing an Online Contaminant
Monitoring System. U.S. EPA Cooperative Research and Development Agreement, X-83128301-0,
December 9, 2004.
AwwaRF, Data Processing and Analysis for On-line Distribution System Monitoring (AwwaRF Project
3035), Presentation at the Distribution System On-line Monitoring and Security Summit, Charleston,
S.C., October 3-4, 2005.
Clark, R.M. and R.A. Deininger. Protecting the Nation's Critical Infrastructure: The Vulnerability of U.S.
Water Supply Systems. Journal of Contingencies and Crisis Management. 8(2). pp. 73-80, June 2000.
Cook, J., E. Roehl and R. Daamen. Decision Support System for Water Distribution System Monitoring
for Homeland Security. Proceedings of the AWWA Water Security Conference, Oklahoma City, OK,
April 10-12,2005.
Hasan, J., S. States and R. Deininger. Safeguarding the Security of Public Water Supplies Using Early
Warning Systems: A Brief Review. J. of Contemporary Water Research and Education. 129. pp. 27-33.
2004.
Hall, J., A.D. Zaffiro, R.B. Marx, P.C. Kefauver, E.R. Krishnan, RC. Haught and J.G. Herrmann. On-line
Water Quality Parameters as Indicators of Distribution System Contamination. Submitted toJAWWA,
July, 2005.
ICF Consulting, Inc. Technologies and Techniques for Early Warning Systems to Evaluate and Monitor
Drinking Water Quality: A State-of-the-Art Review. Draft report prepared under EPA Contract 68-C-02-
009, WA 3-53, Work Assignment Manager, Jafrul Hasan, July 8, 2005.
King, K. and D. Kroll. Trigger and Detection Method for Threat Agents in Drinking Water. Proceedings
of The International Society for Optical Engineering, Optics and Photonics in Global Homeland Security,
Vol. 5781, pp. 63-74, T. Saito, editor, published online August 5, 2005a.
Kroll, D. Water Distribution Monitoring: Opportunities and Challenges for Enhancing Water Quality and
Security. A Presentation to the National Research Council's Committee on Public Water Supply
Distribution Systems: Assessing and Reducing Risks, April 19, 2005, Washington, D.C.
Kroll, D., Advanced Analytical Systems for Water Quality Security. Proceedings of the AWWA Annual
Conference and Exhibit, Security Session, June 14, 2004.
Kroll, D. and K. King, "Validation and Testing of the Operational Effectiveness of an On-line Security
Platform for the Water Distribution System," presented at the Water Quality Technology Conference,
Quebec City, Canada, November 7, 2005b.
Kroll, D. and K. King. Operational Validation of an On-line System for Enhancing Water Security in the
Distribution System. Proceedings of the AWWA Water Security Congress, Oklahoma City, OK, April
10-12, 2005c.
DRAFT-121205 19
-------
WS Online Water Quality
Schlegel, J.A. Automated distribution system monitoring supports water quality, streamlines system
management, and fortifies security. J.AWWA, January, 2004.
Shaw Environmental Inc. Water Quality Sensor Responses to Chemical and Biological Warfare Agent
Simulants in Water Distribution Systems. EPA Contract No. EP-C-04-034, Work Assignment No. 0.06,
August, 2004.
States, S., J. Newberry, J. Wichterman, J. Kuchta, M. Scheuring and L. Casson. Rapid Analytical
Techniques for Drinking Water Security Investigations. J. AWWA, 96 (1), pp. 52-64, January, 2004.
USEPA. Overview of Event Detection Systems for Water Sentinel, 2005a. For Official Use Only
USEPA. Water Sentinel System Architecture, 2005b. For Official Use Only.
USEPA. WaterSentinel Contaminant Selection, 2005c. SENSITIVE. For Official Use Only, Limited
Distribution.
USEPA. WaterSentinel Contaminant Fact Sheets, 2005d. SENSITIVE. For Official Use Only, Limited
Distribution.
USEPA. Hach Company Water Distribution Monitoring Panel and Event Monitor Trigger System.
Draft Environmental Technology Verification Report, USEPA, September, 2005e.
USEPA. Clarion Sensing Systems, Inc. Sentinal 500 Series Continuous Water Quality Monitors. Draft
Environmental Technology Verification Report, USEPA, July, 2005f.
USEPA, Rosemount Analytical Model WQS Continuous Multi-Parameter Water Quality Monitor. Draft
Environmental Technology Verification Report, USEPA, August, 2005g.
USEPA, Analytical Technology, Inc Q45WQ Water Quality Monitors Continuous Water Quality
Monitors. Draft Environmental Technology Verification Report, USEPA, July, 2005h.
USEPA, Man-Tech Associates Inc. TitraSip SA System Continuous Multi-Parameter Water Quality
Monitor," Draft Environmental Technology Verification Report, USEPA, August, 2005i.
USEPA. Evaluation of Water Quality Sensors as Devices to Warn of Intentional Contamination in Water
Distribution Systems. EPA Report 600/R-05/105, uploaded to WaterlSAC, September 2005J.
USEPA. WaterSentinel Contaminant Selection Document, 2005k. SENSITIVE. For Official Use Only.
USEPA, NHSRC, Homeland Security Research, "News 3/10/05 - Technology Testing and Evaluation
Program" November 2005. http://www.epa.gov/NHSRC/news/news031005.htm.
USEPA, NHSRC, Homeland Security Research, "News 7/23/04 - Multi-Parameter Water Quality Probes"
November 2004. http://www.epa.gov/nhsrc/news/news072304.htm.
USEPA, Environmental Technology Verification Program November 2005n. http://www.epa.gov/etv/.
WaterlSAC, "Water Quality Sensor Responses to Potential Chemical Threats in a Pilot-Scale Water
Distribution System". Report pending upload to WaterlSAC. http://www.waterisac.org.
DRAFT-121205 20
-------
WS Online Water Quality
Appendix A: Acronym List
ASCE American Society of Civil Engineers
AwwaRF American Water Works Association Research Foundation
CRADA Cooperative Research and Development Agreement
CWS contamination warning system
DBF disinfection byproduct
DMSO dimethyl sulfoxide
DO dissolved oxygen
DSS distribution system simulator
ECBC Edge wood Chemical Biological Center
EDS event detection system
EMTS Event Monitor Trigger System
EPA U.S. Environmental Protection Agency
ETV Environmental Technology Verification
Hach HST Hach Homeland Security Technologies
gpm gallons per minute
MCL maximum contaminant level
mV Millivolts
NHSRC National Homeland Security Research Center
ORD Office of Research and Development
ORP oxidation/reduction potential
SCADA Supervisory Control and Data
SWTR Surface Water Treatment Rule
T&E Test and Evaluation
TOC total organic carbon
TTEP Technology Testing and Evaluation Program
WATERS Water Assessment Technology Evaluation Research and Security
WaterlSAC Water Information Sharing and Analysis Center
WDMP Water Distribution Monitoring Panel
WS WaterSentinel
WS-CWS WaterSentinel contamination warning system
WSD Water Security Division
DRAFT-121205
21
------- |